diff options
Diffstat (limited to 'target/linux/generic-2.6/patches-2.6.22')
48 files changed, 0 insertions, 44666 deletions
diff --git a/target/linux/generic-2.6/patches-2.6.22/001-squashfs.patch b/target/linux/generic-2.6/patches-2.6.22/001-squashfs.patch deleted file mode 100644 index 0807d1d..0000000 --- a/target/linux/generic-2.6/patches-2.6.22/001-squashfs.patch +++ /dev/null @@ -1,4170 +0,0 @@ ---- a/fs/Kconfig -+++ b/fs/Kconfig -@@ -1367,6 +1367,71 @@ - - If unsure, say N. - -+config SQUASHFS -+ tristate "SquashFS 3.0 - Squashed file system support" -+ select ZLIB_INFLATE -+ help -+ Saying Y here includes support for SquashFS 3.0 (a Compressed Read-Only File -+ System). Squashfs is a highly compressed read-only filesystem for Linux. -+ It uses zlib compression to compress both files, inodes and directories. -+ Inodes in the system are very small and all blocks are packed to minimise -+ data overhead. Block sizes greater than 4K are supported up to a maximum of 64K. -+ SquashFS 3.0 supports 64 bit filesystems and files (larger than 4GB), full -+ uid/gid information, hard links and timestamps. -+ -+ Squashfs is intended for general read-only filesystem use, for archival -+ use (i.e. in cases where a .tar.gz file may be used), and in embedded -+ systems where low overhead is needed. Further information and filesystem tools -+ are available from http://squashfs.sourceforge.net. -+ -+ If you want to compile this as a module ( = code which can be -+ inserted in and removed from the running kernel whenever you want), -+ say M here and read <file:Documentation/modules.txt>. The module -+ will be called squashfs. Note that the root file system (the one -+ containing the directory /) cannot be compiled as a module. -+ -+ If unsure, say N. -+ -+config SQUASHFS_EMBEDDED -+ -+ bool "Additional options for memory-constrained systems" -+ depends on SQUASHFS -+ default n -+ help -+ Saying Y here allows you to specify cache sizes and how Squashfs -+ allocates memory. This is only intended for memory constrained -+ systems. -+ -+ If unsure, say N. -+ -+config SQUASHFS_FRAGMENT_CACHE_SIZE -+ int "Number of fragments cached" if SQUASHFS_EMBEDDED -+ depends on SQUASHFS -+ default "3" -+ help -+ By default SquashFS caches the last 3 fragments read from -+ the filesystem. Increasing this amount may mean SquashFS -+ has to re-read fragments less often from disk, at the expense -+ of extra system memory. Decreasing this amount will mean -+ SquashFS uses less memory at the expense of extra reads from disk. -+ -+ Note there must be at least one cached fragment. Anything -+ much more than three will probably not make much difference. -+ -+config SQUASHFS_VMALLOC -+ bool "Use Vmalloc rather than Kmalloc" if SQUASHFS_EMBEDDED -+ depends on SQUASHFS -+ default n -+ help -+ By default SquashFS uses kmalloc to obtain fragment cache memory. -+ Kmalloc memory is the standard kernel allocator, but it can fail -+ on memory constrained systems. Because of the way Vmalloc works, -+ Vmalloc can succeed when kmalloc fails. Specifying this option -+ will make SquashFS always use Vmalloc to allocate the -+ fragment cache memory. -+ -+ If unsure, say N. -+ - config VXFS_FS - tristate "FreeVxFS file system support (VERITAS VxFS(TM) compatible)" - depends on BLOCK ---- a/fs/Makefile -+++ b/fs/Makefile -@@ -72,6 +72,7 @@ - obj-$(CONFIG_JBD2) += jbd2/ - obj-$(CONFIG_EXT2_FS) += ext2/ - obj-$(CONFIG_CRAMFS) += cramfs/ -+obj-$(CONFIG_SQUASHFS) += squashfs/ - obj-$(CONFIG_RAMFS) += ramfs/ - obj-$(CONFIG_HUGETLBFS) += hugetlbfs/ - obj-$(CONFIG_CODA_FS) += coda/ ---- /dev/null -+++ b/fs/squashfs/inode.c -@@ -0,0 +1,2122 @@ -+/* -+ * Squashfs - a compressed read only filesystem for Linux -+ * -+ * Copyright (c) 2002, 2003, 2004, 2005, 2006 -+ * Phillip Lougher <phillip@lougher.org.uk> -+ * -+ * This program is free software; you can redistribute it and/or -+ * modify it under the terms of the GNU General Public License -+ * as published by the Free Software Foundation; either version 2, -+ * or (at your option) any later version. -+ * -+ * This program is distributed in the hope that it will be useful, -+ * but WITHOUT ANY WARRANTY; without even the implied warranty of -+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the -+ * GNU General Public License for more details. -+ * -+ * You should have received a copy of the GNU General Public License -+ * along with this program; if not, write to the Free Software -+ * Foundation, 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. -+ * -+ * inode.c -+ */ -+ -+#include <linux/types.h> -+#include <linux/squashfs_fs.h> -+#include <linux/module.h> -+#include <linux/errno.h> -+#include <linux/slab.h> -+#include <linux/fs.h> -+#include <linux/smp_lock.h> -+#include <linux/slab.h> -+#include <linux/squashfs_fs_sb.h> -+#include <linux/squashfs_fs_i.h> -+#include <linux/buffer_head.h> -+#include <linux/vfs.h> -+#include <linux/init.h> -+#include <linux/dcache.h> -+#include <linux/wait.h> -+#include <linux/zlib.h> -+#include <linux/blkdev.h> -+#include <linux/vmalloc.h> -+#include <asm/uaccess.h> -+#include <asm/semaphore.h> -+ -+#include "squashfs.h" -+ -+static void squashfs_put_super(struct super_block *); -+static int squashfs_statfs(struct dentry *, struct kstatfs *); -+static int squashfs_symlink_readpage(struct file *file, struct page *page); -+static int squashfs_readpage(struct file *file, struct page *page); -+static int squashfs_readpage4K(struct file *file, struct page *page); -+static int squashfs_readdir(struct file *, void *, filldir_t); -+static struct inode *squashfs_alloc_inode(struct super_block *sb); -+static void squashfs_destroy_inode(struct inode *inode); -+static int init_inodecache(void); -+static void destroy_inodecache(void); -+static struct dentry *squashfs_lookup(struct inode *, struct dentry *, -+ struct nameidata *); -+static struct inode *squashfs_iget(struct super_block *s, squashfs_inode_t inode); -+static long long read_blocklist(struct inode *inode, int index, -+ int readahead_blks, char *block_list, -+ unsigned short **block_p, unsigned int *bsize); -+static int squashfs_get_sb(struct file_system_type *, int, -+ const char *, void *, struct vfsmount *); -+ -+ -+static z_stream stream; -+ -+static struct file_system_type squashfs_fs_type = { -+ .owner = THIS_MODULE, -+ .name = "squashfs", -+ .get_sb = squashfs_get_sb, -+ .kill_sb = kill_block_super, -+ .fs_flags = FS_REQUIRES_DEV -+}; -+ -+static unsigned char squashfs_filetype_table[] = { -+ DT_UNKNOWN, DT_DIR, DT_REG, DT_LNK, DT_BLK, DT_CHR, DT_FIFO, DT_SOCK -+}; -+ -+static struct super_operations squashfs_ops = { -+ .alloc_inode = squashfs_alloc_inode, -+ .destroy_inode = squashfs_destroy_inode, -+ .statfs = squashfs_statfs, -+ .put_super = squashfs_put_super, -+}; -+ -+SQSH_EXTERN struct address_space_operations squashfs_symlink_aops = { -+ .readpage = squashfs_symlink_readpage -+}; -+ -+SQSH_EXTERN struct address_space_operations squashfs_aops = { -+ .readpage = squashfs_readpage -+}; -+ -+SQSH_EXTERN struct address_space_operations squashfs_aops_4K = { -+ .readpage = squashfs_readpage4K -+}; -+ -+static struct file_operations squashfs_dir_ops = { -+ .read = generic_read_dir, -+ .readdir = squashfs_readdir -+}; -+ -+SQSH_EXTERN struct inode_operations squashfs_dir_inode_ops = { -+ .lookup = squashfs_lookup -+}; -+ -+ -+static struct buffer_head *get_block_length(struct super_block *s, -+ int *cur_index, int *offset, int *c_byte) -+{ -+ struct squashfs_sb_info *msblk = s->s_fs_info; -+ unsigned short temp; -+ struct buffer_head *bh; -+ -+ if (!(bh = sb_bread(s, *cur_index))) -+ goto out; -+ -+ if (msblk->devblksize - *offset == 1) { -+ if (msblk->swap) -+ ((unsigned char *) &temp)[1] = *((unsigned char *) -+ (bh->b_data + *offset)); -+ else -+ ((unsigned char *) &temp)[0] = *((unsigned char *) -+ (bh->b_data + *offset)); -+ brelse(bh); -+ if (!(bh = sb_bread(s, ++(*cur_index)))) -+ goto out; -+ if (msblk->swap) -+ ((unsigned char *) &temp)[0] = *((unsigned char *) -+ bh->b_data); -+ else -+ ((unsigned char *) &temp)[1] = *((unsigned char *) -+ bh->b_data); -+ *c_byte = temp; -+ *offset = 1; -+ } else { -+ if (msblk->swap) { -+ ((unsigned char *) &temp)[1] = *((unsigned char *) -+ (bh->b_data + *offset)); -+ ((unsigned char *) &temp)[0] = *((unsigned char *) -+ (bh->b_data + *offset + 1)); -+ } else { -+ ((unsigned char *) &temp)[0] = *((unsigned char *) -+ (bh->b_data + *offset)); -+ ((unsigned char *) &temp)[1] = *((unsigned char *) -+ (bh->b_data + *offset + 1)); -+ } -+ *c_byte = temp; -+ *offset += 2; -+ } -+ -+ if (SQUASHFS_CHECK_DATA(msblk->sblk.flags)) { -+ if (*offset == msblk->devblksize) { -+ brelse(bh); -+ if (!(bh = sb_bread(s, ++(*cur_index)))) -+ goto out; -+ *offset = 0; -+ } -+ if (*((unsigned char *) (bh->b_data + *offset)) != -+ SQUASHFS_MARKER_BYTE) { -+ ERROR("Metadata block marker corrupt @ %x\n", -+ *cur_index); -+ brelse(bh); -+ goto out; -+ } -+ (*offset)++; -+ } -+ return bh; -+ -+out: -+ return NULL; -+} -+ -+ -+SQSH_EXTERN unsigned int squashfs_read_data(struct super_block *s, char *buffer, -+ long long index, unsigned int length, -+ long long *next_index) -+{ -+ struct squashfs_sb_info *msblk = s->s_fs_info; -+ struct buffer_head *bh[((SQUASHFS_FILE_MAX_SIZE - 1) >> -+ msblk->devblksize_log2) + 2]; -+ unsigned int offset = index & ((1 << msblk->devblksize_log2) - 1); -+ unsigned int cur_index = index >> msblk->devblksize_log2; -+ int bytes, avail_bytes, b = 0, k; -+ char *c_buffer; -+ unsigned int compressed; -+ unsigned int c_byte = length; -+ -+ if (c_byte) { -+ bytes = msblk->devblksize - offset; -+ compressed = SQUASHFS_COMPRESSED_BLOCK(c_byte); -+ c_buffer = compressed ? msblk->read_data : buffer; -+ c_byte = SQUASHFS_COMPRESSED_SIZE_BLOCK(c_byte); -+ -+ TRACE("Block @ 0x%llx, %scompressed size %d\n", index, compressed -+ ? "" : "un", (unsigned int) c_byte); -+ -+ if (!(bh[0] = sb_getblk(s, cur_index))) -+ goto block_release; -+ -+ for (b = 1; bytes < c_byte; b++) { -+ if (!(bh[b] = sb_getblk(s, ++cur_index))) -+ goto block_release; -+ bytes += msblk->devblksize; -+ } -+ ll_rw_block(READ, b, bh); -+ } else { -+ if (!(bh[0] = get_block_length(s, &cur_index, &offset, -+ &c_byte))) -+ goto read_failure; -+ -+ bytes = msblk->devblksize - offset; -+ compressed = SQUASHFS_COMPRESSED(c_byte); -+ c_buffer = compressed ? msblk->read_data : buffer; -+ c_byte = SQUASHFS_COMPRESSED_SIZE(c_byte); -+ -+ TRACE("Block @ 0x%llx, %scompressed size %d\n", index, compressed -+ ? "" : "un", (unsigned int) c_byte); -+ -+ for (b = 1; bytes < c_byte; b++) { -+ if (!(bh[b] = sb_getblk(s, ++cur_index))) -+ goto block_release; -+ bytes += msblk->devblksize; -+ } -+ ll_rw_block(READ, b - 1, bh + 1); -+ } -+ -+ if (compressed) -+ down(&msblk->read_data_mutex); -+ -+ for (bytes = 0, k = 0; k < b; k++) { -+ avail_bytes = (c_byte - bytes) > (msblk->devblksize - offset) ? -+ msblk->devblksize - offset : -+ c_byte - bytes; -+ wait_on_buffer(bh[k]); -+ if (!buffer_uptodate(bh[k])) -+ goto block_release; -+ memcpy(c_buffer + bytes, bh[k]->b_data + offset, avail_bytes); -+ bytes += avail_bytes; -+ offset = 0; -+ brelse(bh[k]); -+ } -+ -+ /* -+ * uncompress block -+ */ -+ if (compressed) { -+ int zlib_err; -+ -+ stream.next_in = c_buffer; -+ stream.avail_in = c_byte; -+ stream.next_out = buffer; -+ stream.avail_out = msblk->read_size; -+ -+ if (((zlib_err = zlib_inflateInit(&stream)) != Z_OK) || -+ ((zlib_err = zlib_inflate(&stream, Z_FINISH)) -+ != Z_STREAM_END) || ((zlib_err = -+ zlib_inflateEnd(&stream)) != Z_OK)) { -+ ERROR("zlib_fs returned unexpected result 0x%x\n", -+ zlib_err); -+ bytes = 0; -+ } else -+ bytes = stream.total_out; -+ -+ up(&msblk->read_data_mutex); -+ } -+ -+ if (next_index) -+ *next_index = index + c_byte + (length ? 0 : -+ (SQUASHFS_CHECK_DATA(msblk->sblk.flags) -+ ? 3 : 2)); -+ return bytes; -+ -+block_release: -+ while (--b >= 0) -+ brelse(bh[b]); -+ -+read_failure: -+ ERROR("sb_bread failed reading block 0x%x\n", cur_index); -+ return 0; -+} -+ -+ -+SQSH_EXTERN int squashfs_get_cached_block(struct super_block *s, char *buffer, -+ long long block, unsigned int offset, -+ int length, long long *next_block, -+ unsigned int *next_offset) -+{ -+ struct squashfs_sb_info *msblk = s->s_fs_info; -+ int n, i, bytes, return_length = length; -+ long long next_index; -+ -+ TRACE("Entered squashfs_get_cached_block [%llx:%x]\n", block, offset); -+ -+ while ( 1 ) { -+ for (i = 0; i < SQUASHFS_CACHED_BLKS; i++) -+ if (msblk->block_cache[i].block == block) -+ break; -+ -+ down(&msblk->block_cache_mutex); -+ -+ if (i == SQUASHFS_CACHED_BLKS) { -+ /* read inode header block */ -+ for (i = msblk->next_cache, n = SQUASHFS_CACHED_BLKS; -+ n ; n --, i = (i + 1) % -+ SQUASHFS_CACHED_BLKS) -+ if (msblk->block_cache[i].block != -+ SQUASHFS_USED_BLK) -+ break; -+ -+ if (n == 0) { -+ wait_queue_t wait; -+ -+ init_waitqueue_entry(&wait, current); -+ add_wait_queue(&msblk->waitq, &wait); -+ set_current_state(TASK_UNINTERRUPTIBLE); -+ up(&msblk->block_cache_mutex); -+ schedule(); -+ set_current_state(TASK_RUNNING); -+ remove_wait_queue(&msblk->waitq, &wait); -+ continue; -+ } -+ msblk->next_cache = (i + 1) % SQUASHFS_CACHED_BLKS; -+ -+ if (msblk->block_cache[i].block == -+ SQUASHFS_INVALID_BLK) { -+ if (!(msblk->block_cache[i].data = -+ kmalloc(SQUASHFS_METADATA_SIZE, -+ GFP_KERNEL))) { -+ ERROR("Failed to allocate cache" -+ "block\n"); -+ up(&msblk->block_cache_mutex); -+ goto out; -+ } -+ } -+ -+ msblk->block_cache[i].block = SQUASHFS_USED_BLK; -+ up(&msblk->block_cache_mutex); -+ -+ if (!(msblk->block_cache[i].length = -+ squashfs_read_data(s, -+ msblk->block_cache[i].data, -+ block, 0, &next_index))) { -+ ERROR("Unable to read cache block [%llx:%x]\n", -+ block, offset); -+ goto out; -+ } -+ -+ down(&msblk->block_cache_mutex); -+ wake_up(&msblk->waitq); -+ msblk->block_cache[i].block = block; -+ msblk->block_cache[i].next_index = next_index; -+ TRACE("Read cache block [%llx:%x]\n", block, offset); -+ } -+ -+ if (msblk->block_cache[i].block != block) { -+ up(&msblk->block_cache_mutex); -+ continue; -+ } -+ -+ if ((bytes = msblk->block_cache[i].length - offset) >= length) { -+ if (buffer) -+ memcpy(buffer, msblk->block_cache[i].data + -+ offset, length); -+ if (msblk->block_cache[i].length - offset == length) { -+ *next_block = msblk->block_cache[i].next_index; -+ *next_offset = 0; -+ } else { -+ *next_block = block; -+ *next_offset = offset + length; -+ } -+ up(&msblk->block_cache_mutex); -+ goto finish; -+ } else { -+ if (buffer) { -+ memcpy(buffer, msblk->block_cache[i].data + -+ offset, bytes); -+ buffer += bytes; -+ } -+ block = msblk->block_cache[i].next_index; -+ up(&msblk->block_cache_mutex); -+ length -= bytes; -+ offset = 0; -+ } -+ } -+ -+finish: -+ return return_length; -+out: -+ return 0; -+} -+ -+ -+static int get_fragment_location(struct super_block *s, unsigned int fragment, -+ long long *fragment_start_block, -+ unsigned int *fragment_size) -+{ -+ struct squashfs_sb_info *msblk = s->s_fs_info; -+ long long start_block = -+ msblk->fragment_index[SQUASHFS_FRAGMENT_INDEX(fragment)]; -+ int offset = SQUASHFS_FRAGMENT_INDEX_OFFSET(fragment); -+ struct squashfs_fragment_entry fragment_entry; -+ -+ if (msblk->swap) { -+ struct squashfs_fragment_entry sfragment_entry; -+ -+ if (!squashfs_get_cached_block(s, (char *) &sfragment_entry, -+ start_block, offset, -+ sizeof(sfragment_entry), &start_block, -+ &offset)) -+ goto out; -+ SQUASHFS_SWAP_FRAGMENT_ENTRY(&fragment_entry, &sfragment_entry); -+ } else -+ if (!squashfs_get_cached_block(s, (char *) &fragment_entry, -+ start_block, offset, -+ sizeof(fragment_entry), &start_block, -+ &offset)) -+ goto out; -+ -+ *fragment_start_block = fragment_entry.start_block; -+ *fragment_size = fragment_entry.size; -+ -+ return 1; -+ -+out: -+ return 0; -+} -+ -+ -+SQSH_EXTERN void release_cached_fragment(struct squashfs_sb_info *msblk, struct -+ squashfs_fragment_cache *fragment) -+{ -+ down(&msblk->fragment_mutex); -+ fragment->locked --; -+ wake_up(&msblk->fragment_wait_queue); -+ up(&msblk->fragment_mutex); -+} -+ -+ -+SQSH_EXTERN struct squashfs_fragment_cache *get_cached_fragment(struct super_block -+ *s, long long start_block, -+ int length) -+{ -+ int i, n; -+ struct squashfs_sb_info *msblk = s->s_fs_info; -+ -+ while ( 1 ) { -+ down(&msblk->fragment_mutex); -+ -+ for (i = 0; i < SQUASHFS_CACHED_FRAGMENTS && -+ msblk->fragment[i].block != start_block; i++); -+ -+ if (i == SQUASHFS_CACHED_FRAGMENTS) { -+ for (i = msblk->next_fragment, n = -+ SQUASHFS_CACHED_FRAGMENTS; n && -+ msblk->fragment[i].locked; n--, i = (i + 1) % -+ SQUASHFS_CACHED_FRAGMENTS); -+ -+ if (n == 0) { -+ wait_queue_t wait; -+ -+ init_waitqueue_entry(&wait, current); -+ add_wait_queue(&msblk->fragment_wait_queue, -+ &wait); -+ set_current_state(TASK_UNINTERRUPTIBLE); -+ up(&msblk->fragment_mutex); -+ schedule(); -+ set_current_state(TASK_RUNNING); -+ remove_wait_queue(&msblk->fragment_wait_queue, -+ &wait); -+ continue; -+ } -+ msblk->next_fragment = (msblk->next_fragment + 1) % -+ SQUASHFS_CACHED_FRAGMENTS; -+ -+ if (msblk->fragment[i].data == NULL) -+ if (!(msblk->fragment[i].data = SQUASHFS_ALLOC -+ (SQUASHFS_FILE_MAX_SIZE))) { -+ ERROR("Failed to allocate fragment " -+ "cache block\n"); -+ up(&msblk->fragment_mutex); -+ goto out; -+ } -+ -+ msblk->fragment[i].block = SQUASHFS_INVALID_BLK; -+ msblk->fragment[i].locked = 1; -+ up(&msblk->fragment_mutex); -+ -+ if (!(msblk->fragment[i].length = squashfs_read_data(s, -+ msblk->fragment[i].data, -+ start_block, length, NULL))) { -+ ERROR("Unable to read fragment cache block " -+ "[%llx]\n", start_block); -+ msblk->fragment[i].locked = 0; -+ goto out; -+ } -+ -+ msblk->fragment[i].block = start_block; -+ TRACE("New fragment %d, start block %lld, locked %d\n", -+ i, msblk->fragment[i].block, -+ msblk->fragment[i].locked); -+ break; -+ } -+ -+ msblk->fragment[i].locked++; -+ up(&msblk->fragment_mutex); -+ TRACE("Got fragment %d, start block %lld, locked %d\n", i, -+ msblk->fragment[i].block, -+ msblk->fragment[i].locked); -+ break; -+ } -+ -+ return &msblk->fragment[i]; -+ -+out: -+ return NULL; -+} -+ -+ -+static struct inode *squashfs_new_inode(struct super_block *s, -+ struct squashfs_base_inode_header *inodeb) -+{ -+ struct squashfs_sb_info *msblk = s->s_fs_info; -+ struct inode *i = new_inode(s); -+ -+ if (i) { -+ i->i_ino = inodeb->inode_number; -+ i->i_mtime.tv_sec = inodeb->mtime; -+ i->i_atime.tv_sec = inodeb->mtime; -+ i->i_ctime.tv_sec = inodeb->mtime; -+ i->i_uid = msblk->uid[inodeb->uid]; -+ i->i_mode = inodeb->mode; -+ i->i_size = 0; -+ if (inodeb->guid == SQUASHFS_GUIDS) -+ i->i_gid = i->i_uid; -+ else -+ i->i_gid = msblk->guid[inodeb->guid]; -+ } -+ -+ return i; -+} -+ -+ -+static struct inode *squashfs_iget(struct super_block *s, squashfs_inode_t inode) -+{ -+ struct inode *i; -+ struct squashfs_sb_info *msblk = s->s_fs_info; -+ struct squashfs_super_block *sblk = &msblk->sblk; -+ long long block = SQUASHFS_INODE_BLK(inode) + -+ sblk->inode_table_start; -+ unsigned int offset = SQUASHFS_INODE_OFFSET(inode); -+ long long next_block; -+ unsigned int next_offset; -+ union squashfs_inode_header id, sid; -+ struct squashfs_base_inode_header *inodeb = &id.base, -+ *sinodeb = &sid.base; -+ -+ TRACE("Entered squashfs_iget\n"); -+ -+ if (msblk->swap) { -+ if (!squashfs_get_cached_block(s, (char *) sinodeb, block, -+ offset, sizeof(*sinodeb), &next_block, -+ &next_offset)) -+ goto failed_read; -+ SQUASHFS_SWAP_BASE_INODE_HEADER(inodeb, sinodeb, -+ sizeof(*sinodeb)); -+ } else -+ if (!squashfs_get_cached_block(s, (char *) inodeb, block, -+ offset, sizeof(*inodeb), &next_block, -+ &next_offset)) -+ goto failed_read; -+ -+ switch(inodeb->inode_type) { -+ case SQUASHFS_FILE_TYPE: { -+ unsigned int frag_size; -+ long long frag_blk; -+ struct squashfs_reg_inode_header *inodep = &id.reg; -+ struct squashfs_reg_inode_header *sinodep = &sid.reg; -+ -+ if (msblk->swap) { -+ if (!squashfs_get_cached_block(s, (char *) -+ sinodep, block, offset, -+ sizeof(*sinodep), &next_block, -+ &next_offset)) -+ goto failed_read; -+ SQUASHFS_SWAP_REG_INODE_HEADER(inodep, sinodep); -+ } else -+ if (!squashfs_get_cached_block(s, (char *) -+ inodep, block, offset, -+ sizeof(*inodep), &next_block, -+ &next_offset)) -+ goto failed_read; -+ -+ frag_blk = SQUASHFS_INVALID_BLK; -+ if (inodep->fragment != SQUASHFS_INVALID_FRAG && -+ !get_fragment_location(s, -+ inodep->fragment, &frag_blk, &frag_size)) -+ goto failed_read; -+ -+ if((i = squashfs_new_inode(s, inodeb)) == NULL) -+ goto failed_read1; -+ -+ i->i_nlink = 1; -+ i->i_size = inodep->file_size; -+ i->i_fop = &generic_ro_fops; -+ i->i_mode |= S_IFREG; -+ i->i_blocks = ((i->i_size - 1) >> 9) + 1; -+ SQUASHFS_I(i)->u.s1.fragment_start_block = frag_blk; -+ SQUASHFS_I(i)->u.s1.fragment_size = frag_size; -+ SQUASHFS_I(i)->u.s1.fragment_offset = inodep->offset; -+ SQUASHFS_I(i)->start_block = inodep->start_block; -+ SQUASHFS_I(i)->u.s1.block_list_start = next_block; -+ SQUASHFS_I(i)->offset = next_offset; -+ if (sblk->block_size > 4096) -+ i->i_data.a_ops = &squashfs_aops; -+ else -+ i->i_data.a_ops = &squashfs_aops_4K; -+ -+ TRACE("File inode %x:%x, start_block %llx, " -+ "block_list_start %llx, offset %x\n", -+ SQUASHFS_INODE_BLK(inode), offset, -+ inodep->start_block, next_block, -+ next_offset); -+ break; -+ } -+ case SQUASHFS_LREG_TYPE: { -+ unsigned int frag_size; -+ long long frag_blk; -+ struct squashfs_lreg_inode_header *inodep = &id.lreg; -+ struct squashfs_lreg_inode_header *sinodep = &sid.lreg; -+ -+ if (msblk->swap) { -+ if (!squashfs_get_cached_block(s, (char *) -+ sinodep, block, offset, -+ sizeof(*sinodep), &next_block, -+ &next_offset)) -+ goto failed_read; -+ SQUASHFS_SWAP_LREG_INODE_HEADER(inodep, sinodep); -+ } else -+ if (!squashfs_get_cached_block(s, (char *) -+ inodep, block, offset, -+ sizeof(*inodep), &next_block, -+ &next_offset)) -+ goto failed_read; -+ -+ frag_blk = SQUASHFS_INVALID_BLK; -+ if (inodep->fragment != SQUASHFS_INVALID_FRAG && -+ !get_fragment_location(s, -+ inodep->fragment, &frag_blk, &frag_size)) -+ goto failed_read; -+ -+ if((i = squashfs_new_inode(s, inodeb)) == NULL) -+ goto failed_read1; -+ -+ i->i_nlink = inodep->nlink; -+ i->i_size = inodep->file_size; -+ i->i_fop = &generic_ro_fops; -+ i->i_mode |= S_IFREG; -+ i->i_blocks = ((i->i_size - 1) >> 9) + 1; -+ SQUASHFS_I(i)->u.s1.fragment_start_block = frag_blk; -+ SQUASHFS_I(i)->u.s1.fragment_size = frag_size; -+ SQUASHFS_I(i)->u.s1.fragment_offset = inodep->offset; -+ SQUASHFS_I(i)->start_block = inodep->start_block; -+ SQUASHFS_I(i)->u.s1.block_list_start = next_block; -+ SQUASHFS_I(i)->offset = next_offset; -+ if (sblk->block_size > 4096) -+ i->i_data.a_ops = &squashfs_aops; -+ else -+ i->i_data.a_ops = &squashfs_aops_4K; -+ -+ TRACE("File inode %x:%x, start_block %llx, " -+ "block_list_start %llx, offset %x\n", -+ SQUASHFS_INODE_BLK(inode), offset, -+ inodep->start_block, next_block, -+ next_offset); -+ break; -+ } -+ case SQUASHFS_DIR_TYPE: { -+ struct squashfs_dir_inode_header *inodep = &id.dir; -+ struct squashfs_dir_inode_header *sinodep = &sid.dir; -+ -+ if (msblk->swap) { -+ if (!squashfs_get_cached_block(s, (char *) -+ sinodep, block, offset, -+ sizeof(*sinodep), &next_block, -+ &next_offset)) -+ goto failed_read; -+ SQUASHFS_SWAP_DIR_INODE_HEADER(inodep, sinodep); -+ } else -+ if (!squashfs_get_cached_block(s, (char *) -+ inodep, block, offset, -+ sizeof(*inodep), &next_block, -+ &next_offset)) -+ goto failed_read; -+ -+ if((i = squashfs_new_inode(s, inodeb)) == NULL) -+ goto failed_read1; -+ -+ i->i_nlink = inodep->nlink; -+ i->i_size = inodep->file_size; -+ i->i_op = &squashfs_dir_inode_ops; -+ i->i_fop = &squashfs_dir_ops; -+ i->i_mode |= S_IFDIR; -+ SQUASHFS_I(i)->start_block = inodep->start_block; -+ SQUASHFS_I(i)->offset = inodep->offset; -+ SQUASHFS_I(i)->u.s2.directory_index_count = 0; -+ SQUASHFS_I(i)->u.s2.parent_inode = inodep->parent_inode; -+ -+ TRACE("Directory inode %x:%x, start_block %x, offset " -+ "%x\n", SQUASHFS_INODE_BLK(inode), -+ offset, inodep->start_block, -+ inodep->offset); -+ break; -+ } -+ case SQUASHFS_LDIR_TYPE: { -+ struct squashfs_ldir_inode_header *inodep = &id.ldir; -+ struct squashfs_ldir_inode_header *sinodep = &sid.ldir; -+ -+ if (msblk->swap) { -+ if (!squashfs_get_cached_block(s, (char *) -+ sinodep, block, offset, -+ sizeof(*sinodep), &next_block, -+ &next_offset)) -+ goto failed_read; -+ SQUASHFS_SWAP_LDIR_INODE_HEADER(inodep, -+ sinodep); -+ } else -+ if (!squashfs_get_cached_block(s, (char *) -+ inodep, block, offset, -+ sizeof(*inodep), &next_block, -+ &next_offset)) -+ goto failed_read; -+ -+ if((i = squashfs_new_inode(s, inodeb)) == NULL) -+ goto failed_read1; -+ -+ i->i_nlink = inodep->nlink; -+ i->i_size = inodep->file_size; -+ i->i_op = &squashfs_dir_inode_ops; -+ i->i_fop = &squashfs_dir_ops; -+ i->i_mode |= S_IFDIR; -+ SQUASHFS_I(i)->start_block = inodep->start_block; -+ SQUASHFS_I(i)->offset = inodep->offset; -+ SQUASHFS_I(i)->u.s2.directory_index_start = next_block; -+ SQUASHFS_I(i)->u.s2.directory_index_offset = -+ next_offset; -+ SQUASHFS_I(i)->u.s2.directory_index_count = -+ inodep->i_count; -+ SQUASHFS_I(i)->u.s2.parent_inode = inodep->parent_inode; -+ -+ TRACE("Long directory inode %x:%x, start_block %x, " -+ "offset %x\n", -+ SQUASHFS_INODE_BLK(inode), offset, -+ inodep->start_block, inodep->offset); -+ break; -+ } -+ case SQUASHFS_SYMLINK_TYPE: { -+ struct squashfs_symlink_inode_header *inodep = -+ &id.symlink; -+ struct squashfs_symlink_inode_header *sinodep = -+ &sid.symlink; -+ -+ if (msblk->swap) { -+ if (!squashfs_get_cached_block(s, (char *) -+ sinodep, block, offset, -+ sizeof(*sinodep), &next_block, -+ &next_offset)) -+ goto failed_read; -+ SQUASHFS_SWAP_SYMLINK_INODE_HEADER(inodep, -+ sinodep); -+ } else -+ if (!squashfs_get_cached_block(s, (char *) -+ inodep, block, offset, -+ sizeof(*inodep), &next_block, -+ &next_offset)) -+ goto failed_read; -+ -+ if((i = squashfs_new_inode(s, inodeb)) == NULL) -+ goto failed_read1; -+ -+ i->i_nlink = inodep->nlink; -+ i->i_size = inodep->symlink_size; -+ i->i_op = &page_symlink_inode_operations; -+ i->i_data.a_ops = &squashfs_symlink_aops; -+ i->i_mode |= S_IFLNK; -+ SQUASHFS_I(i)->start_block = next_block; -+ SQUASHFS_I(i)->offset = next_offset; -+ -+ TRACE("Symbolic link inode %x:%x, start_block %llx, " -+ "offset %x\n", -+ SQUASHFS_INODE_BLK(inode), offset, -+ next_block, next_offset); -+ break; -+ } -+ case SQUASHFS_BLKDEV_TYPE: -+ case SQUASHFS_CHRDEV_TYPE: { -+ struct squashfs_dev_inode_header *inodep = &id.dev; -+ struct squashfs_dev_inode_header *sinodep = &sid.dev; -+ -+ if (msblk->swap) { -+ if (!squashfs_get_cached_block(s, (char *) -+ sinodep, block, offset, -+ sizeof(*sinodep), &next_block, -+ &next_offset)) -+ goto failed_read; -+ SQUASHFS_SWAP_DEV_INODE_HEADER(inodep, sinodep); -+ } else -+ if (!squashfs_get_cached_block(s, (char *) -+ inodep, block, offset, -+ sizeof(*inodep), &next_block, -+ &next_offset)) -+ goto failed_read; -+ -+ if ((i = squashfs_new_inode(s, inodeb)) == NULL) -+ goto failed_read1; -+ -+ i->i_nlink = inodep->nlink; -+ i->i_mode |= (inodeb->inode_type == -+ SQUASHFS_CHRDEV_TYPE) ? S_IFCHR : -+ S_IFBLK; -+ init_special_inode(i, i->i_mode, -+ old_decode_dev(inodep->rdev)); -+ -+ TRACE("Device inode %x:%x, rdev %x\n", -+ SQUASHFS_INODE_BLK(inode), offset, -+ inodep->rdev); -+ break; -+ } -+ case SQUASHFS_FIFO_TYPE: -+ case SQUASHFS_SOCKET_TYPE: { -+ struct squashfs_ipc_inode_header *inodep = &id.ipc; -+ struct squashfs_ipc_inode_header *sinodep = &sid.ipc; -+ -+ if (msblk->swap) { -+ if (!squashfs_get_cached_block(s, (char *) -+ sinodep, block, offset, -+ sizeof(*sinodep), &next_block, -+ &next_offset)) -+ goto failed_read; -+ SQUASHFS_SWAP_IPC_INODE_HEADER(inodep, sinodep); -+ } else -+ if (!squashfs_get_cached_block(s, (char *) -+ inodep, block, offset, -+ sizeof(*inodep), &next_block, -+ &next_offset)) -+ goto failed_read; -+ -+ if ((i = squashfs_new_inode(s, inodeb)) == NULL) -+ goto failed_read1; -+ -+ i->i_nlink = inodep->nlink; -+ i->i_mode |= (inodeb->inode_type == SQUASHFS_FIFO_TYPE) -+ ? S_IFIFO : S_IFSOCK; -+ init_special_inode(i, i->i_mode, 0); -+ break; -+ } -+ default: -+ ERROR("Unknown inode type %d in squashfs_iget!\n", -+ inodeb->inode_type); -+ goto failed_read1; -+ } -+ -+ insert_inode_hash(i); -+ return i; -+ -+failed_read: -+ ERROR("Unable to read inode [%llx:%x]\n", block, offset); -+ -+failed_read1: -+ return NULL; -+} -+ -+ -+static int read_fragment_index_table(struct super_block *s) -+{ -+ struct squashfs_sb_info *msblk = s->s_fs_info; -+ struct squashfs_super_block *sblk = &msblk->sblk; -+ -+ /* Allocate fragment index table */ -+ if (!(msblk->fragment_index = kmalloc(SQUASHFS_FRAGMENT_INDEX_BYTES -+ (sblk->fragments), GFP_KERNEL))) { -+ ERROR("Failed to allocate uid/gid table\n"); -+ return 0; -+ } -+ -+ if (SQUASHFS_FRAGMENT_INDEX_BYTES(sblk->fragments) && -+ !squashfs_read_data(s, (char *) -+ msblk->fragment_index, -+ sblk->fragment_table_start, -+ SQUASHFS_FRAGMENT_INDEX_BYTES -+ (sblk->fragments) | -+ SQUASHFS_COMPRESSED_BIT_BLOCK, NULL)) { -+ ERROR("unable to read fragment index table\n"); -+ return 0; -+ } -+ -+ if (msblk->swap) { -+ int i; -+ long long fragment; -+ -+ for (i = 0; i < SQUASHFS_FRAGMENT_INDEXES(sblk->fragments); -+ i++) { -+ SQUASHFS_SWAP_FRAGMENT_INDEXES((&fragment), -+ &msblk->fragment_index[i], 1); -+ msblk->fragment_index[i] = fragment; -+ } -+ } -+ -+ return 1; -+} -+ -+ -+static int supported_squashfs_filesystem(struct squashfs_sb_info *msblk, int silent) -+{ -+ struct squashfs_super_block *sblk = &msblk->sblk; -+ -+ msblk->iget = squashfs_iget; -+ msblk->read_blocklist = read_blocklist; -+ msblk->read_fragment_index_table = read_fragment_index_table; -+ -+ if (sblk->s_major == 1) { -+ if (!squashfs_1_0_supported(msblk)) { -+ SERROR("Major/Minor mismatch, Squashfs 1.0 filesystems " -+ "are unsupported\n"); -+ SERROR("Please recompile with " -+ "Squashfs 1.0 support enabled\n"); -+ return 0; -+ } -+ } else if (sblk->s_major == 2) { -+ if (!squashfs_2_0_supported(msblk)) { -+ SERROR("Major/Minor mismatch, Squashfs 2.0 filesystems " -+ "are unsupported\n"); -+ SERROR("Please recompile with " -+ "Squashfs 2.0 support enabled\n"); -+ return 0; -+ } -+ } else if(sblk->s_major != SQUASHFS_MAJOR || sblk->s_minor > -+ SQUASHFS_MINOR) { -+ SERROR("Major/Minor mismatch, trying to mount newer %d.%d " -+ "filesystem\n", sblk->s_major, sblk->s_minor); -+ SERROR("Please update your kernel\n"); -+ return 0; -+ } -+ -+ return 1; -+} -+ -+ -+static int squashfs_fill_super(struct super_block *s, void *data, int silent) -+{ -+ struct squashfs_sb_info *msblk; -+ struct squashfs_super_block *sblk; -+ int i; -+ char b[BDEVNAME_SIZE]; -+ struct inode *root; -+ -+ TRACE("Entered squashfs_read_superblock\n"); -+ -+ if (!(s->s_fs_info = kmalloc(sizeof(struct squashfs_sb_info), -+ GFP_KERNEL))) { -+ ERROR("Failed to allocate superblock\n"); -+ goto failure; -+ } -+ memset(s->s_fs_info, 0, sizeof(struct squashfs_sb_info)); -+ msblk = s->s_fs_info; -+ sblk = &msblk->sblk; -+ -+ msblk->devblksize = sb_min_blocksize(s, BLOCK_SIZE); -+ msblk->devblksize_log2 = ffz(~msblk->devblksize); -+ -+ init_MUTEX(&msblk->read_data_mutex); -+ init_MUTEX(&msblk->read_page_mutex); -+ init_MUTEX(&msblk->block_cache_mutex); -+ init_MUTEX(&msblk->fragment_mutex); -+ init_MUTEX(&msblk->meta_index_mutex); -+ -+ init_waitqueue_head(&msblk->waitq); -+ init_waitqueue_head(&msblk->fragment_wait_queue); -+ -+ if (!squashfs_read_data(s, (char *) sblk, SQUASHFS_START, -+ sizeof(struct squashfs_super_block) | -+ SQUASHFS_COMPRESSED_BIT_BLOCK, NULL)) { -+ SERROR("unable to read superblock\n"); -+ goto failed_mount; -+ } -+ -+ /* Check it is a SQUASHFS superblock */ -+ msblk->swap = 0; -+ if ((s->s_magic = sblk->s_magic) != SQUASHFS_MAGIC) { -+ if (sblk->s_magic == SQUASHFS_MAGIC_SWAP) { -+ struct squashfs_super_block ssblk; -+ -+ WARNING("Mounting a different endian SQUASHFS " -+ "filesystem on %s\n", bdevname(s->s_bdev, b)); -+ -+ SQUASHFS_SWAP_SUPER_BLOCK(&ssblk, sblk); -+ memcpy(sblk, &ssblk, sizeof(struct squashfs_super_block)); -+ msblk->swap = 1; -+ } else { -+ SERROR("Can't find a SQUASHFS superblock on %s\n", -+ bdevname(s->s_bdev, b)); -+ goto failed_mount; -+ } -+ } -+ -+ /* Check the MAJOR & MINOR versions */ -+ if(!supported_squashfs_filesystem(msblk, silent)) -+ goto failed_mount; -+ -+ TRACE("Found valid superblock on %s\n", bdevname(s->s_bdev, b)); -+ TRACE("Inodes are %scompressed\n", -+ SQUASHFS_UNCOMPRESSED_INODES -+ (sblk->flags) ? "un" : ""); -+ TRACE("Data is %scompressed\n", -+ SQUASHFS_UNCOMPRESSED_DATA(sblk->flags) -+ ? "un" : ""); -+ TRACE("Check data is %s present in the filesystem\n", -+ SQUASHFS_CHECK_DATA(sblk->flags) ? -+ "" : "not"); -+ TRACE("Filesystem size %lld bytes\n", sblk->bytes_used); -+ TRACE("Block size %d\n", sblk->block_size); -+ TRACE("Number of inodes %d\n", sblk->inodes); -+ if (sblk->s_major > 1) -+ TRACE("Number of fragments %d\n", sblk->fragments); -+ TRACE("Number of uids %d\n", sblk->no_uids); -+ TRACE("Number of gids %d\n", sblk->no_guids); -+ TRACE("sblk->inode_table_start %llx\n", sblk->inode_table_start); -+ TRACE("sblk->directory_table_start %llx\n", sblk->directory_table_start); -+ if (sblk->s_major > 1) -+ TRACE("sblk->fragment_table_start %llx\n", -+ sblk->fragment_table_start); -+ TRACE("sblk->uid_start %llx\n", sblk->uid_start); -+ -+ s->s_flags |= MS_RDONLY; -+ s->s_op = &squashfs_ops; -+ -+ /* Init inode_table block pointer array */ -+ if (!(msblk->block_cache = kmalloc(sizeof(struct squashfs_cache) * -+ SQUASHFS_CACHED_BLKS, GFP_KERNEL))) { -+ ERROR("Failed to allocate block cache\n"); -+ goto failed_mount; -+ } -+ -+ for (i = 0; i < SQUASHFS_CACHED_BLKS; i++) -+ msblk->block_cache[i].block = SQUASHFS_INVALID_BLK; -+ -+ msblk->next_cache = 0; -+ -+ /* Allocate read_data block */ -+ msblk->read_size = (sblk->block_size < SQUASHFS_METADATA_SIZE) ? -+ SQUASHFS_METADATA_SIZE : -+ sblk->block_size; -+ -+ if (!(msblk->read_data = kmalloc(msblk->read_size, GFP_KERNEL))) { -+ ERROR("Failed to allocate read_data block\n"); -+ goto failed_mount; -+ } -+ -+ /* Allocate read_page block */ -+ if (!(msblk->read_page = kmalloc(sblk->block_size, GFP_KERNEL))) { -+ ERROR("Failed to allocate read_page block\n"); -+ goto failed_mount; -+ } -+ -+ /* Allocate uid and gid tables */ -+ if (!(msblk->uid = kmalloc((sblk->no_uids + sblk->no_guids) * -+ sizeof(unsigned int), GFP_KERNEL))) { -+ ERROR("Failed to allocate uid/gid table\n"); -+ goto failed_mount; -+ } -+ msblk->guid = msblk->uid + sblk->no_uids; -+ -+ if (msblk->swap) { -+ unsigned int suid[sblk->no_uids + sblk->no_guids]; -+ -+ if (!squashfs_read_data(s, (char *) &suid, sblk->uid_start, -+ ((sblk->no_uids + sblk->no_guids) * -+ sizeof(unsigned int)) | -+ SQUASHFS_COMPRESSED_BIT_BLOCK, NULL)) { -+ ERROR("unable to read uid/gid table\n"); -+ goto failed_mount; -+ } -+ -+ SQUASHFS_SWAP_DATA(msblk->uid, suid, (sblk->no_uids + -+ sblk->no_guids), (sizeof(unsigned int) * 8)); -+ } else -+ if (!squashfs_read_data(s, (char *) msblk->uid, sblk->uid_start, -+ ((sblk->no_uids + sblk->no_guids) * -+ sizeof(unsigned int)) | -+ SQUASHFS_COMPRESSED_BIT_BLOCK, NULL)) { -+ ERROR("unable to read uid/gid table\n"); -+ goto failed_mount; -+ } -+ -+ -+ if (sblk->s_major == 1 && squashfs_1_0_supported(msblk)) -+ goto allocate_root; -+ -+ if (!(msblk->fragment = kmalloc(sizeof(struct squashfs_fragment_cache) * -+ SQUASHFS_CACHED_FRAGMENTS, GFP_KERNEL))) { -+ ERROR("Failed to allocate fragment block cache\n"); -+ goto failed_mount; -+ } -+ -+ for (i = 0; i < SQUASHFS_CACHED_FRAGMENTS; i++) { -+ msblk->fragment[i].locked = 0; -+ msblk->fragment[i].block = SQUASHFS_INVALID_BLK; -+ msblk->fragment[i].data = NULL; -+ } -+ -+ msblk->next_fragment = 0; -+ -+ /* Allocate fragment index table */ -+ if (msblk->read_fragment_index_table(s) == 0) -+ goto failed_mount; -+ -+allocate_root: -+ if ((root = (msblk->iget)(s, sblk->root_inode)) == NULL) -+ goto failed_mount; -+ -+ if ((s->s_root = d_alloc_root(root)) == NULL) { -+ ERROR("Root inode create failed\n"); -+ iput(root); -+ goto failed_mount; -+ } -+ -+ TRACE("Leaving squashfs_read_super\n"); -+ return 0; -+ -+failed_mount: -+ kfree(msblk->fragment_index); -+ kfree(msblk->fragment); -+ kfree(msblk->uid); -+ kfree(msblk->read_page); -+ kfree(msblk->read_data); -+ kfree(msblk->block_cache); -+ kfree(msblk->fragment_index_2); -+ kfree(s->s_fs_info); -+ s->s_fs_info = NULL; -+ return -EINVAL; -+ -+failure: -+ return -ENOMEM; -+} -+ -+ -+static int squashfs_statfs(struct dentry *dentry, struct kstatfs *buf) -+{ -+ struct squashfs_sb_info *msblk = dentry->d_inode->i_sb->s_fs_info; -+ struct squashfs_super_block *sblk = &msblk->sblk; -+ -+ TRACE("Entered squashfs_statfs\n"); -+ -+ buf->f_type = SQUASHFS_MAGIC; -+ buf->f_bsize = sblk->block_size; -+ buf->f_blocks = ((sblk->bytes_used - 1) >> sblk->block_log) + 1; -+ buf->f_bfree = buf->f_bavail = 0; -+ buf->f_files = sblk->inodes; -+ buf->f_ffree = 0; -+ buf->f_namelen = SQUASHFS_NAME_LEN; -+ -+ return 0; -+} -+ -+ -+static int squashfs_symlink_readpage(struct file *file, struct page *page) -+{ -+ struct inode *inode = page->mapping->host; -+ int index = page->index << PAGE_CACHE_SHIFT, length, bytes; -+ long long block = SQUASHFS_I(inode)->start_block; -+ int offset = SQUASHFS_I(inode)->offset; -+ void *pageaddr = kmap(page); -+ -+ TRACE("Entered squashfs_symlink_readpage, page index %ld, start block " -+ "%llx, offset %x\n", page->index, -+ SQUASHFS_I(inode)->start_block, -+ SQUASHFS_I(inode)->offset); -+ -+ for (length = 0; length < index; length += bytes) { -+ if (!(bytes = squashfs_get_cached_block(inode->i_sb, NULL, -+ block, offset, PAGE_CACHE_SIZE, &block, -+ &offset))) { -+ ERROR("Unable to read symbolic link [%llx:%x]\n", block, -+ offset); -+ goto skip_read; -+ } -+ } -+ -+ if (length != index) { -+ ERROR("(squashfs_symlink_readpage) length != index\n"); -+ bytes = 0; -+ goto skip_read; -+ } -+ -+ bytes = (i_size_read(inode) - length) > PAGE_CACHE_SIZE ? PAGE_CACHE_SIZE : -+ i_size_read(inode) - length; -+ -+ if (!(bytes = squashfs_get_cached_block(inode->i_sb, pageaddr, block, -+ offset, bytes, &block, &offset))) -+ ERROR("Unable to read symbolic link [%llx:%x]\n", block, offset); -+ -+skip_read: -+ memset(pageaddr + bytes, 0, PAGE_CACHE_SIZE - bytes); -+ kunmap(page); -+ SetPageUptodate(page); -+ unlock_page(page); -+ -+ return 0; -+} -+ -+ -+struct meta_index *locate_meta_index(struct inode *inode, int index, int offset) -+{ -+ struct meta_index *meta = NULL; -+ struct squashfs_sb_info *msblk = inode->i_sb->s_fs_info; -+ int i; -+ -+ down(&msblk->meta_index_mutex); -+ -+ TRACE("locate_meta_index: index %d, offset %d\n", index, offset); -+ -+ if(msblk->meta_index == NULL) -+ goto not_allocated; -+ -+ for (i = 0; i < SQUASHFS_META_NUMBER; i ++) -+ if (msblk->meta_index[i].inode_number == inode->i_ino && -+ msblk->meta_index[i].offset >= offset && -+ msblk->meta_index[i].offset <= index && -+ msblk->meta_index[i].locked == 0) { -+ TRACE("locate_meta_index: entry %d, offset %d\n", i, -+ msblk->meta_index[i].offset); -+ meta = &msblk->meta_index[i]; -+ offset = meta->offset; -+ } -+ -+ if (meta) -+ meta->locked = 1; -+ -+not_allocated: -+ up(&msblk->meta_index_mutex); -+ -+ return meta; -+} -+ -+ -+struct meta_index *empty_meta_index(struct inode *inode, int offset, int skip) -+{ -+ struct squashfs_sb_info *msblk = inode->i_sb->s_fs_info; -+ struct meta_index *meta = NULL; -+ int i; -+ -+ down(&msblk->meta_index_mutex); -+ -+ TRACE("empty_meta_index: offset %d, skip %d\n", offset, skip); -+ -+ if(msblk->meta_index == NULL) { -+ if (!(msblk->meta_index = kmalloc(sizeof(struct meta_index) * -+ SQUASHFS_META_NUMBER, GFP_KERNEL))) { -+ ERROR("Failed to allocate meta_index\n"); -+ goto failed; -+ } -+ for(i = 0; i < SQUASHFS_META_NUMBER; i++) { -+ msblk->meta_index[i].inode_number = 0; -+ msblk->meta_index[i].locked = 0; -+ } -+ msblk->next_meta_index = 0; -+ } -+ -+ for(i = SQUASHFS_META_NUMBER; i && -+ msblk->meta_index[msblk->next_meta_index].locked; i --) -+ msblk->next_meta_index = (msblk->next_meta_index + 1) % -+ SQUASHFS_META_NUMBER; -+ -+ if(i == 0) { -+ TRACE("empty_meta_index: failed!\n"); -+ goto failed; -+ } -+ -+ TRACE("empty_meta_index: returned meta entry %d, %p\n", -+ msblk->next_meta_index, -+ &msblk->meta_index[msblk->next_meta_index]); -+ -+ meta = &msblk->meta_index[msblk->next_meta_index]; -+ msblk->next_meta_index = (msblk->next_meta_index + 1) % -+ SQUASHFS_META_NUMBER; -+ -+ meta->inode_number = inode->i_ino; -+ meta->offset = offset; -+ meta->skip = skip; -+ meta->entries = 0; -+ meta->locked = 1; -+ -+failed: -+ up(&msblk->meta_index_mutex); -+ return meta; -+} -+ -+ -+void release_meta_index(struct inode *inode, struct meta_index *meta) -+{ -+ meta->locked = 0; -+} -+ -+ -+static int read_block_index(struct super_block *s, int blocks, char *block_list, -+ long long *start_block, int *offset) -+{ -+ struct squashfs_sb_info *msblk = s->s_fs_info; -+ unsigned int *block_listp; -+ int block = 0; -+ -+ if (msblk->swap) { -+ char sblock_list[blocks << 2]; -+ -+ if (!squashfs_get_cached_block(s, sblock_list, *start_block, -+ *offset, blocks << 2, start_block, offset)) { -+ ERROR("Unable to read block list [%llx:%x]\n", -+ *start_block, *offset); -+ goto failure; -+ } -+ SQUASHFS_SWAP_INTS(((unsigned int *)block_list), -+ ((unsigned int *)sblock_list), blocks); -+ } else -+ if (!squashfs_get_cached_block(s, block_list, *start_block, -+ *offset, blocks << 2, start_block, offset)) { -+ ERROR("Unable to read block list [%llx:%x]\n", -+ *start_block, *offset); -+ goto failure; -+ } -+ -+ for (block_listp = (unsigned int *) block_list; blocks; -+ block_listp++, blocks --) -+ block += SQUASHFS_COMPRESSED_SIZE_BLOCK(*block_listp); -+ -+ return block; -+ -+failure: -+ return -1; -+} -+ -+ -+#define SIZE 256 -+ -+static inline int calculate_skip(int blocks) { -+ int skip = (blocks - 1) / ((SQUASHFS_SLOTS * SQUASHFS_META_ENTRIES + 1) * SQUASHFS_META_INDEXES); -+ return skip >= 7 ? 7 : skip + 1; -+} -+ -+ -+static int get_meta_index(struct inode *inode, int index, -+ long long *index_block, int *index_offset, -+ long long *data_block, char *block_list) -+{ -+ struct squashfs_sb_info *msblk = inode->i_sb->s_fs_info; -+ struct squashfs_super_block *sblk = &msblk->sblk; -+ int skip = calculate_skip(i_size_read(inode) >> sblk->block_log); -+ int offset = 0; -+ struct meta_index *meta; -+ struct meta_entry *meta_entry; -+ long long cur_index_block = SQUASHFS_I(inode)->u.s1.block_list_start; -+ int cur_offset = SQUASHFS_I(inode)->offset; -+ long long cur_data_block = SQUASHFS_I(inode)->start_block; -+ int i; -+ -+ index /= SQUASHFS_META_INDEXES * skip; -+ -+ while ( offset < index ) { -+ meta = locate_meta_index(inode, index, offset + 1); -+ -+ if (meta == NULL) { -+ if ((meta = empty_meta_index(inode, offset + 1, -+ skip)) == NULL) -+ goto all_done; -+ } else { -+ offset = index < meta->offset + meta->entries ? index : -+ meta->offset + meta->entries - 1; -+ meta_entry = &meta->meta_entry[offset - meta->offset]; -+ cur_index_block = meta_entry->index_block + sblk->inode_table_start; -+ cur_offset = meta_entry->offset; -+ cur_data_block = meta_entry->data_block; -+ TRACE("get_meta_index: offset %d, meta->offset %d, " -+ "meta->entries %d\n", offset, meta->offset, -+ meta->entries); -+ TRACE("get_meta_index: index_block 0x%llx, offset 0x%x" -+ " data_block 0x%llx\n", cur_index_block, -+ cur_offset, cur_data_block); -+ } -+ -+ for (i = meta->offset + meta->entries; i <= index && -+ i < meta->offset + SQUASHFS_META_ENTRIES; i++) { -+ int blocks = skip * SQUASHFS_META_INDEXES; -+ -+ while (blocks) { -+ int block = blocks > (SIZE >> 2) ? (SIZE >> 2) : -+ blocks; -+ int res = read_block_index(inode->i_sb, block, -+ block_list, &cur_index_block, -+ &cur_offset); -+ -+ if (res == -1) -+ goto failed; -+ -+ cur_data_block += res; -+ blocks -= block; -+ } -+ -+ meta_entry = &meta->meta_entry[i - meta->offset]; -+ meta_entry->index_block = cur_index_block - sblk->inode_table_start; -+ meta_entry->offset = cur_offset; -+ meta_entry->data_block = cur_data_block; -+ meta->entries ++; -+ offset ++; -+ } -+ -+ TRACE("get_meta_index: meta->offset %d, meta->entries %d\n", -+ meta->offset, meta->entries); -+ -+ release_meta_index(inode, meta); -+ } -+ -+all_done: -+ *index_block = cur_index_block; -+ *index_offset = cur_offset; -+ *data_block = cur_data_block; -+ -+ return offset * SQUASHFS_META_INDEXES * skip; -+ -+failed: -+ release_meta_index(inode, meta); -+ return -1; -+} -+ -+ -+static long long read_blocklist(struct inode *inode, int index, -+ int readahead_blks, char *block_list, -+ unsigned short **block_p, unsigned int *bsize) -+{ -+ long long block_ptr; -+ int offset; -+ long long block; -+ int res = get_meta_index(inode, index, &block_ptr, &offset, &block, -+ block_list); -+ -+ TRACE("read_blocklist: res %d, index %d, block_ptr 0x%llx, offset" -+ " 0x%x, block 0x%llx\n", res, index, block_ptr, offset, -+ block); -+ -+ if(res == -1) -+ goto failure; -+ -+ index -= res; -+ -+ while ( index ) { -+ int blocks = index > (SIZE >> 2) ? (SIZE >> 2) : index; -+ int res = read_block_index(inode->i_sb, blocks, block_list, -+ &block_ptr, &offset); -+ if (res == -1) -+ goto failure; -+ block += res; -+ index -= blocks; -+ } -+ -+ if (read_block_index(inode->i_sb, 1, block_list, -+ &block_ptr, &offset) == -1) -+ goto failure; -+ *bsize = *((unsigned int *) block_list); -+ -+ return block; -+ -+failure: -+ return 0; -+} -+ -+ -+static int squashfs_readpage(struct file *file, struct page *page) -+{ -+ struct inode *inode = page->mapping->host; -+ struct squashfs_sb_info *msblk = inode->i_sb->s_fs_info; -+ struct squashfs_super_block *sblk = &msblk->sblk; -+ unsigned char block_list[SIZE]; -+ long long block; -+ unsigned int bsize, i = 0, bytes = 0, byte_offset = 0; -+ int index = page->index >> (sblk->block_log - PAGE_CACHE_SHIFT); -+ void *pageaddr; -+ struct squashfs_fragment_cache *fragment = NULL; -+ char *data_ptr = msblk->read_page; -+ -+ int mask = (1 << (sblk->block_log - PAGE_CACHE_SHIFT)) - 1; -+ int start_index = page->index & ~mask; -+ int end_index = start_index | mask; -+ -+ TRACE("Entered squashfs_readpage, page index %lx, start block %llx\n", -+ page->index, -+ SQUASHFS_I(inode)->start_block); -+ -+ if (page->index >= ((i_size_read(inode) + PAGE_CACHE_SIZE - 1) >> -+ PAGE_CACHE_SHIFT)) -+ goto skip_read; -+ -+ if (SQUASHFS_I(inode)->u.s1.fragment_start_block == SQUASHFS_INVALID_BLK -+ || index < (i_size_read(inode) >> -+ sblk->block_log)) { -+ if ((block = (msblk->read_blocklist)(inode, index, 1, -+ block_list, NULL, &bsize)) == 0) -+ goto skip_read; -+ -+ down(&msblk->read_page_mutex); -+ -+ if (!(bytes = squashfs_read_data(inode->i_sb, msblk->read_page, -+ block, bsize, NULL))) { -+ ERROR("Unable to read page, block %llx, size %x\n", block, -+ bsize); -+ up(&msblk->read_page_mutex); -+ goto skip_read; -+ } -+ } else { -+ if ((fragment = get_cached_fragment(inode->i_sb, -+ SQUASHFS_I(inode)-> -+ u.s1.fragment_start_block, -+ SQUASHFS_I(inode)->u.s1.fragment_size)) -+ == NULL) { -+ ERROR("Unable to read page, block %llx, size %x\n", -+ SQUASHFS_I(inode)-> -+ u.s1.fragment_start_block, -+ (int) SQUASHFS_I(inode)-> -+ u.s1.fragment_size); -+ goto skip_read; -+ } -+ bytes = SQUASHFS_I(inode)->u.s1.fragment_offset + -+ (i_size_read(inode) & (sblk->block_size -+ - 1)); -+ byte_offset = SQUASHFS_I(inode)->u.s1.fragment_offset; -+ data_ptr = fragment->data; -+ } -+ -+ for (i = start_index; i <= end_index && byte_offset < bytes; -+ i++, byte_offset += PAGE_CACHE_SIZE) { -+ struct page *push_page; -+ int available_bytes = (bytes - byte_offset) > PAGE_CACHE_SIZE ? -+ PAGE_CACHE_SIZE : bytes - byte_offset; -+ -+ TRACE("bytes %d, i %d, byte_offset %d, available_bytes %d\n", -+ bytes, i, byte_offset, available_bytes); -+ -+ if (i == page->index) { -+ pageaddr = kmap_atomic(page, KM_USER0); -+ memcpy(pageaddr, data_ptr + byte_offset, -+ available_bytes); -+ memset(pageaddr + available_bytes, 0, -+ PAGE_CACHE_SIZE - available_bytes); -+ kunmap_atomic(pageaddr, KM_USER0); -+ flush_dcache_page(page); -+ SetPageUptodate(page); -+ unlock_page(page); -+ } else if ((push_page = -+ grab_cache_page_nowait(page->mapping, i))) { -+ pageaddr = kmap_atomic(push_page, KM_USER0); -+ -+ memcpy(pageaddr, data_ptr + byte_offset, -+ available_bytes); -+ memset(pageaddr + available_bytes, 0, -+ PAGE_CACHE_SIZE - available_bytes); -+ kunmap_atomic(pageaddr, KM_USER0); -+ flush_dcache_page(push_page); -+ SetPageUptodate(push_page); -+ unlock_page(push_page); -+ page_cache_release(push_page); -+ } -+ } -+ -+ if (SQUASHFS_I(inode)->u.s1.fragment_start_block == SQUASHFS_INVALID_BLK -+ || index < (i_size_read(inode) >> -+ sblk->block_log)) -+ up(&msblk->read_page_mutex); -+ else -+ release_cached_fragment(msblk, fragment); -+ -+ return 0; -+ -+skip_read: -+ pageaddr = kmap_atomic(page, KM_USER0); -+ memset(pageaddr + bytes, 0, PAGE_CACHE_SIZE - bytes); -+ kunmap_atomic(pageaddr, KM_USER0); -+ flush_dcache_page(page); -+ SetPageUptodate(page); -+ unlock_page(page); -+ -+ return 0; -+} -+ -+ -+static int squashfs_readpage4K(struct file *file, struct page *page) -+{ -+ struct inode *inode = page->mapping->host; -+ struct squashfs_sb_info *msblk = inode->i_sb->s_fs_info; -+ struct squashfs_super_block *sblk = &msblk->sblk; -+ unsigned char block_list[SIZE]; -+ long long block; -+ unsigned int bsize, bytes = 0; -+ void *pageaddr; -+ -+ TRACE("Entered squashfs_readpage4K, page index %lx, start block %llx\n", -+ page->index, -+ SQUASHFS_I(inode)->start_block); -+ -+ if (page->index >= ((i_size_read(inode) + PAGE_CACHE_SIZE - 1) >> -+ PAGE_CACHE_SHIFT)) { -+ pageaddr = kmap_atomic(page, KM_USER0); -+ goto skip_read; -+ } -+ -+ if (SQUASHFS_I(inode)->u.s1.fragment_start_block == SQUASHFS_INVALID_BLK -+ || page->index < (i_size_read(inode) >> -+ sblk->block_log)) { -+ block = (msblk->read_blocklist)(inode, page->index, 1, -+ block_list, NULL, &bsize); -+ -+ down(&msblk->read_page_mutex); -+ bytes = squashfs_read_data(inode->i_sb, msblk->read_page, block, -+ bsize, NULL); -+ pageaddr = kmap_atomic(page, KM_USER0); -+ if (bytes) -+ memcpy(pageaddr, msblk->read_page, bytes); -+ else -+ ERROR("Unable to read page, block %llx, size %x\n", -+ block, bsize); -+ up(&msblk->read_page_mutex); -+ } else { -+ struct squashfs_fragment_cache *fragment = -+ get_cached_fragment(inode->i_sb, -+ SQUASHFS_I(inode)-> -+ u.s1.fragment_start_block, -+ SQUASHFS_I(inode)-> u.s1.fragment_size); -+ pageaddr = kmap_atomic(page, KM_USER0); -+ if (fragment) { -+ bytes = i_size_read(inode) & (sblk->block_size - 1); -+ memcpy(pageaddr, fragment->data + SQUASHFS_I(inode)-> -+ u.s1.fragment_offset, bytes); -+ release_cached_fragment(msblk, fragment); -+ } else -+ ERROR("Unable to read page, block %llx, size %x\n", -+ SQUASHFS_I(inode)-> -+ u.s1.fragment_start_block, (int) -+ SQUASHFS_I(inode)-> u.s1.fragment_size); -+ } -+ -+skip_read: -+ memset(pageaddr + bytes, 0, PAGE_CACHE_SIZE - bytes); -+ kunmap_atomic(pageaddr, KM_USER0); -+ flush_dcache_page(page); -+ SetPageUptodate(page); -+ unlock_page(page); -+ -+ return 0; -+} -+ -+ -+static int get_dir_index_using_offset(struct super_block *s, long long -+ *next_block, unsigned int *next_offset, -+ long long index_start, -+ unsigned int index_offset, int i_count, -+ long long f_pos) -+{ -+ struct squashfs_sb_info *msblk = s->s_fs_info; -+ struct squashfs_super_block *sblk = &msblk->sblk; -+ int i, length = 0; -+ struct squashfs_dir_index index; -+ -+ TRACE("Entered get_dir_index_using_offset, i_count %d, f_pos %d\n", -+ i_count, (unsigned int) f_pos); -+ -+ f_pos =- 3; -+ if (f_pos == 0) -+ goto finish; -+ -+ for (i = 0; i < i_count; i++) { -+ if (msblk->swap) { -+ struct squashfs_dir_index sindex; -+ squashfs_get_cached_block(s, (char *) &sindex, -+ index_start, index_offset, -+ sizeof(sindex), &index_start, -+ &index_offset); -+ SQUASHFS_SWAP_DIR_INDEX(&index, &sindex); -+ } else -+ squashfs_get_cached_block(s, (char *) &index, -+ index_start, index_offset, -+ sizeof(index), &index_start, -+ &index_offset); -+ -+ if (index.index > f_pos) -+ break; -+ -+ squashfs_get_cached_block(s, NULL, index_start, index_offset, -+ index.size + 1, &index_start, -+ &index_offset); -+ -+ length = index.index; -+ *next_block = index.start_block + sblk->directory_table_start; -+ } -+ -+ *next_offset = (length + *next_offset) % SQUASHFS_METADATA_SIZE; -+ -+finish: -+ return length + 3; -+} -+ -+ -+static int get_dir_index_using_name(struct super_block *s, long long -+ *next_block, unsigned int *next_offset, -+ long long index_start, -+ unsigned int index_offset, int i_count, -+ const char *name, int size) -+{ -+ struct squashfs_sb_info *msblk = s->s_fs_info; -+ struct squashfs_super_block *sblk = &msblk->sblk; -+ int i, length = 0; -+ char buffer[sizeof(struct squashfs_dir_index) + SQUASHFS_NAME_LEN + 1]; -+ struct squashfs_dir_index *index = (struct squashfs_dir_index *) buffer; -+ char str[SQUASHFS_NAME_LEN + 1]; -+ -+ TRACE("Entered get_dir_index_using_name, i_count %d\n", i_count); -+ -+ strncpy(str, name, size); -+ str[size] = '\0'; -+ -+ for (i = 0; i < i_count; i++) { -+ if (msblk->swap) { -+ struct squashfs_dir_index sindex; -+ squashfs_get_cached_block(s, (char *) &sindex, -+ index_start, index_offset, -+ sizeof(sindex), &index_start, -+ &index_offset); -+ SQUASHFS_SWAP_DIR_INDEX(index, &sindex); -+ } else -+ squashfs_get_cached_block(s, (char *) index, -+ index_start, index_offset, -+ sizeof(struct squashfs_dir_index), -+ &index_start, &index_offset); -+ -+ squashfs_get_cached_block(s, index->name, index_start, -+ index_offset, index->size + 1, -+ &index_start, &index_offset); -+ -+ index->name[index->size + 1] = '\0'; -+ -+ if (strcmp(index->name, str) > 0) -+ break; -+ -+ length = index->index; -+ *next_block = index->start_block + sblk->directory_table_start; -+ } -+ -+ *next_offset = (length + *next_offset) % SQUASHFS_METADATA_SIZE; -+ return length + 3; -+} -+ -+ -+static int squashfs_readdir(struct file *file, void *dirent, filldir_t filldir) -+{ -+ struct inode *i = file->f_dentry->d_inode; -+ struct squashfs_sb_info *msblk = i->i_sb->s_fs_info; -+ struct squashfs_super_block *sblk = &msblk->sblk; -+ long long next_block = SQUASHFS_I(i)->start_block + -+ sblk->directory_table_start; -+ int next_offset = SQUASHFS_I(i)->offset, length = 0, dirs_read = 0, -+ dir_count; -+ struct squashfs_dir_header dirh; -+ char buffer[sizeof(struct squashfs_dir_entry) + SQUASHFS_NAME_LEN + 1]; -+ struct squashfs_dir_entry *dire = (struct squashfs_dir_entry *) buffer; -+ -+ TRACE("Entered squashfs_readdir [%llx:%x]\n", next_block, next_offset); -+ -+ while(file->f_pos < 3) { -+ char *name; -+ int size, i_ino; -+ -+ if(file->f_pos == 0) { -+ name = "."; -+ size = 1; -+ i_ino = i->i_ino; -+ } else { -+ name = ".."; -+ size = 2; -+ i_ino = SQUASHFS_I(i)->u.s2.parent_inode; -+ } -+ TRACE("Calling filldir(%x, %s, %d, %d, %d, %d)\n", -+ (unsigned int) dirent, name, size, (int) -+ file->f_pos, i_ino, -+ squashfs_filetype_table[1]); -+ -+ if (filldir(dirent, name, size, -+ file->f_pos, i_ino, -+ squashfs_filetype_table[1]) < 0) { -+ TRACE("Filldir returned less than 0\n"); -+ goto finish; -+ } -+ file->f_pos += size; -+ dirs_read++; -+ } -+ -+ length = get_dir_index_using_offset(i->i_sb, &next_block, &next_offset, -+ SQUASHFS_I(i)->u.s2.directory_index_start, -+ SQUASHFS_I(i)->u.s2.directory_index_offset, -+ SQUASHFS_I(i)->u.s2.directory_index_count, -+ file->f_pos); -+ -+ while (length < i_size_read(i)) { -+ /* read directory header */ -+ if (msblk->swap) { -+ struct squashfs_dir_header sdirh; -+ -+ if (!squashfs_get_cached_block(i->i_sb, (char *) &sdirh, -+ next_block, next_offset, sizeof(sdirh), -+ &next_block, &next_offset)) -+ goto failed_read; -+ -+ length += sizeof(sdirh); -+ SQUASHFS_SWAP_DIR_HEADER(&dirh, &sdirh); -+ } else { -+ if (!squashfs_get_cached_block(i->i_sb, (char *) &dirh, -+ next_block, next_offset, sizeof(dirh), -+ &next_block, &next_offset)) -+ goto failed_read; -+ -+ length += sizeof(dirh); -+ } -+ -+ dir_count = dirh.count + 1; -+ while (dir_count--) { -+ if (msblk->swap) { -+ struct squashfs_dir_entry sdire; -+ if (!squashfs_get_cached_block(i->i_sb, (char *) -+ &sdire, next_block, next_offset, -+ sizeof(sdire), &next_block, -+ &next_offset)) -+ goto failed_read; -+ -+ length += sizeof(sdire); -+ SQUASHFS_SWAP_DIR_ENTRY(dire, &sdire); -+ } else { -+ if (!squashfs_get_cached_block(i->i_sb, (char *) -+ dire, next_block, next_offset, -+ sizeof(*dire), &next_block, -+ &next_offset)) -+ goto failed_read; -+ -+ length += sizeof(*dire); -+ } -+ -+ if (!squashfs_get_cached_block(i->i_sb, dire->name, -+ next_block, next_offset, -+ dire->size + 1, &next_block, -+ &next_offset)) -+ goto failed_read; -+ -+ length += dire->size + 1; -+ -+ if (file->f_pos >= length) -+ continue; -+ -+ dire->name[dire->size + 1] = '\0'; -+ -+ TRACE("Calling filldir(%x, %s, %d, %d, %x:%x, %d, %d)\n", -+ (unsigned int) dirent, dire->name, -+ dire->size + 1, (int) file->f_pos, -+ dirh.start_block, dire->offset, -+ dirh.inode_number + dire->inode_number, -+ squashfs_filetype_table[dire->type]); -+ -+ if (filldir(dirent, dire->name, dire->size + 1, -+ file->f_pos, -+ dirh.inode_number + dire->inode_number, -+ squashfs_filetype_table[dire->type]) -+ < 0) { -+ TRACE("Filldir returned less than 0\n"); -+ goto finish; -+ } -+ file->f_pos = length; -+ dirs_read++; -+ } -+ } -+ -+finish: -+ return dirs_read; -+ -+failed_read: -+ ERROR("Unable to read directory block [%llx:%x]\n", next_block, -+ next_offset); -+ return 0; -+} -+ -+ -+static struct dentry *squashfs_lookup(struct inode *i, struct dentry *dentry, -+ struct nameidata *nd) -+{ -+ const unsigned char *name = dentry->d_name.name; -+ int len = dentry->d_name.len; -+ struct inode *inode = NULL; -+ struct squashfs_sb_info *msblk = i->i_sb->s_fs_info; -+ struct squashfs_super_block *sblk = &msblk->sblk; -+ long long next_block = SQUASHFS_I(i)->start_block + -+ sblk->directory_table_start; -+ int next_offset = SQUASHFS_I(i)->offset, length = 0, -+ dir_count; -+ struct squashfs_dir_header dirh; -+ char buffer[sizeof(struct squashfs_dir_entry) + SQUASHFS_NAME_LEN]; -+ struct squashfs_dir_entry *dire = (struct squashfs_dir_entry *) buffer; -+ -+ TRACE("Entered squashfs_lookup [%llx:%x]\n", next_block, next_offset); -+ -+ if (len > SQUASHFS_NAME_LEN) -+ goto exit_loop; -+ -+ length = get_dir_index_using_name(i->i_sb, &next_block, &next_offset, -+ SQUASHFS_I(i)->u.s2.directory_index_start, -+ SQUASHFS_I(i)->u.s2.directory_index_offset, -+ SQUASHFS_I(i)->u.s2.directory_index_count, name, -+ len); -+ -+ while (length < i_size_read(i)) { -+ /* read directory header */ -+ if (msblk->swap) { -+ struct squashfs_dir_header sdirh; -+ if (!squashfs_get_cached_block(i->i_sb, (char *) &sdirh, -+ next_block, next_offset, sizeof(sdirh), -+ &next_block, &next_offset)) -+ goto failed_read; -+ -+ length += sizeof(sdirh); -+ SQUASHFS_SWAP_DIR_HEADER(&dirh, &sdirh); -+ } else { -+ if (!squashfs_get_cached_block(i->i_sb, (char *) &dirh, -+ next_block, next_offset, sizeof(dirh), -+ &next_block, &next_offset)) -+ goto failed_read; -+ -+ length += sizeof(dirh); -+ } -+ -+ dir_count = dirh.count + 1; -+ while (dir_count--) { -+ if (msblk->swap) { -+ struct squashfs_dir_entry sdire; -+ if (!squashfs_get_cached_block(i->i_sb, (char *) -+ &sdire, next_block,next_offset, -+ sizeof(sdire), &next_block, -+ &next_offset)) -+ goto failed_read; -+ -+ length += sizeof(sdire); -+ SQUASHFS_SWAP_DIR_ENTRY(dire, &sdire); -+ } else { -+ if (!squashfs_get_cached_block(i->i_sb, (char *) -+ dire, next_block,next_offset, -+ sizeof(*dire), &next_block, -+ &next_offset)) -+ goto failed_read; -+ -+ length += sizeof(*dire); -+ } -+ -+ if (!squashfs_get_cached_block(i->i_sb, dire->name, -+ next_block, next_offset, dire->size + 1, -+ &next_block, &next_offset)) -+ goto failed_read; -+ -+ length += dire->size + 1; -+ -+ if (name[0] < dire->name[0]) -+ goto exit_loop; -+ -+ if ((len == dire->size + 1) && !strncmp(name, -+ dire->name, len)) { -+ squashfs_inode_t ino = -+ SQUASHFS_MKINODE(dirh.start_block, -+ dire->offset); -+ -+ TRACE("calling squashfs_iget for directory " -+ "entry %s, inode %x:%x, %d\n", name, -+ dirh.start_block, dire->offset, -+ dirh.inode_number + dire->inode_number); -+ -+ inode = (msblk->iget)(i->i_sb, ino); -+ -+ goto exit_loop; -+ } -+ } -+ } -+ -+exit_loop: -+ d_add(dentry, inode); -+ return ERR_PTR(0); -+ -+failed_read: -+ ERROR("Unable to read directory block [%llx:%x]\n", next_block, -+ next_offset); -+ goto exit_loop; -+} -+ -+ -+static void squashfs_put_super(struct super_block *s) -+{ -+ int i; -+ -+ if (s->s_fs_info) { -+ struct squashfs_sb_info *sbi = s->s_fs_info; -+ if (sbi->block_cache) -+ for (i = 0; i < SQUASHFS_CACHED_BLKS; i++) -+ if (sbi->block_cache[i].block != -+ SQUASHFS_INVALID_BLK) -+ kfree(sbi->block_cache[i].data); -+ if (sbi->fragment) -+ for (i = 0; i < SQUASHFS_CACHED_FRAGMENTS; i++) -+ SQUASHFS_FREE(sbi->fragment[i].data); -+ kfree(sbi->fragment); -+ kfree(sbi->block_cache); -+ kfree(sbi->read_data); -+ kfree(sbi->read_page); -+ kfree(sbi->uid); -+ kfree(sbi->fragment_index); -+ kfree(sbi->fragment_index_2); -+ kfree(sbi->meta_index); -+ kfree(s->s_fs_info); -+ s->s_fs_info = NULL; -+ } -+} -+ -+ -+static int squashfs_get_sb(struct file_system_type *fs_type, -+ int flags, const char *dev_name, void *data, -+ struct vfsmount *mnt) -+{ -+ return get_sb_bdev(fs_type, flags, dev_name, data, squashfs_fill_super, mnt); -+} -+ -+ -+static int __init init_squashfs_fs(void) -+{ -+ int err = init_inodecache(); -+ if (err) -+ goto out; -+ -+ printk(KERN_INFO "squashfs: version 3.0 (2006/03/15) " -+ "Phillip Lougher\n"); -+ -+ if (!(stream.workspace = vmalloc(zlib_inflate_workspacesize()))) { -+ ERROR("Failed to allocate zlib workspace\n"); -+ destroy_inodecache(); -+ err = -ENOMEM; -+ goto out; -+ } -+ -+ if ((err = register_filesystem(&squashfs_fs_type))) { -+ vfree(stream.workspace); -+ destroy_inodecache(); -+ } -+ -+out: -+ return err; -+} -+ -+ -+static void __exit exit_squashfs_fs(void) -+{ -+ vfree(stream.workspace); -+ unregister_filesystem(&squashfs_fs_type); -+ destroy_inodecache(); -+} -+ -+ -+static struct kmem_cache * squashfs_inode_cachep; -+ -+ -+static struct inode *squashfs_alloc_inode(struct super_block *sb) -+{ -+ struct squashfs_inode_info *ei; -+ ei = kmem_cache_alloc(squashfs_inode_cachep, GFP_KERNEL); -+ if (!ei) -+ return NULL; -+ return &ei->vfs_inode; -+} -+ -+ -+static void squashfs_destroy_inode(struct inode *inode) -+{ -+ kmem_cache_free(squashfs_inode_cachep, SQUASHFS_I(inode)); -+} -+ -+ -+static void init_once(void * foo, struct kmem_cache * cachep, unsigned long flags) -+{ -+ struct squashfs_inode_info *ei = foo; -+ -+ inode_init_once(&ei->vfs_inode); -+} -+ -+ -+static int __init init_inodecache(void) -+{ -+ squashfs_inode_cachep = kmem_cache_create("squashfs_inode_cache", -+ sizeof(struct squashfs_inode_info), -+ 0, SLAB_HWCACHE_ALIGN|SLAB_RECLAIM_ACCOUNT, -+ init_once, NULL); -+ if (squashfs_inode_cachep == NULL) -+ return -ENOMEM; -+ return 0; -+} -+ -+ -+static void destroy_inodecache(void) -+{ -+ kmem_cache_destroy(squashfs_inode_cachep); -+} -+ -+ -+module_init(init_squashfs_fs); -+module_exit(exit_squashfs_fs); -+MODULE_DESCRIPTION("squashfs, a compressed read-only filesystem"); -+MODULE_AUTHOR("Phillip Lougher <phillip@lougher.org.uk>"); -+MODULE_LICENSE("GPL"); ---- /dev/null -+++ b/fs/squashfs/Makefile -@@ -0,0 +1,7 @@ -+# -+# Makefile for the linux squashfs routines. -+# -+ -+obj-$(CONFIG_SQUASHFS) += squashfs.o -+squashfs-y += inode.o -+squashfs-y += squashfs2_0.o ---- /dev/null -+++ b/fs/squashfs/squashfs2_0.c -@@ -0,0 +1,758 @@ -+/* -+ * Squashfs - a compressed read only filesystem for Linux -+ * -+ * Copyright (c) 2002, 2003, 2004, 2005, 2006 -+ * Phillip Lougher <phillip@lougher.org.uk> -+ * -+ * This program is free software; you can redistribute it and/or -+ * modify it under the terms of the GNU General Public License -+ * as published by the Free Software Foundation; either version 2, -+ * or (at your option) any later version. -+ * -+ * This program is distributed in the hope that it will be useful, -+ * but WITHOUT ANY WARRANTY; without even the implied warranty of -+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the -+ * GNU General Public License for more details. -+ * -+ * You should have received a copy of the GNU General Public License -+ * along with this program; if not, write to the Free Software -+ * Foundation, 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. -+ * -+ * squashfs2_0.c -+ */ -+ -+#include <linux/types.h> -+#include <linux/squashfs_fs.h> -+#include <linux/module.h> -+#include <linux/errno.h> -+#include <linux/slab.h> -+#include <linux/fs.h> -+#include <linux/smp_lock.h> -+#include <linux/slab.h> -+#include <linux/squashfs_fs_sb.h> -+#include <linux/squashfs_fs_i.h> -+#include <linux/buffer_head.h> -+#include <linux/vfs.h> -+#include <linux/init.h> -+#include <linux/dcache.h> -+#include <linux/wait.h> -+#include <linux/zlib.h> -+#include <linux/blkdev.h> -+#include <linux/vmalloc.h> -+#include <asm/uaccess.h> -+#include <asm/semaphore.h> -+ -+#include "squashfs.h" -+static int squashfs_readdir_2(struct file *file, void *dirent, filldir_t filldir); -+static struct dentry *squashfs_lookup_2(struct inode *, struct dentry *, -+ struct nameidata *); -+ -+static struct file_operations squashfs_dir_ops_2 = { -+ .read = generic_read_dir, -+ .readdir = squashfs_readdir_2 -+}; -+ -+static struct inode_operations squashfs_dir_inode_ops_2 = { -+ .lookup = squashfs_lookup_2 -+}; -+ -+static unsigned char squashfs_filetype_table[] = { -+ DT_UNKNOWN, DT_DIR, DT_REG, DT_LNK, DT_BLK, DT_CHR, DT_FIFO, DT_SOCK -+}; -+ -+static int read_fragment_index_table_2(struct super_block *s) -+{ -+ struct squashfs_sb_info *msblk = s->s_fs_info; -+ struct squashfs_super_block *sblk = &msblk->sblk; -+ -+ if (!(msblk->fragment_index_2 = kmalloc(SQUASHFS_FRAGMENT_INDEX_BYTES_2 -+ (sblk->fragments), GFP_KERNEL))) { -+ ERROR("Failed to allocate uid/gid table\n"); -+ return 0; -+ } -+ -+ if (SQUASHFS_FRAGMENT_INDEX_BYTES_2(sblk->fragments) && -+ !squashfs_read_data(s, (char *) -+ msblk->fragment_index_2, -+ sblk->fragment_table_start, -+ SQUASHFS_FRAGMENT_INDEX_BYTES_2 -+ (sblk->fragments) | -+ SQUASHFS_COMPRESSED_BIT_BLOCK, NULL)) { -+ ERROR("unable to read fragment index table\n"); -+ return 0; -+ } -+ -+ if (msblk->swap) { -+ int i; -+ unsigned int fragment; -+ -+ for (i = 0; i < SQUASHFS_FRAGMENT_INDEXES_2(sblk->fragments); -+ i++) { -+ SQUASHFS_SWAP_FRAGMENT_INDEXES_2((&fragment), -+ &msblk->fragment_index_2[i], 1); -+ msblk->fragment_index_2[i] = fragment; -+ } -+ } -+ -+ return 1; -+} -+ -+ -+static int get_fragment_location_2(struct super_block *s, unsigned int fragment, -+ long long *fragment_start_block, -+ unsigned int *fragment_size) -+{ -+ struct squashfs_sb_info *msblk = s->s_fs_info; -+ long long start_block = -+ msblk->fragment_index_2[SQUASHFS_FRAGMENT_INDEX_2(fragment)]; -+ int offset = SQUASHFS_FRAGMENT_INDEX_OFFSET_2(fragment); -+ struct squashfs_fragment_entry_2 fragment_entry; -+ -+ if (msblk->swap) { -+ struct squashfs_fragment_entry_2 sfragment_entry; -+ -+ if (!squashfs_get_cached_block(s, (char *) &sfragment_entry, -+ start_block, offset, -+ sizeof(sfragment_entry), &start_block, -+ &offset)) -+ goto out; -+ SQUASHFS_SWAP_FRAGMENT_ENTRY_2(&fragment_entry, &sfragment_entry); -+ } else -+ if (!squashfs_get_cached_block(s, (char *) &fragment_entry, -+ start_block, offset, -+ sizeof(fragment_entry), &start_block, -+ &offset)) -+ goto out; -+ -+ *fragment_start_block = fragment_entry.start_block; -+ *fragment_size = fragment_entry.size; -+ -+ return 1; -+ -+out: -+ return 0; -+} -+ -+ -+static struct inode *squashfs_new_inode(struct super_block *s, -+ struct squashfs_base_inode_header_2 *inodeb, unsigned int ino) -+{ -+ struct squashfs_sb_info *msblk = s->s_fs_info; -+ struct squashfs_super_block *sblk = &msblk->sblk; -+ struct inode *i = new_inode(s); -+ -+ if (i) { -+ i->i_ino = ino; -+ i->i_mtime.tv_sec = sblk->mkfs_time; -+ i->i_atime.tv_sec = sblk->mkfs_time; -+ i->i_ctime.tv_sec = sblk->mkfs_time; -+ i->i_uid = msblk->uid[inodeb->uid]; -+ i->i_mode = inodeb->mode; -+ i->i_nlink = 1; -+ i->i_size = 0; -+ if (inodeb->guid == SQUASHFS_GUIDS) -+ i->i_gid = i->i_uid; -+ else -+ i->i_gid = msblk->guid[inodeb->guid]; -+ } -+ -+ return i; -+} -+ -+ -+static struct inode *squashfs_iget_2(struct super_block *s, squashfs_inode_t inode) -+{ -+ struct inode *i; -+ struct squashfs_sb_info *msblk = s->s_fs_info; -+ struct squashfs_super_block *sblk = &msblk->sblk; -+ unsigned int block = SQUASHFS_INODE_BLK(inode) + -+ sblk->inode_table_start; -+ unsigned int offset = SQUASHFS_INODE_OFFSET(inode); -+ unsigned int ino = SQUASHFS_MK_VFS_INODE(block -+ - sblk->inode_table_start, offset); -+ long long next_block; -+ unsigned int next_offset; -+ union squashfs_inode_header_2 id, sid; -+ struct squashfs_base_inode_header_2 *inodeb = &id.base, -+ *sinodeb = &sid.base; -+ -+ TRACE("Entered squashfs_iget\n"); -+ -+ if (msblk->swap) { -+ if (!squashfs_get_cached_block(s, (char *) sinodeb, block, -+ offset, sizeof(*sinodeb), &next_block, -+ &next_offset)) -+ goto failed_read; -+ SQUASHFS_SWAP_BASE_INODE_HEADER_2(inodeb, sinodeb, -+ sizeof(*sinodeb)); -+ } else -+ if (!squashfs_get_cached_block(s, (char *) inodeb, block, -+ offset, sizeof(*inodeb), &next_block, -+ &next_offset)) -+ goto failed_read; -+ -+ switch(inodeb->inode_type) { -+ case SQUASHFS_FILE_TYPE: { -+ struct squashfs_reg_inode_header_2 *inodep = &id.reg; -+ struct squashfs_reg_inode_header_2 *sinodep = &sid.reg; -+ long long frag_blk; -+ unsigned int frag_size; -+ -+ if (msblk->swap) { -+ if (!squashfs_get_cached_block(s, (char *) -+ sinodep, block, offset, -+ sizeof(*sinodep), &next_block, -+ &next_offset)) -+ goto failed_read; -+ SQUASHFS_SWAP_REG_INODE_HEADER_2(inodep, sinodep); -+ } else -+ if (!squashfs_get_cached_block(s, (char *) -+ inodep, block, offset, -+ sizeof(*inodep), &next_block, -+ &next_offset)) -+ goto failed_read; -+ -+ frag_blk = SQUASHFS_INVALID_BLK; -+ if (inodep->fragment != SQUASHFS_INVALID_FRAG && -+ !get_fragment_location_2(s, -+ inodep->fragment, &frag_blk, &frag_size)) -+ goto failed_read; -+ -+ if((i = squashfs_new_inode(s, inodeb, ino)) == NULL) -+ goto failed_read1; -+ -+ i->i_size = inodep->file_size; -+ i->i_fop = &generic_ro_fops; -+ i->i_mode |= S_IFREG; -+ i->i_mtime.tv_sec = inodep->mtime; -+ i->i_atime.tv_sec = inodep->mtime; -+ i->i_ctime.tv_sec = inodep->mtime; -+ i->i_blocks = ((i->i_size - 1) >> 9) + 1; -+ i->i_blksize = PAGE_CACHE_SIZE; -+ SQUASHFS_I(i)->u.s1.fragment_start_block = frag_blk; -+ SQUASHFS_I(i)->u.s1.fragment_size = frag_size; -+ SQUASHFS_I(i)->u.s1.fragment_offset = inodep->offset; -+ SQUASHFS_I(i)->start_block = inodep->start_block; -+ SQUASHFS_I(i)->u.s1.block_list_start = next_block; -+ SQUASHFS_I(i)->offset = next_offset; -+ if (sblk->block_size > 4096) -+ i->i_data.a_ops = &squashfs_aops; -+ else -+ i->i_data.a_ops = &squashfs_aops_4K; -+ -+ TRACE("File inode %x:%x, start_block %x, " -+ "block_list_start %llx, offset %x\n", -+ SQUASHFS_INODE_BLK(inode), offset, -+ inodep->start_block, next_block, -+ next_offset); -+ break; -+ } -+ case SQUASHFS_DIR_TYPE: { -+ struct squashfs_dir_inode_header_2 *inodep = &id.dir; -+ struct squashfs_dir_inode_header_2 *sinodep = &sid.dir; -+ -+ if (msblk->swap) { -+ if (!squashfs_get_cached_block(s, (char *) -+ sinodep, block, offset, -+ sizeof(*sinodep), &next_block, -+ &next_offset)) -+ goto failed_read; -+ SQUASHFS_SWAP_DIR_INODE_HEADER_2(inodep, sinodep); -+ } else -+ if (!squashfs_get_cached_block(s, (char *) -+ inodep, block, offset, -+ sizeof(*inodep), &next_block, -+ &next_offset)) -+ goto failed_read; -+ -+ if((i = squashfs_new_inode(s, inodeb, ino)) == NULL) -+ goto failed_read1; -+ -+ i->i_size = inodep->file_size; -+ i->i_op = &squashfs_dir_inode_ops_2; -+ i->i_fop = &squashfs_dir_ops_2; -+ i->i_mode |= S_IFDIR; -+ i->i_mtime.tv_sec = inodep->mtime; -+ i->i_atime.tv_sec = inodep->mtime; -+ i->i_ctime.tv_sec = inodep->mtime; -+ SQUASHFS_I(i)->start_block = inodep->start_block; -+ SQUASHFS_I(i)->offset = inodep->offset; -+ SQUASHFS_I(i)->u.s2.directory_index_count = 0; -+ SQUASHFS_I(i)->u.s2.parent_inode = 0; -+ -+ TRACE("Directory inode %x:%x, start_block %x, offset " -+ "%x\n", SQUASHFS_INODE_BLK(inode), -+ offset, inodep->start_block, -+ inodep->offset); -+ break; -+ } -+ case SQUASHFS_LDIR_TYPE: { -+ struct squashfs_ldir_inode_header_2 *inodep = &id.ldir; -+ struct squashfs_ldir_inode_header_2 *sinodep = &sid.ldir; -+ -+ if (msblk->swap) { -+ if (!squashfs_get_cached_block(s, (char *) -+ sinodep, block, offset, -+ sizeof(*sinodep), &next_block, -+ &next_offset)) -+ goto failed_read; -+ SQUASHFS_SWAP_LDIR_INODE_HEADER_2(inodep, -+ sinodep); -+ } else -+ if (!squashfs_get_cached_block(s, (char *) -+ inodep, block, offset, -+ sizeof(*inodep), &next_block, -+ &next_offset)) -+ goto failed_read; -+ -+ if((i = squashfs_new_inode(s, inodeb, ino)) == NULL) -+ goto failed_read1; -+ -+ i->i_size = inodep->file_size; -+ i->i_op = &squashfs_dir_inode_ops_2; -+ i->i_fop = &squashfs_dir_ops_2; -+ i->i_mode |= S_IFDIR; -+ i->i_mtime.tv_sec = inodep->mtime; -+ i->i_atime.tv_sec = inodep->mtime; -+ i->i_ctime.tv_sec = inodep->mtime; -+ SQUASHFS_I(i)->start_block = inodep->start_block; -+ SQUASHFS_I(i)->offset = inodep->offset; -+ SQUASHFS_I(i)->u.s2.directory_index_start = next_block; -+ SQUASHFS_I(i)->u.s2.directory_index_offset = -+ next_offset; -+ SQUASHFS_I(i)->u.s2.directory_index_count = -+ inodep->i_count; -+ SQUASHFS_I(i)->u.s2.parent_inode = 0; -+ -+ TRACE("Long directory inode %x:%x, start_block %x, " -+ "offset %x\n", -+ SQUASHFS_INODE_BLK(inode), offset, -+ inodep->start_block, inodep->offset); -+ break; -+ } -+ case SQUASHFS_SYMLINK_TYPE: { -+ struct squashfs_symlink_inode_header_2 *inodep = -+ &id.symlink; -+ struct squashfs_symlink_inode_header_2 *sinodep = -+ &sid.symlink; -+ -+ if (msblk->swap) { -+ if (!squashfs_get_cached_block(s, (char *) -+ sinodep, block, offset, -+ sizeof(*sinodep), &next_block, -+ &next_offset)) -+ goto failed_read; -+ SQUASHFS_SWAP_SYMLINK_INODE_HEADER_2(inodep, -+ sinodep); -+ } else -+ if (!squashfs_get_cached_block(s, (char *) -+ inodep, block, offset, -+ sizeof(*inodep), &next_block, -+ &next_offset)) -+ goto failed_read; -+ -+ if((i = squashfs_new_inode(s, inodeb, ino)) == NULL) -+ goto failed_read1; -+ -+ i->i_size = inodep->symlink_size; -+ i->i_op = &page_symlink_inode_operations; -+ i->i_data.a_ops = &squashfs_symlink_aops; -+ i->i_mode |= S_IFLNK; -+ SQUASHFS_I(i)->start_block = next_block; -+ SQUASHFS_I(i)->offset = next_offset; -+ -+ TRACE("Symbolic link inode %x:%x, start_block %llx, " -+ "offset %x\n", -+ SQUASHFS_INODE_BLK(inode), offset, -+ next_block, next_offset); -+ break; -+ } -+ case SQUASHFS_BLKDEV_TYPE: -+ case SQUASHFS_CHRDEV_TYPE: { -+ struct squashfs_dev_inode_header_2 *inodep = &id.dev; -+ struct squashfs_dev_inode_header_2 *sinodep = &sid.dev; -+ -+ if (msblk->swap) { -+ if (!squashfs_get_cached_block(s, (char *) -+ sinodep, block, offset, -+ sizeof(*sinodep), &next_block, -+ &next_offset)) -+ goto failed_read; -+ SQUASHFS_SWAP_DEV_INODE_HEADER_2(inodep, sinodep); -+ } else -+ if (!squashfs_get_cached_block(s, (char *) -+ inodep, block, offset, -+ sizeof(*inodep), &next_block, -+ &next_offset)) -+ goto failed_read; -+ -+ if ((i = squashfs_new_inode(s, inodeb, ino)) == NULL) -+ goto failed_read1; -+ -+ i->i_mode |= (inodeb->inode_type == -+ SQUASHFS_CHRDEV_TYPE) ? S_IFCHR : -+ S_IFBLK; -+ init_special_inode(i, i->i_mode, -+ old_decode_dev(inodep->rdev)); -+ -+ TRACE("Device inode %x:%x, rdev %x\n", -+ SQUASHFS_INODE_BLK(inode), offset, -+ inodep->rdev); -+ break; -+ } -+ case SQUASHFS_FIFO_TYPE: -+ case SQUASHFS_SOCKET_TYPE: { -+ if ((i = squashfs_new_inode(s, inodeb, ino)) == NULL) -+ goto failed_read1; -+ -+ i->i_mode |= (inodeb->inode_type == SQUASHFS_FIFO_TYPE) -+ ? S_IFIFO : S_IFSOCK; -+ init_special_inode(i, i->i_mode, 0); -+ break; -+ } -+ default: -+ ERROR("Unknown inode type %d in squashfs_iget!\n", -+ inodeb->inode_type); -+ goto failed_read1; -+ } -+ -+ insert_inode_hash(i); -+ return i; -+ -+failed_read: -+ ERROR("Unable to read inode [%x:%x]\n", block, offset); -+ -+failed_read1: -+ return NULL; -+} -+ -+ -+static int get_dir_index_using_offset(struct super_block *s, long long -+ *next_block, unsigned int *next_offset, -+ long long index_start, -+ unsigned int index_offset, int i_count, -+ long long f_pos) -+{ -+ struct squashfs_sb_info *msblk = s->s_fs_info; -+ struct squashfs_super_block *sblk = &msblk->sblk; -+ int i, length = 0; -+ struct squashfs_dir_index_2 index; -+ -+ TRACE("Entered get_dir_index_using_offset, i_count %d, f_pos %d\n", -+ i_count, (unsigned int) f_pos); -+ -+ if (f_pos == 0) -+ goto finish; -+ -+ for (i = 0; i < i_count; i++) { -+ if (msblk->swap) { -+ struct squashfs_dir_index_2 sindex; -+ squashfs_get_cached_block(s, (char *) &sindex, -+ index_start, index_offset, -+ sizeof(sindex), &index_start, -+ &index_offset); -+ SQUASHFS_SWAP_DIR_INDEX_2(&index, &sindex); -+ } else -+ squashfs_get_cached_block(s, (char *) &index, -+ index_start, index_offset, -+ sizeof(index), &index_start, -+ &index_offset); -+ -+ if (index.index > f_pos) -+ break; -+ -+ squashfs_get_cached_block(s, NULL, index_start, index_offset, -+ index.size + 1, &index_start, -+ &index_offset); -+ -+ length = index.index; -+ *next_block = index.start_block + sblk->directory_table_start; -+ } -+ -+ *next_offset = (length + *next_offset) % SQUASHFS_METADATA_SIZE; -+ -+finish: -+ return length; -+} -+ -+ -+static int get_dir_index_using_name(struct super_block *s, long long -+ *next_block, unsigned int *next_offset, -+ long long index_start, -+ unsigned int index_offset, int i_count, -+ const char *name, int size) -+{ -+ struct squashfs_sb_info *msblk = s->s_fs_info; -+ struct squashfs_super_block *sblk = &msblk->sblk; -+ int i, length = 0; -+ char buffer[sizeof(struct squashfs_dir_index_2) + SQUASHFS_NAME_LEN + 1]; -+ struct squashfs_dir_index_2 *index = (struct squashfs_dir_index_2 *) buffer; -+ char str[SQUASHFS_NAME_LEN + 1]; -+ -+ TRACE("Entered get_dir_index_using_name, i_count %d\n", i_count); -+ -+ strncpy(str, name, size); -+ str[size] = '\0'; -+ -+ for (i = 0; i < i_count; i++) { -+ if (msblk->swap) { -+ struct squashfs_dir_index_2 sindex; -+ squashfs_get_cached_block(s, (char *) &sindex, -+ index_start, index_offset, -+ sizeof(sindex), &index_start, -+ &index_offset); -+ SQUASHFS_SWAP_DIR_INDEX_2(index, &sindex); -+ } else -+ squashfs_get_cached_block(s, (char *) index, -+ index_start, index_offset, -+ sizeof(struct squashfs_dir_index_2), -+ &index_start, &index_offset); -+ -+ squashfs_get_cached_block(s, index->name, index_start, -+ index_offset, index->size + 1, -+ &index_start, &index_offset); -+ -+ index->name[index->size + 1] = '\0'; -+ -+ if (strcmp(index->name, str) > 0) -+ break; -+ -+ length = index->index; -+ *next_block = index->start_block + sblk->directory_table_start; -+ } -+ -+ *next_offset = (length + *next_offset) % SQUASHFS_METADATA_SIZE; -+ return length; -+} -+ -+ -+static int squashfs_readdir_2(struct file *file, void *dirent, filldir_t filldir) -+{ -+ struct inode *i = file->f_dentry->d_inode; -+ struct squashfs_sb_info *msblk = i->i_sb->s_fs_info; -+ struct squashfs_super_block *sblk = &msblk->sblk; -+ long long next_block = SQUASHFS_I(i)->start_block + -+ sblk->directory_table_start; -+ int next_offset = SQUASHFS_I(i)->offset, length = 0, dirs_read = 0, -+ dir_count; -+ struct squashfs_dir_header_2 dirh; -+ char buffer[sizeof(struct squashfs_dir_entry_2) + SQUASHFS_NAME_LEN + 1]; -+ struct squashfs_dir_entry_2 *dire = (struct squashfs_dir_entry_2 *) buffer; -+ -+ TRACE("Entered squashfs_readdir_2 [%llx:%x]\n", next_block, next_offset); -+ -+ length = get_dir_index_using_offset(i->i_sb, &next_block, &next_offset, -+ SQUASHFS_I(i)->u.s2.directory_index_start, -+ SQUASHFS_I(i)->u.s2.directory_index_offset, -+ SQUASHFS_I(i)->u.s2.directory_index_count, -+ file->f_pos); -+ -+ while (length < i_size_read(i)) { -+ /* read directory header */ -+ if (msblk->swap) { -+ struct squashfs_dir_header_2 sdirh; -+ -+ if (!squashfs_get_cached_block(i->i_sb, (char *) &sdirh, -+ next_block, next_offset, sizeof(sdirh), -+ &next_block, &next_offset)) -+ goto failed_read; -+ -+ length += sizeof(sdirh); -+ SQUASHFS_SWAP_DIR_HEADER_2(&dirh, &sdirh); -+ } else { -+ if (!squashfs_get_cached_block(i->i_sb, (char *) &dirh, -+ next_block, next_offset, sizeof(dirh), -+ &next_block, &next_offset)) -+ goto failed_read; -+ -+ length += sizeof(dirh); -+ } -+ -+ dir_count = dirh.count + 1; -+ while (dir_count--) { -+ if (msblk->swap) { -+ struct squashfs_dir_entry_2 sdire; -+ if (!squashfs_get_cached_block(i->i_sb, (char *) -+ &sdire, next_block, next_offset, -+ sizeof(sdire), &next_block, -+ &next_offset)) -+ goto failed_read; -+ -+ length += sizeof(sdire); -+ SQUASHFS_SWAP_DIR_ENTRY_2(dire, &sdire); -+ } else { -+ if (!squashfs_get_cached_block(i->i_sb, (char *) -+ dire, next_block, next_offset, -+ sizeof(*dire), &next_block, -+ &next_offset)) -+ goto failed_read; -+ -+ length += sizeof(*dire); -+ } -+ -+ if (!squashfs_get_cached_block(i->i_sb, dire->name, -+ next_block, next_offset, -+ dire->size + 1, &next_block, -+ &next_offset)) -+ goto failed_read; -+ -+ length += dire->size + 1; -+ -+ if (file->f_pos >= length) -+ continue; -+ -+ dire->name[dire->size + 1] = '\0'; -+ -+ TRACE("Calling filldir(%x, %s, %d, %d, %x:%x, %d)\n", -+ (unsigned int) dirent, dire->name, -+ dire->size + 1, (int) file->f_pos, -+ dirh.start_block, dire->offset, -+ squashfs_filetype_table[dire->type]); -+ -+ if (filldir(dirent, dire->name, dire->size + 1, -+ file->f_pos, SQUASHFS_MK_VFS_INODE( -+ dirh.start_block, dire->offset), -+ squashfs_filetype_table[dire->type]) -+ < 0) { -+ TRACE("Filldir returned less than 0\n"); -+ goto finish; -+ } -+ file->f_pos = length; -+ dirs_read++; -+ } -+ } -+ -+finish: -+ return dirs_read; -+ -+failed_read: -+ ERROR("Unable to read directory block [%llx:%x]\n", next_block, -+ next_offset); -+ return 0; -+} -+ -+ -+static struct dentry *squashfs_lookup_2(struct inode *i, struct dentry *dentry, -+ struct nameidata *nd) -+{ -+ const unsigned char *name = dentry->d_name.name; -+ int len = dentry->d_name.len; -+ struct inode *inode = NULL; -+ struct squashfs_sb_info *msblk = i->i_sb->s_fs_info; -+ struct squashfs_super_block *sblk = &msblk->sblk; -+ long long next_block = SQUASHFS_I(i)->start_block + -+ sblk->directory_table_start; -+ int next_offset = SQUASHFS_I(i)->offset, length = 0, -+ dir_count; -+ struct squashfs_dir_header_2 dirh; -+ char buffer[sizeof(struct squashfs_dir_entry_2) + SQUASHFS_NAME_LEN]; -+ struct squashfs_dir_entry_2 *dire = (struct squashfs_dir_entry_2 *) buffer; -+ int sorted = sblk->s_major == 2 && sblk->s_minor >= 1; -+ -+ TRACE("Entered squashfs_lookup [%llx:%x]\n", next_block, next_offset); -+ -+ if (len > SQUASHFS_NAME_LEN) -+ goto exit_loop; -+ -+ length = get_dir_index_using_name(i->i_sb, &next_block, &next_offset, -+ SQUASHFS_I(i)->u.s2.directory_index_start, -+ SQUASHFS_I(i)->u.s2.directory_index_offset, -+ SQUASHFS_I(i)->u.s2.directory_index_count, name, -+ len); -+ -+ while (length < i_size_read(i)) { -+ /* read directory header */ -+ if (msblk->swap) { -+ struct squashfs_dir_header_2 sdirh; -+ if (!squashfs_get_cached_block(i->i_sb, (char *) &sdirh, -+ next_block, next_offset, sizeof(sdirh), -+ &next_block, &next_offset)) -+ goto failed_read; -+ -+ length += sizeof(sdirh); -+ SQUASHFS_SWAP_DIR_HEADER_2(&dirh, &sdirh); -+ } else { -+ if (!squashfs_get_cached_block(i->i_sb, (char *) &dirh, -+ next_block, next_offset, sizeof(dirh), -+ &next_block, &next_offset)) -+ goto failed_read; -+ -+ length += sizeof(dirh); -+ } -+ -+ dir_count = dirh.count + 1; -+ while (dir_count--) { -+ if (msblk->swap) { -+ struct squashfs_dir_entry_2 sdire; -+ if (!squashfs_get_cached_block(i->i_sb, (char *) -+ &sdire, next_block,next_offset, -+ sizeof(sdire), &next_block, -+ &next_offset)) -+ goto failed_read; -+ -+ length += sizeof(sdire); -+ SQUASHFS_SWAP_DIR_ENTRY_2(dire, &sdire); -+ } else { -+ if (!squashfs_get_cached_block(i->i_sb, (char *) -+ dire, next_block,next_offset, -+ sizeof(*dire), &next_block, -+ &next_offset)) -+ goto failed_read; -+ -+ length += sizeof(*dire); -+ } -+ -+ if (!squashfs_get_cached_block(i->i_sb, dire->name, -+ next_block, next_offset, dire->size + 1, -+ &next_block, &next_offset)) -+ goto failed_read; -+ -+ length += dire->size + 1; -+ -+ if (sorted && name[0] < dire->name[0]) -+ goto exit_loop; -+ -+ if ((len == dire->size + 1) && !strncmp(name, -+ dire->name, len)) { -+ squashfs_inode_t ino = -+ SQUASHFS_MKINODE(dirh.start_block, -+ dire->offset); -+ -+ TRACE("calling squashfs_iget for directory " -+ "entry %s, inode %x:%x, %lld\n", name, -+ dirh.start_block, dire->offset, ino); -+ -+ inode = (msblk->iget)(i->i_sb, ino); -+ -+ goto exit_loop; -+ } -+ } -+ } -+ -+exit_loop: -+ d_add(dentry, inode); -+ return ERR_PTR(0); -+ -+failed_read: -+ ERROR("Unable to read directory block [%llx:%x]\n", next_block, -+ next_offset); -+ goto exit_loop; -+} -+ -+ -+int squashfs_2_0_supported(struct squashfs_sb_info *msblk) -+{ -+ struct squashfs_super_block *sblk = &msblk->sblk; -+ -+ msblk->iget = squashfs_iget_2; -+ msblk->read_fragment_index_table = read_fragment_index_table_2; -+ -+ sblk->bytes_used = sblk->bytes_used_2; -+ sblk->uid_start = sblk->uid_start_2; -+ sblk->guid_start = sblk->guid_start_2; -+ sblk->inode_table_start = sblk->inode_table_start_2; -+ sblk->directory_table_start = sblk->directory_table_start_2; -+ sblk->fragment_table_start = sblk->fragment_table_start_2; -+ -+ return 1; -+} ---- /dev/null -+++ b/fs/squashfs/squashfs.h -@@ -0,0 +1,86 @@ -+/* -+ * Squashfs - a compressed read only filesystem for Linux -+ * -+ * Copyright (c) 2002, 2003, 2004, 2005, 2006 -+ * Phillip Lougher <phillip@lougher.org.uk> -+ * -+ * This program is free software; you can redistribute it and/or -+ * modify it under the terms of the GNU General Public License -+ * as published by the Free Software Foundation; either version 2, -+ * or (at your option) any later version. -+ * -+ * This program is distributed in the hope that it will be useful, -+ * but WITHOUT ANY WARRANTY; without even the implied warranty of -+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the -+ * GNU General Public License for more details. -+ * -+ * You should have received a copy of the GNU General Public License -+ * along with this program; if not, write to the Free Software -+ * Foundation, 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. -+ * -+ * squashfs.h -+ */ -+ -+#ifdef CONFIG_SQUASHFS_1_0_COMPATIBILITY -+#undef CONFIG_SQUASHFS_1_0_COMPATIBILITY -+#endif -+ -+#ifdef SQUASHFS_TRACE -+#define TRACE(s, args...) printk(KERN_NOTICE "SQUASHFS: "s, ## args) -+#else -+#define TRACE(s, args...) {} -+#endif -+ -+#define ERROR(s, args...) printk(KERN_ERR "SQUASHFS error: "s, ## args) -+ -+#define SERROR(s, args...) do { \ -+ if (!silent) \ -+ printk(KERN_ERR "SQUASHFS error: "s, ## args);\ -+ } while(0) -+ -+#define WARNING(s, args...) printk(KERN_WARNING "SQUASHFS: "s, ## args) -+ -+static inline struct squashfs_inode_info *SQUASHFS_I(struct inode *inode) -+{ -+ return list_entry(inode, struct squashfs_inode_info, vfs_inode); -+} -+ -+#if defined(CONFIG_SQUASHFS_1_0_COMPATIBILITY ) || defined(CONFIG_SQUASHFS_2_0_COMPATIBILITY) -+#define SQSH_EXTERN -+extern unsigned int squashfs_read_data(struct super_block *s, char *buffer, -+ long long index, unsigned int length, -+ long long *next_index); -+extern int squashfs_get_cached_block(struct super_block *s, char *buffer, -+ long long block, unsigned int offset, -+ int length, long long *next_block, -+ unsigned int *next_offset); -+extern void release_cached_fragment(struct squashfs_sb_info *msblk, struct -+ squashfs_fragment_cache *fragment); -+extern struct squashfs_fragment_cache *get_cached_fragment(struct super_block -+ *s, long long start_block, -+ int length); -+extern struct address_space_operations squashfs_symlink_aops; -+extern struct address_space_operations squashfs_aops; -+extern struct address_space_operations squashfs_aops_4K; -+extern struct inode_operations squashfs_dir_inode_ops; -+#else -+#define SQSH_EXTERN static -+#endif -+ -+#ifdef CONFIG_SQUASHFS_1_0_COMPATIBILITY -+extern int squashfs_1_0_supported(struct squashfs_sb_info *msblk); -+#else -+static inline int squashfs_1_0_supported(struct squashfs_sb_info *msblk) -+{ -+ return 0; -+} -+#endif -+ -+#ifdef CONFIG_SQUASHFS_2_0_COMPATIBILITY -+extern int squashfs_2_0_supported(struct squashfs_sb_info *msblk); -+#else -+static inline int squashfs_2_0_supported(struct squashfs_sb_info *msblk) -+{ -+ return 0; -+} -+#endif ---- a/include/linux/magic.h -+++ b/include/linux/magic.h -@@ -36,6 +36,9 @@ - #define REISER2FS_SUPER_MAGIC_STRING "ReIsEr2Fs" - #define REISER2FS_JR_SUPER_MAGIC_STRING "ReIsEr3Fs" - -+#define SQUASHFS_MAGIC 0x73717368 -+#define SQUASHFS_MAGIC_SWAP 0x68737173 -+ - #define SMB_SUPER_MAGIC 0x517B - #define USBDEVICE_SUPER_MAGIC 0x9fa2 - ---- /dev/null -+++ b/include/linux/squashfs_fs.h -@@ -0,0 +1,911 @@ -+#ifndef SQUASHFS_FS -+#define SQUASHFS_FS -+ -+/* -+ * Squashfs -+ * -+ * Copyright (c) 2002, 2003, 2004, 2005, 2006 -+ * Phillip Lougher <phillip@lougher.org.uk> -+ * -+ * This program is free software; you can redistribute it and/or -+ * modify it under the terms of the GNU General Public License -+ * as published by the Free Software Foundation; either version 2, -+ * or (at your option) any later version. -+ * -+ * This program is distributed in the hope that it will be useful, -+ * but WITHOUT ANY WARRANTY; without even the implied warranty of -+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the -+ * GNU General Public License for more details. -+ * -+ * You should have received a copy of the GNU General Public License -+ * along with this program; if not, write to the Free Software -+ * Foundation, 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. -+ * -+ * squashfs_fs.h -+ */ -+ -+#ifndef CONFIG_SQUASHFS_2_0_COMPATIBILITY -+#define CONFIG_SQUASHFS_2_0_COMPATIBILITY -+#endif -+ -+#ifdef CONFIG_SQUASHFS_VMALLOC -+#define SQUASHFS_ALLOC(a) vmalloc(a) -+#define SQUASHFS_FREE(a) vfree(a) -+#else -+#define SQUASHFS_ALLOC(a) kmalloc(a, GFP_KERNEL) -+#define SQUASHFS_FREE(a) kfree(a) -+#endif -+#define SQUASHFS_CACHED_FRAGMENTS CONFIG_SQUASHFS_FRAGMENT_CACHE_SIZE -+#define SQUASHFS_MAJOR 3 -+#define SQUASHFS_MINOR 0 -+#define SQUASHFS_START 0 -+ -+/* size of metadata (inode and directory) blocks */ -+#define SQUASHFS_METADATA_SIZE 8192 -+#define SQUASHFS_METADATA_LOG 13 -+ -+/* default size of data blocks */ -+#define SQUASHFS_FILE_SIZE 65536 -+#define SQUASHFS_FILE_LOG 16 -+ -+#define SQUASHFS_FILE_MAX_SIZE 65536 -+ -+/* Max number of uids and gids */ -+#define SQUASHFS_UIDS 256 -+#define SQUASHFS_GUIDS 255 -+ -+/* Max length of filename (not 255) */ -+#define SQUASHFS_NAME_LEN 256 -+ -+#define SQUASHFS_INVALID ((long long) 0xffffffffffff) -+#define SQUASHFS_INVALID_FRAG ((unsigned int) 0xffffffff) -+#define SQUASHFS_INVALID_BLK ((long long) -1) -+#define SQUASHFS_USED_BLK ((long long) -2) -+ -+/* Filesystem flags */ -+#define SQUASHFS_NOI 0 -+#define SQUASHFS_NOD 1 -+#define SQUASHFS_CHECK 2 -+#define SQUASHFS_NOF 3 -+#define SQUASHFS_NO_FRAG 4 -+#define SQUASHFS_ALWAYS_FRAG 5 -+#define SQUASHFS_DUPLICATE 6 -+ -+#define SQUASHFS_BIT(flag, bit) ((flag >> bit) & 1) -+ -+#define SQUASHFS_UNCOMPRESSED_INODES(flags) SQUASHFS_BIT(flags, \ -+ SQUASHFS_NOI) -+ -+#define SQUASHFS_UNCOMPRESSED_DATA(flags) SQUASHFS_BIT(flags, \ -+ SQUASHFS_NOD) -+ -+#define SQUASHFS_UNCOMPRESSED_FRAGMENTS(flags) SQUASHFS_BIT(flags, \ -+ SQUASHFS_NOF) -+ -+#define SQUASHFS_NO_FRAGMENTS(flags) SQUASHFS_BIT(flags, \ -+ SQUASHFS_NO_FRAG) -+ -+#define SQUASHFS_ALWAYS_FRAGMENTS(flags) SQUASHFS_BIT(flags, \ -+ SQUASHFS_ALWAYS_FRAG) -+ -+#define SQUASHFS_DUPLICATES(flags) SQUASHFS_BIT(flags, \ -+ SQUASHFS_DUPLICATE) -+ -+#define SQUASHFS_CHECK_DATA(flags) SQUASHFS_BIT(flags, \ -+ SQUASHFS_CHECK) -+ -+#define SQUASHFS_MKFLAGS(noi, nod, check_data, nof, no_frag, always_frag, \ -+ duplicate_checking) (noi | (nod << 1) | (check_data << 2) \ -+ | (nof << 3) | (no_frag << 4) | (always_frag << 5) | \ -+ (duplicate_checking << 6)) -+ -+/* Max number of types and file types */ -+#define SQUASHFS_DIR_TYPE 1 -+#define SQUASHFS_FILE_TYPE 2 -+#define SQUASHFS_SYMLINK_TYPE 3 -+#define SQUASHFS_BLKDEV_TYPE 4 -+#define SQUASHFS_CHRDEV_TYPE 5 -+#define SQUASHFS_FIFO_TYPE 6 -+#define SQUASHFS_SOCKET_TYPE 7 -+#define SQUASHFS_LDIR_TYPE 8 -+#define SQUASHFS_LREG_TYPE 9 -+ -+/* 1.0 filesystem type definitions */ -+#define SQUASHFS_TYPES 5 -+#define SQUASHFS_IPC_TYPE 0 -+ -+/* Flag whether block is compressed or uncompressed, bit is set if block is -+ * uncompressed */ -+#define SQUASHFS_COMPRESSED_BIT (1 << 15) -+ -+#define SQUASHFS_COMPRESSED_SIZE(B) (((B) & ~SQUASHFS_COMPRESSED_BIT) ? \ -+ (B) & ~SQUASHFS_COMPRESSED_BIT : SQUASHFS_COMPRESSED_BIT) -+ -+#define SQUASHFS_COMPRESSED(B) (!((B) & SQUASHFS_COMPRESSED_BIT)) -+ -+#define SQUASHFS_COMPRESSED_BIT_BLOCK (1 << 24) -+ -+#define SQUASHFS_COMPRESSED_SIZE_BLOCK(B) (((B) & \ -+ ~SQUASHFS_COMPRESSED_BIT_BLOCK) ? (B) & \ -+ ~SQUASHFS_COMPRESSED_BIT_BLOCK : SQUASHFS_COMPRESSED_BIT_BLOCK) -+ -+#define SQUASHFS_COMPRESSED_BLOCK(B) (!((B) & SQUASHFS_COMPRESSED_BIT_BLOCK)) -+ -+/* -+ * Inode number ops. Inodes consist of a compressed block number, and an -+ * uncompressed offset within that block -+ */ -+#define SQUASHFS_INODE_BLK(a) ((unsigned int) ((a) >> 16)) -+ -+#define SQUASHFS_INODE_OFFSET(a) ((unsigned int) ((a) & 0xffff)) -+ -+#define SQUASHFS_MKINODE(A, B) ((squashfs_inode_t)(((squashfs_inode_t) (A)\ -+ << 16) + (B))) -+ -+/* Compute 32 bit VFS inode number from squashfs inode number */ -+#define SQUASHFS_MK_VFS_INODE(a, b) ((unsigned int) (((a) << 8) + \ -+ ((b) >> 2) + 1)) -+/* XXX */ -+ -+/* Translate between VFS mode and squashfs mode */ -+#define SQUASHFS_MODE(a) ((a) & 0xfff) -+ -+/* fragment and fragment table defines */ -+#define SQUASHFS_FRAGMENT_BYTES(A) (A * sizeof(struct squashfs_fragment_entry)) -+ -+#define SQUASHFS_FRAGMENT_INDEX(A) (SQUASHFS_FRAGMENT_BYTES(A) / \ -+ SQUASHFS_METADATA_SIZE) -+ -+#define SQUASHFS_FRAGMENT_INDEX_OFFSET(A) (SQUASHFS_FRAGMENT_BYTES(A) % \ -+ SQUASHFS_METADATA_SIZE) -+ -+#define SQUASHFS_FRAGMENT_INDEXES(A) ((SQUASHFS_FRAGMENT_BYTES(A) + \ -+ SQUASHFS_METADATA_SIZE - 1) / \ -+ SQUASHFS_METADATA_SIZE) -+ -+#define SQUASHFS_FRAGMENT_INDEX_BYTES(A) (SQUASHFS_FRAGMENT_INDEXES(A) *\ -+ sizeof(long long)) -+ -+/* cached data constants for filesystem */ -+#define SQUASHFS_CACHED_BLKS 8 -+ -+#define SQUASHFS_MAX_FILE_SIZE_LOG 64 -+ -+#define SQUASHFS_MAX_FILE_SIZE ((long long) 1 << \ -+ (SQUASHFS_MAX_FILE_SIZE_LOG - 2)) -+ -+#define SQUASHFS_MARKER_BYTE 0xff -+ -+/* meta index cache */ -+#define SQUASHFS_META_INDEXES (SQUASHFS_METADATA_SIZE / sizeof(unsigned int)) -+#define SQUASHFS_META_ENTRIES 31 -+#define SQUASHFS_META_NUMBER 8 -+#define SQUASHFS_SLOTS 4 -+ -+#include <linux/magic.h> -+ -+struct meta_entry { -+ long long data_block; -+ unsigned int index_block; -+ unsigned short offset; -+ unsigned short pad; -+}; -+ -+struct meta_index { -+ unsigned int inode_number; -+ unsigned int offset; -+ unsigned short entries; -+ unsigned short skip; -+ unsigned short locked; -+ unsigned short pad; -+ struct meta_entry meta_entry[SQUASHFS_META_ENTRIES]; -+}; -+ -+ -+/* -+ * definitions for structures on disk -+ */ -+ -+typedef long long squashfs_block_t; -+typedef long long squashfs_inode_t; -+ -+struct squashfs_super_block { -+ unsigned int s_magic; -+ unsigned int inodes; -+ unsigned int bytes_used_2; -+ unsigned int uid_start_2; -+ unsigned int guid_start_2; -+ unsigned int inode_table_start_2; -+ unsigned int directory_table_start_2; -+ unsigned int s_major:16; -+ unsigned int s_minor:16; -+ unsigned int block_size_1:16; -+ unsigned int block_log:16; -+ unsigned int flags:8; -+ unsigned int no_uids:8; -+ unsigned int no_guids:8; -+ unsigned int mkfs_time /* time of filesystem creation */; -+ squashfs_inode_t root_inode; -+ unsigned int block_size; -+ unsigned int fragments; -+ unsigned int fragment_table_start_2; -+ long long bytes_used; -+ long long uid_start; -+ long long guid_start; -+ long long inode_table_start; -+ long long directory_table_start; -+ long long fragment_table_start; -+ long long unused; -+} __attribute__ ((packed)); -+ -+struct squashfs_dir_index { -+ unsigned int index; -+ unsigned int start_block; -+ unsigned char size; -+ unsigned char name[0]; -+} __attribute__ ((packed)); -+ -+#define SQUASHFS_BASE_INODE_HEADER \ -+ unsigned int inode_type:4; \ -+ unsigned int mode:12; \ -+ unsigned int uid:8; \ -+ unsigned int guid:8; \ -+ unsigned int mtime; \ -+ unsigned int inode_number; -+ -+struct squashfs_base_inode_header { -+ SQUASHFS_BASE_INODE_HEADER; -+} __attribute__ ((packed)); -+ -+struct squashfs_ipc_inode_header { -+ SQUASHFS_BASE_INODE_HEADER; -+ unsigned int nlink; -+} __attribute__ ((packed)); -+ -+struct squashfs_dev_inode_header { -+ SQUASHFS_BASE_INODE_HEADER; -+ unsigned int nlink; -+ unsigned short rdev; -+} __attribute__ ((packed)); -+ -+struct squashfs_symlink_inode_header { -+ SQUASHFS_BASE_INODE_HEADER; -+ unsigned int nlink; -+ unsigned short symlink_size; -+ char symlink[0]; -+} __attribute__ ((packed)); -+ -+struct squashfs_reg_inode_header { -+ SQUASHFS_BASE_INODE_HEADER; -+ squashfs_block_t start_block; -+ unsigned int fragment; -+ unsigned int offset; -+ unsigned int file_size; -+ unsigned short block_list[0]; -+} __attribute__ ((packed)); -+ -+struct squashfs_lreg_inode_header { -+ SQUASHFS_BASE_INODE_HEADER; -+ unsigned int nlink; -+ squashfs_block_t start_block; -+ unsigned int fragment; -+ unsigned int offset; -+ long long file_size; -+ unsigned short block_list[0]; -+} __attribute__ ((packed)); -+ -+struct squashfs_dir_inode_header { -+ SQUASHFS_BASE_INODE_HEADER; -+ unsigned int nlink; -+ unsigned int file_size:19; -+ unsigned int offset:13; -+ unsigned int start_block; -+ unsigned int parent_inode; -+} __attribute__ ((packed)); -+ -+struct squashfs_ldir_inode_header { -+ SQUASHFS_BASE_INODE_HEADER; -+ unsigned int nlink; -+ unsigned int file_size:27; -+ unsigned int offset:13; -+ unsigned int start_block; -+ unsigned int i_count:16; -+ unsigned int parent_inode; -+ struct squashfs_dir_index index[0]; -+} __attribute__ ((packed)); -+ -+union squashfs_inode_header { -+ struct squashfs_base_inode_header base; -+ struct squashfs_dev_inode_header dev; -+ struct squashfs_symlink_inode_header symlink; -+ struct squashfs_reg_inode_header reg; -+ struct squashfs_lreg_inode_header lreg; -+ struct squashfs_dir_inode_header dir; -+ struct squashfs_ldir_inode_header ldir; -+ struct squashfs_ipc_inode_header ipc; -+}; -+ -+struct squashfs_dir_entry { -+ unsigned int offset:13; -+ unsigned int type:3; -+ unsigned int size:8; -+ int inode_number:16; -+ char name[0]; -+} __attribute__ ((packed)); -+ -+struct squashfs_dir_header { -+ unsigned int count:8; -+ unsigned int start_block; -+ unsigned int inode_number; -+} __attribute__ ((packed)); -+ -+struct squashfs_fragment_entry { -+ long long start_block; -+ unsigned int size; -+ unsigned int unused; -+} __attribute__ ((packed)); -+ -+extern int squashfs_uncompress_block(void *d, int dstlen, void *s, int srclen); -+extern int squashfs_uncompress_init(void); -+extern int squashfs_uncompress_exit(void); -+ -+/* -+ * macros to convert each packed bitfield structure from little endian to big -+ * endian and vice versa. These are needed when creating or using a filesystem -+ * on a machine with different byte ordering to the target architecture. -+ * -+ */ -+ -+#define SQUASHFS_SWAP_START \ -+ int bits;\ -+ int b_pos;\ -+ unsigned long long val;\ -+ unsigned char *s;\ -+ unsigned char *d; -+ -+#define SQUASHFS_SWAP_SUPER_BLOCK(s, d) {\ -+ SQUASHFS_SWAP_START\ -+ SQUASHFS_MEMSET(s, d, sizeof(struct squashfs_super_block));\ -+ SQUASHFS_SWAP((s)->s_magic, d, 0, 32);\ -+ SQUASHFS_SWAP((s)->inodes, d, 32, 32);\ -+ SQUASHFS_SWAP((s)->bytes_used_2, d, 64, 32);\ -+ SQUASHFS_SWAP((s)->uid_start_2, d, 96, 32);\ -+ SQUASHFS_SWAP((s)->guid_start_2, d, 128, 32);\ -+ SQUASHFS_SWAP((s)->inode_table_start_2, d, 160, 32);\ -+ SQUASHFS_SWAP((s)->directory_table_start_2, d, 192, 32);\ -+ SQUASHFS_SWAP((s)->s_major, d, 224, 16);\ -+ SQUASHFS_SWAP((s)->s_minor, d, 240, 16);\ -+ SQUASHFS_SWAP((s)->block_size_1, d, 256, 16);\ -+ SQUASHFS_SWAP((s)->block_log, d, 272, 16);\ -+ SQUASHFS_SWAP((s)->flags, d, 288, 8);\ -+ SQUASHFS_SWAP((s)->no_uids, d, 296, 8);\ -+ SQUASHFS_SWAP((s)->no_guids, d, 304, 8);\ -+ SQUASHFS_SWAP((s)->mkfs_time, d, 312, 32);\ -+ SQUASHFS_SWAP((s)->root_inode, d, 344, 64);\ -+ SQUASHFS_SWAP((s)->block_size, d, 408, 32);\ -+ SQUASHFS_SWAP((s)->fragments, d, 440, 32);\ -+ SQUASHFS_SWAP((s)->fragment_table_start_2, d, 472, 32);\ -+ SQUASHFS_SWAP((s)->bytes_used, d, 504, 64);\ -+ SQUASHFS_SWAP((s)->uid_start, d, 568, 64);\ -+ SQUASHFS_SWAP((s)->guid_start, d, 632, 64);\ -+ SQUASHFS_SWAP((s)->inode_table_start, d, 696, 64);\ -+ SQUASHFS_SWAP((s)->directory_table_start, d, 760, 64);\ -+ SQUASHFS_SWAP((s)->fragment_table_start, d, 824, 64);\ -+ SQUASHFS_SWAP((s)->unused, d, 888, 64);\ -+} -+ -+#define SQUASHFS_SWAP_BASE_INODE_CORE(s, d, n)\ -+ SQUASHFS_MEMSET(s, d, n);\ -+ SQUASHFS_SWAP((s)->inode_type, d, 0, 4);\ -+ SQUASHFS_SWAP((s)->mode, d, 4, 12);\ -+ SQUASHFS_SWAP((s)->uid, d, 16, 8);\ -+ SQUASHFS_SWAP((s)->guid, d, 24, 8);\ -+ SQUASHFS_SWAP((s)->mtime, d, 32, 32);\ -+ SQUASHFS_SWAP((s)->inode_number, d, 64, 32); -+ -+#define SQUASHFS_SWAP_BASE_INODE_HEADER(s, d, n) {\ -+ SQUASHFS_SWAP_START\ -+ SQUASHFS_SWAP_BASE_INODE_CORE(s, d, n)\ -+} -+ -+#define SQUASHFS_SWAP_IPC_INODE_HEADER(s, d) {\ -+ SQUASHFS_SWAP_START\ -+ SQUASHFS_SWAP_BASE_INODE_CORE(s, d, \ -+ sizeof(struct squashfs_ipc_inode_header))\ -+ SQUASHFS_SWAP((s)->nlink, d, 96, 32);\ -+} -+ -+#define SQUASHFS_SWAP_DEV_INODE_HEADER(s, d) {\ -+ SQUASHFS_SWAP_START\ -+ SQUASHFS_SWAP_BASE_INODE_CORE(s, d, \ -+ sizeof(struct squashfs_dev_inode_header)); \ -+ SQUASHFS_SWAP((s)->nlink, d, 96, 32);\ -+ SQUASHFS_SWAP((s)->rdev, d, 128, 16);\ -+} -+ -+#define SQUASHFS_SWAP_SYMLINK_INODE_HEADER(s, d) {\ -+ SQUASHFS_SWAP_START\ -+ SQUASHFS_SWAP_BASE_INODE_CORE(s, d, \ -+ sizeof(struct squashfs_symlink_inode_header));\ -+ SQUASHFS_SWAP((s)->nlink, d, 96, 32);\ -+ SQUASHFS_SWAP((s)->symlink_size, d, 128, 16);\ -+} -+ -+#define SQUASHFS_SWAP_REG_INODE_HEADER(s, d) {\ -+ SQUASHFS_SWAP_START\ -+ SQUASHFS_SWAP_BASE_INODE_CORE(s, d, \ -+ sizeof(struct squashfs_reg_inode_header));\ -+ SQUASHFS_SWAP((s)->start_block, d, 96, 64);\ -+ SQUASHFS_SWAP((s)->fragment, d, 160, 32);\ -+ SQUASHFS_SWAP((s)->offset, d, 192, 32);\ -+ SQUASHFS_SWAP((s)->file_size, d, 224, 32);\ -+} -+ -+#define SQUASHFS_SWAP_LREG_INODE_HEADER(s, d) {\ -+ SQUASHFS_SWAP_START\ -+ SQUASHFS_SWAP_BASE_INODE_CORE(s, d, \ -+ sizeof(struct squashfs_lreg_inode_header));\ -+ SQUASHFS_SWAP((s)->nlink, d, 96, 32);\ -+ SQUASHFS_SWAP((s)->start_block, d, 128, 64);\ -+ SQUASHFS_SWAP((s)->fragment, d, 192, 32);\ -+ SQUASHFS_SWAP((s)->offset, d, 224, 32);\ -+ SQUASHFS_SWAP((s)->file_size, d, 256, 64);\ -+} -+ -+#define SQUASHFS_SWAP_DIR_INODE_HEADER(s, d) {\ -+ SQUASHFS_SWAP_START\ -+ SQUASHFS_SWAP_BASE_INODE_CORE(s, d, \ -+ sizeof(struct squashfs_dir_inode_header));\ -+ SQUASHFS_SWAP((s)->nlink, d, 96, 32);\ -+ SQUASHFS_SWAP((s)->file_size, d, 128, 19);\ -+ SQUASHFS_SWAP((s)->offset, d, 147, 13);\ -+ SQUASHFS_SWAP((s)->start_block, d, 160, 32);\ -+ SQUASHFS_SWAP((s)->parent_inode, d, 192, 32);\ -+} -+ -+#define SQUASHFS_SWAP_LDIR_INODE_HEADER(s, d) {\ -+ SQUASHFS_SWAP_START\ -+ SQUASHFS_SWAP_BASE_INODE_CORE(s, d, \ -+ sizeof(struct squashfs_ldir_inode_header));\ -+ SQUASHFS_SWAP((s)->nlink, d, 96, 32);\ -+ SQUASHFS_SWAP((s)->file_size, d, 128, 27);\ -+ SQUASHFS_SWAP((s)->offset, d, 155, 13);\ -+ SQUASHFS_SWAP((s)->start_block, d, 168, 32);\ -+ SQUASHFS_SWAP((s)->i_count, d, 200, 16);\ -+ SQUASHFS_SWAP((s)->parent_inode, d, 216, 32);\ -+} -+ -+#define SQUASHFS_SWAP_DIR_INDEX(s, d) {\ -+ SQUASHFS_SWAP_START\ -+ SQUASHFS_MEMSET(s, d, sizeof(struct squashfs_dir_index));\ -+ SQUASHFS_SWAP((s)->index, d, 0, 32);\ -+ SQUASHFS_SWAP((s)->start_block, d, 32, 32);\ -+ SQUASHFS_SWAP((s)->size, d, 64, 8);\ -+} -+ -+#define SQUASHFS_SWAP_DIR_HEADER(s, d) {\ -+ SQUASHFS_SWAP_START\ -+ SQUASHFS_MEMSET(s, d, sizeof(struct squashfs_dir_header));\ -+ SQUASHFS_SWAP((s)->count, d, 0, 8);\ -+ SQUASHFS_SWAP((s)->start_block, d, 8, 32);\ -+ SQUASHFS_SWAP((s)->inode_number, d, 40, 32);\ -+} -+ -+#define SQUASHFS_SWAP_DIR_ENTRY(s, d) {\ -+ SQUASHFS_SWAP_START\ -+ SQUASHFS_MEMSET(s, d, sizeof(struct squashfs_dir_entry));\ -+ SQUASHFS_SWAP((s)->offset, d, 0, 13);\ -+ SQUASHFS_SWAP((s)->type, d, 13, 3);\ -+ SQUASHFS_SWAP((s)->size, d, 16, 8);\ -+ SQUASHFS_SWAP((s)->inode_number, d, 24, 16);\ -+} -+ -+#define SQUASHFS_SWAP_FRAGMENT_ENTRY(s, d) {\ -+ SQUASHFS_SWAP_START\ -+ SQUASHFS_MEMSET(s, d, sizeof(struct squashfs_fragment_entry));\ -+ SQUASHFS_SWAP((s)->start_block, d, 0, 64);\ -+ SQUASHFS_SWAP((s)->size, d, 64, 32);\ -+} -+ -+#define SQUASHFS_SWAP_SHORTS(s, d, n) {\ -+ int entry;\ -+ int bit_position;\ -+ SQUASHFS_SWAP_START\ -+ SQUASHFS_MEMSET(s, d, n * 2);\ -+ for(entry = 0, bit_position = 0; entry < n; entry++, bit_position += \ -+ 16)\ -+ SQUASHFS_SWAP(s[entry], d, bit_position, 16);\ -+} -+ -+#define SQUASHFS_SWAP_INTS(s, d, n) {\ -+ int entry;\ -+ int bit_position;\ -+ SQUASHFS_SWAP_START\ -+ SQUASHFS_MEMSET(s, d, n * 4);\ -+ for(entry = 0, bit_position = 0; entry < n; entry++, bit_position += \ -+ 32)\ -+ SQUASHFS_SWAP(s[entry], d, bit_position, 32);\ -+} -+ -+#define SQUASHFS_SWAP_LONG_LONGS(s, d, n) {\ -+ int entry;\ -+ int bit_position;\ -+ SQUASHFS_SWAP_START\ -+ SQUASHFS_MEMSET(s, d, n * 8);\ -+ for(entry = 0, bit_position = 0; entry < n; entry++, bit_position += \ -+ 64)\ -+ SQUASHFS_SWAP(s[entry], d, bit_position, 64);\ -+} -+ -+#define SQUASHFS_SWAP_DATA(s, d, n, bits) {\ -+ int entry;\ -+ int bit_position;\ -+ SQUASHFS_SWAP_START\ -+ SQUASHFS_MEMSET(s, d, n * bits / 8);\ -+ for(entry = 0, bit_position = 0; entry < n; entry++, bit_position += \ -+ bits)\ -+ SQUASHFS_SWAP(s[entry], d, bit_position, bits);\ -+} -+ -+#define SQUASHFS_SWAP_FRAGMENT_INDEXES(s, d, n) SQUASHFS_SWAP_LONG_LONGS(s, d, n) -+ -+#ifdef CONFIG_SQUASHFS_1_0_COMPATIBILITY -+ -+struct squashfs_base_inode_header_1 { -+ unsigned int inode_type:4; -+ unsigned int mode:12; /* protection */ -+ unsigned int uid:4; /* index into uid table */ -+ unsigned int guid:4; /* index into guid table */ -+} __attribute__ ((packed)); -+ -+struct squashfs_ipc_inode_header_1 { -+ unsigned int inode_type:4; -+ unsigned int mode:12; /* protection */ -+ unsigned int uid:4; /* index into uid table */ -+ unsigned int guid:4; /* index into guid table */ -+ unsigned int type:4; -+ unsigned int offset:4; -+} __attribute__ ((packed)); -+ -+struct squashfs_dev_inode_header_1 { -+ unsigned int inode_type:4; -+ unsigned int mode:12; /* protection */ -+ unsigned int uid:4; /* index into uid table */ -+ unsigned int guid:4; /* index into guid table */ -+ unsigned short rdev; -+} __attribute__ ((packed)); -+ -+struct squashfs_symlink_inode_header_1 { -+ unsigned int inode_type:4; -+ unsigned int mode:12; /* protection */ -+ unsigned int uid:4; /* index into uid table */ -+ unsigned int guid:4; /* index into guid table */ -+ unsigned short symlink_size; -+ char symlink[0]; -+} __attribute__ ((packed)); -+ -+struct squashfs_reg_inode_header_1 { -+ unsigned int inode_type:4; -+ unsigned int mode:12; /* protection */ -+ unsigned int uid:4; /* index into uid table */ -+ unsigned int guid:4; /* index into guid table */ -+ unsigned int mtime; -+ unsigned int start_block; -+ unsigned int file_size:32; -+ unsigned short block_list[0]; -+} __attribute__ ((packed)); -+ -+struct squashfs_dir_inode_header_1 { -+ unsigned int inode_type:4; -+ unsigned int mode:12; /* protection */ -+ unsigned int uid:4; /* index into uid table */ -+ unsigned int guid:4; /* index into guid table */ -+ unsigned int file_size:19; -+ unsigned int offset:13; -+ unsigned int mtime; -+ unsigned int start_block:24; -+} __attribute__ ((packed)); -+ -+#define SQUASHFS_SWAP_BASE_INODE_CORE_1(s, d, n) \ -+ SQUASHFS_MEMSET(s, d, n);\ -+ SQUASHFS_SWAP((s)->inode_type, d, 0, 4);\ -+ SQUASHFS_SWAP((s)->mode, d, 4, 12);\ -+ SQUASHFS_SWAP((s)->uid, d, 16, 4);\ -+ SQUASHFS_SWAP((s)->guid, d, 20, 4); -+ -+#define SQUASHFS_SWAP_BASE_INODE_HEADER_1(s, d, n) {\ -+ SQUASHFS_SWAP_START\ -+ SQUASHFS_SWAP_BASE_INODE_CORE_1(s, d, n)\ -+} -+ -+#define SQUASHFS_SWAP_IPC_INODE_HEADER_1(s, d) {\ -+ SQUASHFS_SWAP_START\ -+ SQUASHFS_SWAP_BASE_INODE_CORE_1(s, d, \ -+ sizeof(struct squashfs_ipc_inode_header_1));\ -+ SQUASHFS_SWAP((s)->type, d, 24, 4);\ -+ SQUASHFS_SWAP((s)->offset, d, 28, 4);\ -+} -+ -+#define SQUASHFS_SWAP_DEV_INODE_HEADER_1(s, d) {\ -+ SQUASHFS_SWAP_START\ -+ SQUASHFS_SWAP_BASE_INODE_CORE_1(s, d, \ -+ sizeof(struct squashfs_dev_inode_header_1));\ -+ SQUASHFS_SWAP((s)->rdev, d, 24, 16);\ -+} -+ -+#define SQUASHFS_SWAP_SYMLINK_INODE_HEADER_1(s, d) {\ -+ SQUASHFS_SWAP_START\ -+ SQUASHFS_SWAP_BASE_INODE_CORE_1(s, d, \ -+ sizeof(struct squashfs_symlink_inode_header_1));\ -+ SQUASHFS_SWAP((s)->symlink_size, d, 24, 16);\ -+} -+ -+#define SQUASHFS_SWAP_REG_INODE_HEADER_1(s, d) {\ -+ SQUASHFS_SWAP_START\ -+ SQUASHFS_SWAP_BASE_INODE_CORE_1(s, d, \ -+ sizeof(struct squashfs_reg_inode_header_1));\ -+ SQUASHFS_SWAP((s)->mtime, d, 24, 32);\ -+ SQUASHFS_SWAP((s)->start_block, d, 56, 32);\ -+ SQUASHFS_SWAP((s)->file_size, d, 88, 32);\ -+} -+ -+#define SQUASHFS_SWAP_DIR_INODE_HEADER_1(s, d) {\ -+ SQUASHFS_SWAP_START\ -+ SQUASHFS_SWAP_BASE_INODE_CORE_1(s, d, \ -+ sizeof(struct squashfs_dir_inode_header_1));\ -+ SQUASHFS_SWAP((s)->file_size, d, 24, 19);\ -+ SQUASHFS_SWAP((s)->offset, d, 43, 13);\ -+ SQUASHFS_SWAP((s)->mtime, d, 56, 32);\ -+ SQUASHFS_SWAP((s)->start_block, d, 88, 24);\ -+} -+ -+#endif -+ -+#ifdef CONFIG_SQUASHFS_2_0_COMPATIBILITY -+ -+struct squashfs_dir_index_2 { -+ unsigned int index:27; -+ unsigned int start_block:29; -+ unsigned char size; -+ unsigned char name[0]; -+} __attribute__ ((packed)); -+ -+struct squashfs_base_inode_header_2 { -+ unsigned int inode_type:4; -+ unsigned int mode:12; /* protection */ -+ unsigned int uid:8; /* index into uid table */ -+ unsigned int guid:8; /* index into guid table */ -+} __attribute__ ((packed)); -+ -+struct squashfs_ipc_inode_header_2 { -+ unsigned int inode_type:4; -+ unsigned int mode:12; /* protection */ -+ unsigned int uid:8; /* index into uid table */ -+ unsigned int guid:8; /* index into guid table */ -+} __attribute__ ((packed)); -+ -+struct squashfs_dev_inode_header_2 { -+ unsigned int inode_type:4; -+ unsigned int mode:12; /* protection */ -+ unsigned int uid:8; /* index into uid table */ -+ unsigned int guid:8; /* index into guid table */ -+ unsigned short rdev; -+} __attribute__ ((packed)); -+ -+struct squashfs_symlink_inode_header_2 { -+ unsigned int inode_type:4; -+ unsigned int mode:12; /* protection */ -+ unsigned int uid:8; /* index into uid table */ -+ unsigned int guid:8; /* index into guid table */ -+ unsigned short symlink_size; -+ char symlink[0]; -+} __attribute__ ((packed)); -+ -+struct squashfs_reg_inode_header_2 { -+ unsigned int inode_type:4; -+ unsigned int mode:12; /* protection */ -+ unsigned int uid:8; /* index into uid table */ -+ unsigned int guid:8; /* index into guid table */ -+ unsigned int mtime; -+ unsigned int start_block; -+ unsigned int fragment; -+ unsigned int offset; -+ unsigned int file_size:32; -+ unsigned short block_list[0]; -+} __attribute__ ((packed)); -+ -+struct squashfs_dir_inode_header_2 { -+ unsigned int inode_type:4; -+ unsigned int mode:12; /* protection */ -+ unsigned int uid:8; /* index into uid table */ -+ unsigned int guid:8; /* index into guid table */ -+ unsigned int file_size:19; -+ unsigned int offset:13; -+ unsigned int mtime; -+ unsigned int start_block:24; -+} __attribute__ ((packed)); -+ -+struct squashfs_ldir_inode_header_2 { -+ unsigned int inode_type:4; -+ unsigned int mode:12; /* protection */ -+ unsigned int uid:8; /* index into uid table */ -+ unsigned int guid:8; /* index into guid table */ -+ unsigned int file_size:27; -+ unsigned int offset:13; -+ unsigned int mtime; -+ unsigned int start_block:24; -+ unsigned int i_count:16; -+ struct squashfs_dir_index_2 index[0]; -+} __attribute__ ((packed)); -+ -+union squashfs_inode_header_2 { -+ struct squashfs_base_inode_header_2 base; -+ struct squashfs_dev_inode_header_2 dev; -+ struct squashfs_symlink_inode_header_2 symlink; -+ struct squashfs_reg_inode_header_2 reg; -+ struct squashfs_dir_inode_header_2 dir; -+ struct squashfs_ldir_inode_header_2 ldir; -+ struct squashfs_ipc_inode_header_2 ipc; -+}; -+ -+struct squashfs_dir_header_2 { -+ unsigned int count:8; -+ unsigned int start_block:24; -+} __attribute__ ((packed)); -+ -+struct squashfs_dir_entry_2 { -+ unsigned int offset:13; -+ unsigned int type:3; -+ unsigned int size:8; -+ char name[0]; -+} __attribute__ ((packed)); -+ -+struct squashfs_fragment_entry_2 { -+ unsigned int start_block; -+ unsigned int size; -+} __attribute__ ((packed)); -+ -+#define SQUASHFS_SWAP_BASE_INODE_CORE_2(s, d, n)\ -+ SQUASHFS_MEMSET(s, d, n);\ -+ SQUASHFS_SWAP((s)->inode_type, d, 0, 4);\ -+ SQUASHFS_SWAP((s)->mode, d, 4, 12);\ -+ SQUASHFS_SWAP((s)->uid, d, 16, 8);\ -+ SQUASHFS_SWAP((s)->guid, d, 24, 8);\ -+ -+#define SQUASHFS_SWAP_BASE_INODE_HEADER_2(s, d, n) {\ -+ SQUASHFS_SWAP_START\ -+ SQUASHFS_SWAP_BASE_INODE_CORE_2(s, d, n)\ -+} -+ -+#define SQUASHFS_SWAP_IPC_INODE_HEADER_2(s, d) \ -+ SQUASHFS_SWAP_BASE_INODE_HEADER_2(s, d, sizeof(struct squashfs_ipc_inode_header_2)) -+ -+#define SQUASHFS_SWAP_DEV_INODE_HEADER_2(s, d) {\ -+ SQUASHFS_SWAP_START\ -+ SQUASHFS_SWAP_BASE_INODE_CORE_2(s, d, \ -+ sizeof(struct squashfs_dev_inode_header_2)); \ -+ SQUASHFS_SWAP((s)->rdev, d, 32, 16);\ -+} -+ -+#define SQUASHFS_SWAP_SYMLINK_INODE_HEADER_2(s, d) {\ -+ SQUASHFS_SWAP_START\ -+ SQUASHFS_SWAP_BASE_INODE_CORE_2(s, d, \ -+ sizeof(struct squashfs_symlink_inode_header_2));\ -+ SQUASHFS_SWAP((s)->symlink_size, d, 32, 16);\ -+} -+ -+#define SQUASHFS_SWAP_REG_INODE_HEADER_2(s, d) {\ -+ SQUASHFS_SWAP_START\ -+ SQUASHFS_SWAP_BASE_INODE_CORE_2(s, d, \ -+ sizeof(struct squashfs_reg_inode_header_2));\ -+ SQUASHFS_SWAP((s)->mtime, d, 32, 32);\ -+ SQUASHFS_SWAP((s)->start_block, d, 64, 32);\ -+ SQUASHFS_SWAP((s)->fragment, d, 96, 32);\ -+ SQUASHFS_SWAP((s)->offset, d, 128, 32);\ -+ SQUASHFS_SWAP((s)->file_size, d, 160, 32);\ -+} -+ -+#define SQUASHFS_SWAP_DIR_INODE_HEADER_2(s, d) {\ -+ SQUASHFS_SWAP_START\ -+ SQUASHFS_SWAP_BASE_INODE_CORE_2(s, d, \ -+ sizeof(struct squashfs_dir_inode_header_2));\ -+ SQUASHFS_SWAP((s)->file_size, d, 32, 19);\ -+ SQUASHFS_SWAP((s)->offset, d, 51, 13);\ -+ SQUASHFS_SWAP((s)->mtime, d, 64, 32);\ -+ SQUASHFS_SWAP((s)->start_block, d, 96, 24);\ -+} -+ -+#define SQUASHFS_SWAP_LDIR_INODE_HEADER_2(s, d) {\ -+ SQUASHFS_SWAP_START\ -+ SQUASHFS_SWAP_BASE_INODE_CORE_2(s, d, \ -+ sizeof(struct squashfs_ldir_inode_header_2));\ -+ SQUASHFS_SWAP((s)->file_size, d, 32, 27);\ -+ SQUASHFS_SWAP((s)->offset, d, 59, 13);\ -+ SQUASHFS_SWAP((s)->mtime, d, 72, 32);\ -+ SQUASHFS_SWAP((s)->start_block, d, 104, 24);\ -+ SQUASHFS_SWAP((s)->i_count, d, 128, 16);\ -+} -+ -+#define SQUASHFS_SWAP_DIR_INDEX_2(s, d) {\ -+ SQUASHFS_SWAP_START\ -+ SQUASHFS_MEMSET(s, d, sizeof(struct squashfs_dir_index_2));\ -+ SQUASHFS_SWAP((s)->index, d, 0, 27);\ -+ SQUASHFS_SWAP((s)->start_block, d, 27, 29);\ -+ SQUASHFS_SWAP((s)->size, d, 56, 8);\ -+} -+#define SQUASHFS_SWAP_DIR_HEADER_2(s, d) {\ -+ SQUASHFS_SWAP_START\ -+ SQUASHFS_MEMSET(s, d, sizeof(struct squashfs_dir_header_2));\ -+ SQUASHFS_SWAP((s)->count, d, 0, 8);\ -+ SQUASHFS_SWAP((s)->start_block, d, 8, 24);\ -+} -+ -+#define SQUASHFS_SWAP_DIR_ENTRY_2(s, d) {\ -+ SQUASHFS_SWAP_START\ -+ SQUASHFS_MEMSET(s, d, sizeof(struct squashfs_dir_entry_2));\ -+ SQUASHFS_SWAP((s)->offset, d, 0, 13);\ -+ SQUASHFS_SWAP((s)->type, d, 13, 3);\ -+ SQUASHFS_SWAP((s)->size, d, 16, 8);\ -+} -+ -+#define SQUASHFS_SWAP_FRAGMENT_ENTRY_2(s, d) {\ -+ SQUASHFS_SWAP_START\ -+ SQUASHFS_MEMSET(s, d, sizeof(struct squashfs_fragment_entry_2));\ -+ SQUASHFS_SWAP((s)->start_block, d, 0, 32);\ -+ SQUASHFS_SWAP((s)->size, d, 32, 32);\ -+} -+ -+#define SQUASHFS_SWAP_FRAGMENT_INDEXES_2(s, d, n) SQUASHFS_SWAP_INTS(s, d, n) -+ -+/* fragment and fragment table defines */ -+#define SQUASHFS_FRAGMENT_BYTES_2(A) (A * sizeof(struct squashfs_fragment_entry_2)) -+ -+#define SQUASHFS_FRAGMENT_INDEX_2(A) (SQUASHFS_FRAGMENT_BYTES_2(A) / \ -+ SQUASHFS_METADATA_SIZE) -+ -+#define SQUASHFS_FRAGMENT_INDEX_OFFSET_2(A) (SQUASHFS_FRAGMENT_BYTES_2(A) % \ -+ SQUASHFS_METADATA_SIZE) -+ -+#define SQUASHFS_FRAGMENT_INDEXES_2(A) ((SQUASHFS_FRAGMENT_BYTES_2(A) + \ -+ SQUASHFS_METADATA_SIZE - 1) / \ -+ SQUASHFS_METADATA_SIZE) -+ -+#define SQUASHFS_FRAGMENT_INDEX_BYTES_2(A) (SQUASHFS_FRAGMENT_INDEXES_2(A) *\ -+ sizeof(int)) -+ -+#endif -+ -+#ifdef __KERNEL__ -+ -+/* -+ * macros used to swap each structure entry, taking into account -+ * bitfields and different bitfield placing conventions on differing -+ * architectures -+ */ -+ -+#include <asm/byteorder.h> -+ -+#ifdef __BIG_ENDIAN -+ /* convert from little endian to big endian */ -+#define SQUASHFS_SWAP(value, p, pos, tbits) _SQUASHFS_SWAP(value, p, pos, \ -+ tbits, b_pos) -+#else -+ /* convert from big endian to little endian */ -+#define SQUASHFS_SWAP(value, p, pos, tbits) _SQUASHFS_SWAP(value, p, pos, \ -+ tbits, 64 - tbits - b_pos) -+#endif -+ -+#define _SQUASHFS_SWAP(value, p, pos, tbits, SHIFT) {\ -+ b_pos = pos % 8;\ -+ val = 0;\ -+ s = (unsigned char *)p + (pos / 8);\ -+ d = ((unsigned char *) &val) + 7;\ -+ for(bits = 0; bits < (tbits + b_pos); bits += 8) \ -+ *d-- = *s++;\ -+ value = (val >> (SHIFT))/* & ((1 << tbits) - 1)*/;\ -+} -+ -+#define SQUASHFS_MEMSET(s, d, n) memset(s, 0, n); -+ -+#endif -+#endif ---- /dev/null -+++ b/include/linux/squashfs_fs_i.h -@@ -0,0 +1,45 @@ -+#ifndef SQUASHFS_FS_I -+#define SQUASHFS_FS_I -+/* -+ * Squashfs -+ * -+ * Copyright (c) 2002, 2003, 2004, 2005, 2006 -+ * Phillip Lougher <phillip@lougher.org.uk> -+ * -+ * This program is free software; you can redistribute it and/or -+ * modify it under the terms of the GNU General Public License -+ * as published by the Free Software Foundation; either version 2, -+ * or (at your option) any later version. -+ * -+ * This program is distributed in the hope that it will be useful, -+ * but WITHOUT ANY WARRANTY; without even the implied warranty of -+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the -+ * GNU General Public License for more details. -+ * -+ * You should have received a copy of the GNU General Public License -+ * along with this program; if not, write to the Free Software -+ * Foundation, 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. -+ * -+ * squashfs_fs_i.h -+ */ -+ -+struct squashfs_inode_info { -+ long long start_block; -+ unsigned int offset; -+ union { -+ struct { -+ long long fragment_start_block; -+ unsigned int fragment_size; -+ unsigned int fragment_offset; -+ long long block_list_start; -+ } s1; -+ struct { -+ long long directory_index_start; -+ unsigned int directory_index_offset; -+ unsigned int directory_index_count; -+ unsigned int parent_inode; -+ } s2; -+ } u; -+ struct inode vfs_inode; -+}; -+#endif ---- /dev/null -+++ b/include/linux/squashfs_fs_sb.h -@@ -0,0 +1,74 @@ -+#ifndef SQUASHFS_FS_SB -+#define SQUASHFS_FS_SB -+/* -+ * Squashfs -+ * -+ * Copyright (c) 2002, 2003, 2004, 2005, 2006 -+ * Phillip Lougher <phillip@lougher.org.uk> -+ * -+ * This program is free software; you can redistribute it and/or -+ * modify it under the terms of the GNU General Public License -+ * as published by the Free Software Foundation; either version 2, -+ * or (at your option) any later version. -+ * -+ * This program is distributed in the hope that it will be useful, -+ * but WITHOUT ANY WARRANTY; without even the implied warranty of -+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the -+ * GNU General Public License for more details. -+ * -+ * You should have received a copy of the GNU General Public License -+ * along with this program; if not, write to the Free Software -+ * Foundation, 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. -+ * -+ * squashfs_fs_sb.h -+ */ -+ -+#include <linux/squashfs_fs.h> -+ -+struct squashfs_cache { -+ long long block; -+ int length; -+ long long next_index; -+ char *data; -+}; -+ -+struct squashfs_fragment_cache { -+ long long block; -+ int length; -+ unsigned int locked; -+ char *data; -+}; -+ -+struct squashfs_sb_info { -+ struct squashfs_super_block sblk; -+ int devblksize; -+ int devblksize_log2; -+ int swap; -+ struct squashfs_cache *block_cache; -+ struct squashfs_fragment_cache *fragment; -+ int next_cache; -+ int next_fragment; -+ int next_meta_index; -+ unsigned int *uid; -+ unsigned int *guid; -+ long long *fragment_index; -+ unsigned int *fragment_index_2; -+ unsigned int read_size; -+ char *read_data; -+ char *read_page; -+ struct semaphore read_data_mutex; -+ struct semaphore read_page_mutex; -+ struct semaphore block_cache_mutex; -+ struct semaphore fragment_mutex; -+ struct semaphore meta_index_mutex; -+ wait_queue_head_t waitq; -+ wait_queue_head_t fragment_wait_queue; -+ struct meta_index *meta_index; -+ struct inode *(*iget)(struct super_block *s, squashfs_inode_t \ -+ inode); -+ long long (*read_blocklist)(struct inode *inode, int \ -+ index, int readahead_blks, char *block_list, \ -+ unsigned short **block_p, unsigned int *bsize); -+ int (*read_fragment_index_table)(struct super_block *s); -+}; -+#endif ---- a/init/do_mounts_rd.c -+++ b/init/do_mounts_rd.c -@@ -5,6 +5,7 @@ - #include <linux/ext2_fs.h> - #include <linux/romfs_fs.h> - #include <linux/cramfs_fs.h> -+#include <linux/squashfs_fs.h> - #include <linux/initrd.h> - #include <linux/string.h> - -@@ -39,6 +40,7 @@ - * numbers could not be found. - * - * We currently check for the following magic numbers: -+ * squashfs - * minix - * ext2 - * romfs -@@ -53,6 +55,7 @@ - struct ext2_super_block *ext2sb; - struct romfs_super_block *romfsb; - struct cramfs_super *cramfsb; -+ struct squashfs_super_block *squashfsb; - int nblocks = -1; - unsigned char *buf; - -@@ -64,6 +67,7 @@ - ext2sb = (struct ext2_super_block *) buf; - romfsb = (struct romfs_super_block *) buf; - cramfsb = (struct cramfs_super *) buf; -+ squashfsb = (struct squashfs_super_block *) buf; - memset(buf, 0xe5, size); - - /* -@@ -101,6 +105,15 @@ - goto done; - } - -+ /* squashfs is at block zero too */ -+ if (squashfsb->s_magic == SQUASHFS_MAGIC) { -+ printk(KERN_NOTICE -+ "RAMDISK: squashfs filesystem found at block %d\n", -+ start_block); -+ nblocks = (squashfsb->bytes_used+BLOCK_SIZE-1)>>BLOCK_SIZE_BITS; -+ goto done; -+ } -+ - /* - * Read block 1 to test for minix and ext2 superblock - */ diff --git a/target/linux/generic-2.6/patches-2.6.22/002-lzma_decompress.patch b/target/linux/generic-2.6/patches-2.6.22/002-lzma_decompress.patch deleted file mode 100644 index f2f31f9..0000000 --- a/target/linux/generic-2.6/patches-2.6.22/002-lzma_decompress.patch +++ /dev/null @@ -1,788 +0,0 @@ ---- /dev/null -+++ b/include/linux/LzmaDecode.h -@@ -0,0 +1,100 @@ -+/* -+ LzmaDecode.h -+ LZMA Decoder interface -+ -+ LZMA SDK 4.05 Copyright (c) 1999-2004 Igor Pavlov (2004-08-25) -+ http://www.7-zip.org/ -+ -+ LZMA SDK is licensed under two licenses: -+ 1) GNU Lesser General Public License (GNU LGPL) -+ 2) Common Public License (CPL) -+ It means that you can select one of these two licenses and -+ follow rules of that license. -+ -+ SPECIAL EXCEPTION: -+ Igor Pavlov, as the author of this code, expressly permits you to -+ statically or dynamically link your code (or bind by name) to the -+ interfaces of this file without subjecting your linked code to the -+ terms of the CPL or GNU LGPL. Any modifications or additions -+ to this file, however, are subject to the LGPL or CPL terms. -+*/ -+ -+#ifndef __LZMADECODE_H -+#define __LZMADECODE_H -+ -+/* #define _LZMA_IN_CB */ -+/* Use callback for input data */ -+ -+/* #define _LZMA_OUT_READ */ -+/* Use read function for output data */ -+ -+/* #define _LZMA_PROB32 */ -+/* It can increase speed on some 32-bit CPUs, -+ but memory usage will be doubled in that case */ -+ -+/* #define _LZMA_LOC_OPT */ -+/* Enable local speed optimizations inside code */ -+ -+#ifndef UInt32 -+#ifdef _LZMA_UINT32_IS_ULONG -+#define UInt32 unsigned long -+#else -+#define UInt32 unsigned int -+#endif -+#endif -+ -+#ifdef _LZMA_PROB32 -+#define CProb UInt32 -+#else -+#define CProb unsigned short -+#endif -+ -+#define LZMA_RESULT_OK 0 -+#define LZMA_RESULT_DATA_ERROR 1 -+#define LZMA_RESULT_NOT_ENOUGH_MEM 2 -+ -+#ifdef _LZMA_IN_CB -+typedef struct _ILzmaInCallback -+{ -+ int (*Read)(void *object, unsigned char **buffer, UInt32 *bufferSize); -+} ILzmaInCallback; -+#endif -+ -+#define LZMA_BASE_SIZE 1846 -+#define LZMA_LIT_SIZE 768 -+ -+/* -+bufferSize = (LZMA_BASE_SIZE + (LZMA_LIT_SIZE << (lc + lp)))* sizeof(CProb) -+bufferSize += 100 in case of _LZMA_OUT_READ -+by default CProb is unsigned short, -+but if specify _LZMA_PROB_32, CProb will be UInt32(unsigned int) -+*/ -+ -+#ifdef _LZMA_OUT_READ -+int LzmaDecoderInit( -+ unsigned char *buffer, UInt32 bufferSize, -+ int lc, int lp, int pb, -+ unsigned char *dictionary, UInt32 dictionarySize, -+ #ifdef _LZMA_IN_CB -+ ILzmaInCallback *inCallback -+ #else -+ unsigned char *inStream, UInt32 inSize -+ #endif -+); -+#endif -+ -+int LzmaDecode( -+ unsigned char *buffer, -+ #ifndef _LZMA_OUT_READ -+ UInt32 bufferSize, -+ int lc, int lp, int pb, -+ #ifdef _LZMA_IN_CB -+ ILzmaInCallback *inCallback, -+ #else -+ unsigned char *inStream, UInt32 inSize, -+ #endif -+ #endif -+ unsigned char *outStream, UInt32 outSize, -+ UInt32 *outSizeProcessed); -+ -+#endif ---- /dev/null -+++ b/lib/LzmaDecode.c -@@ -0,0 +1,663 @@ -+/* -+ LzmaDecode.c -+ LZMA Decoder -+ -+ LZMA SDK 4.05 Copyright (c) 1999-2004 Igor Pavlov (2004-08-25) -+ http://www.7-zip.org/ -+ -+ LZMA SDK is licensed under two licenses: -+ 1) GNU Lesser General Public License (GNU LGPL) -+ 2) Common Public License (CPL) -+ It means that you can select one of these two licenses and -+ follow rules of that license. -+ -+ SPECIAL EXCEPTION: -+ Igor Pavlov, as the author of this code, expressly permits you to -+ statically or dynamically link your code (or bind by name) to the -+ interfaces of this file without subjecting your linked code to the -+ terms of the CPL or GNU LGPL. Any modifications or additions -+ to this file, however, are subject to the LGPL or CPL terms. -+*/ -+ -+#include <linux/LzmaDecode.h> -+ -+#ifndef Byte -+#define Byte unsigned char -+#endif -+ -+#define kNumTopBits 24 -+#define kTopValue ((UInt32)1 << kNumTopBits) -+ -+#define kNumBitModelTotalBits 11 -+#define kBitModelTotal (1 << kNumBitModelTotalBits) -+#define kNumMoveBits 5 -+ -+typedef struct _CRangeDecoder -+{ -+ Byte *Buffer; -+ Byte *BufferLim; -+ UInt32 Range; -+ UInt32 Code; -+ #ifdef _LZMA_IN_CB -+ ILzmaInCallback *InCallback; -+ int Result; -+ #endif -+ int ExtraBytes; -+} CRangeDecoder; -+ -+Byte RangeDecoderReadByte(CRangeDecoder *rd) -+{ -+ if (rd->Buffer == rd->BufferLim) -+ { -+ #ifdef _LZMA_IN_CB -+ UInt32 size; -+ rd->Result = rd->InCallback->Read(rd->InCallback, &rd->Buffer, &size); -+ rd->BufferLim = rd->Buffer + size; -+ if (size == 0) -+ #endif -+ { -+ rd->ExtraBytes = 1; -+ return 0xFF; -+ } -+ } -+ return (*rd->Buffer++); -+} -+ -+/* #define ReadByte (*rd->Buffer++) */ -+#define ReadByte (RangeDecoderReadByte(rd)) -+ -+void RangeDecoderInit(CRangeDecoder *rd, -+ #ifdef _LZMA_IN_CB -+ ILzmaInCallback *inCallback -+ #else -+ Byte *stream, UInt32 bufferSize -+ #endif -+ ) -+{ -+ int i; -+ #ifdef _LZMA_IN_CB -+ rd->InCallback = inCallback; -+ rd->Buffer = rd->BufferLim = 0; -+ #else -+ rd->Buffer = stream; -+ rd->BufferLim = stream + bufferSize; -+ #endif -+ rd->ExtraBytes = 0; -+ rd->Code = 0; -+ rd->Range = (0xFFFFFFFF); -+ for(i = 0; i < 5; i++) -+ rd->Code = (rd->Code << 8) | ReadByte; -+} -+ -+#define RC_INIT_VAR UInt32 range = rd->Range; UInt32 code = rd->Code; -+#define RC_FLUSH_VAR rd->Range = range; rd->Code = code; -+#define RC_NORMALIZE if (range < kTopValue) { range <<= 8; code = (code << 8) | ReadByte; } -+ -+UInt32 RangeDecoderDecodeDirectBits(CRangeDecoder *rd, int numTotalBits) -+{ -+ RC_INIT_VAR -+ UInt32 result = 0; -+ int i; -+ for (i = numTotalBits; i > 0; i--) -+ { -+ /* UInt32 t; */ -+ range >>= 1; -+ -+ result <<= 1; -+ if (code >= range) -+ { -+ code -= range; -+ result |= 1; -+ } -+ /* -+ t = (code - range) >> 31; -+ t &= 1; -+ code -= range & (t - 1); -+ result = (result + result) | (1 - t); -+ */ -+ RC_NORMALIZE -+ } -+ RC_FLUSH_VAR -+ return result; -+} -+ -+int RangeDecoderBitDecode(CProb *prob, CRangeDecoder *rd) -+{ -+ UInt32 bound = (rd->Range >> kNumBitModelTotalBits) * *prob; -+ if (rd->Code < bound) -+ { -+ rd->Range = bound; -+ *prob += (kBitModelTotal - *prob) >> kNumMoveBits; -+ if (rd->Range < kTopValue) -+ { -+ rd->Code = (rd->Code << 8) | ReadByte; -+ rd->Range <<= 8; -+ } -+ return 0; -+ } -+ else -+ { -+ rd->Range -= bound; -+ rd->Code -= bound; -+ *prob -= (*prob) >> kNumMoveBits; -+ if (rd->Range < kTopValue) -+ { -+ rd->Code = (rd->Code << 8) | ReadByte; -+ rd->Range <<= 8; -+ } -+ return 1; -+ } -+} -+ -+#define RC_GET_BIT2(prob, mi, A0, A1) \ -+ UInt32 bound = (range >> kNumBitModelTotalBits) * *prob; \ -+ if (code < bound) \ -+ { A0; range = bound; *prob += (kBitModelTotal - *prob) >> kNumMoveBits; mi <<= 1; } \ -+ else \ -+ { A1; range -= bound; code -= bound; *prob -= (*prob) >> kNumMoveBits; mi = (mi + mi) + 1; } \ -+ RC_NORMALIZE -+ -+#define RC_GET_BIT(prob, mi) RC_GET_BIT2(prob, mi, ; , ;) -+ -+int RangeDecoderBitTreeDecode(CProb *probs, int numLevels, CRangeDecoder *rd) -+{ -+ int mi = 1; -+ int i; -+ #ifdef _LZMA_LOC_OPT -+ RC_INIT_VAR -+ #endif -+ for(i = numLevels; i > 0; i--) -+ { -+ #ifdef _LZMA_LOC_OPT -+ CProb *prob = probs + mi; -+ RC_GET_BIT(prob, mi) -+ #else -+ mi = (mi + mi) + RangeDecoderBitDecode(probs + mi, rd); -+ #endif -+ } -+ #ifdef _LZMA_LOC_OPT -+ RC_FLUSH_VAR -+ #endif -+ return mi - (1 << numLevels); -+} -+ -+int RangeDecoderReverseBitTreeDecode(CProb *probs, int numLevels, CRangeDecoder *rd) -+{ -+ int mi = 1; -+ int i; -+ int symbol = 0; -+ #ifdef _LZMA_LOC_OPT -+ RC_INIT_VAR -+ #endif -+ for(i = 0; i < numLevels; i++) -+ { -+ #ifdef _LZMA_LOC_OPT -+ CProb *prob = probs + mi; -+ RC_GET_BIT2(prob, mi, ; , symbol |= (1 << i)) -+ #else -+ int bit = RangeDecoderBitDecode(probs + mi, rd); -+ mi = mi + mi + bit; -+ symbol |= (bit << i); -+ #endif -+ } -+ #ifdef _LZMA_LOC_OPT -+ RC_FLUSH_VAR -+ #endif -+ return symbol; -+} -+ -+Byte LzmaLiteralDecode(CProb *probs, CRangeDecoder *rd) -+{ -+ int symbol = 1; -+ #ifdef _LZMA_LOC_OPT -+ RC_INIT_VAR -+ #endif -+ do -+ { -+ #ifdef _LZMA_LOC_OPT -+ CProb *prob = probs + symbol; -+ RC_GET_BIT(prob, symbol) -+ #else -+ symbol = (symbol + symbol) | RangeDecoderBitDecode(probs + symbol, rd); -+ #endif -+ } -+ while (symbol < 0x100); -+ #ifdef _LZMA_LOC_OPT -+ RC_FLUSH_VAR -+ #endif -+ return symbol; -+} -+ -+Byte LzmaLiteralDecodeMatch(CProb *probs, CRangeDecoder *rd, Byte matchByte) -+{ -+ int symbol = 1; -+ #ifdef _LZMA_LOC_OPT -+ RC_INIT_VAR -+ #endif -+ do -+ { -+ int bit; -+ int matchBit = (matchByte >> 7) & 1; -+ matchByte <<= 1; -+ #ifdef _LZMA_LOC_OPT -+ { -+ CProb *prob = probs + ((1 + matchBit) << 8) + symbol; -+ RC_GET_BIT2(prob, symbol, bit = 0, bit = 1) -+ } -+ #else -+ bit = RangeDecoderBitDecode(probs + ((1 + matchBit) << 8) + symbol, rd); -+ symbol = (symbol << 1) | bit; -+ #endif -+ if (matchBit != bit) -+ { -+ while (symbol < 0x100) -+ { -+ #ifdef _LZMA_LOC_OPT -+ CProb *prob = probs + symbol; -+ RC_GET_BIT(prob, symbol) -+ #else -+ symbol = (symbol + symbol) | RangeDecoderBitDecode(probs + symbol, rd); -+ #endif -+ } -+ break; -+ } -+ } -+ while (symbol < 0x100); -+ #ifdef _LZMA_LOC_OPT -+ RC_FLUSH_VAR -+ #endif -+ return symbol; -+} -+ -+#define kNumPosBitsMax 4 -+#define kNumPosStatesMax (1 << kNumPosBitsMax) -+ -+#define kLenNumLowBits 3 -+#define kLenNumLowSymbols (1 << kLenNumLowBits) -+#define kLenNumMidBits 3 -+#define kLenNumMidSymbols (1 << kLenNumMidBits) -+#define kLenNumHighBits 8 -+#define kLenNumHighSymbols (1 << kLenNumHighBits) -+ -+#define LenChoice 0 -+#define LenChoice2 (LenChoice + 1) -+#define LenLow (LenChoice2 + 1) -+#define LenMid (LenLow + (kNumPosStatesMax << kLenNumLowBits)) -+#define LenHigh (LenMid + (kNumPosStatesMax << kLenNumMidBits)) -+#define kNumLenProbs (LenHigh + kLenNumHighSymbols) -+ -+int LzmaLenDecode(CProb *p, CRangeDecoder *rd, int posState) -+{ -+ if(RangeDecoderBitDecode(p + LenChoice, rd) == 0) -+ return RangeDecoderBitTreeDecode(p + LenLow + -+ (posState << kLenNumLowBits), kLenNumLowBits, rd); -+ if(RangeDecoderBitDecode(p + LenChoice2, rd) == 0) -+ return kLenNumLowSymbols + RangeDecoderBitTreeDecode(p + LenMid + -+ (posState << kLenNumMidBits), kLenNumMidBits, rd); -+ return kLenNumLowSymbols + kLenNumMidSymbols + -+ RangeDecoderBitTreeDecode(p + LenHigh, kLenNumHighBits, rd); -+} -+ -+#define kNumStates 12 -+ -+#define kStartPosModelIndex 4 -+#define kEndPosModelIndex 14 -+#define kNumFullDistances (1 << (kEndPosModelIndex >> 1)) -+ -+#define kNumPosSlotBits 6 -+#define kNumLenToPosStates 4 -+ -+#define kNumAlignBits 4 -+#define kAlignTableSize (1 << kNumAlignBits) -+ -+#define kMatchMinLen 2 -+ -+#define IsMatch 0 -+#define IsRep (IsMatch + (kNumStates << kNumPosBitsMax)) -+#define IsRepG0 (IsRep + kNumStates) -+#define IsRepG1 (IsRepG0 + kNumStates) -+#define IsRepG2 (IsRepG1 + kNumStates) -+#define IsRep0Long (IsRepG2 + kNumStates) -+#define PosSlot (IsRep0Long + (kNumStates << kNumPosBitsMax)) -+#define SpecPos (PosSlot + (kNumLenToPosStates << kNumPosSlotBits)) -+#define Align (SpecPos + kNumFullDistances - kEndPosModelIndex) -+#define LenCoder (Align + kAlignTableSize) -+#define RepLenCoder (LenCoder + kNumLenProbs) -+#define Literal (RepLenCoder + kNumLenProbs) -+ -+#if Literal != LZMA_BASE_SIZE -+StopCompilingDueBUG -+#endif -+ -+#ifdef _LZMA_OUT_READ -+ -+typedef struct _LzmaVarState -+{ -+ CRangeDecoder RangeDecoder; -+ Byte *Dictionary; -+ UInt32 DictionarySize; -+ UInt32 DictionaryPos; -+ UInt32 GlobalPos; -+ UInt32 Reps[4]; -+ int lc; -+ int lp; -+ int pb; -+ int State; -+ int PreviousIsMatch; -+ int RemainLen; -+} LzmaVarState; -+ -+int LzmaDecoderInit( -+ unsigned char *buffer, UInt32 bufferSize, -+ int lc, int lp, int pb, -+ unsigned char *dictionary, UInt32 dictionarySize, -+ #ifdef _LZMA_IN_CB -+ ILzmaInCallback *inCallback -+ #else -+ unsigned char *inStream, UInt32 inSize -+ #endif -+ ) -+{ -+ LzmaVarState *vs = (LzmaVarState *)buffer; -+ CProb *p = (CProb *)(buffer + sizeof(LzmaVarState)); -+ UInt32 numProbs = Literal + ((UInt32)LZMA_LIT_SIZE << (lc + lp)); -+ UInt32 i; -+ if (bufferSize < numProbs * sizeof(CProb) + sizeof(LzmaVarState)) -+ return LZMA_RESULT_NOT_ENOUGH_MEM; -+ vs->Dictionary = dictionary; -+ vs->DictionarySize = dictionarySize; -+ vs->DictionaryPos = 0; -+ vs->GlobalPos = 0; -+ vs->Reps[0] = vs->Reps[1] = vs->Reps[2] = vs->Reps[3] = 1; -+ vs->lc = lc; -+ vs->lp = lp; -+ vs->pb = pb; -+ vs->State = 0; -+ vs->PreviousIsMatch = 0; -+ vs->RemainLen = 0; -+ dictionary[dictionarySize - 1] = 0; -+ for (i = 0; i < numProbs; i++) -+ p[i] = kBitModelTotal >> 1; -+ RangeDecoderInit(&vs->RangeDecoder, -+ #ifdef _LZMA_IN_CB -+ inCallback -+ #else -+ inStream, inSize -+ #endif -+ ); -+ return LZMA_RESULT_OK; -+} -+ -+int LzmaDecode(unsigned char *buffer, -+ unsigned char *outStream, UInt32 outSize, -+ UInt32 *outSizeProcessed) -+{ -+ LzmaVarState *vs = (LzmaVarState *)buffer; -+ CProb *p = (CProb *)(buffer + sizeof(LzmaVarState)); -+ CRangeDecoder rd = vs->RangeDecoder; -+ int state = vs->State; -+ int previousIsMatch = vs->PreviousIsMatch; -+ Byte previousByte; -+ UInt32 rep0 = vs->Reps[0], rep1 = vs->Reps[1], rep2 = vs->Reps[2], rep3 = vs->Reps[3]; -+ UInt32 nowPos = 0; -+ UInt32 posStateMask = (1 << (vs->pb)) - 1; -+ UInt32 literalPosMask = (1 << (vs->lp)) - 1; -+ int lc = vs->lc; -+ int len = vs->RemainLen; -+ UInt32 globalPos = vs->GlobalPos; -+ -+ Byte *dictionary = vs->Dictionary; -+ UInt32 dictionarySize = vs->DictionarySize; -+ UInt32 dictionaryPos = vs->DictionaryPos; -+ -+ if (len == -1) -+ { -+ *outSizeProcessed = 0; -+ return LZMA_RESULT_OK; -+ } -+ -+ while(len > 0 && nowPos < outSize) -+ { -+ UInt32 pos = dictionaryPos - rep0; -+ if (pos >= dictionarySize) -+ pos += dictionarySize; -+ outStream[nowPos++] = dictionary[dictionaryPos] = dictionary[pos]; -+ if (++dictionaryPos == dictionarySize) -+ dictionaryPos = 0; -+ len--; -+ } -+ if (dictionaryPos == 0) -+ previousByte = dictionary[dictionarySize - 1]; -+ else -+ previousByte = dictionary[dictionaryPos - 1]; -+#else -+ -+int LzmaDecode( -+ Byte *buffer, UInt32 bufferSize, -+ int lc, int lp, int pb, -+ #ifdef _LZMA_IN_CB -+ ILzmaInCallback *inCallback, -+ #else -+ unsigned char *inStream, UInt32 inSize, -+ #endif -+ unsigned char *outStream, UInt32 outSize, -+ UInt32 *outSizeProcessed) -+{ -+ UInt32 numProbs = Literal + ((UInt32)LZMA_LIT_SIZE << (lc + lp)); -+ CProb *p = (CProb *)buffer; -+ CRangeDecoder rd; -+ UInt32 i; -+ int state = 0; -+ int previousIsMatch = 0; -+ Byte previousByte = 0; -+ UInt32 rep0 = 1, rep1 = 1, rep2 = 1, rep3 = 1; -+ UInt32 nowPos = 0; -+ UInt32 posStateMask = (1 << pb) - 1; -+ UInt32 literalPosMask = (1 << lp) - 1; -+ int len = 0; -+ if (bufferSize < numProbs * sizeof(CProb)) -+ return LZMA_RESULT_NOT_ENOUGH_MEM; -+ for (i = 0; i < numProbs; i++) -+ p[i] = kBitModelTotal >> 1; -+ RangeDecoderInit(&rd, -+ #ifdef _LZMA_IN_CB -+ inCallback -+ #else -+ inStream, inSize -+ #endif -+ ); -+#endif -+ -+ *outSizeProcessed = 0; -+ while(nowPos < outSize) -+ { -+ int posState = (int)( -+ (nowPos -+ #ifdef _LZMA_OUT_READ -+ + globalPos -+ #endif -+ ) -+ & posStateMask); -+ #ifdef _LZMA_IN_CB -+ if (rd.Result != LZMA_RESULT_OK) -+ return rd.Result; -+ #endif -+ if (rd.ExtraBytes != 0) -+ return LZMA_RESULT_DATA_ERROR; -+ if (RangeDecoderBitDecode(p + IsMatch + (state << kNumPosBitsMax) + posState, &rd) == 0) -+ { -+ CProb *probs = p + Literal + (LZMA_LIT_SIZE * -+ ((( -+ (nowPos -+ #ifdef _LZMA_OUT_READ -+ + globalPos -+ #endif -+ ) -+ & literalPosMask) << lc) + (previousByte >> (8 - lc)))); -+ -+ if (state < 4) state = 0; -+ else if (state < 10) state -= 3; -+ else state -= 6; -+ if (previousIsMatch) -+ { -+ Byte matchByte; -+ #ifdef _LZMA_OUT_READ -+ UInt32 pos = dictionaryPos - rep0; -+ if (pos >= dictionarySize) -+ pos += dictionarySize; -+ matchByte = dictionary[pos]; -+ #else -+ matchByte = outStream[nowPos - rep0]; -+ #endif -+ previousByte = LzmaLiteralDecodeMatch(probs, &rd, matchByte); -+ previousIsMatch = 0; -+ } -+ else -+ previousByte = LzmaLiteralDecode(probs, &rd); -+ outStream[nowPos++] = previousByte; -+ #ifdef _LZMA_OUT_READ -+ dictionary[dictionaryPos] = previousByte; -+ if (++dictionaryPos == dictionarySize) -+ dictionaryPos = 0; -+ #endif -+ } -+ else -+ { -+ previousIsMatch = 1; -+ if (RangeDecoderBitDecode(p + IsRep + state, &rd) == 1) -+ { -+ if (RangeDecoderBitDecode(p + IsRepG0 + state, &rd) == 0) -+ { -+ if (RangeDecoderBitDecode(p + IsRep0Long + (state << kNumPosBitsMax) + posState, &rd) == 0) -+ { -+ #ifdef _LZMA_OUT_READ -+ UInt32 pos; -+ #endif -+ if ( -+ (nowPos -+ #ifdef _LZMA_OUT_READ -+ + globalPos -+ #endif -+ ) -+ == 0) -+ return LZMA_RESULT_DATA_ERROR; -+ state = state < 7 ? 9 : 11; -+ #ifdef _LZMA_OUT_READ -+ pos = dictionaryPos - rep0; -+ if (pos >= dictionarySize) -+ pos += dictionarySize; -+ previousByte = dictionary[pos]; -+ dictionary[dictionaryPos] = previousByte; -+ if (++dictionaryPos == dictionarySize) -+ dictionaryPos = 0; -+ #else -+ previousByte = outStream[nowPos - rep0]; -+ #endif -+ outStream[nowPos++] = previousByte; -+ continue; -+ } -+ } -+ else -+ { -+ UInt32 distance; -+ if(RangeDecoderBitDecode(p + IsRepG1 + state, &rd) == 0) -+ distance = rep1; -+ else -+ { -+ if(RangeDecoderBitDecode(p + IsRepG2 + state, &rd) == 0) -+ distance = rep2; -+ else -+ { -+ distance = rep3; -+ rep3 = rep2; -+ } -+ rep2 = rep1; -+ } -+ rep1 = rep0; -+ rep0 = distance; -+ } -+ len = LzmaLenDecode(p + RepLenCoder, &rd, posState); -+ state = state < 7 ? 8 : 11; -+ } -+ else -+ { -+ int posSlot; -+ rep3 = rep2; -+ rep2 = rep1; -+ rep1 = rep0; -+ state = state < 7 ? 7 : 10; -+ len = LzmaLenDecode(p + LenCoder, &rd, posState); -+ posSlot = RangeDecoderBitTreeDecode(p + PosSlot + -+ ((len < kNumLenToPosStates ? len : kNumLenToPosStates - 1) << -+ kNumPosSlotBits), kNumPosSlotBits, &rd); -+ if (posSlot >= kStartPosModelIndex) -+ { -+ int numDirectBits = ((posSlot >> 1) - 1); -+ rep0 = ((2 | ((UInt32)posSlot & 1)) << numDirectBits); -+ if (posSlot < kEndPosModelIndex) -+ { -+ rep0 += RangeDecoderReverseBitTreeDecode( -+ p + SpecPos + rep0 - posSlot - 1, numDirectBits, &rd); -+ } -+ else -+ { -+ rep0 += RangeDecoderDecodeDirectBits(&rd, -+ numDirectBits - kNumAlignBits) << kNumAlignBits; -+ rep0 += RangeDecoderReverseBitTreeDecode(p + Align, kNumAlignBits, &rd); -+ } -+ } -+ else -+ rep0 = posSlot; -+ rep0++; -+ } -+ if (rep0 == (UInt32)(0)) -+ { -+ /* it's for stream version */ -+ len = -1; -+ break; -+ } -+ if (rep0 > nowPos -+ #ifdef _LZMA_OUT_READ -+ + globalPos -+ #endif -+ ) -+ { -+ return LZMA_RESULT_DATA_ERROR; -+ } -+ len += kMatchMinLen; -+ do -+ { -+ #ifdef _LZMA_OUT_READ -+ UInt32 pos = dictionaryPos - rep0; -+ if (pos >= dictionarySize) -+ pos += dictionarySize; -+ previousByte = dictionary[pos]; -+ dictionary[dictionaryPos] = previousByte; -+ if (++dictionaryPos == dictionarySize) -+ dictionaryPos = 0; -+ #else -+ previousByte = outStream[nowPos - rep0]; -+ #endif -+ outStream[nowPos++] = previousByte; -+ len--; -+ } -+ while(len > 0 && nowPos < outSize); -+ } -+ } -+ -+ #ifdef _LZMA_OUT_READ -+ vs->RangeDecoder = rd; -+ vs->DictionaryPos = dictionaryPos; -+ vs->GlobalPos = globalPos + nowPos; -+ vs->Reps[0] = rep0; -+ vs->Reps[1] = rep1; -+ vs->Reps[2] = rep2; -+ vs->Reps[3] = rep3; -+ vs->State = state; -+ vs->PreviousIsMatch = previousIsMatch; -+ vs->RemainLen = len; -+ #endif -+ -+ *outSizeProcessed = nowPos; -+ return LZMA_RESULT_OK; -+} ---- a/lib/Makefile -+++ b/lib/Makefile -@@ -13,7 +13,7 @@ - lib-y += kobject.o kref.o kobject_uevent.o klist.o - - obj-y += div64.o sort.o parser.o halfmd4.o debug_locks.o random32.o \ -- bust_spinlocks.o hexdump.o -+ bust_spinlocks.o hexdump.o LzmaDecode.o - - ifeq ($(CONFIG_DEBUG_KOBJECT),y) - CFLAGS_kobject.o += -DDEBUG -@@ -58,6 +58,7 @@ - obj-$(CONFIG_AUDIT_GENERIC) += audit.o - - obj-$(CONFIG_SWIOTLB) += swiotlb.o -+ - obj-$(CONFIG_FAULT_INJECTION) += fault-inject.o - - lib-$(CONFIG_GENERIC_BUG) += bug.o diff --git a/target/linux/generic-2.6/patches-2.6.22/003-squashfs_lzma.patch b/target/linux/generic-2.6/patches-2.6.22/003-squashfs_lzma.patch deleted file mode 100644 index 16cc873..0000000 --- a/target/linux/generic-2.6/patches-2.6.22/003-squashfs_lzma.patch +++ /dev/null @@ -1,107 +0,0 @@ ---- a/fs/squashfs/inode.c -+++ b/fs/squashfs/inode.c -@@ -4,6 +4,9 @@ - * Copyright (c) 2002, 2003, 2004, 2005, 2006 - * Phillip Lougher <phillip@lougher.org.uk> - * -+ * LZMA decompressor support added by Oleg I. Vdovikin -+ * Copyright (c) 2005 Oleg I.Vdovikin <oleg@cs.msu.su> -+ * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version 2, -@@ -21,6 +24,7 @@ - * inode.c - */ - -+#define SQUASHFS_LZMA - #include <linux/types.h> - #include <linux/squashfs_fs.h> - #include <linux/module.h> -@@ -44,6 +48,19 @@ - - #include "squashfs.h" - -+#ifdef SQUASHFS_LZMA -+#include <linux/LzmaDecode.h> -+ -+/* default LZMA settings, should be in sync with mksquashfs */ -+#define LZMA_LC 3 -+#define LZMA_LP 0 -+#define LZMA_PB 2 -+ -+#define LZMA_WORKSPACE_SIZE ((LZMA_BASE_SIZE + \ -+ (LZMA_LIT_SIZE << (LZMA_LC + LZMA_LP))) * sizeof(CProb)) -+ -+#endif -+ - static void squashfs_put_super(struct super_block *); - static int squashfs_statfs(struct dentry *, struct kstatfs *); - static int squashfs_symlink_readpage(struct file *file, struct page *page); -@@ -64,7 +81,11 @@ - const char *, void *, struct vfsmount *); - - -+#ifdef SQUASHFS_LZMA -+static unsigned char lzma_workspace[LZMA_WORKSPACE_SIZE]; -+#else - static z_stream stream; -+#endif - - static struct file_system_type squashfs_fs_type = { - .owner = THIS_MODULE, -@@ -249,6 +270,15 @@ - if (compressed) { - int zlib_err; - -+#ifdef SQUASHFS_LZMA -+ if ((zlib_err = LzmaDecode(lzma_workspace, -+ LZMA_WORKSPACE_SIZE, LZMA_LC, LZMA_LP, LZMA_PB, -+ c_buffer, c_byte, buffer, msblk->read_size, &bytes)) != LZMA_RESULT_OK) -+ { -+ ERROR("lzma returned unexpected result 0x%x\n", zlib_err); -+ bytes = 0; -+ } -+#else - stream.next_in = c_buffer; - stream.avail_in = c_byte; - stream.next_out = buffer; -@@ -263,7 +293,7 @@ - bytes = 0; - } else - bytes = stream.total_out; -- -+#endif - up(&msblk->read_data_mutex); - } - -@@ -2045,15 +2075,19 @@ - printk(KERN_INFO "squashfs: version 3.0 (2006/03/15) " - "Phillip Lougher\n"); - -+#ifndef SQUASHFS_LZMA - if (!(stream.workspace = vmalloc(zlib_inflate_workspacesize()))) { - ERROR("Failed to allocate zlib workspace\n"); - destroy_inodecache(); - err = -ENOMEM; - goto out; - } -+#endif - - if ((err = register_filesystem(&squashfs_fs_type))) { -+#ifndef SQUASHFS_LZMA - vfree(stream.workspace); -+#endif - destroy_inodecache(); - } - -@@ -2064,7 +2098,9 @@ - - static void __exit exit_squashfs_fs(void) - { -+#ifndef SQUASHFS_LZMA - vfree(stream.workspace); -+#endif - unregister_filesystem(&squashfs_fs_type); - destroy_inodecache(); - } diff --git a/target/linux/generic-2.6/patches-2.6.22/004-extra_optimization.patch b/target/linux/generic-2.6/patches-2.6.22/004-extra_optimization.patch deleted file mode 100644 index 5e70e1b..0000000 --- a/target/linux/generic-2.6/patches-2.6.22/004-extra_optimization.patch +++ /dev/null @@ -1,12 +0,0 @@ ---- a/Makefile -+++ b/Makefile -@@ -507,6 +507,9 @@ - NOSTDINC_FLAGS += -nostdinc -isystem $(shell $(CC) -print-file-name=include) - CHECKFLAGS += $(NOSTDINC_FLAGS) - -+# improve gcc optimization -+CFLAGS += $(call cc-option,-funit-at-a-time,) -+ - # warn about C99 declaration after statement - CFLAGS += $(call cc-option,-Wdeclaration-after-statement,) - diff --git a/target/linux/generic-2.6/patches-2.6.22/006-gcc4_inline_fix.patch b/target/linux/generic-2.6/patches-2.6.22/006-gcc4_inline_fix.patch deleted file mode 100644 index 0059088..0000000 --- a/target/linux/generic-2.6/patches-2.6.22/006-gcc4_inline_fix.patch +++ /dev/null @@ -1,11 +0,0 @@ ---- a/include/asm-mips/system.h -+++ b/include/asm-mips/system.h -@@ -188,7 +188,7 @@ - if something tries to do an invalid xchg(). */ - extern void __xchg_called_with_bad_pointer(void); - --static inline unsigned long __xchg(unsigned long x, volatile void * ptr, int size) -+static __always_inline unsigned long __xchg(unsigned long x, volatile void * ptr, int size) - { - switch (size) { - case 4: diff --git a/target/linux/generic-2.6/patches-2.6.22/007-samsung_flash.patch b/target/linux/generic-2.6/patches-2.6.22/007-samsung_flash.patch deleted file mode 100644 index 335ffd9..0000000 --- a/target/linux/generic-2.6/patches-2.6.22/007-samsung_flash.patch +++ /dev/null @@ -1,36 +0,0 @@ ---- a/drivers/mtd/chips/cfi_cmdset_0002.c -+++ b/drivers/mtd/chips/cfi_cmdset_0002.c -@@ -51,6 +51,7 @@ - #define SST49LF040B 0x0050 - #define SST49LF008A 0x005a - #define AT49BV6416 0x00d6 -+#define MANUFACTURER_SAMSUNG 0x00ec - - static int cfi_amdstd_read (struct mtd_info *, loff_t, size_t, size_t *, u_char *); - static int cfi_amdstd_write_words(struct mtd_info *, loff_t, size_t, size_t *, const u_char *); -@@ -294,12 +295,19 @@ - - if (extp->MajorVersion != '1' || - (extp->MinorVersion < '0' || extp->MinorVersion > '4')) { -- printk(KERN_ERR " Unknown Amd/Fujitsu Extended Query " -- "version %c.%c.\n", extp->MajorVersion, -- extp->MinorVersion); -- kfree(extp); -- kfree(mtd); -- return NULL; -+ if (cfi->mfr == MANUFACTURER_SAMSUNG && -+ (extp->MajorVersion == '3' && extp->MinorVersion == '3')) { -+ printk(KERN_NOTICE " Newer Samsung flash detected, " -+ "should be compatibile with Amd/Fujitsu.\n"); -+ } -+ else { -+ printk(KERN_ERR " Unknown Amd/Fujitsu Extended Query " -+ "version %c.%c.\n", extp->MajorVersion, -+ extp->MinorVersion); -+ kfree(extp); -+ kfree(mtd); -+ return NULL; -+ } - } - - /* Install our own private info structure */ diff --git a/target/linux/generic-2.6/patches-2.6.22/009-revert_intel_flash_breakage.patch b/target/linux/generic-2.6/patches-2.6.22/009-revert_intel_flash_breakage.patch deleted file mode 100644 index 51b84a7..0000000 --- a/target/linux/generic-2.6/patches-2.6.22/009-revert_intel_flash_breakage.patch +++ /dev/null @@ -1,169 +0,0 @@ ---- a/drivers/mtd/chips/cfi_cmdset_0001.c -+++ b/drivers/mtd/chips/cfi_cmdset_0001.c -@@ -933,7 +933,7 @@ - - static int __xipram xip_wait_for_operation( - struct map_info *map, struct flchip *chip, -- unsigned long adr, unsigned int chip_op_time ) -+ unsigned long adr, int *chip_op_time ) - { - struct cfi_private *cfi = map->fldrv_priv; - struct cfi_pri_intelext *cfip = cfi->cmdset_priv; -@@ -942,7 +942,7 @@ - flstate_t oldstate, newstate; - - start = xip_currtime(); -- usec = chip_op_time * 8; -+ usec = *chip_op_time * 8; - if (usec == 0) - usec = 500000; - done = 0; -@@ -1052,8 +1052,8 @@ - #define XIP_INVAL_CACHED_RANGE(map, from, size) \ - INVALIDATE_CACHED_RANGE(map, from, size) - --#define INVAL_CACHE_AND_WAIT(map, chip, cmd_adr, inval_adr, inval_len, usec) \ -- xip_wait_for_operation(map, chip, cmd_adr, usec) -+#define INVAL_CACHE_AND_WAIT(map, chip, cmd_adr, inval_adr, inval_len, p_usec) \ -+ xip_wait_for_operation(map, chip, cmd_adr, p_usec) - - #else - -@@ -1065,65 +1065,65 @@ - static int inval_cache_and_wait_for_operation( - struct map_info *map, struct flchip *chip, - unsigned long cmd_adr, unsigned long inval_adr, int inval_len, -- unsigned int chip_op_time) -+ int *chip_op_time ) - { - struct cfi_private *cfi = map->fldrv_priv; - map_word status, status_OK = CMD(0x80); -- int chip_state = chip->state; -- unsigned int timeo, sleep_time; -+ int z, chip_state = chip->state; -+ unsigned long timeo; - - spin_unlock(chip->mutex); - if (inval_len) - INVALIDATE_CACHED_RANGE(map, inval_adr, inval_len); -+ if (*chip_op_time) -+ cfi_udelay(*chip_op_time); - spin_lock(chip->mutex); - -- /* set our timeout to 8 times the expected delay */ -- timeo = chip_op_time * 8; -- if (!timeo) -- timeo = 500000; -- sleep_time = chip_op_time / 2; -+ timeo = *chip_op_time * 8 * HZ / 1000000; -+ if (timeo < HZ/2) -+ timeo = HZ/2; -+ timeo += jiffies; - -+ z = 0; - for (;;) { -+ if (chip->state != chip_state) { -+ /* Someone's suspended the operation: sleep */ -+ DECLARE_WAITQUEUE(wait, current); -+ -+ set_current_state(TASK_UNINTERRUPTIBLE); -+ add_wait_queue(&chip->wq, &wait); -+ spin_unlock(chip->mutex); -+ schedule(); -+ remove_wait_queue(&chip->wq, &wait); -+ timeo = jiffies + (HZ / 2); /* FIXME */ -+ spin_lock(chip->mutex); -+ continue; -+ } -+ - status = map_read(map, cmd_adr); - if (map_word_andequal(map, status, status_OK, status_OK)) - break; - -- if (!timeo) { -+ /* OK Still waiting */ -+ if (time_after(jiffies, timeo)) { - map_write(map, CMD(0x70), cmd_adr); - chip->state = FL_STATUS; - return -ETIME; - } - -- /* OK Still waiting. Drop the lock, wait a while and retry. */ -+ /* Latency issues. Drop the lock, wait a while and retry */ -+ z++; - spin_unlock(chip->mutex); -- if (sleep_time >= 1000000/HZ) { -- /* -- * Half of the normal delay still remaining -- * can be performed with a sleeping delay instead -- * of busy waiting. -- */ -- msleep(sleep_time/1000); -- timeo -= sleep_time; -- sleep_time = 1000000/HZ; -- } else { -- udelay(1); -- cond_resched(); -- timeo--; -- } -+ cfi_udelay(1); - spin_lock(chip->mutex); -- -- while (chip->state != chip_state) { -- /* Someone's suspended the operation: sleep */ -- DECLARE_WAITQUEUE(wait, current); -- set_current_state(TASK_UNINTERRUPTIBLE); -- add_wait_queue(&chip->wq, &wait); -- spin_unlock(chip->mutex); -- schedule(); -- remove_wait_queue(&chip->wq, &wait); -- spin_lock(chip->mutex); -- } - } - -+ if (!z) { -+ if (!--(*chip_op_time)) -+ *chip_op_time = 1; -+ } else if (z > 1) -+ ++(*chip_op_time); -+ - /* Done and happy. */ - chip->state = FL_STATUS; - return 0; -@@ -1132,7 +1132,8 @@ - #endif - - #define WAIT_TIMEOUT(map, chip, adr, udelay) \ -- INVAL_CACHE_AND_WAIT(map, chip, adr, 0, 0, udelay); -+ ({ int __udelay = (udelay); \ -+ INVAL_CACHE_AND_WAIT(map, chip, adr, 0, 0, &__udelay); }) - - - static int do_point_onechip (struct map_info *map, struct flchip *chip, loff_t adr, size_t len) -@@ -1356,7 +1357,7 @@ - - ret = INVAL_CACHE_AND_WAIT(map, chip, adr, - adr, map_bankwidth(map), -- chip->word_write_time); -+ &chip->word_write_time); - if (ret) { - xip_enable(map, chip, adr); - printk(KERN_ERR "%s: word write error (status timeout)\n", map->name); -@@ -1593,7 +1594,7 @@ - - ret = INVAL_CACHE_AND_WAIT(map, chip, cmd_adr, - adr, len, -- chip->buffer_write_time); -+ &chip->buffer_write_time); - if (ret) { - map_write(map, CMD(0x70), cmd_adr); - chip->state = FL_STATUS; -@@ -1728,7 +1729,7 @@ - - ret = INVAL_CACHE_AND_WAIT(map, chip, adr, - adr, len, -- chip->erase_time); -+ &chip->erase_time); - if (ret) { - map_write(map, CMD(0x70), adr); - chip->state = FL_STATUS; diff --git a/target/linux/generic-2.6/patches-2.6.22/010-disable_old_squashfs_compatibility.patch b/target/linux/generic-2.6/patches-2.6.22/010-disable_old_squashfs_compatibility.patch deleted file mode 100644 index 01e2757..0000000 --- a/target/linux/generic-2.6/patches-2.6.22/010-disable_old_squashfs_compatibility.patch +++ /dev/null @@ -1,19 +0,0 @@ ---- a/fs/squashfs/Makefile -+++ b/fs/squashfs/Makefile -@@ -4,4 +4,3 @@ - - obj-$(CONFIG_SQUASHFS) += squashfs.o - squashfs-y += inode.o --squashfs-y += squashfs2_0.o ---- a/fs/squashfs/squashfs.h -+++ b/fs/squashfs/squashfs.h -@@ -24,6 +24,9 @@ - #ifdef CONFIG_SQUASHFS_1_0_COMPATIBILITY - #undef CONFIG_SQUASHFS_1_0_COMPATIBILITY - #endif -+#ifdef CONFIG_SQUASHFS_2_0_COMPATIBILITY -+#undef CONFIG_SQUASHFS_2_0_COMPATIBILITY -+#endif - - #ifdef SQUASHFS_TRACE - #define TRACE(s, args...) printk(KERN_NOTICE "SQUASHFS: "s, ## args) diff --git a/target/linux/generic-2.6/patches-2.6.22/011-mips_boot.patch b/target/linux/generic-2.6/patches-2.6.22/011-mips_boot.patch deleted file mode 100644 index 6b1caca..0000000 --- a/target/linux/generic-2.6/patches-2.6.22/011-mips_boot.patch +++ /dev/null @@ -1,19 +0,0 @@ ---- a/arch/mips/kernel/head.S -+++ b/arch/mips/kernel/head.S -@@ -129,11 +129,15 @@ - #endif - .endm - -+ -+ j kernel_entry -+ nop -+ - /* - * Reserved space for exception handlers. - * Necessary for machines which link their kernels at KSEG0. - */ -- .fill 0x400 -+ .align 10 - - EXPORT(stext) # used for profiling - EXPORT(_stext) diff --git a/target/linux/generic-2.6/patches-2.6.22/012-mips_cpu_tlb.patch b/target/linux/generic-2.6/patches-2.6.22/012-mips_cpu_tlb.patch deleted file mode 100644 index 47316d2..0000000 --- a/target/linux/generic-2.6/patches-2.6.22/012-mips_cpu_tlb.patch +++ /dev/null @@ -1,18 +0,0 @@ ---- a/arch/mips/mm/tlbex.c -+++ b/arch/mips/mm/tlbex.c -@@ -887,7 +887,6 @@ - case CPU_R10000: - case CPU_R12000: - case CPU_R14000: -- case CPU_4KC: - case CPU_SB1: - case CPU_SB1A: - case CPU_4KSC: -@@ -915,6 +914,7 @@ - tlbw(p); - break; - -+ case CPU_4KC: - case CPU_4KEC: - case CPU_24K: - case CPU_34K: diff --git a/target/linux/generic-2.6/patches-2.6.22/013-mips_generic_gpio_support.patch b/target/linux/generic-2.6/patches-2.6.22/013-mips_generic_gpio_support.patch deleted file mode 100644 index 8f2632b..0000000 --- a/target/linux/generic-2.6/patches-2.6.22/013-mips_generic_gpio_support.patch +++ /dev/null @@ -1,32 +0,0 @@ ---- a/arch/mips/defconfig -+++ b/arch/mips/defconfig -@@ -69,6 +69,7 @@ - CONFIG_GENERIC_HWEIGHT=y - CONFIG_GENERIC_CALIBRATE_DELAY=y - CONFIG_GENERIC_TIME=y -+CONFIG_GENERIC_GPIO=n - CONFIG_SCHED_NO_NO_OMIT_FRAME_POINTER=y - # CONFIG_GENERIC_HARDIRQS_NO__DO_IRQ is not set - CONFIG_ARC=y ---- a/arch/mips/Kconfig -+++ b/arch/mips/Kconfig -@@ -704,6 +704,10 @@ - bool - default y - -+config GENERIC_GPIO -+ bool -+ default n -+ - config SCHED_NO_NO_OMIT_FRAME_POINTER - bool - default y ---- /dev/null -+++ b/include/asm-mips/gpio.h -@@ -0,0 +1,6 @@ -+#ifndef _ASM_MIPS_GPIO_H -+#define _ASM_MIPS_GPIO_H -+ -+#include <gpio.h> -+ -+#endif /* _ASM_MIPS_GPIO_H */ diff --git a/target/linux/generic-2.6/patches-2.6.22/014-x86_newsetup.patch b/target/linux/generic-2.6/patches-2.6.22/014-x86_newsetup.patch deleted file mode 100644 index 9d77c83..0000000 --- a/target/linux/generic-2.6/patches-2.6.22/014-x86_newsetup.patch +++ /dev/null @@ -1,11865 +0,0 @@ -GIT 1783e2f0f21444020e3dee1be46b1e34af0ea3e7 git+ssh://master.kernel.org/pub/scm/linux/kernel/git/hpa/linux-2.6-newsetup.git - -commit 1783e2f0f21444020e3dee1be46b1e34af0ea3e7 -Author: Venki Pallipadi <venkatesh.pallipadi@intel.com> -Date: Wed Jun 20 14:12:39 2007 -0700 - - Use a new CPU feature word to cover all Intel features that are spread around - - in different CPUID leafs like 0x5, 0x6 and 0xA. Make this - feature detection code common across i386 and x86_64. - - Display Intel Dynamic Acceleration feature in /proc/cpuinfo. This feature - will be enabled automatically by current acpi-cpufreq driver. - - Refer to Intel Software Developer's Manual for more details about the feature. - - Thanks to hpa (H Peter Anvin) for the making the actual code detecting the - scattered features data-driven. - - Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com> - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit cd19eb67cd6636a4e5c9df99631422c7c7286f59 -Author: H. Peter Anvin <hpa@zytor.com> -Date: Wed Jun 20 14:33:17 2007 -0700 - - x86 setup: move __bss_start into the .bss segment - - Move __bss_start into the .bss segment, and create __bss_end. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit 100327ad6b609cd28970219be57d293847d1261d -Author: H. Peter Anvin <hpa@zytor.com> -Date: Wed Jun 6 22:07:01 2007 -0700 - - x86 setup: remove TSC as a required feature - - Remove TSC as a required feature, in anticipation of CONFIG_X86_TSC - removal. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit 7c91a172b8af7d4ba087f1f88ed5b155ed459ca3 -Author: Antonino A. Daplas <adaplas@gmail.com> -Date: Tue Jun 5 19:21:05 2007 +0800 - - i386: Set 6-bit DAC channel properties in vesa video setup - - If the video BIOS is not capable of switching or failed to switch the - hardware to 8-bit DAC, the channel properties are not set. This leads - to a blank (all black) display with vesafb at 8 bpp. Fix by defaulting - to a 6-bit DAC. - - Signed-off-by: Antonino Daplas <adaplas@gmail.com> - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit 6eac2d442de8d87eac94a4ca8600bd87219fa06b -Author: H. Peter Anvin <hpa@zytor.com> -Date: Tue Jun 5 16:19:36 2007 -0700 - - x86 setup: arch/i386/boot/cpucheck.c whitespace cleanup - - Remove stealth whitespace - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit f7d89f05a30433034a1b4651143afdbb2a8a9c92 -Author: H. Peter Anvin <hpa@zytor.com> -Date: Thu May 24 16:56:44 2007 -0700 - - hd.c: remove BIOS/CMOS queries - - An ST-506 disk these days is pretty much someone trying to pull ancient - data using an auxilliary controller. Pulling data from the BIOS or CMOS - is just plain wrong, since it's likely to be the primary OS disk... and - would be user-entered data anyway. Instead, require the user enters it - on the command line. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit 14c2fdb3bbfd6a9a774980e446c2443150749891 -Author: H. Peter Anvin <hpa@zytor.com> -Date: Thu May 24 15:25:10 2007 -0700 - - x86: add back pbe bit to visible CPUID flags - - Add pbe back to the visible CPUID flags. We *do* correctly filter abuses - of this bit for 3DNow! in all the appropriate paths. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit e071b068a3b9f318be314f0378e655e2eb50ac89 -Author: H. Peter Anvin <hpa@zytor.com> -Date: Wed May 23 14:52:34 2007 -0700 - - x86 setup: VIA feature mask MSR doesn't just apply to model <= 9 - - The VIA feature mask MSR is known to be present on model 10, and it - seems likely it will continue to be supported. Since we only touch the - MSR if we're about to print an error message anyway, go ahead and be - aggressive. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit abe0c5aa1827932cda9c754a3842ec22b278d704 -Author: H. Peter Anvin <hpa@zytor.com> -Date: Tue May 22 17:17:41 2007 -0700 - - x86 setup: correct inline assembly constraints in edd.c - - Fix the inline assembly constraints in edd.c. In particular, "driveno" - was getting clobbered on some (buggy?) BIOSes. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit c2e5887ad275aab90673a3e33344f09946159cf7 -Author: H. Peter Anvin <hpa@zytor.com> -Date: Fri May 18 10:02:55 2007 -0700 - - x86 setup: force the assembler to generate a 2-byte jump in header - - The jump instruction in the header only has two bytes available, so - it *better* be a 2-byte jump! Unfortunately, the assembler will - always generate a 3/5-byte jump when the target is in a different - section. Deal with that by generating the jump instruction - explicitly from .byte's, just like we do elsewhere when we need a - specific binary representation of a certain instruction. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit ce82e3b93eba48b6852822a03efa73c74e165d4f -Author: H. Peter Anvin <hpa@zytor.com> -Date: Thu May 17 15:44:48 2007 -0700 - - x86 setup: move the symbol start_of_setup into the proper section. - - start_of_setup is the beginning of the executable code and should be - located in the appropriate section. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit e5f3a529457a5bfaf8f8783fb86013221279a81c -Author: H. Peter Anvin <hpa@zytor.com> -Date: Thu May 17 15:43:19 2007 -0700 - - x86 setup: add an ASSERT that the header ends up in the right place - - Just in case we have funnies involving the linker or people putting - inappropriate align statements, make the linker abort if the setup - header ends up in the wrong place. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit d9dbde725687ab99d1f529f49f14d1e280cc5cac -Author: Alexander van Heukelum <heukelum@mailshack.com> -Date: Thu May 17 20:54:25 2007 +0200 - - x86 new setup: use appropriate sections for code and data - - An intermediate elf file is generated for the 16-bit setup code. - The generated code can be viewed using objdump -m i8086 -d. As it - stands, it also tries to disassemble the bugger_off_msg, which - results in garbage. This introduces two new sections to separate - the code and the data part of the bootsector stub. It also moves - some code from the .header section (a data section) to .inittext. - - Signed-off-by: Alexander van Heukelum <heukelum@mailshack.com> - -commit 0d7558a81cf61e9fd2332a54897c5fd18df0d7f2 -Author: H. Peter Anvin <hpa@zytor.com> -Date: Wed May 16 22:03:16 2007 -0700 - - x86 setup: use -include code16gcc.h instead of explicit #include - - Use -include in the Makefile instead of #include to include code16gcc.h. - This really is more of a compiler switch than anything else, and is a lot - cleaner to do implicitly. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit 017ce54e8a4a9628a76d6b510c7309a7e4e111a8 -Author: H. Peter Anvin <hpa@zytor.com> -Date: Wed May 16 18:48:06 2007 -0700 - - x86 setup: enable features on Centaur (VIA) and Transmeta processors - - AMD are not the only ones who sometimes mask features which the kernel - may very well depend on. VIA and Transmeta do, too. Add code to enable - these features during checking. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit b794f5f9c5089709f3df38c6d91869fa38a9c1a4 -Author: H. Peter Anvin <hpa@zytor.com> -Date: Wed May 16 16:37:47 2007 -0700 - - x86 setup: in older versions of ld, ASSERT() is an expression - - Older versions of ld (pre-2.15 or so) need: - - . = ASSERT(foo, "msg"); - - instead of: - - ASSERT(foo, "msg") - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit 21c2b7c99c417d07015ee8e516a634ec3d98c4ee -Author: H. Peter Anvin <hpa@zytor.com> -Date: Wed May 16 10:52:41 2007 -0700 - - x86 setup: print a warning message if the bootloader gave us no heap. - - If the bootloader is so old it doesn't set the CAN_USE_HEAP flag, - a lot of functionality will by necessity be disabled, so print a - warning message. This means either a 2.00 protocol bootloader or - a buggy bootloader; the Qemu bootloader falls in this category. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit 52ca0431390d389a2a2246f02fe652ea84c1ddd8 -Author: H. Peter Anvin <hpa@zytor.com> -Date: Wed May 16 10:51:03 2007 -0700 - - x86 setup: rely on a compiled-in default for load high/load low - - When deciding if we should move the kernel from 0x10000 to 0x1000, as - is required for a zImage kernel, rely on a compiled-in default since - Qemu unconditionally zeroes the loadflags. This, of course, is a bug - in Qemu, but still... - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit 4db77a97793104a32e5fb83e62b943fa144b329d -Author: H. Peter Anvin <hpa@zytor.com> -Date: Wed May 16 08:45:37 2007 -0700 - - x86 setup: correct assembly constraints. - - Double use of "d" in an asm() constraints; most gcc versions correctly - detect and avoid using it, but some version of gcc runs itself into - a brick wall instead. Fix the one "d" which should have been "a". - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit 4fbccbc1457d6710d3a9ce55ad70ec6cb0b75fc5 -Author: H. Peter Anvin <hpa@zytor.com> -Date: Tue May 15 09:16:29 2007 -0700 - - x86 setup: include <asm/msr-index.h> not <asm/msr.h> - - <asm/msr.h> brings in the accessor functions, which may potentially - bring in all other kinds of kernel headers which are inappropriate for - the setup code. For the setup code, include <asm/msr-index.h> - instead, which only includes the numeric constants. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit 839cafa9c0020e7506722dd2a4fd82a71c2939cc -Author: H. Peter Anvin <hpa@zytor.com> -Date: Mon May 14 15:49:01 2007 -0700 - - x86 setup: protocol 2.0[01]: base for CL_OFFSET depends on setup_move_size - - Handle the use of boot protocol 2.00 and 2.01: the base segment for - CL_OFFSET depends on the value of setup_move_size. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit d60357ad68a694b03e9b952eadba5ac277c31df0 -Author: H. Peter Anvin <hpa@zytor.com> -Date: Sat May 12 12:18:53 2007 -0700 - - x86 setup: remove unused variable - - Remove unused variable - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit e21a2030b01081612259847321bcce13eae1e883 -Author: Sam Ravnborg <sam@ravnborg.org> -Date: Sat May 12 12:17:30 2007 -0700 - - x86 setup: share i386 Makefile with x86_64 - - The boot Makefile for i386 and x86_64 are equal - except for the CFLAGS setting. - Teach x86_64 to use the Makefile from i386 and - make CFLAGS setting arch dependent in i386 Makefile. - - Signed-off-by: Sam Ravnborg <sam@ravnborg.org> - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit 8618d92339d0d106045f98f34833d863c3235cdb -Author: H. Peter Anvin <hpa@zytor.com> -Date: Sat May 12 00:32:12 2007 -0700 - - x86 setup: video-bios.c missed the pointer to the set_mode method! - - We need the actual pointer to the set_mode method (oops!) - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit 85dfc374ea9aad33b9e0315f07a4b2722dc11e3e -Author: H. Peter Anvin <hpa@zytor.com> -Date: Sat May 12 00:14:43 2007 -0700 - - x86 setup: when setting unknown BIOS modes and failing, try to revert - - If we set an unknown BIOS mode and fail, then explicitly try to revert - to the original mode. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit f4f7949f126d2f152b09fa9367b1ec693f2ea818 -Author: H. Peter Anvin <hpa@zytor.com> -Date: Fri May 11 11:20:59 2007 -0700 - - x86 setup: fix typo "video_bios" should be "video-bios" - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit 51ba7113ea5b07189b7f8a0534d400a072535197 -Author: H. Peter Anvin <hpa@zytor.com> -Date: Fri May 11 11:09:55 2007 -0700 - - x86 setup: allow setting VESA modes "blind" - - Apparently, people really do set VESA modes "blind". As a result, make - the framework for settting blind modes more general, to remove some - special cases. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit 1b8f73d9b2bf7630a2914ddab606db16fddb509e -Author: H. Peter Anvin <hpa@zytor.com> -Date: Thu May 10 22:08:45 2007 -0700 - - x86_64: CONFIG_PHYSICAL_ALIGN should be 2 MB - - It's not actually used yet, but set CONFIG_PHYSICAL_ALIGN to 2 MB - as it should be, to prevent conflicts with other works in progress. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit b81f3c88923e4470cd0942d4596fafc0fb1cf4fd -Author: H. Peter Anvin <hpa@zytor.com> -Date: Thu May 10 19:11:32 2007 -0700 - - x86 setup: remove debugging statements - - Remove debugging statements in video.c that were not meant for - production. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit be58b6d7e9c14e482bce495e8343955999dea77f -Author: H. Peter Anvin <hpa@zytor.com> -Date: Thu May 10 18:49:40 2007 -0700 - - x86 setup: only restore the screen image when needed - - Only restore the screen image when needed. This is how the original - code behaves, so it's presumably the desired behaviour. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit 22f6bd8cc23b512af28e34ae7d40036982a0ac63 -Author: H. Peter Anvin <hpa@zytor.com> -Date: Thu May 10 18:44:08 2007 -0700 - - x86 setup: correct the definition of the GDT limit - - Like all other x86 segment limits, the GDT limit points to the last byte - that is *permitted* to access, so it needs to be sizeof(boot_gdt)-1. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit 7f73f1f4aa4c97745bffe07a3ebcf226a4965b00 -Author: H. Peter Anvin <hpa@zytor.com> -Date: Thu May 10 15:47:48 2007 -0700 - - x86 setup: Re-implement scanning for hidden video modes - - Re-implement scanning for hidden video modes. Every now and then, - apparently, you can find them hidden like easter eggs. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit 6770176714bc12ec92372311ac02c14f0d22776e -Author: H. Peter Anvin <hpa@zytor.com> -Date: Thu May 10 15:24:27 2007 -0700 - - x86 setup: whitespace cleanup - - Clean up stealth whitespace. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit ba0480a3537cf471b08bdb99dae6d0780cfb1972 -Author: H. Peter Anvin <hpa@zytor.com> -Date: Wed May 9 16:54:42 2007 -0700 - - x86: sync the CPU feature string arrays - - With <asm/cpufeature.h> unified, synchronize the CPU feature string - arrays. The whole kernel/cpu directory really needs to be unified. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit ecb53b84efddbad3d9aa49e95598550831324348 -Author: H. Peter Anvin <hpa@zytor.com> -Date: Tue May 8 22:06:04 2007 -0700 - - x86 setup: need to set orig_video_isVGA - - After detecting a VGA console, we need to set - boot_params.screen_info.orig_video_isVGA. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit dc97fc053faff17b984ec962686caea52bd27628 -Author: H. Peter Anvin <hpa@zytor.com> -Date: Tue May 8 20:51:17 2007 -0700 - - x86 setup: boot sector should use ljmp, not jmpl - - We have an "jmpl" instruction in the boot sector, which was meant - to be an "ljmp" instruction. It worked anyway because gas interpreted - a two-argument "jmpl" as an "ljmpl" instruction, however, use plain - "ljmp" (i.e. "ljmpw".) - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit 7907f05e9692557c53c9ac13647db5e5343c7c76 -Author: H. Peter Anvin <hpa@zytor.com> -Date: Tue May 8 20:27:10 2007 -0700 - - x86 setup: only make VESA graphics modes selectable if CONFIG_FB - - If we select a VESA graphics mode, we better have framebuffer support - or the user will have no console. Therefore, make these modes - non-selectable if CONFIG_FB is not set. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit 8e509f9ebc44f45544d231454e84f10bf78d5772 -Author: H. Peter Anvin <hpa@zytor.com> -Date: Tue May 8 20:24:11 2007 -0700 - - x86 setup: need to probe VESA EDID block 0 only - - The VESA EDID BIOS call takes the EDID block number in %dx, and may - corrupt it by spec. Pass it in properly. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit 9912b9aed7943773d1fadaa2e2e52f42af395048 -Author: H. Peter Anvin <hpa@zytor.com> -Date: Mon May 7 18:22:04 2007 -0700 - - x86 setup: add missing file "bitops.h" missing from previous checkins - - The file "bitops.h" was missing from previous checkins. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit 732eb3fac2d772980e6555b8c69902c8107c72aa -Author: H. Peter Anvin <hpa@zytor.com> -Date: Mon May 7 14:59:43 2007 -0700 - - x86 setup: add -fno-stack-protector; other Makefile fixes - - Add -fno-stack-protector for the gcc's that need that; - Use -ffreestanding consistently; - Use $(LINUXINCLUDE); - Handle linker scripts consistently with other Makefiles. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit 2d5e47f21202e156fe97aba0a88d158d5c157a33 -Author: H. Peter Anvin <hpa@zytor.com> -Date: Mon May 7 14:45:25 2007 -0700 - - x86 setup: swap cpu.c and cpucheck.c; rename functions - - Make cpucheck.c the reusable component; the generically-named cpu.c - gets to be the wrapper. Accordingly, rename functions to make it - less confusing. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit bf2a428a4e7c1ee3ab9acb23cfafb45e818887a1 -Author: H. Peter Anvin <hpa@zytor.com> -Date: Mon May 7 14:09:38 2007 -0700 - - x86 setup: remove code moved from cpucheck.c -> cpu.c - - Move all info about requirements into cpu.c. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit 9ea8429fabe5df6aed6393ac3a00d0b64445ba6a -Author: H. Peter Anvin <hpa@zytor.com> -Date: Mon May 7 09:42:51 2007 -0700 - - x86 setup: remove double nesting of a20_test() - - a20_test() was invoked as either a20_test() or a20_wait(), where the - latter was simply a loop around a loop. Make the count a parameter - instead; this is clearer and saves a couple of bytes. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit 9edc55718f57195c664ee3175514d652f651cfd2 -Author: H. Peter Anvin <hpa@zytor.com> -Date: Mon May 7 09:30:41 2007 -0700 - - x86 setup: compile with -fomit-frame-pointer - - Compiling with -fomit-frame-pointer reduces the size by about 2%. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit e1003433f2d491bf17c79437cd75268da220dab5 -Author: H. Peter Anvin <hpa@zytor.com> -Date: Mon May 7 09:30:04 2007 -0700 - - x86 setup: be more paranoid about the stack setup in header.S - - In particular, deal correctly with the stack pointer being zero on entry. - While we're at it, align the stack. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit 853499c3dc3fcbeb192a613ac241d150ebc7c5a0 -Author: H. Peter Anvin <hpa@zytor.com> -Date: Sun May 6 23:25:10 2007 -0700 - - x86 setup: Factor out the environment-independent part of the CPU check. - - Factor out the environment-independent part of the CPU check so it can - be invoked from other parts of the kernel as well. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit f235a61f6d6dff57883efad351d746540bcb8caf -Author: H. Peter Anvin <hpa@zytor.com> -Date: Sat May 5 22:16:54 2007 -0700 - - x86 setup: when watching the setup size, take the stack into account - - When watching the setup size, we have to take the stack into account. - In particular, the stack is used not only by the setup code itself, but - by BIOS interrupt handlers and system calls. Reserve a minimum of - 512 bytes. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit 0d0e10091be48f7e4c8888e9d5c2836c704994f5 -Author: H. Peter Anvin <hpa@zytor.com> -Date: Sat May 5 19:25:51 2007 -0700 - - x86 setup: actually check the end of the heap. - - Keep track of where the heap ends and actually watch out for it. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit 47aab0b8f4d012fad3c42b5b0754d3cb87961b37 -Author: H. Peter Anvin <hpa@zytor.com> -Date: Sat May 5 15:47:58 2007 -0700 - - x86 setup: coppyright rPath, Inc. - - This work was done on the dime of rPath, Inc.; they own the copyright. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit d22571534d7eabf9408f29d9da423e1c6e04445f -Author: H. Peter Anvin <hpa@zytor.com> -Date: Sat May 5 15:21:11 2007 -0700 - - x86 setup: implement screen contents save/restore - - The old setup code had screen contents save and restore, so implement - it for the new one as well. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit e5145601a752bd998e783d159c187d3017d45d6d -Author: H. Peter Anvin <hpa@zytor.com> -Date: Sat May 5 15:20:19 2007 -0700 - - x86 setup: whitespace cleanup - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit 045ecb52f91a74eecad93ffc8791eefe59cf7fd1 -Author: H. Peter Anvin <hpa@zytor.com> -Date: Sat May 5 14:22:39 2007 -0700 - - x86 setup: allow setting of VESA graphics modes; cleanups - - - Allow setting of VESA graphics modes (used by vesafb) - - Clean up the macros related to the heap - - #if 0 copy functions that aren't actually currently being used - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit 58c04ed7e2d7d5979e1917a74b49bdc0f3dde211 -Author: H. Peter Anvin <hpa@zytor.com> -Date: Sat May 5 12:06:14 2007 -0700 - - x86 setup: move all VESA-related code into video-vesa.c; add EDID - - - Move all VESA-related code into video-vesa.c - - Add VESA EDID query support - - Remove some totally obsolete definitions from video.h - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit 07bc3931175fb98256140275c03194426d441b74 -Author: H. Peter Anvin <hpa@zytor.com> -Date: Sat May 5 12:04:40 2007 -0700 - - x86-64: remove -traditional from AFLAGS - - In arch/x86_64/boot/compressed, remove -traditional from AFLAGS. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit a830f615eeef838d461cbf7bbbee8c1c84708ec8 -Author: H. Peter Anvin <hpa@zytor.com> -Date: Fri May 4 18:44:38 2007 -0700 - - x86 setup: share code between i386 and x86-64 - - Share the boot (setup) code and tools between i386 and x86-64. - The compression code is now running in 64-bit mode in order to support - relocation, so do *not* share that code. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit 3e159a323bdfa5d5a7be2c1f6be089ca22d598e0 -Author: H. Peter Anvin <hpa@zytor.com> -Date: Fri May 4 18:43:35 2007 -0700 - - x86-64: use 0x1b4 as the scratch area in boot_params, not 0x3c - - Use 0x1b4 as the scratch area in boot_params rather than 0x3c. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit 4cf4424e7a0f29f251b781f9b5e3655b0645cb7f -Author: H. Peter Anvin <hpa@zytor.com> -Date: Fri May 4 18:26:18 2007 -0700 - - Revert "x86-64: Make arch/x86-64/boot a symlink to arch/i386/boot" - - This reverts commit b2ad90f4969226fe8cf3edc5330711ed5fc20105. - - Restore arch/x86_64/boot as a separate directory hierarchy. - - Conflicts: - -commit 8ed1ae1d2f94410811b7cca4b1a426e37652457f -Author: H. Peter Anvin <hpa@zytor.com> -Date: Fri May 4 17:00:33 2007 -0700 - - x86-64: It appears MTRR isn't a required feature after all. - - MTRR was documented as a required feature, but appears to boot fine - without it (tested since Bochs doesn't have MTRR support.) - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit 7c616d098579fb790662cdc703f2a0f26ea1668c -Author: H. Peter Anvin <hpa@zytor.com> -Date: Fri May 4 16:22:57 2007 -0700 - - x86 setup: use 0x1e4 as scratch, instead of 0x3c - - The compressed relocation needs a 4-byte scratch area to obtain - its own address. - - 0x3c is at the end of the video area, which is quite constrained -- it - only has 6 bytes left (12 if we recycle the obsolete fields which invade - this space.) Define 0x1e4 as a scratch field, and use it. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit 5bc1019227e94576e4876d05ee920f59195bce90 -Author: H. Peter Anvin <hpa@zytor.com> -Date: Fri May 4 16:09:15 2007 -0700 - - x86 setup: boot_params.e820_map is just the map, not the count; adjust - - boot_params.e820_map is just a list of entries, whereas - "struct e820map" contains a count as well. Thus, don't use - "struct e820map" to describe struct boot_params. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit 0f96b52497f444be2d52d1184ca90be49f713ea3 -Author: H. Peter Anvin <hpa@zytor.com> -Date: Fri May 4 15:49:03 2007 -0700 - - x86 setup: E820MAX is a definitional constant; no need to use sizeof hacks - - Now when we're using the standard headers for the setup code, we can use - E820MAX instead of playing sizeof games. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit 3a23a428b20cbb31fd7ff5516a053b99afc447f7 -Author: H. Peter Anvin <hpa@zytor.com> -Date: Fri May 4 12:08:46 2007 -0700 - - x86: move the bootparam structure definition into include/ - - Move the bootparam structure definition into include/, and make other - things use it. Haven't cleaned up all the macros yet, though. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit e93ec58911995971aa059990f8a91a02b05f6c8f -Author: H. Peter Anvin <hpa@zytor.com> -Date: Fri May 4 12:07:50 2007 -0700 - - i386: change %lu to %u in arch/i386/kernel/e820.h - - It's an u32, print it with %u - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit 2f47f004f614e2744867c0df274c55d8af2a42d5 -Author: H. Peter Anvin <hpa@zytor.com> -Date: Fri May 4 12:06:04 2007 -0700 - - x86: fix differences between i386 and x86-64 <asm/e820.h> - - Fix minor differences between i386 and x86-64 <asm/e820.h> - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit 56ec52f14e948f430af941052adee98019a617b7 -Author: H. Peter Anvin <hpa@zytor.com> -Date: Fri May 4 11:45:17 2007 -0700 - - x86: fix the definition of struct screen_info - - Name the fields that aren't really struct screen_info, and declare - the structure packed (the "capabilities" field is misaligned.) - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit 1d4429eaa564b0085d9ee3aa2de57e87a093a14e -Author: H. Peter Anvin <hpa@zytor.com> -Date: Fri May 4 11:43:10 2007 -0700 - - x86-64: Make arch/x86-64/boot a symlink to arch/i386/boot - - Until such time that Kbuild allows for a cleaner solution, make - arch/x86-64/boot a symlink to arch/i386/boot. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit 6a85f1b5fd041ea99d8604782559ce0502a60cc0 -Author: H. Peter Anvin <hpa@zytor.com> -Date: Fri May 4 10:42:06 2007 -0700 - - x86-64: rearrange includes due to unifications and inclusion from setup - - Unification caused a circular dependency between <asm/alternative.h> - and <asm/cpufeature.h>; resolve this. - - Add #ifndef _SETUP in <asm/e820.h> so it can be included from the boot - code. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit f6bbdc254bdbd5f7cf7a40c4cd6f9844af90824a -Author: H. Peter Anvin <hpa@zytor.com> -Date: Fri May 4 10:40:26 2007 -0700 - - x86: Complete <asm/cpufeature.h> with the union of i386 and x86-64 - - Add a feature to <asm/cpufeature.h> which was previously present - in x86-64 but missing in i386. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit 1a0819281060489901732914f67869e0aa8f26fd -Author: H. Peter Anvin <hpa@zytor.com> -Date: Fri May 4 10:39:32 2007 -0700 - - x86: unify <asm/boot.h> - - Unify <asm/boot.h> between i386 and x86-64 - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit 8d9c54585f4623e0310f970fb5c6eda7ec1614df -Author: H. Peter Anvin <hpa@zytor.com> -Date: Fri May 4 10:38:35 2007 -0700 - - x86-64: verify_cpu.S: use new masks - - Use the <asm/required-features.h> masks. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit 6cf3308646bb7a3210f0f76bcb895b2dea76a93c -Author: H. Peter Anvin <hpa@zytor.com> -Date: Fri May 4 10:37:26 2007 -0700 - - x86-64: fix compilation errors due to required-features.h change - - Fix compilation errors induced by required-features.h change. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit 1324201a93ce380b46a3128826ecbd794e617e59 -Author: H. Peter Anvin <hpa@zytor.com> -Date: Fri May 4 10:35:37 2007 -0700 - - x86-64: <asm/segment.h>: add boot segment descriptors - - Add boot segment descriptors to <asm/segment.h> to match i386. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit a0b15a9e79ed0310813709cd0690d6838917fe82 -Author: H. Peter Anvin <hpa@zytor.com> -Date: Fri May 4 10:34:37 2007 -0700 - - x86-64: add CONFIG_PHYSICAL_ALIGN to match i386 - - Add CONFIG_PHYSICAL_ALIGN to match i386, even though we don't use it. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit 8f5d14d11a7318e257351ae477392c7f7e314602 -Author: H. Peter Anvin <hpa@zytor.com> -Date: Fri May 4 10:33:54 2007 -0700 - - x86 setup: cleanups for compatibility with x86-64 - - These changes are necessary to compile on x86-64. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit a32f68b6d4023c1c6b1e62e8561189516c571ab9 -Author: H. Peter Anvin <hpa@zytor.com> -Date: Fri May 4 08:40:07 2007 -0700 - - x86 setup: add missing linker script - - Add linker script for the setup code, apparently missing from previous - checkins. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit 4f34ca8e926b2d0bf3a7502b99f8dfced8cdba9d -Author: H. Peter Anvin <hpa@zytor.com> -Date: Thu May 3 17:42:29 2007 -0700 - - x86 setup: paranoia: clear the high half of %esp - - We're invoked in 16-bit mode from an unknown bootloader. Make sure - we explicitly zero the upper half of %esp to avoid nasty surprises. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit 19eb9b73cc1632a923003a002108b242af7a6080 -Author: H. Peter Anvin <hpa@zytor.com> -Date: Thu May 3 17:35:41 2007 -0700 - - x86 setup: bootlin is *so* dead... - - Bootlin was never able to load bzImage kernels, so who cares about it. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit 3b9fb73c65151ee043bc74c333d9e3c9b1872125 -Author: H. Peter Anvin <hpa@zytor.com> -Date: Thu May 3 10:56:40 2007 -0700 - - x86 setup: apparently $(src) is insufficient, needs $(srctree)/$(src) - - For some unfanthomable reason the location of the source tree that - corresponds to the current directory has to be written as - $(srctree)/$(src) apparently. There might be a good reason for it, - but shorthand would be appreciated, and $(src) really should be the - short form. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit a6d01d375a2269be1e3a6b31bcc4d426ad5a473d -Author: H. Peter Anvin <hpa@zytor.com> -Date: Thu May 3 10:51:45 2007 -0700 - - x86 setup: remove reference to obsolete cpureq.c - - cpureq.c has been removed; remove it from the Makefile too. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit cbe5b7585d800435080bcbf1b1fd242926982674 -Author: H. Peter Anvin <hpa@zytor.com> -Date: Thu May 3 10:33:12 2007 -0700 - - x86 setup: use the required masks from <asm/required-features.h> - - Use the now-uniform features from <asm/required-features.h>. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit 99ed30180ecc1bb4e93f6edda5f6bad1adf3e630 -Author: H. Peter Anvin <hpa@zytor.com> -Date: Thu May 3 10:31:12 2007 -0700 - - x86: make the handling of required features consistent - - Make the handling of required features consistent between i386 - and x86-64. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit 1120d70a2be8f2deb6bda64047da288d8f86dad3 -Author: H. Peter Anvin <hpa@zytor.com> -Date: Thu May 3 00:09:53 2007 -0700 - - x86: Kconfig.cpu: the minimum CPU model is always 3; WP_WORKS_OK = i486 - - The minimum CPU model number is always 3 (i386), and if we have - WP_WORKS_OK it means we need an i486. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit ebc308c204149b86984ae2216f5b9b2e63932028 -Author: H. Peter Anvin <hpa@zytor.com> -Date: Thu May 3 00:08:48 2007 -0700 - - x86 setup: use CONFIG_X86_MINIMUM_CPU_MODEL - - Use CONFIG_X86_MINIMUM_CPU_MODEL as defined in Kconfig.cpu. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit 8b50b640e015bf5d0f65502437da6fcab46c391b -Author: H. Peter Anvin <hpa@zytor.com> -Date: Wed May 2 23:45:42 2007 -0700 - - x86 setup: remove bogus "static" - - Remove invalid "static" declarations in cpu.c - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit 35d23b60dfb110da81c24bcbfcda089cfc4fd264 -Author: H. Peter Anvin <hpa@zytor.com> -Date: Wed May 2 23:37:50 2007 -0700 - - x86 setup: cpu detection cleanups - - - Use <asm/processor-flags.h> - - Make sure %cr0 isn't in a dangerous configuration before probing the FPU - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit a1150a03247b355d11a4bb696b8aae1f46612992 -Author: H. Peter Anvin <hpa@zytor.com> -Date: Wed May 2 23:36:55 2007 -0700 - - x86 setup: compile with -DSETUP - - Define SETUP to make it easier to share code with the rest of the kernel. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit 7eb52e8ad1bdf01886023d1a13b3313084cd7db6 -Author: H. Peter Anvin <hpa@zytor.com> -Date: Wed May 2 23:34:57 2007 -0700 - - x86 setup: remove unused verify_cpu.S - - verify_cpu.S is obsoleted by boot/cpu.c. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit e90317a027c30176968220d18eb18bd6a9d9cc74 -Author: H. Peter Anvin <hpa@zytor.com> -Date: Wed May 2 20:07:43 2007 -0700 - - x86 setup: files missing from previous checkin (cpu.c, cpureq.c) - - These files were missing from a previous checkin; CPU feature-checking - code and the list of CPU features to check for. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit 060f9b3db33c67b5344b2b4110bc823eb776e5cd -Author: H. Peter Anvin <hpa@zytor.com> -Date: Wed May 2 19:51:34 2007 -0700 - - x86 setup: whitespace cleanup - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit 9f997a5569ec8fceaa15c2e9cf28e728e2ce118d -Author: H. Peter Anvin <hpa@zytor.com> -Date: Wed May 2 19:07:14 2007 -0700 - - x86 setup: add CPU feature detect/abort on insufficient featurage - - The x86 setup is the right place to check features and abort if they - are not present, since we can still get a message to the user via the - firmware. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit de4e976376fddec340651ef40b16a45f6189619d -Author: H. Peter Anvin <hpa@zytor.com> -Date: Wed May 2 19:05:34 2007 -0700 - - x86 setup: whitespace cleanup - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit bcd2d2f8de5d4568b6628aa133fce1ac40ece526 -Author: H. Peter Anvin <hpa@zytor.com> -Date: Wed May 2 16:19:59 2007 -0700 - - x86 setup: tag functions noreturn; error message on A20 failure - - Tag appropriate functions noreturn. - If the A20 gate fails, output an error message and refuse to boot. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit 752aef90cbfc888084bf11fd83f8f72b6a668fc9 -Author: H. Peter Anvin <hpa@zytor.com> -Date: Wed May 2 15:45:08 2007 -0700 - - x86 setup: clobber registers in keyboard BIOS call - - Keyboard BIOS call to set repeat rate is known to clobber registers on - "many" BIOSes. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit dde94003e4759aab275732cf9f1834440cd381d0 -Author: H. Peter Anvin <hpa@zytor.com> -Date: Wed May 2 15:44:21 2007 -0700 - - x86 setup: implement APM BIOS probe - - APM BIOS probe ported from assembly - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit 9403917d79e3349184318704476fa080836bd52c -Author: H. Peter Anvin <hpa@zytor.com> -Date: Wed May 2 15:17:14 2007 -0700 - - x86 setup: remove references to obsolete probes - - Remove "Hello, World!" as well as references to probes which are no - longer used... - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit 712f65ffbd1d4b55b4c55d68b4dcd32406c28fb8 -Author: H. Peter Anvin <hpa@zytor.com> -Date: Wed May 2 12:17:15 2007 -0700 - - x86 setup: video.c: correct the handling of special mode numbers - - Special mode numbers with the high bit set need to be handled *before* - masking out the high bit. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit 9cf083204fe14cda3b09840eba8d131d2e48ccdf -Author: H. Peter Anvin <hpa@zytor.com> -Date: Wed May 2 11:44:16 2007 -0700 - - x86 setup: Modern ATI cards pass the probe but lacks the modes. - - It appears modern ATI cards pass the probe for ATI-ness but lack the - modes. Kill off the driver as being incorrect. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit 890cbe950589e30af17eac9da800efc76e35e01d -Author: H. Peter Anvin <hpa@zytor.com> -Date: Wed May 2 11:32:21 2007 -0700 - - x86 setup: a20.c: make empty_8042() return status - - Make functions which could reasonably return status do so. It may - be relevant in the future, and it's a lot better if the programmer - doesn't have to figure out where everything should hook in. - - Just on principle. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit 08a44dc655e0086d23fc3c70cb93eb51eaeec259 -Author: H. Peter Anvin <hpa@zytor.com> -Date: Wed May 2 11:31:03 2007 -0700 - - x86 setup: video.c: clean up unused stuff - - Clean up unused variables that we have no intent on using, as well - as other cruft. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit 57e69acff1f577de430cae1523fd49a5d113e885 -Author: H. Peter Anvin <hpa@zytor.com> -Date: Wed May 2 11:18:13 2007 -0700 - - x86 setup: drop video mode range checking - - Drop video mode range checking. If someone really has, say, 12x40 mode - visible through the BIOS then allow them to select it... odds are low - that it will actually conflict with the very sparse allocation we have - anyway. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit c0dda0b90f92d43872d55d295630a71cd357cfa6 -Author: H. Peter Anvin <hpa@zytor.com> -Date: Wed May 2 11:15:53 2007 -0700 - - x86 setup: if no specific video mode ID is given, generate one - - If we don't specify a certain video mode ID in the driver, then - generate the 0xRRCC mode ID automatically. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit 0db5086e79810e7c5d560006b1c9a7501a02d80c -Author: H. Peter Anvin <hpa@zytor.com> -Date: Wed May 2 11:10:28 2007 -0700 - - x86 setup: Sadly, Cirrus removed extended text modes from their BIOS. - - In the later era of the Cirrus 54xx series, Cirrus removed extended text - modes from their BIOS. Neither Qemu nor Bochs implement them in the BIOS. - If we can find a direct-register-poking method of setting them that - works in Bochs/Qemu it might be worthwhile to resurrect this; the probing - routine *does* work. - - Of course, the Right Thing[TM] would be to submit such a routine to the - Bochs/Qemu BIOS as a VESA text mode. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit 41f3fddeeb764687bf3fb0cf77fd858128571d58 -Author: H. Peter Anvin <hpa@zytor.com> -Date: Wed May 2 10:18:07 2007 -0700 - - x86 setup: remove assembly implementation of putchar and puts - - Already unused, remove assembly implementation of putchar and puts. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit dfa94cd86aca2c01d2f5e14b6e7c3e8258276195 -Author: H. Peter Anvin <hpa@zytor.com> -Date: Tue May 1 21:41:28 2007 -0700 - - x86 setup: Call INT 15h AX=E820h properly - - The calling convention for BIOS call 15:E820 was messed up. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit 2487575a36435c0a983febbb4f3751331bd2df7a -Author: H. Peter Anvin <hpa@zytor.com> -Date: Tue May 1 21:34:12 2007 -0700 - - x86 setup: advance one e820 descriptor at a time... - - Adding sizeof(foo) to a foo * is not just useless, it's pretty damaging... - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit 530d4f4f1732335ae8725c0b8c332a618e63ea1d -Author: H. Peter Anvin <hpa@zytor.com> -Date: Tue May 1 21:33:28 2007 -0700 - - x86 setup: fix memcmp_[fg]s() - - Actually return a value from memcmp_[fg]s()... - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit 8617cd56ff2e43303147da012b26c9dd46af726e -Author: H. Peter Anvin <hpa@zytor.com> -Date: Tue May 1 21:32:47 2007 -0700 - - x86 setup: fix missing semicolon in video-ati.c - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit 7bbf7fa3e199b9cef4877c5a56128faff8636cc9 -Author: H. Peter Anvin <hpa@zytor.com> -Date: Tue May 1 21:26:48 2007 -0700 - - x86 setup: make the video setup code actually do something... - - Basic video setup now works (there is still work to be done, however.) - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit 45bcd4406e4b812b32d317d9b3b8db2e5f135a3c -Author: H. Peter Anvin <hpa@zytor.com> -Date: Tue May 1 21:25:20 2007 -0700 - - x86 setup: segment descriptors need to be Present - - The segment descriptors were missing the Present bit. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit a39479d4ccf4dceffb623ad2ec7e2d708c38c637 -Author: H. Peter Anvin <hpa@zytor.com> -Date: Tue May 1 21:24:32 2007 -0700 - - build: setup sectors doesn't include the boot sector - - The "setup sectors" field doesn't include the old boot sector, - even though the two are now one module. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit d8f3d4928ead72e8febe2fcd740d0fee71a61f42 -Author: H. Peter Anvin <hpa@zytor.com> -Date: Tue May 1 21:23:44 2007 -0700 - - x86 setup: in tty.c, actually tell it what character to print - - putchar() was missing the actual passing of the character code to the - BIOS call, with very silly-looking results. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit 4f1462ed0377e180484a223e622d62432baa64b7 -Author: H. Peter Anvin <hpa@zytor.com> -Date: Tue May 1 21:22:46 2007 -0700 - - x86 setup: printf.c needs code16gcc.h - - printf.c was missing code16gcc.h, with predictable consequences. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit a5ba7e6df198bd204b0f87fc6e3f68388b9d14c1 -Author: H. Peter Anvin <hpa@zytor.com> -Date: Mon Apr 30 20:56:42 2007 -0700 - - MAINTAINERS: formally take responsibility for the i386 boot code - - Change MAINTAINERS to formally take responsibility for the i386 boot code. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> - -commit 6c821fc005655a99eff6e86c2e4b13654de94dea -Author: H. Peter Anvin <hpa@zytor.com> -Date: Mon Apr 30 20:54:07 2007 -0700 - - x86 setup code rewrite: initial development snapshot - - Clean up the setup code and rewrite it in C. - This is an initial development snapshot, not a working tree. - - Signed-off-by: H. Peter Anvin <hpa@zytor.com> -Signed-off-by: Andrew Morton <akpm@linux-foundation.org> ---- - - MAINTAINERS | 4 - arch/i386/Kconfig.cpu | 4 - arch/i386/boot/Makefile | 45 - arch/i386/boot/a20.c | 161 + - arch/i386/boot/apm.c | 97 - arch/i386/boot/bitops.h | 45 - arch/i386/boot/boot.h | 290 ++ - arch/i386/boot/bootsect.S | 98 - arch/i386/boot/cmdline.c | 97 - arch/i386/boot/code16gcc.h | 9 - arch/i386/boot/compressed/Makefile | 7 - arch/i386/boot/compressed/head.S | 6 - arch/i386/boot/compressed/misc.c | 3 - arch/i386/boot/copy.S | 101 - arch/i386/boot/cpu.c | 69 - arch/i386/boot/cpucheck.c | 266 ++ - arch/i386/boot/edd.S | 231 -- - arch/i386/boot/edd.c | 196 + - arch/i386/boot/header.S | 283 ++ - arch/i386/boot/main.c | 161 + - arch/i386/boot/mca.c | 43 - arch/i386/boot/memory.c | 99 - arch/i386/boot/pm.c | 170 + - arch/i386/boot/pmjump.S | 54 - arch/i386/boot/printf.c | 331 ++ - arch/i386/boot/setup.S | 1075 --------- - arch/i386/boot/setup.ld | 54 - arch/i386/boot/string.c | 34 - arch/i386/boot/tools/build.c | 156 - - arch/i386/boot/tty.c | 112 - arch/i386/boot/version.c | 23 - arch/i386/boot/vesa.h | 79 - arch/i386/boot/video-bios.c | 125 + - arch/i386/boot/video-vesa.c | 283 ++ - arch/i386/boot/video-vga.c | 260 ++ - arch/i386/boot/video.S | 2043 ------------------ - arch/i386/boot/video.c | 456 ++++ - arch/i386/boot/video.h | 145 + - arch/i386/boot/voyager.c | 46 - arch/i386/kernel/cpu/addon_cpuid_features.c | 50 - arch/i386/kernel/cpu/common.c | 2 - arch/i386/kernel/cpu/proc.c | 21 - arch/i386/kernel/e820.c | 2 - arch/i386/kernel/setup.c | 12 - arch/i386/kernel/verify_cpu.S | 94 - arch/x86_64/Kconfig | 4 - arch/x86_64/boot/Makefile | 136 - - arch/x86_64/boot/bootsect.S | 98 - arch/x86_64/boot/compressed/Makefile | 9 - arch/x86_64/boot/compressed/head.S | 6 - arch/x86_64/boot/install.sh | 2 - arch/x86_64/boot/mtools.conf.in | 17 - arch/x86_64/boot/setup.S | 826 ------- - arch/x86_64/boot/tools/build.c | 185 - - arch/x86_64/kernel/Makefile | 2 - arch/x86_64/kernel/setup.c | 21 - arch/x86_64/kernel/verify_cpu.S | 22 - drivers/ide/legacy/hd.c | 73 - include/asm-i386/boot.h | 6 - include/asm-i386/bootparam.h | 85 - include/asm-i386/cpufeature.h | 26 - include/asm-i386/e820.h | 14 - include/asm-i386/processor.h | 1 - include/asm-i386/required-features.h | 37 - include/asm-i386/setup.h | 10 - include/asm-x86_64/alternative.h | 68 - include/asm-x86_64/boot.h | 16 - include/asm-x86_64/bootparam.h | 1 - include/asm-x86_64/cpufeature.h | 115 - - include/asm-x86_64/e820.h | 6 - include/asm-x86_64/processor.h | 3 - include/asm-x86_64/required-features.h | 46 - include/asm-x86_64/segment.h | 8 - include/linux/edd.h | 4 - include/linux/screen_info.h | 9 - 75 files changed, 4594 insertions(+), 5204 deletions(-) - ---- a/MAINTAINERS -+++ b/MAINTAINERS -@@ -1750,8 +1750,8 @@ - S: Maintained - - i386 BOOT CODE --P: Riley H. Williams --M: Riley@Williams.Name -+P: H. Peter Anvin -+M: hpa@zytor.com - L: Linux-Kernel@vger.kernel.org - S: Maintained - ---- a/arch/i386/Kconfig.cpu -+++ b/arch/i386/Kconfig.cpu -@@ -346,6 +346,6 @@ - - config X86_MINIMUM_CPU_MODEL - int -- default "4" if X86_XADD || X86_CMPXCHG || X86_BSWAP -- default "0" -+ default "4" if X86_XADD || X86_CMPXCHG || X86_BSWAP || X86_WP_WORKS_OK -+ default "3" - ---- a/arch/i386/boot/Makefile -+++ b/arch/i386/boot/Makefile -@@ -25,27 +25,53 @@ - - #RAMDISK := -DRAMDISK=512 - --targets := vmlinux.bin bootsect bootsect.o \ -- setup setup.o zImage bzImage -+targets := vmlinux.bin setup.bin setup.elf zImage bzImage - subdir- := compressed - -+setup-y += a20.o apm.o cmdline.o copy.o cpu.o cpucheck.o edd.o -+setup-y += header.o main.o mca.o memory.o pm.o pmjump.o -+setup-y += printf.o string.o tty.o video.o version.o voyager.o -+ -+# The link order of the video-*.o modules can matter. In particular, -+# video-vga.o *must* be listed first, followed by video-vesa.o. -+# Hardware-specific drivers should follow in the order they should be -+# probed, and video-bios.o should typically be last. -+setup-y += video-vga.o -+setup-y += video-vesa.o -+setup-y += video-bios.o -+ - hostprogs-y := tools/build - - HOSTCFLAGS_build.o := $(LINUXINCLUDE) - - # --------------------------------------------------------------------------- - -+# How to compile the 16-bit code. Note we always compile for -march=i386, -+# that way we can complain to the user if the CPU is insufficient. -+cflags-i386 := -+cflags-x86_64 := -m32 -+CFLAGS := $(LINUXINCLUDE) -g -Os -D_SETUP -D__KERNEL__ \ -+ $(cflags-$(ARCH)) \ -+ -Wall -Wstrict-prototypes \ -+ -march=i386 -mregparm=3 \ -+ -include $(srctree)/$(src)/code16gcc.h \ -+ -fno-strict-aliasing -fomit-frame-pointer \ -+ $(call cc-option, -ffreestanding) \ -+ $(call cc-option, -fno-stack-protector) -+AFLAGS := $(CFLAGS) -D__ASSEMBLY__ -+ - $(obj)/zImage: IMAGE_OFFSET := 0x1000 - $(obj)/zImage: EXTRA_AFLAGS := $(SVGA_MODE) $(RAMDISK) - $(obj)/bzImage: IMAGE_OFFSET := 0x100000 -+$(obj)/bzImage: EXTRA_CFLAGS := -D__BIG_KERNEL__ - $(obj)/bzImage: EXTRA_AFLAGS := $(SVGA_MODE) $(RAMDISK) -D__BIG_KERNEL__ - $(obj)/bzImage: BUILDFLAGS := -b - - quiet_cmd_image = BUILD $@ --cmd_image = $(obj)/tools/build $(BUILDFLAGS) $(obj)/bootsect $(obj)/setup \ -+cmd_image = $(obj)/tools/build $(BUILDFLAGS) $(obj)/setup.bin \ - $(obj)/vmlinux.bin $(ROOT_DEV) > $@ - --$(obj)/zImage $(obj)/bzImage: $(obj)/bootsect $(obj)/setup \ -+$(obj)/zImage $(obj)/bzImage: $(obj)/setup.bin \ - $(obj)/vmlinux.bin $(obj)/tools/build FORCE - $(call if_changed,image) - @echo 'Kernel: $@ is ready' ' (#'`cat .version`')' -@@ -53,12 +79,17 @@ - $(obj)/vmlinux.bin: $(obj)/compressed/vmlinux FORCE - $(call if_changed,objcopy) - --LDFLAGS_bootsect := -Ttext 0x0 -s --oformat binary --LDFLAGS_setup := -Ttext 0x0 -s --oformat binary -e begtext -+SETUP_OBJS = $(addprefix $(obj)/,$(setup-y)) - --$(obj)/setup $(obj)/bootsect: %: %.o FORCE -+LDFLAGS_setup.elf := -T -+$(obj)/setup.elf: $(src)/setup.ld $(SETUP_OBJS) FORCE - $(call if_changed,ld) - -+OBJCOPYFLAGS_setup.bin := -O binary -+ -+$(obj)/setup.bin: $(obj)/setup.elf FORCE -+ $(call if_changed,objcopy) -+ - $(obj)/compressed/vmlinux: FORCE - $(Q)$(MAKE) $(build)=$(obj)/compressed IMAGE_OFFSET=$(IMAGE_OFFSET) $@ - ---- /dev/null -+++ b/arch/i386/boot/a20.c -@@ -0,0 +1,161 @@ -+/* -*- linux-c -*- ------------------------------------------------------- * -+ * -+ * Copyright (C) 1991, 1992 Linus Torvalds -+ * Copyright 2007 rPath, Inc. - All Rights Reserved -+ * -+ * This file is part of the Linux kernel, and is made available under -+ * the terms of the GNU General Public License version 2. -+ * -+ * ----------------------------------------------------------------------- */ -+ -+/* -+ * arch/i386/boot/a20.c -+ * -+ * Enable A20 gate (return -1 on failure) -+ */ -+ -+#include "boot.h" -+ -+#define MAX_8042_LOOPS 100000 -+ -+static int empty_8042(void) -+{ -+ u8 status; -+ int loops = MAX_8042_LOOPS; -+ -+ while (loops--) { -+ io_delay(); -+ -+ status = inb(0x64); -+ if (status & 1) { -+ /* Read and discard input data */ -+ io_delay(); -+ (void)inb(0x60); -+ } else if (!(status & 2)) { -+ /* Buffers empty, finished! */ -+ return 0; -+ } -+ } -+ -+ return -1; -+} -+ -+/* Returns nonzero if the A20 line is enabled. The memory address -+ used as a test is the int $0x80 vector, which should be safe. */ -+ -+#define A20_TEST_ADDR (4*0x80) -+#define A20_TEST_SHORT 32 -+#define A20_TEST_LONG 2097152 /* 2^21 */ -+ -+static int a20_test(int loops) -+{ -+ int ok = 0; -+ int saved, ctr; -+ -+ set_fs(0x0000); -+ set_gs(0xffff); -+ -+ saved = ctr = rdfs32(A20_TEST_ADDR); -+ -+ while (loops--) { -+ wrfs32(++ctr, A20_TEST_ADDR); -+ io_delay(); /* Serialize and make delay constant */ -+ ok = rdgs32(A20_TEST_ADDR+0x10) ^ ctr; -+ if (ok) -+ break; -+ } -+ -+ wrfs32(saved, A20_TEST_ADDR); -+ return ok; -+} -+ -+/* Quick test to see if A20 is already enabled */ -+static int a20_test_short(void) -+{ -+ return a20_test(A20_TEST_SHORT); -+} -+ -+/* Longer test that actually waits for A20 to come on line; this -+ is useful when dealing with the KBC or other slow external circuitry. */ -+static int a20_test_long(void) -+{ -+ return a20_test(A20_TEST_LONG); -+} -+ -+static void enable_a20_bios(void) -+{ -+ asm volatile("pushfl; int $0x15; popfl" -+ : : "a" ((u16)0x2401)); -+} -+ -+static void enable_a20_kbc(void) -+{ -+ empty_8042(); -+ -+ outb(0xd1, 0x64); /* Command write */ -+ empty_8042(); -+ -+ outb(0xdf, 0x60); /* A20 on */ -+ empty_8042(); -+} -+ -+static void enable_a20_fast(void) -+{ -+ u8 port_a; -+ -+ port_a = inb(0x92); /* Configuration port A */ -+ port_a |= 0x02; /* Enable A20 */ -+ port_a &= ~0x01; /* Do not reset machine */ -+ outb(port_a, 0x92); -+} -+ -+/* -+ * Actual routine to enable A20; return 0 on ok, -1 on failure -+ */ -+ -+#define A20_ENABLE_LOOPS 255 /* Number of times to try */ -+ -+int enable_a20(void) -+{ -+ int loops = A20_ENABLE_LOOPS; -+ -+#if defined(CONFIG_X86_ELAN) -+ /* Elan croaks if we try to touch the KBC */ -+ enable_a20_fast(); -+ while (!a20_test_long()) -+ ; -+ return 0; -+#elif defined(CONFIG_X86_VOYAGER) -+ /* On Voyager, a20_test() is unsafe? */ -+ enable_a20_kbc(); -+ return 0; -+#else -+ while (loops--) { -+ /* First, check to see if A20 is already enabled -+ (legacy free, etc.) */ -+ if (a20_test_short()) -+ return 0; -+ -+ /* Next, try the BIOS (INT 0x15, AX=0x2401) */ -+ enable_a20_bios(); -+ if (a20_test_short()) -+ return 0; -+ -+ /* Try enabling A20 through the keyboard controller */ -+ empty_8042(); -+ if (a20_test_short()) -+ return 0; /* BIOS worked, but with delayed reaction */ -+ -+ enable_a20_kbc(); -+ if (a20_test_long()) -+ return 0; -+ -+ /* Finally, try enabling the "fast A20 gate" */ -+ enable_a20_fast(); -+ if (a20_test_long()) -+ return 0; -+ } -+ -+ return -1; -+#endif -+} ---- /dev/null -+++ b/arch/i386/boot/apm.c -@@ -0,0 +1,97 @@ -+/* -*- linux-c -*- ------------------------------------------------------- * -+ * -+ * Copyright (C) 1991, 1992 Linus Torvalds -+ * Copyright 2007 rPath, Inc. - All Rights Reserved -+ * -+ * Original APM BIOS checking by Stephen Rothwell, May 1994 -+ * (sfr@canb.auug.org.au) -+ * -+ * This file is part of the Linux kernel, and is made available under -+ * the terms of the GNU General Public License version 2. -+ * -+ * ----------------------------------------------------------------------- */ -+ -+/* -+ * arch/i386/boot/apm.c -+ * -+ * Get APM BIOS information -+ */ -+ -+#include "boot.h" -+ -+#if defined(CONFIG_APM) || defined(CONFIG_APM_MODULE) -+ -+int query_apm_bios(void) -+{ -+ u16 ax, bx, cx, dx, di; -+ u32 ebx, esi; -+ u8 err; -+ -+ /* APM BIOS installation check */ -+ ax = 0x5300; -+ bx = cx = 0; -+ asm volatile("pushl %%ebp ; int $0x15 ; popl %%ebp ; setc %0" -+ : "=d" (err), "+a" (ax), "+b" (bx), "+c" (cx) -+ : : "esi", "edi"); -+ -+ if (err) -+ return -1; /* No APM BIOS */ -+ -+ if (bx != 0x504d) /* "PM" signature */ -+ return -1; -+ -+ if (cx & 0x02) /* 32 bits supported? */ -+ return -1; -+ -+ /* Disconnect first, just in case */ -+ ax = 0x5304; -+ asm volatile("pushl %%ebp ; int $0x15 ; popl %%ebp" -+ : "+a" (ax) -+ : : "ebx", "ecx", "edx", "esi", "edi"); -+ -+ /* Paranoia */ -+ ebx = esi = 0; -+ cx = dx = di = 0; -+ -+ /* 32-bit connect */ -+ asm volatile("pushl %%ebp ; int $0x15 ; popl %%ebp ; setc %6" -+ : "=a" (ax), "+b" (ebx), "+c" (cx), "+d" (dx), -+ "+S" (esi), "+D" (di), "=m" (err) -+ : "a" (0x5303)); -+ -+ boot_params.apm_bios_info.cseg = ax; -+ boot_params.apm_bios_info.offset = ebx; -+ boot_params.apm_bios_info.cseg_16 = cx; -+ boot_params.apm_bios_info.dseg = dx; -+ boot_params.apm_bios_info.cseg_len = (u16)esi; -+ boot_params.apm_bios_info.cseg_16_len = esi >> 16; -+ boot_params.apm_bios_info.dseg_len = di; -+ -+ if (err) -+ return -1; -+ -+ /* Redo the installation check as the 32-bit connect; -+ some BIOSes return different flags this way... */ -+ -+ ax = 0x5300; -+ bx = cx = 0; -+ asm volatile("pushl %%ebp ; int $0x15 ; popl %%ebp ; setc %0" -+ : "=d" (err), "+a" (ax), "+b" (bx), "+c" (cx) -+ : : "esi", "edi"); -+ -+ if (err || bx != 0x504d) { -+ /* Failure with 32-bit connect, try to disconect and ignore */ -+ ax = 0x5304; -+ bx = 0; -+ asm volatile("pushl %%ebp ; int $0x15 ; popl %%ebp" -+ : "+a" (ax), "+b" (bx) -+ : : "ecx", "edx", "esi", "edi"); -+ return -1; -+ } -+ -+ boot_params.apm_bios_info.version = ax; -+ boot_params.apm_bios_info.flags = cx; -+ return 0; -+} -+ -+#endif ---- /dev/null -+++ b/arch/i386/boot/bitops.h -@@ -0,0 +1,45 @@ -+/* -*- linux-c -*- ------------------------------------------------------- * -+ * -+ * Copyright (C) 1991, 1992 Linus Torvalds -+ * Copyright 2007 rPath, Inc. - All Rights Reserved -+ * -+ * This file is part of the Linux kernel, and is made available under -+ * the terms of the GNU General Public License version 2. -+ * -+ * ----------------------------------------------------------------------- */ -+ -+/* -+ * arch/i386/boot/bitops.h -+ * -+ * Very simple bitops for the boot code. -+ */ -+ -+#ifndef BOOT_BITOPS_H -+#define BOOT_BITOPS_H -+#define _LINUX_BITOPS_H /* Inhibit inclusion of <linux/bitops.h> */ -+ -+static inline int constant_test_bit(int nr, const void *addr) -+{ -+ const u32 *p = (const u32 *)addr; -+ return ((1UL << (nr & 31)) & (p[nr >> 5])) != 0; -+} -+static inline int variable_test_bit(int nr, const void *addr) -+{ -+ u8 v; -+ const u32 *p = (const u32 *)addr; -+ -+ asm("btl %2,%1; setc %0" : "=qm" (v) : "m" (*p), "Ir" (nr)); -+ return v; -+} -+ -+#define test_bit(nr,addr) \ -+(__builtin_constant_p(nr) ? \ -+ constant_test_bit((nr),(addr)) : \ -+ variable_test_bit((nr),(addr))) -+ -+static inline void set_bit(int nr, void *addr) -+{ -+ asm("btsl %1,%0" : "+m" (*(u32 *)addr) : "Ir" (nr)); -+} -+ -+#endif /* BOOT_BITOPS_H */ ---- /dev/null -+++ b/arch/i386/boot/boot.h -@@ -0,0 +1,290 @@ -+/* -*- linux-c -*- ------------------------------------------------------- * -+ * -+ * Copyright (C) 1991, 1992 Linus Torvalds -+ * Copyright 2007 rPath, Inc. - All Rights Reserved -+ * -+ * This file is part of the Linux kernel, and is made available under -+ * the terms of the GNU General Public License version 2. -+ * -+ * ----------------------------------------------------------------------- */ -+ -+/* -+ * arch/i386/boot/boot.h -+ * -+ * Header file for the real-mode kernel code -+ */ -+ -+#ifndef BOOT_BOOT_H -+#define BOOT_BOOT_H -+ -+#ifndef __ASSEMBLY__ -+ -+#include <stdarg.h> -+#include <linux/types.h> -+#include <linux/edd.h> -+#include <asm/boot.h> -+#include <asm/bootparam.h> -+ -+/* Useful macros */ -+#define BUILD_BUG_ON(condition) ((void)sizeof(char[1 - 2*!!(condition)])) -+ -+extern struct setup_header hdr; -+extern struct boot_params boot_params; -+ -+/* Basic port I/O */ -+static inline void outb(u8 v, u16 port) -+{ -+ asm volatile("outb %0,%1" : : "a" (v), "dN" (port)); -+} -+static inline u8 inb(u16 port) -+{ -+ u8 v; -+ asm volatile("inb %1,%0" : "=a" (v) : "dN" (port)); -+ return v; -+} -+ -+static inline void outw(u16 v, u16 port) -+{ -+ asm volatile("outw %0,%1" : : "a" (v), "dN" (port)); -+} -+static inline u16 inw(u16 port) -+{ -+ u16 v; -+ asm volatile("inw %1,%0" : "=a" (v) : "dN" (port)); -+ return v; -+} -+ -+static inline void outl(u32 v, u16 port) -+{ -+ asm volatile("outl %0,%1" : : "a" (v), "dn" (port)); -+} -+static inline u32 inl(u32 port) -+{ -+ u32 v; -+ asm volatile("inl %1,%0" : "=a" (v) : "dN" (port)); -+ return v; -+} -+ -+static inline void io_delay(void) -+{ -+ const u16 DELAY_PORT = 0x80; -+ asm volatile("outb %%al,%0" : : "dN" (DELAY_PORT)); -+} -+ -+/* These functions are used to reference data in other segments. */ -+ -+static inline u16 ds(void) -+{ -+ u16 seg; -+ asm("movw %%ds,%0" : "=rm" (seg)); -+ return seg; -+} -+ -+static inline void set_fs(u16 seg) -+{ -+ asm volatile("movw %0,%%fs" : : "rm" (seg)); -+} -+static inline u16 fs(void) -+{ -+ u16 seg; -+ asm("movw %%fs,%0" : "=rm" (seg)); -+ return seg; -+} -+ -+static inline void set_gs(u16 seg) -+{ -+ asm volatile("movw %0,%%gs" : : "rm" (seg)); -+} -+static inline u16 gs(void) -+{ -+ u16 seg; -+ asm("movw %%gs,%0" : "=rm" (seg)); -+ return seg; -+} -+ -+typedef unsigned int addr_t; -+ -+static inline u8 rdfs8(addr_t addr) -+{ -+ u8 v; -+ asm("movb %%fs:%1,%0" : "=r" (v) : "m" (*(u8 *)addr)); -+ return v; -+} -+static inline u16 rdfs16(addr_t addr) -+{ -+ u16 v; -+ asm("movw %%fs:%1,%0" : "=r" (v) : "m" (*(u16 *)addr)); -+ return v; -+} -+static inline u32 rdfs32(addr_t addr) -+{ -+ u32 v; -+ asm("movl %%fs:%1,%0" : "=r" (v) : "m" (*(u32 *)addr)); -+ return v; -+} -+ -+static inline void wrfs8(u8 v, addr_t addr) -+{ -+ asm volatile("movb %1,%%fs:%0" : "+m" (*(u8 *)addr) : "r" (v)); -+} -+static inline void wrfs16(u16 v, addr_t addr) -+{ -+ asm volatile("movw %1,%%fs:%0" : "+m" (*(u16 *)addr) : "r" (v)); -+} -+static inline void wrfs32(u32 v, addr_t addr) -+{ -+ asm volatile("movl %1,%%fs:%0" : "+m" (*(u32 *)addr) : "r" (v)); -+} -+ -+static inline u8 rdgs8(addr_t addr) -+{ -+ u8 v; -+ asm("movb %%gs:%1,%0" : "=r" (v) : "m" (*(u8 *)addr)); -+ return v; -+} -+static inline u16 rdgs16(addr_t addr) -+{ -+ u16 v; -+ asm("movw %%gs:%1,%0" : "=r" (v) : "m" (*(u16 *)addr)); -+ return v; -+} -+static inline u32 rdgs32(addr_t addr) -+{ -+ u32 v; -+ asm("movl %%gs:%1,%0" : "=r" (v) : "m" (*(u32 *)addr)); -+ return v; -+} -+ -+static inline void wrgs8(u8 v, addr_t addr) -+{ -+ asm volatile("movb %1,%%gs:%0" : "+m" (*(u8 *)addr) : "r" (v)); -+} -+static inline void wrgs16(u16 v, addr_t addr) -+{ -+ asm volatile("movw %1,%%gs:%0" : "+m" (*(u16 *)addr) : "r" (v)); -+} -+static inline void wrgs32(u32 v, addr_t addr) -+{ -+ asm volatile("movl %1,%%gs:%0" : "+m" (*(u32 *)addr) : "r" (v)); -+} -+ -+/* Note: these only return true/false, not a signed return value! */ -+static inline int memcmp(const void *s1, const void *s2, size_t len) -+{ -+ u8 diff; -+ asm("repe; cmpsb; setnz %0" -+ : "=qm" (diff), "+D" (s1), "+S" (s2), "+c" (len)); -+ return diff; -+} -+ -+static inline int memcmp_fs(const void *s1, addr_t s2, size_t len) -+{ -+ u8 diff; -+ asm("fs; repe; cmpsb; setnz %0" -+ : "=qm" (diff), "+D" (s1), "+S" (s2), "+c" (len)); -+ return diff; -+} -+static inline int memcmp_gs(const void *s1, addr_t s2, size_t len) -+{ -+ u8 diff; -+ asm("gs; repe; cmpsb; setnz %0" -+ : "=qm" (diff), "+D" (s1), "+S" (s2), "+c" (len)); -+ return diff; -+} -+ -+/* Heap -- available for dynamic lists. */ -+#define STACK_SIZE 512 /* Minimum number of bytes for stack */ -+ -+extern char _end[]; -+extern char *HEAP; -+extern char *heap_end; -+#define RESET_HEAP() ((void *)( HEAP = _end )) -+static inline char *__get_heap(size_t s, size_t a, size_t n) -+{ -+ char *tmp; -+ -+ HEAP = (char *)(((size_t)HEAP+(a-1)) & ~(a-1)); -+ tmp = HEAP; -+ HEAP += s*n; -+ return tmp; -+} -+#define GET_HEAP(type, n) \ -+ ((type *)__get_heap(sizeof(type),__alignof__(type),(n))) -+ -+static inline int heap_free(void) -+{ -+ return heap_end-HEAP; -+} -+ -+/* copy.S */ -+ -+void copy_to_fs(addr_t dst, void *src, size_t len); -+void *copy_from_fs(void *dst, addr_t src, size_t len); -+void copy_to_gs(addr_t dst, void *src, size_t len); -+void *copy_from_gs(void *dst, addr_t src, size_t len); -+void *memcpy(void *dst, void *src, size_t len); -+void *memset(void *dst, int c, size_t len); -+ -+#define memcpy(d,s,l) __builtin_memcpy(d,s,l) -+#define memset(d,c,l) __builtin_memset(d,c,l) -+ -+/* a20.c */ -+int enable_a20(void); -+ -+/* apm.c */ -+int query_apm_bios(void); -+ -+/* cmdline.c */ -+int cmdline_find_option(const char *option, char *buffer, int bufsize); -+ -+/* cpu.c, cpucheck.c */ -+int check_cpu(int *cpu_level_ptr, int *req_level_ptr, u32 **err_flags_ptr); -+int validate_cpu(void); -+ -+/* edd.c */ -+void query_edd(void); -+ -+/* header.S */ -+void __attribute__((noreturn)) die(void); -+ -+/* mca.c */ -+int query_mca(void); -+ -+/* memory.c */ -+int detect_memory(void); -+ -+/* pm.c */ -+void __attribute__((noreturn)) go_to_protected_mode(void); -+ -+/* pmjump.S */ -+void __attribute__((noreturn)) -+ protected_mode_jump(u32 entrypoint, u32 bootparams); -+ -+/* printf.c */ -+unsigned int atou(const char *s); -+int sprintf(char *buf, const char *fmt, ...); -+int vsprintf(char *buf, const char *fmt, va_list args); -+int printf(const char *fmt, ...); -+ -+/* string.c */ -+int strcmp(const char *str1, const char *str2); -+ -+/* tty.c */ -+void puts(const char *); -+void putchar(int); -+int getchar(void); -+void kbd_flush(void); -+int getchar_timeout(void); -+ -+/* video.c */ -+void set_video(void); -+ -+/* video-vesa.c */ -+void vesa_store_edid(void); -+ -+/* voyager.c */ -+int query_voyager(void); -+ -+#endif /* __ASSEMBLY__ */ -+ -+#endif /* BOOT_BOOT_H */ ---- a/arch/i386/boot/bootsect.S -+++ /dev/null -@@ -1,98 +0,0 @@ --/* -- * bootsect.S Copyright (C) 1991, 1992 Linus Torvalds -- * -- * modified by Drew Eckhardt -- * modified by Bruce Evans (bde) -- * modified by Chris Noe (May 1999) (as86 -> gas) -- * gutted by H. Peter Anvin (Jan 2003) -- * -- * BIG FAT NOTE: We're in real mode using 64k segments. Therefore segment -- * addresses must be multiplied by 16 to obtain their respective linear -- * addresses. To avoid confusion, linear addresses are written using leading -- * hex while segment addresses are written as segment:offset. -- * -- */ -- --#include <asm/boot.h> -- --SETUPSECTS = 4 /* default nr of setup-sectors */ --BOOTSEG = 0x07C0 /* original address of boot-sector */ --INITSEG = DEF_INITSEG /* we move boot here - out of the way */ --SETUPSEG = DEF_SETUPSEG /* setup starts here */ --SYSSEG = DEF_SYSSEG /* system loaded at 0x10000 (65536) */ --SYSSIZE = DEF_SYSSIZE /* system size: # of 16-byte clicks */ -- /* to be loaded */ --ROOT_DEV = 0 /* ROOT_DEV is now written by "build" */ --SWAP_DEV = 0 /* SWAP_DEV is now written by "build" */ -- --#ifndef SVGA_MODE --#define SVGA_MODE ASK_VGA --#endif -- --#ifndef RAMDISK --#define RAMDISK 0 --#endif -- --#ifndef ROOT_RDONLY --#define ROOT_RDONLY 1 --#endif -- --.code16 --.text -- --.global _start --_start: -- -- # Normalize the start address -- jmpl $BOOTSEG, $start2 -- --start2: -- movw %cs, %ax -- movw %ax, %ds -- movw %ax, %es -- movw %ax, %ss -- movw $0x7c00, %sp -- sti -- cld -- -- movw $bugger_off_msg, %si -- --msg_loop: -- lodsb -- andb %al, %al -- jz die -- movb $0xe, %ah -- movw $7, %bx -- int $0x10 -- jmp msg_loop -- --die: -- # Allow the user to press a key, then reboot -- xorw %ax, %ax -- int $0x16 -- int $0x19 -- -- # int 0x19 should never return. In case it does anyway, -- # invoke the BIOS reset code... -- ljmp $0xf000,$0xfff0 -- -- --bugger_off_msg: -- .ascii "Direct booting from floppy is no longer supported.\r\n" -- .ascii "Please use a boot loader program instead.\r\n" -- .ascii "\n" -- .ascii "Remove disk and press any key to reboot . . .\r\n" -- .byte 0 -- -- -- # Kernel attributes; used by setup -- -- .org 497 --setup_sects: .byte SETUPSECTS --root_flags: .word ROOT_RDONLY --syssize: .word SYSSIZE --swap_dev: .word SWAP_DEV --ram_size: .word RAMDISK --vid_mode: .word SVGA_MODE --root_dev: .word ROOT_DEV --boot_flag: .word 0xAA55 ---- /dev/null -+++ b/arch/i386/boot/cmdline.c -@@ -0,0 +1,97 @@ -+/* -*- linux-c -*- ------------------------------------------------------- * -+ * -+ * Copyright (C) 1991, 1992 Linus Torvalds -+ * Copyright 2007 rPath, Inc. - All Rights Reserved -+ * -+ * This file is part of the Linux kernel, and is made available under -+ * the terms of the GNU General Public License version 2. -+ * -+ * ----------------------------------------------------------------------- */ -+ -+/* -+ * arch/i386/boot/cmdline.c -+ * -+ * Simple command-line parser for early boot. -+ */ -+ -+#include "boot.h" -+ -+static inline int myisspace(u8 c) -+{ -+ return c <= ' '; /* Close enough approximation */ -+} -+ -+/* -+ * Find a non-boolean option, that is, "option=argument". In accordance -+ * with standard Linux practice, if this option is repeated, this returns -+ * the last instance on the command line. -+ * -+ * Returns the length of the argument (regardless of if it was -+ * truncated to fit in the buffer), or -1 on not found. -+ */ -+int cmdline_find_option(const char *option, char *buffer, int bufsize) -+{ -+ u32 cmdline_ptr = boot_params.hdr.cmd_line_ptr; -+ addr_t cptr; -+ char c; -+ int len = -1; -+ const char *opptr = NULL; -+ char *bufptr = buffer; -+ enum { -+ st_wordstart, /* Start of word/after whitespace */ -+ st_wordcmp, /* Comparing this word */ -+ st_wordskip, /* Miscompare, skip */ -+ st_bufcpy /* Copying this to buffer */ -+ } state = st_wordstart; -+ -+ if (!cmdline_ptr || cmdline_ptr >= 0x100000) -+ return -1; /* No command line, or inaccessible */ -+ -+ cptr = cmdline_ptr & 0xf; -+ set_fs(cmdline_ptr >> 4); -+ -+ while (cptr < 0x10000 && (c = rdfs8(cptr++))) { -+ switch (state) { -+ case st_wordstart: -+ if (myisspace(c)) -+ break; -+ -+ /* else */ -+ state = st_wordcmp; -+ opptr = option; -+ /* fall through */ -+ -+ case st_wordcmp: -+ if (c == '=' && !*opptr) { -+ len = 0; -+ bufptr = buffer; -+ state = st_bufcpy; -+ } else if (myisspace(c)) { -+ state = st_wordstart; -+ } else if (c != *opptr++) { -+ state = st_wordskip; -+ } -+ break; -+ -+ case st_wordskip: -+ if (myisspace(c)) -+ state = st_wordstart; -+ break; -+ -+ case st_bufcpy: -+ if (myisspace(c)) { -+ state = st_wordstart; -+ } else { -+ if (len < bufsize-1) -+ *bufptr++ = c; -+ len++; -+ } -+ break; -+ } -+ } -+ -+ if (bufsize) -+ *bufptr = '\0'; -+ -+ return len; -+} ---- /dev/null -+++ b/arch/i386/boot/code16gcc.h -@@ -0,0 +1,9 @@ -+/* -+ * code16gcc.h -+ * -+ * This file is -include'd when compiling 16-bit C code. -+ */ -+ -+#ifndef __ASSEMBLY__ -+asm(".code16gcc"); -+#endif ---- a/arch/i386/boot/compressed/Makefile -+++ b/arch/i386/boot/compressed/Makefile -@@ -9,9 +9,14 @@ - EXTRA_AFLAGS := -traditional - - LDFLAGS_vmlinux := -T --CFLAGS_misc.o += -fPIC - hostprogs-y := relocs - -+CFLAGS := -m32 -D__KERNEL__ $(LINUX_INCLUDE) -O2 \ -+ -fno-strict-aliasing -fPIC \ -+ $(call cc-option,-ffreestanding) \ -+ $(call cc-option,-fno-stack-protector) -+LDFLAGS := -m elf_i386 -+ - $(obj)/vmlinux: $(src)/vmlinux.lds $(obj)/head.o $(obj)/misc.o $(obj)/piggy.o FORCE - $(call if_changed,ld) - @: ---- a/arch/i386/boot/compressed/head.S -+++ b/arch/i386/boot/compressed/head.S -@@ -45,10 +45,10 @@ - * at and where we were actually loaded at. This can only be done - * with a short local call on x86. Nothing else will tell us what - * address we are running at. The reserved chunk of the real-mode -- * data at 0x34-0x3f are used as the stack for this calculation. -- * Only 4 bytes are needed. -+ * data at 0x1e4 (defined as a scratch field) are used as the stack -+ * for this calculation. Only 4 bytes are needed. - */ -- leal 0x40(%esi), %esp -+ leal (0x1e4+4)(%esi), %esp - call 1f - 1: popl %ebp - subl $1b, %ebp ---- a/arch/i386/boot/compressed/misc.c -+++ b/arch/i386/boot/compressed/misc.c -@@ -11,7 +11,6 @@ - - #undef CONFIG_PARAVIRT - #include <linux/linkage.h> --#include <linux/vmalloc.h> - #include <linux/screen_info.h> - #include <asm/io.h> - #include <asm/page.h> -@@ -364,8 +363,10 @@ - - if ((u32)output & (CONFIG_PHYSICAL_ALIGN -1)) - error("Destination address not CONFIG_PHYSICAL_ALIGN aligned"); -+#ifndef CONFIG_X86_64 - if (end > ((-__PAGE_OFFSET-(512 <<20)-1) & 0x7fffffff)) - error("Destination address too large"); -+#endif - #ifndef CONFIG_RELOCATABLE - if ((u32)output != LOAD_PHYSICAL_ADDR) - error("Wrong destination address"); ---- /dev/null -+++ b/arch/i386/boot/copy.S -@@ -0,0 +1,101 @@ -+/* ----------------------------------------------------------------------- * -+ * -+ * Copyright (C) 1991, 1992 Linus Torvalds -+ * Copyright 2007 rPath, Inc. - All Rights Reserved -+ * -+ * This file is part of the Linux kernel, and is made available under -+ * the terms of the GNU General Public License version 2. -+ * -+ * ----------------------------------------------------------------------- */ -+ -+/* -+ * arch/i386/boot/copy.S -+ * -+ * Memory copy routines -+ */ -+ -+ .code16gcc -+ .text -+ -+ .globl memcpy -+ .type memcpy, @function -+memcpy: -+ pushw %si -+ pushw %di -+ movw %ax, %di -+ movw %dx, %si -+ pushw %cx -+ shrw $2, %cx -+ rep; movsl -+ popw %cx -+ andw $3, %cx -+ rep; movsb -+ popw %di -+ popw %si -+ ret -+ .size memcpy, .-memcpy -+ -+ .globl memset -+ .type memset, @function -+memset: -+ pushw %di -+ movw %ax, %di -+ movzbl %dl, %eax -+ imull $0x01010101,%eax -+ pushw %cx -+ shrw $2, %cx -+ rep; stosl -+ popw %cx -+ andw $3, %cx -+ rep; stosb -+ popw %di -+ ret -+ .size memset, .-memset -+ -+ .globl copy_from_fs -+ .type copy_from_fs, @function -+copy_from_fs: -+ pushw %ds -+ pushw %fs -+ popw %ds -+ call memcpy -+ popw %ds -+ ret -+ .size copy_from_fs, .-copy_from_fs -+ -+ .globl copy_to_fs -+ .type copy_to_fs, @function -+copy_to_fs: -+ pushw %es -+ pushw %fs -+ popw %es -+ call memcpy -+ popw %es -+ ret -+ .size copy_to_fs, .-copy_to_fs -+ -+#if 0 /* Not currently used, but can be enabled as needed */ -+ -+ .globl copy_from_gs -+ .type copy_from_gs, @function -+copy_from_gs: -+ pushw %ds -+ pushw %gs -+ popw %ds -+ call memcpy -+ popw %ds -+ ret -+ .size copy_from_gs, .-copy_from_gs -+ .globl copy_to_gs -+ -+ .type copy_to_gs, @function -+copy_to_gs: -+ pushw %es -+ pushw %gs -+ popw %es -+ call memcpy -+ popw %es -+ ret -+ .size copy_to_gs, .-copy_to_gs -+ -+#endif ---- /dev/null -+++ b/arch/i386/boot/cpu.c -@@ -0,0 +1,69 @@ -+/* -*- linux-c -*- ------------------------------------------------------- * -+ * -+ * Copyright (C) 1991, 1992 Linus Torvalds -+ * Copyright 2007 rPath, Inc. - All Rights Reserved -+ * -+ * This file is part of the Linux kernel, and is made available under -+ * the terms of the GNU General Public License version 2. -+ * -+ * ----------------------------------------------------------------------- */ -+ -+/* -+ * arch/i386/boot/cpucheck.c -+ * -+ * Check for obligatory CPU features and abort if the features are not -+ * present. -+ */ -+ -+#include "boot.h" -+#include "bitops.h" -+#include <asm/cpufeature.h> -+ -+static char *cpu_name(int level) -+{ -+ static char buf[6]; -+ -+ if (level == 64) { -+ return "x86-64"; -+ } else { -+ sprintf(buf, "i%d86", level); -+ return buf; -+ } -+} -+ -+int validate_cpu(void) -+{ -+ u32 *err_flags; -+ int cpu_level, req_level; -+ -+ check_cpu(&cpu_level, &req_level, &err_flags); -+ -+ if (cpu_level < req_level) { -+ printf("This kernel requires an %s CPU, ", -+ cpu_name(req_level)); -+ printf("but only detected an %s CPU.\n", -+ cpu_name(cpu_level)); -+ return -1; -+ } -+ -+ if (err_flags) { -+ int i, j; -+ puts("This kernel requires the following features " -+ "not present on the CPU:\n"); -+ -+ for (i = 0; i < NCAPINTS; i++) { -+ u32 e = err_flags[i]; -+ -+ for (j = 0; j < 32; j++) { -+ if (e & 1) -+ printf("%d:%d ", i, j); -+ -+ e >>= 1; -+ } -+ } -+ putchar('\n'); -+ return -1; -+ } else { -+ return 0; -+ } -+} ---- /dev/null -+++ b/arch/i386/boot/cpucheck.c -@@ -0,0 +1,266 @@ -+/* -*- linux-c -*- ------------------------------------------------------- * -+ * -+ * Copyright (C) 1991, 1992 Linus Torvalds -+ * Copyright 2007 rPath, Inc. - All Rights Reserved -+ * -+ * This file is part of the Linux kernel, and is made available under -+ * the terms of the GNU General Public License version 2. -+ * -+ * ----------------------------------------------------------------------- */ -+ -+/* -+ * arch/i386/boot/cpu.c -+ * -+ * Check for obligatory CPU features and abort if the features are not -+ * present. This code should be compilable as 16-, 32- or 64-bit -+ * code, so be very careful with types and inline assembly. -+ * -+ * This code should not contain any messages; that requires an -+ * additional wrapper. -+ * -+ * As written, this code is not safe for inclusion into the kernel -+ * proper (after FPU initialization, in particular). -+ */ -+ -+#ifdef _SETUP -+# include "boot.h" -+# include "bitops.h" -+#endif -+#include <linux/types.h> -+#include <asm/cpufeature.h> -+#include <asm/processor-flags.h> -+#include <asm/required-features.h> -+#include <asm/msr-index.h> -+ -+struct cpu_features { -+ int level; -+ int model; -+ u32 flags[NCAPINTS]; -+}; -+ -+static struct cpu_features cpu; -+static u32 cpu_vendor[3]; -+static u32 err_flags[NCAPINTS]; -+ -+#ifdef CONFIG_X86_64 -+static const int req_level = 64; -+#elif defined(CONFIG_X86_MINIMUM_CPU_MODEL) -+static const int req_level = CONFIG_X86_MINIMUM_CPU_MODEL; -+#else -+static const int req_level = 3; -+#endif -+ -+static const u32 req_flags[NCAPINTS] = -+{ -+ REQUIRED_MASK0, -+ REQUIRED_MASK1, -+ REQUIRED_MASK2, -+ REQUIRED_MASK3, -+ REQUIRED_MASK4, -+ REQUIRED_MASK5, -+ REQUIRED_MASK6, -+}; -+ -+#define A32(a,b,c,d) (((d) << 24)+((c) << 16)+((b) << 8)+(a)) -+ -+static int is_amd(void) -+{ -+ return cpu_vendor[0] == A32('A','u','t','h') && -+ cpu_vendor[1] == A32('e','n','t','i') && -+ cpu_vendor[2] == A32('c','A','M','D'); -+} -+ -+static int is_centaur(void) -+{ -+ return cpu_vendor[0] == A32('C','e','n','t') && -+ cpu_vendor[1] == A32('a','u','r','H') && -+ cpu_vendor[2] == A32('a','u','l','s'); -+} -+ -+static int is_transmeta(void) -+{ -+ return cpu_vendor[0] == A32('G','e','n','u') && -+ cpu_vendor[1] == A32('i','n','e','T') && -+ cpu_vendor[2] == A32('M','x','8','6'); -+} -+ -+static int has_fpu(void) -+{ -+ u16 fcw = -1, fsw = -1; -+ u32 cr0; -+ -+ asm("movl %%cr0,%0" : "=r" (cr0)); -+ if (cr0 & (X86_CR0_EM|X86_CR0_TS)) { -+ cr0 &= ~(X86_CR0_EM|X86_CR0_TS); -+ asm volatile("movl %0,%%cr0" : : "r" (cr0)); -+ } -+ -+ asm("fninit ; fnstsw %0 ; fnstcw %1" : "+m" (fsw), "+m" (fcw)); -+ -+ return fsw == 0 && (fcw & 0x103f) == 0x003f; -+} -+ -+static int has_eflag(u32 mask) -+{ -+ u32 f0, f1; -+ -+ asm("pushfl ; " -+ "pushfl ; " -+ "popl %0 ; " -+ "movl %0,%1 ; " -+ "xorl %2,%1 ; " -+ "pushl %1 ; " -+ "popfl ; " -+ "pushfl ; " -+ "popl %1 ; " -+ "popfl" -+ : "=r" (f0), "=r" (f1) -+ : "g" (mask)); -+ -+ return !!((f0^f1) & mask); -+} -+ -+static void get_flags(void) -+{ -+ u32 max_intel_level, max_amd_level; -+ u32 tfms; -+ -+ if (has_fpu()) -+ set_bit(X86_FEATURE_FPU, cpu.flags); -+ -+ if (has_eflag(X86_EFLAGS_ID)) { -+ asm("cpuid" -+ : "=a" (max_intel_level), -+ "=b" (cpu_vendor[0]), -+ "=d" (cpu_vendor[1]), -+ "=c" (cpu_vendor[2]) -+ : "a" (0)); -+ -+ if (max_intel_level >= 0x00000001 && -+ max_intel_level <= 0x0000ffff) { -+ asm("cpuid" -+ : "=a" (tfms), -+ "=c" (cpu.flags[4]), -+ "=d" (cpu.flags[0]) -+ : "a" (0x00000001) -+ : "ebx"); -+ cpu.level = (tfms >> 8) & 15; -+ cpu.model = (tfms >> 4) & 15; -+ if (cpu.level >= 6) -+ cpu.model += ((tfms >> 16) & 0xf) << 4; -+ } -+ -+ asm("cpuid" -+ : "=a" (max_amd_level) -+ : "a" (0x80000000) -+ : "ebx", "ecx", "edx"); -+ -+ if (max_amd_level >= 0x80000001 && -+ max_amd_level <= 0x8000ffff) { -+ u32 eax = 0x80000001; -+ asm("cpuid" -+ : "+a" (eax), -+ "=c" (cpu.flags[6]), -+ "=d" (cpu.flags[1]) -+ : : "ebx"); -+ } -+ } -+} -+ -+/* Returns a bitmask of which words we have error bits in */ -+static int check_flags(void) -+{ -+ u32 err; -+ int i; -+ -+ err = 0; -+ for (i = 0; i < NCAPINTS; i++) { -+ err_flags[i] = req_flags[i] & ~cpu.flags[i]; -+ if (err_flags[i]) -+ err |= 1 << i; -+ } -+ -+ return err; -+} -+ -+/* -+ * Returns -1 on error. -+ * -+ * *cpu_level is set to the current CPU level; *req_level to the required -+ * level. x86-64 is considered level 64 for this purpose. -+ * -+ * *err_flags_ptr is set to the flags error array if there are flags missing. -+ */ -+int check_cpu(int *cpu_level_ptr, int *req_level_ptr, u32 **err_flags_ptr) -+{ -+ int err; -+ -+ memset(&cpu.flags, 0, sizeof cpu.flags); -+ cpu.level = 3; -+ -+ if (has_eflag(X86_EFLAGS_AC)) -+ cpu.level = 4; -+ -+ get_flags(); -+ err = check_flags(); -+ -+ if (test_bit(X86_FEATURE_LM, cpu.flags)) -+ cpu.level = 64; -+ -+ if (err == 0x01 && -+ !(err_flags[0] & -+ ~((1 << X86_FEATURE_XMM)|(1 << X86_FEATURE_XMM2))) && -+ is_amd()) { -+ /* If this is an AMD and we're only missing SSE+SSE2, try to -+ turn them on */ -+ -+ u32 ecx = MSR_K7_HWCR; -+ u32 eax, edx; -+ -+ asm("rdmsr" : "=a" (eax), "=d" (edx) : "c" (ecx)); -+ eax &= ~(1 << 15); -+ asm("wrmsr" : : "a" (eax), "d" (edx), "c" (ecx)); -+ -+ get_flags(); /* Make sure it really did something */ -+ err = check_flags(); -+ } else if (err == 0x01 && -+ !(err_flags[0] & ~(1 << X86_FEATURE_CX8)) && -+ is_centaur() && cpu.model >= 6) { -+ /* If this is a VIA C3, we might have to enable CX8 -+ explicitly */ -+ -+ u32 ecx = MSR_VIA_FCR; -+ u32 eax, edx; -+ -+ asm("rdmsr" : "=a" (eax), "=d" (edx) : "c" (ecx)); -+ eax |= (1<<1)|(1<<7); -+ asm("wrmsr" : : "a" (eax), "d" (edx), "c" (ecx)); -+ -+ set_bit(X86_FEATURE_CX8, cpu.flags); -+ err = check_flags(); -+ } else if (err == 0x01 && is_transmeta()) { -+ /* Transmeta might have masked feature bits in word 0 */ -+ -+ u32 ecx = 0x80860004; -+ u32 eax, edx; -+ u32 level = 1; -+ -+ asm("rdmsr" : "=a" (eax), "=d" (edx) : "c" (ecx)); -+ asm("wrmsr" : : "a" (~0), "d" (edx), "c" (ecx)); -+ asm("cpuid" -+ : "+a" (level), "=d" (cpu.flags[0]) -+ : : "ecx", "ebx"); -+ asm("wrmsr" : : "a" (eax), "d" (edx), "c" (ecx)); -+ -+ err = check_flags(); -+ } -+ -+ if (err_flags_ptr) -+ *err_flags_ptr = err ? err_flags : NULL; -+ if (cpu_level_ptr) -+ *cpu_level_ptr = cpu.level; -+ if (req_level_ptr) -+ *req_level_ptr = req_level; -+ -+ return (cpu.level < req_level || err) ? -1 : 0; -+} ---- a/arch/i386/boot/edd.S -+++ /dev/null -@@ -1,231 +0,0 @@ --/* -- * BIOS Enhanced Disk Drive support -- * Copyright (C) 2002, 2003, 2004 Dell, Inc. -- * by Matt Domsch <Matt_Domsch@dell.com> October 2002 -- * conformant to T13 Committee www.t13.org -- * projects 1572D, 1484D, 1386D, 1226DT -- * disk signature read by Matt Domsch <Matt_Domsch@dell.com> -- * and Andrew Wilks <Andrew_Wilks@dell.com> September 2003, June 2004 -- * legacy CHS retrieval by Patrick J. LoPresti <patl@users.sourceforge.net> -- * March 2004 -- * Command line option parsing, Matt Domsch, November 2004 -- */ -- --#include <linux/edd.h> --#include <asm/setup.h> -- --#if defined(CONFIG_EDD) || defined(CONFIG_EDD_MODULE) -- --# It is assumed that %ds == INITSEG here -- -- movb $0, (EDD_MBR_SIG_NR_BUF) -- movb $0, (EDDNR) -- --# Check the command line for options: --# edd=of disables EDD completely (edd=off) --# edd=sk skips the MBR test (edd=skipmbr) --# edd=on re-enables EDD (edd=on) -- -- pushl %esi -- movw $edd_mbr_sig_start, %di # Default to edd=on -- -- movl %cs:(cmd_line_ptr), %esi -- andl %esi, %esi -- jz old_cl # Old boot protocol? -- --# Convert to a real-mode pointer in fs:si -- movl %esi, %eax -- shrl $4, %eax -- movw %ax, %fs -- andw $0xf, %si -- jmp have_cl_pointer -- --# Old-style boot protocol? --old_cl: -- push %ds # aka INITSEG -- pop %fs -- -- cmpw $0xa33f, (0x20) -- jne done_cl # No command line at all? -- movw (0x22), %si # Pointer relative to INITSEG -- --# fs:si has the pointer to the command line now --have_cl_pointer: -- --# Loop through kernel command line one byte at a time. Just in --# case the loader is buggy and failed to null-terminate the command line --# terminate if we get close enough to the end of the segment that we --# cannot fit "edd=XX"... --cl_atspace: -- cmpw $-5, %si # Watch for segment wraparound -- jae done_cl -- movl %fs:(%si), %eax -- andb %al, %al # End of line? -- jz done_cl -- cmpl $EDD_CL_EQUALS, %eax -- jz found_edd_equals -- cmpb $0x20, %al # <= space consider whitespace -- ja cl_skipword -- incw %si -- jmp cl_atspace -- --cl_skipword: -- cmpw $-5, %si # Watch for segment wraparound -- jae done_cl -- movb %fs:(%si), %al # End of string? -- andb %al, %al -- jz done_cl -- cmpb $0x20, %al -- jbe cl_atspace -- incw %si -- jmp cl_skipword -- --found_edd_equals: --# only looking at first two characters after equals --# late overrides early on the command line, so keep going after finding something -- movw %fs:4(%si), %ax -- cmpw $EDD_CL_OFF, %ax # edd=of -- je do_edd_off -- cmpw $EDD_CL_SKIP, %ax # edd=sk -- je do_edd_skipmbr -- cmpw $EDD_CL_ON, %ax # edd=on -- je do_edd_on -- jmp cl_skipword --do_edd_skipmbr: -- movw $edd_start, %di -- jmp cl_skipword --do_edd_off: -- movw $edd_done, %di -- jmp cl_skipword --do_edd_on: -- movw $edd_mbr_sig_start, %di -- jmp cl_skipword -- --done_cl: -- popl %esi -- jmpw *%di -- --# Read the first sector of each BIOS disk device and store the 4-byte signature --edd_mbr_sig_start: -- movb $0x80, %dl # from device 80 -- movw $EDD_MBR_SIG_BUF, %bx # store buffer ptr in bx --edd_mbr_sig_read: -- movl $0xFFFFFFFF, %eax -- movl %eax, (%bx) # assume failure -- pushw %bx -- movb $READ_SECTORS, %ah -- movb $1, %al # read 1 sector -- movb $0, %dh # at head 0 -- movw $1, %cx # cylinder 0, sector 0 -- pushw %es -- pushw %ds -- popw %es -- movw $EDDBUF, %bx # disk's data goes into EDDBUF -- pushw %dx # work around buggy BIOSes -- stc # work around buggy BIOSes -- int $0x13 -- sti # work around buggy BIOSes -- popw %dx -- popw %es -- popw %bx -- jc edd_mbr_sig_done # on failure, we're done. -- cmpb $0, %ah # some BIOSes do not set CF -- jne edd_mbr_sig_done # on failure, we're done. -- movl (EDDBUF+EDD_MBR_SIG_OFFSET), %eax # read sig out of the MBR -- movl %eax, (%bx) # store success -- incb (EDD_MBR_SIG_NR_BUF) # note that we stored something -- incb %dl # increment to next device -- addw $4, %bx # increment sig buffer ptr -- cmpb $EDD_MBR_SIG_MAX, (EDD_MBR_SIG_NR_BUF) # Out of space? -- jb edd_mbr_sig_read # keep looping --edd_mbr_sig_done: -- --# Do the BIOS Enhanced Disk Drive calls --# This consists of two calls: --# int 13h ah=41h "Check Extensions Present" --# int 13h ah=48h "Get Device Parameters" --# int 13h ah=08h "Legacy Get Device Parameters" --# --# A buffer of size EDDMAXNR*(EDDEXTSIZE+EDDPARMSIZE) is reserved for our use --# in the boot_params at EDDBUF. The first four bytes of which are --# used to store the device number, interface support map and version --# results from fn41. The next four bytes are used to store the legacy --# cylinders, heads, and sectors from fn08. The following 74 bytes are used to --# store the results from fn48. Starting from device 80h, fn41, then fn48 --# are called and their results stored in EDDBUF+n*(EDDEXTSIZE+EDDPARMIZE). --# Then the pointer is incremented to store the data for the next call. --# This repeats until either a device doesn't exist, or until EDDMAXNR --# devices have been stored. --# The one tricky part is that ds:si always points EDDEXTSIZE bytes into --# the structure, and the fn41 and fn08 results are stored at offsets --# from there. This removes the need to increment the pointer for --# every store, and leaves it ready for the fn48 call. --# A second one-byte buffer, EDDNR, in the boot_params stores --# the number of BIOS devices which exist, up to EDDMAXNR. --# In setup.c, copy_edd() stores both boot_params buffers away --# for later use, as they would get overwritten otherwise. --# This code is sensitive to the size of the structs in edd.h --edd_start: -- # %ds points to the bootsector -- # result buffer for fn48 -- movw $EDDBUF+EDDEXTSIZE, %si # in ds:si, fn41 results -- # kept just before that -- movb $0x80, %dl # BIOS device 0x80 -- --edd_check_ext: -- movb $CHECKEXTENSIONSPRESENT, %ah # Function 41 -- movw $EDDMAGIC1, %bx # magic -- int $0x13 # make the call -- jc edd_done # no more BIOS devices -- -- cmpw $EDDMAGIC2, %bx # is magic right? -- jne edd_next # nope, next... -- -- movb %dl, %ds:-8(%si) # store device number -- movb %ah, %ds:-7(%si) # store version -- movw %cx, %ds:-6(%si) # store extensions -- incb (EDDNR) # note that we stored something -- --edd_get_device_params: -- movw $EDDPARMSIZE, %ds:(%si) # put size -- movw $0x0, %ds:2(%si) # work around buggy BIOSes -- movb $GETDEVICEPARAMETERS, %ah # Function 48 -- int $0x13 # make the call -- # Don't check for fail return -- # it doesn't matter. --edd_get_legacy_chs: -- xorw %ax, %ax -- movw %ax, %ds:-4(%si) -- movw %ax, %ds:-2(%si) -- # Ralf Brown's Interrupt List says to set ES:DI to -- # 0000h:0000h "to guard against BIOS bugs" -- pushw %es -- movw %ax, %es -- movw %ax, %di -- pushw %dx # legacy call clobbers %dl -- movb $LEGACYGETDEVICEPARAMETERS, %ah # Function 08 -- int $0x13 # make the call -- jc edd_legacy_done # failed -- movb %cl, %al # Low 6 bits are max -- andb $0x3F, %al # sector number -- movb %al, %ds:-1(%si) # Record max sect -- movb %dh, %ds:-2(%si) # Record max head number -- movb %ch, %al # Low 8 bits of max cyl -- shr $6, %cl -- movb %cl, %ah # High 2 bits of max cyl -- movw %ax, %ds:-4(%si) -- --edd_legacy_done: -- popw %dx -- popw %es -- movw %si, %ax # increment si -- addw $EDDPARMSIZE+EDDEXTSIZE, %ax -- movw %ax, %si -- --edd_next: -- incb %dl # increment to next device -- cmpb $EDDMAXNR, (EDDNR) # Out of space? -- jb edd_check_ext # keep looping -- --edd_done: --#endif ---- /dev/null -+++ b/arch/i386/boot/edd.c -@@ -0,0 +1,196 @@ -+/* -*- linux-c -*- ------------------------------------------------------- * -+ * -+ * Copyright (C) 1991, 1992 Linus Torvalds -+ * Copyright 2007 rPath, Inc. - All Rights Reserved -+ * -+ * This file is part of the Linux kernel, and is made available under -+ * the terms of the GNU General Public License version 2. -+ * -+ * ----------------------------------------------------------------------- */ -+ -+/* -+ * arch/i386/boot/edd.c -+ * -+ * Get EDD BIOS disk information -+ */ -+ -+#include "boot.h" -+#include <linux/edd.h> -+ -+#if defined(CONFIG_EDD) || defined(CONFIG_EDD_MODULE) -+ -+struct edd_dapa { -+ u8 pkt_size; -+ u8 rsvd; -+ u16 sector_cnt; -+ u16 buf_off, buf_seg; -+ u64 lba; -+ u64 buf_lin_addr; -+}; -+ -+/* -+ * Note: this uses the heap to hold the loaded sector. -+ */ -+static int read_sector(u8 devno, u64 lba, void *buf) -+{ -+ struct edd_dapa dapa; -+ u16 ax, bx, cx, dx, si; -+ -+ memset(&dapa, 0, sizeof dapa); -+ dapa.pkt_size = sizeof(dapa); -+ dapa.sector_cnt = 1; -+ dapa.buf_off = (size_t)buf; -+ dapa.buf_seg = ds(); -+ dapa.lba = lba; -+ -+ ax = 0x4200; /* Extended Read */ -+ si = (size_t)&dapa; -+ dx = devno; -+ asm("pushfl; stc; int $0x13; setc %%al; popfl" -+ : "+a" (ax), "+S" (si), "+d" (dx) -+ : "m" (dapa) -+ : "ebx", "ecx", "edi", "memory"); -+ -+ if (!(u8)ax) -+ return 0; /* OK */ -+ -+ ax = 0x0201; /* Legacy Read, one sector */ -+ cx = 0x0001; /* Sector 0-0-1 */ -+ dx = devno; -+ bx = (size_t)buf; -+ asm("pushfl; stc; int $0x13; setc %%al; popfl" -+ : "+a" (ax), "+c" (cx), "+d" (dx), "+b" (bx) -+ : : "esi", "edi", "memory"); -+ -+ return -(u8)ax; /* 0 or -1 */ -+} -+ -+static u32 read_mbr_sig(u8 devno, struct edd_info *ei) -+{ -+ int sector_size; -+ char *mbrbuf_ptr, *mbrbuf_end; -+ u32 mbrsig; -+ u32 buf_base, mbr_base; -+ extern char _end[]; -+ static char mbr_buf[1024]; -+ -+ sector_size = ei->params.bytes_per_sector; -+ if (!sector_size) -+ sector_size = 512; /* Best available guess */ -+ -+ buf_base = (ds() << 4) + (u32)&_end; -+ mbr_base = (buf_base+sector_size-1) & ~(sector_size-1); -+ mbrbuf_ptr = mbr_buf + (mbr_base-buf_base); -+ mbrbuf_end = mbrbuf_ptr + sector_size; -+ -+ if (!(boot_params.hdr.loadflags & CAN_USE_HEAP)) -+ return 0; -+ if (mbrbuf_end > (char *)(size_t)boot_params.hdr.heap_end_ptr) -+ return 0; -+ -+ if (read_sector(devno, 0, mbrbuf_ptr)) -+ return 0; -+ -+ mbrsig = *(u32 *)&mbrbuf_ptr[EDD_MBR_SIG_OFFSET]; -+ return mbrsig; -+} -+ -+static int get_edd_info(u8 devno, struct edd_info *ei) -+{ -+ u16 ax, bx, cx, dx, di; -+ -+ memset(ei, 0, sizeof *ei); -+ -+ /* Check Extensions Present */ -+ -+ ax = 0x4100; -+ bx = EDDMAGIC1; -+ dx = devno; -+ asm("pushfl; stc; int $0x13; setc %%al; popfl" -+ : "+a" (ax), "+b" (bx), "=c" (cx), "+d" (dx) -+ : : "esi", "edi"); -+ -+ if ((u8)ax) -+ return -1; /* No extended information */ -+ -+ if (bx != EDDMAGIC2) -+ return -1; -+ -+ ei->device = devno; -+ ei->version = ax >> 8; /* EDD version number */ -+ ei->interface_support = cx; /* EDD functionality subsets */ -+ -+ /* Extended Get Device Parameters */ -+ -+ ei->params.length = sizeof(ei->params); -+ ax = 0x4800; -+ dx = devno; -+ asm("pushfl; int $0x13; popfl" -+ : "+a" (ax), "+d" (dx) -+ : "S" (&ei->params) -+ : "ebx", "ecx", "edi"); -+ -+ /* Get legacy CHS parameters */ -+ -+ /* Ralf Brown recommends setting ES:DI to 0:0 */ -+ ax = 0x0800; -+ dx = devno; -+ di = 0; -+ asm("pushw %%es; " -+ "movw %%di,%%es; " -+ "pushfl; stc; int $0x13; setc %%al; popfl; " -+ "popw %%es" -+ : "+a" (ax), "=b" (bx), "=c" (cx), "+d" (dx), "+D" (di) -+ : : "esi"); -+ -+ if ((u8)ax == 0) { -+ ei->legacy_max_cylinder = (cx >> 8) + ((cx & 0xc0) << 2); -+ ei->legacy_max_head = dx >> 8; -+ ei->legacy_sectors_per_track = cx & 0x3f; -+ } -+ -+ return 0; -+} -+ -+void query_edd(void) -+{ -+ char eddarg[8]; -+ int do_mbr = 1; -+ int do_edd = 1; -+ int devno; -+ struct edd_info ei, *edp; -+ -+ if (cmdline_find_option("edd", eddarg, sizeof eddarg) > 0) { -+ if (!strcmp(eddarg, "skipmbr") || !strcmp(eddarg, "skip")) -+ do_mbr = 0; -+ else if (!strcmp(eddarg, "off")) -+ do_edd = 0; -+ } -+ -+ edp = (struct edd_info *)boot_params.eddbuf; -+ -+ if (!do_edd) -+ return; -+ -+ for (devno = 0x80; devno < 0x80+EDD_MBR_SIG_MAX; devno++) { -+ /* -+ * Scan the BIOS-supported hard disks and query EDD -+ * information... -+ */ -+ get_edd_info(devno, &ei); -+ -+ if (boot_params.eddbuf_entries < EDDMAXNR) { -+ memcpy(edp, &ei, sizeof ei); -+ edp++; -+ boot_params.eddbuf_entries++; -+ } -+ -+ if (do_mbr) { -+ u32 mbr_sig; -+ mbr_sig = read_mbr_sig(devno, &ei); -+ boot_params.edd_mbr_sig_buffer[devno-0x80] = mbr_sig; -+ } -+ } -+} -+ -+#endif ---- /dev/null -+++ b/arch/i386/boot/header.S -@@ -0,0 +1,283 @@ -+/* -+ * header.S -+ * -+ * Copyright (C) 1991, 1992 Linus Torvalds -+ * -+ * Based on bootsect.S and setup.S -+ * modified by more people than can be counted -+ * -+ * Rewritten as a common file by H. Peter Anvin (Apr 2007) -+ * -+ * BIG FAT NOTE: We're in real mode using 64k segments. Therefore segment -+ * addresses must be multiplied by 16 to obtain their respective linear -+ * addresses. To avoid confusion, linear addresses are written using leading -+ * hex while segment addresses are written as segment:offset. -+ * -+ */ -+ -+#include <asm/segment.h> -+#include <linux/utsrelease.h> -+#include <asm/boot.h> -+#include <asm/e820.h> -+#include <asm/page.h> -+#include <asm/setup.h> -+#include "boot.h" -+ -+SETUPSECTS = 4 /* default nr of setup-sectors */ -+BOOTSEG = 0x07C0 /* original address of boot-sector */ -+SYSSEG = DEF_SYSSEG /* system loaded at 0x10000 (65536) */ -+SYSSIZE = DEF_SYSSIZE /* system size: # of 16-byte clicks */ -+ /* to be loaded */ -+ROOT_DEV = 0 /* ROOT_DEV is now written by "build" */ -+SWAP_DEV = 0 /* SWAP_DEV is now written by "build" */ -+ -+#ifndef SVGA_MODE -+#define SVGA_MODE ASK_VGA -+#endif -+ -+#ifndef RAMDISK -+#define RAMDISK 0 -+#endif -+ -+#ifndef ROOT_RDONLY -+#define ROOT_RDONLY 1 -+#endif -+ -+ .code16 -+ .section ".bstext", "ax" -+ -+ .global bootsect_start -+bootsect_start: -+ -+ # Normalize the start address -+ ljmp $BOOTSEG, $start2 -+ -+start2: -+ movw %cs, %ax -+ movw %ax, %ds -+ movw %ax, %es -+ movw %ax, %ss -+ xorw %sp, %sp -+ sti -+ cld -+ -+ movw $bugger_off_msg, %si -+ -+msg_loop: -+ lodsb -+ andb %al, %al -+ jz bs_die -+ movb $0xe, %ah -+ movw $7, %bx -+ int $0x10 -+ jmp msg_loop -+ -+bs_die: -+ # Allow the user to press a key, then reboot -+ xorw %ax, %ax -+ int $0x16 -+ int $0x19 -+ -+ # int 0x19 should never return. In case it does anyway, -+ # invoke the BIOS reset code... -+ ljmp $0xf000,$0xfff0 -+ -+ .section ".bsdata", "a" -+bugger_off_msg: -+ .ascii "Direct booting from floppy is no longer supported.\r\n" -+ .ascii "Please use a boot loader program instead.\r\n" -+ .ascii "\n" -+ .ascii "Remove disk and press any key to reboot . . .\r\n" -+ .byte 0 -+ -+ -+ # Kernel attributes; used by setup. This is part 1 of the -+ # header, from the old boot sector. -+ -+ .section ".header", "a" -+ .globl hdr -+hdr: -+setup_sects: .byte SETUPSECTS -+root_flags: .word ROOT_RDONLY -+syssize: .long SYSSIZE -+ram_size: .word RAMDISK -+vid_mode: .word SVGA_MODE -+root_dev: .word ROOT_DEV -+boot_flag: .word 0xAA55 -+ -+ # offset 512, entry point -+ -+ .globl _start -+_start: -+ # Explicitly enter this as bytes, or the assembler -+ # tries to generate a 3-byte jump here, which causes -+ # everything else to push off to the wrong offset. -+ .byte 0xeb # short (2-byte) jump -+ .byte start_of_setup-1f -+1: -+ -+ # Part 2 of the header, from the old setup.S -+ -+ .ascii "HdrS" # header signature -+ .word 0x0206 # header version number (>= 0x0105) -+ # or else old loadlin-1.5 will fail) -+ .globl realmode_swtch -+realmode_swtch: .word 0, 0 # default_switch, SETUPSEG -+start_sys_seg: .word SYSSEG -+ .word kernel_version-512 # pointing to kernel version string -+ # above section of header is compatible -+ # with loadlin-1.5 (header v1.5). Don't -+ # change it. -+ -+type_of_loader: .byte 0 # = 0, old one (LILO, Loadlin, -+ # Bootlin, SYSLX, bootsect...) -+ # See Documentation/i386/boot.txt for -+ # assigned ids -+ -+# flags, unused bits must be zero (RFU) bit within loadflags -+loadflags: -+LOADED_HIGH = 1 # If set, the kernel is loaded high -+CAN_USE_HEAP = 0x80 # If set, the loader also has set -+ # heap_end_ptr to tell how much -+ # space behind setup.S can be used for -+ # heap purposes. -+ # Only the loader knows what is free -+#ifndef __BIG_KERNEL__ -+ .byte 0 -+#else -+ .byte LOADED_HIGH -+#endif -+ -+setup_move_size: .word 0x8000 # size to move, when setup is not -+ # loaded at 0x90000. We will move setup -+ # to 0x90000 then just before jumping -+ # into the kernel. However, only the -+ # loader knows how much data behind -+ # us also needs to be loaded. -+ -+code32_start: # here loaders can put a different -+ # start address for 32-bit code. -+#ifndef __BIG_KERNEL__ -+ .long 0x1000 # 0x1000 = default for zImage -+#else -+ .long 0x100000 # 0x100000 = default for big kernel -+#endif -+ -+ramdisk_image: .long 0 # address of loaded ramdisk image -+ # Here the loader puts the 32-bit -+ # address where it loaded the image. -+ # This only will be read by the kernel. -+ -+ramdisk_size: .long 0 # its size in bytes -+ -+bootsect_kludge: -+ .long 0 # obsolete -+ -+heap_end_ptr: .word _end+1024 # (Header version 0x0201 or later) -+ # space from here (exclusive) down to -+ # end of setup code can be used by setup -+ # for local heap purposes. -+ -+pad1: .word 0 -+cmd_line_ptr: .long 0 # (Header version 0x0202 or later) -+ # If nonzero, a 32-bit pointer -+ # to the kernel command line. -+ # The command line should be -+ # located between the start of -+ # setup and the end of low -+ # memory (0xa0000), or it may -+ # get overwritten before it -+ # gets read. If this field is -+ # used, there is no longer -+ # anything magical about the -+ # 0x90000 segment; the setup -+ # can be located anywhere in -+ # low memory 0x10000 or higher. -+ -+ramdisk_max: .long (-__PAGE_OFFSET-(512 << 20)-1) & 0x7fffffff -+ # (Header version 0x0203 or later) -+ # The highest safe address for -+ # the contents of an initrd -+ -+kernel_alignment: .long CONFIG_PHYSICAL_ALIGN #physical addr alignment -+ #required for protected mode -+ #kernel -+#ifdef CONFIG_RELOCATABLE -+relocatable_kernel: .byte 1 -+#else -+relocatable_kernel: .byte 0 -+#endif -+pad2: .byte 0 -+pad3: .word 0 -+ -+cmdline_size: .long COMMAND_LINE_SIZE-1 #length of the command line, -+ #added with boot protocol -+ #version 2.06 -+ -+# End of setup header ##################################################### -+ -+ .section ".inittext", "ax" -+start_of_setup: -+#ifdef SAFE_RESET_DISK_CONTROLLER -+# Reset the disk controller. -+ movw $0x0000, %ax # Reset disk controller -+ movb $0x80, %dl # All disks -+ int $0x13 -+#endif -+ -+# We will have entired with %cs = %ds+0x20, normalize %cs so -+# it is on par with the other segments. -+ pushw %ds -+ pushw $setup2 -+ lretw -+ -+setup2: -+# Force %es = %ds -+ movw %ds, %ax -+ movw %ax, %es -+ cld -+ -+# Stack paranoia: align the stack and make sure it is good -+# for both 16- and 32-bit references. In particular, if we -+# were meant to have been using the full 16-bit segment, the -+# caller might have set %sp to zero, which breaks %esp-based -+# references. -+ andw $~3, %sp # dword align (might as well...) -+ jnz 1f -+ movw $0xfffc, %sp # Make sure we're not zero -+1: movzwl %sp, %esp # Clear upper half of %esp -+ sti -+ -+# Check signature at end of setup -+ cmpl $0x5a5aaa55, setup_sig -+ jne setup_bad -+ -+# Zero the bss -+ movw $__bss_start, %di -+ movw $_end+3, %cx -+ xorl %eax, %eax -+ subw %di, %cx -+ shrw $2, %cx -+ rep; stosl -+ -+# Jump to C code (should not return) -+ calll main -+ -+# Setup corrupt somehow... -+setup_bad: -+ movl $setup_corrupt, %eax -+ calll puts -+ # Fall through... -+ -+ .globl die -+ .type die, @function -+die: -+ hlt -+ jmp die -+ -+ .size die, .-due -+ -+ .section ".initdata", "a" -+setup_corrupt: -+ .byte 7 -+ .string "No setup signature found..." ---- /dev/null -+++ b/arch/i386/boot/main.c -@@ -0,0 +1,161 @@ -+/* -*- linux-c -*- ------------------------------------------------------- * -+ * -+ * Copyright (C) 1991, 1992 Linus Torvalds -+ * Copyright 2007 rPath, Inc. - All Rights Reserved -+ * -+ * This file is part of the Linux kernel, and is made available under -+ * the terms of the GNU General Public License version 2. -+ * -+ * ----------------------------------------------------------------------- */ -+ -+/* -+ * arch/i386/boot/main.c -+ * -+ * Main module for the real-mode kernel code -+ */ -+ -+#include "boot.h" -+ -+struct boot_params boot_params __attribute__((aligned(16))); -+ -+char *HEAP = _end; -+char *heap_end = _end; /* Default end of heap = no heap */ -+ -+/* -+ * Copy the header into the boot parameter block. Since this -+ * screws up the old-style command line protocol, adjust by -+ * filling in the new-style command line pointer instead. -+ */ -+#define OLD_CL_MAGIC 0xA33F -+#define OLD_CL_ADDRESS 0x20 -+ -+static void copy_boot_params(void) -+{ -+ struct old_cmdline { -+ u16 cl_magic; -+ u16 cl_offset; -+ }; -+ const struct old_cmdline * const oldcmd = -+ (const struct old_cmdline *)OLD_CL_ADDRESS; -+ -+ BUILD_BUG_ON(sizeof boot_params != 4096); -+ memcpy(&boot_params.hdr, &hdr, sizeof hdr); -+ -+ if (!boot_params.hdr.cmd_line_ptr && -+ oldcmd->cl_magic == OLD_CL_MAGIC) { -+ /* Old-style command line protocol. */ -+ u16 cmdline_seg; -+ -+ /* Figure out if the command line falls in the region -+ of memory that an old kernel would have copied up -+ to 0x90000... */ -+ if (oldcmd->cl_offset < boot_params.hdr.setup_move_size) -+ cmdline_seg = ds(); -+ else -+ cmdline_seg = 0x9000; -+ -+ boot_params.hdr.cmd_line_ptr = -+ (cmdline_seg << 4) + oldcmd->cl_offset; -+ } -+} -+ -+/* -+ * Set the keyboard repeat rate to maximum. Unclear why this -+ * is done here; this might be possible to kill off as stale code. -+ */ -+static void keyboard_set_repeat(void) -+{ -+ u16 ax = 0x0305; -+ u16 bx = 0; -+ asm volatile("int $0x16" -+ : "+a" (ax), "+b" (bx) -+ : : "ecx", "edx", "esi", "edi"); -+} -+ -+/* -+ * Get Intel SpeedStep IST information. -+ */ -+static void query_speedstep_ist(void) -+{ -+ asm("int $0x15" -+ : "=a" (boot_params.speedstep_info[0]), -+ "=b" (boot_params.speedstep_info[1]), -+ "=c" (boot_params.speedstep_info[2]), -+ "=d" (boot_params.speedstep_info[3]) -+ : "a" (0x0000e980), /* IST Support */ -+ "d" (0x47534943)); /* Request value */ -+} -+ -+/* -+ * Tell the BIOS what CPU mode we intend to run in. -+ */ -+static void set_bios_mode(void) -+{ -+#ifdef CONFIG_X86_64 -+ u32 eax, ebx; -+ -+ eax = 0xec00; -+ ebx = 2; -+ asm volatile("int $0x15" -+ : "+a" (eax), "+b" (ebx) -+ : : "ecx", "edx", "esi", "edi"); -+#endif -+} -+ -+void main(void) -+{ -+ /* First, copy the boot header into the "zeropage" */ -+ copy_boot_params(); -+ -+ /* End of heap check */ -+ if (boot_params.hdr.loadflags & CAN_USE_HEAP) { -+ heap_end = (char *)(boot_params.hdr.heap_end_ptr -+ +0x200-STACK_SIZE); -+ } else { -+ /* Boot protocol 2.00 only, no heap available */ -+ puts("WARNING: Ancient bootloader, some functionality " -+ "may be limited!\n"); -+ } -+ -+ /* Make sure we have all the proper CPU support */ -+ if (validate_cpu()) { -+ puts("Unable to boot - please use a kernel appropriate " -+ "for your CPU.\n"); -+ die(); -+ } -+ -+ /* Tell the BIOS what CPU mode we intend to run in. */ -+ set_bios_mode(); -+ -+ /* Detect memory layout */ -+ detect_memory(); -+ -+ /* Set keyboard repeat rate (why?) */ -+ keyboard_set_repeat(); -+ -+ /* Set the video mode */ -+ set_video(); -+ -+ /* Query MCA information */ -+ query_mca(); -+ -+ /* Voyager */ -+#ifdef CONFIG_X86_VOYAGER -+ query_voyager(); -+#endif -+ -+ /* Query SpeedStep IST information */ -+ query_speedstep_ist(); -+ -+ /* Query APM information */ -+#if defined(CONFIG_APM) || defined(CONFIG_APM_MODULE) -+ query_apm_bios(); -+#endif -+ -+ /* Query EDD information */ -+#if defined(CONFIG_EDD) || defined(CONFIG_EDD_MODULE) -+ query_edd(); -+#endif -+ /* Do the last things and invoke protected mode */ -+ go_to_protected_mode(); -+} ---- /dev/null -+++ b/arch/i386/boot/mca.c -@@ -0,0 +1,43 @@ -+/* -*- linux-c -*- ------------------------------------------------------- * -+ * -+ * Copyright (C) 1991, 1992 Linus Torvalds -+ * Copyright 2007 rPath, Inc. - All Rights Reserved -+ * -+ * This file is part of the Linux kernel, and is made available under -+ * the terms of the GNU General Public License version 2. -+ * -+ * ----------------------------------------------------------------------- */ -+ -+/* -+ * arch/i386/boot/mca.c -+ * -+ * Get the MCA system description table -+ */ -+ -+#include "boot.h" -+ -+int query_mca(void) -+{ -+ u8 err; -+ u16 es, bx, len; -+ -+ asm("pushw %%es ; " -+ "int $0x15 ; " -+ "setc %0 ; " -+ "movw %%es, %1 ; " -+ "popw %%es" -+ : "=acdSDm" (err), "=acdSDm" (es), "=b" (bx) -+ : "a" (0xc000)); -+ -+ if (err) -+ return -1; /* No MCA present */ -+ -+ set_fs(es); -+ len = rdfs16(bx); -+ -+ if (len > sizeof(boot_params.sys_desc_table)) -+ len = sizeof(boot_params.sys_desc_table); -+ -+ copy_from_fs(&boot_params.sys_desc_table, bx, len); -+ return 0; -+} ---- /dev/null -+++ b/arch/i386/boot/memory.c -@@ -0,0 +1,99 @@ -+/* -*- linux-c -*- ------------------------------------------------------- * -+ * -+ * Copyright (C) 1991, 1992 Linus Torvalds -+ * Copyright 2007 rPath, Inc. - All Rights Reserved -+ * -+ * This file is part of the Linux kernel, and is made available under -+ * the terms of the GNU General Public License version 2. -+ * -+ * ----------------------------------------------------------------------- */ -+ -+/* -+ * arch/i386/boot/memory.c -+ * -+ * Memory detection code -+ */ -+ -+#include "boot.h" -+ -+#define SMAP 0x534d4150 /* ASCII "SMAP" */ -+ -+static int detect_memory_e820(void) -+{ -+ u32 next = 0; -+ u32 size, id; -+ u8 err; -+ struct e820entry *desc = boot_params.e820_map; -+ -+ do { -+ size = sizeof(struct e820entry); -+ id = SMAP; -+ asm("int $0x15; setc %0" -+ : "=am" (err), "+b" (next), "+d" (id), "+c" (size), -+ "=m" (*desc) -+ : "D" (desc), "a" (0xe820)); -+ -+ if (err || id != SMAP) -+ break; -+ -+ boot_params.e820_entries++; -+ desc++; -+ } while (next && boot_params.e820_entries < E820MAX); -+ -+ return boot_params.e820_entries; -+} -+ -+static int detect_memory_e801(void) -+{ -+ u16 ax, bx, cx, dx; -+ u8 err; -+ -+ bx = cx = dx = 0; -+ ax = 0xe801; -+ asm("stc; int $0x15; setc %0" -+ : "=m" (err), "+a" (ax), "+b" (bx), "+c" (cx), "+d" (dx)); -+ -+ if (err) -+ return -1; -+ -+ /* Do we really need to do this? */ -+ if (cx || dx) { -+ ax = cx; -+ bx = dx; -+ } -+ -+ if (ax > 15*1024) -+ return -1; /* Bogus! */ -+ -+ /* This ignores memory above 16MB if we have a memory hole -+ there. If someone actually finds a machine with a memory -+ hole at 16MB and no support for 0E820h they should probably -+ generate a fake e820 map. */ -+ boot_params.alt_mem_k = (ax == 15*1024) ? (dx << 6)+ax : ax; -+ -+ return 0; -+} -+ -+static int detect_memory_88(void) -+{ -+ u16 ax; -+ u8 err; -+ -+ ax = 0x8800; -+ asm("stc; int $0x15; setc %0" : "=bcdm" (err), "+a" (ax)); -+ -+ boot_params.screen_info.ext_mem_k = ax; -+ -+ return -err; -+} -+ -+int detect_memory(void) -+{ -+ if (detect_memory_e820() > 0) -+ return 0; -+ -+ if (!detect_memory_e801()) -+ return 0; -+ -+ return detect_memory_88(); -+} ---- /dev/null -+++ b/arch/i386/boot/pm.c -@@ -0,0 +1,170 @@ -+/* -*- linux-c -*- ------------------------------------------------------- * -+ * -+ * Copyright (C) 1991, 1992 Linus Torvalds -+ * Copyright 2007 rPath, Inc. - All Rights Reserved -+ * -+ * This file is part of the Linux kernel, and is made available under -+ * the terms of the GNU General Public License version 2. -+ * -+ * ----------------------------------------------------------------------- */ -+ -+/* -+ * arch/i386/boot/pm.c -+ * -+ * Prepare the machine for transition to protected mode. -+ */ -+ -+#include "boot.h" -+#include <asm/segment.h> -+ -+/* -+ * Invoke the realmode switch hook if present; otherwise -+ * disable all interrupts. -+ */ -+static void realmode_switch_hook(void) -+{ -+ if (boot_params.hdr.realmode_swtch) { -+ asm volatile("lcallw *%0" -+ : : "m" (boot_params.hdr.realmode_swtch) -+ : "eax", "ebx", "ecx", "edx"); -+ } else { -+ asm volatile("cli"); -+ outb(0x80, 0x70); /* Disable NMI */ -+ io_delay(); -+ } -+} -+ -+/* -+ * A zImage kernel is loaded at 0x10000 but wants to run at 0x1000. -+ * A bzImage kernel is loaded and runs at 0x100000. -+ */ -+static void move_kernel_around(void) -+{ -+ /* Note: rely on the compile-time option here rather than -+ the LOADED_HIGH flag. The Qemu kernel loader unconditionally -+ sets the loadflags to zero. */ -+#ifndef __BIG_KERNEL__ -+ u16 dst_seg, src_seg; -+ u32 syssize; -+ -+ dst_seg = 0x1000 >> 4; -+ src_seg = 0x10000 >> 4; -+ syssize = boot_params.hdr.syssize; /* Size in 16-byte paragraps */ -+ -+ while (syssize) { -+ int paras = (syssize >= 0x1000) ? 0x1000 : syssize; -+ int dwords = paras << 2; -+ -+ asm volatile("pushw %%es ; " -+ "pushw %%ds ; " -+ "movw %1,%%es ; " -+ "movw %2,%%ds ; " -+ "xorw %%di,%%di ; " -+ "xorw %%si,%%si ; " -+ "rep;movsl ; " -+ "popw %%ds ; " -+ "popw %%es" -+ : "+c" (dwords) -+ : "rm" (dst_seg), "rm" (src_seg) -+ : "esi", "edi"); -+ -+ syssize -= paras; -+ dst_seg += paras; -+ src_seg += paras; -+ } -+#endif -+} -+ -+/* -+ * Disable all interrupts at the legacy PIC. -+ */ -+static void mask_all_interrupts(void) -+{ -+ outb(0xff, 0xa1); /* Mask all interrupts on the seconday PIC */ -+ io_delay(); -+ outb(0xfb, 0x21); /* Mask all but cascade on the primary PIC */ -+ io_delay(); -+} -+ -+/* -+ * Reset IGNNE# if asserted in the FPU. -+ */ -+static void reset_coprocessor(void) -+{ -+ outb(0, 0xf0); -+ io_delay(); -+ outb(0, 0xf1); -+ io_delay(); -+} -+ -+/* -+ * Set up the GDT -+ */ -+#define GDT_ENTRY(flags,base,limit) \ -+ (((u64)(base & 0xff000000) << 32) | \ -+ ((u64)flags << 40) | \ -+ ((u64)(limit & 0x00ff0000) << 32) | \ -+ ((u64)(base & 0x00ffff00) << 16) | \ -+ ((u64)(limit & 0x0000ffff))) -+ -+struct gdt_ptr { -+ u16 len; -+ u32 ptr; -+} __attribute__((packed)); -+ -+static void setup_gdt(void) -+{ -+ /* There are machines which are known to not boot with the GDT -+ being 8-byte unaligned. Intel recommends 16 byte alignment. */ -+ static const u64 boot_gdt[] __attribute__((aligned(16))) = { -+ /* CS: code, read/execute, 4 GB, base 0 */ -+ [GDT_ENTRY_BOOT_CS] = GDT_ENTRY(0xc09b, 0, 0xfffff), -+ /* DS: data, read/write, 4 GB, base 0 */ -+ [GDT_ENTRY_BOOT_DS] = GDT_ENTRY(0xc093, 0, 0xfffff), -+ }; -+ struct gdt_ptr gdt; -+ -+ gdt.len = sizeof(boot_gdt)-1; -+ gdt.ptr = (u32)&boot_gdt + (ds() << 4); -+ -+ asm volatile("lgdtl %0" : : "m" (gdt)); -+} -+ -+/* -+ * Set up the IDT -+ */ -+static void setup_idt(void) -+{ -+ static const struct gdt_ptr null_idt = {0, 0}; -+ asm volatile("lidtl %0" : : "m" (null_idt)); -+} -+ -+/* -+ * Actual invocation sequence -+ */ -+void go_to_protected_mode(void) -+{ -+ /* Hook before leaving real mode, also disables interrupts */ -+ realmode_switch_hook(); -+ -+ /* Move the kernel/setup to their final resting places */ -+ move_kernel_around(); -+ -+ /* Enable the A20 gate */ -+ if (enable_a20()) { -+ puts("A20 gate not responding, unable to boot...\n"); -+ die(); -+ } -+ -+ /* Reset coprocessor (IGNNE#) */ -+ reset_coprocessor(); -+ -+ /* Mask all interrupts in the PIC */ -+ mask_all_interrupts(); -+ -+ /* Actual transition to protected mode... */ -+ setup_idt(); -+ setup_gdt(); -+ protected_mode_jump(boot_params.hdr.code32_start, -+ (u32)&boot_params + (ds() << 4)); -+} ---- /dev/null -+++ b/arch/i386/boot/pmjump.S -@@ -0,0 +1,54 @@ -+/* ----------------------------------------------------------------------- * -+ * -+ * Copyright (C) 1991, 1992 Linus Torvalds -+ * Copyright 2007 rPath, Inc. - All Rights Reserved -+ * -+ * This file is part of the Linux kernel, and is made available under -+ * the terms of the GNU General Public License version 2. -+ * -+ * ----------------------------------------------------------------------- */ -+ -+/* -+ * arch/i386/boot/pmjump.S -+ * -+ * The actual transition into protected mode -+ */ -+ -+#include <asm/boot.h> -+#include <asm/segment.h> -+ -+ .text -+ -+ .globl protected_mode_jump -+ .type protected_mode_jump, @function -+ -+ .code16 -+ -+/* -+ * void protected_mode_jump(u32 entrypoint, u32 bootparams); -+ */ -+protected_mode_jump: -+ xorl %ebx, %ebx # Flag to indicate this is a boot -+ movl %edx, %esi # Pointer to boot_params table -+ movl %eax, 2f # Patch ljmpl instruction -+ jmp 1f # Short jump to flush instruction q. -+ -+1: -+ movw $__BOOT_DS, %cx -+ -+ movl %cr0, %edx -+ orb $1, %dl # Protected mode (PE) bit -+ movl %edx, %cr0 -+ -+ movw %cx, %ds -+ movw %cx, %es -+ movw %cx, %fs -+ movw %cx, %gs -+ movw %cx, %ss -+ -+ # Jump to the 32-bit entrypoint -+ .byte 0x66, 0xea # ljmpl opcode -+2: .long 0 # offset -+ .word __BOOT_CS # segment -+ -+ .size protected_mode_jump, .-protected_mode_jump ---- /dev/null -+++ b/arch/i386/boot/printf.c -@@ -0,0 +1,331 @@ -+/* -*- linux-c -*- ------------------------------------------------------- * -+ * -+ * Copyright (C) 1991, 1992 Linus Torvalds -+ * Copyright 2007 rPath, Inc. - All Rights Reserved -+ * -+ * This file is part of the Linux kernel, and is made available under -+ * the terms of the GNU General Public License version 2. -+ * -+ * ----------------------------------------------------------------------- */ -+ -+/* -+ * arch/i386/boot/printf.c -+ * -+ * Oh, it's a waste of space, but oh-so-yummy for debugging. This -+ * version of printf() does not include 64-bit support. "Live with -+ * it." -+ * -+ */ -+ -+#include "boot.h" -+ -+static inline int isdigit(int ch) -+{ -+ return (ch >= '0') && (ch <= '9'); -+} -+ -+static int skip_atoi(const char **s) -+{ -+ int i = 0; -+ -+ while (isdigit(**s)) -+ i = i * 10 + *((*s)++) - '0'; -+ return i; -+} -+ -+unsigned int atou(const char *s) -+{ -+ unsigned int i = 0; -+ while (isdigit(*s)) -+ i = i * 10 + (*s++ - '0'); -+ return i; -+} -+ -+static int strnlen(const char *s, int maxlen) -+{ -+ const char *es = s; -+ while (*es && maxlen) { -+ es++; -+ maxlen--; -+ } -+ -+ return (es - s); -+} -+ -+#define ZEROPAD 1 /* pad with zero */ -+#define SIGN 2 /* unsigned/signed long */ -+#define PLUS 4 /* show plus */ -+#define SPACE 8 /* space if plus */ -+#define LEFT 16 /* left justified */ -+#define SPECIAL 32 /* 0x */ -+#define LARGE 64 /* use 'ABCDEF' instead of 'abcdef' */ -+ -+#define do_div(n,base) ({ \ -+int __res; \ -+__res = ((unsigned long) n) % (unsigned) base; \ -+n = ((unsigned long) n) / (unsigned) base; \ -+__res; }) -+ -+static char *number(char *str, long num, int base, int size, int precision, -+ int type) -+{ -+ char c, sign, tmp[66]; -+ const char *digits = "0123456789abcdefghijklmnopqrstuvwxyz"; -+ int i; -+ -+ if (type & LARGE) -+ digits = "0123456789ABCDEFGHIJKLMNOPQRSTUVWXYZ"; -+ if (type & LEFT) -+ type &= ~ZEROPAD; -+ if (base < 2 || base > 36) -+ return 0; -+ c = (type & ZEROPAD) ? '0' : ' '; -+ sign = 0; -+ if (type & SIGN) { -+ if (num < 0) { -+ sign = '-'; -+ num = -num; -+ size--; -+ } else if (type & PLUS) { -+ sign = '+'; -+ size--; -+ } else if (type & SPACE) { -+ sign = ' '; -+ size--; -+ } -+ } -+ if (type & SPECIAL) { -+ if (base == 16) -+ size -= 2; -+ else if (base == 8) -+ size--; -+ } -+ i = 0; -+ if (num == 0) -+ tmp[i++] = '0'; -+ else -+ while (num != 0) -+ tmp[i++] = digits[do_div(num, base)]; -+ if (i > precision) -+ precision = i; -+ size -= precision; -+ if (!(type & (ZEROPAD + LEFT))) -+ while (size-- > 0) -+ *str++ = ' '; -+ if (sign) -+ *str++ = sign; -+ if (type & SPECIAL) { -+ if (base == 8) -+ *str++ = '0'; -+ else if (base == 16) { -+ *str++ = '0'; -+ *str++ = digits[33]; -+ } -+ } -+ if (!(type & LEFT)) -+ while (size-- > 0) -+ *str++ = c; -+ while (i < precision--) -+ *str++ = '0'; -+ while (i-- > 0) -+ *str++ = tmp[i]; -+ while (size-- > 0) -+ *str++ = ' '; -+ return str; -+} -+ -+int vsprintf(char *buf, const char *fmt, va_list args) -+{ -+ int len; -+ unsigned long num; -+ int i, base; -+ char *str; -+ const char *s; -+ -+ int flags; /* flags to number() */ -+ -+ int field_width; /* width of output field */ -+ int precision; /* min. # of digits for integers; max -+ number of chars for from string */ -+ int qualifier; /* 'h', 'l', or 'L' for integer fields */ -+ -+ for (str = buf; *fmt; ++fmt) { -+ if (*fmt != '%') { -+ *str++ = *fmt; -+ continue; -+ } -+ -+ /* process flags */ -+ flags = 0; -+ repeat: -+ ++fmt; /* this also skips first '%' */ -+ switch (*fmt) { -+ case '-': -+ flags |= LEFT; -+ goto repeat; -+ case '+': -+ flags |= PLUS; -+ goto repeat; -+ case ' ': -+ flags |= SPACE; -+ goto repeat; -+ case '#': -+ flags |= SPECIAL; -+ goto repeat; -+ case '0': -+ flags |= ZEROPAD; -+ goto repeat; -+ } -+ -+ /* get field width */ -+ field_width = -1; -+ if (isdigit(*fmt)) -+ field_width = skip_atoi(&fmt); -+ else if (*fmt == '*') { -+ ++fmt; -+ /* it's the next argument */ -+ field_width = va_arg(args, int); -+ if (field_width < 0) { -+ field_width = -field_width; -+ flags |= LEFT; -+ } -+ } -+ -+ /* get the precision */ -+ precision = -1; -+ if (*fmt == '.') { -+ ++fmt; -+ if (isdigit(*fmt)) -+ precision = skip_atoi(&fmt); -+ else if (*fmt == '*') { -+ ++fmt; -+ /* it's the next argument */ -+ precision = va_arg(args, int); -+ } -+ if (precision < 0) -+ precision = 0; -+ } -+ -+ /* get the conversion qualifier */ -+ qualifier = -1; -+ if (*fmt == 'h' || *fmt == 'l' || *fmt == 'L') { -+ qualifier = *fmt; -+ ++fmt; -+ } -+ -+ /* default base */ -+ base = 10; -+ -+ switch (*fmt) { -+ case 'c': -+ if (!(flags & LEFT)) -+ while (--field_width > 0) -+ *str++ = ' '; -+ *str++ = (unsigned char)va_arg(args, int); -+ while (--field_width > 0) -+ *str++ = ' '; -+ continue; -+ -+ case 's': -+ s = va_arg(args, char *); -+ len = strnlen(s, precision); -+ -+ if (!(flags & LEFT)) -+ while (len < field_width--) -+ *str++ = ' '; -+ for (i = 0; i < len; ++i) -+ *str++ = *s++; -+ while (len < field_width--) -+ *str++ = ' '; -+ continue; -+ -+ case 'p': -+ if (field_width == -1) { -+ field_width = 2 * sizeof(void *); -+ flags |= ZEROPAD; -+ } -+ str = number(str, -+ (unsigned long)va_arg(args, void *), 16, -+ field_width, precision, flags); -+ continue; -+ -+ case 'n': -+ if (qualifier == 'l') { -+ long *ip = va_arg(args, long *); -+ *ip = (str - buf); -+ } else { -+ int *ip = va_arg(args, int *); -+ *ip = (str - buf); -+ } -+ continue; -+ -+ case '%': -+ *str++ = '%'; -+ continue; -+ -+ /* integer number formats - set up the flags and "break" */ -+ case 'o': -+ base = 8; -+ break; -+ -+ case 'X': -+ flags |= LARGE; -+ case 'x': -+ base = 16; -+ break; -+ -+ case 'd': -+ case 'i': -+ flags |= SIGN; -+ case 'u': -+ break; -+ -+ default: -+ *str++ = '%'; -+ if (*fmt) -+ *str++ = *fmt; -+ else -+ --fmt; -+ continue; -+ } -+ if (qualifier == 'l') -+ num = va_arg(args, unsigned long); -+ else if (qualifier == 'h') { -+ num = (unsigned short)va_arg(args, int); -+ if (flags & SIGN) -+ num = (short)num; -+ } else if (flags & SIGN) -+ num = va_arg(args, int); -+ else -+ num = va_arg(args, unsigned int); -+ str = number(str, num, base, field_width, precision, flags); -+ } -+ *str = '\0'; -+ return str - buf; -+} -+ -+int sprintf(char *buf, const char *fmt, ...) -+{ -+ va_list args; -+ int i; -+ -+ va_start(args, fmt); -+ i = vsprintf(buf, fmt, args); -+ va_end(args); -+ return i; -+} -+ -+int printf(const char *fmt, ...) -+{ -+ char printf_buf[1024]; -+ va_list args; -+ int printed; -+ -+ va_start(args, fmt); -+ printed = vsprintf(printf_buf, fmt, args); -+ va_end(args); -+ -+ puts(printf_buf); -+ -+ return printed; -+} ---- a/arch/i386/boot/setup.S -+++ /dev/null -@@ -1,1075 +0,0 @@ --/* -- * setup.S Copyright (C) 1991, 1992 Linus Torvalds -- * -- * setup.s is responsible for getting the system data from the BIOS, -- * and putting them into the appropriate places in system memory. -- * both setup.s and system has been loaded by the bootblock. -- * -- * This code asks the bios for memory/disk/other parameters, and -- * puts them in a "safe" place: 0x90000-0x901FF, ie where the -- * boot-block used to be. It is then up to the protected mode -- * system to read them from there before the area is overwritten -- * for buffer-blocks. -- * -- * Move PS/2 aux init code to psaux.c -- * (troyer@saifr00.cfsat.Honeywell.COM) 03Oct92 -- * -- * some changes and additional features by Christoph Niemann, -- * March 1993/June 1994 (Christoph.Niemann@linux.org) -- * -- * add APM BIOS checking by Stephen Rothwell, May 1994 -- * (sfr@canb.auug.org.au) -- * -- * High load stuff, initrd support and position independency -- * by Hans Lermen & Werner Almesberger, February 1996 -- * <lermen@elserv.ffm.fgan.de>, <almesber@lrc.epfl.ch> -- * -- * Video handling moved to video.S by Martin Mares, March 1996 -- * <mj@k332.feld.cvut.cz> -- * -- * Extended memory detection scheme retwiddled by orc@pell.chi.il.us (david -- * parsons) to avoid loadlin confusion, July 1997 -- * -- * Transcribed from Intel (as86) -> AT&T (gas) by Chris Noe, May 1999. -- * <stiker@northlink.com> -- * -- * Fix to work around buggy BIOSes which don't use carry bit correctly -- * and/or report extended memory in CX/DX for e801h memory size detection -- * call. As a result the kernel got wrong figures. The int15/e801h docs -- * from Ralf Brown interrupt list seem to indicate AX/BX should be used -- * anyway. So to avoid breaking many machines (presumably there was a reason -- * to orginally use CX/DX instead of AX/BX), we do a kludge to see -- * if CX/DX have been changed in the e801 call and if so use AX/BX . -- * Michael Miller, April 2001 <michaelm@mjmm.org> -- * -- * New A20 code ported from SYSLINUX by H. Peter Anvin. AMD Elan bugfixes -- * by Robert Schwebel, December 2001 <robert@schwebel.de> -- */ -- --#include <asm/segment.h> --#include <linux/utsrelease.h> --#include <linux/compile.h> --#include <asm/boot.h> --#include <asm/e820.h> --#include <asm/page.h> --#include <asm/setup.h> -- --/* Signature words to ensure LILO loaded us right */ --#define SIG1 0xAA55 --#define SIG2 0x5A5A -- --INITSEG = DEF_INITSEG # 0x9000, we move boot here, out of the way --SYSSEG = DEF_SYSSEG # 0x1000, system loaded at 0x10000 (65536). --SETUPSEG = DEF_SETUPSEG # 0x9020, this is the current segment -- # ... and the former contents of CS -- --DELTA_INITSEG = SETUPSEG - INITSEG # 0x0020 -- --.code16 --.globl begtext, begdata, begbss, endtext, enddata, endbss -- --.text --begtext: --.data --begdata: --.bss --begbss: --.text -- --start: -- jmp trampoline -- --# This is the setup header, and it must start at %cs:2 (old 0x9020:2) -- -- .ascii "HdrS" # header signature -- .word 0x0206 # header version number (>= 0x0105) -- # or else old loadlin-1.5 will fail) --realmode_swtch: .word 0, 0 # default_switch, SETUPSEG --start_sys_seg: .word SYSSEG -- .word kernel_version # pointing to kernel version string -- # above section of header is compatible -- # with loadlin-1.5 (header v1.5). Don't -- # change it. -- --type_of_loader: .byte 0 # = 0, old one (LILO, Loadlin, -- # Bootlin, SYSLX, bootsect...) -- # See Documentation/i386/boot.txt for -- # assigned ids -- --# flags, unused bits must be zero (RFU) bit within loadflags --loadflags: --LOADED_HIGH = 1 # If set, the kernel is loaded high --CAN_USE_HEAP = 0x80 # If set, the loader also has set -- # heap_end_ptr to tell how much -- # space behind setup.S can be used for -- # heap purposes. -- # Only the loader knows what is free --#ifndef __BIG_KERNEL__ -- .byte 0 --#else -- .byte LOADED_HIGH --#endif -- --setup_move_size: .word 0x8000 # size to move, when setup is not -- # loaded at 0x90000. We will move setup -- # to 0x90000 then just before jumping -- # into the kernel. However, only the -- # loader knows how much data behind -- # us also needs to be loaded. -- --code32_start: # here loaders can put a different -- # start address for 32-bit code. --#ifndef __BIG_KERNEL__ -- .long 0x1000 # 0x1000 = default for zImage --#else -- .long 0x100000 # 0x100000 = default for big kernel --#endif -- --ramdisk_image: .long 0 # address of loaded ramdisk image -- # Here the loader puts the 32-bit -- # address where it loaded the image. -- # This only will be read by the kernel. -- --ramdisk_size: .long 0 # its size in bytes -- --bootsect_kludge: -- .long 0 # obsolete -- --heap_end_ptr: .word modelist+1024 # (Header version 0x0201 or later) -- # space from here (exclusive) down to -- # end of setup code can be used by setup -- # for local heap purposes. -- --pad1: .word 0 --cmd_line_ptr: .long 0 # (Header version 0x0202 or later) -- # If nonzero, a 32-bit pointer -- # to the kernel command line. -- # The command line should be -- # located between the start of -- # setup and the end of low -- # memory (0xa0000), or it may -- # get overwritten before it -- # gets read. If this field is -- # used, there is no longer -- # anything magical about the -- # 0x90000 segment; the setup -- # can be located anywhere in -- # low memory 0x10000 or higher. -- --ramdisk_max: .long (-__PAGE_OFFSET-(512 << 20)-1) & 0x7fffffff -- # (Header version 0x0203 or later) -- # The highest safe address for -- # the contents of an initrd -- --kernel_alignment: .long CONFIG_PHYSICAL_ALIGN #physical addr alignment -- #required for protected mode -- #kernel --#ifdef CONFIG_RELOCATABLE --relocatable_kernel: .byte 1 --#else --relocatable_kernel: .byte 0 --#endif --pad2: .byte 0 --pad3: .word 0 -- --cmdline_size: .long COMMAND_LINE_SIZE-1 #length of the command line, -- #added with boot protocol -- #version 2.06 -- --trampoline: call start_of_setup -- .align 16 -- # The offset at this point is 0x240 -- .space (0xeff-0x240+1) # E820 & EDD space (ending at 0xeff) --# End of setup header ##################################################### -- --start_of_setup: --# Bootlin depends on this being done early -- movw $0x01500, %ax -- movb $0x81, %dl -- int $0x13 -- --#ifdef SAFE_RESET_DISK_CONTROLLER --# Reset the disk controller. -- movw $0x0000, %ax -- movb $0x80, %dl -- int $0x13 --#endif -- --# Set %ds = %cs, we know that SETUPSEG = %cs at this point -- movw %cs, %ax # aka SETUPSEG -- movw %ax, %ds --# Check signature at end of setup -- cmpw $SIG1, setup_sig1 -- jne bad_sig -- -- cmpw $SIG2, setup_sig2 -- jne bad_sig -- -- jmp good_sig1 -- --# Routine to print asciiz string at ds:si --prtstr: -- lodsb -- andb %al, %al -- jz fin -- -- call prtchr -- jmp prtstr -- --fin: ret -- --# Space printing --prtsp2: call prtspc # Print double space --prtspc: movb $0x20, %al # Print single space (note: fall-thru) -- --# Part of above routine, this one just prints ascii al --prtchr: pushw %ax -- pushw %cx -- movw $7,%bx -- movw $0x01, %cx -- movb $0x0e, %ah -- int $0x10 -- popw %cx -- popw %ax -- ret -- --beep: movb $0x07, %al -- jmp prtchr -- --no_sig_mess: .string "No setup signature found ..." -- --good_sig1: -- jmp good_sig -- --# We now have to find the rest of the setup code/data --bad_sig: -- movw %cs, %ax # SETUPSEG -- subw $DELTA_INITSEG, %ax # INITSEG -- movw %ax, %ds -- xorb %bh, %bh -- movb (497), %bl # get setup sect from bootsect -- subw $4, %bx # LILO loads 4 sectors of setup -- shlw $8, %bx # convert to words (1sect=2^8 words) -- movw %bx, %cx -- shrw $3, %bx # convert to segment -- addw $SYSSEG, %bx -- movw %bx, %cs:start_sys_seg --# Move rest of setup code/data to here -- movw $2048, %di # four sectors loaded by LILO -- subw %si, %si -- pushw %cs -- popw %es -- movw $SYSSEG, %ax -- movw %ax, %ds -- rep -- movsw -- movw %cs, %ax # aka SETUPSEG -- movw %ax, %ds -- cmpw $SIG1, setup_sig1 -- jne no_sig -- -- cmpw $SIG2, setup_sig2 -- jne no_sig -- -- jmp good_sig -- --no_sig: -- lea no_sig_mess, %si -- call prtstr -- --no_sig_loop: -- hlt -- jmp no_sig_loop -- --good_sig: -- movw %cs, %ax # aka SETUPSEG -- subw $DELTA_INITSEG, %ax # aka INITSEG -- movw %ax, %ds --# Check if an old loader tries to load a big-kernel -- testb $LOADED_HIGH, %cs:loadflags # Do we have a big kernel? -- jz loader_ok # No, no danger for old loaders. -- -- cmpb $0, %cs:type_of_loader # Do we have a loader that -- # can deal with us? -- jnz loader_ok # Yes, continue. -- -- pushw %cs # No, we have an old loader, -- popw %ds # die. -- lea loader_panic_mess, %si -- call prtstr -- -- jmp no_sig_loop -- --loader_panic_mess: .string "Wrong loader, giving up..." -- --# check minimum cpuid --# we do this here because it is the last place we can actually --# show a user visible error message. Later the video modus --# might be already messed up. --loader_ok: -- call verify_cpu -- testl %eax,%eax -- jz cpu_ok -- movw %cs,%ax # aka SETUPSEG -- movw %ax,%ds -- lea cpu_panic_mess,%si -- call prtstr --1: jmp 1b -- --cpu_panic_mess: -- .asciz "PANIC: CPU too old for this kernel." -- --#include "../kernel/verify_cpu.S" -- --cpu_ok: --# Get memory size (extended mem, kB) -- -- xorl %eax, %eax -- movl %eax, (0x1e0) --#ifndef STANDARD_MEMORY_BIOS_CALL -- movb %al, (E820NR) --# Try three different memory detection schemes. First, try --# e820h, which lets us assemble a memory map, then try e801h, --# which returns a 32-bit memory size, and finally 88h, which --# returns 0-64m -- --# method E820H: --# the memory map from hell. e820h returns memory classified into --# a whole bunch of different types, and allows memory holes and --# everything. We scan through this memory map and build a list --# of the first 32 memory areas, which we return at [E820MAP]. --# This is documented at http://www.acpi.info/, in the ACPI 2.0 specification. -- --#define SMAP 0x534d4150 -- --meme820: -- xorl %ebx, %ebx # continuation counter -- movw $E820MAP, %di # point into the whitelist -- # so we can have the bios -- # directly write into it. -- --jmpe820: -- movl $0x0000e820, %eax # e820, upper word zeroed -- movl $SMAP, %edx # ascii 'SMAP' -- movl $20, %ecx # size of the e820rec -- pushw %ds # data record. -- popw %es -- int $0x15 # make the call -- jc bail820 # fall to e801 if it fails -- -- cmpl $SMAP, %eax # check the return is `SMAP' -- jne bail820 # fall to e801 if it fails -- --# cmpl $1, 16(%di) # is this usable memory? --# jne again820 -- -- # If this is usable memory, we save it by simply advancing %di by -- # sizeof(e820rec). -- # --good820: -- movb (E820NR), %al # up to 128 entries -- cmpb $E820MAX, %al -- jae bail820 -- -- incb (E820NR) -- movw %di, %ax -- addw $20, %ax -- movw %ax, %di --again820: -- cmpl $0, %ebx # check to see if -- jne jmpe820 # %ebx is set to EOF --bail820: -- -- --# method E801H: --# memory size is in 1k chunksizes, to avoid confusing loadlin. --# we store the 0xe801 memory size in a completely different place, --# because it will most likely be longer than 16 bits. --# (use 1e0 because that's what Larry Augustine uses in his --# alternative new memory detection scheme, and it's sensible --# to write everything into the same place.) -- --meme801: -- stc # fix to work around buggy -- xorw %cx,%cx # BIOSes which don't clear/set -- xorw %dx,%dx # carry on pass/error of -- # e801h memory size call -- # or merely pass cx,dx though -- # without changing them. -- movw $0xe801, %ax -- int $0x15 -- jc mem88 -- -- cmpw $0x0, %cx # Kludge to handle BIOSes -- jne e801usecxdx # which report their extended -- cmpw $0x0, %dx # memory in AX/BX rather than -- jne e801usecxdx # CX/DX. The spec I have read -- movw %ax, %cx # seems to indicate AX/BX -- movw %bx, %dx # are more reasonable anyway... -- --e801usecxdx: -- andl $0xffff, %edx # clear sign extend -- shll $6, %edx # and go from 64k to 1k chunks -- movl %edx, (0x1e0) # store extended memory size -- andl $0xffff, %ecx # clear sign extend -- addl %ecx, (0x1e0) # and add lower memory into -- # total size. -- --# Ye Olde Traditional Methode. Returns the memory size (up to 16mb or --# 64mb, depending on the bios) in ax. --mem88: -- --#endif -- movb $0x88, %ah -- int $0x15 -- movw %ax, (2) -- --# Set the keyboard repeat rate to the max -- movw $0x0305, %ax -- xorw %bx, %bx -- int $0x16 -- --# Check for video adapter and its parameters and allow the --# user to browse video modes. -- call video # NOTE: we need %ds pointing -- # to bootsector -- --# Get hd0 data... -- xorw %ax, %ax -- movw %ax, %ds -- ldsw (4 * 0x41), %si -- movw %cs, %ax # aka SETUPSEG -- subw $DELTA_INITSEG, %ax # aka INITSEG -- pushw %ax -- movw %ax, %es -- movw $0x0080, %di -- movw $0x10, %cx -- pushw %cx -- cld -- rep -- movsb --# Get hd1 data... -- xorw %ax, %ax -- movw %ax, %ds -- ldsw (4 * 0x46), %si -- popw %cx -- popw %es -- movw $0x0090, %di -- rep -- movsb --# Check that there IS a hd1 :-) -- movw $0x01500, %ax -- movb $0x81, %dl -- int $0x13 -- jc no_disk1 -- -- cmpb $3, %ah -- je is_disk1 -- --no_disk1: -- movw %cs, %ax # aka SETUPSEG -- subw $DELTA_INITSEG, %ax # aka INITSEG -- movw %ax, %es -- movw $0x0090, %di -- movw $0x10, %cx -- xorw %ax, %ax -- cld -- rep -- stosb --is_disk1: --# check for Micro Channel (MCA) bus -- movw %cs, %ax # aka SETUPSEG -- subw $DELTA_INITSEG, %ax # aka INITSEG -- movw %ax, %ds -- xorw %ax, %ax -- movw %ax, (0xa0) # set table length to 0 -- movb $0xc0, %ah -- stc -- int $0x15 # moves feature table to es:bx -- jc no_mca -- -- pushw %ds -- movw %es, %ax -- movw %ax, %ds -- movw %cs, %ax # aka SETUPSEG -- subw $DELTA_INITSEG, %ax # aka INITSEG -- movw %ax, %es -- movw %bx, %si -- movw $0xa0, %di -- movw (%si), %cx -- addw $2, %cx # table length is a short -- cmpw $0x10, %cx -- jc sysdesc_ok -- -- movw $0x10, %cx # we keep only first 16 bytes --sysdesc_ok: -- rep -- movsb -- popw %ds --no_mca: --#ifdef CONFIG_X86_VOYAGER -- movb $0xff, 0x40 # flag on config found -- movb $0xc0, %al -- mov $0xff, %ah -- int $0x15 # put voyager config info at es:di -- jc no_voyager -- movw $0x40, %si # place voyager info in apm table -- cld -- movw $7, %cx --voyager_rep: -- movb %es:(%di), %al -- movb %al,(%si) -- incw %di -- incw %si -- decw %cx -- jnz voyager_rep --no_voyager: --#endif --# Check for PS/2 pointing device -- movw %cs, %ax # aka SETUPSEG -- subw $DELTA_INITSEG, %ax # aka INITSEG -- movw %ax, %ds -- movb $0, (0x1ff) # default is no pointing device -- int $0x11 # int 0x11: equipment list -- testb $0x04, %al # check if mouse installed -- jz no_psmouse -- -- movb $0xAA, (0x1ff) # device present --no_psmouse: -- --#if defined(CONFIG_X86_SPEEDSTEP_SMI) || defined(CONFIG_X86_SPEEDSTEP_SMI_MODULE) -- movl $0x0000E980, %eax # IST Support -- movl $0x47534943, %edx # Request value -- int $0x15 -- -- movl %eax, (96) -- movl %ebx, (100) -- movl %ecx, (104) -- movl %edx, (108) --#endif -- --#if defined(CONFIG_APM) || defined(CONFIG_APM_MODULE) --# Then check for an APM BIOS... -- # %ds points to the bootsector -- movw $0, 0x40 # version = 0 means no APM BIOS -- movw $0x05300, %ax # APM BIOS installation check -- xorw %bx, %bx -- int $0x15 -- jc done_apm_bios # Nope, no APM BIOS -- -- cmpw $0x0504d, %bx # Check for "PM" signature -- jne done_apm_bios # No signature, no APM BIOS -- -- andw $0x02, %cx # Is 32 bit supported? -- je done_apm_bios # No 32-bit, no (good) APM BIOS -- -- movw $0x05304, %ax # Disconnect first just in case -- xorw %bx, %bx -- int $0x15 # ignore return code -- movw $0x05303, %ax # 32 bit connect -- xorl %ebx, %ebx -- xorw %cx, %cx # paranoia :-) -- xorw %dx, %dx # ... -- xorl %esi, %esi # ... -- xorw %di, %di # ... -- int $0x15 -- jc no_32_apm_bios # Ack, error. -- -- movw %ax, (66) # BIOS code segment -- movl %ebx, (68) # BIOS entry point offset -- movw %cx, (72) # BIOS 16 bit code segment -- movw %dx, (74) # BIOS data segment -- movl %esi, (78) # BIOS code segment lengths -- movw %di, (82) # BIOS data segment length --# Redo the installation check as the 32 bit connect --# modifies the flags returned on some BIOSs -- movw $0x05300, %ax # APM BIOS installation check -- xorw %bx, %bx -- xorw %cx, %cx # paranoia -- int $0x15 -- jc apm_disconnect # error -> shouldn't happen -- -- cmpw $0x0504d, %bx # check for "PM" signature -- jne apm_disconnect # no sig -> shouldn't happen -- -- movw %ax, (64) # record the APM BIOS version -- movw %cx, (76) # and flags -- jmp done_apm_bios -- --apm_disconnect: # Tidy up -- movw $0x05304, %ax # Disconnect -- xorw %bx, %bx -- int $0x15 # ignore return code -- -- jmp done_apm_bios -- --no_32_apm_bios: -- andw $0xfffd, (76) # remove 32 bit support bit --done_apm_bios: --#endif -- --#include "edd.S" -- --# Now we want to move to protected mode ... -- cmpw $0, %cs:realmode_swtch -- jz rmodeswtch_normal -- -- lcall *%cs:realmode_swtch -- -- jmp rmodeswtch_end -- --rmodeswtch_normal: -- pushw %cs -- call default_switch -- --rmodeswtch_end: --# Now we move the system to its rightful place ... but we check if we have a --# big-kernel. In that case we *must* not move it ... -- testb $LOADED_HIGH, %cs:loadflags -- jz do_move0 # .. then we have a normal low -- # loaded zImage -- # .. or else we have a high -- # loaded bzImage -- jmp end_move # ... and we skip moving -- --do_move0: -- movw $0x100, %ax # start of destination segment -- movw %cs, %bp # aka SETUPSEG -- subw $DELTA_INITSEG, %bp # aka INITSEG -- movw %cs:start_sys_seg, %bx # start of source segment -- cld --do_move: -- movw %ax, %es # destination segment -- incb %ah # instead of add ax,#0x100 -- movw %bx, %ds # source segment -- addw $0x100, %bx -- subw %di, %di -- subw %si, %si -- movw $0x800, %cx -- rep -- movsw -- cmpw %bp, %bx # assume start_sys_seg > 0x200, -- # so we will perhaps read one -- # page more than needed, but -- # never overwrite INITSEG -- # because destination is a -- # minimum one page below source -- jb do_move -- --end_move: --# then we load the segment descriptors -- movw %cs, %ax # aka SETUPSEG -- movw %ax, %ds -- --# Check whether we need to be downward compatible with version <=201 -- cmpl $0, cmd_line_ptr -- jne end_move_self # loader uses version >=202 features -- cmpb $0x20, type_of_loader -- je end_move_self # bootsect loader, we know of it -- --# Boot loader doesnt support boot protocol version 2.02. --# If we have our code not at 0x90000, we need to move it there now. --# We also then need to move the params behind it (commandline) --# Because we would overwrite the code on the current IP, we move --# it in two steps, jumping high after the first one. -- movw %cs, %ax -- cmpw $SETUPSEG, %ax -- je end_move_self -- -- cli # make sure we really have -- # interrupts disabled ! -- # because after this the stack -- # should not be used -- subw $DELTA_INITSEG, %ax # aka INITSEG -- movw %ss, %dx -- cmpw %ax, %dx -- jb move_self_1 -- -- addw $INITSEG, %dx -- subw %ax, %dx # this will go into %ss after -- # the move --move_self_1: -- movw %ax, %ds -- movw $INITSEG, %ax # real INITSEG -- movw %ax, %es -- movw %cs:setup_move_size, %cx -- std # we have to move up, so we use -- # direction down because the -- # areas may overlap -- movw %cx, %di -- decw %di -- movw %di, %si -- subw $move_self_here+0x200, %cx -- rep -- movsb -- ljmp $SETUPSEG, $move_self_here -- --move_self_here: -- movw $move_self_here+0x200, %cx -- rep -- movsb -- movw $SETUPSEG, %ax -- movw %ax, %ds -- movw %dx, %ss --end_move_self: # now we are at the right place -- --# --# Enable A20. This is at the very best an annoying procedure. --# A20 code ported from SYSLINUX 1.52-1.63 by H. Peter Anvin. --# AMD Elan bug fix by Robert Schwebel. --# -- --#if defined(CONFIG_X86_ELAN) -- movb $0x02, %al # alternate A20 gate -- outb %al, $0x92 # this works on SC410/SC520 --a20_elan_wait: -- call a20_test -- jz a20_elan_wait -- jmp a20_done --#endif -- -- --A20_TEST_LOOPS = 32 # Iterations per wait --A20_ENABLE_LOOPS = 255 # Total loops to try -- -- --#ifndef CONFIG_X86_VOYAGER --a20_try_loop: -- -- # First, see if we are on a system with no A20 gate. --a20_none: -- call a20_test -- jnz a20_done -- -- # Next, try the BIOS (INT 0x15, AX=0x2401) --a20_bios: -- movw $0x2401, %ax -- pushfl # Be paranoid about flags -- int $0x15 -- popfl -- -- call a20_test -- jnz a20_done -- -- # Try enabling A20 through the keyboard controller --#endif /* CONFIG_X86_VOYAGER */ --a20_kbc: -- call empty_8042 -- --#ifndef CONFIG_X86_VOYAGER -- call a20_test # Just in case the BIOS worked -- jnz a20_done # but had a delayed reaction. --#endif -- -- movb $0xD1, %al # command write -- outb %al, $0x64 -- call empty_8042 -- -- movb $0xDF, %al # A20 on -- outb %al, $0x60 -- call empty_8042 -- --#ifndef CONFIG_X86_VOYAGER -- # Wait until a20 really *is* enabled; it can take a fair amount of -- # time on certain systems; Toshiba Tecras are known to have this -- # problem. --a20_kbc_wait: -- xorw %cx, %cx --a20_kbc_wait_loop: -- call a20_test -- jnz a20_done -- loop a20_kbc_wait_loop -- -- # Final attempt: use "configuration port A" --a20_fast: -- inb $0x92, %al # Configuration Port A -- orb $0x02, %al # "fast A20" version -- andb $0xFE, %al # don't accidentally reset -- outb %al, $0x92 -- -- # Wait for configuration port A to take effect --a20_fast_wait: -- xorw %cx, %cx --a20_fast_wait_loop: -- call a20_test -- jnz a20_done -- loop a20_fast_wait_loop -- -- # A20 is still not responding. Try frobbing it again. -- # -- decb (a20_tries) -- jnz a20_try_loop -- -- movw $a20_err_msg, %si -- call prtstr -- --a20_die: -- hlt -- jmp a20_die -- --a20_tries: -- .byte A20_ENABLE_LOOPS -- --a20_err_msg: -- .ascii "linux: fatal error: A20 gate not responding!" -- .byte 13, 10, 0 -- -- # If we get here, all is good --a20_done: -- --#endif /* CONFIG_X86_VOYAGER */ --# set up gdt and idt and 32bit start address -- lidt idt_48 # load idt with 0,0 -- xorl %eax, %eax # Compute gdt_base -- movw %ds, %ax # (Convert %ds:gdt to a linear ptr) -- shll $4, %eax -- addl %eax, code32 -- addl $gdt, %eax -- movl %eax, (gdt_48+2) -- lgdt gdt_48 # load gdt with whatever is -- # appropriate -- --# make sure any possible coprocessor is properly reset.. -- xorw %ax, %ax -- outb %al, $0xf0 -- call delay -- -- outb %al, $0xf1 -- call delay -- --# well, that went ok, I hope. Now we mask all interrupts - the rest --# is done in init_IRQ(). -- movb $0xFF, %al # mask all interrupts for now -- outb %al, $0xA1 -- call delay -- -- movb $0xFB, %al # mask all irq's but irq2 which -- outb %al, $0x21 # is cascaded -- --# Well, that certainly wasn't fun :-(. Hopefully it works, and we don't --# need no steenking BIOS anyway (except for the initial loading :-). --# The BIOS-routine wants lots of unnecessary data, and it's less --# "interesting" anyway. This is how REAL programmers do it. --# --# Well, now's the time to actually move into protected mode. To make --# things as simple as possible, we do no register set-up or anything, --# we let the gnu-compiled 32-bit programs do that. We just jump to --# absolute address 0x1000 (or the loader supplied one), --# in 32-bit protected mode. --# --# Note that the short jump isn't strictly needed, although there are --# reasons why it might be a good idea. It won't hurt in any case. -- movw $1, %ax # protected mode (PE) bit -- lmsw %ax # This is it! -- jmp flush_instr -- --flush_instr: -- xorw %bx, %bx # Flag to indicate a boot -- xorl %esi, %esi # Pointer to real-mode code -- movw %cs, %si -- subw $DELTA_INITSEG, %si -- shll $4, %esi # Convert to 32-bit pointer -- --# jump to startup_32 in arch/i386/boot/compressed/head.S --# --# NOTE: For high loaded big kernels we need a --# jmpi 0x100000,__BOOT_CS --# --# but we yet haven't reloaded the CS register, so the default size --# of the target offset still is 16 bit. --# However, using an operand prefix (0x66), the CPU will properly --# take our 48 bit far pointer. (INTeL 80386 Programmer's Reference --# Manual, Mixing 16-bit and 32-bit code, page 16-6) -- -- .byte 0x66, 0xea # prefix + jmpi-opcode --code32: .long startup_32 # will be set to %cs+startup_32 -- .word __BOOT_CS --.code32 --startup_32: -- movl $(__BOOT_DS), %eax -- movl %eax, %ds -- movl %eax, %es -- movl %eax, %fs -- movl %eax, %gs -- movl %eax, %ss -- -- xorl %eax, %eax --1: incl %eax # check that A20 really IS enabled -- movl %eax, 0x00000000 # loop forever if it isn't -- cmpl %eax, 0x00100000 -- je 1b -- -- # Jump to the 32bit entry point -- jmpl *(code32_start - start + (DELTA_INITSEG << 4))(%esi) --.code16 -- --# Here's a bunch of information about your current kernel.. --kernel_version: .ascii UTS_RELEASE -- .ascii " (" -- .ascii LINUX_COMPILE_BY -- .ascii "@" -- .ascii LINUX_COMPILE_HOST -- .ascii ") " -- .ascii UTS_VERSION -- .byte 0 -- --# This is the default real mode switch routine. --# to be called just before protected mode transition --default_switch: -- cli # no interrupts allowed ! -- movb $0x80, %al # disable NMI for bootup -- # sequence -- outb %al, $0x70 -- lret -- -- --#ifndef CONFIG_X86_VOYAGER --# This routine tests whether or not A20 is enabled. If so, it --# exits with zf = 0. --# --# The memory address used, 0x200, is the int $0x80 vector, which --# should be safe. -- --A20_TEST_ADDR = 4*0x80 -- --a20_test: -- pushw %cx -- pushw %ax -- xorw %cx, %cx -- movw %cx, %fs # Low memory -- decw %cx -- movw %cx, %gs # High memory area -- movw $A20_TEST_LOOPS, %cx -- movw %fs:(A20_TEST_ADDR), %ax -- pushw %ax --a20_test_wait: -- incw %ax -- movw %ax, %fs:(A20_TEST_ADDR) -- call delay # Serialize and make delay constant -- cmpw %gs:(A20_TEST_ADDR+0x10), %ax -- loope a20_test_wait -- -- popw %fs:(A20_TEST_ADDR) -- popw %ax -- popw %cx -- ret -- --#endif /* CONFIG_X86_VOYAGER */ -- --# This routine checks that the keyboard command queue is empty --# (after emptying the output buffers) --# --# Some machines have delusions that the keyboard buffer is always full --# with no keyboard attached... --# --# If there is no keyboard controller, we will usually get 0xff --# to all the reads. With each IO taking a microsecond and --# a timeout of 100,000 iterations, this can take about half a --# second ("delay" == outb to port 0x80). That should be ok, --# and should also be plenty of time for a real keyboard controller --# to empty. --# -- --empty_8042: -- pushl %ecx -- movl $100000, %ecx -- --empty_8042_loop: -- decl %ecx -- jz empty_8042_end_loop -- -- call delay -- -- inb $0x64, %al # 8042 status port -- testb $1, %al # output buffer? -- jz no_output -- -- call delay -- inb $0x60, %al # read it -- jmp empty_8042_loop -- --no_output: -- testb $2, %al # is input buffer full? -- jnz empty_8042_loop # yes - loop --empty_8042_end_loop: -- popl %ecx -- ret -- --# Read the cmos clock. Return the seconds in al --gettime: -- pushw %cx -- movb $0x02, %ah -- int $0x1a -- movb %dh, %al # %dh contains the seconds -- andb $0x0f, %al -- movb %dh, %ah -- movb $0x04, %cl -- shrb %cl, %ah -- aad -- popw %cx -- ret -- --# Delay is needed after doing I/O --delay: -- outb %al,$0x80 -- ret -- --# Descriptor tables --# --# NOTE: The intel manual says gdt should be sixteen bytes aligned for --# efficiency reasons. However, there are machines which are known not --# to boot with misaligned GDTs, so alter this at your peril! If you alter --# GDT_ENTRY_BOOT_CS (in asm/segment.h) remember to leave at least two --# empty GDT entries (one for NULL and one reserved). --# --# NOTE: On some CPUs, the GDT must be 8 byte aligned. This is --# true for the Voyager Quad CPU card which will not boot without --# This directive. 16 byte aligment is recommended by intel. --# -- .align 16 --gdt: -- .fill GDT_ENTRY_BOOT_CS,8,0 -- -- .word 0xFFFF # 4Gb - (0x100000*0x1000 = 4Gb) -- .word 0 # base address = 0 -- .word 0x9A00 # code read/exec -- .word 0x00CF # granularity = 4096, 386 -- # (+5th nibble of limit) -- -- .word 0xFFFF # 4Gb - (0x100000*0x1000 = 4Gb) -- .word 0 # base address = 0 -- .word 0x9200 # data read/write -- .word 0x00CF # granularity = 4096, 386 -- # (+5th nibble of limit) --gdt_end: -- .align 4 -- -- .word 0 # alignment byte --idt_48: -- .word 0 # idt limit = 0 -- .word 0, 0 # idt base = 0L -- -- .word 0 # alignment byte --gdt_48: -- .word gdt_end - gdt - 1 # gdt limit -- .word 0, 0 # gdt base (filled in later) -- --# Include video setup & detection code -- --#include "video.S" -- --# Setup signature -- must be last --setup_sig1: .word SIG1 --setup_sig2: .word SIG2 -- --# After this point, there is some free space which is used by the video mode --# handling code to store the temporary mode table (not used by the kernel). -- --modelist: -- --.text --endtext: --.data --enddata: --.bss --endbss: ---- /dev/null -+++ b/arch/i386/boot/setup.ld -@@ -0,0 +1,54 @@ -+/* -+ * setup.ld -+ * -+ * Linker script for the i386 setup code -+ */ -+OUTPUT_FORMAT("elf32-i386", "elf32-i386", "elf32-i386") -+OUTPUT_ARCH(i386) -+ENTRY(_start) -+ -+SECTIONS -+{ -+ . = 0; -+ .bstext : { *(.bstext) } -+ .bsdata : { *(.bsdata) } -+ -+ . = 497; -+ .header : { *(.header) } -+ .inittext : { *(.inittext) } -+ .initdata : { *(.initdata) } -+ .text : { *(.text*) } -+ -+ . = ALIGN(16); -+ .rodata : { *(.rodata*) } -+ -+ .videocards : { -+ video_cards = .; -+ *(.videocards) -+ video_cards_end = .; -+ } -+ -+ . = ALIGN(16); -+ .data : { *(.data*) } -+ -+ .signature : { -+ setup_sig = .; -+ LONG(0x5a5aaa55) -+ } -+ -+ -+ . = ALIGN(16); -+ .bss : -+ { -+ __bss_start = .; -+ *(.bss) -+ __bss_end = .; -+ } -+ . = ALIGN(16); -+ _end = .; -+ -+ /DISCARD/ : { *(.note*) } -+ -+ . = ASSERT(_end <= 0x8000, "Setup too big!"); -+ . = ASSERT(hdr == 0x1f1, "The setup header has the wrong offset!"); -+} ---- /dev/null -+++ b/arch/i386/boot/string.c -@@ -0,0 +1,34 @@ -+/* -*- linux-c -*- ------------------------------------------------------- * -+ * -+ * Copyright (C) 1991, 1992 Linus Torvalds -+ * Copyright 2007 rPath, Inc. - All Rights Reserved -+ * -+ * This file is part of the Linux kernel, and is made available under -+ * the terms of the GNU General Public License version 2. -+ * -+ * ----------------------------------------------------------------------- */ -+ -+/* -+ * arch/i386/boot/string.c -+ * -+ * Very basic string functions -+ */ -+ -+#include "boot.h" -+#include <linux/edd.h> -+ -+int strcmp(const char *str1, const char *str2) -+{ -+ const unsigned char *s1 = (const unsigned char *)str1; -+ const unsigned char *s2 = (const unsigned char *)str2; -+ int delta = 0; -+ -+ while (*s1 || *s2) { -+ delta = *s2 - *s1; -+ if (delta) -+ return delta; -+ s1++; -+ s2++; -+ } -+ return 0; -+} ---- a/arch/i386/boot/tools/build.c -+++ b/arch/i386/boot/tools/build.c -@@ -1,13 +1,12 @@ - /* - * Copyright (C) 1991, 1992 Linus Torvalds - * Copyright (C) 1997 Martin Mares -+ * Copyright (C) 2007 H. Peter Anvin - */ - - /* - * This file builds a disk-image from three different files: - * -- * - bootsect: compatibility mbr which prints an error message if -- * someone tries to boot the kernel directly. - * - setup: 8086 machine code, sets up system parm - * - system: 80386 code for actual system - * -@@ -21,6 +20,7 @@ - * High loaded stuff by Hans Lermen & Werner Almesberger, Feb. 1996 - * Cross compiling fixes by Gertjan van Wingerde, July 1996 - * Rewritten by Martin Mares, April 1997 -+ * Substantially overhauled by H. Peter Anvin, April 2007 - */ - - #include <stdio.h> -@@ -32,23 +32,25 @@ - #include <sys/sysmacros.h> - #include <unistd.h> - #include <fcntl.h> -+#include <sys/mman.h> - #include <asm/boot.h> - --typedef unsigned char byte; --typedef unsigned short word; --typedef unsigned long u32; -+typedef unsigned char u8; -+typedef unsigned short u16; -+typedef unsigned long u32; - - #define DEFAULT_MAJOR_ROOT 0 - #define DEFAULT_MINOR_ROOT 0 - --/* Minimal number of setup sectors (see also bootsect.S) */ --#define SETUP_SECTS 4 -+/* Minimal number of setup sectors */ -+#define SETUP_SECT_MIN 5 -+#define SETUP_SECT_MAX 64 - --byte buf[1024]; --int fd; -+/* This must be large enough to hold the entire setup */ -+u8 buf[SETUP_SECT_MAX*512]; - int is_big_kernel; - --void die(const char * str, ...) -+static void die(const char * str, ...) - { - va_list args; - va_start(args, str); -@@ -57,15 +59,9 @@ - exit(1); - } - --void file_open(const char *name) -+static void usage(void) - { -- if ((fd = open(name, O_RDONLY, 0)) < 0) -- die("Unable to open `%s': %m", name); --} -- --void usage(void) --{ -- die("Usage: build [-b] bootsect setup system [rootdev] [> image]"); -+ die("Usage: build [-b] setup system [rootdev] [> image]"); - } - - int main(int argc, char ** argv) -@@ -73,27 +69,30 @@ - unsigned int i, sz, setup_sectors; - int c; - u32 sys_size; -- byte major_root, minor_root; -+ u8 major_root, minor_root; - struct stat sb; -+ FILE *file; -+ int fd; -+ void *kernel; - - if (argc > 2 && !strcmp(argv[1], "-b")) - { - is_big_kernel = 1; - argc--, argv++; - } -- if ((argc < 4) || (argc > 5)) -+ if ((argc < 3) || (argc > 4)) - usage(); -- if (argc > 4) { -- if (!strcmp(argv[4], "CURRENT")) { -+ if (argc > 3) { -+ if (!strcmp(argv[3], "CURRENT")) { - if (stat("/", &sb)) { - perror("/"); - die("Couldn't stat /"); - } - major_root = major(sb.st_dev); - minor_root = minor(sb.st_dev); -- } else if (strcmp(argv[4], "FLOPPY")) { -- if (stat(argv[4], &sb)) { -- perror(argv[4]); -+ } else if (strcmp(argv[3], "FLOPPY")) { -+ if (stat(argv[3], &sb)) { -+ perror(argv[3]); - die("Couldn't stat root device."); - } - major_root = major(sb.st_rdev); -@@ -108,79 +107,62 @@ - } - fprintf(stderr, "Root device is (%d, %d)\n", major_root, minor_root); - -- file_open(argv[1]); -- i = read(fd, buf, sizeof(buf)); -- fprintf(stderr,"Boot sector %d bytes.\n",i); -- if (i != 512) -- die("Boot block must be exactly 512 bytes"); -+ /* Copy the setup code */ -+ file = fopen(argv[1], "r"); -+ if (!file) -+ die("Unable to open `%s': %m", argv[1]); -+ c = fread(buf, 1, sizeof(buf), file); -+ if (ferror(file)) -+ die("read-error on `setup'"); -+ if (c < 1024) -+ die("The setup must be at least 1024 bytes"); - if (buf[510] != 0x55 || buf[511] != 0xaa) - die("Boot block hasn't got boot flag (0xAA55)"); -+ fclose(file); -+ -+ /* Pad unused space with zeros */ -+ setup_sectors = (c + 511) / 512; -+ if (setup_sectors < SETUP_SECT_MIN) -+ setup_sectors = SETUP_SECT_MIN; -+ i = setup_sectors*512; -+ memset(buf+c, 0, i-c); -+ -+ /* Set the default root device */ - buf[508] = minor_root; - buf[509] = major_root; -- if (write(1, buf, 512) != 512) -- die("Write call failed"); -- close (fd); -- -- file_open(argv[2]); /* Copy the setup code */ -- for (i=0 ; (c=read(fd, buf, sizeof(buf)))>0 ; i+=c ) -- if (write(1, buf, c) != c) -- die("Write call failed"); -- if (c != 0) -- die("read-error on `setup'"); -- close (fd); - -- setup_sectors = (i + 511) / 512; /* Pad unused space with zeros */ -- /* for compatibility with ancient versions of LILO. */ -- if (setup_sectors < SETUP_SECTS) -- setup_sectors = SETUP_SECTS; -- fprintf(stderr, "Setup is %d bytes.\n", i); -- memset(buf, 0, sizeof(buf)); -- while (i < setup_sectors * 512) { -- c = setup_sectors * 512 - i; -- if (c > sizeof(buf)) -- c = sizeof(buf); -- if (write(1, buf, c) != c) -- die("Write call failed"); -- i += c; -- } -+ fprintf(stderr, "Setup is %d bytes (padded to %d bytes).\n", c, i); - -- file_open(argv[3]); -- if (fstat (fd, &sb)) -- die("Unable to stat `%s': %m", argv[3]); -+ /* Open and stat the kernel file */ -+ fd = open(argv[2], O_RDONLY); -+ if (fd < 0) -+ die("Unable to open `%s': %m", argv[2]); -+ if (fstat(fd, &sb)) -+ die("Unable to stat `%s': %m", argv[2]); - sz = sb.st_size; -- fprintf (stderr, "System is %d kB\n", sz/1024); -+ fprintf (stderr, "System is %d kB\n", (sz+1023)/1024); -+ kernel = mmap(NULL, sz, PROT_READ, MAP_SHARED, fd, 0); -+ if (kernel == MAP_FAILED) -+ die("Unable to mmap '%s': %m", argv[2]); - sys_size = (sz + 15) / 16; - if (!is_big_kernel && sys_size > DEF_SYSSIZE) - die("System is too big. Try using bzImage or modules."); -- while (sz > 0) { -- int l, n; - -- l = (sz > sizeof(buf)) ? sizeof(buf) : sz; -- if ((n=read(fd, buf, l)) != l) { -- if (n < 0) -- die("Error reading %s: %m", argv[3]); -- else -- die("%s: Unexpected EOF", argv[3]); -- } -- if (write(1, buf, l) != l) -- die("Write failed"); -- sz -= l; -- } -+ /* Patch the setup code with the appropriate size parameters */ -+ buf[0x1f1] = setup_sectors-1; -+ buf[0x1f4] = sys_size; -+ buf[0x1f5] = sys_size >> 8; -+ buf[0x1f6] = sys_size >> 16; -+ buf[0x1f7] = sys_size >> 24; -+ -+ if (fwrite(buf, 1, i, stdout) != i) -+ die("Writing setup failed"); -+ -+ /* Copy the kernel code */ -+ if (fwrite(kernel, 1, sz, stdout) != sz) -+ die("Writing kernel failed"); - close(fd); - -- if (lseek(1, 497, SEEK_SET) != 497) /* Write sizes to the bootsector */ -- die("Output: seek failed"); -- buf[0] = setup_sectors; -- if (write(1, buf, 1) != 1) -- die("Write of setup sector count failed"); -- if (lseek(1, 500, SEEK_SET) != 500) -- die("Output: seek failed"); -- buf[0] = (sys_size & 0xff); -- buf[1] = ((sys_size >> 8) & 0xff); -- buf[2] = ((sys_size >> 16) & 0xff); -- buf[3] = ((sys_size >> 24) & 0xff); -- if (write(1, buf, 4) != 4) -- die("Write of image length failed"); -- -- return 0; /* Everything is OK */ -+ /* Everything is OK */ -+ return 0; - } ---- /dev/null -+++ b/arch/i386/boot/tty.c -@@ -0,0 +1,112 @@ -+/* -*- linux-c -*- ------------------------------------------------------- * -+ * -+ * Copyright (C) 1991, 1992 Linus Torvalds -+ * Copyright 2007 rPath, Inc. - All Rights Reserved -+ * -+ * This file is part of the Linux kernel, and is made available under -+ * the terms of the GNU General Public License version 2. -+ * -+ * ----------------------------------------------------------------------- */ -+ -+/* -+ * arch/i386/boot/tty.c -+ * -+ * Very simple screen I/O -+ * XXX: Probably should add very simple serial I/O? -+ */ -+ -+#include "boot.h" -+ -+/* -+ * These functions are in .inittext so they can be used to signal -+ * error during initialization. -+ */ -+ -+void __attribute__((section(".inittext"))) putchar(int ch) -+{ -+ unsigned char c = ch; -+ -+ if (c == '\n') -+ putchar('\r'); /* \n -> \r\n */ -+ -+ /* int $0x10 is known to have bugs involving touching registers -+ it shouldn't. Be extra conservative... */ -+ asm volatile("pushal; int $0x10; popal" -+ : : "b" (0x0007), "c" (0x0001), "a" (0x0e00|ch)); -+} -+ -+void __attribute__((section(".inittext"))) puts(const char *str) -+{ -+ int n = 0; -+ while (*str) { -+ putchar(*str++); -+ n++; -+ } -+} -+ -+/* -+ * Read the CMOS clock through the BIOS, and return the -+ * seconds in BCD. -+ */ -+ -+static u8 gettime(void) -+{ -+ u16 ax = 0x0200; -+ u16 cx, dx; -+ -+ asm("int $0x1a" -+ : "+a" (ax), "=c" (cx), "=d" (dx) -+ : : "ebx", "esi", "edi"); -+ -+ return dx >> 8; -+} -+ -+/* -+ * Read from the keyboard -+ */ -+int getchar(void) -+{ -+ u16 ax = 0; -+ asm("int $0x16" : "+a" (ax)); -+ -+ return ax & 0xff; -+} -+ -+static int kbd_pending(void) -+{ -+ u8 pending; -+ asm("int $0x16; setnz %0" -+ : "=rm" (pending) -+ : "a" (0x0100)); -+ return pending; -+} -+ -+void kbd_flush(void) -+{ -+ for (;;) { -+ if (!kbd_pending()) -+ break; -+ getchar(); -+ } -+} -+ -+int getchar_timeout(void) -+{ -+ int cnt = 30; -+ int t0, t1; -+ -+ t0 = gettime(); -+ -+ while (cnt) { -+ if (kbd_pending()) -+ return getchar(); -+ -+ t1 = gettime(); -+ if (t0 != t1) { -+ cnt--; -+ t0 = t1; -+ } -+ } -+ -+ return 0; /* Timeout! */ -+} ---- /dev/null -+++ b/arch/i386/boot/version.c -@@ -0,0 +1,23 @@ -+/* -*- linux-c -*- ------------------------------------------------------- * -+ * -+ * Copyright (C) 1991, 1992 Linus Torvalds -+ * Copyright 2007 rPath, Inc. - All Rights Reserved -+ * -+ * This file is part of the Linux kernel, and is made available under -+ * the terms of the GNU General Public License version 2. -+ * -+ * ----------------------------------------------------------------------- */ -+ -+/* -+ * arch/i386/boot/version.c -+ * -+ * Kernel version string -+ */ -+ -+#include "boot.h" -+#include <linux/utsrelease.h> -+#include <linux/compile.h> -+ -+const char kernel_version[] = -+ UTS_RELEASE " (" LINUX_COMPILE_BY "@" LINUX_COMPILE_HOST ") " -+ UTS_VERSION; ---- /dev/null -+++ b/arch/i386/boot/vesa.h -@@ -0,0 +1,79 @@ -+/* ----------------------------------------------------------------------- * -+ * -+ * Copyright 1999-2007 H. Peter Anvin - All Rights Reserved -+ * -+ * This program is free software; you can redistribute it and/or modify -+ * it under the terms of the GNU General Public License as published by -+ * the Free Software Foundation, Inc., 53 Temple Place Ste 330, -+ * Boston MA 02111-1307, USA; either version 2 of the License, or -+ * (at your option) any later version; incorporated herein by reference. -+ * -+ * ----------------------------------------------------------------------- */ -+ -+#ifndef BOOT_VESA_H -+#define BOOT_VESA_H -+ -+typedef struct { -+ u16 off, seg; -+} far_ptr; -+ -+/* VESA General Information table */ -+struct vesa_general_info { -+ u32 signature; /* 0 Magic number = "VESA" */ -+ u16 version; /* 4 */ -+ far_ptr vendor_string; /* 6 */ -+ u32 capabilities; /* 10 */ -+ far_ptr video_mode_ptr; /* 14 */ -+ u16 total_memory; /* 18 */ -+ -+ u16 oem_software_rev; /* 20 */ -+ far_ptr oem_vendor_name_ptr; /* 22 */ -+ far_ptr oem_product_name_ptr; /* 26 */ -+ far_ptr oem_product_rev_ptr; /* 30 */ -+ -+ u8 reserved[222]; /* 34 */ -+ u8 oem_data[256]; /* 256 */ -+} __attribute__((packed)); -+ -+#define VESA_MAGIC ('V' + ('E' << 8) + ('S' << 16) + ('A' << 24)) -+#define VBE2_MAGIC ('V' + ('B' << 8) + ('E' << 16) + ('2' << 24)) -+ -+struct vesa_mode_info { -+ u16 mode_attr; /* 0 */ -+ u8 win_attr[2]; /* 2 */ -+ u16 win_grain; /* 4 */ -+ u16 win_size; /* 6 */ -+ u16 win_seg[2]; /* 8 */ -+ far_ptr win_scheme; /* 12 */ -+ u16 logical_scan; /* 16 */ -+ -+ u16 h_res; /* 18 */ -+ u16 v_res; /* 20 */ -+ u8 char_width; /* 22 */ -+ u8 char_height; /* 23 */ -+ u8 memory_planes; /* 24 */ -+ u8 bpp; /* 25 */ -+ u8 banks; /* 26 */ -+ u8 memory_layout; /* 27 */ -+ u8 bank_size; /* 28 */ -+ u8 image_planes; /* 29 */ -+ u8 page_function; /* 30 */ -+ -+ u8 rmask; /* 31 */ -+ u8 rpos; /* 32 */ -+ u8 gmask; /* 33 */ -+ u8 gpos; /* 34 */ -+ u8 bmask; /* 35 */ -+ u8 bpos; /* 36 */ -+ u8 resv_mask; /* 37 */ -+ u8 resv_pos; /* 38 */ -+ u8 dcm_info; /* 39 */ -+ -+ u32 lfb_ptr; /* 40 Linear frame buffer address */ -+ u32 offscreen_ptr; /* 44 Offscreen memory address */ -+ u16 offscreen_size; /* 48 */ -+ -+ u8 reserved[206]; /* 50 */ -+} __attribute__((packed)); -+ -+#endif /* LIB_SYS_VESA_H */ ---- /dev/null -+++ b/arch/i386/boot/video-bios.c -@@ -0,0 +1,125 @@ -+/* -*- linux-c -*- ------------------------------------------------------- * -+ * -+ * Copyright (C) 1991, 1992 Linus Torvalds -+ * Copyright 2007 rPath, Inc. - All Rights Reserved -+ * -+ * This file is part of the Linux kernel, and is made available under -+ * the terms of the GNU General Public License version 2. -+ * -+ * ----------------------------------------------------------------------- */ -+ -+/* -+ * arch/i386/boot/video-bios.c -+ * -+ * Standard video BIOS modes -+ * -+ * We have two options for this; silent and scanned. -+ */ -+ -+#include "boot.h" -+#include "video.h" -+ -+__videocard video_bios; -+ -+/* Set a conventional BIOS mode */ -+static int set_bios_mode(u8 mode); -+ -+static int bios_set_mode(struct mode_info *mi) -+{ -+ return set_bios_mode(mi->mode - VIDEO_FIRST_BIOS); -+} -+ -+static int set_bios_mode(u8 mode) -+{ -+ u16 ax; -+ u8 new_mode; -+ -+ ax = mode; /* AH=0x00 Set Video Mode */ -+ asm volatile(INT10 -+ : "+a" (ax) -+ : : "ebx", "ecx", "edx", "esi", "edi"); -+ -+ ax = 0x0f00; /* Get Current Video Mode */ -+ asm volatile(INT10 -+ : "+a" (ax) -+ : : "ebx", "ecx", "edx", "esi", "edi"); -+ -+ do_restore = 1; /* Assume video contents was lost */ -+ new_mode = ax & 0x7f; /* Not all BIOSes are clean with the top bit */ -+ -+ if (new_mode == mode) -+ return 0; /* Mode change OK */ -+ -+ if (new_mode != boot_params.screen_info.orig_video_mode) { -+ /* Mode setting failed, but we didn't end up where we -+ started. That's bad. Try to revert to the original -+ video mode. */ -+ ax = boot_params.screen_info.orig_video_mode; -+ asm volatile(INT10 -+ : "+a" (ax) -+ : : "ebx", "ecx", "edx", "esi", "edi"); -+ } -+ return -1; -+} -+ -+static int bios_probe(void) -+{ -+ u8 mode; -+ u8 saved_mode = boot_params.screen_info.orig_video_mode; -+ u16 crtc; -+ struct mode_info *mi; -+ int nmodes = 0; -+ -+ if (adapter != ADAPTER_EGA && adapter != ADAPTER_VGA) -+ return 0; -+ -+ set_fs(0); -+ crtc = vga_crtc(); -+ -+ video_bios.modes = GET_HEAP(struct mode_info, 0); -+ -+ for (mode = 0x14; mode <= 0x7f; mode++) { -+ if (heap_free() < sizeof(struct mode_info)) -+ break; -+ -+ if (mode_defined(VIDEO_FIRST_BIOS+mode)) -+ continue; -+ -+ if (set_bios_mode(mode)) -+ continue; -+ -+ /* Try to verify that it's a text mode. */ -+ -+ /* Attribute Controller: make graphics controller disabled */ -+ if (in_idx(0x3c0, 0x10) & 0x01) -+ continue; -+ -+ /* Graphics Controller: verify Alpha addressing enabled */ -+ if (in_idx(0x3ce, 0x06) & 0x01) -+ continue; -+ -+ /* CRTC cursor location low should be zero(?) */ -+ if (in_idx(crtc, 0x0f)) -+ continue; -+ -+ mi = GET_HEAP(struct mode_info, 1); -+ mi->mode = VIDEO_FIRST_BIOS+mode; -+ mi->x = rdfs16(0x44a); -+ mi->y = rdfs8(0x484)+1; -+ nmodes++; -+ } -+ -+ set_bios_mode(saved_mode); -+ -+ return nmodes; -+} -+ -+__videocard video_bios = -+{ -+ .card_name = "BIOS (scanned)", -+ .probe = bios_probe, -+ .set_mode = bios_set_mode, -+ .unsafe = 1, -+ .xmode_first = VIDEO_FIRST_BIOS, -+ .xmode_n = 0x80, -+}; ---- /dev/null -+++ b/arch/i386/boot/video-vesa.c -@@ -0,0 +1,283 @@ -+/* -*- linux-c -*- ------------------------------------------------------- * -+ * -+ * Copyright (C) 1991, 1992 Linus Torvalds -+ * Copyright 2007 rPath, Inc. - All Rights Reserved -+ * -+ * This file is part of the Linux kernel, and is made available under -+ * the terms of the GNU General Public License version 2. -+ * -+ * ----------------------------------------------------------------------- */ -+ -+/* -+ * arch/i386/boot/video-vesa.c -+ * -+ * VESA text modes -+ */ -+ -+#include "boot.h" -+#include "video.h" -+#include "vesa.h" -+ -+/* VESA information */ -+static struct vesa_general_info vginfo; -+static struct vesa_mode_info vminfo; -+ -+__videocard video_vesa; -+ -+static void vesa_store_mode_params_graphics(void); -+ -+static int vesa_probe(void) -+{ -+#ifdef CONFIG_VIDEO_VESA -+ u16 ax; -+ u16 mode; -+ addr_t mode_ptr; -+ struct mode_info *mi; -+ int nmodes = 0; -+ -+ video_vesa.modes = GET_HEAP(struct mode_info, 0); -+ -+ vginfo.signature = VBE2_MAGIC; -+ -+ /* Optimistically assume a VESA BIOS is register-clean... */ -+ ax = 0x4f00; -+ asm("int $0x10" : "+a" (ax), "=m" (vginfo) : "D" (&vginfo)); -+ -+ if (ax != 0x004f || -+ vginfo.signature != VESA_MAGIC || -+ vginfo.version < 0x0102) -+ return 0; /* Not present */ -+ -+ set_fs(vginfo.video_mode_ptr.seg); -+ mode_ptr = vginfo.video_mode_ptr.off; -+ -+ while ((mode = rdfs16(mode_ptr)) != 0xffff) { -+ mode_ptr += 2; -+ -+ if (heap_free() < sizeof(struct mode_info)) -+ break; /* Heap full, can't save mode info */ -+ -+ if (mode & ~0x1ff) -+ continue; -+ -+ memset(&vminfo, 0, sizeof vminfo); /* Just in case... */ -+ -+ ax = 0x4f01; -+ asm("int $0x10" -+ : "+a" (ax), "=m" (vminfo) -+ : "c" (mode), "D" (&vminfo)); -+ -+ if (ax != 0x004f) -+ continue; -+ -+ if ((vminfo.mode_attr & 0x15) == 0x05) { -+ /* Text Mode, TTY BIOS supported, -+ supported by hardware */ -+ mi = GET_HEAP(struct mode_info, 1); -+ mi->mode = mode + VIDEO_FIRST_VESA; -+ mi->x = vminfo.h_res; -+ mi->y = vminfo.v_res; -+ nmodes++; -+ } else if ((vminfo.mode_attr & 0x99) == 0x99) { -+#ifdef CONFIG_FB -+ /* Graphics mode, color, linear frame buffer -+ supported -- register the mode but hide from -+ the menu. Only do this if framebuffer is -+ configured, however, otherwise the user will -+ be left without a screen. */ -+ mi = GET_HEAP(struct mode_info, 1); -+ mi->mode = mode + VIDEO_FIRST_VESA; -+ mi->x = mi->y = 0; -+ nmodes++; -+#endif -+ } -+ } -+ -+ return nmodes; -+#else -+ return 0; -+#endif -+} -+ -+static int vesa_set_mode(struct mode_info *mode) -+{ -+ u16 ax; -+ int is_graphic; -+ u16 vesa_mode = mode->mode - VIDEO_FIRST_VESA; -+ -+ memset(&vminfo, 0, sizeof vminfo); /* Just in case... */ -+ -+ ax = 0x4f01; -+ asm("int $0x10" -+ : "+a" (ax), "=m" (vminfo) -+ : "c" (vesa_mode), "D" (&vminfo)); -+ -+ if (ax != 0x004f) -+ return -1; -+ -+ if ((vminfo.mode_attr & 0x15) == 0x05) { -+ /* It's a supported text mode */ -+ is_graphic = 0; -+ } else if ((vminfo.mode_attr & 0x99) == 0x99) { -+ /* It's a graphics mode with linear frame buffer */ -+ is_graphic = 1; -+ vesa_mode |= 0x4000; /* Request linear frame buffer */ -+ } else { -+ return -1; /* Invalid mode */ -+ } -+ -+ -+ ax = 0x4f02; -+ asm volatile("int $0x10" -+ : "+a" (ax) -+ : "b" (vesa_mode), "D" (0)); -+ -+ if (ax != 0x004f) -+ return -1; -+ -+ graphic_mode = is_graphic; -+ if (!is_graphic) { -+ /* Text mode */ -+ force_x = mode->x; -+ force_y = mode->y; -+ do_restore = 1; -+ } else { -+ /* Graphics mode */ -+ vesa_store_mode_params_graphics(); -+ } -+ -+ return 0; -+} -+ -+ -+/* Switch DAC to 8-bit mode */ -+static void vesa_dac_set_8bits(void) -+{ -+ u8 dac_size = 6; -+ -+ /* If possible, switch the DAC to 8-bit mode */ -+ if (vginfo.capabilities & 1) { -+ u16 ax, bx; -+ -+ ax = 0x4f08; -+ bx = 0x0800; -+ asm volatile(INT10 -+ : "+a" (ax), "+b" (bx) -+ : : "ecx", "edx", "esi", "edi"); -+ -+ if (ax == 0x004f) -+ dac_size = bx >> 8; -+ } -+ -+ /* Set the color sizes to the DAC size, and offsets to 0 */ -+ boot_params.screen_info.red_size = dac_size; -+ boot_params.screen_info.green_size = dac_size; -+ boot_params.screen_info.blue_size = dac_size; -+ boot_params.screen_info.rsvd_size = dac_size; -+ -+ boot_params.screen_info.red_pos = 0; -+ boot_params.screen_info.green_pos = 0; -+ boot_params.screen_info.blue_pos = 0; -+ boot_params.screen_info.rsvd_pos = 0; -+} -+ -+/* Save the VESA protected mode info */ -+static void vesa_store_pm_info(void) -+{ -+ u16 ax, bx, di, es; -+ -+ ax = 0x4f0a; -+ bx = di = 0; -+ asm("pushw %%es; "INT10"; movw %%es,%0; popw %%es" -+ : "=d" (es), "+a" (ax), "+b" (bx), "+D" (di) -+ : : "ecx", "esi"); -+ -+ if (ax != 0x004f) -+ return; -+ -+ boot_params.screen_info.vesapm_seg = es; -+ boot_params.screen_info.vesapm_off = di; -+} -+ -+/* -+ * Save video mode parameters for graphics mode -+ */ -+static void vesa_store_mode_params_graphics(void) -+{ -+ /* Tell the kernel we're in VESA graphics mode */ -+ boot_params.screen_info.orig_video_isVGA = 0x23; -+ -+ /* Mode parameters */ -+ boot_params.screen_info.vesa_attributes = vminfo.mode_attr; -+ boot_params.screen_info.lfb_linelength = vminfo.logical_scan; -+ boot_params.screen_info.lfb_width = vminfo.h_res; -+ boot_params.screen_info.lfb_height = vminfo.v_res; -+ boot_params.screen_info.lfb_depth = vminfo.bpp; -+ boot_params.screen_info.pages = vminfo.image_planes; -+ boot_params.screen_info.lfb_base = vminfo.lfb_ptr; -+ memcpy(&boot_params.screen_info.red_size, -+ &vminfo.rmask, 8); -+ -+ /* General parameters */ -+ boot_params.screen_info.lfb_size = vginfo.total_memory; -+ -+ if (vminfo.bpp <= 8) -+ vesa_dac_set_8bits(); -+ -+ vesa_store_pm_info(); -+} -+ -+/* -+ * Save EDID information for the kernel; this is invoked, separately, -+ * after mode-setting. -+ */ -+void vesa_store_edid(void) -+{ -+#ifdef CONFIG_FIRMWARE_EDID -+ u16 ax, bx, cx, dx, di; -+ -+ /* Apparently used as a nonsense token... */ -+ memset(&boot_params.edid_info, 0x13, sizeof boot_params.edid_info); -+ -+ if (vginfo.version < 0x0200) -+ return; /* EDID requires VBE 2.0+ */ -+ -+ ax = 0x4f15; /* VBE DDC */ -+ bx = 0x0000; /* Report DDC capabilities */ -+ cx = 0; /* Controller 0 */ -+ di = 0; /* ES:DI must be 0 by spec */ -+ -+ /* Note: The VBE DDC spec is different from the main VESA spec; -+ we genuinely have to assume all registers are destroyed here. */ -+ -+ asm("pushw %%es; movw %2,%%es; "INT10"; popw %%es" -+ : "+a" (ax), "+b" (bx) -+ : "c" (cx), "D" (di) -+ : "esi"); -+ -+ if (ax != 0x004f) -+ return; /* No EDID */ -+ -+ /* BH = time in seconds to transfer EDD information */ -+ /* BL = DDC level supported */ -+ -+ ax = 0x4f15; /* VBE DDC */ -+ bx = 0x0001; /* Read EDID */ -+ cx = 0; /* Controller 0 */ -+ dx = 0; /* EDID block number */ -+ di =(size_t) &boot_params.edid_info; /* (ES:)Pointer to block */ -+ asm(INT10 -+ : "+a" (ax), "+b" (bx), "+d" (dx) -+ : "c" (cx), "D" (di) -+ : "esi"); -+#endif /* CONFIG_FIRMWARE_EDID */ -+} -+ -+__videocard video_vesa = -+{ -+ .card_name = "VESA", -+ .probe = vesa_probe, -+ .set_mode = vesa_set_mode, -+ .xmode_first = VIDEO_FIRST_VESA, -+ .xmode_n = 0x200, -+}; ---- /dev/null -+++ b/arch/i386/boot/video-vga.c -@@ -0,0 +1,260 @@ -+/* -*- linux-c -*- ------------------------------------------------------- * -+ * -+ * Copyright (C) 1991, 1992 Linus Torvalds -+ * Copyright 2007 rPath, Inc. - All Rights Reserved -+ * -+ * This file is part of the Linux kernel, and is made available under -+ * the terms of the GNU General Public License version 2. -+ * -+ * ----------------------------------------------------------------------- */ -+ -+/* -+ * arch/i386/boot/video-vga.c -+ * -+ * Common all-VGA modes -+ */ -+ -+#include "boot.h" -+#include "video.h" -+ -+static struct mode_info vga_modes[] = { -+ { VIDEO_80x25, 80, 25 }, -+ { VIDEO_8POINT, 80, 50 }, -+ { VIDEO_80x43, 80, 43 }, -+ { VIDEO_80x28, 80, 28 }, -+ { VIDEO_80x30, 80, 30 }, -+ { VIDEO_80x34, 80, 34 }, -+ { VIDEO_80x60, 80, 60 }, -+}; -+ -+static struct mode_info ega_modes[] = { -+ { VIDEO_80x25, 80, 25 }, -+ { VIDEO_8POINT, 80, 43 }, -+}; -+ -+static struct mode_info cga_modes[] = { -+ { VIDEO_80x25, 80, 25 }, -+}; -+ -+__videocard video_vga; -+ -+/* Set basic 80x25 mode */ -+static u8 vga_set_basic_mode(void) -+{ -+ u16 ax; -+ u8 rows; -+ u8 mode; -+ -+#ifdef CONFIG_VIDEO_400_HACK -+ if (adapter >= ADAPTER_VGA) { -+ asm(INT10 -+ : : "a" (0x1202), "b" (0x0030) -+ : "ecx", "edx", "esi", "edi"); -+ } -+#endif -+ -+ ax = 0x0f00; -+ asm(INT10 -+ : "+a" (ax) -+ : : "ebx", "ecx", "edx", "esi", "edi"); -+ -+ mode = (u8)ax; -+ -+ set_fs(0); -+ rows = rdfs8(0x484); /* rows minus one */ -+ -+#ifndef CONFIG_VIDEO_400_HACK -+ if ((ax == 0x5003 || ax == 0x5007) && -+ (rows == 0 || rows == 24)) -+ return mode; -+#endif -+ -+ if (mode != 3 && mode != 7) -+ mode = 3; -+ -+ /* Set the mode */ -+ asm volatile(INT10 -+ : : "a" (mode) -+ : "ebx", "ecx", "edx", "esi", "edi"); -+ do_restore = 1; -+ return mode; -+} -+ -+static void vga_set_8font(void) -+{ -+ /* Set 8x8 font - 80x43 on EGA, 80x50 on VGA */ -+ -+ /* Set 8x8 font */ -+ asm volatile(INT10 : : "a" (0x1112), "b" (0)); -+ -+ /* Use alternate print screen */ -+ asm volatile(INT10 : : "a" (0x1200), "b" (0x20)); -+ -+ /* Turn off cursor emulation */ -+ asm volatile(INT10 : : "a" (0x1201), "b" (0x34)); -+ -+ /* Cursor is scan lines 6-7 */ -+ asm volatile(INT10 : : "a" (0x0100), "c" (0x0607)); -+} -+ -+static void vga_set_14font(void) -+{ -+ /* Set 9x14 font - 80x28 on VGA */ -+ -+ /* Set 9x14 font */ -+ asm volatile(INT10 : : "a" (0x1111), "b" (0)); -+ -+ /* Turn off cursor emulation */ -+ asm volatile(INT10 : : "a" (0x1201), "b" (0x34)); -+ -+ /* Cursor is scan lines 11-12 */ -+ asm volatile(INT10 : : "a" (0x0100), "c" (0x0b0c)); -+} -+ -+static void vga_set_80x43(void) -+{ -+ /* Set 80x43 mode on VGA (not EGA) */ -+ -+ /* Set 350 scans */ -+ asm volatile(INT10 : : "a" (0x1201), "b" (0x30)); -+ -+ /* Reset video mode */ -+ asm volatile(INT10 : : "a" (0x0003)); -+ -+ vga_set_8font(); -+} -+ -+/* I/O address of the VGA CRTC */ -+u16 vga_crtc(void) -+{ -+ return (inb(0x3cc) & 1) ? 0x3d4 : 0x3b4; -+} -+ -+static void vga_set_480_scanlines(int end) -+{ -+ u16 crtc; -+ u8 csel; -+ -+ crtc = vga_crtc(); -+ -+ out_idx(0x0c, crtc, 0x11); /* Vertical sync end, unlock CR0-7 */ -+ out_idx(0x0b, crtc, 0x06); /* Vertical total */ -+ out_idx(0x3e, crtc, 0x07); /* Vertical overflow */ -+ out_idx(0xea, crtc, 0x10); /* Vertical sync start */ -+ out_idx(end, crtc, 0x12); /* Vertical display end */ -+ out_idx(0xe7, crtc, 0x15); /* Vertical blank start */ -+ out_idx(0x04, crtc, 0x16); /* Vertical blank end */ -+ csel = inb(0x3cc); -+ csel &= 0x0d; -+ csel |= 0xe2; -+ outb(csel, 0x3cc); -+} -+ -+static void vga_set_80x30(void) -+{ -+ vga_set_480_scanlines(0xdf); -+} -+ -+static void vga_set_80x34(void) -+{ -+ vga_set_14font(); -+ vga_set_480_scanlines(0xdb); -+} -+ -+static void vga_set_80x60(void) -+{ -+ vga_set_8font(); -+ vga_set_480_scanlines(0xdf); -+} -+ -+static int vga_set_mode(struct mode_info *mode) -+{ -+ /* Set the basic mode */ -+ vga_set_basic_mode(); -+ -+ /* Override a possibly broken BIOS */ -+ force_x = mode->x; -+ force_y = mode->y; -+ -+ switch (mode->mode) { -+ case VIDEO_80x25: -+ break; -+ case VIDEO_8POINT: -+ vga_set_8font(); -+ break; -+ case VIDEO_80x43: -+ vga_set_80x43(); -+ break; -+ case VIDEO_80x28: -+ vga_set_14font(); -+ break; -+ case VIDEO_80x30: -+ vga_set_80x30(); -+ break; -+ case VIDEO_80x34: -+ vga_set_80x34(); -+ break; -+ case VIDEO_80x60: -+ vga_set_80x60(); -+ break; -+ } -+ -+ return 0; -+} -+ -+/* -+ * Note: this probe includes basic information required by all -+ * systems. It should be executed first, by making sure -+ * video-vga.c is listed first in the Makefile. -+ */ -+static int vga_probe(void) -+{ -+ static const char *card_name[] = { -+ "CGA/MDA/HGC", "EGA", "VGA" -+ }; -+ static struct mode_info *mode_lists[] = { -+ cga_modes, -+ ega_modes, -+ vga_modes, -+ }; -+ static int mode_count[] = { -+ sizeof(cga_modes)/sizeof(struct mode_info), -+ sizeof(ega_modes)/sizeof(struct mode_info), -+ sizeof(vga_modes)/sizeof(struct mode_info), -+ }; -+ u8 vga_flag; -+ -+ asm(INT10 -+ : "=b" (boot_params.screen_info.orig_video_ega_bx) -+ : "a" (0x1200), "b" (0x10) /* Check EGA/VGA */ -+ : "ecx", "edx", "esi", "edi"); -+ -+ /* If we have MDA/CGA/HGC then BL will be unchanged at 0x10 */ -+ if ((u8)boot_params.screen_info.orig_video_ega_bx != 0x10) { -+ /* EGA/VGA */ -+ asm(INT10 -+ : "=a" (vga_flag) -+ : "a" (0x1a00) -+ : "ebx", "ecx", "edx", "esi", "edi"); -+ -+ if (vga_flag == 0x1a) { -+ adapter = ADAPTER_VGA; -+ boot_params.screen_info.orig_video_isVGA = 1; -+ } else { -+ adapter = ADAPTER_EGA; -+ } -+ } else { -+ adapter = ADAPTER_CGA; -+ } -+ -+ video_vga.modes = mode_lists[adapter]; -+ video_vga.card_name = card_name[adapter]; -+ return mode_count[adapter]; -+} -+ -+__videocard video_vga = -+{ -+ .card_name = "VGA", -+ .probe = vga_probe, -+ .set_mode = vga_set_mode, -+}; ---- a/arch/i386/boot/video.S -+++ /dev/null -@@ -1,2043 +0,0 @@ --/* video.S -- * -- * Display adapter & video mode setup, version 2.13 (14-May-99) -- * -- * Copyright (C) 1995 -- 1998 Martin Mares <mj@ucw.cz> -- * Based on the original setup.S code (C) Linus Torvalds and Mats Anderson -- * -- * Rewritten to use GNU 'as' by Chris Noe <stiker@northlink.com> May 1999 -- * -- * For further information, look at Documentation/svga.txt. -- * -- */ -- --/* Enable autodetection of SVGA adapters and modes. */ --#undef CONFIG_VIDEO_SVGA -- --/* Enable autodetection of VESA modes */ --#define CONFIG_VIDEO_VESA -- --/* Enable compacting of mode table */ --#define CONFIG_VIDEO_COMPACT -- --/* Retain screen contents when switching modes */ --#define CONFIG_VIDEO_RETAIN -- --/* Enable local mode list */ --#undef CONFIG_VIDEO_LOCAL -- --/* Force 400 scan lines for standard modes (hack to fix bad BIOS behaviour */ --#undef CONFIG_VIDEO_400_HACK -- --/* Hack that lets you force specific BIOS mode ID and specific dimensions */ --#undef CONFIG_VIDEO_GFX_HACK --#define VIDEO_GFX_BIOS_AX 0x4f02 /* 800x600 on ThinkPad */ --#define VIDEO_GFX_BIOS_BX 0x0102 --#define VIDEO_GFX_DUMMY_RESOLUTION 0x6425 /* 100x37 */ -- --/* This code uses an extended set of video mode numbers. These include: -- * Aliases for standard modes -- * NORMAL_VGA (-1) -- * EXTENDED_VGA (-2) -- * ASK_VGA (-3) -- * Video modes numbered by menu position -- NOT RECOMMENDED because of lack -- * of compatibility when extending the table. These are between 0x00 and 0xff. -- */ --#define VIDEO_FIRST_MENU 0x0000 -- --/* Standard BIOS video modes (BIOS number + 0x0100) */ --#define VIDEO_FIRST_BIOS 0x0100 -- --/* VESA BIOS video modes (VESA number + 0x0200) */ --#define VIDEO_FIRST_VESA 0x0200 -- --/* Video7 special modes (BIOS number + 0x0900) */ --#define VIDEO_FIRST_V7 0x0900 -- --/* Special video modes */ --#define VIDEO_FIRST_SPECIAL 0x0f00 --#define VIDEO_80x25 0x0f00 --#define VIDEO_8POINT 0x0f01 --#define VIDEO_80x43 0x0f02 --#define VIDEO_80x28 0x0f03 --#define VIDEO_CURRENT_MODE 0x0f04 --#define VIDEO_80x30 0x0f05 --#define VIDEO_80x34 0x0f06 --#define VIDEO_80x60 0x0f07 --#define VIDEO_GFX_HACK 0x0f08 --#define VIDEO_LAST_SPECIAL 0x0f09 -- --/* Video modes given by resolution */ --#define VIDEO_FIRST_RESOLUTION 0x1000 -- --/* The "recalculate timings" flag */ --#define VIDEO_RECALC 0x8000 -- --/* Positions of various video parameters passed to the kernel */ --/* (see also include/linux/tty.h) */ --#define PARAM_CURSOR_POS 0x00 --#define PARAM_VIDEO_PAGE 0x04 --#define PARAM_VIDEO_MODE 0x06 --#define PARAM_VIDEO_COLS 0x07 --#define PARAM_VIDEO_EGA_BX 0x0a --#define PARAM_VIDEO_LINES 0x0e --#define PARAM_HAVE_VGA 0x0f --#define PARAM_FONT_POINTS 0x10 -- --#define PARAM_LFB_WIDTH 0x12 --#define PARAM_LFB_HEIGHT 0x14 --#define PARAM_LFB_DEPTH 0x16 --#define PARAM_LFB_BASE 0x18 --#define PARAM_LFB_SIZE 0x1c --#define PARAM_LFB_LINELENGTH 0x24 --#define PARAM_LFB_COLORS 0x26 --#define PARAM_VESAPM_SEG 0x2e --#define PARAM_VESAPM_OFF 0x30 --#define PARAM_LFB_PAGES 0x32 --#define PARAM_VESA_ATTRIB 0x34 --#define PARAM_CAPABILITIES 0x36 -- --/* Define DO_STORE according to CONFIG_VIDEO_RETAIN */ --#ifdef CONFIG_VIDEO_RETAIN --#define DO_STORE call store_screen --#else --#define DO_STORE --#endif /* CONFIG_VIDEO_RETAIN */ -- --# This is the main entry point called by setup.S --# %ds *must* be pointing to the bootsector --video: pushw %ds # We use different segments -- pushw %ds # FS contains original DS -- popw %fs -- pushw %cs # DS is equal to CS -- popw %ds -- pushw %cs # ES is equal to CS -- popw %es -- xorw %ax, %ax -- movw %ax, %gs # GS is zero -- cld -- call basic_detect # Basic adapter type testing (EGA/VGA/MDA/CGA) --#ifdef CONFIG_VIDEO_SELECT -- movw %fs:(0x01fa), %ax # User selected video mode -- cmpw $ASK_VGA, %ax # Bring up the menu -- jz vid2 -- -- call mode_set # Set the mode -- jc vid1 -- -- leaw badmdt, %si # Invalid mode ID -- call prtstr --vid2: call mode_menu --vid1: --#ifdef CONFIG_VIDEO_RETAIN -- call restore_screen # Restore screen contents --#endif /* CONFIG_VIDEO_RETAIN */ -- call store_edid --#endif /* CONFIG_VIDEO_SELECT */ -- call mode_params # Store mode parameters -- popw %ds # Restore original DS -- ret -- --# Detect if we have CGA, MDA, EGA or VGA and pass it to the kernel. --basic_detect: -- movb $0, %fs:(PARAM_HAVE_VGA) -- movb $0x12, %ah # Check EGA/VGA -- movb $0x10, %bl -- int $0x10 -- movw %bx, %fs:(PARAM_VIDEO_EGA_BX) # Identifies EGA to the kernel -- cmpb $0x10, %bl # No, it's a CGA/MDA/HGA card. -- je basret -- -- incb adapter -- movw $0x1a00, %ax # Check EGA or VGA? -- int $0x10 -- cmpb $0x1a, %al # 1a means VGA... -- jne basret # anything else is EGA. -- -- incb %fs:(PARAM_HAVE_VGA) # We've detected a VGA -- incb adapter --basret: ret -- --# Store the video mode parameters for later usage by the kernel. --# This is done by asking the BIOS except for the rows/columns --# parameters in the default 80x25 mode -- these are set directly, --# because some very obscure BIOSes supply insane values. --mode_params: --#ifdef CONFIG_VIDEO_SELECT -- cmpb $0, graphic_mode -- jnz mopar_gr --#endif -- movb $0x03, %ah # Read cursor position -- xorb %bh, %bh -- int $0x10 -- movw %dx, %fs:(PARAM_CURSOR_POS) -- movb $0x0f, %ah # Read page/mode/width -- int $0x10 -- movw %bx, %fs:(PARAM_VIDEO_PAGE) -- movw %ax, %fs:(PARAM_VIDEO_MODE) # Video mode and screen width -- cmpb $0x7, %al # MDA/HGA => segment differs -- jnz mopar0 -- -- movw $0xb000, video_segment --mopar0: movw %gs:(0x485), %ax # Font size -- movw %ax, %fs:(PARAM_FONT_POINTS) # (valid only on EGA/VGA) -- movw force_size, %ax # Forced size? -- orw %ax, %ax -- jz mopar1 -- -- movb %ah, %fs:(PARAM_VIDEO_COLS) -- movb %al, %fs:(PARAM_VIDEO_LINES) -- ret -- --mopar1: movb $25, %al -- cmpb $0, adapter # If we are on CGA/MDA/HGA, the -- jz mopar2 # screen must have 25 lines. -- -- movb %gs:(0x484), %al # On EGA/VGA, use the EGA+ BIOS -- incb %al # location of max lines. --mopar2: movb %al, %fs:(PARAM_VIDEO_LINES) -- ret -- --#ifdef CONFIG_VIDEO_SELECT --# Fetching of VESA frame buffer parameters --mopar_gr: -- leaw modelist+1024, %di -- movb $0x23, %fs:(PARAM_HAVE_VGA) -- movw 16(%di), %ax -- movw %ax, %fs:(PARAM_LFB_LINELENGTH) -- movw 18(%di), %ax -- movw %ax, %fs:(PARAM_LFB_WIDTH) -- movw 20(%di), %ax -- movw %ax, %fs:(PARAM_LFB_HEIGHT) -- movb 25(%di), %al -- movb $0, %ah -- movw %ax, %fs:(PARAM_LFB_DEPTH) -- movb 29(%di), %al -- movb $0, %ah -- movw %ax, %fs:(PARAM_LFB_PAGES) -- movl 40(%di), %eax -- movl %eax, %fs:(PARAM_LFB_BASE) -- movl 31(%di), %eax -- movl %eax, %fs:(PARAM_LFB_COLORS) -- movl 35(%di), %eax -- movl %eax, %fs:(PARAM_LFB_COLORS+4) -- movw 0(%di), %ax -- movw %ax, %fs:(PARAM_VESA_ATTRIB) -- --# get video mem size -- leaw modelist+1024, %di -- movw $0x4f00, %ax -- int $0x10 -- xorl %eax, %eax -- movw 18(%di), %ax -- movl %eax, %fs:(PARAM_LFB_SIZE) -- --# store mode capabilities -- movl 10(%di), %eax -- movl %eax, %fs:(PARAM_CAPABILITIES) -- --# switching the DAC to 8-bit is for <= 8 bpp only -- movw %fs:(PARAM_LFB_DEPTH), %ax -- cmpw $8, %ax -- jg dac_done -- --# get DAC switching capability -- xorl %eax, %eax -- movb 10(%di), %al -- testb $1, %al -- jz dac_set -- --# attempt to switch DAC to 8-bit -- movw $0x4f08, %ax -- movw $0x0800, %bx -- int $0x10 -- cmpw $0x004f, %ax -- jne dac_set -- movb %bh, dac_size # store actual DAC size -- --dac_set: --# set color size to DAC size -- movb dac_size, %al -- movb %al, %fs:(PARAM_LFB_COLORS+0) -- movb %al, %fs:(PARAM_LFB_COLORS+2) -- movb %al, %fs:(PARAM_LFB_COLORS+4) -- movb %al, %fs:(PARAM_LFB_COLORS+6) -- --# set color offsets to 0 -- movb $0, %fs:(PARAM_LFB_COLORS+1) -- movb $0, %fs:(PARAM_LFB_COLORS+3) -- movb $0, %fs:(PARAM_LFB_COLORS+5) -- movb $0, %fs:(PARAM_LFB_COLORS+7) -- --dac_done: --# get protected mode interface informations -- movw $0x4f0a, %ax -- xorw %bx, %bx -- xorw %di, %di -- int $0x10 -- cmp $0x004f, %ax -- jnz no_pm -- -- movw %es, %fs:(PARAM_VESAPM_SEG) -- movw %di, %fs:(PARAM_VESAPM_OFF) --no_pm: ret -- --# The video mode menu --mode_menu: -- leaw keymsg, %si # "Return/Space/Timeout" message -- call prtstr -- call flush --nokey: call getkt -- -- cmpb $0x0d, %al # ENTER ? -- je listm # yes - manual mode selection -- -- cmpb $0x20, %al # SPACE ? -- je defmd1 # no - repeat -- -- call beep -- jmp nokey -- --defmd1: ret # No mode chosen? Default 80x25 -- --listm: call mode_table # List mode table --listm0: leaw name_bann, %si # Print adapter name -- call prtstr -- movw card_name, %si -- orw %si, %si -- jnz an2 -- -- movb adapter, %al -- leaw old_name, %si -- orb %al, %al -- jz an1 -- -- leaw ega_name, %si -- decb %al -- jz an1 -- -- leaw vga_name, %si -- jmp an1 -- --an2: call prtstr -- leaw svga_name, %si --an1: call prtstr -- leaw listhdr, %si # Table header -- call prtstr -- movb $0x30, %dl # DL holds mode number -- leaw modelist, %si --lm1: cmpw $ASK_VGA, (%si) # End? -- jz lm2 -- -- movb %dl, %al # Menu selection number -- call prtchr -- call prtsp2 -- lodsw -- call prthw # Mode ID -- call prtsp2 -- movb 0x1(%si), %al -- call prtdec # Rows -- movb $0x78, %al # the letter 'x' -- call prtchr -- lodsw -- call prtdec # Columns -- movb $0x0d, %al # New line -- call prtchr -- movb $0x0a, %al -- call prtchr -- incb %dl # Next character -- cmpb $0x3a, %dl -- jnz lm1 -- -- movb $0x61, %dl -- jmp lm1 -- --lm2: leaw prompt, %si # Mode prompt -- call prtstr -- leaw edit_buf, %di # Editor buffer --lm3: call getkey -- cmpb $0x0d, %al # Enter? -- jz lment -- -- cmpb $0x08, %al # Backspace? -- jz lmbs -- -- cmpb $0x20, %al # Printable? -- jc lm3 -- -- cmpw $edit_buf+4, %di # Enough space? -- jz lm3 -- -- stosb -- call prtchr -- jmp lm3 -- --lmbs: cmpw $edit_buf, %di # Backspace -- jz lm3 -- -- decw %di -- movb $0x08, %al -- call prtchr -- call prtspc -- movb $0x08, %al -- call prtchr -- jmp lm3 -- --lment: movb $0, (%di) -- leaw crlft, %si -- call prtstr -- leaw edit_buf, %si -- cmpb $0, (%si) # Empty string = default mode -- jz lmdef -- -- cmpb $0, 1(%si) # One character = menu selection -- jz mnusel -- -- cmpw $0x6373, (%si) # "scan" => mode scanning -- jnz lmhx -- -- cmpw $0x6e61, 2(%si) -- jz lmscan -- --lmhx: xorw %bx, %bx # Else => mode ID in hex --lmhex: lodsb -- orb %al, %al -- jz lmuse1 -- -- subb $0x30, %al -- jc lmbad -- -- cmpb $10, %al -- jc lmhx1 -- -- subb $7, %al -- andb $0xdf, %al -- cmpb $10, %al -- jc lmbad -- -- cmpb $16, %al -- jnc lmbad -- --lmhx1: shlw $4, %bx -- orb %al, %bl -- jmp lmhex -- --lmuse1: movw %bx, %ax -- jmp lmuse -- --mnusel: lodsb # Menu selection -- xorb %ah, %ah -- subb $0x30, %al -- jc lmbad -- -- cmpb $10, %al -- jc lmuse -- -- cmpb $0x61-0x30, %al -- jc lmbad -- -- subb $0x61-0x30-10, %al -- cmpb $36, %al -- jnc lmbad -- --lmuse: call mode_set -- jc lmdef -- --lmbad: leaw unknt, %si -- call prtstr -- jmp lm2 --lmscan: cmpb $0, adapter # Scanning only on EGA/VGA -- jz lmbad -- -- movw $0, mt_end # Scanning of modes is -- movb $1, scanning # done as new autodetection. -- call mode_table -- jmp listm0 --lmdef: ret -- --# Additional parts of mode_set... (relative jumps, you know) --setv7: # Video7 extended modes -- DO_STORE -- subb $VIDEO_FIRST_V7>>8, %bh -- movw $0x6f05, %ax -- int $0x10 -- stc -- ret -- --_setrec: jmp setrec # Ugly... --_set_80x25: jmp set_80x25 -- --# Aliases for backward compatibility. --setalias: -- movw $VIDEO_80x25, %ax -- incw %bx -- jz mode_set -- -- movb $VIDEO_8POINT-VIDEO_FIRST_SPECIAL, %al -- incw %bx -- jnz setbad # Fall-through! -- --# Setting of user mode (AX=mode ID) => CF=success --mode_set: -- movw %ax, %fs:(0x01fa) # Store mode for use in acpi_wakeup.S -- movw %ax, %bx -- cmpb $0xff, %ah -- jz setalias -- -- testb $VIDEO_RECALC>>8, %ah -- jnz _setrec -- -- cmpb $VIDEO_FIRST_RESOLUTION>>8, %ah -- jnc setres -- -- cmpb $VIDEO_FIRST_SPECIAL>>8, %ah -- jz setspc -- -- cmpb $VIDEO_FIRST_V7>>8, %ah -- jz setv7 -- -- cmpb $VIDEO_FIRST_VESA>>8, %ah -- jnc check_vesa -- -- orb %ah, %ah -- jz setmenu -- -- decb %ah -- jz setbios -- --setbad: clc -- movb $0, do_restore # The screen needn't be restored -- ret -- --setvesa: -- DO_STORE -- subb $VIDEO_FIRST_VESA>>8, %bh -- movw $0x4f02, %ax # VESA BIOS mode set call -- int $0x10 -- cmpw $0x004f, %ax # AL=4f if implemented -- jnz setbad # AH=0 if OK -- -- stc -- ret -- --setbios: -- DO_STORE -- int $0x10 # Standard BIOS mode set call -- pushw %bx -- movb $0x0f, %ah # Check if really set -- int $0x10 -- popw %bx -- cmpb %bl, %al -- jnz setbad -- -- stc -- ret -- --setspc: xorb %bh, %bh # Set special mode -- cmpb $VIDEO_LAST_SPECIAL-VIDEO_FIRST_SPECIAL, %bl -- jnc setbad -- -- addw %bx, %bx -- jmp *spec_inits(%bx) -- --setmenu: -- orb %al, %al # 80x25 is an exception -- jz _set_80x25 -- -- pushw %bx # Set mode chosen from menu -- call mode_table # Build the mode table -- popw %ax -- shlw $2, %ax -- addw %ax, %si -- cmpw %di, %si -- jnc setbad -- -- movw (%si), %ax # Fetch mode ID --_m_s: jmp mode_set -- --setres: pushw %bx # Set mode chosen by resolution -- call mode_table -- popw %bx -- xchgb %bl, %bh --setr1: lodsw -- cmpw $ASK_VGA, %ax # End of the list? -- jz setbad -- -- lodsw -- cmpw %bx, %ax -- jnz setr1 -- -- movw -4(%si), %ax # Fetch mode ID -- jmp _m_s -- --check_vesa: --#ifdef CONFIG_FIRMWARE_EDID -- leaw modelist+1024, %di -- movw $0x4f00, %ax -- int $0x10 -- cmpw $0x004f, %ax -- jnz setbad -- -- movw 4(%di), %ax -- movw %ax, vbe_version --#endif -- leaw modelist+1024, %di -- subb $VIDEO_FIRST_VESA>>8, %bh -- movw %bx, %cx # Get mode information structure -- movw $0x4f01, %ax -- int $0x10 -- addb $VIDEO_FIRST_VESA>>8, %bh -- cmpw $0x004f, %ax -- jnz setbad -- -- movb (%di), %al # Check capabilities. -- andb $0x19, %al -- cmpb $0x09, %al -- jz setvesa # This is a text mode -- -- movb (%di), %al # Check capabilities. -- andb $0x99, %al -- cmpb $0x99, %al -- jnz _setbad # Doh! No linear frame buffer. -- -- subb $VIDEO_FIRST_VESA>>8, %bh -- orw $0x4000, %bx # Use linear frame buffer -- movw $0x4f02, %ax # VESA BIOS mode set call -- int $0x10 -- cmpw $0x004f, %ax # AL=4f if implemented -- jnz _setbad # AH=0 if OK -- -- movb $1, graphic_mode # flag graphic mode -- movb $0, do_restore # no screen restore -- stc -- ret -- --_setbad: jmp setbad # Ugly... -- --# Recalculate vertical display end registers -- this fixes various --# inconsistencies of extended modes on many adapters. Called when --# the VIDEO_RECALC flag is set in the mode ID. -- --setrec: subb $VIDEO_RECALC>>8, %ah # Set the base mode -- call mode_set -- jnc rct3 -- -- movw %gs:(0x485), %ax # Font size in pixels -- movb %gs:(0x484), %bl # Number of rows -- incb %bl -- mulb %bl # Number of visible -- decw %ax # scan lines - 1 -- movw $0x3d4, %dx -- movw %ax, %bx -- movb $0x12, %al # Lower 8 bits -- movb %bl, %ah -- outw %ax, %dx -- movb $0x07, %al # Bits 8 and 9 in the overflow register -- call inidx -- xchgb %al, %ah -- andb $0xbd, %ah -- shrb %bh -- jnc rct1 -- orb $0x02, %ah --rct1: shrb %bh -- jnc rct2 -- orb $0x40, %ah --rct2: movb $0x07, %al -- outw %ax, %dx -- stc --rct3: ret -- --# Table of routines for setting of the special modes. --spec_inits: -- .word set_80x25 -- .word set_8pixel -- .word set_80x43 -- .word set_80x28 -- .word set_current -- .word set_80x30 -- .word set_80x34 -- .word set_80x60 -- .word set_gfx -- --# Set the 80x25 mode. If already set, do nothing. --set_80x25: -- movw $0x5019, force_size # Override possibly broken BIOS --use_80x25: --#ifdef CONFIG_VIDEO_400_HACK -- movw $0x1202, %ax # Force 400 scan lines -- movb $0x30, %bl -- int $0x10 --#else -- movb $0x0f, %ah # Get current mode ID -- int $0x10 -- cmpw $0x5007, %ax # Mode 7 (80x25 mono) is the only one available -- jz st80 # on CGA/MDA/HGA and is also available on EGAM -- -- cmpw $0x5003, %ax # Unknown mode, force 80x25 color -- jnz force3 -- --st80: cmpb $0, adapter # CGA/MDA/HGA => mode 3/7 is always 80x25 -- jz set80 -- -- movb %gs:(0x0484), %al # This is EGA+ -- beware of 80x50 etc. -- orb %al, %al # Some buggy BIOS'es set 0 rows -- jz set80 -- -- cmpb $24, %al # It's hopefully correct -- jz set80 --#endif /* CONFIG_VIDEO_400_HACK */ --force3: DO_STORE -- movw $0x0003, %ax # Forced set -- int $0x10 --set80: stc -- ret -- --# Set the 80x50/80x43 8-pixel mode. Simple BIOS calls. --set_8pixel: -- DO_STORE -- call use_80x25 # The base is 80x25 --set_8pt: -- movw $0x1112, %ax # Use 8x8 font -- xorb %bl, %bl -- int $0x10 -- movw $0x1200, %ax # Use alternate print screen -- movb $0x20, %bl -- int $0x10 -- movw $0x1201, %ax # Turn off cursor emulation -- movb $0x34, %bl -- int $0x10 -- movb $0x01, %ah # Define cursor scan lines 6-7 -- movw $0x0607, %cx -- int $0x10 --set_current: -- stc -- ret -- --# Set the 80x28 mode. This mode works on all VGA's, because it's a standard --# 80x25 mode with 14-point fonts instead of 16-point. --set_80x28: -- DO_STORE -- call use_80x25 # The base is 80x25 --set14: movw $0x1111, %ax # Use 9x14 font -- xorb %bl, %bl -- int $0x10 -- movb $0x01, %ah # Define cursor scan lines 11-12 -- movw $0x0b0c, %cx -- int $0x10 -- stc -- ret -- --# Set the 80x43 mode. This mode is works on all VGA's. --# It's a 350-scanline mode with 8-pixel font. --set_80x43: -- DO_STORE -- movw $0x1201, %ax # Set 350 scans -- movb $0x30, %bl -- int $0x10 -- movw $0x0003, %ax # Reset video mode -- int $0x10 -- jmp set_8pt # Use 8-pixel font -- --# Set the 80x30 mode (all VGA's). 480 scanlines, 16-pixel font. --set_80x30: -- call use_80x25 # Start with real 80x25 -- DO_STORE -- movw $0x3cc, %dx # Get CRTC port -- inb %dx, %al -- movb $0xd4, %dl -- rorb %al # Mono or color? -- jc set48a -- -- movb $0xb4, %dl --set48a: movw $0x0c11, %ax # Vertical sync end (also unlocks CR0-7) -- call outidx -- movw $0x0b06, %ax # Vertical total -- call outidx -- movw $0x3e07, %ax # (Vertical) overflow -- call outidx -- movw $0xea10, %ax # Vertical sync start -- call outidx -- movw $0xdf12, %ax # Vertical display end -- call outidx -- movw $0xe715, %ax # Vertical blank start -- call outidx -- movw $0x0416, %ax # Vertical blank end -- call outidx -- pushw %dx -- movb $0xcc, %dl # Misc output register (read) -- inb %dx, %al -- movb $0xc2, %dl # (write) -- andb $0x0d, %al # Preserve clock select bits and color bit -- orb $0xe2, %al # Set correct sync polarity -- outb %al, %dx -- popw %dx -- movw $0x501e, force_size -- stc # That's all. -- ret -- --# Set the 80x34 mode (all VGA's). 480 scans, 14-pixel font. --set_80x34: -- call set_80x30 # Set 480 scans -- call set14 # And 14-pt font -- movw $0xdb12, %ax # VGA vertical display end -- movw $0x5022, force_size --setvde: call outidx -- stc -- ret -- --# Set the 80x60 mode (all VGA's). 480 scans, 8-pixel font. --set_80x60: -- call set_80x30 # Set 480 scans -- call set_8pt # And 8-pt font -- movw $0xdf12, %ax # VGA vertical display end -- movw $0x503c, force_size -- jmp setvde -- --# Special hack for ThinkPad graphics --set_gfx: --#ifdef CONFIG_VIDEO_GFX_HACK -- movw $VIDEO_GFX_BIOS_AX, %ax -- movw $VIDEO_GFX_BIOS_BX, %bx -- int $0x10 -- movw $VIDEO_GFX_DUMMY_RESOLUTION, force_size -- stc --#endif -- ret -- --#ifdef CONFIG_VIDEO_RETAIN -- --# Store screen contents to temporary buffer. --store_screen: -- cmpb $0, do_restore # Already stored? -- jnz stsr -- -- testb $CAN_USE_HEAP, loadflags # Have we space for storing? -- jz stsr -- -- pushw %ax -- pushw %bx -- pushw force_size # Don't force specific size -- movw $0, force_size -- call mode_params # Obtain params of current mode -- popw force_size -- movb %fs:(PARAM_VIDEO_LINES), %ah -- movb %fs:(PARAM_VIDEO_COLS), %al -- movw %ax, %bx # BX=dimensions -- mulb %ah -- movw %ax, %cx # CX=number of characters -- addw %ax, %ax # Calculate image size -- addw $modelist+1024+4, %ax -- cmpw heap_end_ptr, %ax -- jnc sts1 # Unfortunately, out of memory -- -- movw %fs:(PARAM_CURSOR_POS), %ax # Store mode params -- leaw modelist+1024, %di -- stosw -- movw %bx, %ax -- stosw -- pushw %ds # Store the screen -- movw video_segment, %ds -- xorw %si, %si -- rep -- movsw -- popw %ds -- incb do_restore # Screen will be restored later --sts1: popw %bx -- popw %ax --stsr: ret -- --# Restore screen contents from temporary buffer. --restore_screen: -- cmpb $0, do_restore # Has the screen been stored? -- jz res1 -- -- call mode_params # Get parameters of current mode -- movb %fs:(PARAM_VIDEO_LINES), %cl -- movb %fs:(PARAM_VIDEO_COLS), %ch -- leaw modelist+1024, %si # Screen buffer -- lodsw # Set cursor position -- movw %ax, %dx -- cmpb %cl, %dh -- jc res2 -- -- movb %cl, %dh -- decb %dh --res2: cmpb %ch, %dl -- jc res3 -- -- movb %ch, %dl -- decb %dl --res3: movb $0x02, %ah -- movb $0x00, %bh -- int $0x10 -- lodsw # Display size -- movb %ah, %dl # DL=number of lines -- movb $0, %ah # BX=phys. length of orig. line -- movw %ax, %bx -- cmpb %cl, %dl # Too many? -- jc res4 -- -- pushw %ax -- movb %dl, %al -- subb %cl, %al -- mulb %bl -- addw %ax, %si -- addw %ax, %si -- popw %ax -- movb %cl, %dl --res4: cmpb %ch, %al # Too wide? -- jc res5 -- -- movb %ch, %al # AX=width of src. line --res5: movb $0, %cl -- xchgb %ch, %cl -- movw %cx, %bp # BP=width of dest. line -- pushw %es -- movw video_segment, %es -- xorw %di, %di # Move the data -- addw %bx, %bx # Convert BX and BP to _bytes_ -- addw %bp, %bp --res6: pushw %si -- pushw %di -- movw %ax, %cx -- rep -- movsw -- popw %di -- popw %si -- addw %bp, %di -- addw %bx, %si -- decb %dl -- jnz res6 -- -- popw %es # Done --res1: ret --#endif /* CONFIG_VIDEO_RETAIN */ -- --# Write to indexed VGA register (AL=index, AH=data, DX=index reg. port) --outidx: outb %al, %dx -- pushw %ax -- movb %ah, %al -- incw %dx -- outb %al, %dx -- decw %dx -- popw %ax -- ret -- --# Build the table of video modes (stored after the setup.S code at the --# `modelist' label. Each video mode record looks like: --# .word MODE-ID (our special mode ID (see above)) --# .byte rows (number of rows) --# .byte columns (number of columns) --# Returns address of the end of the table in DI, the end is marked --# with a ASK_VGA ID. --mode_table: -- movw mt_end, %di # Already filled? -- orw %di, %di -- jnz mtab1x -- -- leaw modelist, %di # Store standard modes: -- movl $VIDEO_80x25 + 0x50190000, %eax # The 80x25 mode (ALL) -- stosl -- movb adapter, %al # CGA/MDA/HGA -- no more modes -- orb %al, %al -- jz mtabe -- -- decb %al -- jnz mtabv -- -- movl $VIDEO_8POINT + 0x502b0000, %eax # The 80x43 EGA mode -- stosl -- jmp mtabe -- --mtab1x: jmp mtab1 -- --mtabv: leaw vga_modes, %si # All modes for std VGA -- movw $vga_modes_end-vga_modes, %cx -- rep # I'm unable to use movsw as I don't know how to store a half -- movsb # of the expression above to cx without using explicit shr. -- -- cmpb $0, scanning # Mode scan requested? -- jz mscan1 -- -- call mode_scan --mscan1: -- --#ifdef CONFIG_VIDEO_LOCAL -- call local_modes --#endif /* CONFIG_VIDEO_LOCAL */ -- --#ifdef CONFIG_VIDEO_VESA -- call vesa_modes # Detect VESA VGA modes --#endif /* CONFIG_VIDEO_VESA */ -- --#ifdef CONFIG_VIDEO_SVGA -- cmpb $0, scanning # Bypass when scanning -- jnz mscan2 -- -- call svga_modes # Detect SVGA cards & modes --mscan2: --#endif /* CONFIG_VIDEO_SVGA */ -- --mtabe: -- --#ifdef CONFIG_VIDEO_COMPACT -- leaw modelist, %si -- movw %di, %dx -- movw %si, %di --cmt1: cmpw %dx, %si # Scan all modes -- jz cmt2 -- -- leaw modelist, %bx # Find in previous entries -- movw 2(%si), %cx --cmt3: cmpw %bx, %si -- jz cmt4 -- -- cmpw 2(%bx), %cx # Found => don't copy this entry -- jz cmt5 -- -- addw $4, %bx -- jmp cmt3 -- --cmt4: movsl # Copy entry -- jmp cmt1 -- --cmt5: addw $4, %si # Skip entry -- jmp cmt1 -- --cmt2: --#endif /* CONFIG_VIDEO_COMPACT */ -- -- movw $ASK_VGA, (%di) # End marker -- movw %di, mt_end --mtab1: leaw modelist, %si # SI=mode list, DI=list end --ret0: ret -- --# Modes usable on all standard VGAs --vga_modes: -- .word VIDEO_8POINT -- .word 0x5032 # 80x50 -- .word VIDEO_80x43 -- .word 0x502b # 80x43 -- .word VIDEO_80x28 -- .word 0x501c # 80x28 -- .word VIDEO_80x30 -- .word 0x501e # 80x30 -- .word VIDEO_80x34 -- .word 0x5022 # 80x34 -- .word VIDEO_80x60 -- .word 0x503c # 80x60 --#ifdef CONFIG_VIDEO_GFX_HACK -- .word VIDEO_GFX_HACK -- .word VIDEO_GFX_DUMMY_RESOLUTION --#endif -- --vga_modes_end: --# Detect VESA modes. -- --#ifdef CONFIG_VIDEO_VESA --vesa_modes: -- cmpb $2, adapter # VGA only -- jnz ret0 -- -- movw %di, %bp # BP=original mode table end -- addw $0x200, %di # Buffer space -- movw $0x4f00, %ax # VESA Get card info call -- int $0x10 -- movw %bp, %di -- cmpw $0x004f, %ax # Successful? -- jnz ret0 -- -- cmpw $0x4556, 0x200(%di) -- jnz ret0 -- -- cmpw $0x4153, 0x202(%di) -- jnz ret0 -- -- movw $vesa_name, card_name # Set name to "VESA VGA" -- pushw %gs -- lgsw 0x20e(%di), %si # GS:SI=mode list -- movw $128, %cx # Iteration limit --vesa1: --# gas version 2.9.1, using BFD version 2.9.1.0.23 buggers the next inst. --# XXX: lodsw %gs:(%si), %ax # Get next mode in the list -- gs; lodsw -- cmpw $0xffff, %ax # End of the table? -- jz vesar -- -- cmpw $0x0080, %ax # Check validity of mode ID -- jc vesa2 -- -- orb %ah, %ah # Valid IDs: 0x0000-0x007f/0x0100-0x07ff -- jz vesan # Certain BIOSes report 0x80-0xff! -- -- cmpw $0x0800, %ax -- jnc vesae -- --vesa2: pushw %cx -- movw %ax, %cx # Get mode information structure -- movw $0x4f01, %ax -- int $0x10 -- movw %cx, %bx # BX=mode number -- addb $VIDEO_FIRST_VESA>>8, %bh -- popw %cx -- cmpw $0x004f, %ax -- jnz vesan # Don't report errors (buggy BIOSES) -- -- movb (%di), %al # Check capabilities. We require -- andb $0x19, %al # a color text mode. -- cmpb $0x09, %al -- jnz vesan -- -- cmpw $0xb800, 8(%di) # Standard video memory address required -- jnz vesan -- -- testb $2, (%di) # Mode characteristics supplied? -- movw %bx, (%di) # Store mode number -- jz vesa3 -- -- xorw %dx, %dx -- movw 0x12(%di), %bx # Width -- orb %bh, %bh -- jnz vesan -- -- movb %bl, 0x3(%di) -- movw 0x14(%di), %ax # Height -- orb %ah, %ah -- jnz vesan -- -- movb %al, 2(%di) -- mulb %bl -- cmpw $8193, %ax # Small enough for Linux console driver? -- jnc vesan -- -- jmp vesaok -- --vesa3: subw $0x8108, %bx # This mode has no detailed info specified, -- jc vesan # so it must be a standard VESA mode. -- -- cmpw $5, %bx -- jnc vesan -- -- movw vesa_text_mode_table(%bx), %ax -- movw %ax, 2(%di) --vesaok: addw $4, %di # The mode is valid. Store it. --vesan: loop vesa1 # Next mode. Limit exceeded => error --vesae: leaw vesaer, %si -- call prtstr -- movw %bp, %di # Discard already found modes. --vesar: popw %gs -- ret -- --# Dimensions of standard VESA text modes --vesa_text_mode_table: -- .byte 60, 80 # 0108 -- .byte 25, 132 # 0109 -- .byte 43, 132 # 010A -- .byte 50, 132 # 010B -- .byte 60, 132 # 010C --#endif /* CONFIG_VIDEO_VESA */ -- --# Scan for video modes. A bit dirty, but should work. --mode_scan: -- movw $0x0100, %cx # Start with mode 0 --scm1: movb $0, %ah # Test the mode -- movb %cl, %al -- int $0x10 -- movb $0x0f, %ah -- int $0x10 -- cmpb %cl, %al -- jnz scm2 # Mode not set -- -- movw $0x3c0, %dx # Test if it's a text mode -- movb $0x10, %al # Mode bits -- call inidx -- andb $0x03, %al -- jnz scm2 -- -- movb $0xce, %dl # Another set of mode bits -- movb $0x06, %al -- call inidx -- shrb %al -- jc scm2 -- -- movb $0xd4, %dl # Cursor location -- movb $0x0f, %al -- call inidx -- orb %al, %al -- jnz scm2 -- -- movw %cx, %ax # Ok, store the mode -- stosw -- movb %gs:(0x484), %al # Number of rows -- incb %al -- stosb -- movw %gs:(0x44a), %ax # Number of columns -- stosb --scm2: incb %cl -- jns scm1 -- -- movw $0x0003, %ax # Return back to mode 3 -- int $0x10 -- ret -- --tstidx: outw %ax, %dx # OUT DX,AX and inidx --inidx: outb %al, %dx # Read from indexed VGA register -- incw %dx # AL=index, DX=index reg port -> AL=data -- inb %dx, %al -- decw %dx -- ret -- --# Try to detect type of SVGA card and supply (usually approximate) video --# mode table for it. -- --#ifdef CONFIG_VIDEO_SVGA --svga_modes: -- leaw svga_table, %si # Test all known SVGA adapters --dosvga: lodsw -- movw %ax, %bp # Default mode table -- orw %ax, %ax -- jz didsv1 -- -- lodsw # Pointer to test routine -- pushw %si -- pushw %di -- pushw %es -- movw $0xc000, %bx -- movw %bx, %es -- call *%ax # Call test routine -- popw %es -- popw %di -- popw %si -- orw %bp, %bp -- jz dosvga -- -- movw %bp, %si # Found, copy the modes -- movb svga_prefix, %ah --cpsvga: lodsb -- orb %al, %al -- jz didsv -- -- stosw -- movsw -- jmp cpsvga -- --didsv: movw %si, card_name # Store pointer to card name --didsv1: ret -- --# Table of all known SVGA cards. For each card, we store a pointer to --# a table of video modes supported by the card and a pointer to a routine --# used for testing of presence of the card. The video mode table is always --# followed by the name of the card or the chipset. --svga_table: -- .word ati_md, ati_test -- .word oak_md, oak_test -- .word paradise_md, paradise_test -- .word realtek_md, realtek_test -- .word s3_md, s3_test -- .word chips_md, chips_test -- .word video7_md, video7_test -- .word cirrus5_md, cirrus5_test -- .word cirrus6_md, cirrus6_test -- .word cirrus1_md, cirrus1_test -- .word ahead_md, ahead_test -- .word everex_md, everex_test -- .word genoa_md, genoa_test -- .word trident_md, trident_test -- .word tseng_md, tseng_test -- .word 0 -- --# Test routines and mode tables: -- --# S3 - The test algorithm was taken from the SuperProbe package --# for XFree86 1.2.1. Report bugs to Christoph.Niemann@linux.org --s3_test: -- movw $0x0f35, %cx # we store some constants in cl/ch -- movw $0x03d4, %dx -- movb $0x38, %al -- call inidx -- movb %al, %bh # store current CRT-register 0x38 -- movw $0x0038, %ax -- call outidx # disable writing to special regs -- movb %cl, %al # check whether we can write special reg 0x35 -- call inidx -- movb %al, %bl # save the current value of CRT reg 0x35 -- andb $0xf0, %al # clear bits 0-3 -- movb %al, %ah -- movb %cl, %al # and write it to CRT reg 0x35 -- call outidx -- call inidx # now read it back -- andb %ch, %al # clear the upper 4 bits -- jz s3_2 # the first test failed. But we have a -- -- movb %bl, %ah # second chance -- movb %cl, %al -- call outidx -- jmp s3_1 # do the other tests -- --s3_2: movw %cx, %ax # load ah with 0xf and al with 0x35 -- orb %bl, %ah # set the upper 4 bits of ah with the orig value -- call outidx # write ... -- call inidx # ... and reread -- andb %cl, %al # turn off the upper 4 bits -- pushw %ax -- movb %bl, %ah # restore old value in register 0x35 -- movb %cl, %al -- call outidx -- popw %ax -- cmpb %ch, %al # setting lower 4 bits was successful => bad -- je no_s3 # writing is allowed => this is not an S3 -- --s3_1: movw $0x4838, %ax # allow writing to special regs by putting -- call outidx # magic number into CRT-register 0x38 -- movb %cl, %al # check whether we can write special reg 0x35 -- call inidx -- movb %al, %bl -- andb $0xf0, %al -- movb %al, %ah -- movb %cl, %al -- call outidx -- call inidx -- andb %ch, %al -- jnz no_s3 # no, we can't write => no S3 -- -- movw %cx, %ax -- orb %bl, %ah -- call outidx -- call inidx -- andb %ch, %al -- pushw %ax -- movb %bl, %ah # restore old value in register 0x35 -- movb %cl, %al -- call outidx -- popw %ax -- cmpb %ch, %al -- jne no_s31 # writing not possible => no S3 -- movb $0x30, %al -- call inidx # now get the S3 id ... -- leaw idS3, %di -- movw $0x10, %cx -- repne -- scasb -- je no_s31 -- -- movb %bh, %ah -- movb $0x38, %al -- jmp s3rest -- --no_s3: movb $0x35, %al # restore CRT register 0x35 -- movb %bl, %ah -- call outidx --no_s31: xorw %bp, %bp # Detection failed --s3rest: movb %bh, %ah -- movb $0x38, %al # restore old value of CRT register 0x38 -- jmp outidx -- --idS3: .byte 0x81, 0x82, 0x90, 0x91, 0x92, 0x93, 0x94, 0x95 -- .byte 0xa0, 0xa1, 0xa2, 0xa3, 0xa4, 0xa5, 0xa8, 0xb0 -- --s3_md: .byte 0x54, 0x2b, 0x84 -- .byte 0x55, 0x19, 0x84 -- .byte 0 -- .ascii "S3" -- .byte 0 -- --# ATI cards. --ati_test: -- leaw idati, %si -- movw $0x31, %di -- movw $0x09, %cx -- repe -- cmpsb -- je atiok -- -- xorw %bp, %bp --atiok: ret -- --idati: .ascii "761295520" -- --ati_md: .byte 0x23, 0x19, 0x84 -- .byte 0x33, 0x2c, 0x84 -- .byte 0x22, 0x1e, 0x64 -- .byte 0x21, 0x19, 0x64 -- .byte 0x58, 0x21, 0x50 -- .byte 0x5b, 0x1e, 0x50 -- .byte 0 -- .ascii "ATI" -- .byte 0 -- --# AHEAD --ahead_test: -- movw $0x200f, %ax -- movw $0x3ce, %dx -- outw %ax, %dx -- incw %dx -- inb %dx, %al -- cmpb $0x20, %al -- je isahed -- -- cmpb $0x21, %al -- je isahed -- -- xorw %bp, %bp --isahed: ret -- --ahead_md: -- .byte 0x22, 0x2c, 0x84 -- .byte 0x23, 0x19, 0x84 -- .byte 0x24, 0x1c, 0x84 -- .byte 0x2f, 0x32, 0xa0 -- .byte 0x32, 0x22, 0x50 -- .byte 0x34, 0x42, 0x50 -- .byte 0 -- .ascii "Ahead" -- .byte 0 -- --# Chips & Tech. --chips_test: -- movw $0x3c3, %dx -- inb %dx, %al -- orb $0x10, %al -- outb %al, %dx -- movw $0x104, %dx -- inb %dx, %al -- movb %al, %bl -- movw $0x3c3, %dx -- inb %dx, %al -- andb $0xef, %al -- outb %al, %dx -- cmpb $0xa5, %bl -- je cantok -- -- xorw %bp, %bp --cantok: ret -- --chips_md: -- .byte 0x60, 0x19, 0x84 -- .byte 0x61, 0x32, 0x84 -- .byte 0 -- .ascii "Chips & Technologies" -- .byte 0 -- --# Cirrus Logic 5X0 --cirrus1_test: -- movw $0x3d4, %dx -- movb $0x0c, %al -- outb %al, %dx -- incw %dx -- inb %dx, %al -- movb %al, %bl -- xorb %al, %al -- outb %al, %dx -- decw %dx -- movb $0x1f, %al -- outb %al, %dx -- incw %dx -- inb %dx, %al -- movb %al, %bh -- xorb %ah, %ah -- shlb $4, %al -- movw %ax, %cx -- movb %bh, %al -- shrb $4, %al -- addw %ax, %cx -- shlw $8, %cx -- addw $6, %cx -- movw %cx, %ax -- movw $0x3c4, %dx -- outw %ax, %dx -- incw %dx -- inb %dx, %al -- andb %al, %al -- jnz nocirr -- -- movb %bh, %al -- outb %al, %dx -- inb %dx, %al -- cmpb $0x01, %al -- je iscirr -- --nocirr: xorw %bp, %bp --iscirr: movw $0x3d4, %dx -- movb %bl, %al -- xorb %ah, %ah -- shlw $8, %ax -- addw $0x0c, %ax -- outw %ax, %dx -- ret -- --cirrus1_md: -- .byte 0x1f, 0x19, 0x84 -- .byte 0x20, 0x2c, 0x84 -- .byte 0x22, 0x1e, 0x84 -- .byte 0x31, 0x25, 0x64 -- .byte 0 -- .ascii "Cirrus Logic 5X0" -- .byte 0 -- --# Cirrus Logic 54XX --cirrus5_test: -- movw $0x3c4, %dx -- movb $6, %al -- call inidx -- movb %al, %bl # BL=backup -- movw $6, %ax -- call tstidx -- cmpb $0x0f, %al -- jne c5fail -- -- movw $0x1206, %ax -- call tstidx -- cmpb $0x12, %al -- jne c5fail -- -- movb $0x1e, %al -- call inidx -- movb %al, %bh -- movb %bh, %ah -- andb $0xc0, %ah -- movb $0x1e, %al -- call tstidx -- andb $0x3f, %al -- jne c5xx -- -- movb $0x1e, %al -- movb %bh, %ah -- orb $0x3f, %ah -- call tstidx -- xorb $0x3f, %al -- andb $0x3f, %al --c5xx: pushf -- movb $0x1e, %al -- movb %bh, %ah -- outw %ax, %dx -- popf -- je c5done -- --c5fail: xorw %bp, %bp --c5done: movb $6, %al -- movb %bl, %ah -- outw %ax, %dx -- ret -- --cirrus5_md: -- .byte 0x14, 0x19, 0x84 -- .byte 0x54, 0x2b, 0x84 -- .byte 0 -- .ascii "Cirrus Logic 54XX" -- .byte 0 -- --# Cirrus Logic 64XX -- no known extra modes, but must be identified, because --# it's misidentified by the Ahead test. --cirrus6_test: -- movw $0x3ce, %dx -- movb $0x0a, %al -- call inidx -- movb %al, %bl # BL=backup -- movw $0xce0a, %ax -- call tstidx -- orb %al, %al -- jne c2fail -- -- movw $0xec0a, %ax -- call tstidx -- cmpb $0x01, %al -- jne c2fail -- -- movb $0xaa, %al -- call inidx # 4X, 5X, 7X and 8X are valid 64XX chip ID's. -- shrb $4, %al -- subb $4, %al -- jz c6done -- -- decb %al -- jz c6done -- -- subb $2, %al -- jz c6done -- -- decb %al -- jz c6done -- --c2fail: xorw %bp, %bp --c6done: movb $0x0a, %al -- movb %bl, %ah -- outw %ax, %dx -- ret -- --cirrus6_md: -- .byte 0 -- .ascii "Cirrus Logic 64XX" -- .byte 0 -- --# Everex / Trident --everex_test: -- movw $0x7000, %ax -- xorw %bx, %bx -- int $0x10 -- cmpb $0x70, %al -- jne noevrx -- -- shrw $4, %dx -- cmpw $0x678, %dx -- je evtrid -- -- cmpw $0x236, %dx -- jne evrxok -- --evtrid: leaw trident_md, %bp --evrxok: ret -- --noevrx: xorw %bp, %bp -- ret -- --everex_md: -- .byte 0x03, 0x22, 0x50 -- .byte 0x04, 0x3c, 0x50 -- .byte 0x07, 0x2b, 0x64 -- .byte 0x08, 0x4b, 0x64 -- .byte 0x0a, 0x19, 0x84 -- .byte 0x0b, 0x2c, 0x84 -- .byte 0x16, 0x1e, 0x50 -- .byte 0x18, 0x1b, 0x64 -- .byte 0x21, 0x40, 0xa0 -- .byte 0x40, 0x1e, 0x84 -- .byte 0 -- .ascii "Everex/Trident" -- .byte 0 -- --# Genoa. --genoa_test: -- leaw idgenoa, %si # Check Genoa 'clues' -- xorw %ax, %ax -- movb %es:(0x37), %al -- movw %ax, %di -- movw $0x04, %cx -- decw %si -- decw %di --l1: incw %si -- incw %di -- movb (%si), %al -- testb %al, %al -- jz l2 -- -- cmpb %es:(%di), %al --l2: loope l1 -- orw %cx, %cx -- je isgen -- -- xorw %bp, %bp --isgen: ret -- --idgenoa: .byte 0x77, 0x00, 0x99, 0x66 -- --genoa_md: -- .byte 0x58, 0x20, 0x50 -- .byte 0x5a, 0x2a, 0x64 -- .byte 0x60, 0x19, 0x84 -- .byte 0x61, 0x1d, 0x84 -- .byte 0x62, 0x20, 0x84 -- .byte 0x63, 0x2c, 0x84 -- .byte 0x64, 0x3c, 0x84 -- .byte 0x6b, 0x4f, 0x64 -- .byte 0x72, 0x3c, 0x50 -- .byte 0x74, 0x42, 0x50 -- .byte 0x78, 0x4b, 0x64 -- .byte 0 -- .ascii "Genoa" -- .byte 0 -- --# OAK --oak_test: -- leaw idoakvga, %si -- movw $0x08, %di -- movw $0x08, %cx -- repe -- cmpsb -- je isoak -- -- xorw %bp, %bp --isoak: ret -- --idoakvga: .ascii "OAK VGA " -- --oak_md: .byte 0x4e, 0x3c, 0x50 -- .byte 0x4f, 0x3c, 0x84 -- .byte 0x50, 0x19, 0x84 -- .byte 0x51, 0x2b, 0x84 -- .byte 0 -- .ascii "OAK" -- .byte 0 -- --# WD Paradise. --paradise_test: -- leaw idparadise, %si -- movw $0x7d, %di -- movw $0x04, %cx -- repe -- cmpsb -- je ispara -- -- xorw %bp, %bp --ispara: ret -- --idparadise: .ascii "VGA=" -- --paradise_md: -- .byte 0x41, 0x22, 0x50 -- .byte 0x47, 0x1c, 0x84 -- .byte 0x55, 0x19, 0x84 -- .byte 0x54, 0x2c, 0x84 -- .byte 0 -- .ascii "Paradise" -- .byte 0 -- --# Trident. --trident_test: -- movw $0x3c4, %dx -- movb $0x0e, %al -- outb %al, %dx -- incw %dx -- inb %dx, %al -- xchgb %al, %ah -- xorb %al, %al -- outb %al, %dx -- inb %dx, %al -- xchgb %ah, %al -- movb %al, %bl # Strange thing ... in the book this wasn't -- andb $0x02, %bl # necessary but it worked on my card which -- jz setb2 # is a trident. Without it the screen goes -- # blurred ... -- andb $0xfd, %al -- jmp clrb2 -- --setb2: orb $0x02, %al --clrb2: outb %al, %dx -- andb $0x0f, %ah -- cmpb $0x02, %ah -- je istrid -- -- xorw %bp, %bp --istrid: ret -- --trident_md: -- .byte 0x50, 0x1e, 0x50 -- .byte 0x51, 0x2b, 0x50 -- .byte 0x52, 0x3c, 0x50 -- .byte 0x57, 0x19, 0x84 -- .byte 0x58, 0x1e, 0x84 -- .byte 0x59, 0x2b, 0x84 -- .byte 0x5a, 0x3c, 0x84 -- .byte 0 -- .ascii "Trident" -- .byte 0 -- --# Tseng. --tseng_test: -- movw $0x3cd, %dx -- inb %dx, %al # Could things be this simple ! :-) -- movb %al, %bl -- movb $0x55, %al -- outb %al, %dx -- inb %dx, %al -- movb %al, %ah -- movb %bl, %al -- outb %al, %dx -- cmpb $0x55, %ah -- je istsen -- --isnot: xorw %bp, %bp --istsen: ret -- --tseng_md: -- .byte 0x26, 0x3c, 0x50 -- .byte 0x2a, 0x28, 0x64 -- .byte 0x23, 0x19, 0x84 -- .byte 0x24, 0x1c, 0x84 -- .byte 0x22, 0x2c, 0x84 -- .byte 0x21, 0x3c, 0x84 -- .byte 0 -- .ascii "Tseng" -- .byte 0 -- --# Video7. --video7_test: -- movw $0x3cc, %dx -- inb %dx, %al -- movw $0x3b4, %dx -- andb $0x01, %al -- jz even7 -- -- movw $0x3d4, %dx --even7: movb $0x0c, %al -- outb %al, %dx -- incw %dx -- inb %dx, %al -- movb %al, %bl -- movb $0x55, %al -- outb %al, %dx -- inb %dx, %al -- decw %dx -- movb $0x1f, %al -- outb %al, %dx -- incw %dx -- inb %dx, %al -- movb %al, %bh -- decw %dx -- movb $0x0c, %al -- outb %al, %dx -- incw %dx -- movb %bl, %al -- outb %al, %dx -- movb $0x55, %al -- xorb $0xea, %al -- cmpb %bh, %al -- jne isnot -- -- movb $VIDEO_FIRST_V7>>8, svga_prefix # Use special mode switching -- ret -- --video7_md: -- .byte 0x40, 0x2b, 0x50 -- .byte 0x43, 0x3c, 0x50 -- .byte 0x44, 0x3c, 0x64 -- .byte 0x41, 0x19, 0x84 -- .byte 0x42, 0x2c, 0x84 -- .byte 0x45, 0x1c, 0x84 -- .byte 0 -- .ascii "Video 7" -- .byte 0 -- --# Realtek VGA --realtek_test: -- leaw idrtvga, %si -- movw $0x45, %di -- movw $0x0b, %cx -- repe -- cmpsb -- je isrt -- -- xorw %bp, %bp --isrt: ret -- --idrtvga: .ascii "REALTEK VGA" -- --realtek_md: -- .byte 0x1a, 0x3c, 0x50 -- .byte 0x1b, 0x19, 0x84 -- .byte 0x1c, 0x1e, 0x84 -- .byte 0x1d, 0x2b, 0x84 -- .byte 0x1e, 0x3c, 0x84 -- .byte 0 -- .ascii "REALTEK" -- .byte 0 -- --#endif /* CONFIG_VIDEO_SVGA */ -- --# User-defined local mode table (VGA only) --#ifdef CONFIG_VIDEO_LOCAL --local_modes: -- leaw local_mode_table, %si --locm1: lodsw -- orw %ax, %ax -- jz locm2 -- -- stosw -- movsw -- jmp locm1 -- --locm2: ret -- --# This is the table of local video modes which can be supplied manually --# by the user. Each entry consists of mode ID (word) and dimensions --# (byte for column count and another byte for row count). These modes --# are placed before all SVGA and VESA modes and override them if table --# compacting is enabled. The table must end with a zero word followed --# by NUL-terminated video adapter name. --local_mode_table: -- .word 0x0100 # Example: 40x25 -- .byte 25,40 -- .word 0 -- .ascii "Local" -- .byte 0 --#endif /* CONFIG_VIDEO_LOCAL */ -- --# Read a key and return the ASCII code in al, scan code in ah --getkey: xorb %ah, %ah -- int $0x16 -- ret -- --# Read a key with a timeout of 30 seconds. --# The hardware clock is used to get the time. --getkt: call gettime -- addb $30, %al # Wait 30 seconds -- cmpb $60, %al -- jl lminute -- -- subb $60, %al --lminute: -- movb %al, %cl --again: movb $0x01, %ah -- int $0x16 -- jnz getkey # key pressed, so get it -- -- call gettime -- cmpb %cl, %al -- jne again -- -- movb $0x20, %al # timeout, return `space' -- ret -- --# Flush the keyboard buffer --flush: movb $0x01, %ah -- int $0x16 -- jz empty -- -- xorb %ah, %ah -- int $0x16 -- jmp flush -- --empty: ret -- --# Print hexadecimal number. --prthw: pushw %ax -- movb %ah, %al -- call prthb -- popw %ax --prthb: pushw %ax -- shrb $4, %al -- call prthn -- popw %ax -- andb $0x0f, %al --prthn: cmpb $0x0a, %al -- jc prth1 -- -- addb $0x07, %al --prth1: addb $0x30, %al -- jmp prtchr -- --# Print decimal number in al --prtdec: pushw %ax -- pushw %cx -- xorb %ah, %ah -- movb $0x0a, %cl -- idivb %cl -- cmpb $0x09, %al -- jbe lt100 -- -- call prtdec -- jmp skip10 -- --lt100: addb $0x30, %al -- call prtchr --skip10: movb %ah, %al -- addb $0x30, %al -- call prtchr -- popw %cx -- popw %ax -- ret -- --store_edid: --#ifdef CONFIG_FIRMWARE_EDID -- pushw %es # just save all registers -- pushw %ax -- pushw %bx -- pushw %cx -- pushw %dx -- pushw %di -- -- pushw %fs -- popw %es -- -- movl $0x13131313, %eax # memset block with 0x13 -- movw $32, %cx -- movw $0x140, %di -- cld -- rep -- stosl -- -- cmpw $0x0200, vbe_version # only do EDID on >= VBE2.0 -- jl no_edid -- -- pushw %es # save ES -- xorw %di, %di # Report Capability -- pushw %di -- popw %es # ES:DI must be 0:0 -- movw $0x4f15, %ax -- xorw %bx, %bx -- xorw %cx, %cx -- int $0x10 -- popw %es # restore ES -- -- cmpb $0x00, %ah # call successful -- jne no_edid -- -- cmpb $0x4f, %al # function supported -- jne no_edid -- -- movw $0x4f15, %ax # do VBE/DDC -- movw $0x01, %bx -- movw $0x00, %cx -- movw $0x00, %dx -- movw $0x140, %di -- int $0x10 -- --no_edid: -- popw %di # restore all registers -- popw %dx -- popw %cx -- popw %bx -- popw %ax -- popw %es --#endif -- ret -- --# VIDEO_SELECT-only variables --mt_end: .word 0 # End of video mode table if built --edit_buf: .space 6 # Line editor buffer --card_name: .word 0 # Pointer to adapter name --scanning: .byte 0 # Performing mode scan --do_restore: .byte 0 # Screen contents altered during mode change --svga_prefix: .byte VIDEO_FIRST_BIOS>>8 # Default prefix for BIOS modes --graphic_mode: .byte 0 # Graphic mode with a linear frame buffer --dac_size: .byte 6 # DAC bit depth --vbe_version: .word 0 # VBE bios version -- --# Status messages --keymsg: .ascii "Press <RETURN> to see video modes available, " -- .ascii "<SPACE> to continue or wait 30 secs" -- .byte 0x0d, 0x0a, 0 -- --listhdr: .byte 0x0d, 0x0a -- .ascii "Mode: COLSxROWS:" -- --crlft: .byte 0x0d, 0x0a, 0 -- --prompt: .byte 0x0d, 0x0a -- .asciz "Enter mode number or `scan': " -- --unknt: .asciz "Unknown mode ID. Try again." -- --badmdt: .ascii "You passed an undefined mode number." -- .byte 0x0d, 0x0a, 0 -- --vesaer: .ascii "Error: Scanning of VESA modes failed. Please " -- .ascii "report to <mj@ucw.cz>." -- .byte 0x0d, 0x0a, 0 -- --old_name: .asciz "CGA/MDA/HGA" -- --ega_name: .asciz "EGA" -- --svga_name: .ascii " " -- --vga_name: .asciz "VGA" -- --vesa_name: .asciz "VESA" -- --name_bann: .asciz "Video adapter: " --#endif /* CONFIG_VIDEO_SELECT */ -- --# Other variables: --adapter: .byte 0 # Video adapter: 0=CGA/MDA/HGA,1=EGA,2=VGA --video_segment: .word 0xb800 # Video memory segment --force_size: .word 0 # Use this size instead of the one in BIOS vars ---- /dev/null -+++ b/arch/i386/boot/video.c -@@ -0,0 +1,456 @@ -+/* -*- linux-c -*- ------------------------------------------------------- * -+ * -+ * Copyright (C) 1991, 1992 Linus Torvalds -+ * Copyright 2007 rPath, Inc. - All Rights Reserved -+ * -+ * This file is part of the Linux kernel, and is made available under -+ * the terms of the GNU General Public License version 2. -+ * -+ * ----------------------------------------------------------------------- */ -+ -+/* -+ * arch/i386/boot/video.c -+ * -+ * Select video mode -+ */ -+ -+#include "boot.h" -+#include "video.h" -+#include "vesa.h" -+ -+/* -+ * Mode list variables -+ */ -+static struct card_info cards[]; /* List of cards to probe for */ -+ -+/* -+ * Common variables -+ */ -+int adapter; /* 0=CGA/MDA/HGC, 1=EGA, 2=VGA+ */ -+u16 video_segment; -+int force_x, force_y; /* Don't query the BIOS for cols/rows */ -+ -+int do_restore = 0; /* Screen contents changed during mode flip */ -+int graphic_mode; /* Graphic mode with linear frame buffer */ -+ -+static void store_cursor_position(void) -+{ -+ u16 curpos; -+ u16 ax, bx; -+ -+ ax = 0x0300; -+ bx = 0; -+ asm(INT10 -+ : "=d" (curpos), "+a" (ax), "+b" (bx) -+ : : "ecx", "esi", "edi"); -+ -+ boot_params.screen_info.orig_x = curpos; -+ boot_params.screen_info.orig_y = curpos >> 8; -+} -+ -+static void store_video_mode(void) -+{ -+ u16 ax, page; -+ -+ /* N.B.: the saving of the video page here is a bit silly, -+ since we pretty much assume page 0 everywhere. */ -+ ax = 0x0f00; -+ asm(INT10 -+ : "+a" (ax), "=b" (page) -+ : : "ecx", "edx", "esi", "edi"); -+ -+ /* Not all BIOSes are clean with respect to the top bit */ -+ boot_params.screen_info.orig_video_mode = ax & 0x7f; -+ boot_params.screen_info.orig_video_page = page; -+} -+ -+/* -+ * Store the video mode parameters for later usage by the kernel. -+ * This is done by asking the BIOS except for the rows/columns -+ * parameters in the default 80x25 mode -- these are set directly, -+ * because some very obscure BIOSes supply insane values. -+ */ -+static void store_mode_params(void) -+{ -+ u16 font_size; -+ int x, y; -+ -+ /* For graphics mode, it is up to the mode-setting driver -+ (currently only video-vesa.c) to store the parameters */ -+ if (graphic_mode) -+ return; -+ -+ store_cursor_position(); -+ store_video_mode(); -+ -+ if (boot_params.screen_info.orig_video_mode == 0x07) { -+ /* MDA, HGC, or VGA in monochrome mode */ -+ video_segment = 0xb000; -+ } else { -+ /* CGA, EGA, VGA and so forth */ -+ video_segment = 0xb800; -+ } -+ -+ set_fs(0); -+ font_size = rdfs16(0x485); /* Font size, BIOS area */ -+ boot_params.screen_info.orig_video_points = font_size; -+ -+ x = rdfs16(0x44a); -+ y = (adapter == ADAPTER_CGA) ? 25 : rdfs8(0x484)+1; -+ -+ if (force_x) -+ x = force_x; -+ if (force_y) -+ y = force_y; -+ -+ boot_params.screen_info.orig_video_cols = x; -+ boot_params.screen_info.orig_video_lines = y; -+} -+ -+/* Probe the video drivers and have them generate their mode lists. */ -+static void probe_cards(int unsafe) -+{ -+ struct card_info *card; -+ static u8 probed[2]; -+ -+ if (probed[unsafe]) -+ return; -+ -+ probed[unsafe] = 1; -+ -+ for (card = video_cards; card < video_cards_end; card++) { -+ if (card->unsafe == unsafe) { -+ if (card->probe) -+ card->nmodes = card->probe(); -+ else -+ card->nmodes = 0; -+ } -+ } -+} -+ -+/* Test if a mode is defined */ -+int mode_defined(u16 mode) -+{ -+ struct card_info *card; -+ struct mode_info *mi; -+ int i; -+ -+ for (card = video_cards; card < video_cards_end; card++) { -+ mi = card->modes; -+ for (i = 0; i < card->nmodes; i++, mi++) { -+ if (mi->mode == mode) -+ return 1; -+ } -+ } -+ -+ return 0; -+} -+ -+/* Set mode (without recalc) */ -+static int raw_set_mode(u16 mode) -+{ -+ int nmode, i; -+ struct card_info *card; -+ struct mode_info *mi; -+ -+ /* Drop the recalc bit if set */ -+ mode &= ~VIDEO_RECALC; -+ -+ /* Scan for mode based on fixed ID, position, or resolution */ -+ nmode = 0; -+ for (card = video_cards; card < video_cards_end; card++) { -+ mi = card->modes; -+ for (i = 0; i < card->nmodes; i++, mi++) { -+ int visible = mi->x || mi->y; -+ -+ if ((mode == nmode && visible) || -+ mode == mi->mode || -+ mode == (mi->y << 8)+mi->x) -+ return card->set_mode(mi); -+ -+ if (visible) -+ nmode++; -+ } -+ } -+ -+ /* Nothing found? Is it an "exceptional" (unprobed) mode? */ -+ for (card = video_cards; card < video_cards_end; card++) { -+ if (mode >= card->xmode_first && -+ mode < card->xmode_first+card->xmode_n) { -+ struct mode_info mix; -+ mix.mode = mode; -+ mix.x = mix.y = 0; -+ return card->set_mode(&mix); -+ } -+ } -+ -+ /* Otherwise, failure... */ -+ return -1; -+} -+ -+/* -+ * Recalculate the vertical video cutoff (hack!) -+ */ -+static void vga_recalc_vertical(void) -+{ -+ unsigned int font_size, rows; -+ u16 crtc; -+ u8 ov; -+ -+ set_fs(0); -+ font_size = rdfs8(0x485); /* BIOS: font size (pixels) */ -+ rows = force_y ? force_y : rdfs8(0x484)+1; /* Text rows */ -+ -+ rows *= font_size; /* Visible scan lines */ -+ rows--; /* ... minus one */ -+ -+ crtc = vga_crtc(); -+ -+ out_idx((u8)rows, crtc, 0x12); /* Lower height register */ -+ ov = in_idx(crtc, 0x07); /* Overflow register */ -+ ov &= 0xbd; -+ ov |= (rows >> (8-1)) & 0x02; -+ ov |= (rows >> (9-6)) & 0x40; -+ out_idx(ov, crtc, 0x07); -+} -+ -+/* Set mode (with recalc if specified) */ -+static int set_mode(u16 mode) -+{ -+ int rv; -+ -+ /* Very special mode numbers... */ -+ if (mode == VIDEO_CURRENT_MODE) -+ return 0; /* Nothing to do... */ -+ else if (mode == NORMAL_VGA) -+ mode = VIDEO_80x25; -+ else if (mode == EXTENDED_VGA) -+ mode = VIDEO_8POINT; -+ -+ rv = raw_set_mode(mode); -+ if (rv) -+ return rv; -+ -+ if (mode & VIDEO_RECALC) -+ vga_recalc_vertical(); -+ -+ return 0; -+} -+ -+static unsigned int get_entry(void) -+{ -+ char entry_buf[4]; -+ int i, len = 0; -+ int key; -+ unsigned int v; -+ -+ do { -+ key = getchar(); -+ -+ if (key == '\b') { -+ if (len > 0) { -+ puts("\b \b"); -+ len--; -+ } -+ } else if ((key >= '0' && key <= '9') || -+ (key >= 'A' && key <= 'Z') || -+ (key >= 'a' && key <= 'z')) { -+ if (len < sizeof entry_buf) { -+ entry_buf[len++] = key; -+ putchar(key); -+ } -+ } -+ } while (key != '\r'); -+ putchar('\n'); -+ -+ if (len == 0) -+ return VIDEO_CURRENT_MODE; /* Default */ -+ -+ v = 0; -+ for (i = 0; i < len; i++) { -+ v <<= 4; -+ key = entry_buf[i] | 0x20; -+ v += (key > '9') ? key-'a'+10 : key-'0'; -+ } -+ -+ return v; -+} -+ -+static void display_menu(void) -+{ -+ struct card_info *card; -+ struct mode_info *mi; -+ char ch; -+ int i; -+ -+ puts("Mode: COLSxROWS:\n"); -+ -+ ch = '0'; -+ for (card = video_cards; card < video_cards_end; card++) { -+ mi = card->modes; -+ for (i = 0; i < card->nmodes; i++, mi++) { -+ int visible = mi->x && mi->y; -+ u16 mode_id = mi->mode ? mi->mode : -+ (mi->y << 8)+mi->x; -+ -+ if (!visible) -+ continue; /* Hidden mode */ -+ -+ printf("%c %04X %3dx%-3d %s\n", -+ ch, mode_id, mi->x, mi->y, card->card_name); -+ -+ if (ch == '9') -+ ch = 'a'; -+ else if (ch == 'z' || ch == ' ') -+ ch = ' '; /* Out of keys... */ -+ else -+ ch++; -+ } -+ } -+} -+ -+#define H(x) ((x)-'a'+10) -+#define SCAN ((H('s')<<12)+(H('c')<<8)+(H('a')<<4)+H('n')) -+ -+static unsigned int mode_menu(void) -+{ -+ int key; -+ unsigned int sel; -+ -+ puts("Press <ENTER> to see video modes available, " -+ "<SPACE> to continue, or wait 30 sec\n"); -+ -+ kbd_flush(); -+ while (1) { -+ key = getchar_timeout(); -+ if (key == ' ' || key == 0) -+ return VIDEO_CURRENT_MODE; /* Default */ -+ if (key == '\r') -+ break; -+ putchar('\a'); /* Beep! */ -+ } -+ -+ -+ for (;;) { -+ display_menu(); -+ -+ puts("Enter a video mode or \"scan\" to scan for " -+ "additional modes: "); -+ sel = get_entry(); -+ if (sel != SCAN) -+ return sel; -+ -+ probe_cards(1); -+ } -+} -+ -+#ifdef CONFIG_VIDEO_RETAIN -+/* Save screen content to the heap */ -+struct saved_screen { -+ int x, y; -+ int curx, cury; -+ u16 *data; -+} saved; -+ -+static void save_screen(void) -+{ -+ /* Should be called after store_mode_params() */ -+ saved.x = boot_params.screen_info.orig_video_cols; -+ saved.y = boot_params.screen_info.orig_video_lines; -+ saved.curx = boot_params.screen_info.orig_x; -+ saved.cury = boot_params.screen_info.orig_y; -+ -+ if (heap_free() < saved.x*saved.y*sizeof(u16)+512) -+ return; /* Not enough heap to save the screen */ -+ -+ saved.data = GET_HEAP(u16, saved.x*saved.y); -+ -+ set_fs(video_segment); -+ copy_from_fs(saved.data, 0, saved.x*saved.y*sizeof(u16)); -+} -+ -+static void restore_screen(void) -+{ -+ /* Should be called after store_mode_params() */ -+ int xs = boot_params.screen_info.orig_video_cols; -+ int ys = boot_params.screen_info.orig_video_lines; -+ int y; -+ addr_t dst = 0; -+ u16 *src = saved.data; -+ u16 ax, bx, dx; -+ -+ if (graphic_mode) -+ return; /* Can't restore onto a graphic mode */ -+ -+ if (!src) -+ return; /* No saved screen contents */ -+ -+ /* Restore screen contents */ -+ -+ set_fs(video_segment); -+ for (y = 0; y < ys; y++) { -+ int npad; -+ -+ if (y < saved.y) { -+ int copy = (xs < saved.x) ? xs : saved.x; -+ copy_to_fs(dst, src, copy*sizeof(u16)); -+ dst += copy*sizeof(u16); -+ src += saved.x; -+ npad = (xs < saved.x) ? 0 : xs-saved.x; -+ } else { -+ npad = xs; -+ } -+ -+ /* Writes "npad" blank characters to -+ video_segment:dst and advances dst */ -+ asm volatile("pushw %%es ; " -+ "movw %2,%%es ; " -+ "shrw %%cx ; " -+ "jnc 1f ; " -+ "stosw \n\t" -+ "1: rep;stosl ; " -+ "popw %%es" -+ : "+D" (dst), "+c" (npad) -+ : "bdSm" (video_segment), -+ "a" (0x07200720)); -+ } -+ -+ /* Restore cursor position */ -+ ax = 0x0200; /* Set cursor position */ -+ bx = 0; /* Page number (<< 8) */ -+ dx = (saved.cury << 8)+saved.curx; -+ asm volatile(INT10 -+ : "+a" (ax), "+b" (bx), "+d" (dx) -+ : : "ecx", "esi", "edi"); -+} -+#else -+#define save_screen() ((void)0) -+#define restore_screen() ((void)0) -+#endif -+ -+void set_video(void) -+{ -+ u16 mode = boot_params.hdr.vid_mode; -+ -+ RESET_HEAP(); -+ -+ store_mode_params(); -+ save_screen(); -+ probe_cards(0); -+ -+ for (;;) { -+ if (mode == ASK_VGA) -+ mode = mode_menu(); -+ -+ if (!set_mode(mode)) -+ break; -+ -+ printf("Undefined video mode number: %x\n", mode); -+ mode = ASK_VGA; -+ } -+ vesa_store_edid(); -+ store_mode_params(); -+ -+ if (do_restore) -+ restore_screen(); -+} ---- /dev/null -+++ b/arch/i386/boot/video.h -@@ -0,0 +1,145 @@ -+/* -*- linux-c -*- ------------------------------------------------------- * -+ * -+ * Copyright (C) 1991, 1992 Linus Torvalds -+ * Copyright 2007 rPath, Inc. - All Rights Reserved -+ * -+ * This file is part of the Linux kernel, and is made available under -+ * the terms of the GNU General Public License version 2. -+ * -+ * ----------------------------------------------------------------------- */ -+ -+/* -+ * arch/i386/boot/video.h -+ * -+ * Header file for the real-mode video probing code -+ */ -+ -+#ifndef BOOT_VIDEO_H -+#define BOOT_VIDEO_H -+ -+#include <linux/types.h> -+ -+/* Enable autodetection of SVGA adapters and modes. */ -+#undef CONFIG_VIDEO_SVGA -+ -+/* Enable autodetection of VESA modes */ -+#define CONFIG_VIDEO_VESA -+ -+/* Retain screen contents when switching modes */ -+#define CONFIG_VIDEO_RETAIN -+ -+/* Force 400 scan lines for standard modes (hack to fix bad BIOS behaviour */ -+#undef CONFIG_VIDEO_400_HACK -+ -+/* This code uses an extended set of video mode numbers. These include: -+ * Aliases for standard modes -+ * NORMAL_VGA (-1) -+ * EXTENDED_VGA (-2) -+ * ASK_VGA (-3) -+ * Video modes numbered by menu position -- NOT RECOMMENDED because of lack -+ * of compatibility when extending the table. These are between 0x00 and 0xff. -+ */ -+#define VIDEO_FIRST_MENU 0x0000 -+ -+/* Standard BIOS video modes (BIOS number + 0x0100) */ -+#define VIDEO_FIRST_BIOS 0x0100 -+ -+/* VESA BIOS video modes (VESA number + 0x0200) */ -+#define VIDEO_FIRST_VESA 0x0200 -+ -+/* Video7 special modes (BIOS number + 0x0900) */ -+#define VIDEO_FIRST_V7 0x0900 -+ -+/* Special video modes */ -+#define VIDEO_FIRST_SPECIAL 0x0f00 -+#define VIDEO_80x25 0x0f00 -+#define VIDEO_8POINT 0x0f01 -+#define VIDEO_80x43 0x0f02 -+#define VIDEO_80x28 0x0f03 -+#define VIDEO_CURRENT_MODE 0x0f04 -+#define VIDEO_80x30 0x0f05 -+#define VIDEO_80x34 0x0f06 -+#define VIDEO_80x60 0x0f07 -+#define VIDEO_GFX_HACK 0x0f08 -+#define VIDEO_LAST_SPECIAL 0x0f09 -+ -+/* Video modes given by resolution */ -+#define VIDEO_FIRST_RESOLUTION 0x1000 -+ -+/* The "recalculate timings" flag */ -+#define VIDEO_RECALC 0x8000 -+ -+/* Define DO_STORE according to CONFIG_VIDEO_RETAIN */ -+#ifdef CONFIG_VIDEO_RETAIN -+void store_screen(void); -+#define DO_STORE() store_screen() -+#else -+#define DO_STORE() ((void)0) -+#endif /* CONFIG_VIDEO_RETAIN */ -+ -+/* -+ * Mode table structures -+ */ -+ -+struct mode_info { -+ u16 mode; /* Mode number (vga= style) */ -+ u8 x, y; /* Width, height */ -+}; -+ -+struct card_info { -+ const char *card_name; -+ int (*set_mode)(struct mode_info *mode); -+ int (*probe)(void); -+ struct mode_info *modes; -+ int nmodes; /* Number of probed modes so far */ -+ int unsafe; /* Probing is unsafe, only do after "scan" */ -+ u16 xmode_first; /* Unprobed modes to try to call anyway */ -+ u16 xmode_n; /* Size of unprobed mode range */ -+}; -+ -+#define __videocard struct card_info __attribute__((section(".videocards"))) -+extern struct card_info video_cards[], video_cards_end[]; -+ -+int mode_defined(u16 mode); /* video.c */ -+ -+/* Basic video information */ -+#define ADAPTER_CGA 0 /* CGA/MDA/HGC */ -+#define ADAPTER_EGA 1 -+#define ADAPTER_VGA 2 -+ -+extern int adapter; -+extern u16 video_segment; -+extern int force_x, force_y; /* Don't query the BIOS for cols/rows */ -+extern int do_restore; /* Restore screen contents */ -+extern int graphic_mode; /* Graphics mode with linear frame buffer */ -+ -+/* -+ * int $0x10 is notorious for touching registers it shouldn't. -+ * gcc doesn't like %ebp being clobbered, so define it as a push/pop -+ * sequence here. -+ */ -+#define INT10 "pushl %%ebp; int $0x10; popl %%ebp" -+ -+/* Accessing VGA indexed registers */ -+static inline u8 in_idx(u16 port, u8 index) -+{ -+ outb(index, port); -+ return inb(port+1); -+} -+ -+static inline void out_idx(u8 v, u16 port, u8 index) -+{ -+ outw(index+(v << 8), port); -+} -+ -+/* Writes a value to an indexed port and then reads the port again */ -+static inline u8 tst_idx(u8 v, u16 port, u8 index) -+{ -+ out_idx(port, index, v); -+ return in_idx(port, index); -+} -+ -+/* Get the I/O port of the VGA CRTC */ -+u16 vga_crtc(void); /* video-vga.c */ -+ -+#endif /* BOOT_VIDEO_H */ ---- /dev/null -+++ b/arch/i386/boot/voyager.c -@@ -0,0 +1,46 @@ -+/* -*- linux-c -*- ------------------------------------------------------- * -+ * -+ * Copyright (C) 1991, 1992 Linus Torvalds -+ * Copyright 2007 rPath, Inc. - All Rights Reserved -+ * -+ * This file is part of the Linux kernel, and is made available under -+ * the terms of the GNU General Public License version 2. -+ * -+ * ----------------------------------------------------------------------- */ -+ -+/* -+ * arch/i386/boot/voyager.c -+ * -+ * Get the Voyager config information -+ */ -+ -+#include "boot.h" -+ -+#ifdef CONFIG_X86_VOYAGER -+ -+int query_voyager(void) -+{ -+ u8 err; -+ u16 es, di; -+ /* Abuse the apm_bios_info area for this */ -+ u8 *data_ptr = (u8 *)&boot_params.apm_bios_info; -+ -+ data_ptr[0] = 0xff; /* Flag on config not found(?) */ -+ -+ asm("pushw %%es ; " -+ "int $0x15 ; " -+ "setc %0 ; " -+ "movw %%es, %1 ; " -+ "popw %%es" -+ : "=qm" (err), "=rm" (es), "=D" (di) -+ : "a" (0xffc0)); -+ -+ if (err) -+ return -1; /* Not Voyager */ -+ -+ set_fs(es); -+ copy_from_fs(data_ptr, di, 7); /* Table is 7 bytes apparently */ -+ return 0; -+} -+ -+#endif /* CONFIG_X86_VOYAGER */ ---- /dev/null -+++ b/arch/i386/kernel/cpu/addon_cpuid_features.c -@@ -0,0 +1,50 @@ -+ -+/* -+ * Routines to indentify additional cpu features that are scattered in -+ * cpuid space. -+ */ -+ -+#include <linux/cpu.h> -+ -+#include <asm/processor.h> -+ -+struct cpuid_bit { -+ u16 feature; -+ u8 reg; -+ u8 bit; -+ u32 level; -+}; -+ -+enum cpuid_regs { -+ CR_EAX = 0, -+ CR_ECX, -+ CR_EDX, -+ CR_EBX -+}; -+ -+void __cpuinit init_scattered_cpuid_features(struct cpuinfo_x86 *c) -+{ -+ u32 max_level; -+ u32 regs[4]; -+ const struct cpuid_bit *cb; -+ -+ static const struct cpuid_bit cpuid_bits[] = { -+ { X86_FEATURE_IDA, CR_EAX, 1, 0x00000006 }, -+ { 0, 0, 0, 0 } -+ }; -+ -+ for (cb = cpuid_bits; cb->feature; cb++) { -+ -+ /* Verify that the level is valid */ -+ max_level = cpuid_eax(cb->level & 0xffff0000); -+ if (max_level < cb->level || -+ max_level > (cb->level | 0xffff)) -+ continue; -+ -+ cpuid(cb->level, ®s[CR_EAX], ®s[CR_EBX], -+ ®s[CR_ECX], ®s[CR_EDX]); -+ -+ if (regs[cb->reg] & (1 << cb->bit)) -+ set_bit(cb->feature, c->x86_capability); -+ } -+} ---- a/arch/i386/kernel/cpu/common.c -+++ b/arch/i386/kernel/cpu/common.c -@@ -353,6 +353,8 @@ - if ( xlvl >= 0x80000004 ) - get_model_name(c); /* Default name */ - } -+ -+ init_scattered_cpuid_features(c); - } - - early_intel_workaround(c); ---- a/arch/i386/kernel/cpu/proc.c -+++ b/arch/i386/kernel/cpu/proc.c -@@ -29,7 +29,8 @@ - NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, - NULL, NULL, NULL, "syscall", NULL, NULL, NULL, NULL, - NULL, NULL, NULL, "mp", "nx", NULL, "mmxext", NULL, -- NULL, "fxsr_opt", "pdpe1gb", "rdtscp", NULL, "lm", "3dnowext", "3dnow", -+ NULL, "fxsr_opt", "pdpe1gb", "rdtscp", NULL, "lm", -+ "3dnowext", "3dnow", - - /* Transmeta-defined */ - "recovery", "longrun", NULL, "lrti", NULL, NULL, NULL, NULL, -@@ -40,8 +41,9 @@ - /* Other (Linux-defined) */ - "cxmmx", "k6_mtrr", "cyrix_arr", "centaur_mcr", - NULL, NULL, NULL, NULL, -- "constant_tsc", "up", NULL, NULL, NULL, NULL, NULL, NULL, -- NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, -+ "constant_tsc", "up", NULL, "arch_perfmon", -+ "pebs", "bts", NULL, "sync_rdtsc", -+ "rep_good", NULL, NULL, NULL, NULL, NULL, NULL, NULL, - NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, - - /* Intel-defined (#2) */ -@@ -57,9 +59,16 @@ - NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, - - /* AMD-defined (#2) */ -- "lahf_lm", "cmp_legacy", "svm", "extapic", "cr8legacy", "abm", -- "sse4a", "misalignsse", -- "3dnowprefetch", "osvw", "ibs", NULL, NULL, NULL, NULL, NULL, -+ "lahf_lm", "cmp_legacy", "svm", "extapic", "cr8_legacy", -+ "altmovcr8", "abm", "sse4a", -+ "misalignsse", "3dnowprefetch", -+ "osvw", "ibs", NULL, NULL, NULL, NULL, -+ NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, -+ NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, -+ -+ /* Auxiliary (Linux-defined) */ -+ "ida", NULL, NULL, NULL, NULL, NULL, NULL, NULL, -+ NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, - NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, - NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, - }; ---- a/arch/i386/kernel/e820.c -+++ b/arch/i386/kernel/e820.c -@@ -734,7 +734,7 @@ - case E820_NVS: - printk("(ACPI NVS)\n"); - break; -- default: printk("type %lu\n", e820.map[i].type); -+ default: printk("type %u\n", e820.map[i].type); - break; - } - } ---- a/arch/i386/kernel/setup.c -+++ b/arch/i386/kernel/setup.c -@@ -102,19 +102,10 @@ - /* - * Setup options - */ --struct drive_info_struct { char dummy[32]; } drive_info; --#if defined(CONFIG_BLK_DEV_IDE) || defined(CONFIG_BLK_DEV_HD) || \ -- defined(CONFIG_BLK_DEV_IDE_MODULE) || defined(CONFIG_BLK_DEV_HD_MODULE) --EXPORT_SYMBOL(drive_info); --#endif - struct screen_info screen_info; - EXPORT_SYMBOL(screen_info); - struct apm_info apm_info; - EXPORT_SYMBOL(apm_info); --struct sys_desc_table_struct { -- unsigned short length; -- unsigned char table[0]; --}; - struct edid_info edid_info; - EXPORT_SYMBOL_GPL(edid_info); - struct ist_info ist_info; -@@ -134,7 +125,7 @@ - - static char __initdata command_line[COMMAND_LINE_SIZE]; - --unsigned char __initdata boot_params[PARAM_SIZE]; -+struct boot_params __initdata boot_params; - - #if defined(CONFIG_EDD) || defined(CONFIG_EDD_MODULE) - struct edd edd; -@@ -528,7 +519,6 @@ - #endif - - ROOT_DEV = old_decode_dev(ORIG_ROOT_DEV); -- drive_info = DRIVE_INFO; - screen_info = SCREEN_INFO; - edid_info = EDID_INFO; - apm_info.bios = APM_BIOS_INFO; ---- a/arch/i386/kernel/verify_cpu.S -+++ /dev/null -@@ -1,94 +0,0 @@ --/* Check if CPU has some minimum CPUID bits -- This runs in 16bit mode so that the caller can still use the BIOS -- to output errors on the screen */ --#include <asm/cpufeature.h> --#include <asm/msr.h> -- --verify_cpu: -- pushfl # Save caller passed flags -- pushl $0 # Kill any dangerous flags -- popfl -- --#if CONFIG_X86_MINIMUM_CPU_MODEL >= 4 -- pushfl -- pop %eax -- orl $(1<<18),%eax # try setting AC -- push %eax -- popfl -- pushfl -- popl %eax -- testl $(1<<18),%eax -- jz bad --#endif --#if REQUIRED_MASK1 != 0 -- pushfl # standard way to check for cpuid -- popl %eax -- movl %eax,%ebx -- xorl $0x200000,%eax -- pushl %eax -- popfl -- pushfl -- popl %eax -- cmpl %eax,%ebx -- pushfl # standard way to check for cpuid -- popl %eax -- movl %eax,%ebx -- xorl $0x200000,%eax -- pushl %eax -- popfl -- pushfl -- popl %eax -- cmpl %eax,%ebx -- jz bad # REQUIRED_MASK1 != 0 requires CPUID -- -- movl $0x0,%eax # See if cpuid 1 is implemented -- cpuid -- cmpl $0x1,%eax -- jb bad # no cpuid 1 -- --#if REQUIRED_MASK1 & NEED_CMPXCHG64 -- /* Some VIA C3s need magic MSRs to enable CX64. Do this here */ -- cmpl $0x746e6543,%ebx # Cent -- jne 1f -- cmpl $0x48727561,%edx # aurH -- jne 1f -- cmpl $0x736c7561,%ecx # auls -- jne 1f -- movl $1,%eax # check model -- cpuid -- movl %eax,%ebx -- shr $8,%ebx -- andl $0xf,%ebx -- cmp $6,%ebx # check family == 6 -- jne 1f -- shr $4,%eax -- andl $0xf,%eax -- cmpl $6,%eax # check model >= 6 -- jb 1f -- # assume models >= 6 all support this MSR -- movl $MSR_VIA_FCR,%ecx -- rdmsr -- orl $((1<<1)|(1<<7)),%eax # enable CMPXCHG64 and PGE -- wrmsr --1: --#endif -- movl $0x1,%eax # Does the cpu have what it takes -- cpuid -- --#if CONFIG_X86_MINIMUM_CPU_MODEL > 4 --#error add proper model checking here --#endif -- -- andl $REQUIRED_MASK1,%edx -- xorl $REQUIRED_MASK1,%edx -- jnz bad --#endif /* REQUIRED_MASK1 */ -- -- popfl -- xor %eax,%eax -- ret -- --bad: -- popfl -- movl $1,%eax -- ret ---- a/arch/x86_64/Kconfig -+++ b/arch/x86_64/Kconfig -@@ -427,6 +427,10 @@ - This is purely to save memory - each supported CPU requires - memory in the static kernel configuration. - -+config PHYSICAL_ALIGN -+ hex -+ default "0x200000" -+ - config HOTPLUG_CPU - bool "Support for suspend on SMP and hot-pluggable CPUs (EXPERIMENTAL)" - depends on SMP && HOTPLUG && EXPERIMENTAL ---- a/arch/x86_64/boot/Makefile -+++ b/arch/x86_64/boot/Makefile -@@ -1,135 +1,9 @@ - # - # arch/x86_64/boot/Makefile - # --# This file is subject to the terms and conditions of the GNU General Public --# License. See the file "COPYING" in the main directory of this archive --# for more details. --# --# Copyright (C) 1994 by Linus Torvalds --# -- --# ROOT_DEV specifies the default root-device when making the image. --# This can be either FLOPPY, CURRENT, /dev/xxxx or empty, in which case --# the default of FLOPPY is used by 'build'. -- --ROOT_DEV := CURRENT -- --# If you want to preset the SVGA mode, uncomment the next line and --# set SVGA_MODE to whatever number you want. --# Set it to -DSVGA_MODE=NORMAL_VGA if you just want the EGA/VGA mode. --# The number is the same as you would ordinarily press at bootup. -- --SVGA_MODE := -DSVGA_MODE=NORMAL_VGA -- --# If you want the RAM disk device, define this to be the size in blocks. -- --#RAMDISK := -DRAMDISK=512 -- --targets := vmlinux.bin bootsect bootsect.o \ -- setup setup.o bzImage mtools.conf -- --EXTRA_CFLAGS := -m32 -- --hostprogs-y := tools/build --HOST_EXTRACFLAGS += $(LINUXINCLUDE) --subdir- := compressed/ #Let make clean descend in compressed/ --# --------------------------------------------------------------------------- -- --$(obj)/bzImage: IMAGE_OFFSET := 0x100000 --$(obj)/bzImage: EXTRA_AFLAGS := $(SVGA_MODE) $(RAMDISK) -D__BIG_KERNEL__ --$(obj)/bzImage: BUILDFLAGS := -b -- --quiet_cmd_image = BUILD $@ --cmd_image = $(obj)/tools/build $(BUILDFLAGS) $(obj)/bootsect $(obj)/setup \ -- $(obj)/vmlinux.bin $(ROOT_DEV) > $@ -- --$(obj)/bzImage: $(obj)/bootsect $(obj)/setup \ -- $(obj)/vmlinux.bin $(obj)/tools/build FORCE -- $(call if_changed,image) -- @echo 'Kernel: $@ is ready' ' (#'`cat .version`')' -- --$(obj)/vmlinux.bin: $(obj)/compressed/vmlinux FORCE -- $(call if_changed,objcopy) -- --LDFLAGS_bootsect := -Ttext 0x0 -s --oformat binary --LDFLAGS_setup := -Ttext 0x0 -s --oformat binary -e begtext -- --$(obj)/setup $(obj)/bootsect: %: %.o FORCE -- $(call if_changed,ld) -- --$(obj)/compressed/vmlinux: FORCE -- $(Q)$(MAKE) $(build)=$(obj)/compressed IMAGE_OFFSET=$(IMAGE_OFFSET) $@ -- --# Set this if you want to pass append arguments to the zdisk/fdimage/isoimage kernel --FDARGS = --# Set this if you want an initrd included with the zdisk/fdimage/isoimage kernel --FDINITRD = -- --image_cmdline = default linux $(FDARGS) $(if $(FDINITRD),initrd=initrd.img,) -- --$(obj)/mtools.conf: $(src)/mtools.conf.in -- sed -e 's|@OBJ@|$(obj)|g' < $< > $@ -- --# This requires write access to /dev/fd0 --zdisk: $(BOOTIMAGE) $(obj)/mtools.conf -- MTOOLSRC=$(obj)/mtools.conf mformat a: ; sync -- syslinux /dev/fd0 ; sync -- echo '$(image_cmdline)' | \ -- MTOOLSRC=$(obj)/mtools.conf mcopy - a:syslinux.cfg -- if [ -f '$(FDINITRD)' ] ; then \ -- MTOOLSRC=$(obj)/mtools.conf mcopy '$(FDINITRD)' a:initrd.img ; \ -- fi -- MTOOLSRC=$(obj)/mtools.conf mcopy $(BOOTIMAGE) a:linux ; sync -- --# These require being root or having syslinux 2.02 or higher installed --fdimage fdimage144: $(BOOTIMAGE) $(obj)/mtools.conf -- dd if=/dev/zero of=$(obj)/fdimage bs=1024 count=1440 -- MTOOLSRC=$(obj)/mtools.conf mformat v: ; sync -- syslinux $(obj)/fdimage ; sync -- echo '$(image_cmdline)' | \ -- MTOOLSRC=$(obj)/mtools.conf mcopy - v:syslinux.cfg -- if [ -f '$(FDINITRD)' ] ; then \ -- MTOOLSRC=$(obj)/mtools.conf mcopy '$(FDINITRD)' v:initrd.img ; \ -- fi -- MTOOLSRC=$(obj)/mtools.conf mcopy $(BOOTIMAGE) v:linux ; sync -- --fdimage288: $(BOOTIMAGE) $(obj)/mtools.conf -- dd if=/dev/zero of=$(obj)/fdimage bs=1024 count=2880 -- MTOOLSRC=$(obj)/mtools.conf mformat w: ; sync -- syslinux $(obj)/fdimage ; sync -- echo '$(image_cmdline)' | \ -- MTOOLSRC=$(obj)/mtools.conf mcopy - w:syslinux.cfg -- if [ -f '$(FDINITRD)' ] ; then \ -- MTOOLSRC=$(obj)/mtools.conf mcopy '$(FDINITRD)' w:initrd.img ; \ -- fi -- MTOOLSRC=$(obj)/mtools.conf mcopy $(BOOTIMAGE) w:linux ; sync -- --isoimage: $(BOOTIMAGE) -- -rm -rf $(obj)/isoimage -- mkdir $(obj)/isoimage -- for i in lib lib64 share end ; do \ -- if [ -f /usr/$$i/syslinux/isolinux.bin ] ; then \ -- cp /usr/$$i/syslinux/isolinux.bin $(obj)/isoimage ; \ -- break ; \ -- fi ; \ -- if [ $$i = end ] ; then exit 1 ; fi ; \ -- done -- cp $(BOOTIMAGE) $(obj)/isoimage/linux -- echo '$(image_cmdline)' > $(obj)/isoimage/isolinux.cfg -- if [ -f '$(FDINITRD)' ] ; then \ -- cp '$(FDINITRD)' $(obj)/isoimage/initrd.img ; \ -- fi -- mkisofs -J -r -o $(obj)/image.iso -b isolinux.bin -c boot.cat \ -- -no-emul-boot -boot-load-size 4 -boot-info-table \ -- $(obj)/isoimage -- rm -rf $(obj)/isoimage -- --zlilo: $(BOOTIMAGE) -- if [ -f $(INSTALL_PATH)/vmlinuz ]; then mv $(INSTALL_PATH)/vmlinuz $(INSTALL_PATH)/vmlinuz.old; fi -- if [ -f $(INSTALL_PATH)/System.map ]; then mv $(INSTALL_PATH)/System.map $(INSTALL_PATH)/System.old; fi -- cat $(BOOTIMAGE) > $(INSTALL_PATH)/vmlinuz -- cp System.map $(INSTALL_PATH)/ -- if [ -x /sbin/lilo ]; then /sbin/lilo; else /etc/lilo/install; fi -+# The actual boot code is shared with i386 including the Makefile. -+# So tell kbuild that we fetch the code from i386 and include the -+# Makefile from i386 too. - --install: -- sh $(srctree)/$(src)/install.sh $(KERNELRELEASE) $(BOOTIMAGE) System.map "$(INSTALL_PATH)" -+src := arch/i386/boot -+include $(src)/Makefile ---- a/arch/x86_64/boot/bootsect.S -+++ /dev/null -@@ -1,98 +0,0 @@ --/* -- * bootsect.S Copyright (C) 1991, 1992 Linus Torvalds -- * -- * modified by Drew Eckhardt -- * modified by Bruce Evans (bde) -- * modified by Chris Noe (May 1999) (as86 -> gas) -- * gutted by H. Peter Anvin (Jan 2003) -- * -- * BIG FAT NOTE: We're in real mode using 64k segments. Therefore segment -- * addresses must be multiplied by 16 to obtain their respective linear -- * addresses. To avoid confusion, linear addresses are written using leading -- * hex while segment addresses are written as segment:offset. -- * -- */ -- --#include <asm/boot.h> -- --SETUPSECTS = 4 /* default nr of setup-sectors */ --BOOTSEG = 0x07C0 /* original address of boot-sector */ --INITSEG = DEF_INITSEG /* we move boot here - out of the way */ --SETUPSEG = DEF_SETUPSEG /* setup starts here */ --SYSSEG = DEF_SYSSEG /* system loaded at 0x10000 (65536) */ --SYSSIZE = DEF_SYSSIZE /* system size: # of 16-byte clicks */ -- /* to be loaded */ --ROOT_DEV = 0 /* ROOT_DEV is now written by "build" */ --SWAP_DEV = 0 /* SWAP_DEV is now written by "build" */ -- --#ifndef SVGA_MODE --#define SVGA_MODE ASK_VGA --#endif -- --#ifndef RAMDISK --#define RAMDISK 0 --#endif -- --#ifndef ROOT_RDONLY --#define ROOT_RDONLY 1 --#endif -- --.code16 --.text -- --.global _start --_start: -- -- # Normalize the start address -- jmpl $BOOTSEG, $start2 -- --start2: -- movw %cs, %ax -- movw %ax, %ds -- movw %ax, %es -- movw %ax, %ss -- movw $0x7c00, %sp -- sti -- cld -- -- movw $bugger_off_msg, %si -- --msg_loop: -- lodsb -- andb %al, %al -- jz die -- movb $0xe, %ah -- movw $7, %bx -- int $0x10 -- jmp msg_loop -- --die: -- # Allow the user to press a key, then reboot -- xorw %ax, %ax -- int $0x16 -- int $0x19 -- -- # int 0x19 should never return. In case it does anyway, -- # invoke the BIOS reset code... -- ljmp $0xf000,$0xfff0 -- -- --bugger_off_msg: -- .ascii "Direct booting from floppy is no longer supported.\r\n" -- .ascii "Please use a boot loader program instead.\r\n" -- .ascii "\n" -- .ascii "Remove disk and press any key to reboot . . .\r\n" -- .byte 0 -- -- -- # Kernel attributes; used by setup -- -- .org 497 --setup_sects: .byte SETUPSECTS --root_flags: .word ROOT_RDONLY --syssize: .word SYSSIZE --swap_dev: .word SWAP_DEV --ram_size: .word RAMDISK --vid_mode: .word SVGA_MODE --root_dev: .word ROOT_DEV --boot_flag: .word 0xAA55 ---- a/arch/x86_64/boot/compressed/Makefile -+++ b/arch/x86_64/boot/compressed/Makefile -@@ -7,11 +7,12 @@ - # - - targets := vmlinux vmlinux.bin vmlinux.bin.gz head.o misc.o piggy.o --EXTRA_AFLAGS := -traditional - --# cannot use EXTRA_CFLAGS because base CFLAGS contains -mkernel which conflicts with --# -m32 --CFLAGS := -m64 -D__KERNEL__ -Iinclude -O2 -fno-strict-aliasing -fPIC -mcmodel=small -fno-builtin -+CFLAGS := -m64 -D__KERNEL__ $(LINUXINCLUDE) -O2 \ -+ -fno-strict-aliasing -fPIC -mcmodel=small \ -+ $(call cc-option, -ffreestanding) \ -+ $(call cc-option, -fno-stack-protector) -+AFLAGS := $(CFLAGS) -D__ASSEMBLY__ - LDFLAGS := -m elf_x86_64 - - LDFLAGS_vmlinux := -T ---- a/arch/x86_64/boot/compressed/head.S -+++ b/arch/x86_64/boot/compressed/head.S -@@ -46,10 +46,10 @@ - * at and where we were actually loaded at. This can only be done - * with a short local call on x86. Nothing else will tell us what - * address we are running at. The reserved chunk of the real-mode -- * data at 0x34-0x3f are used as the stack for this calculation. -- * Only 4 bytes are needed. -+ * data at 0x1e4 (defined as a scratch field) are used as the stack -+ * for this calculation. Only 4 bytes are needed. - */ -- leal 0x40(%esi), %esp -+ leal (0x1e4+4)(%esi), %esp - call 1f - 1: popl %ebp - subl $1b, %ebp ---- a/arch/x86_64/boot/install.sh -+++ /dev/null -@@ -1,2 +0,0 @@ --#!/bin/sh --. $srctree/arch/i386/boot/install.sh ---- a/arch/x86_64/boot/mtools.conf.in -+++ /dev/null -@@ -1,17 +0,0 @@ --# --# mtools configuration file for "make (b)zdisk" --# -- --# Actual floppy drive --drive a: -- file="/dev/fd0" -- --# 1.44 MB floppy disk image --drive v: -- file="@OBJ@/fdimage" cylinders=80 heads=2 sectors=18 filter -- --# 2.88 MB floppy disk image (mostly for virtual uses) --drive w: -- file="@OBJ@/fdimage" cylinders=80 heads=2 sectors=36 filter -- -- ---- a/arch/x86_64/boot/setup.S -+++ /dev/null -@@ -1,826 +0,0 @@ --/* -- * setup.S Copyright (C) 1991, 1992 Linus Torvalds -- * -- * setup.s is responsible for getting the system data from the BIOS, -- * and putting them into the appropriate places in system memory. -- * both setup.s and system has been loaded by the bootblock. -- * -- * This code asks the bios for memory/disk/other parameters, and -- * puts them in a "safe" place: 0x90000-0x901FF, ie where the -- * boot-block used to be. It is then up to the protected mode -- * system to read them from there before the area is overwritten -- * for buffer-blocks. -- * -- * Move PS/2 aux init code to psaux.c -- * (troyer@saifr00.cfsat.Honeywell.COM) 03Oct92 -- * -- * some changes and additional features by Christoph Niemann, -- * March 1993/June 1994 (Christoph.Niemann@linux.org) -- * -- * add APM BIOS checking by Stephen Rothwell, May 1994 -- * (sfr@canb.auug.org.au) -- * -- * High load stuff, initrd support and position independency -- * by Hans Lermen & Werner Almesberger, February 1996 -- * <lermen@elserv.ffm.fgan.de>, <almesber@lrc.epfl.ch> -- * -- * Video handling moved to video.S by Martin Mares, March 1996 -- * <mj@k332.feld.cvut.cz> -- * -- * Extended memory detection scheme retwiddled by orc@pell.chi.il.us (david -- * parsons) to avoid loadlin confusion, July 1997 -- * -- * Transcribed from Intel (as86) -> AT&T (gas) by Chris Noe, May 1999. -- * <stiker@northlink.com> -- * -- * Fix to work around buggy BIOSes which don't use carry bit correctly -- * and/or report extended memory in CX/DX for e801h memory size detection -- * call. As a result the kernel got wrong figures. The int15/e801h docs -- * from Ralf Brown interrupt list seem to indicate AX/BX should be used -- * anyway. So to avoid breaking many machines (presumably there was a reason -- * to orginally use CX/DX instead of AX/BX), we do a kludge to see -- * if CX/DX have been changed in the e801 call and if so use AX/BX . -- * Michael Miller, April 2001 <michaelm@mjmm.org> -- * -- * Added long mode checking and SSE force. March 2003, Andi Kleen. -- */ -- --#include <asm/segment.h> --#include <linux/utsrelease.h> --#include <linux/compile.h> --#include <asm/boot.h> --#include <asm/e820.h> --#include <asm/page.h> --#include <asm/setup.h> -- --/* Signature words to ensure LILO loaded us right */ --#define SIG1 0xAA55 --#define SIG2 0x5A5A -- --INITSEG = DEF_INITSEG # 0x9000, we move boot here, out of the way --SYSSEG = DEF_SYSSEG # 0x1000, system loaded at 0x10000 (65536). --SETUPSEG = DEF_SETUPSEG # 0x9020, this is the current segment -- # ... and the former contents of CS -- --DELTA_INITSEG = SETUPSEG - INITSEG # 0x0020 -- --.code16 --.globl begtext, begdata, begbss, endtext, enddata, endbss -- --.text --begtext: --.data --begdata: --.bss --begbss: --.text -- --start: -- jmp trampoline -- --# This is the setup header, and it must start at %cs:2 (old 0x9020:2) -- -- .ascii "HdrS" # header signature -- .word 0x0206 # header version number (>= 0x0105) -- # or else old loadlin-1.5 will fail) --realmode_swtch: .word 0, 0 # default_switch, SETUPSEG --start_sys_seg: .word SYSSEG -- .word kernel_version # pointing to kernel version string -- # above section of header is compatible -- # with loadlin-1.5 (header v1.5). Don't -- # change it. -- --type_of_loader: .byte 0 # = 0, old one (LILO, Loadlin, -- # Bootlin, SYSLX, bootsect...) -- # See Documentation/i386/boot.txt for -- # assigned ids -- --# flags, unused bits must be zero (RFU) bit within loadflags --loadflags: --LOADED_HIGH = 1 # If set, the kernel is loaded high --CAN_USE_HEAP = 0x80 # If set, the loader also has set -- # heap_end_ptr to tell how much -- # space behind setup.S can be used for -- # heap purposes. -- # Only the loader knows what is free --#ifndef __BIG_KERNEL__ -- .byte 0 --#else -- .byte LOADED_HIGH --#endif -- --setup_move_size: .word 0x8000 # size to move, when setup is not -- # loaded at 0x90000. We will move setup -- # to 0x90000 then just before jumping -- # into the kernel. However, only the -- # loader knows how much data behind -- # us also needs to be loaded. -- --code32_start: # here loaders can put a different -- # start address for 32-bit code. --#ifndef __BIG_KERNEL__ -- .long 0x1000 # 0x1000 = default for zImage --#else -- .long 0x100000 # 0x100000 = default for big kernel --#endif -- --ramdisk_image: .long 0 # address of loaded ramdisk image -- # Here the loader puts the 32-bit -- # address where it loaded the image. -- # This only will be read by the kernel. -- --ramdisk_size: .long 0 # its size in bytes -- --bootsect_kludge: -- .long 0 # obsolete -- --heap_end_ptr: .word modelist+1024 # (Header version 0x0201 or later) -- # space from here (exclusive) down to -- # end of setup code can be used by setup -- # for local heap purposes. -- --pad1: .word 0 --cmd_line_ptr: .long 0 # (Header version 0x0202 or later) -- # If nonzero, a 32-bit pointer -- # to the kernel command line. -- # The command line should be -- # located between the start of -- # setup and the end of low -- # memory (0xa0000), or it may -- # get overwritten before it -- # gets read. If this field is -- # used, there is no longer -- # anything magical about the -- # 0x90000 segment; the setup -- # can be located anywhere in -- # low memory 0x10000 or higher. -- --ramdisk_max: .long 0xffffffff --kernel_alignment: .long 0x200000 # physical addr alignment required for -- # protected mode relocatable kernel --#ifdef CONFIG_RELOCATABLE --relocatable_kernel: .byte 1 --#else --relocatable_kernel: .byte 0 --#endif --pad2: .byte 0 --pad3: .word 0 -- --cmdline_size: .long COMMAND_LINE_SIZE-1 #length of the command line, -- #added with boot protocol -- #version 2.06 -- --trampoline: call start_of_setup -- .align 16 -- # The offset at this point is 0x240 -- .space (0xeff-0x240+1) # E820 & EDD space (ending at 0xeff) --# End of setup header ##################################################### -- --start_of_setup: --# Bootlin depends on this being done early -- movw $0x01500, %ax -- movb $0x81, %dl -- int $0x13 -- --#ifdef SAFE_RESET_DISK_CONTROLLER --# Reset the disk controller. -- movw $0x0000, %ax -- movb $0x80, %dl -- int $0x13 --#endif -- --# Set %ds = %cs, we know that SETUPSEG = %cs at this point -- movw %cs, %ax # aka SETUPSEG -- movw %ax, %ds --# Check signature at end of setup -- cmpw $SIG1, setup_sig1 -- jne bad_sig -- -- cmpw $SIG2, setup_sig2 -- jne bad_sig -- -- jmp good_sig1 -- --# Routine to print asciiz string at ds:si --prtstr: -- lodsb -- andb %al, %al -- jz fin -- -- call prtchr -- jmp prtstr -- --fin: ret -- --# Space printing --prtsp2: call prtspc # Print double space --prtspc: movb $0x20, %al # Print single space (note: fall-thru) -- --prtchr: -- pushw %ax -- pushw %cx -- movw $0007,%bx -- movw $0x01, %cx -- movb $0x0e, %ah -- int $0x10 -- popw %cx -- popw %ax -- ret -- --beep: movb $0x07, %al -- jmp prtchr -- --no_sig_mess: .string "No setup signature found ..." -- --good_sig1: -- jmp good_sig -- --# We now have to find the rest of the setup code/data --bad_sig: -- movw %cs, %ax # SETUPSEG -- subw $DELTA_INITSEG, %ax # INITSEG -- movw %ax, %ds -- xorb %bh, %bh -- movb (497), %bl # get setup sect from bootsect -- subw $4, %bx # LILO loads 4 sectors of setup -- shlw $8, %bx # convert to words (1sect=2^8 words) -- movw %bx, %cx -- shrw $3, %bx # convert to segment -- addw $SYSSEG, %bx -- movw %bx, %cs:start_sys_seg --# Move rest of setup code/data to here -- movw $2048, %di # four sectors loaded by LILO -- subw %si, %si -- movw %cs, %ax # aka SETUPSEG -- movw %ax, %es -- movw $SYSSEG, %ax -- movw %ax, %ds -- rep -- movsw -- movw %cs, %ax # aka SETUPSEG -- movw %ax, %ds -- cmpw $SIG1, setup_sig1 -- jne no_sig -- -- cmpw $SIG2, setup_sig2 -- jne no_sig -- -- jmp good_sig -- --no_sig: -- lea no_sig_mess, %si -- call prtstr -- --no_sig_loop: -- jmp no_sig_loop -- --good_sig: -- movw %cs, %ax # aka SETUPSEG -- subw $DELTA_INITSEG, %ax # aka INITSEG -- movw %ax, %ds --# Check if an old loader tries to load a big-kernel -- testb $LOADED_HIGH, %cs:loadflags # Do we have a big kernel? -- jz loader_ok # No, no danger for old loaders. -- -- cmpb $0, %cs:type_of_loader # Do we have a loader that -- # can deal with us? -- jnz loader_ok # Yes, continue. -- -- pushw %cs # No, we have an old loader, -- popw %ds # die. -- lea loader_panic_mess, %si -- call prtstr -- -- jmp no_sig_loop -- --loader_panic_mess: .string "Wrong loader, giving up..." -- --loader_ok: -- /* check for long mode. */ -- /* we have to do this before the VESA setup, otherwise the user -- can't see the error message. */ -- -- pushw %ds -- movw %cs,%ax -- movw %ax,%ds -- -- call verify_cpu -- testl %eax,%eax -- jz sse_ok -- --no_longmode: -- call beep -- lea long_mode_panic,%si -- call prtstr --no_longmode_loop: -- jmp no_longmode_loop --long_mode_panic: -- .string "Your CPU does not support long mode. Use a 32bit distribution." -- .byte 0 -- --#include "../kernel/verify_cpu.S" --sse_ok: -- popw %ds -- --# tell BIOS we want to go to long mode -- movl $0xec00,%eax # declare target operating mode -- movl $2,%ebx # long mode -- int $0x15 -- --# Get memory size (extended mem, kB) -- -- xorl %eax, %eax -- movl %eax, (0x1e0) --#ifndef STANDARD_MEMORY_BIOS_CALL -- movb %al, (E820NR) --# Try three different memory detection schemes. First, try --# e820h, which lets us assemble a memory map, then try e801h, --# which returns a 32-bit memory size, and finally 88h, which --# returns 0-64m -- --# method E820H: --# the memory map from hell. e820h returns memory classified into --# a whole bunch of different types, and allows memory holes and --# everything. We scan through this memory map and build a list --# of the first 32 memory areas, which we return at [E820MAP]. --# This is documented at http://www.acpi.info/, in the ACPI 2.0 specification. -- --#define SMAP 0x534d4150 -- --meme820: -- xorl %ebx, %ebx # continuation counter -- movw $E820MAP, %di # point into the whitelist -- # so we can have the bios -- # directly write into it. -- --jmpe820: -- movl $0x0000e820, %eax # e820, upper word zeroed -- movl $SMAP, %edx # ascii 'SMAP' -- movl $20, %ecx # size of the e820rec -- pushw %ds # data record. -- popw %es -- int $0x15 # make the call -- jc bail820 # fall to e801 if it fails -- -- cmpl $SMAP, %eax # check the return is `SMAP' -- jne bail820 # fall to e801 if it fails -- --# cmpl $1, 16(%di) # is this usable memory? --# jne again820 -- -- # If this is usable memory, we save it by simply advancing %di by -- # sizeof(e820rec). -- # --good820: -- movb (E820NR), %al # up to 128 entries -- cmpb $E820MAX, %al -- jae bail820 -- -- incb (E820NR) -- movw %di, %ax -- addw $20, %ax -- movw %ax, %di --again820: -- cmpl $0, %ebx # check to see if -- jne jmpe820 # %ebx is set to EOF --bail820: -- -- --# method E801H: --# memory size is in 1k chunksizes, to avoid confusing loadlin. --# we store the 0xe801 memory size in a completely different place, --# because it will most likely be longer than 16 bits. --# (use 1e0 because that's what Larry Augustine uses in his --# alternative new memory detection scheme, and it's sensible --# to write everything into the same place.) -- --meme801: -- stc # fix to work around buggy -- xorw %cx,%cx # BIOSes which don't clear/set -- xorw %dx,%dx # carry on pass/error of -- # e801h memory size call -- # or merely pass cx,dx though -- # without changing them. -- movw $0xe801, %ax -- int $0x15 -- jc mem88 -- -- cmpw $0x0, %cx # Kludge to handle BIOSes -- jne e801usecxdx # which report their extended -- cmpw $0x0, %dx # memory in AX/BX rather than -- jne e801usecxdx # CX/DX. The spec I have read -- movw %ax, %cx # seems to indicate AX/BX -- movw %bx, %dx # are more reasonable anyway... -- --e801usecxdx: -- andl $0xffff, %edx # clear sign extend -- shll $6, %edx # and go from 64k to 1k chunks -- movl %edx, (0x1e0) # store extended memory size -- andl $0xffff, %ecx # clear sign extend -- addl %ecx, (0x1e0) # and add lower memory into -- # total size. -- --# Ye Olde Traditional Methode. Returns the memory size (up to 16mb or --# 64mb, depending on the bios) in ax. --mem88: -- --#endif -- movb $0x88, %ah -- int $0x15 -- movw %ax, (2) -- --# Set the keyboard repeat rate to the max -- movw $0x0305, %ax -- xorw %bx, %bx -- int $0x16 -- --# Check for video adapter and its parameters and allow the --# user to browse video modes. -- call video # NOTE: we need %ds pointing -- # to bootsector -- --# Get hd0 data... -- xorw %ax, %ax -- movw %ax, %ds -- ldsw (4 * 0x41), %si -- movw %cs, %ax # aka SETUPSEG -- subw $DELTA_INITSEG, %ax # aka INITSEG -- pushw %ax -- movw %ax, %es -- movw $0x0080, %di -- movw $0x10, %cx -- pushw %cx -- cld -- rep -- movsb --# Get hd1 data... -- xorw %ax, %ax -- movw %ax, %ds -- ldsw (4 * 0x46), %si -- popw %cx -- popw %es -- movw $0x0090, %di -- rep -- movsb --# Check that there IS a hd1 :-) -- movw $0x01500, %ax -- movb $0x81, %dl -- int $0x13 -- jc no_disk1 -- -- cmpb $3, %ah -- je is_disk1 -- --no_disk1: -- movw %cs, %ax # aka SETUPSEG -- subw $DELTA_INITSEG, %ax # aka INITSEG -- movw %ax, %es -- movw $0x0090, %di -- movw $0x10, %cx -- xorw %ax, %ax -- cld -- rep -- stosb --is_disk1: -- --# Check for PS/2 pointing device -- movw %cs, %ax # aka SETUPSEG -- subw $DELTA_INITSEG, %ax # aka INITSEG -- movw %ax, %ds -- movb $0, (0x1ff) # default is no pointing device -- int $0x11 # int 0x11: equipment list -- testb $0x04, %al # check if mouse installed -- jz no_psmouse -- -- movb $0xAA, (0x1ff) # device present --no_psmouse: -- --#include "../../i386/boot/edd.S" -- --# Now we want to move to protected mode ... -- cmpw $0, %cs:realmode_swtch -- jz rmodeswtch_normal -- -- lcall *%cs:realmode_swtch -- -- jmp rmodeswtch_end -- --rmodeswtch_normal: -- pushw %cs -- call default_switch -- --rmodeswtch_end: --# we get the code32 start address and modify the below 'jmpi' --# (loader may have changed it) -- movl %cs:code32_start, %eax -- movl %eax, %cs:code32 -- --# Now we move the system to its rightful place ... but we check if we have a --# big-kernel. In that case we *must* not move it ... -- testb $LOADED_HIGH, %cs:loadflags -- jz do_move0 # .. then we have a normal low -- # loaded zImage -- # .. or else we have a high -- # loaded bzImage -- jmp end_move # ... and we skip moving -- --do_move0: -- movw $0x100, %ax # start of destination segment -- movw %cs, %bp # aka SETUPSEG -- subw $DELTA_INITSEG, %bp # aka INITSEG -- movw %cs:start_sys_seg, %bx # start of source segment -- cld --do_move: -- movw %ax, %es # destination segment -- incb %ah # instead of add ax,#0x100 -- movw %bx, %ds # source segment -- addw $0x100, %bx -- subw %di, %di -- subw %si, %si -- movw $0x800, %cx -- rep -- movsw -- cmpw %bp, %bx # assume start_sys_seg > 0x200, -- # so we will perhaps read one -- # page more than needed, but -- # never overwrite INITSEG -- # because destination is a -- # minimum one page below source -- jb do_move -- --end_move: --# then we load the segment descriptors -- movw %cs, %ax # aka SETUPSEG -- movw %ax, %ds -- --# Check whether we need to be downward compatible with version <=201 -- cmpl $0, cmd_line_ptr -- jne end_move_self # loader uses version >=202 features -- cmpb $0x20, type_of_loader -- je end_move_self # bootsect loader, we know of it -- --# Boot loader doesnt support boot protocol version 2.02. --# If we have our code not at 0x90000, we need to move it there now. --# We also then need to move the params behind it (commandline) --# Because we would overwrite the code on the current IP, we move --# it in two steps, jumping high after the first one. -- movw %cs, %ax -- cmpw $SETUPSEG, %ax -- je end_move_self -- -- cli # make sure we really have -- # interrupts disabled ! -- # because after this the stack -- # should not be used -- subw $DELTA_INITSEG, %ax # aka INITSEG -- movw %ss, %dx -- cmpw %ax, %dx -- jb move_self_1 -- -- addw $INITSEG, %dx -- subw %ax, %dx # this will go into %ss after -- # the move --move_self_1: -- movw %ax, %ds -- movw $INITSEG, %ax # real INITSEG -- movw %ax, %es -- movw %cs:setup_move_size, %cx -- std # we have to move up, so we use -- # direction down because the -- # areas may overlap -- movw %cx, %di -- decw %di -- movw %di, %si -- subw $move_self_here+0x200, %cx -- rep -- movsb -- ljmp $SETUPSEG, $move_self_here -- --move_self_here: -- movw $move_self_here+0x200, %cx -- rep -- movsb -- movw $SETUPSEG, %ax -- movw %ax, %ds -- movw %dx, %ss --end_move_self: # now we are at the right place -- lidt idt_48 # load idt with 0,0 -- xorl %eax, %eax # Compute gdt_base -- movw %ds, %ax # (Convert %ds:gdt to a linear ptr) -- shll $4, %eax -- addl $gdt, %eax -- movl %eax, (gdt_48+2) -- lgdt gdt_48 # load gdt with whatever is -- # appropriate -- --# that was painless, now we enable a20 -- call empty_8042 -- -- movb $0xD1, %al # command write -- outb %al, $0x64 -- call empty_8042 -- -- movb $0xDF, %al # A20 on -- outb %al, $0x60 -- call empty_8042 -- --# --# You must preserve the other bits here. Otherwise embarrasing things --# like laptops powering off on boot happen. Corrected version by Kira --# Brown from Linux 2.2 --# -- inb $0x92, %al # -- orb $02, %al # "fast A20" version -- outb %al, $0x92 # some chips have only this -- --# wait until a20 really *is* enabled; it can take a fair amount of --# time on certain systems; Toshiba Tecras are known to have this --# problem. The memory location used here (0x200) is the int 0x80 --# vector, which should be safe to use. -- -- xorw %ax, %ax # segment 0x0000 -- movw %ax, %fs -- decw %ax # segment 0xffff (HMA) -- movw %ax, %gs --a20_wait: -- incw %ax # unused memory location <0xfff0 -- movw %ax, %fs:(0x200) # we use the "int 0x80" vector -- cmpw %gs:(0x210), %ax # and its corresponding HMA addr -- je a20_wait # loop until no longer aliased -- --# make sure any possible coprocessor is properly reset.. -- xorw %ax, %ax -- outb %al, $0xf0 -- call delay -- -- outb %al, $0xf1 -- call delay -- --# well, that went ok, I hope. Now we mask all interrupts - the rest --# is done in init_IRQ(). -- movb $0xFF, %al # mask all interrupts for now -- outb %al, $0xA1 -- call delay -- -- movb $0xFB, %al # mask all irq's but irq2 which -- outb %al, $0x21 # is cascaded -- --# Well, that certainly wasn't fun :-(. Hopefully it works, and we don't --# need no steenking BIOS anyway (except for the initial loading :-). --# The BIOS-routine wants lots of unnecessary data, and it's less --# "interesting" anyway. This is how REAL programmers do it. --# --# Well, now's the time to actually move into protected mode. To make --# things as simple as possible, we do no register set-up or anything, --# we let the gnu-compiled 32-bit programs do that. We just jump to --# absolute address 0x1000 (or the loader supplied one), --# in 32-bit protected mode. --# --# Note that the short jump isn't strictly needed, although there are --# reasons why it might be a good idea. It won't hurt in any case. -- movw $1, %ax # protected mode (PE) bit -- lmsw %ax # This is it! -- jmp flush_instr -- --flush_instr: -- xorw %bx, %bx # Flag to indicate a boot -- xorl %esi, %esi # Pointer to real-mode code -- movw %cs, %si -- subw $DELTA_INITSEG, %si -- shll $4, %esi # Convert to 32-bit pointer --# NOTE: For high loaded big kernels we need a --# jmpi 0x100000,__KERNEL_CS --# --# but we yet haven't reloaded the CS register, so the default size --# of the target offset still is 16 bit. --# However, using an operand prefix (0x66), the CPU will properly --# take our 48 bit far pointer. (INTeL 80386 Programmer's Reference --# Manual, Mixing 16-bit and 32-bit code, page 16-6) -- -- .byte 0x66, 0xea # prefix + jmpi-opcode --code32: .long 0x1000 # will be set to 0x100000 -- # for big kernels -- .word __KERNEL_CS -- --# Here's a bunch of information about your current kernel.. --kernel_version: .ascii UTS_RELEASE -- .ascii " (" -- .ascii LINUX_COMPILE_BY -- .ascii "@" -- .ascii LINUX_COMPILE_HOST -- .ascii ") " -- .ascii UTS_VERSION -- .byte 0 -- --# This is the default real mode switch routine. --# to be called just before protected mode transition --default_switch: -- cli # no interrupts allowed ! -- movb $0x80, %al # disable NMI for bootup -- # sequence -- outb %al, $0x70 -- lret -- -- --# This routine checks that the keyboard command queue is empty --# (after emptying the output buffers) --# --# Some machines have delusions that the keyboard buffer is always full --# with no keyboard attached... --# --# If there is no keyboard controller, we will usually get 0xff --# to all the reads. With each IO taking a microsecond and --# a timeout of 100,000 iterations, this can take about half a --# second ("delay" == outb to port 0x80). That should be ok, --# and should also be plenty of time for a real keyboard controller --# to empty. --# -- --empty_8042: -- pushl %ecx -- movl $100000, %ecx -- --empty_8042_loop: -- decl %ecx -- jz empty_8042_end_loop -- -- call delay -- -- inb $0x64, %al # 8042 status port -- testb $1, %al # output buffer? -- jz no_output -- -- call delay -- inb $0x60, %al # read it -- jmp empty_8042_loop -- --no_output: -- testb $2, %al # is input buffer full? -- jnz empty_8042_loop # yes - loop --empty_8042_end_loop: -- popl %ecx -- ret -- --# Read the cmos clock. Return the seconds in al --gettime: -- pushw %cx -- movb $0x02, %ah -- int $0x1a -- movb %dh, %al # %dh contains the seconds -- andb $0x0f, %al -- movb %dh, %ah -- movb $0x04, %cl -- shrb %cl, %ah -- aad -- popw %cx -- ret -- --# Delay is needed after doing I/O --delay: -- outb %al,$0x80 -- ret -- --# Descriptor tables --gdt: -- .word 0, 0, 0, 0 # dummy -- -- .word 0, 0, 0, 0 # unused -- -- .word 0xFFFF # 4Gb - (0x100000*0x1000 = 4Gb) -- .word 0 # base address = 0 -- .word 0x9A00 # code read/exec -- .word 0x00CF # granularity = 4096, 386 -- # (+5th nibble of limit) -- -- .word 0xFFFF # 4Gb - (0x100000*0x1000 = 4Gb) -- .word 0 # base address = 0 -- .word 0x9200 # data read/write -- .word 0x00CF # granularity = 4096, 386 -- # (+5th nibble of limit) --gdt_end: --idt_48: -- .word 0 # idt limit = 0 -- .word 0, 0 # idt base = 0L --gdt_48: -- .word gdt_end-gdt-1 # gdt limit -- .word 0, 0 # gdt base (filled in later) -- --# Include video setup & detection code -- --#include "../../i386/boot/video.S" -- --# Setup signature -- must be last --setup_sig1: .word SIG1 --setup_sig2: .word SIG2 -- --# After this point, there is some free space which is used by the video mode --# handling code to store the temporary mode table (not used by the kernel). -- --modelist: -- --.text --endtext: --.data --enddata: --.bss --endbss: ---- a/arch/x86_64/boot/tools/build.c -+++ /dev/null -@@ -1,185 +0,0 @@ --/* -- * Copyright (C) 1991, 1992 Linus Torvalds -- * Copyright (C) 1997 Martin Mares -- */ -- --/* -- * This file builds a disk-image from three different files: -- * -- * - bootsect: compatibility mbr which prints an error message if -- * someone tries to boot the kernel directly. -- * - setup: 8086 machine code, sets up system parm -- * - system: 80386 code for actual system -- * -- * It does some checking that all files are of the correct type, and -- * just writes the result to stdout, removing headers and padding to -- * the right amount. It also writes some system data to stderr. -- */ -- --/* -- * Changes by tytso to allow root device specification -- * High loaded stuff by Hans Lermen & Werner Almesberger, Feb. 1996 -- * Cross compiling fixes by Gertjan van Wingerde, July 1996 -- * Rewritten by Martin Mares, April 1997 -- */ -- --#include <stdio.h> --#include <string.h> --#include <stdlib.h> --#include <stdarg.h> --#include <sys/types.h> --#include <sys/stat.h> --#include <sys/sysmacros.h> --#include <unistd.h> --#include <fcntl.h> --#include <asm/boot.h> -- --typedef unsigned char byte; --typedef unsigned short word; --typedef unsigned long u32; -- --#define DEFAULT_MAJOR_ROOT 0 --#define DEFAULT_MINOR_ROOT 0 -- --/* Minimal number of setup sectors (see also bootsect.S) */ --#define SETUP_SECTS 4 -- --byte buf[1024]; --int fd; --int is_big_kernel; -- --void die(const char * str, ...) --{ -- va_list args; -- va_start(args, str); -- vfprintf(stderr, str, args); -- fputc('\n', stderr); -- exit(1); --} -- --void file_open(const char *name) --{ -- if ((fd = open(name, O_RDONLY, 0)) < 0) -- die("Unable to open `%s': %m", name); --} -- --void usage(void) --{ -- die("Usage: build [-b] bootsect setup system [rootdev] [> image]"); --} -- --int main(int argc, char ** argv) --{ -- unsigned int i, c, sz, setup_sectors; -- u32 sys_size; -- byte major_root, minor_root; -- struct stat sb; -- -- if (argc > 2 && !strcmp(argv[1], "-b")) -- { -- is_big_kernel = 1; -- argc--, argv++; -- } -- if ((argc < 4) || (argc > 5)) -- usage(); -- if (argc > 4) { -- if (!strcmp(argv[4], "CURRENT")) { -- if (stat("/", &sb)) { -- perror("/"); -- die("Couldn't stat /"); -- } -- major_root = major(sb.st_dev); -- minor_root = minor(sb.st_dev); -- } else if (strcmp(argv[4], "FLOPPY")) { -- if (stat(argv[4], &sb)) { -- perror(argv[4]); -- die("Couldn't stat root device."); -- } -- major_root = major(sb.st_rdev); -- minor_root = minor(sb.st_rdev); -- } else { -- major_root = 0; -- minor_root = 0; -- } -- } else { -- major_root = DEFAULT_MAJOR_ROOT; -- minor_root = DEFAULT_MINOR_ROOT; -- } -- fprintf(stderr, "Root device is (%d, %d)\n", major_root, minor_root); -- -- file_open(argv[1]); -- i = read(fd, buf, sizeof(buf)); -- fprintf(stderr,"Boot sector %d bytes.\n",i); -- if (i != 512) -- die("Boot block must be exactly 512 bytes"); -- if (buf[510] != 0x55 || buf[511] != 0xaa) -- die("Boot block hasn't got boot flag (0xAA55)"); -- buf[508] = minor_root; -- buf[509] = major_root; -- if (write(1, buf, 512) != 512) -- die("Write call failed"); -- close (fd); -- -- file_open(argv[2]); /* Copy the setup code */ -- for (i=0 ; (c=read(fd, buf, sizeof(buf)))>0 ; i+=c ) -- if (write(1, buf, c) != c) -- die("Write call failed"); -- if (c != 0) -- die("read-error on `setup'"); -- close (fd); -- -- setup_sectors = (i + 511) / 512; /* Pad unused space with zeros */ -- /* for compatibility with ancient versions of LILO. */ -- if (setup_sectors < SETUP_SECTS) -- setup_sectors = SETUP_SECTS; -- fprintf(stderr, "Setup is %d bytes.\n", i); -- memset(buf, 0, sizeof(buf)); -- while (i < setup_sectors * 512) { -- c = setup_sectors * 512 - i; -- if (c > sizeof(buf)) -- c = sizeof(buf); -- if (write(1, buf, c) != c) -- die("Write call failed"); -- i += c; -- } -- -- file_open(argv[3]); -- if (fstat (fd, &sb)) -- die("Unable to stat `%s': %m", argv[3]); -- sz = sb.st_size; -- fprintf (stderr, "System is %d kB\n", sz/1024); -- sys_size = (sz + 15) / 16; -- if (!is_big_kernel && sys_size > DEF_SYSSIZE) -- die("System is too big. Try using bzImage or modules."); -- while (sz > 0) { -- int l, n; -- -- l = (sz > sizeof(buf)) ? sizeof(buf) : sz; -- if ((n=read(fd, buf, l)) != l) { -- if (n < 0) -- die("Error reading %s: %m", argv[3]); -- else -- die("%s: Unexpected EOF", argv[3]); -- } -- if (write(1, buf, l) != l) -- die("Write failed"); -- sz -= l; -- } -- close(fd); -- -- if (lseek(1, 497, SEEK_SET) != 497) /* Write sizes to the bootsector */ -- die("Output: seek failed"); -- buf[0] = setup_sectors; -- if (write(1, buf, 1) != 1) -- die("Write of setup sector count failed"); -- if (lseek(1, 500, SEEK_SET) != 500) -- die("Output: seek failed"); -- buf[0] = (sys_size & 0xff); -- buf[1] = ((sys_size >> 8) & 0xff); -- buf[2] = ((sys_size >> 16) & 0xff); -- buf[3] = ((sys_size >> 24) & 0xff); -- if (write(1, buf, 4) != 4) -- die("Write of image length failed"); -- -- return 0; /* Everything is OK */ --} ---- a/arch/x86_64/kernel/Makefile -+++ b/arch/x86_64/kernel/Makefile -@@ -43,6 +43,7 @@ - - obj-y += topology.o - obj-y += intel_cacheinfo.o -+obj-y += addon_cpuid_features.o - obj-y += pcspeaker.o - - CFLAGS_vsyscall.o := $(PROFILING) -g0 -@@ -53,6 +54,7 @@ - topology-y += ../../i386/kernel/topology.o - microcode-$(subst m,y,$(CONFIG_MICROCODE)) += ../../i386/kernel/microcode.o - intel_cacheinfo-y += ../../i386/kernel/cpu/intel_cacheinfo.o -+addon_cpuid_features-y += ../../i386/kernel/cpu/addon_cpuid_features.o - quirks-y += ../../i386/kernel/quirks.o - i8237-y += ../../i386/kernel/i8237.o - msr-$(subst m,y,$(CONFIG_X86_MSR)) += ../../i386/kernel/msr.o ---- a/arch/x86_64/kernel/setup.c -+++ b/arch/x86_64/kernel/setup.c -@@ -846,6 +846,8 @@ - c->x86_capability[2] = cpuid_edx(0x80860001); - } - -+ init_scattered_cpuid_features(c); -+ - c->apicid = phys_pkg_id(0); - - /* -@@ -931,7 +933,7 @@ - "fpu", "vme", "de", "pse", "tsc", "msr", "pae", "mce", - "cx8", "apic", NULL, "sep", "mtrr", "pge", "mca", "cmov", - "pat", "pse36", "pn", "clflush", NULL, "dts", "acpi", "mmx", -- "fxsr", "sse", "sse2", "ss", "ht", "tm", "ia64", NULL, -+ "fxsr", "sse", "sse2", "ss", "ht", "tm", "ia64", "pbe", - - /* AMD-defined */ - NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, -@@ -947,10 +949,11 @@ - NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, - - /* Other (Linux-defined) */ -- "cxmmx", NULL, "cyrix_arr", "centaur_mcr", NULL, -- "constant_tsc", NULL, NULL, -- "up", NULL, NULL, NULL, NULL, NULL, NULL, NULL, -- NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, -+ "cxmmx", "k6_mtrr", "cyrix_arr", "centaur_mcr", -+ NULL, NULL, NULL, NULL, -+ "constant_tsc", "up", NULL, "arch_perfmon", -+ "pebs", "bts", NULL, "sync_rdtsc", -+ "rep_good", NULL, NULL, NULL, NULL, NULL, NULL, NULL, - NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, - - /* Intel-defined (#2) */ -@@ -961,7 +964,7 @@ - - /* VIA/Cyrix/Centaur-defined */ - NULL, NULL, "rng", "rng_en", NULL, NULL, "ace", "ace_en", -- NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, -+ "ace2", "ace2_en", "phe", "phe_en", "pmm", "pmm_en", NULL, NULL, - NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, - NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, - -@@ -972,6 +975,12 @@ - "osvw", "ibs", NULL, NULL, NULL, NULL, - NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, - NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, -+ -+ /* Auxiliary (Linux-defined) */ -+ "ida", NULL, NULL, NULL, NULL, NULL, NULL, NULL, -+ NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, -+ NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, -+ NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, - }; - static char *x86_power_flags[] = { - "ts", /* temperature sensor */ ---- a/arch/x86_64/kernel/verify_cpu.S -+++ b/arch/x86_64/kernel/verify_cpu.S -@@ -37,20 +37,6 @@ - pushl $0 # Kill any dangerous flags - popfl - -- /* minimum CPUID flags for x86-64 as defined by AMD */ --#define M(x) (1<<(x)) --#define M2(a,b) M(a)|M(b) --#define M4(a,b,c,d) M(a)|M(b)|M(c)|M(d) -- --#define SSE_MASK \ -- (M2(X86_FEATURE_XMM,X86_FEATURE_XMM2)) --#define REQUIRED_MASK1 \ -- (M4(X86_FEATURE_FPU,X86_FEATURE_PSE,X86_FEATURE_TSC,X86_FEATURE_MSR)|\ -- M4(X86_FEATURE_PAE,X86_FEATURE_CX8,X86_FEATURE_PGE,X86_FEATURE_CMOV)|\ -- M(X86_FEATURE_FXSR)) --#define REQUIRED_MASK2 \ -- (M(X86_FEATURE_LM - 32)) -- - pushfl # standard way to check for cpuid - popl %eax - movl %eax,%ebx -@@ -79,8 +65,8 @@ - verify_cpu_noamd: - movl $0x1,%eax # Does the cpu have what it takes - cpuid -- andl $REQUIRED_MASK1,%edx -- xorl $REQUIRED_MASK1,%edx -+ andl $REQUIRED_MASK0,%edx -+ xorl $REQUIRED_MASK0,%edx - jnz verify_cpu_no_longmode - - movl $0x80000000,%eax # See if extended cpuid is implemented -@@ -90,8 +76,8 @@ - - movl $0x80000001,%eax # Does the cpu have what it takes - cpuid -- andl $REQUIRED_MASK2,%edx -- xorl $REQUIRED_MASK2,%edx -+ andl $REQUIRED_MASK1,%edx -+ xorl $REQUIRED_MASK1,%edx - jnz verify_cpu_no_longmode - - verify_cpu_sse_test: ---- a/drivers/ide/legacy/hd.c -+++ b/drivers/ide/legacy/hd.c -@@ -718,74 +718,25 @@ - device_timer.function = hd_times_out; - blk_queue_hardsect_size(hd_queue, 512); - --#ifdef __i386__ - if (!NR_HD) { -- extern struct drive_info drive_info; -- unsigned char *BIOS = (unsigned char *) &drive_info; -- unsigned long flags; -- int cmos_disks; -- -- for (drive=0 ; drive<2 ; drive++) { -- hd_info[drive].cyl = *(unsigned short *) BIOS; -- hd_info[drive].head = *(2+BIOS); -- hd_info[drive].wpcom = *(unsigned short *) (5+BIOS); -- hd_info[drive].ctl = *(8+BIOS); -- hd_info[drive].lzone = *(unsigned short *) (12+BIOS); -- hd_info[drive].sect = *(14+BIOS); --#ifdef does_not_work_for_everybody_with_scsi_but_helps_ibm_vp -- if (hd_info[drive].cyl && NR_HD == drive) -- NR_HD++; --#endif -- BIOS += 16; -- } -- -- /* -- We query CMOS about hard disks : it could be that -- we have a SCSI/ESDI/etc controller that is BIOS -- compatible with ST-506, and thus showing up in our -- BIOS table, but not register compatible, and therefore -- not present in CMOS. -- -- Furthermore, we will assume that our ST-506 drives -- <if any> are the primary drives in the system, and -- the ones reflected as drive 1 or 2. -- -- The first drive is stored in the high nibble of CMOS -- byte 0x12, the second in the low nibble. This will be -- either a 4 bit drive type or 0xf indicating use byte 0x19 -- for an 8 bit type, drive 1, 0x1a for drive 2 in CMOS. -- -- Needless to say, a non-zero value means we have -- an AT controller hard disk for that drive. -- -- Currently the rtc_lock is a bit academic since this -- driver is non-modular, but someday... ? Paul G. -- */ -- -- spin_lock_irqsave(&rtc_lock, flags); -- cmos_disks = CMOS_READ(0x12); -- spin_unlock_irqrestore(&rtc_lock, flags); -- -- if (cmos_disks & 0xf0) { -- if (cmos_disks & 0x0f) -- NR_HD = 2; -- else -- NR_HD = 1; -- } -- } --#endif /* __i386__ */ --#ifdef __arm__ -- if (!NR_HD) { -- /* We don't know anything about the drive. This means -+ /* -+ * We don't know anything about the drive. This means - * that you *MUST* specify the drive parameters to the - * kernel yourself. -+ * -+ * If we were on an i386, we used to read this info from -+ * the BIOS or CMOS. This doesn't work all that well, -+ * since this assumes that this is a primary or secondary -+ * drive, and if we're using this legacy driver, it's -+ * probably an auxilliary controller added to recover -+ * legacy data off an ST-506 drive. Either way, it's -+ * definitely safest to have the user explicitly specify -+ * the information. - */ - printk("hd: no drives specified - use hd=cyl,head,sectors" - " on kernel command line\n"); -- } --#endif -- if (!NR_HD) - goto out; -+ } - - for (drive=0 ; drive < NR_HD ; drive++) { - struct gendisk *disk = alloc_disk(64); ---- a/include/asm-i386/boot.h -+++ b/include/asm-i386/boot.h -@@ -1,5 +1,5 @@ --#ifndef _LINUX_BOOT_H --#define _LINUX_BOOT_H -+#ifndef _ASM_BOOT_H -+#define _ASM_BOOT_H - - /* Don't touch these, unless you really know what you're doing. */ - #define DEF_INITSEG 0x9000 -@@ -17,4 +17,4 @@ - + (CONFIG_PHYSICAL_ALIGN - 1)) \ - & ~(CONFIG_PHYSICAL_ALIGN - 1)) - --#endif /* _LINUX_BOOT_H */ -+#endif /* _ASM_BOOT_H */ ---- /dev/null -+++ b/include/asm-i386/bootparam.h -@@ -0,0 +1,85 @@ -+#ifndef _ASM_BOOTPARAM_H -+#define _ASM_BOOTPARAM_H -+ -+#include <linux/types.h> -+#include <linux/screen_info.h> -+#include <linux/apm_bios.h> -+#include <asm/e820.h> -+#include <linux/edd.h> -+#include <video/edid.h> -+ -+struct setup_header { -+ u8 setup_sects; -+ u16 root_flags; -+ u32 syssize; -+ u16 ram_size; -+ u16 vid_mode; -+ u16 root_dev; -+ u16 boot_flag; -+ u16 jump; -+ u32 header; -+ u16 version; -+ u32 realmode_swtch; -+ u16 start_sys; -+ u16 kernel_version; -+ u8 type_of_loader; -+ u8 loadflags; -+#define LOADED_HIGH 0x01 -+#define CAN_USE_HEAP 0x80 -+ u16 setup_move_size; -+ u32 code32_start; -+ u32 ramdisk_image; -+ u32 ramdisk_size; -+ u32 bootsect_kludge; -+ u16 heap_end_ptr; -+ u16 _pad1; -+ u32 cmd_line_ptr; -+ u32 initrd_addr_max; -+ u32 kernel_alignment; -+ u8 relocatable_kernel; -+} __attribute__((packed)); -+ -+struct sys_desc_table { -+ u16 length; -+ u8 table[14]; -+}; -+ -+struct efi_info { -+ u32 _pad1; -+ u32 efi_systab; -+ u32 efi_memdesc_size; -+ u32 efi_memdec_version; -+ u32 efi_memmap; -+ u32 fi_memmap_size; -+ u32 _pad2[2]; -+}; -+ -+/* The so-called "zeropage" */ -+struct boot_params { -+ struct screen_info screen_info; /* 0x000 */ -+ struct apm_bios_info apm_bios_info; /* 0x040 */ -+ u8 _pad2[12]; /* 0x054 */ -+ u32 speedstep_info[4]; /* 0x060 */ -+ u8 _pad3[16]; /* 0x070 */ -+ u8 hd0_info[16]; /* obsolete! */ /* 0x080 */ -+ u8 hd1_info[16]; /* obsolete! */ /* 0x090 */ -+ struct sys_desc_table sys_desc_table; /* 0x0a0 */ -+ u8 _pad4[144]; /* 0x0b0 */ -+ struct edid_info edid_info; /* 0x140 */ -+ struct efi_info efi_info; /* 0x1c0 */ -+ u32 alt_mem_k; /* 0x1e0 */ -+ u32 scratch; /* Scratch field! */ /* 0x1e4 */ -+ u8 e820_entries; /* 0x1e8 */ -+ u8 eddbuf_entries; /* 0x1e9 */ -+ u8 edd_mbr_sig_buf_entries; /* 0x1ea */ -+ u8 _pad6[6]; /* 0x1eb */ -+ struct setup_header hdr; /* setup header */ /* 0x1f1 */ -+ u8 _pad7[0x290-0x1f1-sizeof(struct setup_header)]; -+ u32 edd_mbr_sig_buffer[EDD_MBR_SIG_MAX]; /* 0x290 */ -+ struct e820entry e820_map[E820MAX]; /* 0x2d0 */ -+ u8 _pad8[48]; /* 0xcd0 */ -+ struct edd_info eddbuf[EDDMAXNR]; /* 0xd00 */ -+ u8 _pad9[276]; /* 0xeec */ -+} __attribute__((packed)); -+ -+#endif /* _ASM_BOOTPARAM_H */ ---- a/include/asm-i386/cpufeature.h -+++ b/include/asm-i386/cpufeature.h -@@ -12,7 +12,7 @@ - #endif - #include <asm/required-features.h> - --#define NCAPINTS 7 /* N 32-bit words worth of info */ -+#define NCAPINTS 8 /* N 32-bit words worth of info */ - - /* Intel-defined CPU features, CPUID level 0x00000001 (edx), word 0 */ - #define X86_FEATURE_FPU (0*32+ 0) /* Onboard FPU */ -@@ -81,6 +81,7 @@ - #define X86_FEATURE_BTS (3*32+13) /* Branch Trace Store */ - /* 14 free */ - #define X86_FEATURE_SYNC_RDTSC (3*32+15) /* RDTSC synchronizes the CPU */ -+#define X86_FEATURE_REP_GOOD (3*32+16) /* rep microcode works well on this CPU */ - - /* Intel-defined CPU features, CPUID level 0x00000001 (ecx), word 4 */ - #define X86_FEATURE_XMM3 (4*32+ 0) /* Streaming SIMD Extensions-3 */ -@@ -108,11 +109,24 @@ - #define X86_FEATURE_LAHF_LM (6*32+ 0) /* LAHF/SAHF in long mode */ - #define X86_FEATURE_CMP_LEGACY (6*32+ 1) /* If yes HyperThreading not valid */ - --#define cpu_has(c, bit) \ -- ((__builtin_constant_p(bit) && (bit) < 32 && \ -- (1UL << (bit)) & REQUIRED_MASK1) ? \ -- 1 : \ -- test_bit(bit, (c)->x86_capability)) -+/* -+ * Auxiliary flags: Linux defined - For features scattered in various -+ * CPUID levels like 0x6, 0xA etc -+ */ -+#define X86_FEATURE_IDA (7*32+ 0) /* Intel Dynamic Acceleration */ -+ -+#define cpu_has(c, bit) \ -+ (__builtin_constant_p(bit) && \ -+ ( (((bit)>>5)==0 && (1UL<<((bit)&31) & REQUIRED_MASK0)) || \ -+ (((bit)>>5)==1 && (1UL<<((bit)&31) & REQUIRED_MASK1)) || \ -+ (((bit)>>5)==2 && (1UL<<((bit)&31) & REQUIRED_MASK2)) || \ -+ (((bit)>>5)==3 && (1UL<<((bit)&31) & REQUIRED_MASK3)) || \ -+ (((bit)>>5)==4 && (1UL<<((bit)&31) & REQUIRED_MASK4)) || \ -+ (((bit)>>5)==5 && (1UL<<((bit)&31) & REQUIRED_MASK5)) || \ -+ (((bit)>>5)==6 && (1UL<<((bit)&31) & REQUIRED_MASK6)) || \ -+ (((bit)>>5)==7 && (1UL<<((bit)&31) & REQUIRED_MASK7)) ) \ -+ ? 1 : \ -+ test_bit(bit, (c)->x86_capability)) - #define boot_cpu_has(bit) cpu_has(&boot_cpu_data, bit) - - #define cpu_has_fpu boot_cpu_has(X86_FEATURE_FPU) ---- a/include/asm-i386/e820.h -+++ b/include/asm-i386/e820.h -@@ -25,13 +25,15 @@ - - #ifndef __ASSEMBLY__ - -+struct e820entry { -+ u64 addr; /* start of memory segment */ -+ u64 size; /* size of memory segment */ -+ u32 type; /* type of memory segment */ -+} __attribute__((packed)); -+ - struct e820map { -- int nr_map; -- struct e820entry { -- unsigned long long addr; /* start of memory segment */ -- unsigned long long size; /* size of memory segment */ -- unsigned long type; /* type of memory segment */ -- } map[E820MAX]; -+ u32 nr_map; -+ struct e820entry map[E820MAX]; - }; - - extern struct e820map e820; ---- a/include/asm-i386/processor.h -+++ b/include/asm-i386/processor.h -@@ -119,6 +119,7 @@ - extern void identify_boot_cpu(void); - extern void identify_secondary_cpu(struct cpuinfo_x86 *); - extern void print_cpu_info(struct cpuinfo_x86 *); -+extern void init_scattered_cpuid_features(struct cpuinfo_x86 *c); - extern unsigned int init_intel_cacheinfo(struct cpuinfo_x86 *c); - extern unsigned short num_cache_leaves; - ---- a/include/asm-i386/required-features.h -+++ b/include/asm-i386/required-features.h -@@ -3,7 +3,7 @@ - - /* Define minimum CPUID feature set for kernel These bits are checked - really early to actually display a visible error message before the -- kernel dies. Only add word 0 bits here -+ kernel dies. Make sure to assign features to the proper mask! - - Some requirements that are not in CPUID yet are also in the - CONFIG_X86_MINIMUM_CPU mode which is checked too. -@@ -11,24 +11,45 @@ - The real information is in arch/i386/Kconfig.cpu, this just converts - the CONFIGs into a bitmask */ - -+#ifndef CONFIG_MATH_EMULATION -+# define NEED_FPU (1<<(X86_FEATURE_FPU & 31)) -+#else -+# define NEED_FPU 0 -+#endif -+ - #ifdef CONFIG_X86_PAE --#define NEED_PAE (1<<X86_FEATURE_PAE) -+# define NEED_PAE (1<<(X86_FEATURE_PAE & 31)) - #else --#define NEED_PAE 0 -+# define NEED_PAE 0 - #endif - - #ifdef CONFIG_X86_CMOV --#define NEED_CMOV (1<<X86_FEATURE_CMOV) -+# define NEED_CMOV (1<<(X86_FEATURE_CMOV & 31)) - #else --#define NEED_CMOV 0 -+# define NEED_CMOV 0 - #endif - - #ifdef CONFIG_X86_CMPXCHG64 --#define NEED_CMPXCHG64 (1<<X86_FEATURE_CX8) -+# define NEED_CX8 (1<<(X86_FEATURE_CX8 & 31)) -+#else -+# define NEED_CX8 0 -+#endif -+ -+#define REQUIRED_MASK0 (NEED_FPU|NEED_PAE|NEED_CMOV|NEED_CX8) -+ -+#ifdef CONFIG_X86_USE_3DNOW -+# define NEED_3DNOW (1<<(X86_FEATURE_3DNOW & 31)) - #else --#define NEED_CMPXCHG64 0 -+# define NEED_3DNOW 0 - #endif - --#define REQUIRED_MASK1 (NEED_PAE|NEED_CMOV|NEED_CMPXCHG64) -+#define REQUIRED_MASK1 (NEED_3DNOW) -+ -+#define REQUIRED_MASK2 0 -+#define REQUIRED_MASK3 0 -+#define REQUIRED_MASK4 0 -+#define REQUIRED_MASK5 0 -+#define REQUIRED_MASK6 0 -+#define REQUIRED_MASK7 0 - - #endif ---- a/include/asm-i386/setup.h -+++ b/include/asm-i386/setup.h -@@ -26,12 +26,15 @@ - #define NEW_CL_POINTER 0x228 /* Relative to real mode data */ - - #ifndef __ASSEMBLY__ -+ -+#include <asm/bootparam.h> -+ - /* - * This is set up by the setup-routine at boot-time - */ --extern unsigned char boot_params[PARAM_SIZE]; -+extern struct boot_params boot_params; - --#define PARAM (boot_params) -+#define PARAM ((char *)&boot_params) - #define SCREEN_INFO (*(struct screen_info *) (PARAM+0)) - #define EXT_MEM_K (*(unsigned short *) (PARAM+2)) - #define ALT_MEM_K (*(unsigned long *) (PARAM+0x1e0)) -@@ -39,8 +42,7 @@ - #define E820_MAP ((struct e820entry *) (PARAM+E820MAP)) - #define APM_BIOS_INFO (*(struct apm_bios_info *) (PARAM+0x40)) - #define IST_INFO (*(struct ist_info *) (PARAM+0x60)) --#define DRIVE_INFO (*(struct drive_info_struct *) (PARAM+0x80)) --#define SYS_DESC_TABLE (*(struct sys_desc_table_struct*)(PARAM+0xa0)) -+#define SYS_DESC_TABLE (*(struct sys_desc_table *)(PARAM+0xa0)) - #define EFI_SYSTAB ((efi_system_table_t *) *((unsigned long *)(PARAM+0x1c4))) - #define EFI_MEMDESC_SIZE (*((unsigned long *) (PARAM+0x1c8))) - #define EFI_MEMDESC_VERSION (*((unsigned long *) (PARAM+0x1cc))) ---- a/include/asm-x86_64/alternative.h -+++ b/include/asm-x86_64/alternative.h -@@ -5,6 +5,41 @@ - - #include <linux/types.h> - #include <linux/stddef.h> -+ -+/* -+ * Alternative inline assembly for SMP. -+ * -+ * The LOCK_PREFIX macro defined here replaces the LOCK and -+ * LOCK_PREFIX macros used everywhere in the source tree. -+ * -+ * SMP alternatives use the same data structures as the other -+ * alternatives and the X86_FEATURE_UP flag to indicate the case of a -+ * UP system running a SMP kernel. The existing apply_alternatives() -+ * works fine for patching a SMP kernel for UP. -+ * -+ * The SMP alternative tables can be kept after boot and contain both -+ * UP and SMP versions of the instructions to allow switching back to -+ * SMP at runtime, when hotplugging in a new CPU, which is especially -+ * useful in virtualized environments. -+ * -+ * The very common lock prefix is handled as special case in a -+ * separate table which is a pure address list without replacement ptr -+ * and size information. That keeps the table sizes small. -+ */ -+ -+#ifdef CONFIG_SMP -+#define LOCK_PREFIX \ -+ ".section .smp_locks,\"a\"\n" \ -+ " .align 8\n" \ -+ " .quad 661f\n" /* address */ \ -+ ".previous\n" \ -+ "661:\n\tlock; " -+ -+#else /* ! CONFIG_SMP */ -+#define LOCK_PREFIX "" -+#endif -+ -+/* This must be included *after* the definition of LOCK_PREFIX */ - #include <asm/cpufeature.h> - - struct alt_instr { -@@ -108,39 +143,6 @@ - */ - #define ASM_OUTPUT2(a, b) a, b - --/* -- * Alternative inline assembly for SMP. -- * -- * The LOCK_PREFIX macro defined here replaces the LOCK and -- * LOCK_PREFIX macros used everywhere in the source tree. -- * -- * SMP alternatives use the same data structures as the other -- * alternatives and the X86_FEATURE_UP flag to indicate the case of a -- * UP system running a SMP kernel. The existing apply_alternatives() -- * works fine for patching a SMP kernel for UP. -- * -- * The SMP alternative tables can be kept after boot and contain both -- * UP and SMP versions of the instructions to allow switching back to -- * SMP at runtime, when hotplugging in a new CPU, which is especially -- * useful in virtualized environments. -- * -- * The very common lock prefix is handled as special case in a -- * separate table which is a pure address list without replacement ptr -- * and size information. That keeps the table sizes small. -- */ -- --#ifdef CONFIG_SMP --#define LOCK_PREFIX \ -- ".section .smp_locks,\"a\"\n" \ -- " .align 8\n" \ -- " .quad 661f\n" /* address */ \ -- ".previous\n" \ -- "661:\n\tlock; " -- --#else /* ! CONFIG_SMP */ --#define LOCK_PREFIX "" --#endif -- - struct paravirt_patch; - #ifdef CONFIG_PARAVIRT - void apply_paravirt(struct paravirt_patch *start, struct paravirt_patch *end); ---- a/include/asm-x86_64/boot.h -+++ b/include/asm-x86_64/boot.h -@@ -1,15 +1 @@ --#ifndef _LINUX_BOOT_H --#define _LINUX_BOOT_H -- --/* Don't touch these, unless you really know what you're doing. */ --#define DEF_INITSEG 0x9000 --#define DEF_SYSSEG 0x1000 --#define DEF_SETUPSEG 0x9020 --#define DEF_SYSSIZE 0x7F00 -- --/* Internal svga startup constants */ --#define NORMAL_VGA 0xffff /* 80x25 mode */ --#define EXTENDED_VGA 0xfffe /* 80x50 mode */ --#define ASK_VGA 0xfffd /* ask for it at bootup */ -- --#endif -+#include <asm-i386/boot.h> ---- /dev/null -+++ b/include/asm-x86_64/bootparam.h -@@ -0,0 +1 @@ -+#include <asm-i386/bootparam.h> ---- a/include/asm-x86_64/cpufeature.h -+++ b/include/asm-x86_64/cpufeature.h -@@ -7,115 +7,24 @@ - #ifndef __ASM_X8664_CPUFEATURE_H - #define __ASM_X8664_CPUFEATURE_H - --#define NCAPINTS 7 /* N 32-bit words worth of info */ -+#include <asm-i386/cpufeature.h> - --/* Intel-defined CPU features, CPUID level 0x00000001, word 0 */ --#define X86_FEATURE_FPU (0*32+ 0) /* Onboard FPU */ --#define X86_FEATURE_VME (0*32+ 1) /* Virtual Mode Extensions */ --#define X86_FEATURE_DE (0*32+ 2) /* Debugging Extensions */ --#define X86_FEATURE_PSE (0*32+ 3) /* Page Size Extensions */ --#define X86_FEATURE_TSC (0*32+ 4) /* Time Stamp Counter */ --#define X86_FEATURE_MSR (0*32+ 5) /* Model-Specific Registers, RDMSR, WRMSR */ --#define X86_FEATURE_PAE (0*32+ 6) /* Physical Address Extensions */ --#define X86_FEATURE_MCE (0*32+ 7) /* Machine Check Architecture */ --#define X86_FEATURE_CX8 (0*32+ 8) /* CMPXCHG8 instruction */ --#define X86_FEATURE_APIC (0*32+ 9) /* Onboard APIC */ --#define X86_FEATURE_SEP (0*32+11) /* SYSENTER/SYSEXIT */ --#define X86_FEATURE_MTRR (0*32+12) /* Memory Type Range Registers */ --#define X86_FEATURE_PGE (0*32+13) /* Page Global Enable */ --#define X86_FEATURE_MCA (0*32+14) /* Machine Check Architecture */ --#define X86_FEATURE_CMOV (0*32+15) /* CMOV instruction (FCMOVCC and FCOMI too if FPU present) */ --#define X86_FEATURE_PAT (0*32+16) /* Page Attribute Table */ --#define X86_FEATURE_PSE36 (0*32+17) /* 36-bit PSEs */ --#define X86_FEATURE_PN (0*32+18) /* Processor serial number */ --#define X86_FEATURE_CLFLSH (0*32+19) /* Supports the CLFLUSH instruction */ --#define X86_FEATURE_DS (0*32+21) /* Debug Store */ --#define X86_FEATURE_ACPI (0*32+22) /* ACPI via MSR */ --#define X86_FEATURE_MMX (0*32+23) /* Multimedia Extensions */ --#define X86_FEATURE_FXSR (0*32+24) /* FXSAVE and FXRSTOR instructions (fast save and restore */ -- /* of FPU context), and CR4.OSFXSR available */ --#define X86_FEATURE_XMM (0*32+25) /* Streaming SIMD Extensions */ --#define X86_FEATURE_XMM2 (0*32+26) /* Streaming SIMD Extensions-2 */ --#define X86_FEATURE_SELFSNOOP (0*32+27) /* CPU self snoop */ --#define X86_FEATURE_HT (0*32+28) /* Hyper-Threading */ --#define X86_FEATURE_ACC (0*32+29) /* Automatic clock control */ --#define X86_FEATURE_IA64 (0*32+30) /* IA-64 processor */ -- --/* AMD-defined CPU features, CPUID level 0x80000001, word 1 */ --/* Don't duplicate feature flags which are redundant with Intel! */ --#define X86_FEATURE_SYSCALL (1*32+11) /* SYSCALL/SYSRET */ --#define X86_FEATURE_MMXEXT (1*32+22) /* AMD MMX extensions */ --#define X86_FEATURE_FXSR_OPT (1*32+25) /* FXSR optimizations */ --#define X86_FEATURE_RDTSCP (1*32+27) /* RDTSCP */ --#define X86_FEATURE_LM (1*32+29) /* Long Mode (x86-64) */ --#define X86_FEATURE_3DNOWEXT (1*32+30) /* AMD 3DNow! extensions */ --#define X86_FEATURE_3DNOW (1*32+31) /* 3DNow! */ -- --/* Transmeta-defined CPU features, CPUID level 0x80860001, word 2 */ --#define X86_FEATURE_RECOVERY (2*32+ 0) /* CPU in recovery mode */ --#define X86_FEATURE_LONGRUN (2*32+ 1) /* Longrun power control */ --#define X86_FEATURE_LRTI (2*32+ 3) /* LongRun table interface */ -- --/* Other features, Linux-defined mapping, word 3 */ --/* This range is used for feature bits which conflict or are synthesized */ --#define X86_FEATURE_CXMMX (3*32+ 0) /* Cyrix MMX extensions */ --#define X86_FEATURE_K6_MTRR (3*32+ 1) /* AMD K6 nonstandard MTRRs */ --#define X86_FEATURE_CYRIX_ARR (3*32+ 2) /* Cyrix ARRs (= MTRRs) */ --#define X86_FEATURE_CENTAUR_MCR (3*32+ 3) /* Centaur MCRs (= MTRRs) */ --#define X86_FEATURE_REP_GOOD (3*32+ 4) /* rep microcode works well on this CPU */ --#define X86_FEATURE_CONSTANT_TSC (3*32+5) /* TSC runs at constant rate */ --#define X86_FEATURE_SYNC_RDTSC (3*32+6) /* RDTSC syncs CPU core */ --#define X86_FEATURE_FXSAVE_LEAK (3*32+7) /* FIP/FOP/FDP leaks through FXSAVE */ --#define X86_FEATURE_UP (3*32+8) /* SMP kernel running on UP */ --#define X86_FEATURE_ARCH_PERFMON (3*32+9) /* Intel Architectural PerfMon */ --#define X86_FEATURE_PEBS (3*32+10) /* Precise-Event Based Sampling */ --#define X86_FEATURE_BTS (3*32+11) /* Branch Trace Store */ -- --/* Intel-defined CPU features, CPUID level 0x00000001 (ecx), word 4 */ --#define X86_FEATURE_XMM3 (4*32+ 0) /* Streaming SIMD Extensions-3 */ --#define X86_FEATURE_MWAIT (4*32+ 3) /* Monitor/Mwait support */ --#define X86_FEATURE_DSCPL (4*32+ 4) /* CPL Qualified Debug Store */ --#define X86_FEATURE_EST (4*32+ 7) /* Enhanced SpeedStep */ --#define X86_FEATURE_TM2 (4*32+ 8) /* Thermal Monitor 2 */ --#define X86_FEATURE_CID (4*32+10) /* Context ID */ --#define X86_FEATURE_CX16 (4*32+13) /* CMPXCHG16B */ --#define X86_FEATURE_XTPR (4*32+14) /* Send Task Priority Messages */ -- --/* VIA/Cyrix/Centaur-defined CPU features, CPUID level 0xC0000001, word 5 */ --#define X86_FEATURE_XSTORE (5*32+ 2) /* on-CPU RNG present (xstore insn) */ --#define X86_FEATURE_XSTORE_EN (5*32+ 3) /* on-CPU RNG enabled */ --#define X86_FEATURE_XCRYPT (5*32+ 6) /* on-CPU crypto (xcrypt insn) */ --#define X86_FEATURE_XCRYPT_EN (5*32+ 7) /* on-CPU crypto enabled */ -- --/* More extended AMD flags: CPUID level 0x80000001, ecx, word 6 */ --#define X86_FEATURE_LAHF_LM (6*32+ 0) /* LAHF/SAHF in long mode */ --#define X86_FEATURE_CMP_LEGACY (6*32+ 1) /* If yes HyperThreading not valid */ -- --#define cpu_has(c, bit) test_bit(bit, (c)->x86_capability) --#define boot_cpu_has(bit) test_bit(bit, boot_cpu_data.x86_capability) -- --#define cpu_has_fpu 1 -+#undef cpu_has_vme - #define cpu_has_vme 0 --#define cpu_has_de 1 --#define cpu_has_pse 1 --#define cpu_has_tsc 1 -+ -+#undef cpu_has_pae - #define cpu_has_pae ___BUG___ --#define cpu_has_pge 1 --#define cpu_has_apic boot_cpu_has(X86_FEATURE_APIC) --#define cpu_has_mtrr 1 --#define cpu_has_mmx 1 --#define cpu_has_fxsr 1 --#define cpu_has_xmm 1 --#define cpu_has_xmm2 1 --#define cpu_has_xmm3 boot_cpu_has(X86_FEATURE_XMM3) --#define cpu_has_ht boot_cpu_has(X86_FEATURE_HT) -+ -+#undef cpu_has_mp - #define cpu_has_mp 1 /* XXX */ -+ -+#undef cpu_has_k6_mtrr - #define cpu_has_k6_mtrr 0 -+ -+#undef cpu_has_cyrix_arr - #define cpu_has_cyrix_arr 0 -+ -+#undef cpu_has_centaur_mcr - #define cpu_has_centaur_mcr 0 --#define cpu_has_clflush boot_cpu_has(X86_FEATURE_CLFLSH) --#define cpu_has_ds boot_cpu_has(X86_FEATURE_DS) --#define cpu_has_pebs boot_cpu_has(X86_FEATURE_PEBS) --#define cpu_has_bts boot_cpu_has(X86_FEATURE_BTS) - - #endif /* __ASM_X8664_CPUFEATURE_H */ ---- a/include/asm-x86_64/e820.h -+++ b/include/asm-x86_64/e820.h -@@ -11,7 +11,9 @@ - #ifndef __E820_HEADER - #define __E820_HEADER - --#include <linux/mmzone.h> -+#ifndef _SETUP -+# include <linux/mmzone.h> -+#endif - - #define E820MAP 0x2d0 /* our map */ - #define E820MAX 128 /* number of entries in E820MAP */ -@@ -30,7 +32,7 @@ - } __attribute__((packed)); - - struct e820map { -- int nr_map; -+ u32 nr_map; - struct e820entry map[E820MAX]; - }; - ---- a/include/asm-x86_64/processor.h -+++ b/include/asm-x86_64/processor.h -@@ -100,6 +100,7 @@ - - extern void identify_cpu(struct cpuinfo_x86 *); - extern void print_cpu_info(struct cpuinfo_x86 *); -+extern void init_scattered_cpuid_features(struct cpuinfo_x86 *c); - extern unsigned int init_intel_cacheinfo(struct cpuinfo_x86 *c); - extern unsigned short num_cache_leaves; - -@@ -368,8 +369,6 @@ - asm volatile("cpuid" : "=a" (tmp) : "0" (1) : "ebx","ecx","edx","memory"); - } - --#define cpu_has_fpu 1 -- - #define ARCH_HAS_PREFETCH - static inline void prefetch(void *x) - { ---- /dev/null -+++ b/include/asm-x86_64/required-features.h -@@ -0,0 +1,46 @@ -+#ifndef _ASM_REQUIRED_FEATURES_H -+#define _ASM_REQUIRED_FEATURES_H 1 -+ -+/* Define minimum CPUID feature set for kernel These bits are checked -+ really early to actually display a visible error message before the -+ kernel dies. Make sure to assign features to the proper mask! -+ -+ The real information is in arch/x86_64/Kconfig.cpu, this just converts -+ the CONFIGs into a bitmask */ -+ -+/* x86-64 baseline features */ -+#define NEED_FPU (1<<(X86_FEATURE_FPU & 31)) -+#define NEED_PSE (1<<(X86_FEATURE_PSE & 31)) -+#define NEED_MSR (1<<(X86_FEATURE_MSR & 31)) -+#define NEED_PAE (1<<(X86_FEATURE_PAE & 31)) -+#define NEED_CX8 (1<<(X86_FEATURE_CX8 & 31)) -+#define NEED_PGE (1<<(X86_FEATURE_PGE & 31)) -+#define NEED_FXSR (1<<(X86_FEATURE_FXSR & 31)) -+#define NEED_CMOV (1<<(X86_FEATURE_CMOV & 31)) -+#define NEED_XMM (1<<(X86_FEATURE_XMM & 31)) -+#define NEED_XMM2 (1<<(X86_FEATURE_XMM2 & 31)) -+ -+#define REQUIRED_MASK0 (NEED_FPU|NEED_PSE|NEED_MSR|NEED_PAE|\ -+ NEED_CX8|NEED_PGE|NEED_FXSR|NEED_CMOV|\ -+ NEED_XMM|NEED_XMM2) -+#define SSE_MASK (NEED_XMM|NEED_XMM2) -+ -+/* x86-64 baseline features */ -+#define NEED_LM (1<<(X86_FEATURE_LM & 31)) -+ -+#ifdef CONFIG_X86_USE_3DNOW -+# define NEED_3DNOW (1<<(X86_FEATURE_3DNOW & 31)) -+#else -+# define NEED_3DNOW 0 -+#endif -+ -+#define REQUIRED_MASK1 (NEED_LM|NEED_3DNOW) -+ -+#define REQUIRED_MASK2 0 -+#define REQUIRED_MASK3 0 -+#define REQUIRED_MASK4 0 -+#define REQUIRED_MASK5 0 -+#define REQUIRED_MASK6 0 -+#define REQUIRED_MASK7 0 -+ -+#endif ---- a/include/asm-x86_64/segment.h -+++ b/include/asm-x86_64/segment.h -@@ -3,6 +3,14 @@ - - #include <asm/cache.h> - -+/* Simple and small GDT entries for booting only */ -+ -+#define GDT_ENTRY_BOOT_CS 2 -+#define __BOOT_CS (GDT_ENTRY_BOOT_CS * 8) -+ -+#define GDT_ENTRY_BOOT_DS (GDT_ENTRY_BOOT_CS + 1) -+#define __BOOT_DS (GDT_ENTRY_BOOT_DS * 8) -+ - #define __KERNEL_CS 0x10 - #define __KERNEL_DS 0x18 - ---- a/include/linux/edd.h -+++ b/include/linux/edd.h -@@ -49,10 +49,6 @@ - #define EDD_MBR_SIG_MAX 16 /* max number of signatures to store */ - #define EDD_MBR_SIG_NR_BUF 0x1ea /* addr of number of MBR signtaures at EDD_MBR_SIG_BUF - in boot_params - treat this as 1 byte */ --#define EDD_CL_EQUALS 0x3d646465 /* "edd=" */ --#define EDD_CL_OFF 0x666f /* "of" for off */ --#define EDD_CL_SKIP 0x6b73 /* "sk" for skipmbr */ --#define EDD_CL_ON 0x6e6f /* "on" for on */ - - #ifndef __ASSEMBLY__ - ---- a/include/linux/screen_info.h -+++ b/include/linux/screen_info.h -@@ -10,7 +10,7 @@ - struct screen_info { - u8 orig_x; /* 0x00 */ - u8 orig_y; /* 0x01 */ -- u16 dontuse1; /* 0x02 -- EXT_MEM_K sits here */ -+ u16 ext_mem_k; /* 0x02 */ - u16 orig_video_page; /* 0x04 */ - u8 orig_video_mode; /* 0x06 */ - u8 orig_video_cols; /* 0x07 */ -@@ -27,7 +27,7 @@ - u16 lfb_depth; /* 0x16 */ - u32 lfb_base; /* 0x18 */ - u32 lfb_size; /* 0x1c */ -- u16 dontuse2, dontuse3; /* 0x20 -- CL_MAGIC and CL_OFFSET here */ -+ u16 cl_magic, cl_offset; /* 0x20 */ - u16 lfb_linelength; /* 0x24 */ - u8 red_size; /* 0x26 */ - u8 red_pos; /* 0x27 */ -@@ -42,9 +42,8 @@ - u16 pages; /* 0x32 */ - u16 vesa_attributes; /* 0x34 */ - u32 capabilities; /* 0x36 */ -- /* 0x3a -- 0x3b reserved for future expansion */ -- /* 0x3c -- 0x3f micro stack for relocatable kernels */ --}; -+ u8 _reserved[6]; /* 0x3a */ -+} __attribute__((packed)); - - extern struct screen_info screen_info; - diff --git a/target/linux/generic-2.6/patches-2.6.22/015-x86_newsetup_fixup.patch b/target/linux/generic-2.6/patches-2.6.22/015-x86_newsetup_fixup.patch deleted file mode 100644 index 3576cff..0000000 --- a/target/linux/generic-2.6/patches-2.6.22/015-x86_newsetup_fixup.patch +++ /dev/null @@ -1,19 +0,0 @@ -From: Andrew Morton <akpm@linux-foundation.org> - -Signed-off-by: Andrew Morton <akpm@linux-foundation.org> ---- - - arch/i386/kernel/cpu/Makefile | 2 +- - 1 file changed, 1 insertion(+), 1 deletion(-) - ---- a/arch/i386/kernel/cpu/Makefile -+++ b/arch/i386/kernel/cpu/Makefile -@@ -8,7 +8,7 @@ - obj-y += cyrix.o - obj-y += centaur.o - obj-y += transmeta.o --obj-y += intel.o intel_cacheinfo.o -+obj-y += intel.o intel_cacheinfo.o addon_cpuid_features.o - obj-y += rise.o - obj-y += nexgen.o - obj-y += umc.o diff --git a/target/linux/generic-2.6/patches-2.6.22/060-block2mtd_init.patch b/target/linux/generic-2.6/patches-2.6.22/060-block2mtd_init.patch deleted file mode 100644 index e92499f..0000000 --- a/target/linux/generic-2.6/patches-2.6.22/060-block2mtd_init.patch +++ /dev/null @@ -1,112 +0,0 @@ ---- a/drivers/mtd/devices/block2mtd.c -+++ b/drivers/mtd/devices/block2mtd.c -@@ -16,6 +16,7 @@ - #include <linux/list.h> - #include <linux/init.h> - #include <linux/mtd/mtd.h> -+#include <linux/mtd/partitions.h> - #include <linux/buffer_head.h> - #include <linux/mutex.h> - #include <linux/mount.h> -@@ -237,10 +238,11 @@ - - - /* FIXME: ensure that mtd->size % erase_size == 0 */ --static struct block2mtd_dev *add_device(char *devname, int erase_size) -+static struct block2mtd_dev *add_device(char *devname, int erase_size, char *mtdname) - { - struct block_device *bdev; - struct block2mtd_dev *dev; -+ struct mtd_partition *part; - - if (!devname) - return NULL; -@@ -279,14 +281,18 @@ - - /* Setup the MTD structure */ - /* make the name contain the block device in */ -- dev->mtd.name = kmalloc(sizeof("block2mtd: ") + strlen(devname), -- GFP_KERNEL); -+ -+ if (!mtdname) -+ mtdname = devname; -+ -+ dev->mtd.name = kmalloc(strlen(mtdname) + 1, GFP_KERNEL); -+ - if (!dev->mtd.name) - goto devinit_err; -+ -+ strcpy(dev->mtd.name, mtdname); - -- sprintf(dev->mtd.name, "block2mtd: %s", devname); -- -- dev->mtd.size = dev->blkdev->bd_inode->i_size & PAGE_MASK; -+ dev->mtd.size = dev->blkdev->bd_inode->i_size & PAGE_MASK & ~(erase_size - 1); - dev->mtd.erasesize = erase_size; - dev->mtd.writesize = 1; - dev->mtd.type = MTD_RAM; -@@ -298,15 +304,18 @@ - dev->mtd.read = block2mtd_read; - dev->mtd.priv = dev; - dev->mtd.owner = THIS_MODULE; -- -- if (add_mtd_device(&dev->mtd)) { -+ -+ part = kzalloc(sizeof(struct mtd_partition), GFP_KERNEL); -+ part->name = dev->mtd.name; -+ part->offset = 0; -+ part->size = dev->mtd.size; -+ if (add_mtd_partitions(&dev->mtd, part, 1)) { - /* Device didnt get added, so free the entry */ - goto devinit_err; - } - list_add(&dev->list, &blkmtd_device_list); - INFO("mtd%d: [%s] erase_size = %dKiB [%d]", dev->mtd.index, -- dev->mtd.name + strlen("blkmtd: "), -- dev->mtd.erasesize >> 10, dev->mtd.erasesize); -+ mtdname, dev->mtd.erasesize >> 10, dev->mtd.erasesize); - return dev; - - devinit_err: -@@ -379,9 +388,9 @@ - - static int block2mtd_setup2(const char *val) - { -- char buf[80 + 12]; /* 80 for device, 12 for erase size */ -+ char buf[80 + 12 + 80]; /* 80 for device, 12 for erase size, 80 for name */ - char *str = buf; -- char *token[2]; -+ char *token[3]; - char *name; - size_t erase_size = PAGE_SIZE; - int i, ret; -@@ -392,7 +401,7 @@ - strcpy(str, val); - kill_final_newline(str); - -- for (i = 0; i < 2; i++) -+ for (i = 0; i < 3; i++) - token[i] = strsep(&str, ","); - - if (str) -@@ -412,8 +421,10 @@ - parse_err("illegal erase size"); - } - } -+ if (token[2] && (strlen(token[2]) + 1 > 80)) -+ parse_err("mtd device name too long"); - -- add_device(name, erase_size); -+ add_device(name, erase_size, token[2]); - - return 0; - } -@@ -447,7 +458,7 @@ - - - module_param_call(block2mtd, block2mtd_setup, NULL, NULL, 0200); --MODULE_PARM_DESC(block2mtd, "Device to use. \"block2mtd=<dev>[,<erasesize>]\""); -+MODULE_PARM_DESC(block2mtd, "Device to use. \"block2mtd=<dev>[,<erasesize>[,<name>]]\""); - - static int __init block2mtd_init(void) - { diff --git a/target/linux/generic-2.6/patches-2.6.22/065-rootfs_split.patch b/target/linux/generic-2.6/patches-2.6.22/065-rootfs_split.patch deleted file mode 100644 index 12723d8..0000000 --- a/target/linux/generic-2.6/patches-2.6.22/065-rootfs_split.patch +++ /dev/null @@ -1,944 +0,0 @@ ---- a/drivers/mtd/Kconfig -+++ b/drivers/mtd/Kconfig -@@ -47,6 +47,16 @@ - devices. Partitioning on NFTL 'devices' is a different - that's the - 'normal' form of partitioning used on a block device. - -+config MTD_ROOTFS_ROOT_DEV -+ bool "Automatically set 'rootfs' partition to be root filesystem" -+ depends on MTD_PARTITIONS -+ default y -+ -+config MTD_ROOTFS_SPLIT -+ bool "Automatically split 'rootfs' partition for squashfs" -+ depends on MTD_PARTITIONS -+ default y -+ - config MTD_REDBOOT_PARTS - tristate "RedBoot partition table parsing" - depends on MTD_PARTITIONS ---- a/drivers/mtd/mtdpart.c -+++ b/drivers/mtd/mtdpart.c -@@ -20,6 +20,8 @@ - #include <linux/mtd/mtd.h> - #include <linux/mtd/partitions.h> - #include <linux/mtd/compatmac.h> -+#include <linux/squashfs_fs.h> -+#include <linux/root_dev.h> - - /* Our partition linked list */ - static LIST_HEAD(mtd_partitions); -@@ -39,7 +41,7 @@ - * the pointer to that structure with this macro. - */ - #define PART(x) ((struct mtd_part *)(x)) -- -+#define IS_PART(mtd) (mtd->read == part_read) - - /* - * MTD methods which simply translate the effective address and pass through -@@ -308,6 +310,312 @@ - return 0; - } - -+static u_int32_t cur_offset = 0; -+static int add_one_partition(struct mtd_info *master, const struct mtd_partition *part, -+ int i, struct mtd_part **slp) -+{ -+ struct mtd_part *slave; -+ -+ /* allocate the partition structure */ -+ slave = kzalloc (sizeof(*slave), GFP_KERNEL); -+ if (!slave) { -+ printk ("memory allocation error while creating partitions for \"%s\"\n", -+ master->name); -+ del_mtd_partitions(master); -+ return -ENOMEM; -+ } -+ list_add(&slave->list, &mtd_partitions); -+ -+ /* set up the MTD object for this partition */ -+ slave->mtd.type = master->type; -+ slave->mtd.flags = master->flags & ~part->mask_flags; -+ slave->mtd.size = part->size; -+ slave->mtd.writesize = master->writesize; -+ slave->mtd.oobsize = master->oobsize; -+ slave->mtd.oobavail = master->oobavail; -+ slave->mtd.subpage_sft = master->subpage_sft; -+ -+ slave->mtd.name = part->name; -+ slave->mtd.owner = master->owner; -+ -+ slave->mtd.read = part_read; -+ slave->mtd.write = part_write; -+ slave->mtd.refresh_device = part->refresh_partition; -+ -+ if(master->point && master->unpoint){ -+ slave->mtd.point = part_point; -+ slave->mtd.unpoint = part_unpoint; -+ } -+ -+ if (master->read_oob) -+ slave->mtd.read_oob = part_read_oob; -+ if (master->write_oob) -+ slave->mtd.write_oob = part_write_oob; -+ if(master->read_user_prot_reg) -+ slave->mtd.read_user_prot_reg = part_read_user_prot_reg; -+ if(master->read_fact_prot_reg) -+ slave->mtd.read_fact_prot_reg = part_read_fact_prot_reg; -+ if(master->write_user_prot_reg) -+ slave->mtd.write_user_prot_reg = part_write_user_prot_reg; -+ if(master->lock_user_prot_reg) -+ slave->mtd.lock_user_prot_reg = part_lock_user_prot_reg; -+ if(master->get_user_prot_info) -+ slave->mtd.get_user_prot_info = part_get_user_prot_info; -+ if(master->get_fact_prot_info) -+ slave->mtd.get_fact_prot_info = part_get_fact_prot_info; -+ if (master->sync) -+ slave->mtd.sync = part_sync; -+ if (!i && master->suspend && master->resume) { -+ slave->mtd.suspend = part_suspend; -+ slave->mtd.resume = part_resume; -+ } -+ if (master->writev) -+ slave->mtd.writev = part_writev; -+ if (master->lock) -+ slave->mtd.lock = part_lock; -+ if (master->unlock) -+ slave->mtd.unlock = part_unlock; -+ if (master->block_isbad) -+ slave->mtd.block_isbad = part_block_isbad; -+ if (master->block_markbad) -+ slave->mtd.block_markbad = part_block_markbad; -+ slave->mtd.erase = part_erase; -+ slave->master = master; -+ slave->offset = part->offset; -+ slave->index = i; -+ -+ if (slave->offset == MTDPART_OFS_APPEND) -+ slave->offset = cur_offset; -+ if (slave->offset == MTDPART_OFS_NXTBLK) { -+ slave->offset = cur_offset; -+ if ((cur_offset % master->erasesize) != 0) { -+ /* Round up to next erasesize */ -+ slave->offset = ((cur_offset / master->erasesize) + 1) * master->erasesize; -+ printk(KERN_NOTICE "Moving partition %d: " -+ "0x%08x -> 0x%08x\n", i, -+ cur_offset, slave->offset); -+ } -+ } -+ if (slave->mtd.size == MTDPART_SIZ_FULL) -+ slave->mtd.size = master->size - slave->offset; -+ cur_offset = slave->offset + slave->mtd.size; -+ -+ printk (KERN_NOTICE "0x%08x-0x%08x : \"%s\"\n", slave->offset, -+ slave->offset + slave->mtd.size, slave->mtd.name); -+ -+ /* let's do some sanity checks */ -+ if (slave->offset >= master->size) { -+ /* let's register it anyway to preserve ordering */ -+ slave->offset = 0; -+ slave->mtd.size = 0; -+ printk ("mtd: partition \"%s\" is out of reach -- disabled\n", -+ part->name); -+ } -+ if (slave->offset + slave->mtd.size > master->size) { -+ slave->mtd.size = master->size - slave->offset; -+ printk ("mtd: partition \"%s\" extends beyond the end of device \"%s\" -- size truncated to %#x\n", -+ part->name, master->name, slave->mtd.size); -+ } -+ if (master->numeraseregions>1) { -+ /* Deal with variable erase size stuff */ -+ int i; -+ struct mtd_erase_region_info *regions = master->eraseregions; -+ -+ /* Find the first erase regions which is part of this partition. */ -+ for (i=0; i < master->numeraseregions && slave->offset >= regions[i].offset; i++) -+ ; -+ -+ for (i--; i < master->numeraseregions && slave->offset + slave->mtd.size > regions[i].offset; i++) { -+ if (slave->mtd.erasesize < regions[i].erasesize) { -+ slave->mtd.erasesize = regions[i].erasesize; -+ } -+ } -+ } else { -+ /* Single erase size */ -+ slave->mtd.erasesize = master->erasesize; -+ } -+ -+ if ((slave->mtd.flags & MTD_WRITEABLE) && -+ (slave->offset % slave->mtd.erasesize)) { -+ /* Doesn't start on a boundary of major erase size */ -+ /* FIXME: Let it be writable if it is on a boundary of _minor_ erase size though */ -+ slave->mtd.flags &= ~MTD_WRITEABLE; -+ printk ("mtd: partition \"%s\" doesn't start on an erase block boundary -- force read-only\n", -+ part->name); -+ } -+ if ((slave->mtd.flags & MTD_WRITEABLE) && -+ (slave->mtd.size % slave->mtd.erasesize)) { -+ slave->mtd.flags &= ~MTD_WRITEABLE; -+ printk ("mtd: partition \"%s\" doesn't end on an erase block -- force read-only\n", -+ part->name); -+ } -+ -+ slave->mtd.ecclayout = master->ecclayout; -+ if (master->block_isbad) { -+ uint32_t offs = 0; -+ -+ while(offs < slave->mtd.size) { -+ if (master->block_isbad(master, -+ offs + slave->offset)) -+ slave->mtd.ecc_stats.badblocks++; -+ offs += slave->mtd.erasesize; -+ } -+ } -+ -+ if(part->mtdp) -+ { /* store the object pointer (caller may or may not register it */ -+ *part->mtdp = &slave->mtd; -+ slave->registered = 0; -+ } -+ else -+ { -+ /* register our partition */ -+ add_mtd_device(&slave->mtd); -+ slave->registered = 1; -+ } -+ -+ if (slp) -+ *slp = slave; -+ -+ return 0; -+} -+ -+#ifdef CONFIG_MTD_ROOTFS_SPLIT -+#define ROOTFS_SPLIT_NAME "rootfs_data" -+#define ROOTFS_REMOVED_NAME "<removed>" -+static int split_squashfs(struct mtd_info *master, int offset, int *split_offset) -+{ -+ char buf[512]; -+ struct squashfs_super_block *sb = (struct squashfs_super_block *) buf; -+ int len, ret; -+ -+ ret = master->read(master, offset, sizeof(*sb), &len, buf); -+ if (ret || (len != sizeof(*sb))) { -+ printk(KERN_ALERT "split_squashfs: error occured while reading " -+ "from \"%s\"\n", master->name); -+ return -EINVAL; -+ } -+ -+ if (*((u32 *) buf) != SQUASHFS_MAGIC) { -+ printk(KERN_ALERT "split_squasfs: no squashfs found in \"%s\"\n", -+ master->name); -+ *split_offset = 0; -+ return 0; -+ } -+ -+ if (sb->bytes_used <= 0) { -+ printk(KERN_ALERT "split_squashfs: squashfs is empty in \"%s\"\n", -+ master->name); -+ *split_offset = 0; -+ return 0; -+ } -+ -+ len = (u32) sb->bytes_used; -+ len += (offset & 0x000fffff); -+ len += (master->erasesize - 1); -+ len &= ~(master->erasesize - 1); -+ len -= (offset & 0x000fffff); -+ *split_offset = offset + len; -+ -+ return 0; -+} -+ -+static int split_rootfs_data(struct mtd_info *master, struct mtd_info *rpart, struct mtd_partition *part, -+ int index) -+{ -+ struct mtd_partition *dpart; -+ struct mtd_part *slave = NULL; -+ int split_offset = 0; -+ int ret; -+ -+ ret = split_squashfs(master, part->offset, &split_offset); -+ if (ret) -+ return ret; -+ -+ if (split_offset <= 0) -+ return 0; -+ -+ dpart = kmalloc(sizeof(*part)+sizeof(ROOTFS_SPLIT_NAME)+1, GFP_KERNEL); -+ if (dpart == NULL) { -+ printk(KERN_INFO "split_squashfs: no memory for partition \"%s\"\n", -+ ROOTFS_SPLIT_NAME); -+ return -ENOMEM; -+ } -+ -+ memcpy(dpart, part, sizeof(*part)); -+ dpart->name = (unsigned char *)&dpart[1]; -+ strcpy(dpart->name, ROOTFS_SPLIT_NAME); -+ -+ dpart->size -= split_offset - dpart->offset; -+ dpart->offset = split_offset; -+ -+ if (dpart == NULL) -+ return 1; -+ -+ printk(KERN_INFO "mtd: partition \"%s\" created automatically, ofs=%X, len=%X \n", -+ ROOTFS_SPLIT_NAME, dpart->offset, dpart->size); -+ -+ ret = add_one_partition(master, dpart, index, &slave); -+ if (ret) -+ kfree(dpart); -+ else if (slave) -+ rpart->split = &slave->mtd; -+ -+ return ret; -+} -+ -+static int refresh_rootfs_split(struct mtd_info *mtd) -+{ -+ struct mtd_partition tpart; -+ struct mtd_part *part; -+ int index = 0; -+ int offset, size; -+ int ret; -+ -+ part = PART(mtd); -+ -+ /* check for the new squashfs offset first */ -+ ret = split_squashfs(part->master, part->offset, &offset); -+ if (ret) -+ return ret; -+ -+ if ((offset > 0) && !mtd->split) { -+ printk(KERN_INFO "%s: creating new split partition for \"%s\"\n", __func__, mtd->name); -+ /* if we don't have a rootfs split partition, create a new one */ -+ tpart.name = mtd->name; -+ tpart.size = mtd->size; -+ tpart.offset = part->offset; -+ -+ /* find the index of the last partition */ -+ if (!list_empty(&mtd_partitions)) -+ index = list_first_entry(&mtd_partitions, struct mtd_part, list)->index + 1; -+ -+ return split_rootfs_data(part->master, &part->mtd, &tpart, index); -+ } else if ((offset > 0) && mtd->split) { -+ /* update the offsets of the existing partition */ -+ size = mtd->size + part->offset - offset; -+ -+ part = PART(mtd->split); -+ part->offset = offset; -+ part->mtd.size = size; -+ printk(KERN_INFO "%s: %s partition \"" ROOTFS_SPLIT_NAME "\", offset: 0x%06x (0x%06x)\n", -+ __func__, (!strcmp(part->mtd.name, ROOTFS_SPLIT_NAME) ? "updating" : "creating"), -+ part->offset, part->mtd.size); -+ strcpy(part->mtd.name, ROOTFS_SPLIT_NAME); -+ } else if ((offset <= 0) && mtd->split) { -+ printk(KERN_INFO "%s: removing partition \"%s\"\n", __func__, mtd->split->name); -+ -+ /* mark existing partition as removed */ -+ part = PART(mtd->split); -+ strcpy(part->mtd.name, ROOTFS_REMOVED_NAME); -+ part->offset = 0; -+ part->mtd.size = 0; -+ } -+ -+ return 0; -+} -+#endif /* CONFIG_MTD_ROOTFS_SPLIT */ -+ - /* - * This function, given a master MTD object and a partition table, creates - * and registers slave MTD objects which are bound to the master according to -@@ -320,168 +628,31 @@ - int nbparts) - { - struct mtd_part *slave; -- u_int32_t cur_offset = 0; -- int i; -+ struct mtd_partition *part; -+ int i, j, ret = 0; - - printk (KERN_NOTICE "Creating %d MTD partitions on \"%s\":\n", nbparts, master->name); - -- for (i = 0; i < nbparts; i++) { -- -- /* allocate the partition structure */ -- slave = kzalloc (sizeof(*slave), GFP_KERNEL); -- if (!slave) { -- printk ("memory allocation error while creating partitions for \"%s\"\n", -- master->name); -- del_mtd_partitions(master); -- return -ENOMEM; -- } -- list_add(&slave->list, &mtd_partitions); -- -- /* set up the MTD object for this partition */ -- slave->mtd.type = master->type; -- slave->mtd.flags = master->flags & ~parts[i].mask_flags; -- slave->mtd.size = parts[i].size; -- slave->mtd.writesize = master->writesize; -- slave->mtd.oobsize = master->oobsize; -- slave->mtd.oobavail = master->oobavail; -- slave->mtd.subpage_sft = master->subpage_sft; -- -- slave->mtd.name = parts[i].name; -- slave->mtd.owner = master->owner; -- -- slave->mtd.read = part_read; -- slave->mtd.write = part_write; -- -- if(master->point && master->unpoint){ -- slave->mtd.point = part_point; -- slave->mtd.unpoint = part_unpoint; -- } -- -- if (master->read_oob) -- slave->mtd.read_oob = part_read_oob; -- if (master->write_oob) -- slave->mtd.write_oob = part_write_oob; -- if(master->read_user_prot_reg) -- slave->mtd.read_user_prot_reg = part_read_user_prot_reg; -- if(master->read_fact_prot_reg) -- slave->mtd.read_fact_prot_reg = part_read_fact_prot_reg; -- if(master->write_user_prot_reg) -- slave->mtd.write_user_prot_reg = part_write_user_prot_reg; -- if(master->lock_user_prot_reg) -- slave->mtd.lock_user_prot_reg = part_lock_user_prot_reg; -- if(master->get_user_prot_info) -- slave->mtd.get_user_prot_info = part_get_user_prot_info; -- if(master->get_fact_prot_info) -- slave->mtd.get_fact_prot_info = part_get_fact_prot_info; -- if (master->sync) -- slave->mtd.sync = part_sync; -- if (!i && master->suspend && master->resume) { -- slave->mtd.suspend = part_suspend; -- slave->mtd.resume = part_resume; -- } -- if (master->writev) -- slave->mtd.writev = part_writev; -- if (master->lock) -- slave->mtd.lock = part_lock; -- if (master->unlock) -- slave->mtd.unlock = part_unlock; -- if (master->block_isbad) -- slave->mtd.block_isbad = part_block_isbad; -- if (master->block_markbad) -- slave->mtd.block_markbad = part_block_markbad; -- slave->mtd.erase = part_erase; -- slave->master = master; -- slave->offset = parts[i].offset; -- slave->index = i; -- -- if (slave->offset == MTDPART_OFS_APPEND) -- slave->offset = cur_offset; -- if (slave->offset == MTDPART_OFS_NXTBLK) { -- slave->offset = cur_offset; -- if ((cur_offset % master->erasesize) != 0) { -- /* Round up to next erasesize */ -- slave->offset = ((cur_offset / master->erasesize) + 1) * master->erasesize; -- printk(KERN_NOTICE "Moving partition %d: " -- "0x%08x -> 0x%08x\n", i, -- cur_offset, slave->offset); -- } -- } -- if (slave->mtd.size == MTDPART_SIZ_FULL) -- slave->mtd.size = master->size - slave->offset; -- cur_offset = slave->offset + slave->mtd.size; -- -- printk (KERN_NOTICE "0x%08x-0x%08x : \"%s\"\n", slave->offset, -- slave->offset + slave->mtd.size, slave->mtd.name); -- -- /* let's do some sanity checks */ -- if (slave->offset >= master->size) { -- /* let's register it anyway to preserve ordering */ -- slave->offset = 0; -- slave->mtd.size = 0; -- printk ("mtd: partition \"%s\" is out of reach -- disabled\n", -- parts[i].name); -- } -- if (slave->offset + slave->mtd.size > master->size) { -- slave->mtd.size = master->size - slave->offset; -- printk ("mtd: partition \"%s\" extends beyond the end of device \"%s\" -- size truncated to %#x\n", -- parts[i].name, master->name, slave->mtd.size); -- } -- if (master->numeraseregions>1) { -- /* Deal with variable erase size stuff */ -- int i; -- struct mtd_erase_region_info *regions = master->eraseregions; -- -- /* Find the first erase regions which is part of this partition. */ -- for (i=0; i < master->numeraseregions && slave->offset >= regions[i].offset; i++) -- ; -- -- for (i--; i < master->numeraseregions && slave->offset + slave->mtd.size > regions[i].offset; i++) { -- if (slave->mtd.erasesize < regions[i].erasesize) { -- slave->mtd.erasesize = regions[i].erasesize; -- } -+ for (i = 0, j = 0; i < nbparts; i++) { -+ part = (struct mtd_partition *) &parts[i]; -+ ret = add_one_partition(master, part, j, &slave); -+ if (ret) -+ return ret; -+ j++; -+ -+ if (strcmp(part->name, "rootfs") == 0 && slave->registered) { -+#ifdef CONFIG_MTD_ROOTFS_ROOT_DEV -+ if (ROOT_DEV == 0) { -+ printk(KERN_NOTICE "mtd: partition \"rootfs\" " -+ "set to be root filesystem\n"); -+ ROOT_DEV = MKDEV(MTD_BLOCK_MAJOR, slave->mtd.index); - } -- } else { -- /* Single erase size */ -- slave->mtd.erasesize = master->erasesize; -- } -- -- if ((slave->mtd.flags & MTD_WRITEABLE) && -- (slave->offset % slave->mtd.erasesize)) { -- /* Doesn't start on a boundary of major erase size */ -- /* FIXME: Let it be writable if it is on a boundary of _minor_ erase size though */ -- slave->mtd.flags &= ~MTD_WRITEABLE; -- printk ("mtd: partition \"%s\" doesn't start on an erase block boundary -- force read-only\n", -- parts[i].name); -- } -- if ((slave->mtd.flags & MTD_WRITEABLE) && -- (slave->mtd.size % slave->mtd.erasesize)) { -- slave->mtd.flags &= ~MTD_WRITEABLE; -- printk ("mtd: partition \"%s\" doesn't end on an erase block -- force read-only\n", -- parts[i].name); -- } -- -- slave->mtd.ecclayout = master->ecclayout; -- if (master->block_isbad) { -- uint32_t offs = 0; -- -- while(offs < slave->mtd.size) { -- if (master->block_isbad(master, -- offs + slave->offset)) -- slave->mtd.ecc_stats.badblocks++; -- offs += slave->mtd.erasesize; -- } -- } -- -- if(parts[i].mtdp) -- { /* store the object pointer (caller may or may not register it */ -- *parts[i].mtdp = &slave->mtd; -- slave->registered = 0; -- } -- else -- { -- /* register our partition */ -- add_mtd_device(&slave->mtd); -- slave->registered = 1; -+#endif -+#ifdef CONFIG_MTD_ROOTFS_SPLIT -+ ret = split_rootfs_data(master, &slave->mtd, part, j); -+ if (ret == 0) -+ j++; -+#endif - } - } - -@@ -557,6 +728,32 @@ - return ret; - } - -+int refresh_mtd_partitions(struct mtd_info *mtd) -+{ -+ int ret = 0; -+ -+ if (IS_PART(mtd)) { -+ struct mtd_part *part; -+ struct mtd_info *master; -+ -+ part = PART(mtd); -+ master = part->master; -+ if (master->refresh_device) -+ ret = master->refresh_device(master); -+ } -+ -+ if (!ret && mtd->refresh_device) -+ ret = mtd->refresh_device(mtd); -+ -+#ifdef CONFIG_MTD_ROOTFS_SPLIT -+ if (!ret && IS_PART(mtd) && !strcmp(mtd->name, "rootfs")) -+ refresh_rootfs_split(mtd); -+#endif -+ -+ return 0; -+} -+ - EXPORT_SYMBOL_GPL(parse_mtd_partitions); -+EXPORT_SYMBOL_GPL(refresh_mtd_partitions); - EXPORT_SYMBOL_GPL(register_mtd_parser); - EXPORT_SYMBOL_GPL(deregister_mtd_parser); ---- a/drivers/mtd/devices/block2mtd.c -+++ b/drivers/mtd/devices/block2mtd.c -@@ -34,6 +34,8 @@ - struct block_device *blkdev; - struct mtd_info mtd; - struct mutex write_mutex; -+ rwlock_t bdev_mutex; -+ char devname[0]; - }; - - -@@ -86,6 +88,12 @@ - size_t len = instr->len; - int err; - -+ read_lock(&dev->bdev_mutex); -+ if (!dev->blkdev) { -+ err = -EINVAL; -+ goto done; -+ } -+ - instr->state = MTD_ERASING; - mutex_lock(&dev->write_mutex); - err = _block2mtd_erase(dev, from, len); -@@ -98,6 +106,10 @@ - - instr->state = MTD_ERASE_DONE; - mtd_erase_callback(instr); -+ -+done: -+ read_unlock(&dev->bdev_mutex); -+ - return err; - } - -@@ -109,10 +121,14 @@ - struct page *page; - int index = from >> PAGE_SHIFT; - int offset = from & (PAGE_SIZE-1); -- int cpylen; -+ int cpylen, err = 0; -+ -+ read_lock(&dev->bdev_mutex); -+ if (!dev->blkdev || (from > mtd->size)) { -+ err = -EINVAL; -+ goto done; -+ } - -- if (from > mtd->size) -- return -EINVAL; - if (from + len > mtd->size) - len = mtd->size - from; - -@@ -127,10 +143,14 @@ - len = len - cpylen; - - page = page_read(dev->blkdev->bd_inode->i_mapping, index); -- if (!page) -- return -ENOMEM; -- if (IS_ERR(page)) -- return PTR_ERR(page); -+ if (!page) { -+ err = -ENOMEM; -+ goto done; -+ } -+ if (IS_ERR(page)) { -+ err = PTR_ERR(page); -+ goto done; -+ } - - memcpy(buf, page_address(page) + offset, cpylen); - page_cache_release(page); -@@ -141,7 +161,10 @@ - offset = 0; - index++; - } -- return 0; -+ -+done: -+ read_unlock(&dev->bdev_mutex); -+ return err; - } - - -@@ -193,12 +216,22 @@ - size_t *retlen, const u_char *buf) - { - struct block2mtd_dev *dev = mtd->priv; -- int err; -+ int err = 0; -+ -+ read_lock(&dev->bdev_mutex); -+ if (!dev->blkdev) { -+ err = -EINVAL; -+ goto done; -+ } - - if (!len) -- return 0; -- if (to >= mtd->size) -- return -ENOSPC; -+ goto done; -+ -+ if (to >= mtd->size) { -+ err = -ENOSPC; -+ goto done; -+ } -+ - if (to + len > mtd->size) - len = mtd->size - to; - -@@ -207,6 +240,9 @@ - mutex_unlock(&dev->write_mutex); - if (err > 0) - err = 0; -+ -+done: -+ read_unlock(&dev->bdev_mutex); - return err; - } - -@@ -215,51 +251,29 @@ - static void block2mtd_sync(struct mtd_info *mtd) - { - struct block2mtd_dev *dev = mtd->priv; -- sync_blockdev(dev->blkdev); -- return; --} -- -- --static void block2mtd_free_device(struct block2mtd_dev *dev) --{ -- if (!dev) -- return; -- -- kfree(dev->mtd.name); - -- if (dev->blkdev) { -- invalidate_mapping_pages(dev->blkdev->bd_inode->i_mapping, -- 0, -1); -- close_bdev_excl(dev->blkdev); -- } -+ read_lock(&dev->bdev_mutex); -+ if (dev->blkdev) -+ sync_blockdev(dev->blkdev); -+ read_unlock(&dev->bdev_mutex); - -- kfree(dev); -+ return; - } - - --/* FIXME: ensure that mtd->size % erase_size == 0 */ --static struct block2mtd_dev *add_device(char *devname, int erase_size, char *mtdname) -+static int _open_bdev(struct block2mtd_dev *dev) - { - struct block_device *bdev; -- struct block2mtd_dev *dev; -- struct mtd_partition *part; -- -- if (!devname) -- return NULL; -- -- dev = kzalloc(sizeof(struct block2mtd_dev), GFP_KERNEL); -- if (!dev) -- return NULL; - - /* Get a handle on the device */ -- bdev = open_bdev_excl(devname, O_RDWR, NULL); -+ bdev = open_bdev_excl(dev->devname, O_RDWR, NULL); - #ifndef MODULE - if (IS_ERR(bdev)) { - - /* We might not have rootfs mounted at this point. Try - to resolve the device name by other means. */ - -- dev_t devt = name_to_dev_t(devname); -+ dev_t devt = name_to_dev_t(dev->devname); - if (devt) { - bdev = open_by_devnum(devt, FMODE_WRITE | FMODE_READ); - } -@@ -267,17 +281,96 @@ - #endif - - if (IS_ERR(bdev)) { -- ERROR("error: cannot open device %s", devname); -- goto devinit_err; -+ ERROR("error: cannot open device %s", dev->devname); -+ return 1; - } - dev->blkdev = bdev; - - if (MAJOR(bdev->bd_dev) == MTD_BLOCK_MAJOR) { - ERROR("attempting to use an MTD device as a block device"); -- goto devinit_err; -+ return 1; - } - -+ return 0; -+} -+ -+static void _close_bdev(struct block2mtd_dev *dev) -+{ -+ struct block_device *bdev; -+ -+ if (!dev->blkdev) -+ return; -+ -+ bdev = dev->blkdev; -+ invalidate_mapping_pages(dev->blkdev->bd_inode->i_mapping, 0, -1); -+ close_bdev_excl(dev->blkdev); -+ dev->blkdev = NULL; -+} -+ -+static void block2mtd_free_device(struct block2mtd_dev *dev) -+{ -+ if (!dev) -+ return; -+ -+ kfree(dev->mtd.name); -+ _close_bdev(dev); -+ kfree(dev); -+} -+ -+ -+static int block2mtd_refresh(struct mtd_info *mtd) -+{ -+ struct block2mtd_dev *dev = mtd->priv; -+ struct block_device *bdev; -+ dev_t devt; -+ int err = 0; -+ -+ /* no other mtd function can run at this point */ -+ write_lock(&dev->bdev_mutex); -+ -+ /* get the device number for the whole disk */ -+ devt = MKDEV(MAJOR(dev->blkdev->bd_dev), 0); -+ -+ /* close the old block device */ -+ _close_bdev(dev); -+ -+ /* open the whole disk, issue a partition rescan, then */ -+ bdev = open_by_devnum(devt, FMODE_WRITE | FMODE_READ); -+ if (!bdev || !bdev->bd_disk) -+ err = -EINVAL; -+ else { -+ err = rescan_partitions(bdev->bd_disk, bdev); -+ } -+ if (bdev) -+ close_bdev_excl(bdev); -+ -+ /* try to open the partition block device again */ -+ _open_bdev(dev); -+ write_unlock(&dev->bdev_mutex); -+ -+ return err; -+} -+ -+/* FIXME: ensure that mtd->size % erase_size == 0 */ -+static struct block2mtd_dev *add_device(char *devname, int erase_size, char *mtdname) -+{ -+ struct block2mtd_dev *dev; -+ struct mtd_partition *part; -+ -+ if (!devname) -+ return NULL; -+ -+ dev = kzalloc(sizeof(struct block2mtd_dev) + strlen(devname) + 1, GFP_KERNEL); -+ if (!dev) -+ return NULL; -+ -+ strcpy(dev->devname, devname); -+ -+ if (_open_bdev(dev)) -+ goto devinit_err; -+ - mutex_init(&dev->write_mutex); -+ rwlock_init(&dev->bdev_mutex); - - /* Setup the MTD structure */ - /* make the name contain the block device in */ -@@ -304,6 +397,7 @@ - dev->mtd.read = block2mtd_read; - dev->mtd.priv = dev; - dev->mtd.owner = THIS_MODULE; -+ dev->mtd.refresh_device = block2mtd_refresh; - - part = kzalloc(sizeof(struct mtd_partition), GFP_KERNEL); - part->name = dev->mtd.name; ---- a/drivers/mtd/mtdchar.c -+++ b/drivers/mtd/mtdchar.c -@@ -16,6 +16,7 @@ - - #include <linux/mtd/mtd.h> - #include <linux/mtd/compatmac.h> -+#include <linux/mtd/partitions.h> - - #include <asm/uaccess.h> - -@@ -752,6 +753,13 @@ - file->f_pos = 0; - break; - } -+#ifdef CONFIG_MTD_PARTITIONS -+ case MTDREFRESH: -+ { -+ ret = refresh_mtd_partitions(mtd); -+ break; -+ } -+#endif - - default: - ret = -ENOTTY; ---- a/include/linux/mtd/mtd.h -+++ b/include/linux/mtd/mtd.h -@@ -98,6 +98,7 @@ - uint8_t *oobbuf; - }; - -+struct mtd_info; - struct mtd_info { - u_char type; - u_int32_t flags; -@@ -195,6 +196,9 @@ - struct module *owner; - int usecount; - -+ int (*refresh_device)(struct mtd_info *mtd); -+ struct mtd_info *split; -+ - /* If the driver is something smart, like UBI, it may need to maintain - * its own reference counting. The below functions are only for driver. - * The driver may register its callbacks. These callbacks are not ---- a/include/linux/mtd/partitions.h -+++ b/include/linux/mtd/partitions.h -@@ -36,6 +36,7 @@ - * erasesize aligned (e.g. use MTDPART_OFS_NEXTBLK). - */ - -+struct mtd_partition; - struct mtd_partition { - char *name; /* identifier string */ - u_int32_t size; /* partition size */ -@@ -43,6 +44,7 @@ - u_int32_t mask_flags; /* master MTD flags to mask out for this partition */ - struct nand_ecclayout *ecclayout; /* out of band layout for this partition (NAND only)*/ - struct mtd_info **mtdp; /* pointer to store the MTD object */ -+ int (*refresh_partition)(struct mtd_info *); - }; - - #define MTDPART_OFS_NXTBLK (-2) -@@ -52,6 +54,7 @@ - - int add_mtd_partitions(struct mtd_info *, const struct mtd_partition *, int); - int del_mtd_partitions(struct mtd_info *); -+int refresh_mtd_partitions(struct mtd_info *); - - /* - * Functions dealing with the various ways of partitioning the space ---- a/include/mtd/mtd-abi.h -+++ b/include/mtd/mtd-abi.h -@@ -95,6 +95,7 @@ - #define ECCGETLAYOUT _IOR('M', 17, struct nand_ecclayout) - #define ECCGETSTATS _IOR('M', 18, struct mtd_ecc_stats) - #define MTDFILEMODE _IO('M', 19) -+#define MTDREFRESH _IO('M', 23) - - /* - * Obsolete legacy interface. Keep it in order not to break userspace diff --git a/target/linux/generic-2.6/patches-2.6.22/100-netfilter_layer7_2.17.patch b/target/linux/generic-2.6/patches-2.6.22/100-netfilter_layer7_2.17.patch deleted file mode 100644 index 6600d12..0000000 --- a/target/linux/generic-2.6/patches-2.6.22/100-netfilter_layer7_2.17.patch +++ /dev/null @@ -1,2101 +0,0 @@ ---- a/net/netfilter/Kconfig -+++ b/net/netfilter/Kconfig -@@ -603,6 +603,27 @@ - - To compile it as a module, choose M here. If unsure, say N. - -+config NETFILTER_XT_MATCH_LAYER7 -+ tristate '"layer7" match support' -+ depends on NETFILTER_XTABLES -+ depends on EXPERIMENTAL && (IP_NF_CONNTRACK || NF_CONNTRACK) -+ depends on NF_CT_ACCT -+ help -+ Say Y if you want to be able to classify connections (and their -+ packets) based on regular expression matching of their application -+ layer data. This is one way to classify applications such as -+ peer-to-peer filesharing systems that do not always use the same -+ port. -+ -+ To compile it as a module, choose M here. If unsure, say N. -+ -+config NETFILTER_XT_MATCH_LAYER7_DEBUG -+ bool 'Layer 7 debugging output' -+ depends on NETFILTER_XT_MATCH_LAYER7 -+ help -+ Say Y to get lots of debugging output. -+ -+ - config NETFILTER_XT_MATCH_STATISTIC - tristate '"statistic" match support' - depends on NETFILTER_XTABLES ---- a/net/netfilter/Makefile -+++ b/net/netfilter/Makefile -@@ -68,6 +68,7 @@ - obj-$(CONFIG_NETFILTER_XT_MATCH_REALM) += xt_realm.o - obj-$(CONFIG_NETFILTER_XT_MATCH_SCTP) += xt_sctp.o - obj-$(CONFIG_NETFILTER_XT_MATCH_STATE) += xt_state.o -+obj-$(CONFIG_NETFILTER_XT_MATCH_LAYER7) += xt_layer7.o - obj-$(CONFIG_NETFILTER_XT_MATCH_STATISTIC) += xt_statistic.o - obj-$(CONFIG_NETFILTER_XT_MATCH_STRING) += xt_string.o - obj-$(CONFIG_NETFILTER_XT_MATCH_TCPMSS) += xt_tcpmss.o ---- /dev/null -+++ b/net/netfilter/xt_layer7.c -@@ -0,0 +1,634 @@ -+/* -+ Kernel module to match application layer (OSI layer 7) data in connections. -+ -+ http://l7-filter.sf.net -+ -+ (C) 2003, 2004, 2005, 2006, 2007 Matthew Strait and Ethan Sommer. -+ -+ This program is free software; you can redistribute it and/or -+ modify it under the terms of the GNU General Public License -+ as published by the Free Software Foundation; either version -+ 2 of the License, or (at your option) any later version. -+ http://www.gnu.org/licenses/gpl.txt -+ -+ Based on ipt_string.c (C) 2000 Emmanuel Roger <winfield@freegates.be>, -+ xt_helper.c (C) 2002 Harald Welte and cls_layer7.c (C) 2003 Matthew Strait, -+ Ethan Sommer, Justin Levandoski. -+*/ -+ -+#include <linux/spinlock.h> -+#include <linux/version.h> -+#include <net/ip.h> -+#include <net/tcp.h> -+#include <linux/module.h> -+#include <linux/skbuff.h> -+#include <linux/netfilter.h> -+#include <net/netfilter/nf_conntrack.h> -+#include <net/netfilter/nf_conntrack_core.h> -+#include <linux/netfilter/x_tables.h> -+#include <linux/netfilter/xt_layer7.h> -+#include <linux/ctype.h> -+#include <linux/proc_fs.h> -+ -+#include "regexp/regexp.c" -+ -+MODULE_LICENSE("GPL"); -+MODULE_AUTHOR("Matthew Strait <quadong@users.sf.net>, Ethan Sommer <sommere@users.sf.net>"); -+MODULE_DESCRIPTION("iptables application layer match module"); -+MODULE_ALIAS("ipt_layer7"); -+MODULE_VERSION("2.17"); -+ -+static int maxdatalen = 2048; // this is the default -+module_param(maxdatalen, int, 0444); -+MODULE_PARM_DESC(maxdatalen, "maximum bytes of data looked at by l7-filter"); -+#ifdef CONFIG_NETFILTER_XT_MATCH_LAYER7_DEBUG -+ #define DPRINTK(format,args...) printk(format,##args) -+#else -+ #define DPRINTK(format,args...) -+#endif -+ -+#define TOTAL_PACKETS master_conntrack->counters[IP_CT_DIR_ORIGINAL].packets + \ -+ master_conntrack->counters[IP_CT_DIR_REPLY].packets -+ -+/* Number of packets whose data we look at. -+This can be modified through /proc/net/layer7_numpackets */ -+static int num_packets = 10; -+ -+static struct pattern_cache { -+ char * regex_string; -+ regexp * pattern; -+ struct pattern_cache * next; -+} * first_pattern_cache = NULL; -+ -+DEFINE_SPINLOCK(l7_lock); -+ -+#ifdef CONFIG_IP_NF_MATCH_LAYER7_DEBUG -+/* Converts an unfriendly string into a friendly one by -+replacing unprintables with periods and all whitespace with " ". */ -+static char * friendly_print(unsigned char * s) -+{ -+ char * f = kmalloc(strlen(s) + 1, GFP_ATOMIC); -+ int i; -+ -+ if(!f) { -+ if (net_ratelimit()) -+ printk(KERN_ERR "layer7: out of memory in " -+ "friendly_print, bailing.\n"); -+ return NULL; -+ } -+ -+ for(i = 0; i < strlen(s); i++){ -+ if(isprint(s[i]) && s[i] < 128) f[i] = s[i]; -+ else if(isspace(s[i])) f[i] = ' '; -+ else f[i] = '.'; -+ } -+ f[i] = '\0'; -+ return f; -+} -+ -+static char dec2hex(int i) -+{ -+ switch (i) { -+ case 0 ... 9: -+ return (i + '0'); -+ break; -+ case 10 ... 15: -+ return (i - 10 + 'a'); -+ break; -+ default: -+ if (net_ratelimit()) -+ printk("layer7: Problem in dec2hex\n"); -+ return '\0'; -+ } -+} -+ -+static char * hex_print(unsigned char * s) -+{ -+ char * g = kmalloc(strlen(s)*3 + 1, GFP_ATOMIC); -+ int i; -+ -+ if(!g) { -+ if (net_ratelimit()) -+ printk(KERN_ERR "layer7: out of memory in hex_print, " -+ "bailing.\n"); -+ return NULL; -+ } -+ -+ for(i = 0; i < strlen(s); i++) { -+ g[i*3 ] = dec2hex(s[i]/16); -+ g[i*3 + 1] = dec2hex(s[i]%16); -+ g[i*3 + 2] = ' '; -+ } -+ g[i*3] = '\0'; -+ -+ return g; -+} -+#endif // DEBUG -+ -+/* Use instead of regcomp. As we expect to be seeing the same regexps over and -+over again, it make sense to cache the results. */ -+static regexp * compile_and_cache(const char * regex_string, -+ const char * protocol) -+{ -+ struct pattern_cache * node = first_pattern_cache; -+ struct pattern_cache * last_pattern_cache = first_pattern_cache; -+ struct pattern_cache * tmp; -+ unsigned int len; -+ -+ while (node != NULL) { -+ if (!strcmp(node->regex_string, regex_string)) -+ return node->pattern; -+ -+ last_pattern_cache = node;/* points at the last non-NULL node */ -+ node = node->next; -+ } -+ -+ /* If we reach the end of the list, then we have not yet cached -+ the pattern for this regex. Let's do that now. -+ Be paranoid about running out of memory to avoid list corruption. */ -+ tmp = kmalloc(sizeof(struct pattern_cache), GFP_ATOMIC); -+ -+ if(!tmp) { -+ if (net_ratelimit()) -+ printk(KERN_ERR "layer7: out of memory in " -+ "compile_and_cache, bailing.\n"); -+ return NULL; -+ } -+ -+ tmp->regex_string = kmalloc(strlen(regex_string) + 1, GFP_ATOMIC); -+ tmp->pattern = kmalloc(sizeof(struct regexp), GFP_ATOMIC); -+ tmp->next = NULL; -+ -+ if(!tmp->regex_string || !tmp->pattern) { -+ if (net_ratelimit()) -+ printk(KERN_ERR "layer7: out of memory in " -+ "compile_and_cache, bailing.\n"); -+ kfree(tmp->regex_string); -+ kfree(tmp->pattern); -+ kfree(tmp); -+ return NULL; -+ } -+ -+ /* Ok. The new node is all ready now. */ -+ node = tmp; -+ -+ if(first_pattern_cache == NULL) /* list is empty */ -+ first_pattern_cache = node; /* make node the beginning */ -+ else -+ last_pattern_cache->next = node; /* attach node to the end */ -+ -+ /* copy the string and compile the regex */ -+ len = strlen(regex_string); -+ DPRINTK("About to compile this: \"%s\"\n", regex_string); -+ node->pattern = regcomp((char *)regex_string, &len); -+ if ( !node->pattern ) { -+ if (net_ratelimit()) -+ printk(KERN_ERR "layer7: Error compiling regexp " -+ "\"%s\" (%s)\n", -+ regex_string, protocol); -+ /* pattern is now cached as NULL, so we won't try again. */ -+ } -+ -+ strcpy(node->regex_string, regex_string); -+ return node->pattern; -+} -+ -+static int can_handle(const struct sk_buff *skb) -+{ -+ if(!ip_hdr(skb)) /* not IP */ -+ return 0; -+ if(ip_hdr(skb)->protocol != IPPROTO_TCP && -+ ip_hdr(skb)->protocol != IPPROTO_UDP && -+ ip_hdr(skb)->protocol != IPPROTO_ICMP) -+ return 0; -+ return 1; -+} -+ -+/* Returns offset the into the skb->data that the application data starts */ -+static int app_data_offset(const struct sk_buff *skb) -+{ -+ /* In case we are ported somewhere (ebtables?) where ip_hdr(skb) -+ isn't set, this can be gotten from 4*(skb->data[0] & 0x0f) as well. */ -+ int ip_hl = 4*ip_hdr(skb)->ihl; -+ -+ if( ip_hdr(skb)->protocol == IPPROTO_TCP ) { -+ /* 12 == offset into TCP header for the header length field. -+ Can't get this with skb->h.th->doff because the tcphdr -+ struct doesn't get set when routing (this is confirmed to be -+ true in Netfilter as well as QoS.) */ -+ int tcp_hl = 4*(skb->data[ip_hl + 12] >> 4); -+ -+ return ip_hl + tcp_hl; -+ } else if( ip_hdr(skb)->protocol == IPPROTO_UDP ) { -+ return ip_hl + 8; /* UDP header is always 8 bytes */ -+ } else if( ip_hdr(skb)->protocol == IPPROTO_ICMP ) { -+ return ip_hl + 8; /* ICMP header is 8 bytes */ -+ } else { -+ if (net_ratelimit()) -+ printk(KERN_ERR "layer7: tried to handle unknown " -+ "protocol!\n"); -+ return ip_hl + 8; /* something reasonable */ -+ } -+} -+ -+/* handles whether there's a match when we aren't appending data anymore */ -+static int match_no_append(struct nf_conn * conntrack, -+ struct nf_conn * master_conntrack, -+ enum ip_conntrack_info ctinfo, -+ enum ip_conntrack_info master_ctinfo, -+ const struct xt_layer7_info * info) -+{ -+ /* If we're in here, throw the app data away */ -+ if(master_conntrack->layer7.app_data != NULL) { -+ -+ #ifdef CONFIG_IP_NF_MATCH_LAYER7_DEBUG -+ if(!master_conntrack->layer7.app_proto) { -+ char * f = -+ friendly_print(master_conntrack->layer7.app_data); -+ char * g = -+ hex_print(master_conntrack->layer7.app_data); -+ DPRINTK("\nl7-filter gave up after %d bytes " -+ "(%d packets):\n%s\n", -+ strlen(f), TOTAL_PACKETS, f); -+ kfree(f); -+ DPRINTK("In hex: %s\n", g); -+ kfree(g); -+ } -+ #endif -+ -+ kfree(master_conntrack->layer7.app_data); -+ master_conntrack->layer7.app_data = NULL; /* don't free again */ -+ } -+ -+ if(master_conntrack->layer7.app_proto){ -+ /* Here child connections set their .app_proto (for /proc) */ -+ if(!conntrack->layer7.app_proto) { -+ conntrack->layer7.app_proto = -+ kmalloc(strlen(master_conntrack->layer7.app_proto)+1, -+ GFP_ATOMIC); -+ if(!conntrack->layer7.app_proto){ -+ if (net_ratelimit()) -+ printk(KERN_ERR "layer7: out of memory " -+ "in match_no_append, " -+ "bailing.\n"); -+ return 1; -+ } -+ strcpy(conntrack->layer7.app_proto, -+ master_conntrack->layer7.app_proto); -+ } -+ -+ return (!strcmp(master_conntrack->layer7.app_proto, -+ info->protocol)); -+ } -+ else { -+ /* If not classified, set to "unknown" to distinguish from -+ connections that are still being tested. */ -+ master_conntrack->layer7.app_proto = -+ kmalloc(strlen("unknown")+1, GFP_ATOMIC); -+ if(!master_conntrack->layer7.app_proto){ -+ if (net_ratelimit()) -+ printk(KERN_ERR "layer7: out of memory in " -+ "match_no_append, bailing.\n"); -+ return 1; -+ } -+ strcpy(master_conntrack->layer7.app_proto, "unknown"); -+ return 0; -+ } -+} -+ -+/* add the new app data to the conntrack. Return number of bytes added. */ -+static int add_data(struct nf_conn * master_conntrack, -+ char * app_data, int appdatalen) -+{ -+ int length = 0, i; -+ int oldlength = master_conntrack->layer7.app_data_len; -+ -+ /* This is a fix for a race condition by Deti Fliegl. However, I'm not -+ clear on whether the race condition exists or whether this really -+ fixes it. I might just be being dense... Anyway, if it's not really -+ a fix, all it does is waste a very small amount of time. */ -+ if(!master_conntrack->layer7.app_data) return 0; -+ -+ /* Strip nulls. Make everything lower case (our regex lib doesn't -+ do case insensitivity). Add it to the end of the current data. */ -+ for(i = 0; i < maxdatalen-oldlength-1 && -+ i < appdatalen; i++) { -+ if(app_data[i] != '\0') { -+ /* the kernel version of tolower mungs 'upper ascii' */ -+ master_conntrack->layer7.app_data[length+oldlength] = -+ isascii(app_data[i])? -+ tolower(app_data[i]) : app_data[i]; -+ length++; -+ } -+ } -+ -+ master_conntrack->layer7.app_data[length+oldlength] = '\0'; -+ master_conntrack->layer7.app_data_len = length + oldlength; -+ -+ return length; -+} -+ -+/* taken from drivers/video/modedb.c */ -+static int my_atoi(const char *s) -+{ -+ int val = 0; -+ -+ for (;; s++) { -+ switch (*s) { -+ case '0'...'9': -+ val = 10*val+(*s-'0'); -+ break; -+ default: -+ return val; -+ } -+ } -+} -+ -+/* write out num_packets to userland. */ -+static int layer7_read_proc(char* page, char ** start, off_t off, int count, -+ int* eof, void * data) -+{ -+ if(num_packets > 99 && net_ratelimit()) -+ printk(KERN_ERR "layer7: NOT REACHED. num_packets too big\n"); -+ -+ page[0] = num_packets/10 + '0'; -+ page[1] = num_packets%10 + '0'; -+ page[2] = '\n'; -+ page[3] = '\0'; -+ -+ *eof=1; -+ -+ return 3; -+} -+ -+/* Read in num_packets from userland */ -+static int layer7_write_proc(struct file* file, const char* buffer, -+ unsigned long count, void *data) -+{ -+ char * foo = kmalloc(count, GFP_ATOMIC); -+ -+ if(!foo){ -+ if (net_ratelimit()) -+ printk(KERN_ERR "layer7: out of memory, bailing. " -+ "num_packets unchanged.\n"); -+ return count; -+ } -+ -+ if(copy_from_user(foo, buffer, count)) { -+ return -EFAULT; -+ } -+ -+ -+ num_packets = my_atoi(foo); -+ kfree (foo); -+ -+ /* This has an arbitrary limit to make the math easier. I'm lazy. -+ But anyway, 99 is a LOT! If you want more, you're doing it wrong! */ -+ if(num_packets > 99) { -+ printk(KERN_WARNING "layer7: num_packets can't be > 99.\n"); -+ num_packets = 99; -+ } else if(num_packets < 1) { -+ printk(KERN_WARNING "layer7: num_packets can't be < 1.\n"); -+ num_packets = 1; -+ } -+ -+ return count; -+} -+ -+static int -+match(const struct sk_buff *skbin, -+ const struct net_device *in, -+ const struct net_device *out, -+ const struct xt_match *match, -+ const void *matchinfo, -+ int offset, -+ unsigned int protoff, -+ int *hotdrop) -+{ -+ /* sidestep const without getting a compiler warning... */ -+ struct sk_buff * skb = (struct sk_buff *)skbin; -+ -+ const struct xt_layer7_info * info = matchinfo; -+ enum ip_conntrack_info master_ctinfo, ctinfo; -+ struct nf_conn *master_conntrack, *conntrack; -+ unsigned char * app_data; -+ unsigned int pattern_result, appdatalen; -+ regexp * comppattern; -+ -+ /* Be paranoid/incompetent - lock the entire match function. */ -+ spin_lock_bh(&l7_lock); -+ -+ if(!can_handle(skb)){ -+ DPRINTK("layer7: This is some protocol I can't handle.\n"); -+ spin_unlock_bh(&l7_lock); -+ return info->invert; -+ } -+ -+ /* Treat parent & all its children together as one connection, except -+ for the purpose of setting conntrack->layer7.app_proto in the actual -+ connection. This makes /proc/net/ip_conntrack more satisfying. */ -+ if(!(conntrack = nf_ct_get(skb, &ctinfo)) || -+ !(master_conntrack=nf_ct_get(skb,&master_ctinfo))){ -+ DPRINTK("layer7: couldn't get conntrack.\n"); -+ spin_unlock_bh(&l7_lock); -+ return info->invert; -+ } -+ -+ /* Try to get a master conntrack (and its master etc) for FTP, etc. */ -+ while (master_ct(master_conntrack) != NULL) -+ master_conntrack = master_ct(master_conntrack); -+ -+ /* if we've classified it or seen too many packets */ -+ if(TOTAL_PACKETS > num_packets || -+ master_conntrack->layer7.app_proto) { -+ -+ pattern_result = match_no_append(conntrack, master_conntrack, -+ ctinfo, master_ctinfo, info); -+ -+ /* skb->cb[0] == seen. Don't do things twice if there are -+ multiple l7 rules. I'm not sure that using cb for this purpose -+ is correct, even though it says "put your private variables -+ there". But it doesn't look like it is being used for anything -+ else in the skbs that make it here. */ -+ skb->cb[0] = 1; /* marking it seen here's probably irrelevant */ -+ -+ spin_unlock_bh(&l7_lock); -+ return (pattern_result ^ info->invert); -+ } -+ -+ if(skb_is_nonlinear(skb)){ -+ if(skb_linearize(skb) != 0){ -+ if (net_ratelimit()) -+ printk(KERN_ERR "layer7: failed to linearize " -+ "packet, bailing.\n"); -+ spin_unlock_bh(&l7_lock); -+ return info->invert; -+ } -+ } -+ -+ /* now that the skb is linearized, it's safe to set these. */ -+ app_data = skb->data + app_data_offset(skb); -+ appdatalen = skb_tail_pointer(skb) - app_data; -+ -+ /* the return value gets checked later, when we're ready to use it */ -+ comppattern = compile_and_cache(info->pattern, info->protocol); -+ -+ /* On the first packet of a connection, allocate space for app data */ -+ if(TOTAL_PACKETS == 1 && !skb->cb[0] && -+ !master_conntrack->layer7.app_data){ -+ master_conntrack->layer7.app_data = -+ kmalloc(maxdatalen, GFP_ATOMIC); -+ if(!master_conntrack->layer7.app_data){ -+ if (net_ratelimit()) -+ printk(KERN_ERR "layer7: out of memory in " -+ "match, bailing.\n"); -+ spin_unlock_bh(&l7_lock); -+ return info->invert; -+ } -+ -+ master_conntrack->layer7.app_data[0] = '\0'; -+ } -+ -+ /* Can be here, but unallocated, if numpackets is increased near -+ the beginning of a connection */ -+ if(master_conntrack->layer7.app_data == NULL){ -+ spin_unlock_bh(&l7_lock); -+ return (info->invert); /* unmatched */ -+ } -+ -+ if(!skb->cb[0]){ -+ int newbytes; -+ newbytes = add_data(master_conntrack, app_data, appdatalen); -+ -+ if(newbytes == 0) { /* didn't add any data */ -+ skb->cb[0] = 1; -+ /* Didn't match before, not going to match now */ -+ spin_unlock_bh(&l7_lock); -+ return info->invert; -+ } -+ } -+ -+ /* If looking for "unknown", then never match. "Unknown" means that -+ we've given up; we're still trying with these packets. */ -+ if(!strcmp(info->protocol, "unknown")) { -+ pattern_result = 0; -+ /* If looking for "unset", then always match. "Unset" means that we -+ haven't yet classified the connection. */ -+ } else if(!strcmp(info->protocol, "unset")) { -+ pattern_result = 2; -+ DPRINTK("layer7: matched unset: not yet classified " -+ "(%d/%d packets)\n", TOTAL_PACKETS, num_packets); -+ /* If the regexp failed to compile, don't bother running it */ -+ } else if(comppattern && -+ regexec(comppattern, master_conntrack->layer7.app_data)){ -+ DPRINTK("layer7: matched %s\n", info->protocol); -+ pattern_result = 1; -+ } else pattern_result = 0; -+ -+ if(pattern_result == 1) { -+ master_conntrack->layer7.app_proto = -+ kmalloc(strlen(info->protocol)+1, GFP_ATOMIC); -+ if(!master_conntrack->layer7.app_proto){ -+ if (net_ratelimit()) -+ printk(KERN_ERR "layer7: out of memory in " -+ "match, bailing.\n"); -+ spin_unlock_bh(&l7_lock); -+ return (pattern_result ^ info->invert); -+ } -+ strcpy(master_conntrack->layer7.app_proto, info->protocol); -+ } else if(pattern_result > 1) { /* cleanup from "unset" */ -+ pattern_result = 1; -+ } -+ -+ /* mark the packet seen */ -+ skb->cb[0] = 1; -+ -+ spin_unlock_bh(&l7_lock); -+ return (pattern_result ^ info->invert); -+} -+ -+static int check(const char *tablename, -+ const void *inf, -+ const struct xt_match *match, -+ void *matchinfo, -+ unsigned int hook_mask) -+ -+{ -+ // load nf_conntrack_ipv4 -+ if (nf_ct_l3proto_try_module_get(match->family) < 0) { -+ printk(KERN_WARNING "can't load conntrack support for " -+ "proto=%d\n", match->family); -+ return false; -+ } -+ return true; -+} -+ -+static void -+destroy(const struct xt_match *match, void *matchinfo) -+{ -+ nf_ct_l3proto_module_put(match->family); -+} -+ -+static struct xt_match xt_layer7_match[] = { -+{ -+ .name = "layer7", -+ .family = AF_INET, -+ .checkentry = check, -+ .match = match, -+ .destroy = destroy, -+ .matchsize = sizeof(struct xt_layer7_info), -+ .me = THIS_MODULE -+} -+}; -+ -+static void layer7_cleanup_proc(void) -+{ -+#if LINUX_VERSION_CODE <= KERNEL_VERSION(2,6,23) -+ remove_proc_entry("layer7_numpackets", proc_net); -+#else -+ remove_proc_entry("layer7_numpackets", init_net.proc_net); -+#endif -+} -+ -+/* register the proc file */ -+static void layer7_init_proc(void) -+{ -+ struct proc_dir_entry* entry; -+#if LINUX_VERSION_CODE <= KERNEL_VERSION(2,6,23) -+ entry = create_proc_entry("layer7_numpackets", 0644, proc_net); -+#else -+ entry = create_proc_entry("layer7_numpackets", 0644, init_net.proc_net); -+#endif -+ entry->read_proc = layer7_read_proc; -+ entry->write_proc = layer7_write_proc; -+} -+ -+static int __init xt_layer7_init(void) -+{ -+ need_conntrack(); -+ -+ layer7_init_proc(); -+ if(maxdatalen < 1) { -+ printk(KERN_WARNING "layer7: maxdatalen can't be < 1, " -+ "using 1\n"); -+ maxdatalen = 1; -+ } -+ /* This is not a hard limit. It's just here to prevent people from -+ bringing their slow machines to a grinding halt. */ -+ else if(maxdatalen > 65536) { -+ printk(KERN_WARNING "layer7: maxdatalen can't be > 65536, " -+ "using 65536\n"); -+ maxdatalen = 65536; -+ } -+ return xt_register_matches(xt_layer7_match, -+ ARRAY_SIZE(xt_layer7_match)); -+} -+ -+static void __exit xt_layer7_fini(void) -+{ -+ layer7_cleanup_proc(); -+ xt_unregister_matches(xt_layer7_match, ARRAY_SIZE(xt_layer7_match)); -+} -+ -+module_init(xt_layer7_init); -+module_exit(xt_layer7_fini); ---- /dev/null -+++ b/net/netfilter/regexp/regexp.c -@@ -0,0 +1,1197 @@ -+/* -+ * regcomp and regexec -- regsub and regerror are elsewhere -+ * @(#)regexp.c 1.3 of 18 April 87 -+ * -+ * Copyright (c) 1986 by University of Toronto. -+ * Written by Henry Spencer. Not derived from licensed software. -+ * -+ * Permission is granted to anyone to use this software for any -+ * purpose on any computer system, and to redistribute it freely, -+ * subject to the following restrictions: -+ * -+ * 1. The author is not responsible for the consequences of use of -+ * this software, no matter how awful, even if they arise -+ * from defects in it. -+ * -+ * 2. The origin of this software must not be misrepresented, either -+ * by explicit claim or by omission. -+ * -+ * 3. Altered versions must be plainly marked as such, and must not -+ * be misrepresented as being the original software. -+ * -+ * Beware that some of this code is subtly aware of the way operator -+ * precedence is structured in regular expressions. Serious changes in -+ * regular-expression syntax might require a total rethink. -+ * -+ * This code was modified by Ethan Sommer to work within the kernel -+ * (it now uses kmalloc etc..) -+ * -+ * Modified slightly by Matthew Strait to use more modern C. -+ */ -+ -+#include "regexp.h" -+#include "regmagic.h" -+ -+/* added by ethan and matt. Lets it work in both kernel and user space. -+(So iptables can use it, for instance.) Yea, it goes both ways... */ -+#if __KERNEL__ -+ #define malloc(foo) kmalloc(foo,GFP_ATOMIC) -+#else -+ #define printk(format,args...) printf(format,##args) -+#endif -+ -+void regerror(char * s) -+{ -+ printk("<3>Regexp: %s\n", s); -+ /* NOTREACHED */ -+} -+ -+/* -+ * The "internal use only" fields in regexp.h are present to pass info from -+ * compile to execute that permits the execute phase to run lots faster on -+ * simple cases. They are: -+ * -+ * regstart char that must begin a match; '\0' if none obvious -+ * reganch is the match anchored (at beginning-of-line only)? -+ * regmust string (pointer into program) that match must include, or NULL -+ * regmlen length of regmust string -+ * -+ * Regstart and reganch permit very fast decisions on suitable starting points -+ * for a match, cutting down the work a lot. Regmust permits fast rejection -+ * of lines that cannot possibly match. The regmust tests are costly enough -+ * that regcomp() supplies a regmust only if the r.e. contains something -+ * potentially expensive (at present, the only such thing detected is * or + -+ * at the start of the r.e., which can involve a lot of backup). Regmlen is -+ * supplied because the test in regexec() needs it and regcomp() is computing -+ * it anyway. -+ */ -+ -+/* -+ * Structure for regexp "program". This is essentially a linear encoding -+ * of a nondeterministic finite-state machine (aka syntax charts or -+ * "railroad normal form" in parsing technology). Each node is an opcode -+ * plus a "next" pointer, possibly plus an operand. "Next" pointers of -+ * all nodes except BRANCH implement concatenation; a "next" pointer with -+ * a BRANCH on both ends of it is connecting two alternatives. (Here we -+ * have one of the subtle syntax dependencies: an individual BRANCH (as -+ * opposed to a collection of them) is never concatenated with anything -+ * because of operator precedence.) The operand of some types of node is -+ * a literal string; for others, it is a node leading into a sub-FSM. In -+ * particular, the operand of a BRANCH node is the first node of the branch. -+ * (NB this is *not* a tree structure: the tail of the branch connects -+ * to the thing following the set of BRANCHes.) The opcodes are: -+ */ -+ -+/* definition number opnd? meaning */ -+#define END 0 /* no End of program. */ -+#define BOL 1 /* no Match "" at beginning of line. */ -+#define EOL 2 /* no Match "" at end of line. */ -+#define ANY 3 /* no Match any one character. */ -+#define ANYOF 4 /* str Match any character in this string. */ -+#define ANYBUT 5 /* str Match any character not in this string. */ -+#define BRANCH 6 /* node Match this alternative, or the next... */ -+#define BACK 7 /* no Match "", "next" ptr points backward. */ -+#define EXACTLY 8 /* str Match this string. */ -+#define NOTHING 9 /* no Match empty string. */ -+#define STAR 10 /* node Match this (simple) thing 0 or more times. */ -+#define PLUS 11 /* node Match this (simple) thing 1 or more times. */ -+#define OPEN 20 /* no Mark this point in input as start of #n. */ -+ /* OPEN+1 is number 1, etc. */ -+#define CLOSE 30 /* no Analogous to OPEN. */ -+ -+/* -+ * Opcode notes: -+ * -+ * BRANCH The set of branches constituting a single choice are hooked -+ * together with their "next" pointers, since precedence prevents -+ * anything being concatenated to any individual branch. The -+ * "next" pointer of the last BRANCH in a choice points to the -+ * thing following the whole choice. This is also where the -+ * final "next" pointer of each individual branch points; each -+ * branch starts with the operand node of a BRANCH node. -+ * -+ * BACK Normal "next" pointers all implicitly point forward; BACK -+ * exists to make loop structures possible. -+ * -+ * STAR,PLUS '?', and complex '*' and '+', are implemented as circular -+ * BRANCH structures using BACK. Simple cases (one character -+ * per match) are implemented with STAR and PLUS for speed -+ * and to minimize recursive plunges. -+ * -+ * OPEN,CLOSE ...are numbered at compile time. -+ */ -+ -+/* -+ * A node is one char of opcode followed by two chars of "next" pointer. -+ * "Next" pointers are stored as two 8-bit pieces, high order first. The -+ * value is a positive offset from the opcode of the node containing it. -+ * An operand, if any, simply follows the node. (Note that much of the -+ * code generation knows about this implicit relationship.) -+ * -+ * Using two bytes for the "next" pointer is vast overkill for most things, -+ * but allows patterns to get big without disasters. -+ */ -+#define OP(p) (*(p)) -+#define NEXT(p) (((*((p)+1)&0377)<<8) + (*((p)+2)&0377)) -+#define OPERAND(p) ((p) + 3) -+ -+/* -+ * See regmagic.h for one further detail of program structure. -+ */ -+ -+ -+/* -+ * Utility definitions. -+ */ -+#ifndef CHARBITS -+#define UCHARAT(p) ((int)*(unsigned char *)(p)) -+#else -+#define UCHARAT(p) ((int)*(p)&CHARBITS) -+#endif -+ -+#define FAIL(m) { regerror(m); return(NULL); } -+#define ISMULT(c) ((c) == '*' || (c) == '+' || (c) == '?') -+#define META "^$.[()|?+*\\" -+ -+/* -+ * Flags to be passed up and down. -+ */ -+#define HASWIDTH 01 /* Known never to match null string. */ -+#define SIMPLE 02 /* Simple enough to be STAR/PLUS operand. */ -+#define SPSTART 04 /* Starts with * or +. */ -+#define WORST 0 /* Worst case. */ -+ -+/* -+ * Global work variables for regcomp(). -+ */ -+struct match_globals { -+char *reginput; /* String-input pointer. */ -+char *regbol; /* Beginning of input, for ^ check. */ -+char **regstartp; /* Pointer to startp array. */ -+char **regendp; /* Ditto for endp. */ -+char *regparse; /* Input-scan pointer. */ -+int regnpar; /* () count. */ -+char regdummy; -+char *regcode; /* Code-emit pointer; ®dummy = don't. */ -+long regsize; /* Code size. */ -+}; -+ -+/* -+ * Forward declarations for regcomp()'s friends. -+ */ -+#ifndef STATIC -+#define STATIC static -+#endif -+STATIC char *reg(struct match_globals *g, int paren,int *flagp); -+STATIC char *regbranch(struct match_globals *g, int *flagp); -+STATIC char *regpiece(struct match_globals *g, int *flagp); -+STATIC char *regatom(struct match_globals *g, int *flagp); -+STATIC char *regnode(struct match_globals *g, char op); -+STATIC char *regnext(struct match_globals *g, char *p); -+STATIC void regc(struct match_globals *g, char b); -+STATIC void reginsert(struct match_globals *g, char op, char *opnd); -+STATIC void regtail(struct match_globals *g, char *p, char *val); -+STATIC void regoptail(struct match_globals *g, char *p, char *val); -+ -+ -+__kernel_size_t my_strcspn(const char *s1,const char *s2) -+{ -+ char *scan1; -+ char *scan2; -+ int count; -+ -+ count = 0; -+ for (scan1 = (char *)s1; *scan1 != '\0'; scan1++) { -+ for (scan2 = (char *)s2; *scan2 != '\0';) /* ++ moved down. */ -+ if (*scan1 == *scan2++) -+ return(count); -+ count++; -+ } -+ return(count); -+} -+ -+/* -+ - regcomp - compile a regular expression into internal code -+ * -+ * We can't allocate space until we know how big the compiled form will be, -+ * but we can't compile it (and thus know how big it is) until we've got a -+ * place to put the code. So we cheat: we compile it twice, once with code -+ * generation turned off and size counting turned on, and once "for real". -+ * This also means that we don't allocate space until we are sure that the -+ * thing really will compile successfully, and we never have to move the -+ * code and thus invalidate pointers into it. (Note that it has to be in -+ * one piece because free() must be able to free it all.) -+ * -+ * Beware that the optimization-preparation code in here knows about some -+ * of the structure of the compiled regexp. -+ */ -+regexp * -+regcomp(char *exp,int *patternsize) -+{ -+ register regexp *r; -+ register char *scan; -+ register char *longest; -+ register int len; -+ int flags; -+ struct match_globals g; -+ -+ /* commented out by ethan -+ extern char *malloc(); -+ */ -+ -+ if (exp == NULL) -+ FAIL("NULL argument"); -+ -+ /* First pass: determine size, legality. */ -+ g.regparse = exp; -+ g.regnpar = 1; -+ g.regsize = 0L; -+ g.regcode = &g.regdummy; -+ regc(&g, MAGIC); -+ if (reg(&g, 0, &flags) == NULL) -+ return(NULL); -+ -+ /* Small enough for pointer-storage convention? */ -+ if (g.regsize >= 32767L) /* Probably could be 65535L. */ -+ FAIL("regexp too big"); -+ -+ /* Allocate space. */ -+ *patternsize=sizeof(regexp) + (unsigned)g.regsize; -+ r = (regexp *)malloc(sizeof(regexp) + (unsigned)g.regsize); -+ if (r == NULL) -+ FAIL("out of space"); -+ -+ /* Second pass: emit code. */ -+ g.regparse = exp; -+ g.regnpar = 1; -+ g.regcode = r->program; -+ regc(&g, MAGIC); -+ if (reg(&g, 0, &flags) == NULL) -+ return(NULL); -+ -+ /* Dig out information for optimizations. */ -+ r->regstart = '\0'; /* Worst-case defaults. */ -+ r->reganch = 0; -+ r->regmust = NULL; -+ r->regmlen = 0; -+ scan = r->program+1; /* First BRANCH. */ -+ if (OP(regnext(&g, scan)) == END) { /* Only one top-level choice. */ -+ scan = OPERAND(scan); -+ -+ /* Starting-point info. */ -+ if (OP(scan) == EXACTLY) -+ r->regstart = *OPERAND(scan); -+ else if (OP(scan) == BOL) -+ r->reganch++; -+ -+ /* -+ * If there's something expensive in the r.e., find the -+ * longest literal string that must appear and make it the -+ * regmust. Resolve ties in favor of later strings, since -+ * the regstart check works with the beginning of the r.e. -+ * and avoiding duplication strengthens checking. Not a -+ * strong reason, but sufficient in the absence of others. -+ */ -+ if (flags&SPSTART) { -+ longest = NULL; -+ len = 0; -+ for (; scan != NULL; scan = regnext(&g, scan)) -+ if (OP(scan) == EXACTLY && strlen(OPERAND(scan)) >= len) { -+ longest = OPERAND(scan); -+ len = strlen(OPERAND(scan)); -+ } -+ r->regmust = longest; -+ r->regmlen = len; -+ } -+ } -+ -+ return(r); -+} -+ -+/* -+ - reg - regular expression, i.e. main body or parenthesized thing -+ * -+ * Caller must absorb opening parenthesis. -+ * -+ * Combining parenthesis handling with the base level of regular expression -+ * is a trifle forced, but the need to tie the tails of the branches to what -+ * follows makes it hard to avoid. -+ */ -+static char * -+reg(struct match_globals *g, int paren, int *flagp /* Parenthesized? */ ) -+{ -+ register char *ret; -+ register char *br; -+ register char *ender; -+ register int parno = 0; /* 0 makes gcc happy */ -+ int flags; -+ -+ *flagp = HASWIDTH; /* Tentatively. */ -+ -+ /* Make an OPEN node, if parenthesized. */ -+ if (paren) { -+ if (g->regnpar >= NSUBEXP) -+ FAIL("too many ()"); -+ parno = g->regnpar; -+ g->regnpar++; -+ ret = regnode(g, OPEN+parno); -+ } else -+ ret = NULL; -+ -+ /* Pick up the branches, linking them together. */ -+ br = regbranch(g, &flags); -+ if (br == NULL) -+ return(NULL); -+ if (ret != NULL) -+ regtail(g, ret, br); /* OPEN -> first. */ -+ else -+ ret = br; -+ if (!(flags&HASWIDTH)) -+ *flagp &= ~HASWIDTH; -+ *flagp |= flags&SPSTART; -+ while (*g->regparse == '|') { -+ g->regparse++; -+ br = regbranch(g, &flags); -+ if (br == NULL) -+ return(NULL); -+ regtail(g, ret, br); /* BRANCH -> BRANCH. */ -+ if (!(flags&HASWIDTH)) -+ *flagp &= ~HASWIDTH; -+ *flagp |= flags&SPSTART; -+ } -+ -+ /* Make a closing node, and hook it on the end. */ -+ ender = regnode(g, (paren) ? CLOSE+parno : END); -+ regtail(g, ret, ender); -+ -+ /* Hook the tails of the branches to the closing node. */ -+ for (br = ret; br != NULL; br = regnext(g, br)) -+ regoptail(g, br, ender); -+ -+ /* Check for proper termination. */ -+ if (paren && *g->regparse++ != ')') { -+ FAIL("unmatched ()"); -+ } else if (!paren && *g->regparse != '\0') { -+ if (*g->regparse == ')') { -+ FAIL("unmatched ()"); -+ } else -+ FAIL("junk on end"); /* "Can't happen". */ -+ /* NOTREACHED */ -+ } -+ -+ return(ret); -+} -+ -+/* -+ - regbranch - one alternative of an | operator -+ * -+ * Implements the concatenation operator. -+ */ -+static char * -+regbranch(struct match_globals *g, int *flagp) -+{ -+ register char *ret; -+ register char *chain; -+ register char *latest; -+ int flags; -+ -+ *flagp = WORST; /* Tentatively. */ -+ -+ ret = regnode(g, BRANCH); -+ chain = NULL; -+ while (*g->regparse != '\0' && *g->regparse != '|' && *g->regparse != ')') { -+ latest = regpiece(g, &flags); -+ if (latest == NULL) -+ return(NULL); -+ *flagp |= flags&HASWIDTH; -+ if (chain == NULL) /* First piece. */ -+ *flagp |= flags&SPSTART; -+ else -+ regtail(g, chain, latest); -+ chain = latest; -+ } -+ if (chain == NULL) /* Loop ran zero times. */ -+ (void) regnode(g, NOTHING); -+ -+ return(ret); -+} -+ -+/* -+ - regpiece - something followed by possible [*+?] -+ * -+ * Note that the branching code sequences used for ? and the general cases -+ * of * and + are somewhat optimized: they use the same NOTHING node as -+ * both the endmarker for their branch list and the body of the last branch. -+ * It might seem that this node could be dispensed with entirely, but the -+ * endmarker role is not redundant. -+ */ -+static char * -+regpiece(struct match_globals *g, int *flagp) -+{ -+ register char *ret; -+ register char op; -+ register char *next; -+ int flags; -+ -+ ret = regatom(g, &flags); -+ if (ret == NULL) -+ return(NULL); -+ -+ op = *g->regparse; -+ if (!ISMULT(op)) { -+ *flagp = flags; -+ return(ret); -+ } -+ -+ if (!(flags&HASWIDTH) && op != '?') -+ FAIL("*+ operand could be empty"); -+ *flagp = (op != '+') ? (WORST|SPSTART) : (WORST|HASWIDTH); -+ -+ if (op == '*' && (flags&SIMPLE)) -+ reginsert(g, STAR, ret); -+ else if (op == '*') { -+ /* Emit x* as (x&|), where & means "self". */ -+ reginsert(g, BRANCH, ret); /* Either x */ -+ regoptail(g, ret, regnode(g, BACK)); /* and loop */ -+ regoptail(g, ret, ret); /* back */ -+ regtail(g, ret, regnode(g, BRANCH)); /* or */ -+ regtail(g, ret, regnode(g, NOTHING)); /* null. */ -+ } else if (op == '+' && (flags&SIMPLE)) -+ reginsert(g, PLUS, ret); -+ else if (op == '+') { -+ /* Emit x+ as x(&|), where & means "self". */ -+ next = regnode(g, BRANCH); /* Either */ -+ regtail(g, ret, next); -+ regtail(g, regnode(g, BACK), ret); /* loop back */ -+ regtail(g, next, regnode(g, BRANCH)); /* or */ -+ regtail(g, ret, regnode(g, NOTHING)); /* null. */ -+ } else if (op == '?') { -+ /* Emit x? as (x|) */ -+ reginsert(g, BRANCH, ret); /* Either x */ -+ regtail(g, ret, regnode(g, BRANCH)); /* or */ -+ next = regnode(g, NOTHING); /* null. */ -+ regtail(g, ret, next); -+ regoptail(g, ret, next); -+ } -+ g->regparse++; -+ if (ISMULT(*g->regparse)) -+ FAIL("nested *?+"); -+ -+ return(ret); -+} -+ -+/* -+ - regatom - the lowest level -+ * -+ * Optimization: gobbles an entire sequence of ordinary characters so that -+ * it can turn them into a single node, which is smaller to store and -+ * faster to run. Backslashed characters are exceptions, each becoming a -+ * separate node; the code is simpler that way and it's not worth fixing. -+ */ -+static char * -+regatom(struct match_globals *g, int *flagp) -+{ -+ register char *ret; -+ int flags; -+ -+ *flagp = WORST; /* Tentatively. */ -+ -+ switch (*g->regparse++) { -+ case '^': -+ ret = regnode(g, BOL); -+ break; -+ case '$': -+ ret = regnode(g, EOL); -+ break; -+ case '.': -+ ret = regnode(g, ANY); -+ *flagp |= HASWIDTH|SIMPLE; -+ break; -+ case '[': { -+ register int class; -+ register int classend; -+ -+ if (*g->regparse == '^') { /* Complement of range. */ -+ ret = regnode(g, ANYBUT); -+ g->regparse++; -+ } else -+ ret = regnode(g, ANYOF); -+ if (*g->regparse == ']' || *g->regparse == '-') -+ regc(g, *g->regparse++); -+ while (*g->regparse != '\0' && *g->regparse != ']') { -+ if (*g->regparse == '-') { -+ g->regparse++; -+ if (*g->regparse == ']' || *g->regparse == '\0') -+ regc(g, '-'); -+ else { -+ class = UCHARAT(g->regparse-2)+1; -+ classend = UCHARAT(g->regparse); -+ if (class > classend+1) -+ FAIL("invalid [] range"); -+ for (; class <= classend; class++) -+ regc(g, class); -+ g->regparse++; -+ } -+ } else -+ regc(g, *g->regparse++); -+ } -+ regc(g, '\0'); -+ if (*g->regparse != ']') -+ FAIL("unmatched []"); -+ g->regparse++; -+ *flagp |= HASWIDTH|SIMPLE; -+ } -+ break; -+ case '(': -+ ret = reg(g, 1, &flags); -+ if (ret == NULL) -+ return(NULL); -+ *flagp |= flags&(HASWIDTH|SPSTART); -+ break; -+ case '\0': -+ case '|': -+ case ')': -+ FAIL("internal urp"); /* Supposed to be caught earlier. */ -+ break; -+ case '?': -+ case '+': -+ case '*': -+ FAIL("?+* follows nothing"); -+ break; -+ case '\\': -+ if (*g->regparse == '\0') -+ FAIL("trailing \\"); -+ ret = regnode(g, EXACTLY); -+ regc(g, *g->regparse++); -+ regc(g, '\0'); -+ *flagp |= HASWIDTH|SIMPLE; -+ break; -+ default: { -+ register int len; -+ register char ender; -+ -+ g->regparse--; -+ len = my_strcspn((const char *)g->regparse, (const char *)META); -+ if (len <= 0) -+ FAIL("internal disaster"); -+ ender = *(g->regparse+len); -+ if (len > 1 && ISMULT(ender)) -+ len--; /* Back off clear of ?+* operand. */ -+ *flagp |= HASWIDTH; -+ if (len == 1) -+ *flagp |= SIMPLE; -+ ret = regnode(g, EXACTLY); -+ while (len > 0) { -+ regc(g, *g->regparse++); -+ len--; -+ } -+ regc(g, '\0'); -+ } -+ break; -+ } -+ -+ return(ret); -+} -+ -+/* -+ - regnode - emit a node -+ */ -+static char * /* Location. */ -+regnode(struct match_globals *g, char op) -+{ -+ register char *ret; -+ register char *ptr; -+ -+ ret = g->regcode; -+ if (ret == &g->regdummy) { -+ g->regsize += 3; -+ return(ret); -+ } -+ -+ ptr = ret; -+ *ptr++ = op; -+ *ptr++ = '\0'; /* Null "next" pointer. */ -+ *ptr++ = '\0'; -+ g->regcode = ptr; -+ -+ return(ret); -+} -+ -+/* -+ - regc - emit (if appropriate) a byte of code -+ */ -+static void -+regc(struct match_globals *g, char b) -+{ -+ if (g->regcode != &g->regdummy) -+ *g->regcode++ = b; -+ else -+ g->regsize++; -+} -+ -+/* -+ - reginsert - insert an operator in front of already-emitted operand -+ * -+ * Means relocating the operand. -+ */ -+static void -+reginsert(struct match_globals *g, char op, char* opnd) -+{ -+ register char *src; -+ register char *dst; -+ register char *place; -+ -+ if (g->regcode == &g->regdummy) { -+ g->regsize += 3; -+ return; -+ } -+ -+ src = g->regcode; -+ g->regcode += 3; -+ dst = g->regcode; -+ while (src > opnd) -+ *--dst = *--src; -+ -+ place = opnd; /* Op node, where operand used to be. */ -+ *place++ = op; -+ *place++ = '\0'; -+ *place++ = '\0'; -+} -+ -+/* -+ - regtail - set the next-pointer at the end of a node chain -+ */ -+static void -+regtail(struct match_globals *g, char *p, char *val) -+{ -+ register char *scan; -+ register char *temp; -+ register int offset; -+ -+ if (p == &g->regdummy) -+ return; -+ -+ /* Find last node. */ -+ scan = p; -+ for (;;) { -+ temp = regnext(g, scan); -+ if (temp == NULL) -+ break; -+ scan = temp; -+ } -+ -+ if (OP(scan) == BACK) -+ offset = scan - val; -+ else -+ offset = val - scan; -+ *(scan+1) = (offset>>8)&0377; -+ *(scan+2) = offset&0377; -+} -+ -+/* -+ - regoptail - regtail on operand of first argument; nop if operandless -+ */ -+static void -+regoptail(struct match_globals *g, char *p, char *val) -+{ -+ /* "Operandless" and "op != BRANCH" are synonymous in practice. */ -+ if (p == NULL || p == &g->regdummy || OP(p) != BRANCH) -+ return; -+ regtail(g, OPERAND(p), val); -+} -+ -+/* -+ * regexec and friends -+ */ -+ -+ -+/* -+ * Forwards. -+ */ -+STATIC int regtry(struct match_globals *g, regexp *prog, char *string); -+STATIC int regmatch(struct match_globals *g, char *prog); -+STATIC int regrepeat(struct match_globals *g, char *p); -+ -+#ifdef DEBUG -+int regnarrate = 0; -+void regdump(); -+STATIC char *regprop(char *op); -+#endif -+ -+/* -+ - regexec - match a regexp against a string -+ */ -+int -+regexec(regexp *prog, char *string) -+{ -+ register char *s; -+ struct match_globals g; -+ -+ /* Be paranoid... */ -+ if (prog == NULL || string == NULL) { -+ printk("<3>Regexp: NULL parameter\n"); -+ return(0); -+ } -+ -+ /* Check validity of program. */ -+ if (UCHARAT(prog->program) != MAGIC) { -+ printk("<3>Regexp: corrupted program\n"); -+ return(0); -+ } -+ -+ /* If there is a "must appear" string, look for it. */ -+ if (prog->regmust != NULL) { -+ s = string; -+ while ((s = strchr(s, prog->regmust[0])) != NULL) { -+ if (strncmp(s, prog->regmust, prog->regmlen) == 0) -+ break; /* Found it. */ -+ s++; -+ } -+ if (s == NULL) /* Not present. */ -+ return(0); -+ } -+ -+ /* Mark beginning of line for ^ . */ -+ g.regbol = string; -+ -+ /* Simplest case: anchored match need be tried only once. */ -+ if (prog->reganch) -+ return(regtry(&g, prog, string)); -+ -+ /* Messy cases: unanchored match. */ -+ s = string; -+ if (prog->regstart != '\0') -+ /* We know what char it must start with. */ -+ while ((s = strchr(s, prog->regstart)) != NULL) { -+ if (regtry(&g, prog, s)) -+ return(1); -+ s++; -+ } -+ else -+ /* We don't -- general case. */ -+ do { -+ if (regtry(&g, prog, s)) -+ return(1); -+ } while (*s++ != '\0'); -+ -+ /* Failure. */ -+ return(0); -+} -+ -+/* -+ - regtry - try match at specific point -+ */ -+static int /* 0 failure, 1 success */ -+regtry(struct match_globals *g, regexp *prog, char *string) -+{ -+ register int i; -+ register char **sp; -+ register char **ep; -+ -+ g->reginput = string; -+ g->regstartp = prog->startp; -+ g->regendp = prog->endp; -+ -+ sp = prog->startp; -+ ep = prog->endp; -+ for (i = NSUBEXP; i > 0; i--) { -+ *sp++ = NULL; -+ *ep++ = NULL; -+ } -+ if (regmatch(g, prog->program + 1)) { -+ prog->startp[0] = string; -+ prog->endp[0] = g->reginput; -+ return(1); -+ } else -+ return(0); -+} -+ -+/* -+ - regmatch - main matching routine -+ * -+ * Conceptually the strategy is simple: check to see whether the current -+ * node matches, call self recursively to see whether the rest matches, -+ * and then act accordingly. In practice we make some effort to avoid -+ * recursion, in particular by going through "ordinary" nodes (that don't -+ * need to know whether the rest of the match failed) by a loop instead of -+ * by recursion. -+ */ -+static int /* 0 failure, 1 success */ -+regmatch(struct match_globals *g, char *prog) -+{ -+ register char *scan = prog; /* Current node. */ -+ char *next; /* Next node. */ -+ -+#ifdef DEBUG -+ if (scan != NULL && regnarrate) -+ fprintf(stderr, "%s(\n", regprop(scan)); -+#endif -+ while (scan != NULL) { -+#ifdef DEBUG -+ if (regnarrate) -+ fprintf(stderr, "%s...\n", regprop(scan)); -+#endif -+ next = regnext(g, scan); -+ -+ switch (OP(scan)) { -+ case BOL: -+ if (g->reginput != g->regbol) -+ return(0); -+ break; -+ case EOL: -+ if (*g->reginput != '\0') -+ return(0); -+ break; -+ case ANY: -+ if (*g->reginput == '\0') -+ return(0); -+ g->reginput++; -+ break; -+ case EXACTLY: { -+ register int len; -+ register char *opnd; -+ -+ opnd = OPERAND(scan); -+ /* Inline the first character, for speed. */ -+ if (*opnd != *g->reginput) -+ return(0); -+ len = strlen(opnd); -+ if (len > 1 && strncmp(opnd, g->reginput, len) != 0) -+ return(0); -+ g->reginput += len; -+ } -+ break; -+ case ANYOF: -+ if (*g->reginput == '\0' || strchr(OPERAND(scan), *g->reginput) == NULL) -+ return(0); -+ g->reginput++; -+ break; -+ case ANYBUT: -+ if (*g->reginput == '\0' || strchr(OPERAND(scan), *g->reginput) != NULL) -+ return(0); -+ g->reginput++; -+ break; -+ case NOTHING: -+ case BACK: -+ break; -+ case OPEN+1: -+ case OPEN+2: -+ case OPEN+3: -+ case OPEN+4: -+ case OPEN+5: -+ case OPEN+6: -+ case OPEN+7: -+ case OPEN+8: -+ case OPEN+9: { -+ register int no; -+ register char *save; -+ -+ no = OP(scan) - OPEN; -+ save = g->reginput; -+ -+ if (regmatch(g, next)) { -+ /* -+ * Don't set startp if some later -+ * invocation of the same parentheses -+ * already has. -+ */ -+ if (g->regstartp[no] == NULL) -+ g->regstartp[no] = save; -+ return(1); -+ } else -+ return(0); -+ } -+ break; -+ case CLOSE+1: -+ case CLOSE+2: -+ case CLOSE+3: -+ case CLOSE+4: -+ case CLOSE+5: -+ case CLOSE+6: -+ case CLOSE+7: -+ case CLOSE+8: -+ case CLOSE+9: -+ { -+ register int no; -+ register char *save; -+ -+ no = OP(scan) - CLOSE; -+ save = g->reginput; -+ -+ if (regmatch(g, next)) { -+ /* -+ * Don't set endp if some later -+ * invocation of the same parentheses -+ * already has. -+ */ -+ if (g->regendp[no] == NULL) -+ g->regendp[no] = save; -+ return(1); -+ } else -+ return(0); -+ } -+ break; -+ case BRANCH: { -+ register char *save; -+ -+ if (OP(next) != BRANCH) /* No choice. */ -+ next = OPERAND(scan); /* Avoid recursion. */ -+ else { -+ do { -+ save = g->reginput; -+ if (regmatch(g, OPERAND(scan))) -+ return(1); -+ g->reginput = save; -+ scan = regnext(g, scan); -+ } while (scan != NULL && OP(scan) == BRANCH); -+ return(0); -+ /* NOTREACHED */ -+ } -+ } -+ break; -+ case STAR: -+ case PLUS: { -+ register char nextch; -+ register int no; -+ register char *save; -+ register int min; -+ -+ /* -+ * Lookahead to avoid useless match attempts -+ * when we know what character comes next. -+ */ -+ nextch = '\0'; -+ if (OP(next) == EXACTLY) -+ nextch = *OPERAND(next); -+ min = (OP(scan) == STAR) ? 0 : 1; -+ save = g->reginput; -+ no = regrepeat(g, OPERAND(scan)); -+ while (no >= min) { -+ /* If it could work, try it. */ -+ if (nextch == '\0' || *g->reginput == nextch) -+ if (regmatch(g, next)) -+ return(1); -+ /* Couldn't or didn't -- back up. */ -+ no--; -+ g->reginput = save + no; -+ } -+ return(0); -+ } -+ break; -+ case END: -+ return(1); /* Success! */ -+ break; -+ default: -+ printk("<3>Regexp: memory corruption\n"); -+ return(0); -+ break; -+ } -+ -+ scan = next; -+ } -+ -+ /* -+ * We get here only if there's trouble -- normally "case END" is -+ * the terminating point. -+ */ -+ printk("<3>Regexp: corrupted pointers\n"); -+ return(0); -+} -+ -+/* -+ - regrepeat - repeatedly match something simple, report how many -+ */ -+static int -+regrepeat(struct match_globals *g, char *p) -+{ -+ register int count = 0; -+ register char *scan; -+ register char *opnd; -+ -+ scan = g->reginput; -+ opnd = OPERAND(p); -+ switch (OP(p)) { -+ case ANY: -+ count = strlen(scan); -+ scan += count; -+ break; -+ case EXACTLY: -+ while (*opnd == *scan) { -+ count++; -+ scan++; -+ } -+ break; -+ case ANYOF: -+ while (*scan != '\0' && strchr(opnd, *scan) != NULL) { -+ count++; -+ scan++; -+ } -+ break; -+ case ANYBUT: -+ while (*scan != '\0' && strchr(opnd, *scan) == NULL) { -+ count++; -+ scan++; -+ } -+ break; -+ default: /* Oh dear. Called inappropriately. */ -+ printk("<3>Regexp: internal foulup\n"); -+ count = 0; /* Best compromise. */ -+ break; -+ } -+ g->reginput = scan; -+ -+ return(count); -+} -+ -+/* -+ - regnext - dig the "next" pointer out of a node -+ */ -+static char* -+regnext(struct match_globals *g, char *p) -+{ -+ register int offset; -+ -+ if (p == &g->regdummy) -+ return(NULL); -+ -+ offset = NEXT(p); -+ if (offset == 0) -+ return(NULL); -+ -+ if (OP(p) == BACK) -+ return(p-offset); -+ else -+ return(p+offset); -+} -+ -+#ifdef DEBUG -+ -+STATIC char *regprop(); -+ -+/* -+ - regdump - dump a regexp onto stdout in vaguely comprehensible form -+ */ -+void -+regdump(regexp *r) -+{ -+ register char *s; -+ register char op = EXACTLY; /* Arbitrary non-END op. */ -+ register char *next; -+ /* extern char *strchr(); */ -+ -+ -+ s = r->program + 1; -+ while (op != END) { /* While that wasn't END last time... */ -+ op = OP(s); -+ printf("%2d%s", s-r->program, regprop(s)); /* Where, what. */ -+ next = regnext(s); -+ if (next == NULL) /* Next ptr. */ -+ printf("(0)"); -+ else -+ printf("(%d)", (s-r->program)+(next-s)); -+ s += 3; -+ if (op == ANYOF || op == ANYBUT || op == EXACTLY) { -+ /* Literal string, where present. */ -+ while (*s != '\0') { -+ putchar(*s); -+ s++; -+ } -+ s++; -+ } -+ putchar('\n'); -+ } -+ -+ /* Header fields of interest. */ -+ if (r->regstart != '\0') -+ printf("start `%c' ", r->regstart); -+ if (r->reganch) -+ printf("anchored "); -+ if (r->regmust != NULL) -+ printf("must have \"%s\"", r->regmust); -+ printf("\n"); -+} -+ -+/* -+ - regprop - printable representation of opcode -+ */ -+static char * -+regprop(char *op) -+{ -+#define BUFLEN 50 -+ register char *p; -+ static char buf[BUFLEN]; -+ -+ strcpy(buf, ":"); -+ -+ switch (OP(op)) { -+ case BOL: -+ p = "BOL"; -+ break; -+ case EOL: -+ p = "EOL"; -+ break; -+ case ANY: -+ p = "ANY"; -+ break; -+ case ANYOF: -+ p = "ANYOF"; -+ break; -+ case ANYBUT: -+ p = "ANYBUT"; -+ break; -+ case BRANCH: -+ p = "BRANCH"; -+ break; -+ case EXACTLY: -+ p = "EXACTLY"; -+ break; -+ case NOTHING: -+ p = "NOTHING"; -+ break; -+ case BACK: -+ p = "BACK"; -+ break; -+ case END: -+ p = "END"; -+ break; -+ case OPEN+1: -+ case OPEN+2: -+ case OPEN+3: -+ case OPEN+4: -+ case OPEN+5: -+ case OPEN+6: -+ case OPEN+7: -+ case OPEN+8: -+ case OPEN+9: -+ snprintf(buf+strlen(buf),BUFLEN-strlen(buf), "OPEN%d", OP(op)-OPEN); -+ p = NULL; -+ break; -+ case CLOSE+1: -+ case CLOSE+2: -+ case CLOSE+3: -+ case CLOSE+4: -+ case CLOSE+5: -+ case CLOSE+6: -+ case CLOSE+7: -+ case CLOSE+8: -+ case CLOSE+9: -+ snprintf(buf+strlen(buf),BUFLEN-strlen(buf), "CLOSE%d", OP(op)-CLOSE); -+ p = NULL; -+ break; -+ case STAR: -+ p = "STAR"; -+ break; -+ case PLUS: -+ p = "PLUS"; -+ break; -+ default: -+ printk("<3>Regexp: corrupted opcode\n"); -+ break; -+ } -+ if (p != NULL) -+ strncat(buf, p, BUFLEN-strlen(buf)); -+ return(buf); -+} -+#endif -+ -+ ---- /dev/null -+++ b/net/netfilter/regexp/regexp.h -@@ -0,0 +1,41 @@ -+/* -+ * Definitions etc. for regexp(3) routines. -+ * -+ * Caveat: this is V8 regexp(3) [actually, a reimplementation thereof], -+ * not the System V one. -+ */ -+ -+#ifndef REGEXP_H -+#define REGEXP_H -+ -+ -+/* -+http://www.opensource.apple.com/darwinsource/10.3/expect-1/expect/expect.h , -+which contains a version of this library, says: -+ -+ * -+ * NSUBEXP must be at least 10, and no greater than 117 or the parser -+ * will not work properly. -+ * -+ -+However, it looks rather like this library is limited to 10. If you think -+otherwise, let us know. -+*/ -+ -+#define NSUBEXP 10 -+typedef struct regexp { -+ char *startp[NSUBEXP]; -+ char *endp[NSUBEXP]; -+ char regstart; /* Internal use only. */ -+ char reganch; /* Internal use only. */ -+ char *regmust; /* Internal use only. */ -+ int regmlen; /* Internal use only. */ -+ char program[1]; /* Unwarranted chumminess with compiler. */ -+} regexp; -+ -+regexp * regcomp(char *exp, int *patternsize); -+int regexec(regexp *prog, char *string); -+void regsub(regexp *prog, char *source, char *dest); -+void regerror(char *s); -+ -+#endif ---- /dev/null -+++ b/net/netfilter/regexp/regmagic.h -@@ -0,0 +1,5 @@ -+/* -+ * The first byte of the regexp internal "program" is actually this magic -+ * number; the start node begins in the second byte. -+ */ -+#define MAGIC 0234 ---- /dev/null -+++ b/net/netfilter/regexp/regsub.c -@@ -0,0 +1,95 @@ -+/* -+ * regsub -+ * @(#)regsub.c 1.3 of 2 April 86 -+ * -+ * Copyright (c) 1986 by University of Toronto. -+ * Written by Henry Spencer. Not derived from licensed software. -+ * -+ * Permission is granted to anyone to use this software for any -+ * purpose on any computer system, and to redistribute it freely, -+ * subject to the following restrictions: -+ * -+ * 1. The author is not responsible for the consequences of use of -+ * this software, no matter how awful, even if they arise -+ * from defects in it. -+ * -+ * 2. The origin of this software must not be misrepresented, either -+ * by explicit claim or by omission. -+ * -+ * 3. Altered versions must be plainly marked as such, and must not -+ * be misrepresented as being the original software. -+ * -+ * -+ * This code was modified by Ethan Sommer to work within the kernel -+ * (it now uses kmalloc etc..) -+ * -+ */ -+#include "regexp.h" -+#include "regmagic.h" -+#include <linux/string.h> -+ -+ -+#ifndef CHARBITS -+#define UCHARAT(p) ((int)*(unsigned char *)(p)) -+#else -+#define UCHARAT(p) ((int)*(p)&CHARBITS) -+#endif -+ -+#if 0 -+//void regerror(char * s) -+//{ -+// printk("regexp(3): %s", s); -+// /* NOTREACHED */ -+//} -+#endif -+ -+/* -+ - regsub - perform substitutions after a regexp match -+ */ -+void -+regsub(regexp * prog, char * source, char * dest) -+{ -+ register char *src; -+ register char *dst; -+ register char c; -+ register int no; -+ register int len; -+ -+ /* Not necessary and gcc doesn't like it -MLS */ -+ /*extern char *strncpy();*/ -+ -+ if (prog == NULL || source == NULL || dest == NULL) { -+ regerror("NULL parm to regsub"); -+ return; -+ } -+ if (UCHARAT(prog->program) != MAGIC) { -+ regerror("damaged regexp fed to regsub"); -+ return; -+ } -+ -+ src = source; -+ dst = dest; -+ while ((c = *src++) != '\0') { -+ if (c == '&') -+ no = 0; -+ else if (c == '\\' && '0' <= *src && *src <= '9') -+ no = *src++ - '0'; -+ else -+ no = -1; -+ -+ if (no < 0) { /* Ordinary character. */ -+ if (c == '\\' && (*src == '\\' || *src == '&')) -+ c = *src++; -+ *dst++ = c; -+ } else if (prog->startp[no] != NULL && prog->endp[no] != NULL) { -+ len = prog->endp[no] - prog->startp[no]; -+ (void) strncpy(dst, prog->startp[no], len); -+ dst += len; -+ if (len != 0 && *(dst-1) == '\0') { /* strncpy hit NUL. */ -+ regerror("damaged match string"); -+ return; -+ } -+ } -+ } -+ *dst++ = '\0'; -+} ---- a/net/netfilter/nf_conntrack_core.c -+++ b/net/netfilter/nf_conntrack_core.c -@@ -330,6 +330,14 @@ - * too. */ - nf_ct_remove_expectations(ct); - -+ #if defined(CONFIG_NETFILTER_XT_MATCH_LAYER7) || defined(CONFIG_NETFILTER_XT_MATCH_LAYER7_MODULE) -+ if(ct->layer7.app_proto) -+ kfree(ct->layer7.app_proto); -+ if(ct->layer7.app_data) -+ kfree(ct->layer7.app_data); -+ #endif -+ -+ - /* We overload first tuple to link into unconfirmed list. */ - if (!nf_ct_is_confirmed(ct)) { - BUG_ON(list_empty(&ct->tuplehash[IP_CT_DIR_ORIGINAL].list)); ---- a/net/netfilter/nf_conntrack_standalone.c -+++ b/net/netfilter/nf_conntrack_standalone.c -@@ -184,7 +184,12 @@ - return -ENOSPC; - #endif - -- if (seq_printf(s, "use=%u\n", atomic_read(&conntrack->ct_general.use))) -+#if defined(CONFIG_NETFILTER_XT_MATCH_LAYER7) || defined(CONFIG_NETFILTER_XT_MATCH_LAYER7_MODULE) -+ if(conntrack->layer7.app_proto) -+ if(seq_printf(s, "l7proto=%s ", conntrack->layer7.app_proto)) -+ return -ENOSPC; -+#endif -+ if (seq_printf(s, "asdfuse=%u\n", atomic_read(&conntrack->ct_general.use))) - return -ENOSPC; - - return 0; ---- a/include/net/netfilter/nf_conntrack.h -+++ b/include/net/netfilter/nf_conntrack.h -@@ -128,6 +128,22 @@ - u_int32_t secmark; - #endif - -+#if defined(CONFIG_NETFILTER_XT_MATCH_LAYER7) || \ -+ defined(CONFIG_NETFILTER_XT_MATCH_LAYER7_MODULE) -+ struct { -+ /* -+ * e.g. "http". NULL before decision. "unknown" after decision -+ * if no match. -+ */ -+ char *app_proto; -+ /* -+ * application layer data so far. NULL after match decision. -+ */ -+ char *app_data; -+ unsigned int app_data_len; -+ } layer7; -+#endif -+ - /* Storage reserved for other modules: */ - union nf_conntrack_proto proto; - ---- /dev/null -+++ b/include/linux/netfilter/xt_layer7.h -@@ -0,0 +1,13 @@ -+#ifndef _XT_LAYER7_H -+#define _XT_LAYER7_H -+ -+#define MAX_PATTERN_LEN 8192 -+#define MAX_PROTOCOL_LEN 256 -+ -+struct xt_layer7_info { -+ char protocol[MAX_PROTOCOL_LEN]; -+ char pattern[MAX_PATTERN_LEN]; -+ u_int8_t invert; -+}; -+ -+#endif /* _XT_LAYER7_H */ diff --git a/target/linux/generic-2.6/patches-2.6.22/101-netfilter_layer7_pktmatch.patch b/target/linux/generic-2.6/patches-2.6.22/101-netfilter_layer7_pktmatch.patch deleted file mode 100644 index 167d578..0000000 --- a/target/linux/generic-2.6/patches-2.6.22/101-netfilter_layer7_pktmatch.patch +++ /dev/null @@ -1,109 +0,0 @@ ---- a/include/linux/netfilter/xt_layer7.h -+++ b/include/linux/netfilter/xt_layer7.h -@@ -8,6 +8,7 @@ - char protocol[MAX_PROTOCOL_LEN]; - char pattern[MAX_PATTERN_LEN]; - u_int8_t invert; -+ u_int8_t pkt; - }; - - #endif /* _XT_LAYER7_H */ ---- a/net/netfilter/xt_layer7.c -+++ b/net/netfilter/xt_layer7.c -@@ -297,34 +297,36 @@ - } - - /* add the new app data to the conntrack. Return number of bytes added. */ --static int add_data(struct nf_conn * master_conntrack, -- char * app_data, int appdatalen) -+static int add_datastr(char *target, int offset, char *app_data, int len) - { - int length = 0, i; -- int oldlength = master_conntrack->layer7.app_data_len; -- -- /* This is a fix for a race condition by Deti Fliegl. However, I'm not -- clear on whether the race condition exists or whether this really -- fixes it. I might just be being dense... Anyway, if it's not really -- a fix, all it does is waste a very small amount of time. */ -- if(!master_conntrack->layer7.app_data) return 0; -+ -+ if (!target) return 0; - - /* Strip nulls. Make everything lower case (our regex lib doesn't - do case insensitivity). Add it to the end of the current data. */ -- for(i = 0; i < maxdatalen-oldlength-1 && -- i < appdatalen; i++) { -+ for(i = 0; i < maxdatalen-offset-1 && i < len; i++) { - if(app_data[i] != '\0') { - /* the kernel version of tolower mungs 'upper ascii' */ -- master_conntrack->layer7.app_data[length+oldlength] = -+ target[length+offset] = - isascii(app_data[i])? - tolower(app_data[i]) : app_data[i]; - length++; - } - } -+ target[length+offset] = '\0'; -+ -+ return length; -+} - -- master_conntrack->layer7.app_data[length+oldlength] = '\0'; -- master_conntrack->layer7.app_data_len = length + oldlength; -+/* add the new app data to the conntrack. Return number of bytes added. */ -+static int add_data(struct nf_conn * master_conntrack, -+ char * app_data, int appdatalen) -+{ -+ int length; - -+ length = add_datastr(master_conntrack->layer7.app_data, master_conntrack->layer7.app_data_len, app_data, appdatalen); -+ master_conntrack->layer7.app_data_len += length; - return length; - } - -@@ -411,7 +413,7 @@ - const struct xt_layer7_info * info = matchinfo; - enum ip_conntrack_info master_ctinfo, ctinfo; - struct nf_conn *master_conntrack, *conntrack; -- unsigned char * app_data; -+ unsigned char *app_data, *tmp_data; - unsigned int pattern_result, appdatalen; - regexp * comppattern; - -@@ -439,8 +441,8 @@ - master_conntrack = master_ct(master_conntrack); - - /* if we've classified it or seen too many packets */ -- if(TOTAL_PACKETS > num_packets || -- master_conntrack->layer7.app_proto) { -+ if(!info->pkt && (TOTAL_PACKETS > num_packets || -+ master_conntrack->layer7.app_proto)) { - - pattern_result = match_no_append(conntrack, master_conntrack, - ctinfo, master_ctinfo, info); -@@ -473,6 +475,25 @@ - /* the return value gets checked later, when we're ready to use it */ - comppattern = compile_and_cache(info->pattern, info->protocol); - -+ if (info->pkt) { -+ tmp_data = kmalloc(maxdatalen, GFP_ATOMIC); -+ if(!tmp_data){ -+ if (net_ratelimit()) -+ printk(KERN_ERR "layer7: out of memory in match, bailing.\n"); -+ return info->invert; -+ } -+ -+ tmp_data[0] = '\0'; -+ add_datastr(tmp_data, 0, app_data, appdatalen); -+ pattern_result = ((comppattern && regexec(comppattern, tmp_data)) ? 1 : 0); -+ -+ kfree(tmp_data); -+ tmp_data = NULL; -+ spin_unlock_bh(&l7_lock); -+ -+ return (pattern_result ^ info->invert); -+ } -+ - /* On the first packet of a connection, allocate space for app data */ - if(TOTAL_PACKETS == 1 && !skb->cb[0] && - !master_conntrack->layer7.app_data){ diff --git a/target/linux/generic-2.6/patches-2.6.22/110-ipp2p_0.8.1rc1.patch b/target/linux/generic-2.6/patches-2.6.22/110-ipp2p_0.8.1rc1.patch deleted file mode 100644 index bad2565..0000000 --- a/target/linux/generic-2.6/patches-2.6.22/110-ipp2p_0.8.1rc1.patch +++ /dev/null @@ -1,944 +0,0 @@ ---- /dev/null -+++ b/include/linux/netfilter_ipv4/ipt_ipp2p.h -@@ -0,0 +1,31 @@ -+#ifndef __IPT_IPP2P_H -+#define __IPT_IPP2P_H -+#define IPP2P_VERSION "0.8.1_rc1" -+ -+struct ipt_p2p_info { -+ int cmd; -+ int debug; -+}; -+ -+#endif //__IPT_IPP2P_H -+ -+#define SHORT_HAND_IPP2P 1 /* --ipp2p switch*/ -+//#define SHORT_HAND_DATA 4 /* --ipp2p-data switch*/ -+#define SHORT_HAND_NONE 5 /* no short hand*/ -+ -+#define IPP2P_EDK (1 << 1) -+#define IPP2P_DATA_KAZAA (1 << 2) -+#define IPP2P_DATA_EDK (1 << 3) -+#define IPP2P_DATA_DC (1 << 4) -+#define IPP2P_DC (1 << 5) -+#define IPP2P_DATA_GNU (1 << 6) -+#define IPP2P_GNU (1 << 7) -+#define IPP2P_KAZAA (1 << 8) -+#define IPP2P_BIT (1 << 9) -+#define IPP2P_APPLE (1 << 10) -+#define IPP2P_SOUL (1 << 11) -+#define IPP2P_WINMX (1 << 12) -+#define IPP2P_ARES (1 << 13) -+#define IPP2P_MUTE (1 << 14) -+#define IPP2P_WASTE (1 << 15) -+#define IPP2P_XDCC (1 << 16) ---- /dev/null -+++ b/net/ipv4/netfilter/ipt_ipp2p.c -@@ -0,0 +1,882 @@ -+#if defined(MODVERSIONS) -+#include <linux/modversions.h> -+#endif -+#include <linux/module.h> -+#include <linux/netfilter_ipv4/ip_tables.h> -+#include <linux/version.h> -+#include <linux/netfilter_ipv4/ipt_ipp2p.h> -+#include <net/tcp.h> -+#include <net/udp.h> -+ -+#define get_u8(X,O) (*(__u8 *)(X + O)) -+#define get_u16(X,O) (*(__u16 *)(X + O)) -+#define get_u32(X,O) (*(__u32 *)(X + O)) -+ -+MODULE_AUTHOR("Eicke Friedrich/Klaus Degner <ipp2p@ipp2p.org>"); -+MODULE_DESCRIPTION("An extension to iptables to identify P2P traffic."); -+MODULE_LICENSE("GPL"); -+ -+ -+/*Search for UDP eDonkey/eMule/Kad commands*/ -+int -+udp_search_edk (unsigned char *haystack, int packet_len) -+{ -+ unsigned char *t = haystack; -+ t += 8; -+ -+ switch (t[0]) { -+ case 0xe3: -+ { /*edonkey*/ -+ switch (t[1]) -+ { -+ /* client -> server status request */ -+ case 0x96: -+ if (packet_len == 14) return ((IPP2P_EDK * 100) + 50); -+ break; -+ /* server -> client status request */ -+ case 0x97: if (packet_len == 42) return ((IPP2P_EDK * 100) + 51); -+ break; -+ /* server description request */ -+ /* e3 2a ff f0 .. | size == 6 */ -+ case 0xa2: if ( (packet_len == 14) && ( get_u16(t,2) == __constant_htons(0xfff0) ) ) return ((IPP2P_EDK * 100) + 52); -+ break; -+ /* server description response */ -+ /* e3 a3 ff f0 .. | size > 40 && size < 200 */ -+ //case 0xa3: return ((IPP2P_EDK * 100) + 53); -+ // break; -+ case 0x9a: if (packet_len==26) return ((IPP2P_EDK * 100) + 54); -+ break; -+ -+ case 0x92: if (packet_len==18) return ((IPP2P_EDK * 100) + 55); -+ break; -+ } -+ break; -+ } -+ case 0xe4: -+ { -+ switch (t[1]) -+ { -+ /* e4 20 .. | size == 43 */ -+ case 0x20: if ((packet_len == 43) && (t[2] != 0x00) && (t[34] != 0x00)) return ((IPP2P_EDK * 100) + 60); -+ break; -+ /* e4 00 .. 00 | size == 35 ? */ -+ case 0x00: if ((packet_len == 35) && (t[26] == 0x00)) return ((IPP2P_EDK * 100) + 61); -+ break; -+ /* e4 10 .. 00 | size == 35 ? */ -+ case 0x10: if ((packet_len == 35) && (t[26] == 0x00)) return ((IPP2P_EDK * 100) + 62); -+ break; -+ /* e4 18 .. 00 | size == 35 ? */ -+ case 0x18: if ((packet_len == 35) && (t[26] == 0x00)) return ((IPP2P_EDK * 100) + 63); -+ break; -+ /* e4 52 .. | size = 44 */ -+ case 0x52: if (packet_len == 44 ) return ((IPP2P_EDK * 100) + 64); -+ break; -+ /* e4 58 .. | size == 6 */ -+ case 0x58: if (packet_len == 14 ) return ((IPP2P_EDK * 100) + 65); -+ break; -+ /* e4 59 .. | size == 2 */ -+ case 0x59: if (packet_len == 10 )return ((IPP2P_EDK * 100) + 66); -+ break; -+ /* e4 28 .. | packet_len == 52,77,102,127... */ -+ case 0x28: if (((packet_len-52) % 25) == 0) return ((IPP2P_EDK * 100) + 67); -+ break; -+ /* e4 50 xx xx | size == 4 */ -+ case 0x50: if (packet_len == 12) return ((IPP2P_EDK * 100) + 68); -+ break; -+ /* e4 40 xx xx | size == 48 */ -+ case 0x40: if (packet_len == 56) return ((IPP2P_EDK * 100) + 69); -+ break; -+ } -+ break; -+ } -+ } /* end of switch (t[0]) */ -+ return 0; -+}/*udp_search_edk*/ -+ -+ -+/*Search for UDP Gnutella commands*/ -+int -+udp_search_gnu (unsigned char *haystack, int packet_len) -+{ -+ unsigned char *t = haystack; -+ t += 8; -+ -+ if (memcmp(t, "GND", 3) == 0) return ((IPP2P_GNU * 100) + 51); -+ if (memcmp(t, "GNUTELLA ", 9) == 0) return ((IPP2P_GNU * 100) + 52); -+ return 0; -+}/*udp_search_gnu*/ -+ -+ -+/*Search for UDP KaZaA commands*/ -+int -+udp_search_kazaa (unsigned char *haystack, int packet_len) -+{ -+ unsigned char *t = haystack; -+ -+ if (t[packet_len-1] == 0x00){ -+ t += (packet_len - 6); -+ if (memcmp(t, "KaZaA", 5) == 0) return (IPP2P_KAZAA * 100 +50); -+ } -+ -+ return 0; -+}/*udp_search_kazaa*/ -+ -+/*Search for UDP DirectConnect commands*/ -+int -+udp_search_directconnect (unsigned char *haystack, int packet_len) -+{ -+ unsigned char *t = haystack; -+ if ((*(t + 8) == 0x24) && (*(t + packet_len - 1) == 0x7c)) { -+ t+=8; -+ if (memcmp(t, "SR ", 3) == 0) return ((IPP2P_DC * 100) + 60); -+ if (memcmp(t, "Ping ", 5) == 0) return ((IPP2P_DC * 100) + 61); -+ } -+ return 0; -+}/*udp_search_directconnect*/ -+ -+ -+ -+/*Search for UDP BitTorrent commands*/ -+int -+udp_search_bit (unsigned char *haystack, int packet_len) -+{ -+ switch(packet_len) -+ { -+ case 24: -+ /* ^ 00 00 04 17 27 10 19 80 */ -+ if ((ntohl(get_u32(haystack, 8)) == 0x00000417) && (ntohl(get_u32(haystack, 12)) == 0x27101980)) -+ return (IPP2P_BIT * 100 + 50); -+ break; -+ case 44: -+ if (get_u32(haystack, 16) == __constant_htonl(0x00000400) && get_u32(haystack, 36) == __constant_htonl(0x00000104)) -+ return (IPP2P_BIT * 100 + 51); -+ if (get_u32(haystack, 16) == __constant_htonl(0x00000400)) -+ return (IPP2P_BIT * 100 + 61); -+ break; -+ case 65: -+ if (get_u32(haystack, 16) == __constant_htonl(0x00000404) && get_u32(haystack, 36) == __constant_htonl(0x00000104)) -+ return (IPP2P_BIT * 100 + 52); -+ if (get_u32(haystack, 16) == __constant_htonl(0x00000404)) -+ return (IPP2P_BIT * 100 + 62); -+ break; -+ case 67: -+ if (get_u32(haystack, 16) == __constant_htonl(0x00000406) && get_u32(haystack, 36) == __constant_htonl(0x00000104)) -+ return (IPP2P_BIT * 100 + 53); -+ if (get_u32(haystack, 16) == __constant_htonl(0x00000406)) -+ return (IPP2P_BIT * 100 + 63); -+ break; -+ case 211: -+ if (get_u32(haystack, 8) == __constant_htonl(0x00000405)) -+ return (IPP2P_BIT * 100 + 54); -+ break; -+ case 29: -+ if ((get_u32(haystack, 8) == __constant_htonl(0x00000401))) -+ return (IPP2P_BIT * 100 + 55); -+ break; -+ case 52: -+ if (get_u32(haystack,8) == __constant_htonl(0x00000827) && -+ get_u32(haystack,12) == __constant_htonl(0x37502950)) -+ return (IPP2P_BIT * 100 + 80); -+ break; -+ default: -+ /* this packet does not have a constant size */ -+ if (packet_len >= 40 && get_u32(haystack, 16) == __constant_htonl(0x00000402) && get_u32(haystack, 36) == __constant_htonl(0x00000104)) -+ return (IPP2P_BIT * 100 + 56); -+ break; -+ } -+ -+ /* some extra-bitcomet rules: -+ * "d1:" [a|r] "d2:id20:" -+ */ -+ if (packet_len > 30 && get_u8(haystack, 8) == 'd' && get_u8(haystack, 9) == '1' && get_u8(haystack, 10) == ':' ) -+ { -+ if (get_u8(haystack, 11) == 'a' || get_u8(haystack, 11) == 'r') -+ { -+ if (memcmp(haystack+12,"d2:id20:",8)==0) -+ return (IPP2P_BIT * 100 + 57); -+ } -+ } -+ -+#if 0 -+ /* bitlord rules */ -+ /* packetlen must be bigger than 40 */ -+ /* first 4 bytes are zero */ -+ if (packet_len > 40 && get_u32(haystack, 8) == 0x00000000) -+ { -+ /* first rule: 00 00 00 00 01 00 00 xx xx xx xx 00 00 00 00*/ -+ if (get_u32(haystack, 12) == 0x00000000 && -+ get_u32(haystack, 16) == 0x00010000 && -+ get_u32(haystack, 24) == 0x00000000 ) -+ return (IPP2P_BIT * 100 + 71); -+ -+ /* 00 01 00 00 0d 00 00 xx xx xx xx 00 00 00 00*/ -+ if (get_u32(haystack, 12) == 0x00000001 && -+ get_u32(haystack, 16) == 0x000d0000 && -+ get_u32(haystack, 24) == 0x00000000 ) -+ return (IPP2P_BIT * 100 + 71); -+ -+ -+ } -+#endif -+ -+ return 0; -+}/*udp_search_bit*/ -+ -+ -+ -+/*Search for Ares commands*/ -+//#define IPP2P_DEBUG_ARES -+int -+search_ares (const unsigned char *payload, const u16 plen) -+//int search_ares (unsigned char *haystack, int packet_len, int head_len) -+{ -+// const unsigned char *t = haystack + head_len; -+ -+ /* all ares packets start with */ -+ if (payload[1] == 0 && (plen - payload[0]) == 3) -+ { -+ switch (payload[2]) -+ { -+ case 0x5a: -+ /* ares connect */ -+ if ( plen == 6 && payload[5] == 0x05 ) return ((IPP2P_ARES * 100) + 1); -+ break; -+ case 0x09: -+ /* ares search, min 3 chars --> 14 bytes -+ * lets define a search can be up to 30 chars --> max 34 bytes -+ */ -+ if ( plen >= 14 && plen <= 34 ) return ((IPP2P_ARES * 100) + 1); -+ break; -+#ifdef IPP2P_DEBUG_ARES -+ default: -+ printk(KERN_DEBUG "Unknown Ares command %x recognized, len: %u \n", (unsigned int) payload[2],plen); -+#endif /* IPP2P_DEBUG_ARES */ -+ } -+ } -+ -+#if 0 -+ /* found connect packet: 03 00 5a 04 03 05 */ -+ /* new version ares 1.8: 03 00 5a xx xx 05 */ -+ if ((plen) == 6){ /* possible connect command*/ -+ if ((payload[0] == 0x03) && (payload[1] == 0x00) && (payload[2] == 0x5a) && (payload[5] == 0x05)) -+ return ((IPP2P_ARES * 100) + 1); -+ } -+ if ((plen) == 60){ /* possible download command*/ -+ if ((payload[59] == 0x0a) && (payload[58] == 0x0a)){ -+ if (memcmp(t, "PUSH SHA1:", 10) == 0) /* found download command */ -+ return ((IPP2P_ARES * 100) + 2); -+ } -+ } -+#endif -+ -+ return 0; -+} /*search_ares*/ -+ -+/*Search for SoulSeek commands*/ -+int -+search_soul (const unsigned char *payload, const u16 plen) -+{ -+//#define IPP2P_DEBUG_SOUL -+ /* match: xx xx xx xx | xx = sizeof(payload) - 4 */ -+ if (get_u32(payload, 0) == (plen - 4)){ -+ const __u32 m=get_u32(payload, 4); -+ /* match 00 yy yy 00, yy can be everything */ -+ if ( get_u8(payload, 4) == 0x00 && get_u8(payload, 7) == 0x00 ) -+ { -+#ifdef IPP2P_DEBUG_SOUL -+ printk(KERN_DEBUG "0: Soulseek command 0x%x recognized\n",get_u32(payload, 4)); -+#endif /* IPP2P_DEBUG_SOUL */ -+ return ((IPP2P_SOUL * 100) + 1); -+ } -+ -+ /* next match: 01 yy 00 00 | yy can be everything */ -+ if ( get_u8(payload, 4) == 0x01 && get_u16(payload, 6) == 0x0000 ) -+ { -+#ifdef IPP2P_DEBUG_SOUL -+ printk(KERN_DEBUG "1: Soulseek command 0x%x recognized\n",get_u16(payload, 4)); -+#endif /* IPP2P_DEBUG_SOUL */ -+ return ((IPP2P_SOUL * 100) + 2); -+ } -+ -+ /* other soulseek commandos are: 1-5,7,9,13-18,22,23,26,28,35-37,40-46,50,51,60,62-69,91,92,1001 */ -+ /* try to do this in an intelligent way */ -+ /* get all small commandos */ -+ switch(m) -+ { -+ case 7: -+ case 9: -+ case 22: -+ case 23: -+ case 26: -+ case 28: -+ case 50: -+ case 51: -+ case 60: -+ case 91: -+ case 92: -+ case 1001: -+#ifdef IPP2P_DEBUG_SOUL -+ printk(KERN_DEBUG "2: Soulseek command 0x%x recognized\n",get_u16(payload, 4)); -+#endif /* IPP2P_DEBUG_SOUL */ -+ return ((IPP2P_SOUL * 100) + 3); -+ } -+ -+ if (m > 0 && m < 6 ) -+ { -+#ifdef IPP2P_DEBUG_SOUL -+ printk(KERN_DEBUG "3: Soulseek command 0x%x recognized\n",get_u16(payload, 4)); -+#endif /* IPP2P_DEBUG_SOUL */ -+ return ((IPP2P_SOUL * 100) + 4); -+ } -+ if (m > 12 && m < 19 ) -+ { -+#ifdef IPP2P_DEBUG_SOUL -+ printk(KERN_DEBUG "4: Soulseek command 0x%x recognized\n",get_u16(payload, 4)); -+#endif /* IPP2P_DEBUG_SOUL */ -+ return ((IPP2P_SOUL * 100) + 5); -+ } -+ -+ if (m > 34 && m < 38 ) -+ { -+#ifdef IPP2P_DEBUG_SOUL -+ printk(KERN_DEBUG "5: Soulseek command 0x%x recognized\n",get_u16(payload, 4)); -+#endif /* IPP2P_DEBUG_SOUL */ -+ return ((IPP2P_SOUL * 100) + 6); -+ } -+ -+ if (m > 39 && m < 47 ) -+ { -+#ifdef IPP2P_DEBUG_SOUL -+ printk(KERN_DEBUG "6: Soulseek command 0x%x recognized\n",get_u16(payload, 4)); -+#endif /* IPP2P_DEBUG_SOUL */ -+ return ((IPP2P_SOUL * 100) + 7); -+ } -+ -+ if (m > 61 && m < 70 ) -+ { -+#ifdef IPP2P_DEBUG_SOUL -+ printk(KERN_DEBUG "7: Soulseek command 0x%x recognized\n",get_u16(payload, 4)); -+#endif /* IPP2P_DEBUG_SOUL */ -+ return ((IPP2P_SOUL * 100) + 8); -+ } -+ -+#ifdef IPP2P_DEBUG_SOUL -+ printk(KERN_DEBUG "unknown SOULSEEK command: 0x%x, first 16 bit: 0x%x, first 8 bit: 0x%x ,soulseek ???\n",get_u32(payload, 4),get_u16(payload, 4) >> 16,get_u8(payload, 4) >> 24); -+#endif /* IPP2P_DEBUG_SOUL */ -+ } -+ -+ /* match 14 00 00 00 01 yy 00 00 00 STRING(YY) 01 00 00 00 00 46|50 00 00 00 00 */ -+ /* without size at the beginning !!! */ -+ if ( get_u32(payload, 0) == 0x14 && get_u8(payload, 4) == 0x01 ) -+ { -+ __u32 y=get_u32(payload, 5); -+ /* we need 19 chars + string */ -+ if ( (y + 19) <= (plen) ) -+ { -+ const unsigned char *w=payload+9+y; -+ if (get_u32(w, 0) == 0x01 && ( get_u16(w, 4) == 0x4600 || get_u16(w, 4) == 0x5000) && get_u32(w, 6) == 0x00); -+#ifdef IPP2P_DEBUG_SOUL -+ printk(KERN_DEBUG "Soulssek special client command recognized\n"); -+#endif /* IPP2P_DEBUG_SOUL */ -+ return ((IPP2P_SOUL * 100) + 9); -+ } -+ } -+ return 0; -+} -+ -+ -+/*Search for WinMX commands*/ -+int -+search_winmx (const unsigned char *payload, const u16 plen) -+{ -+//#define IPP2P_DEBUG_WINMX -+ if (((plen) == 4) && (memcmp(payload, "SEND", 4) == 0)) return ((IPP2P_WINMX * 100) + 1); -+ if (((plen) == 3) && (memcmp(payload, "GET", 3) == 0)) return ((IPP2P_WINMX * 100) + 2); -+ //if (packet_len < (head_len + 10)) return 0; -+ if (plen < 10) return 0; -+ -+ if ((memcmp(payload, "SEND", 4) == 0) || (memcmp(payload, "GET", 3) == 0)){ -+ u16 c=4; -+ const u16 end=plen-2; -+ u8 count=0; -+ while (c < end) -+ { -+ if (payload[c]== 0x20 && payload[c+1] == 0x22) -+ { -+ c++; -+ count++; -+ if (count>=2) return ((IPP2P_WINMX * 100) + 3); -+ } -+ c++; -+ } -+ } -+ -+ if ( plen == 149 && payload[0] == '8' ) -+ { -+#ifdef IPP2P_DEBUG_WINMX -+ printk(KERN_INFO "maybe WinMX\n"); -+#endif -+ if (get_u32(payload,17) == 0 && get_u32(payload,21) == 0 && get_u32(payload,25) == 0 && -+// get_u32(payload,33) == __constant_htonl(0x71182b1a) && get_u32(payload,37) == __constant_htonl(0x05050000) && -+// get_u32(payload,133) == __constant_htonl(0x31097edf) && get_u32(payload,145) == __constant_htonl(0xdcb8f792)) -+ get_u16(payload,39) == 0 && get_u16(payload,135) == __constant_htons(0x7edf) && get_u16(payload,147) == __constant_htons(0xf792)) -+ -+ { -+#ifdef IPP2P_DEBUG_WINMX -+ printk(KERN_INFO "got WinMX\n"); -+#endif -+ return ((IPP2P_WINMX * 100) + 4); -+ } -+ } -+ return 0; -+} /*search_winmx*/ -+ -+ -+/*Search for appleJuice commands*/ -+int -+search_apple (const unsigned char *payload, const u16 plen) -+{ -+ if ( (plen > 7) && (payload[6] == 0x0d) && (payload[7] == 0x0a) && (memcmp(payload, "ajprot", 6) == 0)) return (IPP2P_APPLE * 100); -+ -+ return 0; -+} -+ -+ -+/*Search for BitTorrent commands*/ -+int -+search_bittorrent (const unsigned char *payload, const u16 plen) -+{ -+ if (plen > 20) -+ { -+ /* test for match 0x13+"BitTorrent protocol" */ -+ if (payload[0] == 0x13) -+ { -+ if (memcmp(payload+1, "BitTorrent protocol", 19) == 0) return (IPP2P_BIT * 100); -+ } -+ -+ /* get tracker commandos, all starts with GET / -+ * then it can follow: scrape| announce -+ * and then ?hash_info= -+ */ -+ if (memcmp(payload,"GET /",5) == 0) -+ { -+ /* message scrape */ -+ if ( memcmp(payload+5,"scrape?info_hash=",17)==0 ) return (IPP2P_BIT * 100 + 1); -+ /* message announce */ -+ if ( memcmp(payload+5,"announce?info_hash=",19)==0 ) return (IPP2P_BIT * 100 + 2); -+ } -+ } -+ else -+ { -+ /* bitcomet encryptes the first packet, so we have to detect another -+ * one later in the flow */ -+ /* first try failed, too many missdetections */ -+ //if ( size == 5 && get_u32(t,0) == __constant_htonl(1) && t[4] < 3) return (IPP2P_BIT * 100 + 3); -+ -+ /* second try: block request packets */ -+ if ( plen == 17 && get_u32(payload,0) == __constant_htonl(0x0d) && payload[4] == 0x06 && get_u32(payload,13) == __constant_htonl(0x4000) ) return (IPP2P_BIT * 100 + 3); -+ } -+ -+ return 0; -+} -+ -+ -+ -+/*check for Kazaa get command*/ -+int -+search_kazaa (const unsigned char *payload, const u16 plen) -+ -+{ -+ if ((payload[plen-2] == 0x0d) && (payload[plen-1] == 0x0a) && memcmp(payload, "GET /.hash=", 11) == 0) -+ return (IPP2P_DATA_KAZAA * 100); -+ -+ return 0; -+} -+ -+ -+/*check for gnutella get command*/ -+int -+search_gnu (const unsigned char *payload, const u16 plen) -+{ -+ if ((payload[plen-2] == 0x0d) && (payload[plen-1] == 0x0a)) -+ { -+ if (memcmp(payload, "GET /get/", 9) == 0) return ((IPP2P_DATA_GNU * 100) + 1); -+ if (memcmp(payload, "GET /uri-res/", 13) == 0) return ((IPP2P_DATA_GNU * 100) + 2); -+ } -+ return 0; -+} -+ -+ -+/*check for gnutella get commands and other typical data*/ -+int -+search_all_gnu (const unsigned char *payload, const u16 plen) -+{ -+ -+ if ((payload[plen-2] == 0x0d) && (payload[plen-1] == 0x0a)) -+ { -+ -+ if (memcmp(payload, "GNUTELLA CONNECT/", 17) == 0) return ((IPP2P_GNU * 100) + 1); -+ if (memcmp(payload, "GNUTELLA/", 9) == 0) return ((IPP2P_GNU * 100) + 2); -+ -+ -+ if ((memcmp(payload, "GET /get/", 9) == 0) || (memcmp(payload, "GET /uri-res/", 13) == 0)) -+ { -+ u16 c=8; -+ const u16 end=plen-22; -+ while (c < end) { -+ if ( payload[c] == 0x0a && payload[c+1] == 0x0d && ((memcmp(&payload[c+2], "X-Gnutella-", 11) == 0) || (memcmp(&payload[c+2], "X-Queue:", 8) == 0))) -+ return ((IPP2P_GNU * 100) + 3); -+ c++; -+ } -+ } -+ } -+ return 0; -+} -+ -+ -+/*check for KaZaA download commands and other typical data*/ -+int -+search_all_kazaa (const unsigned char *payload, const u16 plen) -+{ -+ if ((payload[plen-2] == 0x0d) && (payload[plen-1] == 0x0a)) -+ { -+ -+ if (memcmp(payload, "GIVE ", 5) == 0) return ((IPP2P_KAZAA * 100) + 1); -+ -+ if (memcmp(payload, "GET /", 5) == 0) { -+ u16 c = 8; -+ const u16 end=plen-22; -+ while (c < end) { -+ if ( payload[c] == 0x0a && payload[c+1] == 0x0d && ((memcmp(&payload[c+2], "X-Kazaa-Username: ", 18) == 0) || (memcmp(&payload[c+2], "User-Agent: PeerEnabler/", 24) == 0))) -+ return ((IPP2P_KAZAA * 100) + 2); -+ c++; -+ } -+ } -+ } -+ return 0; -+} -+ -+/*fast check for edonkey file segment transfer command*/ -+int -+search_edk (const unsigned char *payload, const u16 plen) -+{ -+ if (payload[0] != 0xe3) -+ return 0; -+ else { -+ if (payload[5] == 0x47) -+ return (IPP2P_DATA_EDK * 100); -+ else -+ return 0; -+ } -+} -+ -+ -+ -+/*intensive but slower search for some edonkey packets including size-check*/ -+int -+search_all_edk (const unsigned char *payload, const u16 plen) -+{ -+ if (payload[0] != 0xe3) -+ return 0; -+ else { -+ //t += head_len; -+ const u16 cmd = get_u16(payload, 1); -+ if (cmd == (plen - 5)) { -+ switch (payload[5]) { -+ case 0x01: return ((IPP2P_EDK * 100) + 1); /*Client: hello or Server:hello*/ -+ case 0x4c: return ((IPP2P_EDK * 100) + 9); /*Client: Hello-Answer*/ -+ } -+ } -+ return 0; -+ } -+} -+ -+ -+/*fast check for Direct Connect send command*/ -+int -+search_dc (const unsigned char *payload, const u16 plen) -+{ -+ -+ if (payload[0] != 0x24 ) -+ return 0; -+ else { -+ if (memcmp(&payload[1], "Send|", 5) == 0) -+ return (IPP2P_DATA_DC * 100); -+ else -+ return 0; -+ } -+ -+} -+ -+ -+/*intensive but slower check for all direct connect packets*/ -+int -+search_all_dc (const unsigned char *payload, const u16 plen) -+{ -+// unsigned char *t = haystack; -+ -+ if (payload[0] == 0x24 && payload[plen-1] == 0x7c) -+ { -+ const unsigned char *t=&payload[1]; -+ /* Client-Hub-Protocol */ -+ if (memcmp(t, "Lock ", 5) == 0) return ((IPP2P_DC * 100) + 1); -+ /* Client-Client-Protocol, some are already recognized by client-hub (like lock) */ -+ if (memcmp(t, "MyNick ", 7) == 0) return ((IPP2P_DC * 100) + 38); -+ } -+ return 0; -+} -+ -+/*check for mute*/ -+int -+search_mute (const unsigned char *payload, const u16 plen) -+{ -+ if ( plen == 209 || plen == 345 || plen == 473 || plen == 609 || plen == 1121 ) -+ { -+ //printk(KERN_DEBUG "size hit: %u",size); -+ if (memcmp(payload,"PublicKey: ",11) == 0 ) -+ { -+ return ((IPP2P_MUTE * 100) + 0); -+ -+/* if (memcmp(t+size-14,"\x0aEndPublicKey\x0a",14) == 0) -+ { -+ printk(KERN_DEBUG "end pubic key hit: %u",size); -+ -+ }*/ -+ } -+ } -+ return 0; -+} -+ -+ -+/* check for xdcc */ -+int -+search_xdcc (const unsigned char *payload, const u16 plen) -+{ -+ /* search in small packets only */ -+ if (plen > 20 && plen < 200 && payload[plen-1] == 0x0a && payload[plen-2] == 0x0d && memcmp(payload,"PRIVMSG ",8) == 0) -+ { -+ -+ u16 x=10; -+ const u16 end=plen - 13; -+ -+ /* is seems to be a irc private massage, chedck for xdcc command */ -+ while (x < end) -+ { -+ if (payload[x] == ':') -+ { -+ if ( memcmp(&payload[x+1],"xdcc send #",11) == 0 ) -+ return ((IPP2P_XDCC * 100) + 0); -+ } -+ x++; -+ } -+ } -+ return 0; -+} -+ -+/* search for waste */ -+int search_waste(const unsigned char *payload, const u16 plen) -+{ -+ if ( plen >= 8 && memcmp(payload,"GET.sha1:",9) == 0) -+ return ((IPP2P_WASTE * 100) + 0); -+ -+ return 0; -+} -+ -+ -+static struct { -+ int command; -+ __u8 short_hand; /*for fucntions included in short hands*/ -+ int packet_len; -+ int (*function_name) (const unsigned char *, const u16); -+} matchlist[] = { -+ {IPP2P_EDK,SHORT_HAND_IPP2P,20, &search_all_edk}, -+// {IPP2P_DATA_KAZAA,SHORT_HAND_DATA,200, &search_kazaa}, -+// {IPP2P_DATA_EDK,SHORT_HAND_DATA,60, &search_edk}, -+// {IPP2P_DATA_DC,SHORT_HAND_DATA,26, &search_dc}, -+ {IPP2P_DC,SHORT_HAND_IPP2P,5, search_all_dc}, -+// {IPP2P_DATA_GNU,SHORT_HAND_DATA,40, &search_gnu}, -+ {IPP2P_GNU,SHORT_HAND_IPP2P,5, &search_all_gnu}, -+ {IPP2P_KAZAA,SHORT_HAND_IPP2P,5, &search_all_kazaa}, -+ {IPP2P_BIT,SHORT_HAND_IPP2P,20, &search_bittorrent}, -+ {IPP2P_APPLE,SHORT_HAND_IPP2P,5, &search_apple}, -+ {IPP2P_SOUL,SHORT_HAND_IPP2P,5, &search_soul}, -+ {IPP2P_WINMX,SHORT_HAND_IPP2P,2, &search_winmx}, -+ {IPP2P_ARES,SHORT_HAND_IPP2P,5, &search_ares}, -+ {IPP2P_MUTE,SHORT_HAND_NONE,200, &search_mute}, -+ {IPP2P_WASTE,SHORT_HAND_NONE,5, &search_waste}, -+ {IPP2P_XDCC,SHORT_HAND_NONE,5, &search_xdcc}, -+ {0,0,0,NULL} -+}; -+ -+ -+static struct { -+ int command; -+ __u8 short_hand; /*for fucntions included in short hands*/ -+ int packet_len; -+ int (*function_name) (unsigned char *, int); -+} udp_list[] = { -+ {IPP2P_KAZAA,SHORT_HAND_IPP2P,14, &udp_search_kazaa}, -+ {IPP2P_BIT,SHORT_HAND_IPP2P,23, &udp_search_bit}, -+ {IPP2P_GNU,SHORT_HAND_IPP2P,11, &udp_search_gnu}, -+ {IPP2P_EDK,SHORT_HAND_IPP2P,9, &udp_search_edk}, -+ {IPP2P_DC,SHORT_HAND_IPP2P,12, &udp_search_directconnect}, -+ {0,0,0,NULL} -+}; -+ -+ -+static int -+match(const struct sk_buff *skb, -+ const struct net_device *in, -+ const struct net_device *out, -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,18) -+ const struct xt_match *match, -+#endif -+ const void *matchinfo, -+ int offset, -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,18) -+ unsigned int protoff, -+#elif LINUX_VERSION_CODE < KERNEL_VERSION(2,6,0) -+ const void *hdr, -+ u_int16_t datalen, -+#endif -+ int *hotdrop) -+{ -+ const struct ipt_p2p_info *info = matchinfo; -+ unsigned char *haystack; -+ struct iphdr *ip = ip_hdr(skb); -+ int p2p_result = 0, i = 0; -+// int head_len; -+ int hlen = ntohs(ip->tot_len)-(ip->ihl*4); /*hlen = packet-data length*/ -+ -+ /*must not be a fragment*/ -+ if (offset) { -+ if (info->debug) printk("IPP2P.match: offset found %i \n",offset); -+ return 0; -+ } -+ -+ /*make sure that skb is linear*/ -+ if(skb_is_nonlinear(skb)){ -+ if (info->debug) printk("IPP2P.match: nonlinear skb found\n"); -+ return 0; -+ } -+ -+ -+ haystack=(char *)ip+(ip->ihl*4); /*haystack = packet data*/ -+ -+ switch (ip->protocol){ -+ case IPPROTO_TCP: /*what to do with a TCP packet*/ -+ { -+ struct tcphdr *tcph = (void *) ip + ip->ihl * 4; -+ -+ if (tcph->fin) return 0; /*if FIN bit is set bail out*/ -+ if (tcph->syn) return 0; /*if SYN bit is set bail out*/ -+ if (tcph->rst) return 0; /*if RST bit is set bail out*/ -+ -+ haystack += tcph->doff * 4; /*get TCP-Header-Size*/ -+ hlen -= tcph->doff * 4; -+ while (matchlist[i].command) { -+ if ((((info->cmd & matchlist[i].command) == matchlist[i].command) || -+ ((info->cmd & matchlist[i].short_hand) == matchlist[i].short_hand)) && -+ (hlen > matchlist[i].packet_len)) { -+ p2p_result = matchlist[i].function_name(haystack, hlen); -+ if (p2p_result) -+ { -+ if (info->debug) printk("IPP2P.debug:TCP-match: %i from: %u.%u.%u.%u:%i to: %u.%u.%u.%u:%i Length: %i\n", -+ p2p_result, NIPQUAD(ip->saddr),ntohs(tcph->source), NIPQUAD(ip->daddr),ntohs(tcph->dest),hlen); -+ return p2p_result; -+ } -+ } -+ i++; -+ } -+ return p2p_result; -+ } -+ -+ case IPPROTO_UDP: /*what to do with an UDP packet*/ -+ { -+ struct udphdr *udph = (void *) ip + ip->ihl * 4; -+ -+ while (udp_list[i].command){ -+ if ((((info->cmd & udp_list[i].command) == udp_list[i].command) || -+ ((info->cmd & udp_list[i].short_hand) == udp_list[i].short_hand)) && -+ (hlen > udp_list[i].packet_len)) { -+ p2p_result = udp_list[i].function_name(haystack, hlen); -+ if (p2p_result){ -+ if (info->debug) printk("IPP2P.debug:UDP-match: %i from: %u.%u.%u.%u:%i to: %u.%u.%u.%u:%i Length: %i\n", -+ p2p_result, NIPQUAD(ip->saddr),ntohs(udph->source), NIPQUAD(ip->daddr),ntohs(udph->dest),hlen); -+ return p2p_result; -+ } -+ } -+ i++; -+ } -+ return p2p_result; -+ } -+ -+ default: return 0; -+ } -+} -+ -+ -+ -+static int -+checkentry(const char *tablename, -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,18) -+ const void *ip, -+ const struct xt_match *match, -+#else -+ const struct ipt_ip *ip, -+#endif -+ void *matchinfo, -+#if LINUX_VERSION_CODE < KERNEL_VERSION(2,6,19) -+ unsigned int matchsize, -+#endif -+ unsigned int hook_mask) -+{ -+ /* Must specify -p tcp */ -+/* if (ip->proto != IPPROTO_TCP || (ip->invflags & IPT_INV_PROTO)) { -+ * printk("ipp2p: Only works on TCP packets, use -p tcp\n"); -+ * return 0; -+ * }*/ -+ return 1; -+} -+ -+ -+ -+ -+static struct ipt_match ipp2p_match = { -+#if LINUX_VERSION_CODE < KERNEL_VERSION(2,6,0) -+ { NULL, NULL }, -+ "ipp2p", -+ &match, -+ &checkentry, -+ NULL, -+ THIS_MODULE -+#endif -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,0) -+ .name = "ipp2p", -+ .match = &match, -+ .family = AF_INET, -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,18) -+ .matchsize = sizeof(struct ipt_p2p_info), -+#endif -+ .checkentry = &checkentry, -+ .me = THIS_MODULE, -+#endif -+}; -+ -+ -+static int __init init(void) -+{ -+ printk(KERN_INFO "IPP2P v%s loading\n", IPP2P_VERSION); -+ return xt_register_match(&ipp2p_match); -+} -+ -+static void __exit fini(void) -+{ -+ xt_unregister_match(&ipp2p_match); -+ printk(KERN_INFO "IPP2P v%s unloaded\n", IPP2P_VERSION); -+} -+ -+module_init(init); -+module_exit(fini); -+ -+ ---- a/net/ipv4/netfilter/Kconfig -+++ b/net/ipv4/netfilter/Kconfig -@@ -63,6 +63,12 @@ - - To compile it as a module, choose M here. If unsure, say N. - -+config IP_NF_MATCH_IPP2P -+ tristate "IPP2P" -+ depends on IP_NF_IPTABLES -+ help -+ Module for matching traffic of various Peer-to-Peer applications -+ - config IP_NF_MATCH_TOS - tristate "TOS match support" - depends on IP_NF_IPTABLES ---- a/net/ipv4/netfilter/Makefile -+++ b/net/ipv4/netfilter/Makefile -@@ -49,6 +49,7 @@ - obj-$(CONFIG_IP_NF_MATCH_AH) += ipt_ah.o - obj-$(CONFIG_IP_NF_MATCH_TTL) += ipt_ttl.o - obj-$(CONFIG_IP_NF_MATCH_ADDRTYPE) += ipt_addrtype.o -+obj-$(CONFIG_IP_NF_MATCH_IPP2P) += ipt_ipp2p.o - - # targets - obj-$(CONFIG_IP_NF_TARGET_REJECT) += ipt_REJECT.o diff --git a/target/linux/generic-2.6/patches-2.6.22/120-openswan-2.4.0.kernel-2.6-natt.patch b/target/linux/generic-2.6/patches-2.6.22/120-openswan-2.4.0.kernel-2.6-natt.patch deleted file mode 100644 index 402d345..0000000 --- a/target/linux/generic-2.6/patches-2.6.22/120-openswan-2.4.0.kernel-2.6-natt.patch +++ /dev/null @@ -1,166 +0,0 @@ ---- /dev/null -+++ b/include/net/xfrmudp.h -@@ -0,0 +1,10 @@ -+/* -+ * pointer to function for type that xfrm4_input wants, to permit -+ * decoupling of XFRM from udp.c -+ */ -+#define HAVE_XFRM4_UDP_REGISTER -+ -+typedef int (*xfrm4_rcv_encap_t)(struct sk_buff *skb, __u16 encap_type); -+extern int udp4_register_esp_rcvencap(xfrm4_rcv_encap_t func -+ , xfrm4_rcv_encap_t *oldfunc); -+extern int udp4_unregister_esp_rcvencap(xfrm4_rcv_encap_t func); ---- a/net/ipv4/Kconfig -+++ b/net/ipv4/Kconfig -@@ -266,6 +266,12 @@ - Network), but can be distributed all over the Internet. If you want - to do that, say Y here and to "IP multicast routing" below. - -+config IPSEC_NAT_TRAVERSAL -+ bool "IPSEC NAT-Traversal (KLIPS compatible)" -+ depends on INET -+ ---help--- -+ Includes support for RFC3947/RFC3948 NAT-Traversal of ESP over UDP. -+ - config IP_MROUTE - bool "IP: multicast routing" - depends on IP_MULTICAST ---- a/net/ipv4/udp.c -+++ b/net/ipv4/udp.c -@@ -101,12 +101,15 @@ - #include <net/route.h> - #include <net/checksum.h> - #include <net/xfrm.h> -+#include <net/xfrmudp.h> - #include "udp_impl.h" - - /* - * Snmp MIB for the UDP layer - */ - -+static xfrm4_rcv_encap_t xfrm4_rcv_encap_func; -+ - DEFINE_SNMP_STAT(struct udp_mib, udp_statistics) __read_mostly; - - struct hlist_head udp_hash[UDP_HTABLE_SIZE]; -@@ -919,6 +922,42 @@ - return 0; - } - -+#if defined(CONFIG_XFRM) || defined(CONFIG_IPSEC_NAT_TRAVERSAL) -+ -+/* if XFRM isn't a module, then register it directly. */ -+#if 0 && !defined(CONFIG_XFRM_MODULE) && !defined(CONFIG_IPSEC_NAT_TRAVERSAL) -+static xfrm4_rcv_encap_t xfrm4_rcv_encap_func = xfrm4_rcv_encap; -+#else -+static xfrm4_rcv_encap_t xfrm4_rcv_encap_func = NULL; -+#endif -+ -+int udp4_register_esp_rcvencap(xfrm4_rcv_encap_t func -+ , xfrm4_rcv_encap_t *oldfunc) -+{ -+ if(oldfunc != NULL) { -+ *oldfunc = xfrm4_rcv_encap_func; -+ } -+ -+#if 0 -+ if(xfrm4_rcv_encap_func != NULL) -+ return -1; -+#endif -+ -+ xfrm4_rcv_encap_func = func; -+ return 0; -+} -+ -+int udp4_unregister_esp_rcvencap(xfrm4_rcv_encap_t func) -+{ -+ if(xfrm4_rcv_encap_func != func) -+ return -1; -+ -+ xfrm4_rcv_encap_func = NULL; -+ return 0; -+} -+#endif /* CONFIG_XFRM_MODULE || CONFIG_IPSEC_NAT_TRAVERSAL */ -+ -+ - /* return: - * 1 if the UDP system should process it - * 0 if we should drop this packet -@@ -926,7 +965,7 @@ - */ - static int udp_encap_rcv(struct sock * sk, struct sk_buff *skb) - { --#ifndef CONFIG_XFRM -+#if !defined(CONFIG_XFRM) && !defined(CONFIG_IPSEC_NAT_TRAVERSAL) - return 1; - #else - struct udp_sock *up = udp_sk(sk); -@@ -941,11 +980,11 @@ - /* if we're overly short, let UDP handle it */ - len = skb->len - sizeof(struct udphdr); - if (len <= 0) -- return 1; -+ return 2; - - /* if this is not encapsulated socket, then just return now */ - if (!encap_type) -- return 1; -+ return 3; - - /* If this is a paged skb, make sure we pull up - * whatever data we need to look at. */ -@@ -968,7 +1007,7 @@ - len = sizeof(struct udphdr); - } else - /* Must be an IKE packet.. pass it through */ -- return 1; -+ return 4; - break; - case UDP_ENCAP_ESPINUDP_NON_IKE: - /* Check if this is a keepalive packet. If so, eat it. */ -@@ -981,7 +1020,7 @@ - len = sizeof(struct udphdr) + 2 * sizeof(u32); - } else - /* Must be an IKE packet.. pass it through */ -- return 1; -+ return 5; - break; - } - -@@ -992,6 +1031,8 @@ - */ - if (skb_cloned(skb) && pskb_expand_head(skb, 0, 0, GFP_ATOMIC)) - return 0; -+ if (skb_cloned(skb) && pskb_expand_head(skb, 0, 0, GFP_ATOMIC)) -+ return 0; - - /* Now we can update and verify the packet length... */ - iph = ip_hdr(skb); -@@ -1056,9 +1097,13 @@ - return 0; - } - if (ret < 0) { -- /* process the ESP packet */ -- ret = xfrm4_rcv_encap(skb, up->encap_type); -- UDP_INC_STATS_BH(UDP_MIB_INDATAGRAMS, up->pcflag); -+ if(xfrm4_rcv_encap_func != NULL) { -+ ret = (*xfrm4_rcv_encap_func)(skb, up->encap_type); -+ UDP_INC_STATS_BH(UDP_MIB_INDATAGRAMS, up->pcflag); -+ } else { -+ UDP_INC_STATS_BH(UDP_MIB_INERRORS, up->pcflag); -+ ret = 1; -+ } - return -ret; - } - /* FALLTHROUGH -- it's a UDP Packet */ -@@ -1742,3 +1787,9 @@ - EXPORT_SYMBOL(udp_proc_register); - EXPORT_SYMBOL(udp_proc_unregister); - #endif -+ -+#if defined(CONFIG_IPSEC_NAT_TRAVERSAL) -+EXPORT_SYMBOL(udp4_register_esp_rcvencap); -+EXPORT_SYMBOL(udp4_unregister_esp_rcvencap); -+#endif -+ diff --git a/target/linux/generic-2.6/patches-2.6.22/130-netfilter_ipset.patch b/target/linux/generic-2.6/patches-2.6.22/130-netfilter_ipset.patch deleted file mode 100644 index 4d30282..0000000 --- a/target/linux/generic-2.6/patches-2.6.22/130-netfilter_ipset.patch +++ /dev/null @@ -1,7664 +0,0 @@ ---- /dev/null -+++ b/include/linux/netfilter_ipv4/ip_set.h -@@ -0,0 +1,498 @@ -+#ifndef _IP_SET_H -+#define _IP_SET_H -+ -+/* Copyright (C) 2000-2002 Joakim Axelsson <gozem@linux.nu> -+ * Patrick Schaaf <bof@bof.de> -+ * Martin Josefsson <gandalf@wlug.westbo.se> -+ * Copyright (C) 2003-2004 Jozsef Kadlecsik <kadlec@blackhole.kfki.hu> -+ * -+ * This program is free software; you can redistribute it and/or modify -+ * it under the terms of the GNU General Public License version 2 as -+ * published by the Free Software Foundation. -+ */ -+ -+#if 0 -+#define IP_SET_DEBUG -+#endif -+ -+/* -+ * A sockopt of such quality has hardly ever been seen before on the open -+ * market! This little beauty, hardly ever used: above 64, so it's -+ * traditionally used for firewalling, not touched (even once!) by the -+ * 2.0, 2.2 and 2.4 kernels! -+ * -+ * Comes with its own certificate of authenticity, valid anywhere in the -+ * Free world! -+ * -+ * Rusty, 19.4.2000 -+ */ -+#define SO_IP_SET 83 -+ -+/* -+ * Heavily modify by Joakim Axelsson 08.03.2002 -+ * - Made it more modulebased -+ * -+ * Additional heavy modifications by Jozsef Kadlecsik 22.02.2004 -+ * - bindings added -+ * - in order to "deal with" backward compatibility, renamed to ipset -+ */ -+ -+/* -+ * Used so that the kernel module and ipset-binary can match their versions -+ */ -+#define IP_SET_PROTOCOL_VERSION 2 -+ -+#define IP_SET_MAXNAMELEN 32 /* set names and set typenames */ -+ -+/* Lets work with our own typedef for representing an IP address. -+ * We hope to make the code more portable, possibly to IPv6... -+ * -+ * The representation works in HOST byte order, because most set types -+ * will perform arithmetic operations and compare operations. -+ * -+ * For now the type is an uint32_t. -+ * -+ * Make sure to ONLY use the functions when translating and parsing -+ * in order to keep the host byte order and make it more portable: -+ * parse_ip() -+ * parse_mask() -+ * parse_ipandmask() -+ * ip_tostring() -+ * (Joakim: where are they???) -+ */ -+ -+typedef uint32_t ip_set_ip_t; -+ -+/* Sets are identified by an id in kernel space. Tweak with ip_set_id_t -+ * and IP_SET_INVALID_ID if you want to increase the max number of sets. -+ */ -+typedef uint16_t ip_set_id_t; -+ -+#define IP_SET_INVALID_ID 65535 -+ -+/* How deep we follow bindings */ -+#define IP_SET_MAX_BINDINGS 6 -+ -+/* -+ * Option flags for kernel operations (ipt_set_info) -+ */ -+#define IPSET_SRC 0x01 /* Source match/add */ -+#define IPSET_DST 0x02 /* Destination match/add */ -+#define IPSET_MATCH_INV 0x04 /* Inverse matching */ -+ -+/* -+ * Set features -+ */ -+#define IPSET_TYPE_IP 0x01 /* IP address type of set */ -+#define IPSET_TYPE_PORT 0x02 /* Port type of set */ -+#define IPSET_DATA_SINGLE 0x04 /* Single data storage */ -+#define IPSET_DATA_DOUBLE 0x08 /* Double data storage */ -+ -+/* Reserved keywords */ -+#define IPSET_TOKEN_DEFAULT ":default:" -+#define IPSET_TOKEN_ALL ":all:" -+ -+/* SO_IP_SET operation constants, and their request struct types. -+ * -+ * Operation ids: -+ * 0-99: commands with version checking -+ * 100-199: add/del/test/bind/unbind -+ * 200-299: list, save, restore -+ */ -+ -+/* Single shot operations: -+ * version, create, destroy, flush, rename and swap -+ * -+ * Sets are identified by name. -+ */ -+ -+#define IP_SET_REQ_STD \ -+ unsigned op; \ -+ unsigned version; \ -+ char name[IP_SET_MAXNAMELEN] -+ -+#define IP_SET_OP_CREATE 0x00000001 /* Create a new (empty) set */ -+struct ip_set_req_create { -+ IP_SET_REQ_STD; -+ char typename[IP_SET_MAXNAMELEN]; -+}; -+ -+#define IP_SET_OP_DESTROY 0x00000002 /* Remove a (empty) set */ -+struct ip_set_req_std { -+ IP_SET_REQ_STD; -+}; -+ -+#define IP_SET_OP_FLUSH 0x00000003 /* Remove all IPs in a set */ -+/* Uses ip_set_req_std */ -+ -+#define IP_SET_OP_RENAME 0x00000004 /* Rename a set */ -+/* Uses ip_set_req_create */ -+ -+#define IP_SET_OP_SWAP 0x00000005 /* Swap two sets */ -+/* Uses ip_set_req_create */ -+ -+union ip_set_name_index { -+ char name[IP_SET_MAXNAMELEN]; -+ ip_set_id_t index; -+}; -+ -+#define IP_SET_OP_GET_BYNAME 0x00000006 /* Get set index by name */ -+struct ip_set_req_get_set { -+ unsigned op; -+ unsigned version; -+ union ip_set_name_index set; -+}; -+ -+#define IP_SET_OP_GET_BYINDEX 0x00000007 /* Get set name by index */ -+/* Uses ip_set_req_get_set */ -+ -+#define IP_SET_OP_VERSION 0x00000100 /* Ask kernel version */ -+struct ip_set_req_version { -+ unsigned op; -+ unsigned version; -+}; -+ -+/* Double shots operations: -+ * add, del, test, bind and unbind. -+ * -+ * First we query the kernel to get the index and type of the target set, -+ * then issue the command. Validity of IP is checked in kernel in order -+ * to minimalize sockopt operations. -+ */ -+ -+/* Get minimal set data for add/del/test/bind/unbind IP */ -+#define IP_SET_OP_ADT_GET 0x00000010 /* Get set and type */ -+struct ip_set_req_adt_get { -+ unsigned op; -+ unsigned version; -+ union ip_set_name_index set; -+ char typename[IP_SET_MAXNAMELEN]; -+}; -+ -+#define IP_SET_REQ_BYINDEX \ -+ unsigned op; \ -+ ip_set_id_t index; -+ -+struct ip_set_req_adt { -+ IP_SET_REQ_BYINDEX; -+}; -+ -+#define IP_SET_OP_ADD_IP 0x00000101 /* Add an IP to a set */ -+/* Uses ip_set_req_adt, with type specific addage */ -+ -+#define IP_SET_OP_DEL_IP 0x00000102 /* Remove an IP from a set */ -+/* Uses ip_set_req_adt, with type specific addage */ -+ -+#define IP_SET_OP_TEST_IP 0x00000103 /* Test an IP in a set */ -+/* Uses ip_set_req_adt, with type specific addage */ -+ -+#define IP_SET_OP_BIND_SET 0x00000104 /* Bind an IP to a set */ -+/* Uses ip_set_req_bind, with type specific addage */ -+struct ip_set_req_bind { -+ IP_SET_REQ_BYINDEX; -+ char binding[IP_SET_MAXNAMELEN]; -+}; -+ -+#define IP_SET_OP_UNBIND_SET 0x00000105 /* Unbind an IP from a set */ -+/* Uses ip_set_req_bind, with type speficic addage -+ * index = 0 means unbinding for all sets */ -+ -+#define IP_SET_OP_TEST_BIND_SET 0x00000106 /* Test binding an IP to a set */ -+/* Uses ip_set_req_bind, with type specific addage */ -+ -+/* Multiple shots operations: list, save, restore. -+ * -+ * - check kernel version and query the max number of sets -+ * - get the basic information on all sets -+ * and size required for the next step -+ * - get actual set data: header, data, bindings -+ */ -+ -+/* Get max_sets and the index of a queried set -+ */ -+#define IP_SET_OP_MAX_SETS 0x00000020 -+struct ip_set_req_max_sets { -+ unsigned op; -+ unsigned version; -+ ip_set_id_t max_sets; /* max_sets */ -+ ip_set_id_t sets; /* real number of sets */ -+ union ip_set_name_index set; /* index of set if name used */ -+}; -+ -+/* Get the id and name of the sets plus size for next step */ -+#define IP_SET_OP_LIST_SIZE 0x00000201 -+#define IP_SET_OP_SAVE_SIZE 0x00000202 -+struct ip_set_req_setnames { -+ unsigned op; -+ ip_set_id_t index; /* set to list/save */ -+ size_t size; /* size to get setdata/bindings */ -+ /* followed by sets number of struct ip_set_name_list */ -+}; -+ -+struct ip_set_name_list { -+ char name[IP_SET_MAXNAMELEN]; -+ char typename[IP_SET_MAXNAMELEN]; -+ ip_set_id_t index; -+ ip_set_id_t id; -+}; -+ -+/* The actual list operation */ -+#define IP_SET_OP_LIST 0x00000203 -+struct ip_set_req_list { -+ IP_SET_REQ_BYINDEX; -+ /* sets number of struct ip_set_list in reply */ -+}; -+ -+struct ip_set_list { -+ ip_set_id_t index; -+ ip_set_id_t binding; -+ u_int32_t ref; -+ size_t header_size; /* Set header data of header_size */ -+ size_t members_size; /* Set members data of members_size */ -+ size_t bindings_size; /* Set bindings data of bindings_size */ -+}; -+ -+struct ip_set_hash_list { -+ ip_set_ip_t ip; -+ ip_set_id_t binding; -+}; -+ -+/* The save operation */ -+#define IP_SET_OP_SAVE 0x00000204 -+/* Uses ip_set_req_list, in the reply replaced by -+ * sets number of struct ip_set_save plus a marker -+ * ip_set_save followed by ip_set_hash_save structures. -+ */ -+struct ip_set_save { -+ ip_set_id_t index; -+ ip_set_id_t binding; -+ size_t header_size; /* Set header data of header_size */ -+ size_t members_size; /* Set members data of members_size */ -+}; -+ -+/* At restoring, ip == 0 means default binding for the given set: */ -+struct ip_set_hash_save { -+ ip_set_ip_t ip; -+ ip_set_id_t id; -+ ip_set_id_t binding; -+}; -+ -+/* The restore operation */ -+#define IP_SET_OP_RESTORE 0x00000205 -+/* Uses ip_set_req_setnames followed by ip_set_restore structures -+ * plus a marker ip_set_restore, followed by ip_set_hash_save -+ * structures. -+ */ -+struct ip_set_restore { -+ char name[IP_SET_MAXNAMELEN]; -+ char typename[IP_SET_MAXNAMELEN]; -+ ip_set_id_t index; -+ size_t header_size; /* Create data of header_size */ -+ size_t members_size; /* Set members data of members_size */ -+}; -+ -+static inline int bitmap_bytes(ip_set_ip_t a, ip_set_ip_t b) -+{ -+ return 4 * ((((b - a + 8) / 8) + 3) / 4); -+} -+ -+#ifdef __KERNEL__ -+ -+#define ip_set_printk(format, args...) \ -+ do { \ -+ printk("%s: %s: ", __FILE__, __FUNCTION__); \ -+ printk(format "\n" , ## args); \ -+ } while (0) -+ -+#if defined(IP_SET_DEBUG) -+#define DP(format, args...) \ -+ do { \ -+ printk("%s: %s (DBG): ", __FILE__, __FUNCTION__);\ -+ printk(format "\n" , ## args); \ -+ } while (0) -+#define IP_SET_ASSERT(x) \ -+ do { \ -+ if (!(x)) \ -+ printk("IP_SET_ASSERT: %s:%i(%s)\n", \ -+ __FILE__, __LINE__, __FUNCTION__); \ -+ } while (0) -+#else -+#define DP(format, args...) -+#define IP_SET_ASSERT(x) -+#endif -+ -+struct ip_set; -+ -+/* -+ * The ip_set_type definition - one per set type, e.g. "ipmap". -+ * -+ * Each individual set has a pointer, set->type, going to one -+ * of these structures. Function pointers inside the structure implement -+ * the real behaviour of the sets. -+ * -+ * If not mentioned differently, the implementation behind the function -+ * pointers of a set_type, is expected to return 0 if ok, and a negative -+ * errno (e.g. -EINVAL) on error. -+ */ -+struct ip_set_type { -+ struct list_head list; /* next in list of set types */ -+ -+ /* test for IP in set (kernel: iptables -m set src|dst) -+ * return 0 if not in set, 1 if in set. -+ */ -+ int (*testip_kernel) (struct ip_set *set, -+ const struct sk_buff * skb, -+ ip_set_ip_t *ip, -+ const u_int32_t *flags, -+ unsigned char index); -+ -+ /* test for IP in set (userspace: ipset -T set IP) -+ * return 0 if not in set, 1 if in set. -+ */ -+ int (*testip) (struct ip_set *set, -+ const void *data, size_t size, -+ ip_set_ip_t *ip); -+ -+ /* -+ * Size of the data structure passed by when -+ * adding/deletin/testing an entry. -+ */ -+ size_t reqsize; -+ -+ /* Add IP into set (userspace: ipset -A set IP) -+ * Return -EEXIST if the address is already in the set, -+ * and -ERANGE if the address lies outside the set bounds. -+ * If the address was not already in the set, 0 is returned. -+ */ -+ int (*addip) (struct ip_set *set, -+ const void *data, size_t size, -+ ip_set_ip_t *ip); -+ -+ /* Add IP into set (kernel: iptables ... -j SET set src|dst) -+ * Return -EEXIST if the address is already in the set, -+ * and -ERANGE if the address lies outside the set bounds. -+ * If the address was not already in the set, 0 is returned. -+ */ -+ int (*addip_kernel) (struct ip_set *set, -+ const struct sk_buff * skb, -+ ip_set_ip_t *ip, -+ const u_int32_t *flags, -+ unsigned char index); -+ -+ /* remove IP from set (userspace: ipset -D set --entry x) -+ * Return -EEXIST if the address is NOT in the set, -+ * and -ERANGE if the address lies outside the set bounds. -+ * If the address really was in the set, 0 is returned. -+ */ -+ int (*delip) (struct ip_set *set, -+ const void *data, size_t size, -+ ip_set_ip_t *ip); -+ -+ /* remove IP from set (kernel: iptables ... -j SET --entry x) -+ * Return -EEXIST if the address is NOT in the set, -+ * and -ERANGE if the address lies outside the set bounds. -+ * If the address really was in the set, 0 is returned. -+ */ -+ int (*delip_kernel) (struct ip_set *set, -+ const struct sk_buff * skb, -+ ip_set_ip_t *ip, -+ const u_int32_t *flags, -+ unsigned char index); -+ -+ /* new set creation - allocated type specific items -+ */ -+ int (*create) (struct ip_set *set, -+ const void *data, size_t size); -+ -+ /* retry the operation after successfully tweaking the set -+ */ -+ int (*retry) (struct ip_set *set); -+ -+ /* set destruction - free type specific items -+ * There is no return value. -+ * Can be called only when child sets are destroyed. -+ */ -+ void (*destroy) (struct ip_set *set); -+ -+ /* set flushing - reset all bits in the set, or something similar. -+ * There is no return value. -+ */ -+ void (*flush) (struct ip_set *set); -+ -+ /* Listing: size needed for header -+ */ -+ size_t header_size; -+ -+ /* Listing: Get the header -+ * -+ * Fill in the information in "data". -+ * This function is always run after list_header_size() under a -+ * writelock on the set. Therefor is the length of "data" always -+ * correct. -+ */ -+ void (*list_header) (const struct ip_set *set, -+ void *data); -+ -+ /* Listing: Get the size for the set members -+ */ -+ int (*list_members_size) (const struct ip_set *set); -+ -+ /* Listing: Get the set members -+ * -+ * Fill in the information in "data". -+ * This function is always run after list_member_size() under a -+ * writelock on the set. Therefor is the length of "data" always -+ * correct. -+ */ -+ void (*list_members) (const struct ip_set *set, -+ void *data); -+ -+ char typename[IP_SET_MAXNAMELEN]; -+ unsigned char features; -+ int protocol_version; -+ -+ /* Set this to THIS_MODULE if you are a module, otherwise NULL */ -+ struct module *me; -+}; -+ -+extern int ip_set_register_set_type(struct ip_set_type *set_type); -+extern void ip_set_unregister_set_type(struct ip_set_type *set_type); -+ -+/* A generic ipset */ -+struct ip_set { -+ char name[IP_SET_MAXNAMELEN]; /* the name of the set */ -+ rwlock_t lock; /* lock for concurrency control */ -+ ip_set_id_t id; /* set id for swapping */ -+ ip_set_id_t binding; /* default binding for the set */ -+ atomic_t ref; /* in kernel and in hash references */ -+ struct ip_set_type *type; /* the set types */ -+ void *data; /* pooltype specific data */ -+}; -+ -+/* Structure to bind set elements to sets */ -+struct ip_set_hash { -+ struct list_head list; /* list of clashing entries in hash */ -+ ip_set_ip_t ip; /* ip from set */ -+ ip_set_id_t id; /* set id */ -+ ip_set_id_t binding; /* set we bind the element to */ -+}; -+ -+/* register and unregister set references */ -+extern ip_set_id_t ip_set_get_byname(const char name[IP_SET_MAXNAMELEN]); -+extern ip_set_id_t ip_set_get_byindex(ip_set_id_t id); -+extern void ip_set_put(ip_set_id_t id); -+ -+/* API for iptables set match, and SET target */ -+extern void ip_set_addip_kernel(ip_set_id_t id, -+ const struct sk_buff *skb, -+ const u_int32_t *flags); -+extern void ip_set_delip_kernel(ip_set_id_t id, -+ const struct sk_buff *skb, -+ const u_int32_t *flags); -+extern int ip_set_testip_kernel(ip_set_id_t id, -+ const struct sk_buff *skb, -+ const u_int32_t *flags); -+ -+#endif /* __KERNEL__ */ -+ -+#endif /*_IP_SET_H*/ ---- /dev/null -+++ b/include/linux/netfilter_ipv4/ip_set_iphash.h -@@ -0,0 +1,30 @@ -+#ifndef __IP_SET_IPHASH_H -+#define __IP_SET_IPHASH_H -+ -+#include <linux/netfilter_ipv4/ip_set.h> -+ -+#define SETTYPE_NAME "iphash" -+#define MAX_RANGE 0x0000FFFF -+ -+struct ip_set_iphash { -+ ip_set_ip_t *members; /* the iphash proper */ -+ uint32_t elements; /* number of elements */ -+ uint32_t hashsize; /* hash size */ -+ uint16_t probes; /* max number of probes */ -+ uint16_t resize; /* resize factor in percent */ -+ ip_set_ip_t netmask; /* netmask */ -+ void *initval[0]; /* initvals for jhash_1word */ -+}; -+ -+struct ip_set_req_iphash_create { -+ uint32_t hashsize; -+ uint16_t probes; -+ uint16_t resize; -+ ip_set_ip_t netmask; -+}; -+ -+struct ip_set_req_iphash { -+ ip_set_ip_t ip; -+}; -+ -+#endif /* __IP_SET_IPHASH_H */ ---- /dev/null -+++ b/include/linux/netfilter_ipv4/ip_set_ipmap.h -@@ -0,0 +1,56 @@ -+#ifndef __IP_SET_IPMAP_H -+#define __IP_SET_IPMAP_H -+ -+#include <linux/netfilter_ipv4/ip_set.h> -+ -+#define SETTYPE_NAME "ipmap" -+#define MAX_RANGE 0x0000FFFF -+ -+struct ip_set_ipmap { -+ void *members; /* the ipmap proper */ -+ ip_set_ip_t first_ip; /* host byte order, included in range */ -+ ip_set_ip_t last_ip; /* host byte order, included in range */ -+ ip_set_ip_t netmask; /* subnet netmask */ -+ ip_set_ip_t sizeid; /* size of set in IPs */ -+ ip_set_ip_t hosts; /* number of hosts in a subnet */ -+}; -+ -+struct ip_set_req_ipmap_create { -+ ip_set_ip_t from; -+ ip_set_ip_t to; -+ ip_set_ip_t netmask; -+}; -+ -+struct ip_set_req_ipmap { -+ ip_set_ip_t ip; -+}; -+ -+unsigned int -+mask_to_bits(ip_set_ip_t mask) -+{ -+ unsigned int bits = 32; -+ ip_set_ip_t maskaddr; -+ -+ if (mask == 0xFFFFFFFF) -+ return bits; -+ -+ maskaddr = 0xFFFFFFFE; -+ while (--bits >= 0 && maskaddr != mask) -+ maskaddr <<= 1; -+ -+ return bits; -+} -+ -+ip_set_ip_t -+range_to_mask(ip_set_ip_t from, ip_set_ip_t to, unsigned int *bits) -+{ -+ ip_set_ip_t mask = 0xFFFFFFFE; -+ -+ *bits = 32; -+ while (--(*bits) >= 0 && mask && (to & mask) != from) -+ mask <<= 1; -+ -+ return mask; -+} -+ -+#endif /* __IP_SET_IPMAP_H */ ---- /dev/null -+++ b/include/linux/netfilter_ipv4/ip_set_ipporthash.h -@@ -0,0 +1,34 @@ -+#ifndef __IP_SET_IPPORTHASH_H -+#define __IP_SET_IPPORTHASH_H -+ -+#include <linux/netfilter_ipv4/ip_set.h> -+ -+#define SETTYPE_NAME "ipporthash" -+#define MAX_RANGE 0x0000FFFF -+#define INVALID_PORT (MAX_RANGE + 1) -+ -+struct ip_set_ipporthash { -+ ip_set_ip_t *members; /* the ipporthash proper */ -+ uint32_t elements; /* number of elements */ -+ uint32_t hashsize; /* hash size */ -+ uint16_t probes; /* max number of probes */ -+ uint16_t resize; /* resize factor in percent */ -+ ip_set_ip_t first_ip; /* host byte order, included in range */ -+ ip_set_ip_t last_ip; /* host byte order, included in range */ -+ void *initval[0]; /* initvals for jhash_1word */ -+}; -+ -+struct ip_set_req_ipporthash_create { -+ uint32_t hashsize; -+ uint16_t probes; -+ uint16_t resize; -+ ip_set_ip_t from; -+ ip_set_ip_t to; -+}; -+ -+struct ip_set_req_ipporthash { -+ ip_set_ip_t ip; -+ ip_set_ip_t port; -+}; -+ -+#endif /* __IP_SET_IPPORTHASH_H */ ---- /dev/null -+++ b/include/linux/netfilter_ipv4/ip_set_iptree.h -@@ -0,0 +1,40 @@ -+#ifndef __IP_SET_IPTREE_H -+#define __IP_SET_IPTREE_H -+ -+#include <linux/netfilter_ipv4/ip_set.h> -+ -+#define SETTYPE_NAME "iptree" -+#define MAX_RANGE 0x0000FFFF -+ -+struct ip_set_iptreed { -+ unsigned long expires[256]; /* x.x.x.ADDR */ -+}; -+ -+struct ip_set_iptreec { -+ struct ip_set_iptreed *tree[256]; /* x.x.ADDR.* */ -+}; -+ -+struct ip_set_iptreeb { -+ struct ip_set_iptreec *tree[256]; /* x.ADDR.*.* */ -+}; -+ -+struct ip_set_iptree { -+ unsigned int timeout; -+ unsigned int gc_interval; -+#ifdef __KERNEL__ -+ uint32_t elements; /* number of elements */ -+ struct timer_list gc; -+ struct ip_set_iptreeb *tree[256]; /* ADDR.*.*.* */ -+#endif -+}; -+ -+struct ip_set_req_iptree_create { -+ unsigned int timeout; -+}; -+ -+struct ip_set_req_iptree { -+ ip_set_ip_t ip; -+ unsigned int timeout; -+}; -+ -+#endif /* __IP_SET_IPTREE_H */ ---- /dev/null -+++ b/include/linux/netfilter_ipv4/ip_set_iptreemap.h -@@ -0,0 +1,40 @@ -+#ifndef __IP_SET_IPTREEMAP_H -+#define __IP_SET_IPTREEMAP_H -+ -+#include <linux/netfilter_ipv4/ip_set.h> -+ -+#define SETTYPE_NAME "iptreemap" -+ -+#ifdef __KERNEL__ -+struct ip_set_iptreemap_d { -+ unsigned char bitmap[32]; /* x.x.x.y */ -+}; -+ -+struct ip_set_iptreemap_c { -+ struct ip_set_iptreemap_d *tree[256]; /* x.x.y.x */ -+}; -+ -+struct ip_set_iptreemap_b { -+ struct ip_set_iptreemap_c *tree[256]; /* x.y.x.x */ -+ unsigned char dirty[32]; -+}; -+#endif -+ -+struct ip_set_iptreemap { -+ unsigned int gc_interval; -+#ifdef __KERNEL__ -+ struct timer_list gc; -+ struct ip_set_iptreemap_b *tree[256]; /* y.x.x.x */ -+#endif -+}; -+ -+struct ip_set_req_iptreemap_create { -+ unsigned int gc_interval; -+}; -+ -+struct ip_set_req_iptreemap { -+ ip_set_ip_t start; -+ ip_set_ip_t end; -+}; -+ -+#endif /* __IP_SET_IPTREEMAP_H */ ---- /dev/null -+++ b/include/linux/netfilter_ipv4/ip_set_jhash.h -@@ -0,0 +1,148 @@ -+#ifndef _LINUX_IPSET_JHASH_H -+#define _LINUX_IPSET_JHASH_H -+ -+/* This is a copy of linux/jhash.h but the types u32/u8 are changed -+ * to __u32/__u8 so that the header file can be included into -+ * userspace code as well. Jozsef Kadlecsik (kadlec@blackhole.kfki.hu) -+ */ -+ -+/* jhash.h: Jenkins hash support. -+ * -+ * Copyright (C) 1996 Bob Jenkins (bob_jenkins@burtleburtle.net) -+ * -+ * http://burtleburtle.net/bob/hash/ -+ * -+ * These are the credits from Bob's sources: -+ * -+ * lookup2.c, by Bob Jenkins, December 1996, Public Domain. -+ * hash(), hash2(), hash3, and mix() are externally useful functions. -+ * Routines to test the hash are included if SELF_TEST is defined. -+ * You can use this free for any purpose. It has no warranty. -+ * -+ * Copyright (C) 2003 David S. Miller (davem@redhat.com) -+ * -+ * I've modified Bob's hash to be useful in the Linux kernel, and -+ * any bugs present are surely my fault. -DaveM -+ */ -+ -+/* NOTE: Arguments are modified. */ -+#define __jhash_mix(a, b, c) \ -+{ \ -+ a -= b; a -= c; a ^= (c>>13); \ -+ b -= c; b -= a; b ^= (a<<8); \ -+ c -= a; c -= b; c ^= (b>>13); \ -+ a -= b; a -= c; a ^= (c>>12); \ -+ b -= c; b -= a; b ^= (a<<16); \ -+ c -= a; c -= b; c ^= (b>>5); \ -+ a -= b; a -= c; a ^= (c>>3); \ -+ b -= c; b -= a; b ^= (a<<10); \ -+ c -= a; c -= b; c ^= (b>>15); \ -+} -+ -+/* The golden ration: an arbitrary value */ -+#define JHASH_GOLDEN_RATIO 0x9e3779b9 -+ -+/* The most generic version, hashes an arbitrary sequence -+ * of bytes. No alignment or length assumptions are made about -+ * the input key. -+ */ -+static inline __u32 jhash(void *key, __u32 length, __u32 initval) -+{ -+ __u32 a, b, c, len; -+ __u8 *k = key; -+ -+ len = length; -+ a = b = JHASH_GOLDEN_RATIO; -+ c = initval; -+ -+ while (len >= 12) { -+ a += (k[0] +((__u32)k[1]<<8) +((__u32)k[2]<<16) +((__u32)k[3]<<24)); -+ b += (k[4] +((__u32)k[5]<<8) +((__u32)k[6]<<16) +((__u32)k[7]<<24)); -+ c += (k[8] +((__u32)k[9]<<8) +((__u32)k[10]<<16)+((__u32)k[11]<<24)); -+ -+ __jhash_mix(a,b,c); -+ -+ k += 12; -+ len -= 12; -+ } -+ -+ c += length; -+ switch (len) { -+ case 11: c += ((__u32)k[10]<<24); -+ case 10: c += ((__u32)k[9]<<16); -+ case 9 : c += ((__u32)k[8]<<8); -+ case 8 : b += ((__u32)k[7]<<24); -+ case 7 : b += ((__u32)k[6]<<16); -+ case 6 : b += ((__u32)k[5]<<8); -+ case 5 : b += k[4]; -+ case 4 : a += ((__u32)k[3]<<24); -+ case 3 : a += ((__u32)k[2]<<16); -+ case 2 : a += ((__u32)k[1]<<8); -+ case 1 : a += k[0]; -+ }; -+ -+ __jhash_mix(a,b,c); -+ -+ return c; -+} -+ -+/* A special optimized version that handles 1 or more of __u32s. -+ * The length parameter here is the number of __u32s in the key. -+ */ -+static inline __u32 jhash2(__u32 *k, __u32 length, __u32 initval) -+{ -+ __u32 a, b, c, len; -+ -+ a = b = JHASH_GOLDEN_RATIO; -+ c = initval; -+ len = length; -+ -+ while (len >= 3) { -+ a += k[0]; -+ b += k[1]; -+ c += k[2]; -+ __jhash_mix(a, b, c); -+ k += 3; len -= 3; -+ } -+ -+ c += length * 4; -+ -+ switch (len) { -+ case 2 : b += k[1]; -+ case 1 : a += k[0]; -+ }; -+ -+ __jhash_mix(a,b,c); -+ -+ return c; -+} -+ -+ -+/* A special ultra-optimized versions that knows they are hashing exactly -+ * 3, 2 or 1 word(s). -+ * -+ * NOTE: In partilar the "c += length; __jhash_mix(a,b,c);" normally -+ * done at the end is not done here. -+ */ -+static inline __u32 jhash_3words(__u32 a, __u32 b, __u32 c, __u32 initval) -+{ -+ a += JHASH_GOLDEN_RATIO; -+ b += JHASH_GOLDEN_RATIO; -+ c += initval; -+ -+ __jhash_mix(a, b, c); -+ -+ return c; -+} -+ -+static inline __u32 jhash_2words(__u32 a, __u32 b, __u32 initval) -+{ -+ return jhash_3words(a, b, 0, initval); -+} -+ -+static inline __u32 jhash_1word(__u32 a, __u32 initval) -+{ -+ return jhash_3words(a, 0, 0, initval); -+} -+ -+#endif /* _LINUX_IPSET_JHASH_H */ ---- /dev/null -+++ b/include/linux/netfilter_ipv4/ip_set_macipmap.h -@@ -0,0 +1,38 @@ -+#ifndef __IP_SET_MACIPMAP_H -+#define __IP_SET_MACIPMAP_H -+ -+#include <linux/netfilter_ipv4/ip_set.h> -+ -+#define SETTYPE_NAME "macipmap" -+#define MAX_RANGE 0x0000FFFF -+ -+/* general flags */ -+#define IPSET_MACIP_MATCHUNSET 1 -+ -+/* per ip flags */ -+#define IPSET_MACIP_ISSET 1 -+ -+struct ip_set_macipmap { -+ void *members; /* the macipmap proper */ -+ ip_set_ip_t first_ip; /* host byte order, included in range */ -+ ip_set_ip_t last_ip; /* host byte order, included in range */ -+ u_int32_t flags; -+}; -+ -+struct ip_set_req_macipmap_create { -+ ip_set_ip_t from; -+ ip_set_ip_t to; -+ u_int32_t flags; -+}; -+ -+struct ip_set_req_macipmap { -+ ip_set_ip_t ip; -+ unsigned char ethernet[ETH_ALEN]; -+}; -+ -+struct ip_set_macip { -+ unsigned short flags; -+ unsigned char ethernet[ETH_ALEN]; -+}; -+ -+#endif /* __IP_SET_MACIPMAP_H */ ---- /dev/null -+++ b/include/linux/netfilter_ipv4/ip_set_malloc.h -@@ -0,0 +1,116 @@ -+#ifndef _IP_SET_MALLOC_H -+#define _IP_SET_MALLOC_H -+ -+#ifdef __KERNEL__ -+ -+/* Memory allocation and deallocation */ -+static size_t max_malloc_size = 0; -+ -+static inline void init_max_malloc_size(void) -+{ -+#define CACHE(x) max_malloc_size = x; -+#include <linux/kmalloc_sizes.h> -+#undef CACHE -+} -+ -+static inline void * ip_set_malloc(size_t bytes) -+{ -+ if (bytes > max_malloc_size) -+ return vmalloc(bytes); -+ else -+ return kmalloc(bytes, GFP_KERNEL); -+} -+ -+static inline void ip_set_free(void * data, size_t bytes) -+{ -+ if (bytes > max_malloc_size) -+ vfree(data); -+ else -+ kfree(data); -+} -+ -+struct harray { -+ size_t max_elements; -+ void *arrays[0]; -+}; -+ -+static inline void * -+harray_malloc(size_t hashsize, size_t typesize, int flags) -+{ -+ struct harray *harray; -+ size_t max_elements, size, i, j; -+ -+ if (!max_malloc_size) -+ init_max_malloc_size(); -+ -+ if (typesize > max_malloc_size) -+ return NULL; -+ -+ max_elements = max_malloc_size/typesize; -+ size = hashsize/max_elements; -+ if (hashsize % max_elements) -+ size++; -+ -+ /* Last pointer signals end of arrays */ -+ harray = kmalloc(sizeof(struct harray) + (size + 1) * sizeof(void *), -+ flags); -+ -+ if (!harray) -+ return NULL; -+ -+ for (i = 0; i < size - 1; i++) { -+ harray->arrays[i] = kmalloc(max_elements * typesize, flags); -+ if (!harray->arrays[i]) -+ goto undo; -+ memset(harray->arrays[i], 0, max_elements * typesize); -+ } -+ harray->arrays[i] = kmalloc((hashsize - i * max_elements) * typesize, -+ flags); -+ if (!harray->arrays[i]) -+ goto undo; -+ memset(harray->arrays[i], 0, (hashsize - i * max_elements) * typesize); -+ -+ harray->max_elements = max_elements; -+ harray->arrays[size] = NULL; -+ -+ return (void *)harray; -+ -+ undo: -+ for (j = 0; j < i; j++) { -+ kfree(harray->arrays[j]); -+ } -+ kfree(harray); -+ return NULL; -+} -+ -+static inline void harray_free(void *h) -+{ -+ struct harray *harray = (struct harray *) h; -+ size_t i; -+ -+ for (i = 0; harray->arrays[i] != NULL; i++) -+ kfree(harray->arrays[i]); -+ kfree(harray); -+} -+ -+static inline void harray_flush(void *h, size_t hashsize, size_t typesize) -+{ -+ struct harray *harray = (struct harray *) h; -+ size_t i; -+ -+ for (i = 0; harray->arrays[i+1] != NULL; i++) -+ memset(harray->arrays[i], 0, harray->max_elements * typesize); -+ memset(harray->arrays[i], 0, -+ (hashsize - i * harray->max_elements) * typesize); -+} -+ -+#define HARRAY_ELEM(h, type, which) \ -+({ \ -+ struct harray *__h = (struct harray *)(h); \ -+ ((type)((__h)->arrays[(which)/(__h)->max_elements]) \ -+ + (which)%(__h)->max_elements); \ -+}) -+ -+#endif /* __KERNEL__ */ -+ -+#endif /*_IP_SET_MALLOC_H*/ ---- /dev/null -+++ b/include/linux/netfilter_ipv4/ip_set_nethash.h -@@ -0,0 +1,55 @@ -+#ifndef __IP_SET_NETHASH_H -+#define __IP_SET_NETHASH_H -+ -+#include <linux/netfilter_ipv4/ip_set.h> -+ -+#define SETTYPE_NAME "nethash" -+#define MAX_RANGE 0x0000FFFF -+ -+struct ip_set_nethash { -+ ip_set_ip_t *members; /* the nethash proper */ -+ uint32_t elements; /* number of elements */ -+ uint32_t hashsize; /* hash size */ -+ uint16_t probes; /* max number of probes */ -+ uint16_t resize; /* resize factor in percent */ -+ unsigned char cidr[30]; /* CIDR sizes */ -+ void *initval[0]; /* initvals for jhash_1word */ -+}; -+ -+struct ip_set_req_nethash_create { -+ uint32_t hashsize; -+ uint16_t probes; -+ uint16_t resize; -+}; -+ -+struct ip_set_req_nethash { -+ ip_set_ip_t ip; -+ unsigned char cidr; -+}; -+ -+static unsigned char shifts[] = {255, 253, 249, 241, 225, 193, 129, 1}; -+ -+static inline ip_set_ip_t -+pack(ip_set_ip_t ip, unsigned char cidr) -+{ -+ ip_set_ip_t addr, *paddr = &addr; -+ unsigned char n, t, *a; -+ -+ addr = htonl(ip & (0xFFFFFFFF << (32 - (cidr)))); -+#ifdef __KERNEL__ -+ DP("ip:%u.%u.%u.%u/%u", NIPQUAD(addr), cidr); -+#endif -+ n = cidr / 8; -+ t = cidr % 8; -+ a = &((unsigned char *)paddr)[n]; -+ *a = *a /(1 << (8 - t)) + shifts[t]; -+#ifdef __KERNEL__ -+ DP("n: %u, t: %u, a: %u", n, t, *a); -+ DP("ip:%u.%u.%u.%u/%u, %u.%u.%u.%u", -+ HIPQUAD(ip), cidr, NIPQUAD(addr)); -+#endif -+ -+ return ntohl(addr); -+} -+ -+#endif /* __IP_SET_NETHASH_H */ ---- /dev/null -+++ b/include/linux/netfilter_ipv4/ip_set_portmap.h -@@ -0,0 +1,25 @@ -+#ifndef __IP_SET_PORTMAP_H -+#define __IP_SET_PORTMAP_H -+ -+#include <linux/netfilter_ipv4/ip_set.h> -+ -+#define SETTYPE_NAME "portmap" -+#define MAX_RANGE 0x0000FFFF -+#define INVALID_PORT (MAX_RANGE + 1) -+ -+struct ip_set_portmap { -+ void *members; /* the portmap proper */ -+ ip_set_ip_t first_port; /* host byte order, included in range */ -+ ip_set_ip_t last_port; /* host byte order, included in range */ -+}; -+ -+struct ip_set_req_portmap_create { -+ ip_set_ip_t from; -+ ip_set_ip_t to; -+}; -+ -+struct ip_set_req_portmap { -+ ip_set_ip_t port; -+}; -+ -+#endif /* __IP_SET_PORTMAP_H */ ---- /dev/null -+++ b/include/linux/netfilter_ipv4/ipt_set.h -@@ -0,0 +1,21 @@ -+#ifndef _IPT_SET_H -+#define _IPT_SET_H -+ -+#include <linux/netfilter_ipv4/ip_set.h> -+ -+struct ipt_set_info { -+ ip_set_id_t index; -+ u_int32_t flags[IP_SET_MAX_BINDINGS + 1]; -+}; -+ -+/* match info */ -+struct ipt_set_info_match { -+ struct ipt_set_info match_set; -+}; -+ -+struct ipt_set_info_target { -+ struct ipt_set_info add_set; -+ struct ipt_set_info del_set; -+}; -+ -+#endif /*_IPT_SET_H*/ ---- /dev/null -+++ b/net/ipv4/netfilter/ip_set.c -@@ -0,0 +1,2003 @@ -+/* Copyright (C) 2000-2002 Joakim Axelsson <gozem@linux.nu> -+ * Patrick Schaaf <bof@bof.de> -+ * Copyright (C) 2003-2004 Jozsef Kadlecsik <kadlec@blackhole.kfki.hu> -+ * -+ * This program is free software; you can redistribute it and/or modify -+ * it under the terms of the GNU General Public License version 2 as -+ * published by the Free Software Foundation. -+ */ -+ -+/* Kernel module for IP set management */ -+ -+#include <linux/version.h> -+#if LINUX_VERSION_CODE < KERNEL_VERSION(2,6,19) -+#include <linux/config.h> -+#endif -+#include <linux/module.h> -+#include <linux/moduleparam.h> -+#include <linux/kmod.h> -+#include <linux/ip.h> -+#include <linux/skbuff.h> -+#include <linux/random.h> -+#include <linux/jhash.h> -+#include <linux/netfilter_ipv4/ip_tables.h> -+#include <linux/errno.h> -+#include <asm/uaccess.h> -+#include <asm/bitops.h> -+#include <asm/semaphore.h> -+#include <linux/spinlock.h> -+#include <linux/vmalloc.h> -+ -+#define ASSERT_READ_LOCK(x) -+#define ASSERT_WRITE_LOCK(x) -+#include <linux/netfilter_ipv4/ip_set.h> -+ -+static struct list_head set_type_list; /* all registered sets */ -+static struct ip_set **ip_set_list; /* all individual sets */ -+static DEFINE_RWLOCK(ip_set_lock); /* protects the lists and the hash */ -+static DECLARE_MUTEX(ip_set_app_mutex); /* serializes user access */ -+static ip_set_id_t ip_set_max = CONFIG_IP_NF_SET_MAX; -+static ip_set_id_t ip_set_bindings_hash_size = CONFIG_IP_NF_SET_HASHSIZE; -+static struct list_head *ip_set_hash; /* hash of bindings */ -+static unsigned int ip_set_hash_random; /* random seed */ -+ -+/* -+ * Sets are identified either by the index in ip_set_list or by id. -+ * The id never changes and is used to find a key in the hash. -+ * The index may change by swapping and used at all other places -+ * (set/SET netfilter modules, binding value, etc.) -+ * -+ * Userspace requests are serialized by ip_set_mutex and sets can -+ * be deleted only from userspace. Therefore ip_set_list locking -+ * must obey the following rules: -+ * -+ * - kernel requests: read and write locking mandatory -+ * - user requests: read locking optional, write locking mandatory -+ */ -+ -+static inline void -+__ip_set_get(ip_set_id_t index) -+{ -+ atomic_inc(&ip_set_list[index]->ref); -+} -+ -+static inline void -+__ip_set_put(ip_set_id_t index) -+{ -+ atomic_dec(&ip_set_list[index]->ref); -+} -+ -+/* -+ * Binding routines -+ */ -+ -+static inline struct ip_set_hash * -+__ip_set_find(u_int32_t key, ip_set_id_t id, ip_set_ip_t ip) -+{ -+ struct ip_set_hash *set_hash; -+ -+ list_for_each_entry(set_hash, &ip_set_hash[key], list) -+ if (set_hash->id == id && set_hash->ip == ip) -+ return set_hash; -+ -+ return NULL; -+} -+ -+static ip_set_id_t -+ip_set_find_in_hash(ip_set_id_t id, ip_set_ip_t ip) -+{ -+ u_int32_t key = jhash_2words(id, ip, ip_set_hash_random) -+ % ip_set_bindings_hash_size; -+ struct ip_set_hash *set_hash; -+ -+ ASSERT_READ_LOCK(&ip_set_lock); -+ IP_SET_ASSERT(ip_set_list[id]); -+ DP("set: %s, ip: %u.%u.%u.%u", ip_set_list[id]->name, HIPQUAD(ip)); -+ -+ set_hash = __ip_set_find(key, id, ip); -+ -+ DP("set: %s, ip: %u.%u.%u.%u, binding: %s", ip_set_list[id]->name, -+ HIPQUAD(ip), -+ set_hash != NULL ? ip_set_list[set_hash->binding]->name : ""); -+ -+ return (set_hash != NULL ? set_hash->binding : IP_SET_INVALID_ID); -+} -+ -+static inline void -+__set_hash_del(struct ip_set_hash *set_hash) -+{ -+ ASSERT_WRITE_LOCK(&ip_set_lock); -+ IP_SET_ASSERT(ip_set_list[set_hash->binding]); -+ -+ __ip_set_put(set_hash->binding); -+ list_del(&set_hash->list); -+ kfree(set_hash); -+} -+ -+static int -+ip_set_hash_del(ip_set_id_t id, ip_set_ip_t ip) -+{ -+ u_int32_t key = jhash_2words(id, ip, ip_set_hash_random) -+ % ip_set_bindings_hash_size; -+ struct ip_set_hash *set_hash; -+ -+ IP_SET_ASSERT(ip_set_list[id]); -+ DP("set: %s, ip: %u.%u.%u.%u", ip_set_list[id]->name, HIPQUAD(ip)); -+ write_lock_bh(&ip_set_lock); -+ set_hash = __ip_set_find(key, id, ip); -+ DP("set: %s, ip: %u.%u.%u.%u, binding: %s", ip_set_list[id]->name, -+ HIPQUAD(ip), -+ set_hash != NULL ? ip_set_list[set_hash->binding]->name : ""); -+ -+ if (set_hash != NULL) -+ __set_hash_del(set_hash); -+ write_unlock_bh(&ip_set_lock); -+ return 0; -+} -+ -+static int -+ip_set_hash_add(ip_set_id_t id, ip_set_ip_t ip, ip_set_id_t binding) -+{ -+ u_int32_t key = jhash_2words(id, ip, ip_set_hash_random) -+ % ip_set_bindings_hash_size; -+ struct ip_set_hash *set_hash; -+ int ret = 0; -+ -+ IP_SET_ASSERT(ip_set_list[id]); -+ IP_SET_ASSERT(ip_set_list[binding]); -+ DP("set: %s, ip: %u.%u.%u.%u, binding: %s", ip_set_list[id]->name, -+ HIPQUAD(ip), ip_set_list[binding]->name); -+ write_lock_bh(&ip_set_lock); -+ set_hash = __ip_set_find(key, id, ip); -+ if (!set_hash) { -+ set_hash = kmalloc(sizeof(struct ip_set_hash), GFP_ATOMIC); -+ if (!set_hash) { -+ ret = -ENOMEM; -+ goto unlock; -+ } -+ INIT_LIST_HEAD(&set_hash->list); -+ set_hash->id = id; -+ set_hash->ip = ip; -+ list_add(&set_hash->list, &ip_set_hash[key]); -+ } else { -+ IP_SET_ASSERT(ip_set_list[set_hash->binding]); -+ DP("overwrite binding: %s", -+ ip_set_list[set_hash->binding]->name); -+ __ip_set_put(set_hash->binding); -+ } -+ set_hash->binding = binding; -+ __ip_set_get(set_hash->binding); -+ DP("stored: key %u, id %u (%s), ip %u.%u.%u.%u, binding %u (%s)", -+ key, id, ip_set_list[id]->name, -+ HIPQUAD(ip), binding, ip_set_list[binding]->name); -+ unlock: -+ write_unlock_bh(&ip_set_lock); -+ return ret; -+} -+ -+#define FOREACH_HASH_DO(fn, args...) \ -+({ \ -+ ip_set_id_t __key; \ -+ struct ip_set_hash *__set_hash; \ -+ \ -+ for (__key = 0; __key < ip_set_bindings_hash_size; __key++) { \ -+ list_for_each_entry(__set_hash, &ip_set_hash[__key], list) \ -+ fn(__set_hash , ## args); \ -+ } \ -+}) -+ -+#define FOREACH_HASH_RW_DO(fn, args...) \ -+({ \ -+ ip_set_id_t __key; \ -+ struct ip_set_hash *__set_hash, *__n; \ -+ \ -+ ASSERT_WRITE_LOCK(&ip_set_lock); \ -+ for (__key = 0; __key < ip_set_bindings_hash_size; __key++) { \ -+ list_for_each_entry_safe(__set_hash, __n, &ip_set_hash[__key], list)\ -+ fn(__set_hash , ## args); \ -+ } \ -+}) -+ -+/* Add, del and test set entries from kernel */ -+ -+#define follow_bindings(index, set, ip) \ -+((index = ip_set_find_in_hash((set)->id, ip)) != IP_SET_INVALID_ID \ -+ || (index = (set)->binding) != IP_SET_INVALID_ID) -+ -+int -+ip_set_testip_kernel(ip_set_id_t index, -+ const struct sk_buff *skb, -+ const u_int32_t *flags) -+{ -+ struct ip_set *set; -+ ip_set_ip_t ip; -+ int res; -+ unsigned char i = 0; -+ -+ IP_SET_ASSERT(flags[i]); -+ read_lock_bh(&ip_set_lock); -+ do { -+ set = ip_set_list[index]; -+ IP_SET_ASSERT(set); -+ DP("set %s, index %u", set->name, index); -+ read_lock_bh(&set->lock); -+ res = set->type->testip_kernel(set, skb, &ip, flags, i++); -+ read_unlock_bh(&set->lock); -+ i += !!(set->type->features & IPSET_DATA_DOUBLE); -+ } while (res > 0 -+ && flags[i] -+ && follow_bindings(index, set, ip)); -+ read_unlock_bh(&ip_set_lock); -+ -+ return res; -+} -+ -+void -+ip_set_addip_kernel(ip_set_id_t index, -+ const struct sk_buff *skb, -+ const u_int32_t *flags) -+{ -+ struct ip_set *set; -+ ip_set_ip_t ip; -+ int res; -+ unsigned char i = 0; -+ -+ IP_SET_ASSERT(flags[i]); -+ retry: -+ read_lock_bh(&ip_set_lock); -+ do { -+ set = ip_set_list[index]; -+ IP_SET_ASSERT(set); -+ DP("set %s, index %u", set->name, index); -+ write_lock_bh(&set->lock); -+ res = set->type->addip_kernel(set, skb, &ip, flags, i++); -+ write_unlock_bh(&set->lock); -+ i += !!(set->type->features & IPSET_DATA_DOUBLE); -+ } while ((res == 0 || res == -EEXIST) -+ && flags[i] -+ && follow_bindings(index, set, ip)); -+ read_unlock_bh(&ip_set_lock); -+ -+ if (res == -EAGAIN -+ && set->type->retry -+ && (res = set->type->retry(set)) == 0) -+ goto retry; -+} -+ -+void -+ip_set_delip_kernel(ip_set_id_t index, -+ const struct sk_buff *skb, -+ const u_int32_t *flags) -+{ -+ struct ip_set *set; -+ ip_set_ip_t ip; -+ int res; -+ unsigned char i = 0; -+ -+ IP_SET_ASSERT(flags[i]); -+ read_lock_bh(&ip_set_lock); -+ do { -+ set = ip_set_list[index]; -+ IP_SET_ASSERT(set); -+ DP("set %s, index %u", set->name, index); -+ write_lock_bh(&set->lock); -+ res = set->type->delip_kernel(set, skb, &ip, flags, i++); -+ write_unlock_bh(&set->lock); -+ i += !!(set->type->features & IPSET_DATA_DOUBLE); -+ } while ((res == 0 || res == -EEXIST) -+ && flags[i] -+ && follow_bindings(index, set, ip)); -+ read_unlock_bh(&ip_set_lock); -+} -+ -+/* Register and deregister settype */ -+ -+static inline struct ip_set_type * -+find_set_type(const char *name) -+{ -+ struct ip_set_type *set_type; -+ -+ list_for_each_entry(set_type, &set_type_list, list) -+ if (!strncmp(set_type->typename, name, IP_SET_MAXNAMELEN - 1)) -+ return set_type; -+ return NULL; -+} -+ -+int -+ip_set_register_set_type(struct ip_set_type *set_type) -+{ -+ int ret = 0; -+ -+ if (set_type->protocol_version != IP_SET_PROTOCOL_VERSION) { -+ ip_set_printk("'%s' uses wrong protocol version %u (want %u)", -+ set_type->typename, -+ set_type->protocol_version, -+ IP_SET_PROTOCOL_VERSION); -+ return -EINVAL; -+ } -+ -+ write_lock_bh(&ip_set_lock); -+ if (find_set_type(set_type->typename)) { -+ /* Duplicate! */ -+ ip_set_printk("'%s' already registered!", -+ set_type->typename); -+ ret = -EINVAL; -+ goto unlock; -+ } -+ if (!try_module_get(THIS_MODULE)) { -+ ret = -EFAULT; -+ goto unlock; -+ } -+ list_add(&set_type->list, &set_type_list); -+ DP("'%s' registered.", set_type->typename); -+ unlock: -+ write_unlock_bh(&ip_set_lock); -+ return ret; -+} -+ -+void -+ip_set_unregister_set_type(struct ip_set_type *set_type) -+{ -+ write_lock_bh(&ip_set_lock); -+ if (!find_set_type(set_type->typename)) { -+ ip_set_printk("'%s' not registered?", -+ set_type->typename); -+ goto unlock; -+ } -+ list_del(&set_type->list); -+ module_put(THIS_MODULE); -+ DP("'%s' unregistered.", set_type->typename); -+ unlock: -+ write_unlock_bh(&ip_set_lock); -+ -+} -+ -+/* -+ * Userspace routines -+ */ -+ -+/* -+ * Find set by name, reference it once. The reference makes sure the -+ * thing pointed to, does not go away under our feet. Drop the reference -+ * later, using ip_set_put(). -+ */ -+ip_set_id_t -+ip_set_get_byname(const char *name) -+{ -+ ip_set_id_t i, index = IP_SET_INVALID_ID; -+ -+ down(&ip_set_app_mutex); -+ for (i = 0; i < ip_set_max; i++) { -+ if (ip_set_list[i] != NULL -+ && strcmp(ip_set_list[i]->name, name) == 0) { -+ __ip_set_get(i); -+ index = i; -+ break; -+ } -+ } -+ up(&ip_set_app_mutex); -+ return index; -+} -+ -+/* -+ * Find set by index, reference it once. The reference makes sure the -+ * thing pointed to, does not go away under our feet. Drop the reference -+ * later, using ip_set_put(). -+ */ -+ip_set_id_t -+ip_set_get_byindex(ip_set_id_t index) -+{ -+ down(&ip_set_app_mutex); -+ -+ if (index >= ip_set_max) -+ return IP_SET_INVALID_ID; -+ -+ if (ip_set_list[index]) -+ __ip_set_get(index); -+ else -+ index = IP_SET_INVALID_ID; -+ -+ up(&ip_set_app_mutex); -+ return index; -+} -+ -+/* -+ * If the given set pointer points to a valid set, decrement -+ * reference count by 1. The caller shall not assume the index -+ * to be valid, after calling this function. -+ */ -+void ip_set_put(ip_set_id_t index) -+{ -+ down(&ip_set_app_mutex); -+ if (ip_set_list[index]) -+ __ip_set_put(index); -+ up(&ip_set_app_mutex); -+} -+ -+/* Find a set by name or index */ -+static ip_set_id_t -+ip_set_find_byname(const char *name) -+{ -+ ip_set_id_t i, index = IP_SET_INVALID_ID; -+ -+ for (i = 0; i < ip_set_max; i++) { -+ if (ip_set_list[i] != NULL -+ && strcmp(ip_set_list[i]->name, name) == 0) { -+ index = i; -+ break; -+ } -+ } -+ return index; -+} -+ -+static ip_set_id_t -+ip_set_find_byindex(ip_set_id_t index) -+{ -+ if (index >= ip_set_max || ip_set_list[index] == NULL) -+ index = IP_SET_INVALID_ID; -+ -+ return index; -+} -+ -+/* -+ * Add, del, test, bind and unbind -+ */ -+ -+static inline int -+__ip_set_testip(struct ip_set *set, -+ const void *data, -+ size_t size, -+ ip_set_ip_t *ip) -+{ -+ int res; -+ -+ read_lock_bh(&set->lock); -+ res = set->type->testip(set, data, size, ip); -+ read_unlock_bh(&set->lock); -+ -+ return res; -+} -+ -+static int -+__ip_set_addip(ip_set_id_t index, -+ const void *data, -+ size_t size) -+{ -+ struct ip_set *set = ip_set_list[index]; -+ ip_set_ip_t ip; -+ int res; -+ -+ IP_SET_ASSERT(set); -+ do { -+ write_lock_bh(&set->lock); -+ res = set->type->addip(set, data, size, &ip); -+ write_unlock_bh(&set->lock); -+ } while (res == -EAGAIN -+ && set->type->retry -+ && (res = set->type->retry(set)) == 0); -+ -+ return res; -+} -+ -+static int -+ip_set_addip(ip_set_id_t index, -+ const void *data, -+ size_t size) -+{ -+ -+ return __ip_set_addip(index, -+ data + sizeof(struct ip_set_req_adt), -+ size - sizeof(struct ip_set_req_adt)); -+} -+ -+static int -+ip_set_delip(ip_set_id_t index, -+ const void *data, -+ size_t size) -+{ -+ struct ip_set *set = ip_set_list[index]; -+ ip_set_ip_t ip; -+ int res; -+ -+ IP_SET_ASSERT(set); -+ write_lock_bh(&set->lock); -+ res = set->type->delip(set, -+ data + sizeof(struct ip_set_req_adt), -+ size - sizeof(struct ip_set_req_adt), -+ &ip); -+ write_unlock_bh(&set->lock); -+ -+ return res; -+} -+ -+static int -+ip_set_testip(ip_set_id_t index, -+ const void *data, -+ size_t size) -+{ -+ struct ip_set *set = ip_set_list[index]; -+ ip_set_ip_t ip; -+ int res; -+ -+ IP_SET_ASSERT(set); -+ res = __ip_set_testip(set, -+ data + sizeof(struct ip_set_req_adt), -+ size - sizeof(struct ip_set_req_adt), -+ &ip); -+ -+ return (res > 0 ? -EEXIST : res); -+} -+ -+static int -+ip_set_bindip(ip_set_id_t index, -+ const void *data, -+ size_t size) -+{ -+ struct ip_set *set = ip_set_list[index]; -+ struct ip_set_req_bind *req_bind; -+ ip_set_id_t binding; -+ ip_set_ip_t ip; -+ int res; -+ -+ IP_SET_ASSERT(set); -+ if (size < sizeof(struct ip_set_req_bind)) -+ return -EINVAL; -+ -+ req_bind = (struct ip_set_req_bind *) data; -+ req_bind->binding[IP_SET_MAXNAMELEN - 1] = '\0'; -+ -+ if (strcmp(req_bind->binding, IPSET_TOKEN_DEFAULT) == 0) { -+ /* Default binding of a set */ -+ char *binding_name; -+ -+ if (size != sizeof(struct ip_set_req_bind) + IP_SET_MAXNAMELEN) -+ return -EINVAL; -+ -+ binding_name = (char *)(data + sizeof(struct ip_set_req_bind)); -+ binding_name[IP_SET_MAXNAMELEN - 1] = '\0'; -+ -+ binding = ip_set_find_byname(binding_name); -+ if (binding == IP_SET_INVALID_ID) -+ return -ENOENT; -+ -+ write_lock_bh(&ip_set_lock); -+ /* Sets as binding values are referenced */ -+ if (set->binding != IP_SET_INVALID_ID) -+ __ip_set_put(set->binding); -+ set->binding = binding; -+ __ip_set_get(set->binding); -+ write_unlock_bh(&ip_set_lock); -+ -+ return 0; -+ } -+ binding = ip_set_find_byname(req_bind->binding); -+ if (binding == IP_SET_INVALID_ID) -+ return -ENOENT; -+ -+ res = __ip_set_testip(set, -+ data + sizeof(struct ip_set_req_bind), -+ size - sizeof(struct ip_set_req_bind), -+ &ip); -+ DP("set %s, ip: %u.%u.%u.%u, binding %s", -+ set->name, HIPQUAD(ip), ip_set_list[binding]->name); -+ -+ if (res >= 0) -+ res = ip_set_hash_add(set->id, ip, binding); -+ -+ return res; -+} -+ -+#define FOREACH_SET_DO(fn, args...) \ -+({ \ -+ ip_set_id_t __i; \ -+ struct ip_set *__set; \ -+ \ -+ for (__i = 0; __i < ip_set_max; __i++) { \ -+ __set = ip_set_list[__i]; \ -+ if (__set != NULL) \ -+ fn(__set , ##args); \ -+ } \ -+}) -+ -+static inline void -+__set_hash_del_byid(struct ip_set_hash *set_hash, ip_set_id_t id) -+{ -+ if (set_hash->id == id) -+ __set_hash_del(set_hash); -+} -+ -+static inline void -+__unbind_default(struct ip_set *set) -+{ -+ if (set->binding != IP_SET_INVALID_ID) { -+ /* Sets as binding values are referenced */ -+ __ip_set_put(set->binding); -+ set->binding = IP_SET_INVALID_ID; -+ } -+} -+ -+static int -+ip_set_unbindip(ip_set_id_t index, -+ const void *data, -+ size_t size) -+{ -+ struct ip_set *set; -+ struct ip_set_req_bind *req_bind; -+ ip_set_ip_t ip; -+ int res; -+ -+ DP(""); -+ if (size < sizeof(struct ip_set_req_bind)) -+ return -EINVAL; -+ -+ req_bind = (struct ip_set_req_bind *) data; -+ req_bind->binding[IP_SET_MAXNAMELEN - 1] = '\0'; -+ -+ DP("%u %s", index, req_bind->binding); -+ if (index == IP_SET_INVALID_ID) { -+ /* unbind :all: */ -+ if (strcmp(req_bind->binding, IPSET_TOKEN_DEFAULT) == 0) { -+ /* Default binding of sets */ -+ write_lock_bh(&ip_set_lock); -+ FOREACH_SET_DO(__unbind_default); -+ write_unlock_bh(&ip_set_lock); -+ return 0; -+ } else if (strcmp(req_bind->binding, IPSET_TOKEN_ALL) == 0) { -+ /* Flush all bindings of all sets*/ -+ write_lock_bh(&ip_set_lock); -+ FOREACH_HASH_RW_DO(__set_hash_del); -+ write_unlock_bh(&ip_set_lock); -+ return 0; -+ } -+ DP("unreachable reached!"); -+ return -EINVAL; -+ } -+ -+ set = ip_set_list[index]; -+ IP_SET_ASSERT(set); -+ if (strcmp(req_bind->binding, IPSET_TOKEN_DEFAULT) == 0) { -+ /* Default binding of set */ -+ ip_set_id_t binding = ip_set_find_byindex(set->binding); -+ -+ if (binding == IP_SET_INVALID_ID) -+ return -ENOENT; -+ -+ write_lock_bh(&ip_set_lock); -+ /* Sets in hash values are referenced */ -+ __ip_set_put(set->binding); -+ set->binding = IP_SET_INVALID_ID; -+ write_unlock_bh(&ip_set_lock); -+ -+ return 0; -+ } else if (strcmp(req_bind->binding, IPSET_TOKEN_ALL) == 0) { -+ /* Flush all bindings */ -+ -+ write_lock_bh(&ip_set_lock); -+ FOREACH_HASH_RW_DO(__set_hash_del_byid, set->id); -+ write_unlock_bh(&ip_set_lock); -+ return 0; -+ } -+ -+ res = __ip_set_testip(set, -+ data + sizeof(struct ip_set_req_bind), -+ size - sizeof(struct ip_set_req_bind), -+ &ip); -+ -+ DP("set %s, ip: %u.%u.%u.%u", set->name, HIPQUAD(ip)); -+ if (res >= 0) -+ res = ip_set_hash_del(set->id, ip); -+ -+ return res; -+} -+ -+static int -+ip_set_testbind(ip_set_id_t index, -+ const void *data, -+ size_t size) -+{ -+ struct ip_set *set = ip_set_list[index]; -+ struct ip_set_req_bind *req_bind; -+ ip_set_id_t binding; -+ ip_set_ip_t ip; -+ int res; -+ -+ IP_SET_ASSERT(set); -+ if (size < sizeof(struct ip_set_req_bind)) -+ return -EINVAL; -+ -+ req_bind = (struct ip_set_req_bind *) data; -+ req_bind->binding[IP_SET_MAXNAMELEN - 1] = '\0'; -+ -+ if (strcmp(req_bind->binding, IPSET_TOKEN_DEFAULT) == 0) { -+ /* Default binding of set */ -+ char *binding_name; -+ -+ if (size != sizeof(struct ip_set_req_bind) + IP_SET_MAXNAMELEN) -+ return -EINVAL; -+ -+ binding_name = (char *)(data + sizeof(struct ip_set_req_bind)); -+ binding_name[IP_SET_MAXNAMELEN - 1] = '\0'; -+ -+ binding = ip_set_find_byname(binding_name); -+ if (binding == IP_SET_INVALID_ID) -+ return -ENOENT; -+ -+ res = (set->binding == binding) ? -EEXIST : 0; -+ -+ return res; -+ } -+ binding = ip_set_find_byname(req_bind->binding); -+ if (binding == IP_SET_INVALID_ID) -+ return -ENOENT; -+ -+ -+ res = __ip_set_testip(set, -+ data + sizeof(struct ip_set_req_bind), -+ size - sizeof(struct ip_set_req_bind), -+ &ip); -+ DP("set %s, ip: %u.%u.%u.%u, binding %s", -+ set->name, HIPQUAD(ip), ip_set_list[binding]->name); -+ -+ if (res >= 0) -+ res = (ip_set_find_in_hash(set->id, ip) == binding) -+ ? -EEXIST : 0; -+ -+ return res; -+} -+ -+static struct ip_set_type * -+find_set_type_rlock(const char *typename) -+{ -+ struct ip_set_type *type; -+ -+ read_lock_bh(&ip_set_lock); -+ type = find_set_type(typename); -+ if (type == NULL) -+ read_unlock_bh(&ip_set_lock); -+ -+ return type; -+} -+ -+static int -+find_free_id(const char *name, -+ ip_set_id_t *index, -+ ip_set_id_t *id) -+{ -+ ip_set_id_t i; -+ -+ *id = IP_SET_INVALID_ID; -+ for (i = 0; i < ip_set_max; i++) { -+ if (ip_set_list[i] == NULL) { -+ if (*id == IP_SET_INVALID_ID) -+ *id = *index = i; -+ } else if (strcmp(name, ip_set_list[i]->name) == 0) -+ /* Name clash */ -+ return -EEXIST; -+ } -+ if (*id == IP_SET_INVALID_ID) -+ /* No free slot remained */ -+ return -ERANGE; -+ /* Check that index is usable as id (swapping) */ -+ check: -+ for (i = 0; i < ip_set_max; i++) { -+ if (ip_set_list[i] != NULL -+ && ip_set_list[i]->id == *id) { -+ *id = i; -+ goto check; -+ } -+ } -+ return 0; -+} -+ -+/* -+ * Create a set -+ */ -+static int -+ip_set_create(const char *name, -+ const char *typename, -+ ip_set_id_t restore, -+ const void *data, -+ size_t size) -+{ -+ struct ip_set *set; -+ ip_set_id_t index = 0, id; -+ int res = 0; -+ -+ DP("setname: %s, typename: %s, id: %u", name, typename, restore); -+ /* -+ * First, and without any locks, allocate and initialize -+ * a normal base set structure. -+ */ -+ set = kmalloc(sizeof(struct ip_set), GFP_KERNEL); -+ if (!set) -+ return -ENOMEM; -+ set->lock = RW_LOCK_UNLOCKED; -+ strncpy(set->name, name, IP_SET_MAXNAMELEN); -+ set->binding = IP_SET_INVALID_ID; -+ atomic_set(&set->ref, 0); -+ -+ /* -+ * Next, take the &ip_set_lock, check that we know the type, -+ * and take a reference on the type, to make sure it -+ * stays available while constructing our new set. -+ * -+ * After referencing the type, we drop the &ip_set_lock, -+ * and let the new set construction run without locks. -+ */ -+ set->type = find_set_type_rlock(typename); -+ if (set->type == NULL) { -+ /* Try loading the module */ -+ char modulename[IP_SET_MAXNAMELEN + strlen("ip_set_") + 1]; -+ strcpy(modulename, "ip_set_"); -+ strcat(modulename, typename); -+ DP("try to load %s", modulename); -+ request_module(modulename); -+ set->type = find_set_type_rlock(typename); -+ } -+ if (set->type == NULL) { -+ ip_set_printk("no set type '%s', set '%s' not created", -+ typename, name); -+ res = -ENOENT; -+ goto out; -+ } -+ if (!try_module_get(set->type->me)) { -+ read_unlock_bh(&ip_set_lock); -+ res = -EFAULT; -+ goto out; -+ } -+ read_unlock_bh(&ip_set_lock); -+ -+ /* -+ * Without holding any locks, create private part. -+ */ -+ res = set->type->create(set, data, size); -+ if (res != 0) -+ goto put_out; -+ -+ /* BTW, res==0 here. */ -+ -+ /* -+ * Here, we have a valid, constructed set. &ip_set_lock again, -+ * find free id/index and check that it is not already in -+ * ip_set_list. -+ */ -+ write_lock_bh(&ip_set_lock); -+ if ((res = find_free_id(set->name, &index, &id)) != 0) { -+ DP("no free id!"); -+ goto cleanup; -+ } -+ -+ /* Make sure restore gets the same index */ -+ if (restore != IP_SET_INVALID_ID && index != restore) { -+ DP("Can't restore, sets are screwed up"); -+ res = -ERANGE; -+ goto cleanup; -+ } -+ -+ /* -+ * Finally! Add our shiny new set to the list, and be done. -+ */ -+ DP("create: '%s' created with index %u, id %u!", set->name, index, id); -+ set->id = id; -+ ip_set_list[index] = set; -+ write_unlock_bh(&ip_set_lock); -+ return res; -+ -+ cleanup: -+ write_unlock_bh(&ip_set_lock); -+ set->type->destroy(set); -+ put_out: -+ module_put(set->type->me); -+ out: -+ kfree(set); -+ return res; -+} -+ -+/* -+ * Destroy a given existing set -+ */ -+static void -+ip_set_destroy_set(ip_set_id_t index) -+{ -+ struct ip_set *set = ip_set_list[index]; -+ -+ IP_SET_ASSERT(set); -+ DP("set: %s", set->name); -+ write_lock_bh(&ip_set_lock); -+ FOREACH_HASH_RW_DO(__set_hash_del_byid, set->id); -+ if (set->binding != IP_SET_INVALID_ID) -+ __ip_set_put(set->binding); -+ ip_set_list[index] = NULL; -+ write_unlock_bh(&ip_set_lock); -+ -+ /* Must call it without holding any lock */ -+ set->type->destroy(set); -+ module_put(set->type->me); -+ kfree(set); -+} -+ -+/* -+ * Destroy a set - or all sets -+ * Sets must not be referenced/used. -+ */ -+static int -+ip_set_destroy(ip_set_id_t index) -+{ -+ ip_set_id_t i; -+ -+ /* ref modification always protected by the mutex */ -+ if (index != IP_SET_INVALID_ID) { -+ if (atomic_read(&ip_set_list[index]->ref)) -+ return -EBUSY; -+ ip_set_destroy_set(index); -+ } else { -+ for (i = 0; i < ip_set_max; i++) { -+ if (ip_set_list[i] != NULL -+ && (atomic_read(&ip_set_list[i]->ref))) -+ return -EBUSY; -+ } -+ -+ for (i = 0; i < ip_set_max; i++) { -+ if (ip_set_list[i] != NULL) -+ ip_set_destroy_set(i); -+ } -+ } -+ return 0; -+} -+ -+static void -+ip_set_flush_set(struct ip_set *set) -+{ -+ DP("set: %s %u", set->name, set->id); -+ -+ write_lock_bh(&set->lock); -+ set->type->flush(set); -+ write_unlock_bh(&set->lock); -+} -+ -+/* -+ * Flush data in a set - or in all sets -+ */ -+static int -+ip_set_flush(ip_set_id_t index) -+{ -+ if (index != IP_SET_INVALID_ID) { -+ IP_SET_ASSERT(ip_set_list[index]); -+ ip_set_flush_set(ip_set_list[index]); -+ } else -+ FOREACH_SET_DO(ip_set_flush_set); -+ -+ return 0; -+} -+ -+/* Rename a set */ -+static int -+ip_set_rename(ip_set_id_t index, const char *name) -+{ -+ struct ip_set *set = ip_set_list[index]; -+ ip_set_id_t i; -+ int res = 0; -+ -+ DP("set: %s to %s", set->name, name); -+ write_lock_bh(&ip_set_lock); -+ for (i = 0; i < ip_set_max; i++) { -+ if (ip_set_list[i] != NULL -+ && strncmp(ip_set_list[i]->name, -+ name, -+ IP_SET_MAXNAMELEN - 1) == 0) { -+ res = -EEXIST; -+ goto unlock; -+ } -+ } -+ strncpy(set->name, name, IP_SET_MAXNAMELEN); -+ unlock: -+ write_unlock_bh(&ip_set_lock); -+ return res; -+} -+ -+/* -+ * Swap two sets so that name/index points to the other. -+ * References are also swapped. -+ */ -+static int -+ip_set_swap(ip_set_id_t from_index, ip_set_id_t to_index) -+{ -+ struct ip_set *from = ip_set_list[from_index]; -+ struct ip_set *to = ip_set_list[to_index]; -+ char from_name[IP_SET_MAXNAMELEN]; -+ u_int32_t from_ref; -+ -+ DP("set: %s to %s", from->name, to->name); -+ /* Features must not change. Artifical restriction. */ -+ if (from->type->features != to->type->features) -+ return -ENOEXEC; -+ -+ /* No magic here: ref munging protected by the mutex */ -+ write_lock_bh(&ip_set_lock); -+ strncpy(from_name, from->name, IP_SET_MAXNAMELEN); -+ from_ref = atomic_read(&from->ref); -+ -+ strncpy(from->name, to->name, IP_SET_MAXNAMELEN); -+ atomic_set(&from->ref, atomic_read(&to->ref)); -+ strncpy(to->name, from_name, IP_SET_MAXNAMELEN); -+ atomic_set(&to->ref, from_ref); -+ -+ ip_set_list[from_index] = to; -+ ip_set_list[to_index] = from; -+ -+ write_unlock_bh(&ip_set_lock); -+ return 0; -+} -+ -+/* -+ * List set data -+ */ -+ -+static inline void -+__set_hash_bindings_size_list(struct ip_set_hash *set_hash, -+ ip_set_id_t id, size_t *size) -+{ -+ if (set_hash->id == id) -+ *size += sizeof(struct ip_set_hash_list); -+} -+ -+static inline void -+__set_hash_bindings_size_save(struct ip_set_hash *set_hash, -+ ip_set_id_t id, size_t *size) -+{ -+ if (set_hash->id == id) -+ *size += sizeof(struct ip_set_hash_save); -+} -+ -+static inline void -+__set_hash_bindings(struct ip_set_hash *set_hash, -+ ip_set_id_t id, void *data, int *used) -+{ -+ if (set_hash->id == id) { -+ struct ip_set_hash_list *hash_list = -+ (struct ip_set_hash_list *)(data + *used); -+ -+ hash_list->ip = set_hash->ip; -+ hash_list->binding = set_hash->binding; -+ *used += sizeof(struct ip_set_hash_list); -+ } -+} -+ -+static int ip_set_list_set(ip_set_id_t index, -+ void *data, -+ int *used, -+ int len) -+{ -+ struct ip_set *set = ip_set_list[index]; -+ struct ip_set_list *set_list; -+ -+ /* Pointer to our header */ -+ set_list = (struct ip_set_list *) (data + *used); -+ -+ DP("set: %s, used: %d %p %p", set->name, *used, data, data + *used); -+ -+ /* Get and ensure header size */ -+ if (*used + sizeof(struct ip_set_list) > len) -+ goto not_enough_mem; -+ *used += sizeof(struct ip_set_list); -+ -+ read_lock_bh(&set->lock); -+ /* Get and ensure set specific header size */ -+ set_list->header_size = set->type->header_size; -+ if (*used + set_list->header_size > len) -+ goto unlock_set; -+ -+ /* Fill in the header */ -+ set_list->index = index; -+ set_list->binding = set->binding; -+ set_list->ref = atomic_read(&set->ref); -+ -+ /* Fill in set spefific header data */ -+ set->type->list_header(set, data + *used); -+ *used += set_list->header_size; -+ -+ /* Get and ensure set specific members size */ -+ set_list->members_size = set->type->list_members_size(set); -+ if (*used + set_list->members_size > len) -+ goto unlock_set; -+ -+ /* Fill in set spefific members data */ -+ set->type->list_members(set, data + *used); -+ *used += set_list->members_size; -+ read_unlock_bh(&set->lock); -+ -+ /* Bindings */ -+ -+ /* Get and ensure set specific bindings size */ -+ set_list->bindings_size = 0; -+ FOREACH_HASH_DO(__set_hash_bindings_size_list, -+ set->id, &set_list->bindings_size); -+ if (*used + set_list->bindings_size > len) -+ goto not_enough_mem; -+ -+ /* Fill in set spefific bindings data */ -+ FOREACH_HASH_DO(__set_hash_bindings, set->id, data, used); -+ -+ return 0; -+ -+ unlock_set: -+ read_unlock_bh(&set->lock); -+ not_enough_mem: -+ DP("not enough mem, try again"); -+ return -EAGAIN; -+} -+ -+/* -+ * Save sets -+ */ -+static int ip_set_save_set(ip_set_id_t index, -+ void *data, -+ int *used, -+ int len) -+{ -+ struct ip_set *set; -+ struct ip_set_save *set_save; -+ -+ /* Pointer to our header */ -+ set_save = (struct ip_set_save *) (data + *used); -+ -+ /* Get and ensure header size */ -+ if (*used + sizeof(struct ip_set_save) > len) -+ goto not_enough_mem; -+ *used += sizeof(struct ip_set_save); -+ -+ set = ip_set_list[index]; -+ DP("set: %s, used: %u(%u) %p %p", set->name, *used, len, -+ data, data + *used); -+ -+ read_lock_bh(&set->lock); -+ /* Get and ensure set specific header size */ -+ set_save->header_size = set->type->header_size; -+ if (*used + set_save->header_size > len) -+ goto unlock_set; -+ -+ /* Fill in the header */ -+ set_save->index = index; -+ set_save->binding = set->binding; -+ -+ /* Fill in set spefific header data */ -+ set->type->list_header(set, data + *used); -+ *used += set_save->header_size; -+ -+ DP("set header filled: %s, used: %u(%u) %p %p", set->name, *used, -+ set_save->header_size, data, data + *used); -+ /* Get and ensure set specific members size */ -+ set_save->members_size = set->type->list_members_size(set); -+ if (*used + set_save->members_size > len) -+ goto unlock_set; -+ -+ /* Fill in set spefific members data */ -+ set->type->list_members(set, data + *used); -+ *used += set_save->members_size; -+ read_unlock_bh(&set->lock); -+ DP("set members filled: %s, used: %u(%u) %p %p", set->name, *used, -+ set_save->members_size, data, data + *used); -+ return 0; -+ -+ unlock_set: -+ read_unlock_bh(&set->lock); -+ not_enough_mem: -+ DP("not enough mem, try again"); -+ return -EAGAIN; -+} -+ -+static inline void -+__set_hash_save_bindings(struct ip_set_hash *set_hash, -+ ip_set_id_t id, -+ void *data, -+ int *used, -+ int len, -+ int *res) -+{ -+ if (*res == 0 -+ && (id == IP_SET_INVALID_ID || set_hash->id == id)) { -+ struct ip_set_hash_save *hash_save = -+ (struct ip_set_hash_save *)(data + *used); -+ /* Ensure bindings size */ -+ if (*used + sizeof(struct ip_set_hash_save) > len) { -+ *res = -ENOMEM; -+ return; -+ } -+ hash_save->id = set_hash->id; -+ hash_save->ip = set_hash->ip; -+ hash_save->binding = set_hash->binding; -+ *used += sizeof(struct ip_set_hash_save); -+ } -+} -+ -+static int ip_set_save_bindings(ip_set_id_t index, -+ void *data, -+ int *used, -+ int len) -+{ -+ int res = 0; -+ struct ip_set_save *set_save; -+ -+ DP("used %u, len %u", *used, len); -+ /* Get and ensure header size */ -+ if (*used + sizeof(struct ip_set_save) > len) -+ return -ENOMEM; -+ -+ /* Marker */ -+ set_save = (struct ip_set_save *) (data + *used); -+ set_save->index = IP_SET_INVALID_ID; -+ set_save->header_size = 0; -+ set_save->members_size = 0; -+ *used += sizeof(struct ip_set_save); -+ -+ DP("marker added used %u, len %u", *used, len); -+ /* Fill in bindings data */ -+ if (index != IP_SET_INVALID_ID) -+ /* Sets are identified by id in hash */ -+ index = ip_set_list[index]->id; -+ FOREACH_HASH_DO(__set_hash_save_bindings, index, data, used, len, &res); -+ -+ return res; -+} -+ -+/* -+ * Restore sets -+ */ -+static int ip_set_restore(void *data, -+ int len) -+{ -+ int res = 0; -+ int line = 0, used = 0, members_size; -+ struct ip_set *set; -+ struct ip_set_hash_save *hash_save; -+ struct ip_set_restore *set_restore; -+ ip_set_id_t index; -+ -+ /* Loop to restore sets */ -+ while (1) { -+ line++; -+ -+ DP("%u %u %u", used, sizeof(struct ip_set_restore), len); -+ /* Get and ensure header size */ -+ if (used + sizeof(struct ip_set_restore) > len) -+ return line; -+ set_restore = (struct ip_set_restore *) (data + used); -+ used += sizeof(struct ip_set_restore); -+ -+ /* Ensure data size */ -+ if (used -+ + set_restore->header_size -+ + set_restore->members_size > len) -+ return line; -+ -+ /* Check marker */ -+ if (set_restore->index == IP_SET_INVALID_ID) { -+ line--; -+ goto bindings; -+ } -+ -+ /* Try to create the set */ -+ DP("restore %s %s", set_restore->name, set_restore->typename); -+ res = ip_set_create(set_restore->name, -+ set_restore->typename, -+ set_restore->index, -+ data + used, -+ set_restore->header_size); -+ -+ if (res != 0) -+ return line; -+ used += set_restore->header_size; -+ -+ index = ip_set_find_byindex(set_restore->index); -+ DP("index %u, restore_index %u", index, set_restore->index); -+ if (index != set_restore->index) -+ return line; -+ /* Try to restore members data */ -+ set = ip_set_list[index]; -+ members_size = 0; -+ DP("members_size %u reqsize %u", -+ set_restore->members_size, set->type->reqsize); -+ while (members_size + set->type->reqsize <= -+ set_restore->members_size) { -+ line++; -+ DP("members: %u, line %u", members_size, line); -+ res = __ip_set_addip(index, -+ data + used + members_size, -+ set->type->reqsize); -+ if (!(res == 0 || res == -EEXIST)) -+ return line; -+ members_size += set->type->reqsize; -+ } -+ -+ DP("members_size %u %u", -+ set_restore->members_size, members_size); -+ if (members_size != set_restore->members_size) -+ return line++; -+ used += set_restore->members_size; -+ } -+ -+ bindings: -+ /* Loop to restore bindings */ -+ while (used < len) { -+ line++; -+ -+ DP("restore binding, line %u", line); -+ /* Get and ensure size */ -+ if (used + sizeof(struct ip_set_hash_save) > len) -+ return line; -+ hash_save = (struct ip_set_hash_save *) (data + used); -+ used += sizeof(struct ip_set_hash_save); -+ -+ /* hash_save->id is used to store the index */ -+ index = ip_set_find_byindex(hash_save->id); -+ DP("restore binding index %u, id %u, %u -> %u", -+ index, hash_save->id, hash_save->ip, hash_save->binding); -+ if (index != hash_save->id) -+ return line; -+ if (ip_set_find_byindex(hash_save->binding) == IP_SET_INVALID_ID) { -+ DP("corrupt binding set index %u", hash_save->binding); -+ return line; -+ } -+ set = ip_set_list[hash_save->id]; -+ /* Null valued IP means default binding */ -+ if (hash_save->ip) -+ res = ip_set_hash_add(set->id, -+ hash_save->ip, -+ hash_save->binding); -+ else { -+ IP_SET_ASSERT(set->binding == IP_SET_INVALID_ID); -+ write_lock_bh(&ip_set_lock); -+ set->binding = hash_save->binding; -+ __ip_set_get(set->binding); -+ write_unlock_bh(&ip_set_lock); -+ DP("default binding: %u", set->binding); -+ } -+ if (res != 0) -+ return line; -+ } -+ if (used != len) -+ return line; -+ -+ return 0; -+} -+ -+static int -+ip_set_sockfn_set(struct sock *sk, int optval, void *user, unsigned int len) -+{ -+ void *data; -+ int res = 0; /* Assume OK */ -+ unsigned *op; -+ struct ip_set_req_adt *req_adt; -+ ip_set_id_t index = IP_SET_INVALID_ID; -+ int (*adtfn)(ip_set_id_t index, -+ const void *data, size_t size); -+ struct fn_table { -+ int (*fn)(ip_set_id_t index, -+ const void *data, size_t size); -+ } adtfn_table[] = -+ { { ip_set_addip }, { ip_set_delip }, { ip_set_testip}, -+ { ip_set_bindip}, { ip_set_unbindip }, { ip_set_testbind }, -+ }; -+ -+ DP("optval=%d, user=%p, len=%d", optval, user, len); -+ if (!capable(CAP_NET_ADMIN)) -+ return -EPERM; -+ if (optval != SO_IP_SET) -+ return -EBADF; -+ if (len <= sizeof(unsigned)) { -+ ip_set_printk("short userdata (want >%zu, got %u)", -+ sizeof(unsigned), len); -+ return -EINVAL; -+ } -+ data = vmalloc(len); -+ if (!data) { -+ DP("out of mem for %u bytes", len); -+ return -ENOMEM; -+ } -+ if (copy_from_user(data, user, len) != 0) { -+ res = -EFAULT; -+ goto done; -+ } -+ if (down_interruptible(&ip_set_app_mutex)) { -+ res = -EINTR; -+ goto done; -+ } -+ -+ op = (unsigned *)data; -+ DP("op=%x", *op); -+ -+ if (*op < IP_SET_OP_VERSION) { -+ /* Check the version at the beginning of operations */ -+ struct ip_set_req_version *req_version = -+ (struct ip_set_req_version *) data; -+ if (req_version->version != IP_SET_PROTOCOL_VERSION) { -+ res = -EPROTO; -+ goto done; -+ } -+ } -+ -+ switch (*op) { -+ case IP_SET_OP_CREATE:{ -+ struct ip_set_req_create *req_create -+ = (struct ip_set_req_create *) data; -+ -+ if (len < sizeof(struct ip_set_req_create)) { -+ ip_set_printk("short CREATE data (want >=%zu, got %u)", -+ sizeof(struct ip_set_req_create), len); -+ res = -EINVAL; -+ goto done; -+ } -+ req_create->name[IP_SET_MAXNAMELEN - 1] = '\0'; -+ req_create->typename[IP_SET_MAXNAMELEN - 1] = '\0'; -+ res = ip_set_create(req_create->name, -+ req_create->typename, -+ IP_SET_INVALID_ID, -+ data + sizeof(struct ip_set_req_create), -+ len - sizeof(struct ip_set_req_create)); -+ goto done; -+ } -+ case IP_SET_OP_DESTROY:{ -+ struct ip_set_req_std *req_destroy -+ = (struct ip_set_req_std *) data; -+ -+ if (len != sizeof(struct ip_set_req_std)) { -+ ip_set_printk("invalid DESTROY data (want %zu, got %u)", -+ sizeof(struct ip_set_req_std), len); -+ res = -EINVAL; -+ goto done; -+ } -+ if (strcmp(req_destroy->name, IPSET_TOKEN_ALL) == 0) { -+ /* Destroy all sets */ -+ index = IP_SET_INVALID_ID; -+ } else { -+ req_destroy->name[IP_SET_MAXNAMELEN - 1] = '\0'; -+ index = ip_set_find_byname(req_destroy->name); -+ -+ if (index == IP_SET_INVALID_ID) { -+ res = -ENOENT; -+ goto done; -+ } -+ } -+ -+ res = ip_set_destroy(index); -+ goto done; -+ } -+ case IP_SET_OP_FLUSH:{ -+ struct ip_set_req_std *req_flush = -+ (struct ip_set_req_std *) data; -+ -+ if (len != sizeof(struct ip_set_req_std)) { -+ ip_set_printk("invalid FLUSH data (want %zu, got %u)", -+ sizeof(struct ip_set_req_std), len); -+ res = -EINVAL; -+ goto done; -+ } -+ if (strcmp(req_flush->name, IPSET_TOKEN_ALL) == 0) { -+ /* Flush all sets */ -+ index = IP_SET_INVALID_ID; -+ } else { -+ req_flush->name[IP_SET_MAXNAMELEN - 1] = '\0'; -+ index = ip_set_find_byname(req_flush->name); -+ -+ if (index == IP_SET_INVALID_ID) { -+ res = -ENOENT; -+ goto done; -+ } -+ } -+ res = ip_set_flush(index); -+ goto done; -+ } -+ case IP_SET_OP_RENAME:{ -+ struct ip_set_req_create *req_rename -+ = (struct ip_set_req_create *) data; -+ -+ if (len != sizeof(struct ip_set_req_create)) { -+ ip_set_printk("invalid RENAME data (want %zu, got %u)", -+ sizeof(struct ip_set_req_create), len); -+ res = -EINVAL; -+ goto done; -+ } -+ -+ req_rename->name[IP_SET_MAXNAMELEN - 1] = '\0'; -+ req_rename->typename[IP_SET_MAXNAMELEN - 1] = '\0'; -+ -+ index = ip_set_find_byname(req_rename->name); -+ if (index == IP_SET_INVALID_ID) { -+ res = -ENOENT; -+ goto done; -+ } -+ res = ip_set_rename(index, req_rename->typename); -+ goto done; -+ } -+ case IP_SET_OP_SWAP:{ -+ struct ip_set_req_create *req_swap -+ = (struct ip_set_req_create *) data; -+ ip_set_id_t to_index; -+ -+ if (len != sizeof(struct ip_set_req_create)) { -+ ip_set_printk("invalid SWAP data (want %zu, got %u)", -+ sizeof(struct ip_set_req_create), len); -+ res = -EINVAL; -+ goto done; -+ } -+ -+ req_swap->name[IP_SET_MAXNAMELEN - 1] = '\0'; -+ req_swap->typename[IP_SET_MAXNAMELEN - 1] = '\0'; -+ -+ index = ip_set_find_byname(req_swap->name); -+ if (index == IP_SET_INVALID_ID) { -+ res = -ENOENT; -+ goto done; -+ } -+ to_index = ip_set_find_byname(req_swap->typename); -+ if (to_index == IP_SET_INVALID_ID) { -+ res = -ENOENT; -+ goto done; -+ } -+ res = ip_set_swap(index, to_index); -+ goto done; -+ } -+ default: -+ break; /* Set identified by id */ -+ } -+ -+ /* There we may have add/del/test/bind/unbind/test_bind operations */ -+ if (*op < IP_SET_OP_ADD_IP || *op > IP_SET_OP_TEST_BIND_SET) { -+ res = -EBADMSG; -+ goto done; -+ } -+ adtfn = adtfn_table[*op - IP_SET_OP_ADD_IP].fn; -+ -+ if (len < sizeof(struct ip_set_req_adt)) { -+ ip_set_printk("short data in adt request (want >=%zu, got %u)", -+ sizeof(struct ip_set_req_adt), len); -+ res = -EINVAL; -+ goto done; -+ } -+ req_adt = (struct ip_set_req_adt *) data; -+ -+ /* -U :all: :all:|:default: uses IP_SET_INVALID_ID */ -+ if (!(*op == IP_SET_OP_UNBIND_SET -+ && req_adt->index == IP_SET_INVALID_ID)) { -+ index = ip_set_find_byindex(req_adt->index); -+ if (index == IP_SET_INVALID_ID) { -+ res = -ENOENT; -+ goto done; -+ } -+ } -+ res = adtfn(index, data, len); -+ -+ done: -+ up(&ip_set_app_mutex); -+ vfree(data); -+ if (res > 0) -+ res = 0; -+ DP("final result %d", res); -+ return res; -+} -+ -+static int -+ip_set_sockfn_get(struct sock *sk, int optval, void *user, int *len) -+{ -+ int res = 0; -+ unsigned *op; -+ ip_set_id_t index = IP_SET_INVALID_ID; -+ void *data; -+ int copylen = *len; -+ -+ DP("optval=%d, user=%p, len=%d", optval, user, *len); -+ if (!capable(CAP_NET_ADMIN)) -+ return -EPERM; -+ if (optval != SO_IP_SET) -+ return -EBADF; -+ if (*len < sizeof(unsigned)) { -+ ip_set_printk("short userdata (want >=%zu, got %d)", -+ sizeof(unsigned), *len); -+ return -EINVAL; -+ } -+ data = vmalloc(*len); -+ if (!data) { -+ DP("out of mem for %d bytes", *len); -+ return -ENOMEM; -+ } -+ if (copy_from_user(data, user, *len) != 0) { -+ res = -EFAULT; -+ goto done; -+ } -+ if (down_interruptible(&ip_set_app_mutex)) { -+ res = -EINTR; -+ goto done; -+ } -+ -+ op = (unsigned *) data; -+ DP("op=%x", *op); -+ -+ if (*op < IP_SET_OP_VERSION) { -+ /* Check the version at the beginning of operations */ -+ struct ip_set_req_version *req_version = -+ (struct ip_set_req_version *) data; -+ if (req_version->version != IP_SET_PROTOCOL_VERSION) { -+ res = -EPROTO; -+ goto done; -+ } -+ } -+ -+ switch (*op) { -+ case IP_SET_OP_VERSION: { -+ struct ip_set_req_version *req_version = -+ (struct ip_set_req_version *) data; -+ -+ if (*len != sizeof(struct ip_set_req_version)) { -+ ip_set_printk("invalid VERSION (want %zu, got %d)", -+ sizeof(struct ip_set_req_version), -+ *len); -+ res = -EINVAL; -+ goto done; -+ } -+ -+ req_version->version = IP_SET_PROTOCOL_VERSION; -+ res = copy_to_user(user, req_version, -+ sizeof(struct ip_set_req_version)); -+ goto done; -+ } -+ case IP_SET_OP_GET_BYNAME: { -+ struct ip_set_req_get_set *req_get -+ = (struct ip_set_req_get_set *) data; -+ -+ if (*len != sizeof(struct ip_set_req_get_set)) { -+ ip_set_printk("invalid GET_BYNAME (want %zu, got %d)", -+ sizeof(struct ip_set_req_get_set), *len); -+ res = -EINVAL; -+ goto done; -+ } -+ req_get->set.name[IP_SET_MAXNAMELEN - 1] = '\0'; -+ index = ip_set_find_byname(req_get->set.name); -+ req_get->set.index = index; -+ goto copy; -+ } -+ case IP_SET_OP_GET_BYINDEX: { -+ struct ip_set_req_get_set *req_get -+ = (struct ip_set_req_get_set *) data; -+ -+ if (*len != sizeof(struct ip_set_req_get_set)) { -+ ip_set_printk("invalid GET_BYINDEX (want %zu, got %d)", -+ sizeof(struct ip_set_req_get_set), *len); -+ res = -EINVAL; -+ goto done; -+ } -+ req_get->set.name[IP_SET_MAXNAMELEN - 1] = '\0'; -+ index = ip_set_find_byindex(req_get->set.index); -+ strncpy(req_get->set.name, -+ index == IP_SET_INVALID_ID ? "" -+ : ip_set_list[index]->name, IP_SET_MAXNAMELEN); -+ goto copy; -+ } -+ case IP_SET_OP_ADT_GET: { -+ struct ip_set_req_adt_get *req_get -+ = (struct ip_set_req_adt_get *) data; -+ -+ if (*len != sizeof(struct ip_set_req_adt_get)) { -+ ip_set_printk("invalid ADT_GET (want %zu, got %d)", -+ sizeof(struct ip_set_req_adt_get), *len); -+ res = -EINVAL; -+ goto done; -+ } -+ req_get->set.name[IP_SET_MAXNAMELEN - 1] = '\0'; -+ index = ip_set_find_byname(req_get->set.name); -+ if (index != IP_SET_INVALID_ID) { -+ req_get->set.index = index; -+ strncpy(req_get->typename, -+ ip_set_list[index]->type->typename, -+ IP_SET_MAXNAMELEN - 1); -+ } else { -+ res = -ENOENT; -+ goto done; -+ } -+ goto copy; -+ } -+ case IP_SET_OP_MAX_SETS: { -+ struct ip_set_req_max_sets *req_max_sets -+ = (struct ip_set_req_max_sets *) data; -+ ip_set_id_t i; -+ -+ if (*len != sizeof(struct ip_set_req_max_sets)) { -+ ip_set_printk("invalid MAX_SETS (want %zu, got %d)", -+ sizeof(struct ip_set_req_max_sets), *len); -+ res = -EINVAL; -+ goto done; -+ } -+ -+ if (strcmp(req_max_sets->set.name, IPSET_TOKEN_ALL) == 0) { -+ req_max_sets->set.index = IP_SET_INVALID_ID; -+ } else { -+ req_max_sets->set.name[IP_SET_MAXNAMELEN - 1] = '\0'; -+ req_max_sets->set.index = -+ ip_set_find_byname(req_max_sets->set.name); -+ if (req_max_sets->set.index == IP_SET_INVALID_ID) { -+ res = -ENOENT; -+ goto done; -+ } -+ } -+ req_max_sets->max_sets = ip_set_max; -+ req_max_sets->sets = 0; -+ for (i = 0; i < ip_set_max; i++) { -+ if (ip_set_list[i] != NULL) -+ req_max_sets->sets++; -+ } -+ goto copy; -+ } -+ case IP_SET_OP_LIST_SIZE: -+ case IP_SET_OP_SAVE_SIZE: { -+ struct ip_set_req_setnames *req_setnames -+ = (struct ip_set_req_setnames *) data; -+ struct ip_set_name_list *name_list; -+ struct ip_set *set; -+ ip_set_id_t i; -+ int used; -+ -+ if (*len < sizeof(struct ip_set_req_setnames)) { -+ ip_set_printk("short LIST_SIZE (want >=%zu, got %d)", -+ sizeof(struct ip_set_req_setnames), *len); -+ res = -EINVAL; -+ goto done; -+ } -+ -+ req_setnames->size = 0; -+ used = sizeof(struct ip_set_req_setnames); -+ for (i = 0; i < ip_set_max; i++) { -+ if (ip_set_list[i] == NULL) -+ continue; -+ name_list = (struct ip_set_name_list *) -+ (data + used); -+ used += sizeof(struct ip_set_name_list); -+ if (used > copylen) { -+ res = -EAGAIN; -+ goto done; -+ } -+ set = ip_set_list[i]; -+ /* Fill in index, name, etc. */ -+ name_list->index = i; -+ name_list->id = set->id; -+ strncpy(name_list->name, -+ set->name, -+ IP_SET_MAXNAMELEN - 1); -+ strncpy(name_list->typename, -+ set->type->typename, -+ IP_SET_MAXNAMELEN - 1); -+ DP("filled %s of type %s, index %u\n", -+ name_list->name, name_list->typename, -+ name_list->index); -+ if (!(req_setnames->index == IP_SET_INVALID_ID -+ || req_setnames->index == i)) -+ continue; -+ /* Update size */ -+ switch (*op) { -+ case IP_SET_OP_LIST_SIZE: { -+ req_setnames->size += sizeof(struct ip_set_list) -+ + set->type->header_size -+ + set->type->list_members_size(set); -+ /* Sets are identified by id in the hash */ -+ FOREACH_HASH_DO(__set_hash_bindings_size_list, -+ set->id, &req_setnames->size); -+ break; -+ } -+ case IP_SET_OP_SAVE_SIZE: { -+ req_setnames->size += sizeof(struct ip_set_save) -+ + set->type->header_size -+ + set->type->list_members_size(set); -+ FOREACH_HASH_DO(__set_hash_bindings_size_save, -+ set->id, &req_setnames->size); -+ break; -+ } -+ default: -+ break; -+ } -+ } -+ if (copylen != used) { -+ res = -EAGAIN; -+ goto done; -+ } -+ goto copy; -+ } -+ case IP_SET_OP_LIST: { -+ struct ip_set_req_list *req_list -+ = (struct ip_set_req_list *) data; -+ ip_set_id_t i; -+ int used; -+ -+ if (*len < sizeof(struct ip_set_req_list)) { -+ ip_set_printk("short LIST (want >=%zu, got %d)", -+ sizeof(struct ip_set_req_list), *len); -+ res = -EINVAL; -+ goto done; -+ } -+ index = req_list->index; -+ if (index != IP_SET_INVALID_ID -+ && ip_set_find_byindex(index) != index) { -+ res = -ENOENT; -+ goto done; -+ } -+ used = 0; -+ if (index == IP_SET_INVALID_ID) { -+ /* List all sets */ -+ for (i = 0; i < ip_set_max && res == 0; i++) { -+ if (ip_set_list[i] != NULL) -+ res = ip_set_list_set(i, data, &used, *len); -+ } -+ } else { -+ /* List an individual set */ -+ res = ip_set_list_set(index, data, &used, *len); -+ } -+ if (res != 0) -+ goto done; -+ else if (copylen != used) { -+ res = -EAGAIN; -+ goto done; -+ } -+ goto copy; -+ } -+ case IP_SET_OP_SAVE: { -+ struct ip_set_req_list *req_save -+ = (struct ip_set_req_list *) data; -+ ip_set_id_t i; -+ int used; -+ -+ if (*len < sizeof(struct ip_set_req_list)) { -+ ip_set_printk("short SAVE (want >=%zu, got %d)", -+ sizeof(struct ip_set_req_list), *len); -+ res = -EINVAL; -+ goto done; -+ } -+ index = req_save->index; -+ if (index != IP_SET_INVALID_ID -+ && ip_set_find_byindex(index) != index) { -+ res = -ENOENT; -+ goto done; -+ } -+ used = 0; -+ if (index == IP_SET_INVALID_ID) { -+ /* Save all sets */ -+ for (i = 0; i < ip_set_max && res == 0; i++) { -+ if (ip_set_list[i] != NULL) -+ res = ip_set_save_set(i, data, &used, *len); -+ } -+ } else { -+ /* Save an individual set */ -+ res = ip_set_save_set(index, data, &used, *len); -+ } -+ if (res == 0) -+ res = ip_set_save_bindings(index, data, &used, *len); -+ -+ if (res != 0) -+ goto done; -+ else if (copylen != used) { -+ res = -EAGAIN; -+ goto done; -+ } -+ goto copy; -+ } -+ case IP_SET_OP_RESTORE: { -+ struct ip_set_req_setnames *req_restore -+ = (struct ip_set_req_setnames *) data; -+ int line; -+ -+ if (*len < sizeof(struct ip_set_req_setnames) -+ || *len != req_restore->size) { -+ ip_set_printk("invalid RESTORE (want =%zu, got %d)", -+ req_restore->size, *len); -+ res = -EINVAL; -+ goto done; -+ } -+ line = ip_set_restore(data + sizeof(struct ip_set_req_setnames), -+ req_restore->size - sizeof(struct ip_set_req_setnames)); -+ DP("ip_set_restore: %u", line); -+ if (line != 0) { -+ res = -EAGAIN; -+ req_restore->size = line; -+ copylen = sizeof(struct ip_set_req_setnames); -+ goto copy; -+ } -+ goto done; -+ } -+ default: -+ res = -EBADMSG; -+ goto done; -+ } /* end of switch(op) */ -+ -+ copy: -+ DP("set %s, copylen %u", index != IP_SET_INVALID_ID -+ && ip_set_list[index] -+ ? ip_set_list[index]->name -+ : ":all:", copylen); -+ res = copy_to_user(user, data, copylen); -+ -+ done: -+ up(&ip_set_app_mutex); -+ vfree(data); -+ if (res > 0) -+ res = 0; -+ DP("final result %d", res); -+ return res; -+} -+ -+static struct nf_sockopt_ops so_set = { -+ .pf = PF_INET, -+ .set_optmin = SO_IP_SET, -+ .set_optmax = SO_IP_SET + 1, -+ .set = &ip_set_sockfn_set, -+ .get_optmin = SO_IP_SET, -+ .get_optmax = SO_IP_SET + 1, -+ .get = &ip_set_sockfn_get, -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,23) -+ .owner = THIS_MODULE, -+#endif -+}; -+ -+static int max_sets, hash_size; -+module_param(max_sets, int, 0600); -+MODULE_PARM_DESC(max_sets, "maximal number of sets"); -+module_param(hash_size, int, 0600); -+MODULE_PARM_DESC(hash_size, "hash size for bindings"); -+MODULE_LICENSE("GPL"); -+MODULE_AUTHOR("Jozsef Kadlecsik <kadlec@blackhole.kfki.hu>"); -+MODULE_DESCRIPTION("module implementing core IP set support"); -+ -+static int __init ip_set_init(void) -+{ -+ int res; -+ ip_set_id_t i; -+ -+ get_random_bytes(&ip_set_hash_random, 4); -+ if (max_sets) -+ ip_set_max = max_sets; -+ ip_set_list = vmalloc(sizeof(struct ip_set *) * ip_set_max); -+ if (!ip_set_list) { -+ printk(KERN_ERR "Unable to create ip_set_list\n"); -+ return -ENOMEM; -+ } -+ memset(ip_set_list, 0, sizeof(struct ip_set *) * ip_set_max); -+ if (hash_size) -+ ip_set_bindings_hash_size = hash_size; -+ ip_set_hash = vmalloc(sizeof(struct list_head) * ip_set_bindings_hash_size); -+ if (!ip_set_hash) { -+ printk(KERN_ERR "Unable to create ip_set_hash\n"); -+ vfree(ip_set_list); -+ return -ENOMEM; -+ } -+ for (i = 0; i < ip_set_bindings_hash_size; i++) -+ INIT_LIST_HEAD(&ip_set_hash[i]); -+ -+ INIT_LIST_HEAD(&set_type_list); -+ -+ res = nf_register_sockopt(&so_set); -+ if (res != 0) { -+ ip_set_printk("SO_SET registry failed: %d", res); -+ vfree(ip_set_list); -+ vfree(ip_set_hash); -+ return res; -+ } -+ return 0; -+} -+ -+static void __exit ip_set_fini(void) -+{ -+ /* There can't be any existing set or binding */ -+ nf_unregister_sockopt(&so_set); -+ vfree(ip_set_list); -+ vfree(ip_set_hash); -+ DP("these are the famous last words"); -+} -+ -+EXPORT_SYMBOL(ip_set_register_set_type); -+EXPORT_SYMBOL(ip_set_unregister_set_type); -+ -+EXPORT_SYMBOL(ip_set_get_byname); -+EXPORT_SYMBOL(ip_set_get_byindex); -+EXPORT_SYMBOL(ip_set_put); -+ -+EXPORT_SYMBOL(ip_set_addip_kernel); -+EXPORT_SYMBOL(ip_set_delip_kernel); -+EXPORT_SYMBOL(ip_set_testip_kernel); -+ -+module_init(ip_set_init); -+module_exit(ip_set_fini); ---- /dev/null -+++ b/net/ipv4/netfilter/ip_set_iphash.c -@@ -0,0 +1,429 @@ -+/* Copyright (C) 2003-2004 Jozsef Kadlecsik <kadlec@blackhole.kfki.hu> -+ * -+ * This program is free software; you can redistribute it and/or modify -+ * it under the terms of the GNU General Public License version 2 as -+ * published by the Free Software Foundation. -+ */ -+ -+/* Kernel module implementing an ip hash set */ -+ -+#include <linux/module.h> -+#include <linux/ip.h> -+#include <linux/skbuff.h> -+#include <linux/version.h> -+#include <linux/jhash.h> -+#include <linux/netfilter_ipv4/ip_tables.h> -+#include <linux/netfilter_ipv4/ip_set.h> -+#include <linux/errno.h> -+#include <asm/uaccess.h> -+#include <asm/bitops.h> -+#include <linux/spinlock.h> -+#include <linux/vmalloc.h> -+#include <linux/random.h> -+ -+#include <net/ip.h> -+ -+#include <linux/netfilter_ipv4/ip_set_malloc.h> -+#include <linux/netfilter_ipv4/ip_set_iphash.h> -+ -+static int limit = MAX_RANGE; -+ -+static inline __u32 -+jhash_ip(const struct ip_set_iphash *map, uint16_t i, ip_set_ip_t ip) -+{ -+ return jhash_1word(ip, *(((uint32_t *) map->initval) + i)); -+} -+ -+static inline __u32 -+hash_id(struct ip_set *set, ip_set_ip_t ip, ip_set_ip_t *hash_ip) -+{ -+ struct ip_set_iphash *map = (struct ip_set_iphash *) set->data; -+ __u32 id; -+ u_int16_t i; -+ ip_set_ip_t *elem; -+ -+ *hash_ip = ip & map->netmask; -+ DP("set: %s, ip:%u.%u.%u.%u, %u.%u.%u.%u, %u.%u.%u.%u", -+ set->name, HIPQUAD(ip), HIPQUAD(*hash_ip), HIPQUAD(map->netmask)); -+ -+ for (i = 0; i < map->probes; i++) { -+ id = jhash_ip(map, i, *hash_ip) % map->hashsize; -+ DP("hash key: %u", id); -+ elem = HARRAY_ELEM(map->members, ip_set_ip_t *, id); -+ if (*elem == *hash_ip) -+ return id; -+ /* No shortcut at testing - there can be deleted -+ * entries. */ -+ } -+ return UINT_MAX; -+} -+ -+static inline int -+__testip(struct ip_set *set, ip_set_ip_t ip, ip_set_ip_t *hash_ip) -+{ -+ return (ip && hash_id(set, ip, hash_ip) != UINT_MAX); -+} -+ -+static int -+testip(struct ip_set *set, const void *data, size_t size, -+ ip_set_ip_t *hash_ip) -+{ -+ struct ip_set_req_iphash *req = -+ (struct ip_set_req_iphash *) data; -+ -+ if (size != sizeof(struct ip_set_req_iphash)) { -+ ip_set_printk("data length wrong (want %zu, have %zu)", -+ sizeof(struct ip_set_req_iphash), -+ size); -+ return -EINVAL; -+ } -+ return __testip(set, req->ip, hash_ip); -+} -+ -+static int -+testip_kernel(struct ip_set *set, -+ const struct sk_buff *skb, -+ ip_set_ip_t *hash_ip, -+ const u_int32_t *flags, -+ unsigned char index) -+{ -+ return __testip(set, -+ ntohl(flags[index] & IPSET_SRC -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,22) -+ ? ip_hdr(skb)->saddr -+ : ip_hdr(skb)->daddr), -+#else -+ ? skb->nh.iph->saddr -+ : skb->nh.iph->daddr), -+#endif -+ hash_ip); -+} -+ -+static inline int -+__addip(struct ip_set_iphash *map, ip_set_ip_t ip, ip_set_ip_t *hash_ip) -+{ -+ __u32 probe; -+ u_int16_t i; -+ ip_set_ip_t *elem; -+ -+ if (!ip || map->elements >= limit) -+ return -ERANGE; -+ -+ *hash_ip = ip & map->netmask; -+ -+ for (i = 0; i < map->probes; i++) { -+ probe = jhash_ip(map, i, *hash_ip) % map->hashsize; -+ elem = HARRAY_ELEM(map->members, ip_set_ip_t *, probe); -+ if (*elem == *hash_ip) -+ return -EEXIST; -+ if (!*elem) { -+ *elem = *hash_ip; -+ map->elements++; -+ return 0; -+ } -+ } -+ /* Trigger rehashing */ -+ return -EAGAIN; -+} -+ -+static int -+addip(struct ip_set *set, const void *data, size_t size, -+ ip_set_ip_t *hash_ip) -+{ -+ struct ip_set_req_iphash *req = -+ (struct ip_set_req_iphash *) data; -+ -+ if (size != sizeof(struct ip_set_req_iphash)) { -+ ip_set_printk("data length wrong (want %zu, have %zu)", -+ sizeof(struct ip_set_req_iphash), -+ size); -+ return -EINVAL; -+ } -+ return __addip((struct ip_set_iphash *) set->data, req->ip, hash_ip); -+} -+ -+static int -+addip_kernel(struct ip_set *set, -+ const struct sk_buff *skb, -+ ip_set_ip_t *hash_ip, -+ const u_int32_t *flags, -+ unsigned char index) -+{ -+ return __addip((struct ip_set_iphash *) set->data, -+ ntohl(flags[index] & IPSET_SRC -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,22) -+ ? ip_hdr(skb)->saddr -+ : ip_hdr(skb)->daddr), -+#else -+ ? skb->nh.iph->saddr -+ : skb->nh.iph->daddr), -+#endif -+ hash_ip); -+} -+ -+static int retry(struct ip_set *set) -+{ -+ struct ip_set_iphash *map = (struct ip_set_iphash *) set->data; -+ ip_set_ip_t hash_ip, *elem; -+ void *members; -+ u_int32_t i, hashsize = map->hashsize; -+ int res; -+ struct ip_set_iphash *tmp; -+ -+ if (map->resize == 0) -+ return -ERANGE; -+ -+ again: -+ res = 0; -+ -+ /* Calculate new hash size */ -+ hashsize += (hashsize * map->resize)/100; -+ if (hashsize == map->hashsize) -+ hashsize++; -+ -+ ip_set_printk("rehashing of set %s triggered: " -+ "hashsize grows from %u to %u", -+ set->name, map->hashsize, hashsize); -+ -+ tmp = kmalloc(sizeof(struct ip_set_iphash) -+ + map->probes * sizeof(uint32_t), GFP_ATOMIC); -+ if (!tmp) { -+ DP("out of memory for %d bytes", -+ sizeof(struct ip_set_iphash) -+ + map->probes * sizeof(uint32_t)); -+ return -ENOMEM; -+ } -+ tmp->members = harray_malloc(hashsize, sizeof(ip_set_ip_t), GFP_ATOMIC); -+ if (!tmp->members) { -+ DP("out of memory for %d bytes", hashsize * sizeof(ip_set_ip_t)); -+ kfree(tmp); -+ return -ENOMEM; -+ } -+ tmp->hashsize = hashsize; -+ tmp->elements = 0; -+ tmp->probes = map->probes; -+ tmp->resize = map->resize; -+ tmp->netmask = map->netmask; -+ memcpy(tmp->initval, map->initval, map->probes * sizeof(uint32_t)); -+ -+ write_lock_bh(&set->lock); -+ map = (struct ip_set_iphash *) set->data; /* Play safe */ -+ for (i = 0; i < map->hashsize && res == 0; i++) { -+ elem = HARRAY_ELEM(map->members, ip_set_ip_t *, i); -+ if (*elem) -+ res = __addip(tmp, *elem, &hash_ip); -+ } -+ if (res) { -+ /* Failure, try again */ -+ write_unlock_bh(&set->lock); -+ harray_free(tmp->members); -+ kfree(tmp); -+ goto again; -+ } -+ -+ /* Success at resizing! */ -+ members = map->members; -+ -+ map->hashsize = tmp->hashsize; -+ map->members = tmp->members; -+ write_unlock_bh(&set->lock); -+ -+ harray_free(members); -+ kfree(tmp); -+ -+ return 0; -+} -+ -+static inline int -+__delip(struct ip_set *set, ip_set_ip_t ip, ip_set_ip_t *hash_ip) -+{ -+ struct ip_set_iphash *map = (struct ip_set_iphash *) set->data; -+ ip_set_ip_t id, *elem; -+ -+ if (!ip) -+ return -ERANGE; -+ -+ id = hash_id(set, ip, hash_ip); -+ if (id == UINT_MAX) -+ return -EEXIST; -+ -+ elem = HARRAY_ELEM(map->members, ip_set_ip_t *, id); -+ *elem = 0; -+ map->elements--; -+ -+ return 0; -+} -+ -+static int -+delip(struct ip_set *set, const void *data, size_t size, -+ ip_set_ip_t *hash_ip) -+{ -+ struct ip_set_req_iphash *req = -+ (struct ip_set_req_iphash *) data; -+ -+ if (size != sizeof(struct ip_set_req_iphash)) { -+ ip_set_printk("data length wrong (want %zu, have %zu)", -+ sizeof(struct ip_set_req_iphash), -+ size); -+ return -EINVAL; -+ } -+ return __delip(set, req->ip, hash_ip); -+} -+ -+static int -+delip_kernel(struct ip_set *set, -+ const struct sk_buff *skb, -+ ip_set_ip_t *hash_ip, -+ const u_int32_t *flags, -+ unsigned char index) -+{ -+ return __delip(set, -+ ntohl(flags[index] & IPSET_SRC -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,22) -+ ? ip_hdr(skb)->saddr -+ : ip_hdr(skb)->daddr), -+#else -+ ? skb->nh.iph->saddr -+ : skb->nh.iph->daddr), -+#endif -+ hash_ip); -+} -+ -+static int create(struct ip_set *set, const void *data, size_t size) -+{ -+ struct ip_set_req_iphash_create *req = -+ (struct ip_set_req_iphash_create *) data; -+ struct ip_set_iphash *map; -+ uint16_t i; -+ -+ if (size != sizeof(struct ip_set_req_iphash_create)) { -+ ip_set_printk("data length wrong (want %zu, have %zu)", -+ sizeof(struct ip_set_req_iphash_create), -+ size); -+ return -EINVAL; -+ } -+ -+ if (req->hashsize < 1) { -+ ip_set_printk("hashsize too small"); -+ return -ENOEXEC; -+ } -+ -+ if (req->probes < 1) { -+ ip_set_printk("probes too small"); -+ return -ENOEXEC; -+ } -+ -+ map = kmalloc(sizeof(struct ip_set_iphash) -+ + req->probes * sizeof(uint32_t), GFP_KERNEL); -+ if (!map) { -+ DP("out of memory for %d bytes", -+ sizeof(struct ip_set_iphash) -+ + req->probes * sizeof(uint32_t)); -+ return -ENOMEM; -+ } -+ for (i = 0; i < req->probes; i++) -+ get_random_bytes(((uint32_t *) map->initval)+i, 4); -+ map->elements = 0; -+ map->hashsize = req->hashsize; -+ map->probes = req->probes; -+ map->resize = req->resize; -+ map->netmask = req->netmask; -+ map->members = harray_malloc(map->hashsize, sizeof(ip_set_ip_t), GFP_KERNEL); -+ if (!map->members) { -+ DP("out of memory for %d bytes", map->hashsize * sizeof(ip_set_ip_t)); -+ kfree(map); -+ return -ENOMEM; -+ } -+ -+ set->data = map; -+ return 0; -+} -+ -+static void destroy(struct ip_set *set) -+{ -+ struct ip_set_iphash *map = (struct ip_set_iphash *) set->data; -+ -+ harray_free(map->members); -+ kfree(map); -+ -+ set->data = NULL; -+} -+ -+static void flush(struct ip_set *set) -+{ -+ struct ip_set_iphash *map = (struct ip_set_iphash *) set->data; -+ harray_flush(map->members, map->hashsize, sizeof(ip_set_ip_t)); -+ map->elements = 0; -+} -+ -+static void list_header(const struct ip_set *set, void *data) -+{ -+ struct ip_set_iphash *map = (struct ip_set_iphash *) set->data; -+ struct ip_set_req_iphash_create *header = -+ (struct ip_set_req_iphash_create *) data; -+ -+ header->hashsize = map->hashsize; -+ header->probes = map->probes; -+ header->resize = map->resize; -+ header->netmask = map->netmask; -+} -+ -+static int list_members_size(const struct ip_set *set) -+{ -+ struct ip_set_iphash *map = (struct ip_set_iphash *) set->data; -+ -+ return (map->hashsize * sizeof(ip_set_ip_t)); -+} -+ -+static void list_members(const struct ip_set *set, void *data) -+{ -+ struct ip_set_iphash *map = (struct ip_set_iphash *) set->data; -+ ip_set_ip_t i, *elem; -+ -+ for (i = 0; i < map->hashsize; i++) { -+ elem = HARRAY_ELEM(map->members, ip_set_ip_t *, i); -+ ((ip_set_ip_t *)data)[i] = *elem; -+ } -+} -+ -+static struct ip_set_type ip_set_iphash = { -+ .typename = SETTYPE_NAME, -+ .features = IPSET_TYPE_IP | IPSET_DATA_SINGLE, -+ .protocol_version = IP_SET_PROTOCOL_VERSION, -+ .create = &create, -+ .destroy = &destroy, -+ .flush = &flush, -+ .reqsize = sizeof(struct ip_set_req_iphash), -+ .addip = &addip, -+ .addip_kernel = &addip_kernel, -+ .retry = &retry, -+ .delip = &delip, -+ .delip_kernel = &delip_kernel, -+ .testip = &testip, -+ .testip_kernel = &testip_kernel, -+ .header_size = sizeof(struct ip_set_req_iphash_create), -+ .list_header = &list_header, -+ .list_members_size = &list_members_size, -+ .list_members = &list_members, -+ .me = THIS_MODULE, -+}; -+ -+MODULE_LICENSE("GPL"); -+MODULE_AUTHOR("Jozsef Kadlecsik <kadlec@blackhole.kfki.hu>"); -+MODULE_DESCRIPTION("iphash type of IP sets"); -+module_param(limit, int, 0600); -+MODULE_PARM_DESC(limit, "maximal number of elements stored in the sets"); -+ -+static int __init ip_set_iphash_init(void) -+{ -+ return ip_set_register_set_type(&ip_set_iphash); -+} -+ -+static void __exit ip_set_iphash_fini(void) -+{ -+ /* FIXME: possible race with ip_set_create() */ -+ ip_set_unregister_set_type(&ip_set_iphash); -+} -+ -+module_init(ip_set_iphash_init); -+module_exit(ip_set_iphash_fini); ---- /dev/null -+++ b/net/ipv4/netfilter/ip_set_ipmap.c -@@ -0,0 +1,336 @@ -+/* Copyright (C) 2000-2002 Joakim Axelsson <gozem@linux.nu> -+ * Patrick Schaaf <bof@bof.de> -+ * Copyright (C) 2003-2004 Jozsef Kadlecsik <kadlec@blackhole.kfki.hu> -+ * -+ * This program is free software; you can redistribute it and/or modify -+ * it under the terms of the GNU General Public License version 2 as -+ * published by the Free Software Foundation. -+ */ -+ -+/* Kernel module implementing an IP set type: the single bitmap type */ -+ -+#include <linux/module.h> -+#include <linux/ip.h> -+#include <linux/skbuff.h> -+#include <linux/version.h> -+#include <linux/netfilter_ipv4/ip_tables.h> -+#include <linux/netfilter_ipv4/ip_set.h> -+#include <linux/errno.h> -+#include <asm/uaccess.h> -+#include <asm/bitops.h> -+#include <linux/spinlock.h> -+ -+#include <linux/netfilter_ipv4/ip_set_ipmap.h> -+ -+static inline ip_set_ip_t -+ip_to_id(const struct ip_set_ipmap *map, ip_set_ip_t ip) -+{ -+ return (ip - map->first_ip)/map->hosts; -+} -+ -+static inline int -+__testip(struct ip_set *set, ip_set_ip_t ip, ip_set_ip_t *hash_ip) -+{ -+ struct ip_set_ipmap *map = (struct ip_set_ipmap *) set->data; -+ -+ if (ip < map->first_ip || ip > map->last_ip) -+ return -ERANGE; -+ -+ *hash_ip = ip & map->netmask; -+ DP("set: %s, ip:%u.%u.%u.%u, %u.%u.%u.%u", -+ set->name, HIPQUAD(ip), HIPQUAD(*hash_ip)); -+ return !!test_bit(ip_to_id(map, *hash_ip), map->members); -+} -+ -+static int -+testip(struct ip_set *set, const void *data, size_t size, -+ ip_set_ip_t *hash_ip) -+{ -+ struct ip_set_req_ipmap *req = -+ (struct ip_set_req_ipmap *) data; -+ -+ if (size != sizeof(struct ip_set_req_ipmap)) { -+ ip_set_printk("data length wrong (want %zu, have %zu)", -+ sizeof(struct ip_set_req_ipmap), -+ size); -+ return -EINVAL; -+ } -+ return __testip(set, req->ip, hash_ip); -+} -+ -+static int -+testip_kernel(struct ip_set *set, -+ const struct sk_buff *skb, -+ ip_set_ip_t *hash_ip, -+ const u_int32_t *flags, -+ unsigned char index) -+{ -+ int res = __testip(set, -+ ntohl(flags[index] & IPSET_SRC -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,22) -+ ? ip_hdr(skb)->saddr -+ : ip_hdr(skb)->daddr), -+#else -+ ? skb->nh.iph->saddr -+ : skb->nh.iph->daddr), -+#endif -+ hash_ip); -+ return (res < 0 ? 0 : res); -+} -+ -+static inline int -+__addip(struct ip_set *set, ip_set_ip_t ip, ip_set_ip_t *hash_ip) -+{ -+ struct ip_set_ipmap *map = (struct ip_set_ipmap *) set->data; -+ -+ if (ip < map->first_ip || ip > map->last_ip) -+ return -ERANGE; -+ -+ *hash_ip = ip & map->netmask; -+ DP("%u.%u.%u.%u, %u.%u.%u.%u", HIPQUAD(ip), HIPQUAD(*hash_ip)); -+ if (test_and_set_bit(ip_to_id(map, *hash_ip), map->members)) -+ return -EEXIST; -+ -+ return 0; -+} -+ -+static int -+addip(struct ip_set *set, const void *data, size_t size, -+ ip_set_ip_t *hash_ip) -+{ -+ struct ip_set_req_ipmap *req = -+ (struct ip_set_req_ipmap *) data; -+ -+ if (size != sizeof(struct ip_set_req_ipmap)) { -+ ip_set_printk("data length wrong (want %zu, have %zu)", -+ sizeof(struct ip_set_req_ipmap), -+ size); -+ return -EINVAL; -+ } -+ DP("%u.%u.%u.%u", HIPQUAD(req->ip)); -+ return __addip(set, req->ip, hash_ip); -+} -+ -+static int -+addip_kernel(struct ip_set *set, -+ const struct sk_buff *skb, -+ ip_set_ip_t *hash_ip, -+ const u_int32_t *flags, -+ unsigned char index) -+{ -+ return __addip(set, -+ ntohl(flags[index] & IPSET_SRC -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,22) -+ ? ip_hdr(skb)->saddr -+ : ip_hdr(skb)->daddr), -+#else -+ ? skb->nh.iph->saddr -+ : skb->nh.iph->daddr), -+#endif -+ hash_ip); -+} -+ -+static inline int -+__delip(struct ip_set *set, ip_set_ip_t ip, ip_set_ip_t *hash_ip) -+{ -+ struct ip_set_ipmap *map = (struct ip_set_ipmap *) set->data; -+ -+ if (ip < map->first_ip || ip > map->last_ip) -+ return -ERANGE; -+ -+ *hash_ip = ip & map->netmask; -+ DP("%u.%u.%u.%u, %u.%u.%u.%u", HIPQUAD(ip), HIPQUAD(*hash_ip)); -+ if (!test_and_clear_bit(ip_to_id(map, *hash_ip), map->members)) -+ return -EEXIST; -+ -+ return 0; -+} -+ -+static int -+delip(struct ip_set *set, const void *data, size_t size, -+ ip_set_ip_t *hash_ip) -+{ -+ struct ip_set_req_ipmap *req = -+ (struct ip_set_req_ipmap *) data; -+ -+ if (size != sizeof(struct ip_set_req_ipmap)) { -+ ip_set_printk("data length wrong (want %zu, have %zu)", -+ sizeof(struct ip_set_req_ipmap), -+ size); -+ return -EINVAL; -+ } -+ return __delip(set, req->ip, hash_ip); -+} -+ -+static int -+delip_kernel(struct ip_set *set, -+ const struct sk_buff *skb, -+ ip_set_ip_t *hash_ip, -+ const u_int32_t *flags, -+ unsigned char index) -+{ -+ return __delip(set, -+ ntohl(flags[index] & IPSET_SRC -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,22) -+ ? ip_hdr(skb)->saddr -+ : ip_hdr(skb)->daddr), -+#else -+ ? skb->nh.iph->saddr -+ : skb->nh.iph->daddr), -+#endif -+ hash_ip); -+} -+ -+static int create(struct ip_set *set, const void *data, size_t size) -+{ -+ int newbytes; -+ struct ip_set_req_ipmap_create *req = -+ (struct ip_set_req_ipmap_create *) data; -+ struct ip_set_ipmap *map; -+ -+ if (size != sizeof(struct ip_set_req_ipmap_create)) { -+ ip_set_printk("data length wrong (want %zu, have %zu)", -+ sizeof(struct ip_set_req_ipmap_create), -+ size); -+ return -EINVAL; -+ } -+ -+ DP("from %u.%u.%u.%u to %u.%u.%u.%u", -+ HIPQUAD(req->from), HIPQUAD(req->to)); -+ -+ if (req->from > req->to) { -+ DP("bad ip range"); -+ return -ENOEXEC; -+ } -+ -+ map = kmalloc(sizeof(struct ip_set_ipmap), GFP_KERNEL); -+ if (!map) { -+ DP("out of memory for %d bytes", -+ sizeof(struct ip_set_ipmap)); -+ return -ENOMEM; -+ } -+ map->first_ip = req->from; -+ map->last_ip = req->to; -+ map->netmask = req->netmask; -+ -+ if (req->netmask == 0xFFFFFFFF) { -+ map->hosts = 1; -+ map->sizeid = map->last_ip - map->first_ip + 1; -+ } else { -+ unsigned int mask_bits, netmask_bits; -+ ip_set_ip_t mask; -+ -+ map->first_ip &= map->netmask; /* Should we better bark? */ -+ -+ mask = range_to_mask(map->first_ip, map->last_ip, &mask_bits); -+ netmask_bits = mask_to_bits(map->netmask); -+ -+ if ((!mask && (map->first_ip || map->last_ip != 0xFFFFFFFF)) -+ || netmask_bits <= mask_bits) -+ return -ENOEXEC; -+ -+ DP("mask_bits %u, netmask_bits %u", -+ mask_bits, netmask_bits); -+ map->hosts = 2 << (32 - netmask_bits - 1); -+ map->sizeid = 2 << (netmask_bits - mask_bits - 1); -+ } -+ if (map->sizeid > MAX_RANGE + 1) { -+ ip_set_printk("range too big (max %d addresses)", -+ MAX_RANGE+1); -+ kfree(map); -+ return -ENOEXEC; -+ } -+ DP("hosts %u, sizeid %u", map->hosts, map->sizeid); -+ newbytes = bitmap_bytes(0, map->sizeid - 1); -+ map->members = kmalloc(newbytes, GFP_KERNEL); -+ if (!map->members) { -+ DP("out of memory for %d bytes", newbytes); -+ kfree(map); -+ return -ENOMEM; -+ } -+ memset(map->members, 0, newbytes); -+ -+ set->data = map; -+ return 0; -+} -+ -+static void destroy(struct ip_set *set) -+{ -+ struct ip_set_ipmap *map = (struct ip_set_ipmap *) set->data; -+ -+ kfree(map->members); -+ kfree(map); -+ -+ set->data = NULL; -+} -+ -+static void flush(struct ip_set *set) -+{ -+ struct ip_set_ipmap *map = (struct ip_set_ipmap *) set->data; -+ memset(map->members, 0, bitmap_bytes(0, map->sizeid - 1)); -+} -+ -+static void list_header(const struct ip_set *set, void *data) -+{ -+ struct ip_set_ipmap *map = (struct ip_set_ipmap *) set->data; -+ struct ip_set_req_ipmap_create *header = -+ (struct ip_set_req_ipmap_create *) data; -+ -+ header->from = map->first_ip; -+ header->to = map->last_ip; -+ header->netmask = map->netmask; -+} -+ -+static int list_members_size(const struct ip_set *set) -+{ -+ struct ip_set_ipmap *map = (struct ip_set_ipmap *) set->data; -+ -+ return bitmap_bytes(0, map->sizeid - 1); -+} -+ -+static void list_members(const struct ip_set *set, void *data) -+{ -+ struct ip_set_ipmap *map = (struct ip_set_ipmap *) set->data; -+ int bytes = bitmap_bytes(0, map->sizeid - 1); -+ -+ memcpy(data, map->members, bytes); -+} -+ -+static struct ip_set_type ip_set_ipmap = { -+ .typename = SETTYPE_NAME, -+ .features = IPSET_TYPE_IP | IPSET_DATA_SINGLE, -+ .protocol_version = IP_SET_PROTOCOL_VERSION, -+ .create = &create, -+ .destroy = &destroy, -+ .flush = &flush, -+ .reqsize = sizeof(struct ip_set_req_ipmap), -+ .addip = &addip, -+ .addip_kernel = &addip_kernel, -+ .delip = &delip, -+ .delip_kernel = &delip_kernel, -+ .testip = &testip, -+ .testip_kernel = &testip_kernel, -+ .header_size = sizeof(struct ip_set_req_ipmap_create), -+ .list_header = &list_header, -+ .list_members_size = &list_members_size, -+ .list_members = &list_members, -+ .me = THIS_MODULE, -+}; -+ -+MODULE_LICENSE("GPL"); -+MODULE_AUTHOR("Jozsef Kadlecsik <kadlec@blackhole.kfki.hu>"); -+MODULE_DESCRIPTION("ipmap type of IP sets"); -+ -+static int __init ip_set_ipmap_init(void) -+{ -+ return ip_set_register_set_type(&ip_set_ipmap); -+} -+ -+static void __exit ip_set_ipmap_fini(void) -+{ -+ /* FIXME: possible race with ip_set_create() */ -+ ip_set_unregister_set_type(&ip_set_ipmap); -+} -+ -+module_init(ip_set_ipmap_init); -+module_exit(ip_set_ipmap_fini); ---- /dev/null -+++ b/net/ipv4/netfilter/ip_set_ipporthash.c -@@ -0,0 +1,581 @@ -+/* Copyright (C) 2003-2004 Jozsef Kadlecsik <kadlec@blackhole.kfki.hu> -+ * -+ * This program is free software; you can redistribute it and/or modify -+ * it under the terms of the GNU General Public License version 2 as -+ * published by the Free Software Foundation. -+ */ -+ -+/* Kernel module implementing an ip+port hash set */ -+ -+#include <linux/module.h> -+#include <linux/ip.h> -+#include <linux/tcp.h> -+#include <linux/udp.h> -+#include <linux/skbuff.h> -+#include <linux/version.h> -+#include <linux/jhash.h> -+#include <linux/netfilter_ipv4/ip_tables.h> -+#include <linux/netfilter_ipv4/ip_set.h> -+#include <linux/errno.h> -+#include <asm/uaccess.h> -+#include <asm/bitops.h> -+#include <linux/spinlock.h> -+#include <linux/vmalloc.h> -+#include <linux/random.h> -+ -+#include <net/ip.h> -+ -+#include <linux/netfilter_ipv4/ip_set_malloc.h> -+#include <linux/netfilter_ipv4/ip_set_ipporthash.h> -+ -+static int limit = MAX_RANGE; -+ -+/* We must handle non-linear skbs */ -+static inline ip_set_ip_t -+get_port(const struct sk_buff *skb, u_int32_t flags) -+{ -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,22) -+ struct iphdr *iph = ip_hdr(skb); -+#else -+ struct iphdr *iph = skb->nh.iph; -+#endif -+ u_int16_t offset = ntohs(iph->frag_off) & IP_OFFSET; -+ -+ switch (iph->protocol) { -+ case IPPROTO_TCP: { -+ struct tcphdr tcph; -+ -+ /* See comments at tcp_match in ip_tables.c */ -+ if (offset) -+ return INVALID_PORT; -+ -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,22) -+ if (skb_copy_bits(skb, ip_hdr(skb)->ihl*4, &tcph, sizeof(tcph)) < 0) -+#else -+ if (skb_copy_bits(skb, skb->nh.iph->ihl*4, &tcph, sizeof(tcph)) < 0) -+#endif -+ /* No choice either */ -+ return INVALID_PORT; -+ -+ return ntohs(flags & IPSET_SRC ? -+ tcph.source : tcph.dest); -+ } -+ case IPPROTO_UDP: { -+ struct udphdr udph; -+ -+ if (offset) -+ return INVALID_PORT; -+ -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,22) -+ if (skb_copy_bits(skb, ip_hdr(skb)->ihl*4, &udph, sizeof(udph)) < 0) -+#else -+ if (skb_copy_bits(skb, skb->nh.iph->ihl*4, &udph, sizeof(udph)) < 0) -+#endif -+ /* No choice either */ -+ return INVALID_PORT; -+ -+ return ntohs(flags & IPSET_SRC ? -+ udph.source : udph.dest); -+ } -+ default: -+ return INVALID_PORT; -+ } -+} -+ -+static inline __u32 -+jhash_ip(const struct ip_set_ipporthash *map, uint16_t i, ip_set_ip_t ip) -+{ -+ return jhash_1word(ip, *(((uint32_t *) map->initval) + i)); -+} -+ -+#define HASH_IP(map, ip, port) (port + ((ip - ((map)->first_ip)) << 16)) -+ -+static inline __u32 -+hash_id(struct ip_set *set, ip_set_ip_t ip, ip_set_ip_t port, -+ ip_set_ip_t *hash_ip) -+{ -+ struct ip_set_ipporthash *map = -+ (struct ip_set_ipporthash *) set->data; -+ __u32 id; -+ u_int16_t i; -+ ip_set_ip_t *elem; -+ -+ *hash_ip = HASH_IP(map, ip, port); -+ DP("set: %s, ipport:%u.%u.%u.%u:%u, %u.%u.%u.%u", -+ set->name, HIPQUAD(ip), port, HIPQUAD(*hash_ip)); -+ -+ for (i = 0; i < map->probes; i++) { -+ id = jhash_ip(map, i, *hash_ip) % map->hashsize; -+ DP("hash key: %u", id); -+ elem = HARRAY_ELEM(map->members, ip_set_ip_t *, id); -+ if (*elem == *hash_ip) -+ return id; -+ /* No shortcut at testing - there can be deleted -+ * entries. */ -+ } -+ return UINT_MAX; -+} -+ -+static inline int -+__testip(struct ip_set *set, ip_set_ip_t ip, ip_set_ip_t port, -+ ip_set_ip_t *hash_ip) -+{ -+ struct ip_set_ipporthash *map = (struct ip_set_ipporthash *) set->data; -+ -+ if (ip < map->first_ip || ip > map->last_ip) -+ return -ERANGE; -+ -+ return (hash_id(set, ip, port, hash_ip) != UINT_MAX); -+} -+ -+static int -+testip(struct ip_set *set, const void *data, size_t size, -+ ip_set_ip_t *hash_ip) -+{ -+ struct ip_set_req_ipporthash *req = -+ (struct ip_set_req_ipporthash *) data; -+ -+ if (size != sizeof(struct ip_set_req_ipporthash)) { -+ ip_set_printk("data length wrong (want %zu, have %zu)", -+ sizeof(struct ip_set_req_ipporthash), -+ size); -+ return -EINVAL; -+ } -+ return __testip(set, req->ip, req->port, hash_ip); -+} -+ -+static int -+testip_kernel(struct ip_set *set, -+ const struct sk_buff *skb, -+ ip_set_ip_t *hash_ip, -+ const u_int32_t *flags, -+ unsigned char index) -+{ -+ ip_set_ip_t port; -+ int res; -+ -+ if (flags[index+1] == 0) -+ return 0; -+ -+ port = get_port(skb, flags[index+1]); -+ -+ DP("flag: %s src: %u.%u.%u.%u dst: %u.%u.%u.%u", -+ flags[index] & IPSET_SRC ? "SRC" : "DST", -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,22) -+ NIPQUAD(ip_hdr(skb)->saddr), -+ NIPQUAD(ip_hdr(skb)->daddr)); -+#else -+ NIPQUAD(skb->nh.iph->saddr), -+ NIPQUAD(skb->nh.iph->daddr)); -+#endif -+ DP("flag %s port %u", -+ flags[index+1] & IPSET_SRC ? "SRC" : "DST", -+ port); -+ if (port == INVALID_PORT) -+ return 0; -+ -+ res = __testip(set, -+ ntohl(flags[index] & IPSET_SRC -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,22) -+ ? ip_hdr(skb)->saddr -+ : ip_hdr(skb)->daddr), -+#else -+ ? skb->nh.iph->saddr -+ : skb->nh.iph->daddr), -+#endif -+ port, -+ hash_ip); -+ return (res < 0 ? 0 : res); -+ -+} -+ -+static inline int -+__add_haship(struct ip_set_ipporthash *map, ip_set_ip_t hash_ip) -+{ -+ __u32 probe; -+ u_int16_t i; -+ ip_set_ip_t *elem; -+ -+ for (i = 0; i < map->probes; i++) { -+ probe = jhash_ip(map, i, hash_ip) % map->hashsize; -+ elem = HARRAY_ELEM(map->members, ip_set_ip_t *, probe); -+ if (*elem == hash_ip) -+ return -EEXIST; -+ if (!*elem) { -+ *elem = hash_ip; -+ map->elements++; -+ return 0; -+ } -+ } -+ /* Trigger rehashing */ -+ return -EAGAIN; -+} -+ -+static inline int -+__addip(struct ip_set_ipporthash *map, ip_set_ip_t ip, ip_set_ip_t port, -+ ip_set_ip_t *hash_ip) -+{ -+ if (map->elements > limit) -+ return -ERANGE; -+ if (ip < map->first_ip || ip > map->last_ip) -+ return -ERANGE; -+ -+ *hash_ip = HASH_IP(map, ip, port); -+ -+ return __add_haship(map, *hash_ip); -+} -+ -+static int -+addip(struct ip_set *set, const void *data, size_t size, -+ ip_set_ip_t *hash_ip) -+{ -+ struct ip_set_req_ipporthash *req = -+ (struct ip_set_req_ipporthash *) data; -+ -+ if (size != sizeof(struct ip_set_req_ipporthash)) { -+ ip_set_printk("data length wrong (want %zu, have %zu)", -+ sizeof(struct ip_set_req_ipporthash), -+ size); -+ return -EINVAL; -+ } -+ return __addip((struct ip_set_ipporthash *) set->data, -+ req->ip, req->port, hash_ip); -+} -+ -+static int -+addip_kernel(struct ip_set *set, -+ const struct sk_buff *skb, -+ ip_set_ip_t *hash_ip, -+ const u_int32_t *flags, -+ unsigned char index) -+{ -+ ip_set_ip_t port; -+ -+ if (flags[index+1] == 0) -+ return -EINVAL; -+ -+ port = get_port(skb, flags[index+1]); -+ -+ DP("flag: %s src: %u.%u.%u.%u dst: %u.%u.%u.%u", -+ flags[index] & IPSET_SRC ? "SRC" : "DST", -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,22) -+ NIPQUAD(ip_hdr(skb)->saddr), -+ NIPQUAD(ip_hdr(skb)->daddr)); -+#else -+ NIPQUAD(skb->nh.iph->saddr), -+ NIPQUAD(skb->nh.iph->daddr)); -+#endif -+ DP("flag %s port %u", -+ flags[index+1] & IPSET_SRC ? "SRC" : "DST", -+ port); -+ if (port == INVALID_PORT) -+ return -EINVAL; -+ -+ return __addip((struct ip_set_ipporthash *) set->data, -+ ntohl(flags[index] & IPSET_SRC -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,22) -+ ? ip_hdr(skb)->saddr -+ : ip_hdr(skb)->daddr), -+#else -+ ? skb->nh.iph->saddr -+ : skb->nh.iph->daddr), -+#endif -+ port, -+ hash_ip); -+} -+ -+static int retry(struct ip_set *set) -+{ -+ struct ip_set_ipporthash *map = (struct ip_set_ipporthash *) set->data; -+ ip_set_ip_t *elem; -+ void *members; -+ u_int32_t i, hashsize = map->hashsize; -+ int res; -+ struct ip_set_ipporthash *tmp; -+ -+ if (map->resize == 0) -+ return -ERANGE; -+ -+ again: -+ res = 0; -+ -+ /* Calculate new hash size */ -+ hashsize += (hashsize * map->resize)/100; -+ if (hashsize == map->hashsize) -+ hashsize++; -+ -+ ip_set_printk("rehashing of set %s triggered: " -+ "hashsize grows from %u to %u", -+ set->name, map->hashsize, hashsize); -+ -+ tmp = kmalloc(sizeof(struct ip_set_ipporthash) -+ + map->probes * sizeof(uint32_t), GFP_ATOMIC); -+ if (!tmp) { -+ DP("out of memory for %d bytes", -+ sizeof(struct ip_set_ipporthash) -+ + map->probes * sizeof(uint32_t)); -+ return -ENOMEM; -+ } -+ tmp->members = harray_malloc(hashsize, sizeof(ip_set_ip_t), GFP_ATOMIC); -+ if (!tmp->members) { -+ DP("out of memory for %d bytes", hashsize * sizeof(ip_set_ip_t)); -+ kfree(tmp); -+ return -ENOMEM; -+ } -+ tmp->hashsize = hashsize; -+ tmp->elements = 0; -+ tmp->probes = map->probes; -+ tmp->resize = map->resize; -+ tmp->first_ip = map->first_ip; -+ tmp->last_ip = map->last_ip; -+ memcpy(tmp->initval, map->initval, map->probes * sizeof(uint32_t)); -+ -+ write_lock_bh(&set->lock); -+ map = (struct ip_set_ipporthash *) set->data; /* Play safe */ -+ for (i = 0; i < map->hashsize && res == 0; i++) { -+ elem = HARRAY_ELEM(map->members, ip_set_ip_t *, i); -+ if (*elem) -+ res = __add_haship(tmp, *elem); -+ } -+ if (res) { -+ /* Failure, try again */ -+ write_unlock_bh(&set->lock); -+ harray_free(tmp->members); -+ kfree(tmp); -+ goto again; -+ } -+ -+ /* Success at resizing! */ -+ members = map->members; -+ -+ map->hashsize = tmp->hashsize; -+ map->members = tmp->members; -+ write_unlock_bh(&set->lock); -+ -+ harray_free(members); -+ kfree(tmp); -+ -+ return 0; -+} -+ -+static inline int -+__delip(struct ip_set *set, ip_set_ip_t ip, ip_set_ip_t port, -+ ip_set_ip_t *hash_ip) -+{ -+ struct ip_set_ipporthash *map = (struct ip_set_ipporthash *) set->data; -+ ip_set_ip_t id; -+ ip_set_ip_t *elem; -+ -+ if (ip < map->first_ip || ip > map->last_ip) -+ return -ERANGE; -+ -+ id = hash_id(set, ip, port, hash_ip); -+ -+ if (id == UINT_MAX) -+ return -EEXIST; -+ -+ elem = HARRAY_ELEM(map->members, ip_set_ip_t *, id); -+ *elem = 0; -+ map->elements--; -+ -+ return 0; -+} -+ -+static int -+delip(struct ip_set *set, const void *data, size_t size, -+ ip_set_ip_t *hash_ip) -+{ -+ struct ip_set_req_ipporthash *req = -+ (struct ip_set_req_ipporthash *) data; -+ -+ if (size != sizeof(struct ip_set_req_ipporthash)) { -+ ip_set_printk("data length wrong (want %zu, have %zu)", -+ sizeof(struct ip_set_req_ipporthash), -+ size); -+ return -EINVAL; -+ } -+ return __delip(set, req->ip, req->port, hash_ip); -+} -+ -+static int -+delip_kernel(struct ip_set *set, -+ const struct sk_buff *skb, -+ ip_set_ip_t *hash_ip, -+ const u_int32_t *flags, -+ unsigned char index) -+{ -+ ip_set_ip_t port; -+ -+ if (flags[index+1] == 0) -+ return -EINVAL; -+ -+ port = get_port(skb, flags[index+1]); -+ -+ DP("flag: %s src: %u.%u.%u.%u dst: %u.%u.%u.%u", -+ flags[index] & IPSET_SRC ? "SRC" : "DST", -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,22) -+ NIPQUAD(ip_hdr(skb)->saddr), -+ NIPQUAD(ip_hdr(skb)->daddr)); -+#else -+ NIPQUAD(skb->nh.iph->saddr), -+ NIPQUAD(skb->nh.iph->daddr)); -+#endif -+ DP("flag %s port %u", -+ flags[index+1] & IPSET_SRC ? "SRC" : "DST", -+ port); -+ if (port == INVALID_PORT) -+ return -EINVAL; -+ -+ return __delip(set, -+ ntohl(flags[index] & IPSET_SRC -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,22) -+ ? ip_hdr(skb)->saddr -+ : ip_hdr(skb)->daddr), -+#else -+ ? skb->nh.iph->saddr -+ : skb->nh.iph->daddr), -+#endif -+ port, -+ hash_ip); -+} -+ -+static int create(struct ip_set *set, const void *data, size_t size) -+{ -+ struct ip_set_req_ipporthash_create *req = -+ (struct ip_set_req_ipporthash_create *) data; -+ struct ip_set_ipporthash *map; -+ uint16_t i; -+ -+ if (size != sizeof(struct ip_set_req_ipporthash_create)) { -+ ip_set_printk("data length wrong (want %zu, have %zu)", -+ sizeof(struct ip_set_req_ipporthash_create), -+ size); -+ return -EINVAL; -+ } -+ -+ if (req->hashsize < 1) { -+ ip_set_printk("hashsize too small"); -+ return -ENOEXEC; -+ } -+ -+ if (req->probes < 1) { -+ ip_set_printk("probes too small"); -+ return -ENOEXEC; -+ } -+ -+ map = kmalloc(sizeof(struct ip_set_ipporthash) -+ + req->probes * sizeof(uint32_t), GFP_KERNEL); -+ if (!map) { -+ DP("out of memory for %d bytes", -+ sizeof(struct ip_set_ipporthash) -+ + req->probes * sizeof(uint32_t)); -+ return -ENOMEM; -+ } -+ for (i = 0; i < req->probes; i++) -+ get_random_bytes(((uint32_t *) map->initval)+i, 4); -+ map->elements = 0; -+ map->hashsize = req->hashsize; -+ map->probes = req->probes; -+ map->resize = req->resize; -+ map->first_ip = req->from; -+ map->last_ip = req->to; -+ map->members = harray_malloc(map->hashsize, sizeof(ip_set_ip_t), GFP_KERNEL); -+ if (!map->members) { -+ DP("out of memory for %d bytes", map->hashsize * sizeof(ip_set_ip_t)); -+ kfree(map); -+ return -ENOMEM; -+ } -+ -+ set->data = map; -+ return 0; -+} -+ -+static void destroy(struct ip_set *set) -+{ -+ struct ip_set_ipporthash *map = (struct ip_set_ipporthash *) set->data; -+ -+ harray_free(map->members); -+ kfree(map); -+ -+ set->data = NULL; -+} -+ -+static void flush(struct ip_set *set) -+{ -+ struct ip_set_ipporthash *map = (struct ip_set_ipporthash *) set->data; -+ harray_flush(map->members, map->hashsize, sizeof(ip_set_ip_t)); -+ map->elements = 0; -+} -+ -+static void list_header(const struct ip_set *set, void *data) -+{ -+ struct ip_set_ipporthash *map = (struct ip_set_ipporthash *) set->data; -+ struct ip_set_req_ipporthash_create *header = -+ (struct ip_set_req_ipporthash_create *) data; -+ -+ header->hashsize = map->hashsize; -+ header->probes = map->probes; -+ header->resize = map->resize; -+ header->from = map->first_ip; -+ header->to = map->last_ip; -+} -+ -+static int list_members_size(const struct ip_set *set) -+{ -+ struct ip_set_ipporthash *map = (struct ip_set_ipporthash *) set->data; -+ -+ return (map->hashsize * sizeof(ip_set_ip_t)); -+} -+ -+static void list_members(const struct ip_set *set, void *data) -+{ -+ struct ip_set_ipporthash *map = (struct ip_set_ipporthash *) set->data; -+ ip_set_ip_t i, *elem; -+ -+ for (i = 0; i < map->hashsize; i++) { -+ elem = HARRAY_ELEM(map->members, ip_set_ip_t *, i); -+ ((ip_set_ip_t *)data)[i] = *elem; -+ } -+} -+ -+static struct ip_set_type ip_set_ipporthash = { -+ .typename = SETTYPE_NAME, -+ .features = IPSET_TYPE_IP | IPSET_TYPE_PORT | IPSET_DATA_DOUBLE, -+ .protocol_version = IP_SET_PROTOCOL_VERSION, -+ .create = &create, -+ .destroy = &destroy, -+ .flush = &flush, -+ .reqsize = sizeof(struct ip_set_req_ipporthash), -+ .addip = &addip, -+ .addip_kernel = &addip_kernel, -+ .retry = &retry, -+ .delip = &delip, -+ .delip_kernel = &delip_kernel, -+ .testip = &testip, -+ .testip_kernel = &testip_kernel, -+ .header_size = sizeof(struct ip_set_req_ipporthash_create), -+ .list_header = &list_header, -+ .list_members_size = &list_members_size, -+ .list_members = &list_members, -+ .me = THIS_MODULE, -+}; -+ -+MODULE_LICENSE("GPL"); -+MODULE_AUTHOR("Jozsef Kadlecsik <kadlec@blackhole.kfki.hu>"); -+MODULE_DESCRIPTION("ipporthash type of IP sets"); -+module_param(limit, int, 0600); -+MODULE_PARM_DESC(limit, "maximal number of elements stored in the sets"); -+ -+static int __init ip_set_ipporthash_init(void) -+{ -+ return ip_set_register_set_type(&ip_set_ipporthash); -+} -+ -+static void __exit ip_set_ipporthash_fini(void) -+{ -+ /* FIXME: possible race with ip_set_create() */ -+ ip_set_unregister_set_type(&ip_set_ipporthash); -+} -+ -+module_init(ip_set_ipporthash_init); -+module_exit(ip_set_ipporthash_fini); ---- /dev/null -+++ b/net/ipv4/netfilter/ip_set_iptree.c -@@ -0,0 +1,612 @@ -+/* Copyright (C) 2005 Jozsef Kadlecsik <kadlec@blackhole.kfki.hu> -+ * -+ * This program is free software; you can redistribute it and/or modify -+ * it under the terms of the GNU General Public License version 2 as -+ * published by the Free Software Foundation. -+ */ -+ -+/* Kernel module implementing an IP set type: the iptree type */ -+ -+#include <linux/version.h> -+#include <linux/module.h> -+#include <linux/ip.h> -+#include <linux/skbuff.h> -+#include <linux/slab.h> -+#include <linux/delay.h> -+#include <linux/netfilter_ipv4/ip_tables.h> -+#include <linux/netfilter_ipv4/ip_set.h> -+#include <linux/errno.h> -+#include <asm/uaccess.h> -+#include <asm/bitops.h> -+#include <linux/spinlock.h> -+ -+/* Backward compatibility */ -+#ifndef __nocast -+#define __nocast -+#endif -+ -+#include <linux/netfilter_ipv4/ip_set_iptree.h> -+ -+static int limit = MAX_RANGE; -+ -+/* Garbage collection interval in seconds: */ -+#define IPTREE_GC_TIME 5*60 -+/* Sleep so many milliseconds before trying again -+ * to delete the gc timer at destroying/flushing a set */ -+#define IPTREE_DESTROY_SLEEP 100 -+ -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,21) -+static struct kmem_cache *branch_cachep; -+static struct kmem_cache *leaf_cachep; -+#else -+static kmem_cache_t *branch_cachep; -+static kmem_cache_t *leaf_cachep; -+#endif -+ -+#if defined(__LITTLE_ENDIAN) -+#define ABCD(a,b,c,d,addrp) do { \ -+ a = ((unsigned char *)addrp)[3]; \ -+ b = ((unsigned char *)addrp)[2]; \ -+ c = ((unsigned char *)addrp)[1]; \ -+ d = ((unsigned char *)addrp)[0]; \ -+} while (0) -+#elif defined(__BIG_ENDIAN) -+#define ABCD(a,b,c,d,addrp) do { \ -+ a = ((unsigned char *)addrp)[0]; \ -+ b = ((unsigned char *)addrp)[1]; \ -+ c = ((unsigned char *)addrp)[2]; \ -+ d = ((unsigned char *)addrp)[3]; \ -+} while (0) -+#else -+#error "Please fix asm/byteorder.h" -+#endif /* __LITTLE_ENDIAN */ -+ -+#define TESTIP_WALK(map, elem, branch) do { \ -+ if ((map)->tree[elem]) { \ -+ branch = (map)->tree[elem]; \ -+ } else \ -+ return 0; \ -+} while (0) -+ -+static inline int -+__testip(struct ip_set *set, ip_set_ip_t ip, ip_set_ip_t *hash_ip) -+{ -+ struct ip_set_iptree *map = (struct ip_set_iptree *) set->data; -+ struct ip_set_iptreeb *btree; -+ struct ip_set_iptreec *ctree; -+ struct ip_set_iptreed *dtree; -+ unsigned char a,b,c,d; -+ -+ if (!ip) -+ return -ERANGE; -+ -+ *hash_ip = ip; -+ ABCD(a, b, c, d, hash_ip); -+ DP("%u %u %u %u timeout %u", a, b, c, d, map->timeout); -+ TESTIP_WALK(map, a, btree); -+ TESTIP_WALK(btree, b, ctree); -+ TESTIP_WALK(ctree, c, dtree); -+ DP("%lu %lu", dtree->expires[d], jiffies); -+ return dtree->expires[d] -+ && (!map->timeout -+ || time_after(dtree->expires[d], jiffies)); -+} -+ -+static int -+testip(struct ip_set *set, const void *data, size_t size, -+ ip_set_ip_t *hash_ip) -+{ -+ struct ip_set_req_iptree *req = -+ (struct ip_set_req_iptree *) data; -+ -+ if (size != sizeof(struct ip_set_req_iptree)) { -+ ip_set_printk("data length wrong (want %zu, have %zu)", -+ sizeof(struct ip_set_req_iptree), -+ size); -+ return -EINVAL; -+ } -+ return __testip(set, req->ip, hash_ip); -+} -+ -+static int -+testip_kernel(struct ip_set *set, -+ const struct sk_buff *skb, -+ ip_set_ip_t *hash_ip, -+ const u_int32_t *flags, -+ unsigned char index) -+{ -+ int res; -+ -+ DP("flag: %s src: %u.%u.%u.%u dst: %u.%u.%u.%u", -+ flags[index] & IPSET_SRC ? "SRC" : "DST", -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,22) -+ NIPQUAD(ip_hdr(skb)->saddr), -+ NIPQUAD(ip_hdr(skb)->daddr)); -+#else -+ NIPQUAD(skb->nh.iph->saddr), -+ NIPQUAD(skb->nh.iph->daddr)); -+#endif -+ -+ res = __testip(set, -+ ntohl(flags[index] & IPSET_SRC -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,22) -+ ? ip_hdr(skb)->saddr -+ : ip_hdr(skb)->daddr), -+#else -+ ? skb->nh.iph->saddr -+ : skb->nh.iph->daddr), -+#endif -+ hash_ip); -+ return (res < 0 ? 0 : res); -+} -+ -+#define ADDIP_WALK(map, elem, branch, type, cachep) do { \ -+ if ((map)->tree[elem]) { \ -+ DP("found %u", elem); \ -+ branch = (map)->tree[elem]; \ -+ } else { \ -+ branch = (type *) \ -+ kmem_cache_alloc(cachep, GFP_ATOMIC); \ -+ if (branch == NULL) \ -+ return -ENOMEM; \ -+ memset(branch, 0, sizeof(*branch)); \ -+ (map)->tree[elem] = branch; \ -+ DP("alloc %u", elem); \ -+ } \ -+} while (0) -+ -+static inline int -+__addip(struct ip_set *set, ip_set_ip_t ip, unsigned int timeout, -+ ip_set_ip_t *hash_ip) -+{ -+ struct ip_set_iptree *map = (struct ip_set_iptree *) set->data; -+ struct ip_set_iptreeb *btree; -+ struct ip_set_iptreec *ctree; -+ struct ip_set_iptreed *dtree; -+ unsigned char a,b,c,d; -+ int ret = 0; -+ -+ if (!ip || map->elements >= limit) -+ /* We could call the garbage collector -+ * but it's probably overkill */ -+ return -ERANGE; -+ -+ *hash_ip = ip; -+ ABCD(a, b, c, d, hash_ip); -+ DP("%u %u %u %u timeout %u", a, b, c, d, timeout); -+ ADDIP_WALK(map, a, btree, struct ip_set_iptreeb, branch_cachep); -+ ADDIP_WALK(btree, b, ctree, struct ip_set_iptreec, branch_cachep); -+ ADDIP_WALK(ctree, c, dtree, struct ip_set_iptreed, leaf_cachep); -+ if (dtree->expires[d] -+ && (!map->timeout || time_after(dtree->expires[d], jiffies))) -+ ret = -EEXIST; -+ dtree->expires[d] = map->timeout ? (timeout * HZ + jiffies) : 1; -+ /* Lottery: I won! */ -+ if (dtree->expires[d] == 0) -+ dtree->expires[d] = 1; -+ DP("%u %lu", d, dtree->expires[d]); -+ if (ret == 0) -+ map->elements++; -+ return ret; -+} -+ -+static int -+addip(struct ip_set *set, const void *data, size_t size, -+ ip_set_ip_t *hash_ip) -+{ -+ struct ip_set_iptree *map = (struct ip_set_iptree *) set->data; -+ struct ip_set_req_iptree *req = -+ (struct ip_set_req_iptree *) data; -+ -+ if (size != sizeof(struct ip_set_req_iptree)) { -+ ip_set_printk("data length wrong (want %zu, have %zu)", -+ sizeof(struct ip_set_req_iptree), -+ size); -+ return -EINVAL; -+ } -+ DP("%u.%u.%u.%u %u", HIPQUAD(req->ip), req->timeout); -+ return __addip(set, req->ip, -+ req->timeout ? req->timeout : map->timeout, -+ hash_ip); -+} -+ -+static int -+addip_kernel(struct ip_set *set, -+ const struct sk_buff *skb, -+ ip_set_ip_t *hash_ip, -+ const u_int32_t *flags, -+ unsigned char index) -+{ -+ struct ip_set_iptree *map = (struct ip_set_iptree *) set->data; -+ -+ return __addip(set, -+ ntohl(flags[index] & IPSET_SRC -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,22) -+ ? ip_hdr(skb)->saddr -+ : ip_hdr(skb)->daddr), -+#else -+ ? skb->nh.iph->saddr -+ : skb->nh.iph->daddr), -+#endif -+ map->timeout, -+ hash_ip); -+} -+ -+#define DELIP_WALK(map, elem, branch) do { \ -+ if ((map)->tree[elem]) { \ -+ branch = (map)->tree[elem]; \ -+ } else \ -+ return -EEXIST; \ -+} while (0) -+ -+static inline int -+__delip(struct ip_set *set, ip_set_ip_t ip, ip_set_ip_t *hash_ip) -+{ -+ struct ip_set_iptree *map = (struct ip_set_iptree *) set->data; -+ struct ip_set_iptreeb *btree; -+ struct ip_set_iptreec *ctree; -+ struct ip_set_iptreed *dtree; -+ unsigned char a,b,c,d; -+ -+ if (!ip) -+ return -ERANGE; -+ -+ *hash_ip = ip; -+ ABCD(a, b, c, d, hash_ip); -+ DELIP_WALK(map, a, btree); -+ DELIP_WALK(btree, b, ctree); -+ DELIP_WALK(ctree, c, dtree); -+ -+ if (dtree->expires[d]) { -+ dtree->expires[d] = 0; -+ map->elements--; -+ return 0; -+ } -+ return -EEXIST; -+} -+ -+static int -+delip(struct ip_set *set, const void *data, size_t size, -+ ip_set_ip_t *hash_ip) -+{ -+ struct ip_set_req_iptree *req = -+ (struct ip_set_req_iptree *) data; -+ -+ if (size != sizeof(struct ip_set_req_iptree)) { -+ ip_set_printk("data length wrong (want %zu, have %zu)", -+ sizeof(struct ip_set_req_iptree), -+ size); -+ return -EINVAL; -+ } -+ return __delip(set, req->ip, hash_ip); -+} -+ -+static int -+delip_kernel(struct ip_set *set, -+ const struct sk_buff *skb, -+ ip_set_ip_t *hash_ip, -+ const u_int32_t *flags, -+ unsigned char index) -+{ -+ return __delip(set, -+ ntohl(flags[index] & IPSET_SRC -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,22) -+ ? ip_hdr(skb)->saddr -+ : ip_hdr(skb)->daddr), -+#else -+ ? skb->nh.iph->saddr -+ : skb->nh.iph->daddr), -+#endif -+ hash_ip); -+} -+ -+#define LOOP_WALK_BEGIN(map, i, branch) \ -+ for (i = 0; i < 256; i++) { \ -+ if (!(map)->tree[i]) \ -+ continue; \ -+ branch = (map)->tree[i] -+ -+#define LOOP_WALK_END } -+ -+static void ip_tree_gc(unsigned long ul_set) -+{ -+ struct ip_set *set = (void *) ul_set; -+ struct ip_set_iptree *map = (struct ip_set_iptree *) set->data; -+ struct ip_set_iptreeb *btree; -+ struct ip_set_iptreec *ctree; -+ struct ip_set_iptreed *dtree; -+ unsigned int a,b,c,d; -+ unsigned char i,j,k; -+ -+ i = j = k = 0; -+ DP("gc: %s", set->name); -+ write_lock_bh(&set->lock); -+ LOOP_WALK_BEGIN(map, a, btree); -+ LOOP_WALK_BEGIN(btree, b, ctree); -+ LOOP_WALK_BEGIN(ctree, c, dtree); -+ for (d = 0; d < 256; d++) { -+ if (dtree->expires[d]) { -+ DP("gc: %u %u %u %u: expires %lu jiffies %lu", -+ a, b, c, d, -+ dtree->expires[d], jiffies); -+ if (map->timeout -+ && time_before(dtree->expires[d], jiffies)) { -+ dtree->expires[d] = 0; -+ map->elements--; -+ } else -+ k = 1; -+ } -+ } -+ if (k == 0) { -+ DP("gc: %s: leaf %u %u %u empty", -+ set->name, a, b, c); -+ kmem_cache_free(leaf_cachep, dtree); -+ ctree->tree[c] = NULL; -+ } else { -+ DP("gc: %s: leaf %u %u %u not empty", -+ set->name, a, b, c); -+ j = 1; -+ k = 0; -+ } -+ LOOP_WALK_END; -+ if (j == 0) { -+ DP("gc: %s: branch %u %u empty", -+ set->name, a, b); -+ kmem_cache_free(branch_cachep, ctree); -+ btree->tree[b] = NULL; -+ } else { -+ DP("gc: %s: branch %u %u not empty", -+ set->name, a, b); -+ i = 1; -+ j = k = 0; -+ } -+ LOOP_WALK_END; -+ if (i == 0) { -+ DP("gc: %s: branch %u empty", -+ set->name, a); -+ kmem_cache_free(branch_cachep, btree); -+ map->tree[a] = NULL; -+ } else { -+ DP("gc: %s: branch %u not empty", -+ set->name, a); -+ i = j = k = 0; -+ } -+ LOOP_WALK_END; -+ write_unlock_bh(&set->lock); -+ -+ map->gc.expires = jiffies + map->gc_interval * HZ; -+ add_timer(&map->gc); -+} -+ -+static inline void init_gc_timer(struct ip_set *set) -+{ -+ struct ip_set_iptree *map = (struct ip_set_iptree *) set->data; -+ -+ /* Even if there is no timeout for the entries, -+ * we still have to call gc because delete -+ * do not clean up empty branches */ -+ map->gc_interval = IPTREE_GC_TIME; -+ init_timer(&map->gc); -+ map->gc.data = (unsigned long) set; -+ map->gc.function = ip_tree_gc; -+ map->gc.expires = jiffies + map->gc_interval * HZ; -+ add_timer(&map->gc); -+} -+ -+static int create(struct ip_set *set, const void *data, size_t size) -+{ -+ struct ip_set_req_iptree_create *req = -+ (struct ip_set_req_iptree_create *) data; -+ struct ip_set_iptree *map; -+ -+ if (size != sizeof(struct ip_set_req_iptree_create)) { -+ ip_set_printk("data length wrong (want %zu, have %zu)", -+ sizeof(struct ip_set_req_iptree_create), -+ size); -+ return -EINVAL; -+ } -+ -+ map = kmalloc(sizeof(struct ip_set_iptree), GFP_KERNEL); -+ if (!map) { -+ DP("out of memory for %d bytes", -+ sizeof(struct ip_set_iptree)); -+ return -ENOMEM; -+ } -+ memset(map, 0, sizeof(*map)); -+ map->timeout = req->timeout; -+ map->elements = 0; -+ set->data = map; -+ -+ init_gc_timer(set); -+ -+ return 0; -+} -+ -+static void __flush(struct ip_set_iptree *map) -+{ -+ struct ip_set_iptreeb *btree; -+ struct ip_set_iptreec *ctree; -+ struct ip_set_iptreed *dtree; -+ unsigned int a,b,c; -+ -+ LOOP_WALK_BEGIN(map, a, btree); -+ LOOP_WALK_BEGIN(btree, b, ctree); -+ LOOP_WALK_BEGIN(ctree, c, dtree); -+ kmem_cache_free(leaf_cachep, dtree); -+ LOOP_WALK_END; -+ kmem_cache_free(branch_cachep, ctree); -+ LOOP_WALK_END; -+ kmem_cache_free(branch_cachep, btree); -+ LOOP_WALK_END; -+ map->elements = 0; -+} -+ -+static void destroy(struct ip_set *set) -+{ -+ struct ip_set_iptree *map = (struct ip_set_iptree *) set->data; -+ -+ /* gc might be running */ -+ while (!del_timer(&map->gc)) -+ msleep(IPTREE_DESTROY_SLEEP); -+ __flush(map); -+ kfree(map); -+ set->data = NULL; -+} -+ -+static void flush(struct ip_set *set) -+{ -+ struct ip_set_iptree *map = (struct ip_set_iptree *) set->data; -+ unsigned int timeout = map->timeout; -+ -+ /* gc might be running */ -+ while (!del_timer(&map->gc)) -+ msleep(IPTREE_DESTROY_SLEEP); -+ __flush(map); -+ memset(map, 0, sizeof(*map)); -+ map->timeout = timeout; -+ -+ init_gc_timer(set); -+} -+ -+static void list_header(const struct ip_set *set, void *data) -+{ -+ struct ip_set_iptree *map = (struct ip_set_iptree *) set->data; -+ struct ip_set_req_iptree_create *header = -+ (struct ip_set_req_iptree_create *) data; -+ -+ header->timeout = map->timeout; -+} -+ -+static int list_members_size(const struct ip_set *set) -+{ -+ struct ip_set_iptree *map = (struct ip_set_iptree *) set->data; -+ struct ip_set_iptreeb *btree; -+ struct ip_set_iptreec *ctree; -+ struct ip_set_iptreed *dtree; -+ unsigned int a,b,c,d; -+ unsigned int count = 0; -+ -+ LOOP_WALK_BEGIN(map, a, btree); -+ LOOP_WALK_BEGIN(btree, b, ctree); -+ LOOP_WALK_BEGIN(ctree, c, dtree); -+ for (d = 0; d < 256; d++) { -+ if (dtree->expires[d] -+ && (!map->timeout || time_after(dtree->expires[d], jiffies))) -+ count++; -+ } -+ LOOP_WALK_END; -+ LOOP_WALK_END; -+ LOOP_WALK_END; -+ -+ DP("members %u", count); -+ return (count * sizeof(struct ip_set_req_iptree)); -+} -+ -+static void list_members(const struct ip_set *set, void *data) -+{ -+ struct ip_set_iptree *map = (struct ip_set_iptree *) set->data; -+ struct ip_set_iptreeb *btree; -+ struct ip_set_iptreec *ctree; -+ struct ip_set_iptreed *dtree; -+ unsigned int a,b,c,d; -+ size_t offset = 0; -+ struct ip_set_req_iptree *entry; -+ -+ LOOP_WALK_BEGIN(map, a, btree); -+ LOOP_WALK_BEGIN(btree, b, ctree); -+ LOOP_WALK_BEGIN(ctree, c, dtree); -+ for (d = 0; d < 256; d++) { -+ if (dtree->expires[d] -+ && (!map->timeout || time_after(dtree->expires[d], jiffies))) { -+ entry = (struct ip_set_req_iptree *)(data + offset); -+ entry->ip = ((a << 24) | (b << 16) | (c << 8) | d); -+ entry->timeout = !map->timeout ? 0 -+ : (dtree->expires[d] - jiffies)/HZ; -+ offset += sizeof(struct ip_set_req_iptree); -+ } -+ } -+ LOOP_WALK_END; -+ LOOP_WALK_END; -+ LOOP_WALK_END; -+} -+ -+static struct ip_set_type ip_set_iptree = { -+ .typename = SETTYPE_NAME, -+ .features = IPSET_TYPE_IP | IPSET_DATA_SINGLE, -+ .protocol_version = IP_SET_PROTOCOL_VERSION, -+ .create = &create, -+ .destroy = &destroy, -+ .flush = &flush, -+ .reqsize = sizeof(struct ip_set_req_iptree), -+ .addip = &addip, -+ .addip_kernel = &addip_kernel, -+ .delip = &delip, -+ .delip_kernel = &delip_kernel, -+ .testip = &testip, -+ .testip_kernel = &testip_kernel, -+ .header_size = sizeof(struct ip_set_req_iptree_create), -+ .list_header = &list_header, -+ .list_members_size = &list_members_size, -+ .list_members = &list_members, -+ .me = THIS_MODULE, -+}; -+ -+MODULE_LICENSE("GPL"); -+MODULE_AUTHOR("Jozsef Kadlecsik <kadlec@blackhole.kfki.hu>"); -+MODULE_DESCRIPTION("iptree type of IP sets"); -+module_param(limit, int, 0600); -+MODULE_PARM_DESC(limit, "maximal number of elements stored in the sets"); -+ -+static int __init ip_set_iptree_init(void) -+{ -+ int ret; -+ -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,23) -+ branch_cachep = kmem_cache_create("ip_set_iptreeb", -+ sizeof(struct ip_set_iptreeb), -+ 0, 0, NULL); -+#else -+ branch_cachep = kmem_cache_create("ip_set_iptreeb", -+ sizeof(struct ip_set_iptreeb), -+ 0, 0, NULL, NULL); -+#endif -+ if (!branch_cachep) { -+ printk(KERN_ERR "Unable to create ip_set_iptreeb slab cache\n"); -+ ret = -ENOMEM; -+ goto out; -+ } -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,23) -+ leaf_cachep = kmem_cache_create("ip_set_iptreed", -+ sizeof(struct ip_set_iptreed), -+ 0, 0, NULL); -+#else -+ leaf_cachep = kmem_cache_create("ip_set_iptreed", -+ sizeof(struct ip_set_iptreed), -+ 0, 0, NULL, NULL); -+#endif -+ if (!leaf_cachep) { -+ printk(KERN_ERR "Unable to create ip_set_iptreed slab cache\n"); -+ ret = -ENOMEM; -+ goto free_branch; -+ } -+ ret = ip_set_register_set_type(&ip_set_iptree); -+ if (ret == 0) -+ goto out; -+ -+ kmem_cache_destroy(leaf_cachep); -+ free_branch: -+ kmem_cache_destroy(branch_cachep); -+ out: -+ return ret; -+} -+ -+static void __exit ip_set_iptree_fini(void) -+{ -+ /* FIXME: possible race with ip_set_create() */ -+ ip_set_unregister_set_type(&ip_set_iptree); -+ kmem_cache_destroy(leaf_cachep); -+ kmem_cache_destroy(branch_cachep); -+} -+ -+module_init(ip_set_iptree_init); -+module_exit(ip_set_iptree_fini); ---- /dev/null -+++ b/net/ipv4/netfilter/ip_set_iptreemap.c -@@ -0,0 +1,829 @@ -+/* Copyright (C) 2007 Sven Wegener <sven.wegener@stealer.net> -+ * -+ * This program is free software; you can redistribute it and/or modify it -+ * under the terms of the GNU General Public License version 2 as published by -+ * the Free Software Foundation. -+ */ -+ -+/* This modules implements the iptreemap ipset type. It uses bitmaps to -+ * represent every single IPv4 address as a single bit. The bitmaps are managed -+ * in a tree structure, where the first three octets of an addresses are used -+ * as an index to find the bitmap and the last octet is used as the bit number. -+ */ -+ -+#include <linux/version.h> -+#include <linux/module.h> -+#include <linux/ip.h> -+#include <linux/skbuff.h> -+#include <linux/slab.h> -+#include <linux/delay.h> -+#include <linux/netfilter_ipv4/ip_tables.h> -+#include <linux/netfilter_ipv4/ip_set.h> -+#include <linux/errno.h> -+#include <asm/uaccess.h> -+#include <asm/bitops.h> -+#include <linux/spinlock.h> -+ -+#include <linux/netfilter_ipv4/ip_set_iptreemap.h> -+ -+#define IPTREEMAP_DEFAULT_GC_TIME (5 * 60) -+#define IPTREEMAP_DESTROY_SLEEP (100) -+ -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,21) -+static struct kmem_cache *cachep_b; -+static struct kmem_cache *cachep_c; -+static struct kmem_cache *cachep_d; -+#else -+static kmem_cache_t *cachep_b; -+static kmem_cache_t *cachep_c; -+static kmem_cache_t *cachep_d; -+#endif -+ -+static struct ip_set_iptreemap_d *fullbitmap_d; -+static struct ip_set_iptreemap_c *fullbitmap_c; -+static struct ip_set_iptreemap_b *fullbitmap_b; -+ -+#if defined(__LITTLE_ENDIAN) -+#define ABCD(a, b, c, d, addr) \ -+ do { \ -+ a = ((unsigned char *)addr)[3]; \ -+ b = ((unsigned char *)addr)[2]; \ -+ c = ((unsigned char *)addr)[1]; \ -+ d = ((unsigned char *)addr)[0]; \ -+ } while (0) -+#elif defined(__BIG_ENDIAN) -+#define ABCD(a,b,c,d,addrp) do { \ -+ a = ((unsigned char *)addrp)[0]; \ -+ b = ((unsigned char *)addrp)[1]; \ -+ c = ((unsigned char *)addrp)[2]; \ -+ d = ((unsigned char *)addrp)[3]; \ -+} while (0) -+#else -+#error "Please fix asm/byteorder.h" -+#endif /* __LITTLE_ENDIAN */ -+ -+#define TESTIP_WALK(map, elem, branch, full) \ -+ do { \ -+ branch = (map)->tree[elem]; \ -+ if (!branch) \ -+ return 0; \ -+ else if (branch == full) \ -+ return 1; \ -+ } while (0) -+ -+#define ADDIP_WALK(map, elem, branch, type, cachep, full) \ -+ do { \ -+ branch = (map)->tree[elem]; \ -+ if (!branch) { \ -+ branch = (type *) kmem_cache_alloc(cachep, GFP_ATOMIC); \ -+ if (!branch) \ -+ return -ENOMEM; \ -+ memset(branch, 0, sizeof(*branch)); \ -+ (map)->tree[elem] = branch; \ -+ } else if (branch == full) { \ -+ return -EEXIST; \ -+ } \ -+ } while (0) -+ -+#define ADDIP_RANGE_LOOP(map, a, a1, a2, hint, branch, full, cachep, free) \ -+ for (a = a1; a <= a2; a++) { \ -+ branch = (map)->tree[a]; \ -+ if (branch != full) { \ -+ if ((a > a1 && a < a2) || (hint)) { \ -+ if (branch) \ -+ free(branch); \ -+ (map)->tree[a] = full; \ -+ continue; \ -+ } else if (!branch) { \ -+ branch = kmem_cache_alloc(cachep, GFP_ATOMIC); \ -+ if (!branch) \ -+ return -ENOMEM; \ -+ memset(branch, 0, sizeof(*branch)); \ -+ (map)->tree[a] = branch; \ -+ } -+ -+#define ADDIP_RANGE_LOOP_END() \ -+ } \ -+ } -+ -+#define DELIP_WALK(map, elem, branch, cachep, full, flags) \ -+ do { \ -+ branch = (map)->tree[elem]; \ -+ if (!branch) { \ -+ return -EEXIST; \ -+ } else if (branch == full) { \ -+ branch = kmem_cache_alloc(cachep, flags); \ -+ if (!branch) \ -+ return -ENOMEM; \ -+ memcpy(branch, full, sizeof(*full)); \ -+ (map)->tree[elem] = branch; \ -+ } \ -+ } while (0) -+ -+#define DELIP_RANGE_LOOP(map, a, a1, a2, hint, branch, full, cachep, free, flags) \ -+ for (a = a1; a <= a2; a++) { \ -+ branch = (map)->tree[a]; \ -+ if (branch) { \ -+ if ((a > a1 && a < a2) || (hint)) { \ -+ if (branch != full) \ -+ free(branch); \ -+ (map)->tree[a] = NULL; \ -+ continue; \ -+ } else if (branch == full) { \ -+ branch = kmem_cache_alloc(cachep, flags); \ -+ if (!branch) \ -+ return -ENOMEM; \ -+ memcpy(branch, full, sizeof(*branch)); \ -+ (map)->tree[a] = branch; \ -+ } -+ -+#define DELIP_RANGE_LOOP_END() \ -+ } \ -+ } -+ -+#define LOOP_WALK_BEGIN(map, i, branch) \ -+ for (i = 0; i < 256; i++) { \ -+ branch = (map)->tree[i]; \ -+ if (likely(!branch)) \ -+ continue; -+ -+#define LOOP_WALK_END() \ -+ } -+ -+#define LOOP_WALK_BEGIN_GC(map, i, branch, full, cachep, count) \ -+ count = -256; \ -+ for (i = 0; i < 256; i++) { \ -+ branch = (map)->tree[i]; \ -+ if (likely(!branch)) \ -+ continue; \ -+ count++; \ -+ if (branch == full) { \ -+ count++; \ -+ continue; \ -+ } -+ -+#define LOOP_WALK_END_GC(map, i, branch, full, cachep, count) \ -+ if (-256 == count) { \ -+ kmem_cache_free(cachep, branch); \ -+ (map)->tree[i] = NULL; \ -+ } else if (256 == count) { \ -+ kmem_cache_free(cachep, branch); \ -+ (map)->tree[i] = full; \ -+ } \ -+ } -+ -+#define LOOP_WALK_BEGIN_COUNT(map, i, branch, inrange, count) \ -+ for (i = 0; i < 256; i++) { \ -+ if (!(map)->tree[i]) { \ -+ if (inrange) { \ -+ count++; \ -+ inrange = 0; \ -+ } \ -+ continue; \ -+ } \ -+ branch = (map)->tree[i]; -+ -+#define LOOP_WALK_END_COUNT() \ -+ } -+ -+#define MIN(a, b) (a < b ? a : b) -+#define MAX(a, b) (a > b ? a : b) -+ -+#define GETVALUE1(a, a1, b1, r) \ -+ (a == a1 ? b1 : r) -+ -+#define GETVALUE2(a, b, a1, b1, c1, r) \ -+ (a == a1 && b == b1 ? c1 : r) -+ -+#define GETVALUE3(a, b, c, a1, b1, c1, d1, r) \ -+ (a == a1 && b == b1 && c == c1 ? d1 : r) -+ -+#define CHECK1(a, a1, a2, b1, b2, c1, c2, d1, d2) \ -+ ( \ -+ GETVALUE1(a, a1, b1, 0) == 0 \ -+ && GETVALUE1(a, a2, b2, 255) == 255 \ -+ && c1 == 0 \ -+ && c2 == 255 \ -+ && d1 == 0 \ -+ && d2 == 255 \ -+ ) -+ -+#define CHECK2(a, b, a1, a2, b1, b2, c1, c2, d1, d2) \ -+ ( \ -+ GETVALUE2(a, b, a1, b1, c1, 0) == 0 \ -+ && GETVALUE2(a, b, a2, b2, c2, 255) == 255 \ -+ && d1 == 0 \ -+ && d2 == 255 \ -+ ) -+ -+#define CHECK3(a, b, c, a1, a2, b1, b2, c1, c2, d1, d2) \ -+ ( \ -+ GETVALUE3(a, b, c, a1, b1, c1, d1, 0) == 0 \ -+ && GETVALUE3(a, b, c, a2, b2, c2, d2, 255) == 255 \ -+ ) -+ -+ -+static inline void -+free_d(struct ip_set_iptreemap_d *map) -+{ -+ kmem_cache_free(cachep_d, map); -+} -+ -+static inline void -+free_c(struct ip_set_iptreemap_c *map) -+{ -+ struct ip_set_iptreemap_d *dtree; -+ unsigned int i; -+ -+ LOOP_WALK_BEGIN(map, i, dtree) { -+ if (dtree != fullbitmap_d) -+ free_d(dtree); -+ } LOOP_WALK_END(); -+ -+ kmem_cache_free(cachep_c, map); -+} -+ -+static inline void -+free_b(struct ip_set_iptreemap_b *map) -+{ -+ struct ip_set_iptreemap_c *ctree; -+ unsigned int i; -+ -+ LOOP_WALK_BEGIN(map, i, ctree) { -+ if (ctree != fullbitmap_c) -+ free_c(ctree); -+ } LOOP_WALK_END(); -+ -+ kmem_cache_free(cachep_b, map); -+} -+ -+static inline int -+__testip(struct ip_set *set, ip_set_ip_t ip, ip_set_ip_t *hash_ip) -+{ -+ struct ip_set_iptreemap *map = (struct ip_set_iptreemap *) set->data; -+ struct ip_set_iptreemap_b *btree; -+ struct ip_set_iptreemap_c *ctree; -+ struct ip_set_iptreemap_d *dtree; -+ unsigned char a, b, c, d; -+ -+ *hash_ip = ip; -+ -+ ABCD(a, b, c, d, hash_ip); -+ -+ TESTIP_WALK(map, a, btree, fullbitmap_b); -+ TESTIP_WALK(btree, b, ctree, fullbitmap_c); -+ TESTIP_WALK(ctree, c, dtree, fullbitmap_d); -+ -+ return !!test_bit(d, (void *) dtree->bitmap); -+} -+ -+static int -+testip(struct ip_set *set, const void *data, size_t size, ip_set_ip_t *hash_ip) -+{ -+ struct ip_set_req_iptreemap *req = (struct ip_set_req_iptreemap *) data; -+ -+ if (size != sizeof(struct ip_set_req_iptreemap)) { -+ ip_set_printk("data length wrong (want %zu, have %zu)", sizeof(struct ip_set_req_iptreemap), size); -+ return -EINVAL; -+ } -+ -+ return __testip(set, req->start, hash_ip); -+} -+ -+static int -+testip_kernel(struct ip_set *set, const struct sk_buff *skb, ip_set_ip_t *hash_ip, const u_int32_t *flags, unsigned char index) -+{ -+ int res; -+ -+ res = __testip(set, -+ ntohl(flags[index] & IPSET_SRC -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,22) -+ ? ip_hdr(skb)->saddr -+ : ip_hdr(skb)->daddr), -+#else -+ ? skb->nh.iph->saddr -+ : skb->nh.iph->daddr), -+#endif -+ hash_ip); -+ -+ return (res < 0 ? 0 : res); -+} -+ -+static inline int -+__addip_single(struct ip_set *set, ip_set_ip_t ip, ip_set_ip_t *hash_ip) -+{ -+ struct ip_set_iptreemap *map = (struct ip_set_iptreemap *) set->data; -+ struct ip_set_iptreemap_b *btree; -+ struct ip_set_iptreemap_c *ctree; -+ struct ip_set_iptreemap_d *dtree; -+ unsigned char a, b, c, d; -+ -+ *hash_ip = ip; -+ -+ ABCD(a, b, c, d, hash_ip); -+ -+ ADDIP_WALK(map, a, btree, struct ip_set_iptreemap_b, cachep_b, fullbitmap_b); -+ ADDIP_WALK(btree, b, ctree, struct ip_set_iptreemap_c, cachep_c, fullbitmap_c); -+ ADDIP_WALK(ctree, c, dtree, struct ip_set_iptreemap_d, cachep_d, fullbitmap_d); -+ -+ if (test_and_set_bit(d, (void *) dtree->bitmap)) -+ return -EEXIST; -+ -+ set_bit(b, (void *) btree->dirty); -+ -+ return 0; -+} -+ -+static inline int -+__addip_range(struct ip_set *set, ip_set_ip_t start, ip_set_ip_t end, ip_set_ip_t *hash_ip) -+{ -+ struct ip_set_iptreemap *map = (struct ip_set_iptreemap *) set->data; -+ struct ip_set_iptreemap_b *btree; -+ struct ip_set_iptreemap_c *ctree; -+ struct ip_set_iptreemap_d *dtree; -+ unsigned int a, b, c, d; -+ unsigned char a1, b1, c1, d1; -+ unsigned char a2, b2, c2, d2; -+ -+ if (start == end) -+ return __addip_single(set, start, hash_ip); -+ -+ *hash_ip = start; -+ -+ ABCD(a1, b1, c1, d1, &start); -+ ABCD(a2, b2, c2, d2, &end); -+ -+ /* This is sooo ugly... */ -+ ADDIP_RANGE_LOOP(map, a, a1, a2, CHECK1(a, a1, a2, b1, b2, c1, c2, d1, d2), btree, fullbitmap_b, cachep_b, free_b) { -+ ADDIP_RANGE_LOOP(btree, b, GETVALUE1(a, a1, b1, 0), GETVALUE1(a, a2, b2, 255), CHECK2(a, b, a1, a2, b1, b2, c1, c2, d1, d2), ctree, fullbitmap_c, cachep_c, free_c) { -+ ADDIP_RANGE_LOOP(ctree, c, GETVALUE2(a, b, a1, b1, c1, 0), GETVALUE2(a, b, a2, b2, c2, 255), CHECK3(a, b, c, a1, a2, b1, b2, c1, c2, d1, d2), dtree, fullbitmap_d, cachep_d, free_d) { -+ for (d = GETVALUE3(a, b, c, a1, b1, c1, d1, 0); d <= GETVALUE3(a, b, c, a2, b2, c2, d2, 255); d++) -+ set_bit(d, (void *) dtree->bitmap); -+ set_bit(b, (void *) btree->dirty); -+ } ADDIP_RANGE_LOOP_END(); -+ } ADDIP_RANGE_LOOP_END(); -+ } ADDIP_RANGE_LOOP_END(); -+ -+ return 0; -+} -+ -+static int -+addip(struct ip_set *set, const void *data, size_t size, ip_set_ip_t *hash_ip) -+{ -+ struct ip_set_req_iptreemap *req = (struct ip_set_req_iptreemap *) data; -+ -+ if (size != sizeof(struct ip_set_req_iptreemap)) { -+ ip_set_printk("data length wrong (want %zu, have %zu)", sizeof(struct ip_set_req_iptreemap), size); -+ return -EINVAL; -+ } -+ -+ return __addip_range(set, MIN(req->start, req->end), MAX(req->start, req->end), hash_ip); -+} -+ -+static int -+addip_kernel(struct ip_set *set, const struct sk_buff *skb, ip_set_ip_t *hash_ip, const u_int32_t *flags, unsigned char index) -+{ -+ -+ return __addip_single(set, -+ ntohl(flags[index] & IPSET_SRC -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,22) -+ ? ip_hdr(skb)->saddr -+ : ip_hdr(skb)->daddr), -+#else -+ ? skb->nh.iph->saddr -+ : skb->nh.iph->daddr), -+#endif -+ hash_ip); -+} -+ -+static inline int -+__delip_single(struct ip_set *set, ip_set_ip_t ip, ip_set_ip_t *hash_ip, unsigned int __nocast flags) -+{ -+ struct ip_set_iptreemap *map = (struct ip_set_iptreemap *) set->data; -+ struct ip_set_iptreemap_b *btree; -+ struct ip_set_iptreemap_c *ctree; -+ struct ip_set_iptreemap_d *dtree; -+ unsigned char a,b,c,d; -+ -+ *hash_ip = ip; -+ -+ ABCD(a, b, c, d, hash_ip); -+ -+ DELIP_WALK(map, a, btree, cachep_b, fullbitmap_b, flags); -+ DELIP_WALK(btree, b, ctree, cachep_c, fullbitmap_c, flags); -+ DELIP_WALK(ctree, c, dtree, cachep_d, fullbitmap_d, flags); -+ -+ if (!test_and_clear_bit(d, (void *) dtree->bitmap)) -+ return -EEXIST; -+ -+ set_bit(b, (void *) btree->dirty); -+ -+ return 0; -+} -+ -+static inline int -+__delip_range(struct ip_set *set, ip_set_ip_t start, ip_set_ip_t end, ip_set_ip_t *hash_ip, unsigned int __nocast flags) -+{ -+ struct ip_set_iptreemap *map = (struct ip_set_iptreemap *) set->data; -+ struct ip_set_iptreemap_b *btree; -+ struct ip_set_iptreemap_c *ctree; -+ struct ip_set_iptreemap_d *dtree; -+ unsigned int a, b, c, d; -+ unsigned char a1, b1, c1, d1; -+ unsigned char a2, b2, c2, d2; -+ -+ if (start == end) -+ return __delip_single(set, start, hash_ip, flags); -+ -+ *hash_ip = start; -+ -+ ABCD(a1, b1, c1, d1, &start); -+ ABCD(a2, b2, c2, d2, &end); -+ -+ /* This is sooo ugly... */ -+ DELIP_RANGE_LOOP(map, a, a1, a2, CHECK1(a, a1, a2, b1, b2, c1, c2, d1, d2), btree, fullbitmap_b, cachep_b, free_b, flags) { -+ DELIP_RANGE_LOOP(btree, b, GETVALUE1(a, a1, b1, 0), GETVALUE1(a, a2, b2, 255), CHECK2(a, b, a1, a2, b1, b2, c1, c2, d1, d2), ctree, fullbitmap_c, cachep_c, free_c, flags) { -+ DELIP_RANGE_LOOP(ctree, c, GETVALUE2(a, b, a1, b1, c1, 0), GETVALUE2(a, b, a2, b2, c2, 255), CHECK3(a, b, c, a1, a2, b1, b2, c1, c2, d1, d2), dtree, fullbitmap_d, cachep_d, free_d, flags) { -+ for (d = GETVALUE3(a, b, c, a1, b1, c1, d1, 0); d <= GETVALUE3(a, b, c, a2, b2, c2, d2, 255); d++) -+ clear_bit(d, (void *) dtree->bitmap); -+ set_bit(b, (void *) btree->dirty); -+ } DELIP_RANGE_LOOP_END(); -+ } DELIP_RANGE_LOOP_END(); -+ } DELIP_RANGE_LOOP_END(); -+ -+ return 0; -+} -+ -+static int -+delip(struct ip_set *set, const void *data, size_t size, ip_set_ip_t *hash_ip) -+{ -+ struct ip_set_req_iptreemap *req = (struct ip_set_req_iptreemap *) data; -+ -+ if (size != sizeof(struct ip_set_req_iptreemap)) { -+ ip_set_printk("data length wrong (want %zu, have %zu)", sizeof(struct ip_set_req_iptreemap), size); -+ return -EINVAL; -+ } -+ -+ return __delip_range(set, MIN(req->start, req->end), MAX(req->start, req->end), hash_ip, GFP_KERNEL); -+} -+ -+static int -+delip_kernel(struct ip_set *set, const struct sk_buff *skb, ip_set_ip_t *hash_ip, const u_int32_t *flags, unsigned char index) -+{ -+ return __delip_single(set, -+ ntohl(flags[index] & IPSET_SRC -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,22) -+ ? ip_hdr(skb)->saddr -+ : ip_hdr(skb)->daddr), -+#else -+ ? skb->nh.iph->saddr -+ : skb->nh.iph->daddr), -+#endif -+ hash_ip, -+ GFP_ATOMIC); -+} -+ -+/* Check the status of the bitmap -+ * -1 == all bits cleared -+ * 1 == all bits set -+ * 0 == anything else -+ */ -+static inline int -+bitmap_status(struct ip_set_iptreemap_d *dtree) -+{ -+ unsigned char first = dtree->bitmap[0]; -+ int a; -+ -+ for (a = 1; a < 32; a++) -+ if (dtree->bitmap[a] != first) -+ return 0; -+ -+ return (first == 0 ? -1 : (first == 255 ? 1 : 0)); -+} -+ -+static void -+gc(unsigned long addr) -+{ -+ struct ip_set *set = (struct ip_set *) addr; -+ struct ip_set_iptreemap *map = (struct ip_set_iptreemap *) set->data; -+ struct ip_set_iptreemap_b *btree; -+ struct ip_set_iptreemap_c *ctree; -+ struct ip_set_iptreemap_d *dtree; -+ unsigned int a, b, c; -+ int i, j, k; -+ -+ write_lock_bh(&set->lock); -+ -+ LOOP_WALK_BEGIN_GC(map, a, btree, fullbitmap_b, cachep_b, i) { -+ LOOP_WALK_BEGIN_GC(btree, b, ctree, fullbitmap_c, cachep_c, j) { -+ if (!test_and_clear_bit(b, (void *) btree->dirty)) -+ continue; -+ LOOP_WALK_BEGIN_GC(ctree, c, dtree, fullbitmap_d, cachep_d, k) { -+ switch (bitmap_status(dtree)) { -+ case -1: -+ kmem_cache_free(cachep_d, dtree); -+ ctree->tree[c] = NULL; -+ k--; -+ break; -+ case 1: -+ kmem_cache_free(cachep_d, dtree); -+ ctree->tree[c] = fullbitmap_d; -+ k++; -+ break; -+ } -+ } LOOP_WALK_END(); -+ } LOOP_WALK_END_GC(btree, b, ctree, fullbitmap_c, cachep_c, k); -+ } LOOP_WALK_END_GC(map, a, btree, fullbitmap_b, cachep_b, j); -+ -+ write_unlock_bh(&set->lock); -+ -+ map->gc.expires = jiffies + map->gc_interval * HZ; -+ add_timer(&map->gc); -+} -+ -+static inline void -+init_gc_timer(struct ip_set *set) -+{ -+ struct ip_set_iptreemap *map = (struct ip_set_iptreemap *) set->data; -+ -+ init_timer(&map->gc); -+ map->gc.data = (unsigned long) set; -+ map->gc.function = gc; -+ map->gc.expires = jiffies + map->gc_interval * HZ; -+ add_timer(&map->gc); -+} -+ -+static int create(struct ip_set *set, const void *data, size_t size) -+{ -+ struct ip_set_req_iptreemap_create *req = (struct ip_set_req_iptreemap_create *) data; -+ struct ip_set_iptreemap *map; -+ -+ if (size != sizeof(struct ip_set_req_iptreemap_create)) { -+ ip_set_printk("data length wrong (want %zu, have %zu)", sizeof(struct ip_set_req_iptreemap_create), size); -+ return -EINVAL; -+ } -+ -+ map = kzalloc(sizeof(*map), GFP_KERNEL); -+ if (!map) -+ return -ENOMEM; -+ -+ map->gc_interval = req->gc_interval ? req->gc_interval : IPTREEMAP_DEFAULT_GC_TIME; -+ set->data = map; -+ -+ init_gc_timer(set); -+ -+ return 0; -+} -+ -+static inline void __flush(struct ip_set_iptreemap *map) -+{ -+ struct ip_set_iptreemap_b *btree; -+ unsigned int a; -+ -+ LOOP_WALK_BEGIN(map, a, btree); -+ if (btree != fullbitmap_b) -+ free_b(btree); -+ LOOP_WALK_END(); -+} -+ -+static void destroy(struct ip_set *set) -+{ -+ struct ip_set_iptreemap *map = (struct ip_set_iptreemap *) set->data; -+ -+ while (!del_timer(&map->gc)) -+ msleep(IPTREEMAP_DESTROY_SLEEP); -+ -+ __flush(map); -+ kfree(map); -+ -+ set->data = NULL; -+} -+ -+static void flush(struct ip_set *set) -+{ -+ struct ip_set_iptreemap *map = (struct ip_set_iptreemap *) set->data; -+ -+ while (!del_timer(&map->gc)) -+ msleep(IPTREEMAP_DESTROY_SLEEP); -+ -+ __flush(map); -+ -+ memset(map, 0, sizeof(*map)); -+ -+ init_gc_timer(set); -+} -+ -+static void list_header(const struct ip_set *set, void *data) -+{ -+ struct ip_set_iptreemap *map = (struct ip_set_iptreemap *) set->data; -+ struct ip_set_req_iptreemap_create *header = (struct ip_set_req_iptreemap_create *) data; -+ -+ header->gc_interval = map->gc_interval; -+} -+ -+static int list_members_size(const struct ip_set *set) -+{ -+ struct ip_set_iptreemap *map = (struct ip_set_iptreemap *) set->data; -+ struct ip_set_iptreemap_b *btree; -+ struct ip_set_iptreemap_c *ctree; -+ struct ip_set_iptreemap_d *dtree; -+ unsigned int a, b, c, d, inrange = 0, count = 0; -+ -+ LOOP_WALK_BEGIN_COUNT(map, a, btree, inrange, count) { -+ LOOP_WALK_BEGIN_COUNT(btree, b, ctree, inrange, count) { -+ LOOP_WALK_BEGIN_COUNT(ctree, c, dtree, inrange, count) { -+ for (d = 0; d < 256; d++) { -+ if (test_bit(d, (void *) dtree->bitmap)) { -+ inrange = 1; -+ } else if (inrange) { -+ count++; -+ inrange = 0; -+ } -+ } -+ } LOOP_WALK_END_COUNT(); -+ } LOOP_WALK_END_COUNT(); -+ } LOOP_WALK_END_COUNT(); -+ -+ if (inrange) -+ count++; -+ -+ return (count * sizeof(struct ip_set_req_iptreemap)); -+} -+ -+static inline size_t add_member(void *data, size_t offset, ip_set_ip_t start, ip_set_ip_t end) -+{ -+ struct ip_set_req_iptreemap *entry = (struct ip_set_req_iptreemap *) (data + offset); -+ -+ entry->start = start; -+ entry->end = end; -+ -+ return sizeof(*entry); -+} -+ -+static void list_members(const struct ip_set *set, void *data) -+{ -+ struct ip_set_iptreemap *map = (struct ip_set_iptreemap *) set->data; -+ struct ip_set_iptreemap_b *btree; -+ struct ip_set_iptreemap_c *ctree; -+ struct ip_set_iptreemap_d *dtree; -+ unsigned int a, b, c, d, inrange = 0; -+ size_t offset = 0; -+ ip_set_ip_t start = 0, end = 0, ip; -+ -+ LOOP_WALK_BEGIN(map, a, btree) { -+ LOOP_WALK_BEGIN(btree, b, ctree) { -+ LOOP_WALK_BEGIN(ctree, c, dtree) { -+ for (d = 0; d < 256; d++) { -+ if (test_bit(d, (void *) dtree->bitmap)) { -+ ip = ((a << 24) | (b << 16) | (c << 8) | d); -+ if (!inrange) { -+ inrange = 1; -+ start = ip; -+ } else if (end < ip - 1) { -+ offset += add_member(data, offset, start, end); -+ start = ip; -+ } -+ end = ip; -+ } else if (inrange) { -+ offset += add_member(data, offset, start, end); -+ inrange = 0; -+ } -+ } -+ } LOOP_WALK_END(); -+ } LOOP_WALK_END(); -+ } LOOP_WALK_END(); -+ -+ if (inrange) -+ add_member(data, offset, start, end); -+} -+ -+static struct ip_set_type ip_set_iptreemap = { -+ .typename = SETTYPE_NAME, -+ .features = IPSET_TYPE_IP | IPSET_DATA_SINGLE, -+ .protocol_version = IP_SET_PROTOCOL_VERSION, -+ .create = create, -+ .destroy = destroy, -+ .flush = flush, -+ .reqsize = sizeof(struct ip_set_req_iptreemap), -+ .addip = addip, -+ .addip_kernel = addip_kernel, -+ .delip = delip, -+ .delip_kernel = delip_kernel, -+ .testip = testip, -+ .testip_kernel = testip_kernel, -+ .header_size = sizeof(struct ip_set_req_iptreemap_create), -+ .list_header = list_header, -+ .list_members_size = list_members_size, -+ .list_members = list_members, -+ .me = THIS_MODULE, -+}; -+ -+MODULE_LICENSE("GPL"); -+MODULE_AUTHOR("Sven Wegener <sven.wegener@stealer.net>"); -+MODULE_DESCRIPTION("iptreemap type of IP sets"); -+ -+static int __init ip_set_iptreemap_init(void) -+{ -+ int ret = -ENOMEM; -+ int a; -+ -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,23) -+ cachep_b = kmem_cache_create("ip_set_iptreemap_b", -+ sizeof(struct ip_set_iptreemap_b), -+ 0, 0, NULL); -+#else -+ cachep_b = kmem_cache_create("ip_set_iptreemap_b", -+ sizeof(struct ip_set_iptreemap_b), -+ 0, 0, NULL, NULL); -+#endif -+ if (!cachep_b) { -+ ip_set_printk("Unable to create ip_set_iptreemap_b slab cache"); -+ goto out; -+ } -+ -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,23) -+ cachep_c = kmem_cache_create("ip_set_iptreemap_c", -+ sizeof(struct ip_set_iptreemap_c), -+ 0, 0, NULL); -+#else -+ cachep_c = kmem_cache_create("ip_set_iptreemap_c", -+ sizeof(struct ip_set_iptreemap_c), -+ 0, 0, NULL, NULL); -+#endif -+ if (!cachep_c) { -+ ip_set_printk("Unable to create ip_set_iptreemap_c slab cache"); -+ goto outb; -+ } -+ -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,23) -+ cachep_d = kmem_cache_create("ip_set_iptreemap_d", -+ sizeof(struct ip_set_iptreemap_d), -+ 0, 0, NULL); -+#else -+ cachep_d = kmem_cache_create("ip_set_iptreemap_d", -+ sizeof(struct ip_set_iptreemap_d), -+ 0, 0, NULL, NULL); -+#endif -+ if (!cachep_d) { -+ ip_set_printk("Unable to create ip_set_iptreemap_d slab cache"); -+ goto outc; -+ } -+ -+ fullbitmap_d = kmem_cache_alloc(cachep_d, GFP_KERNEL); -+ if (!fullbitmap_d) -+ goto outd; -+ -+ fullbitmap_c = kmem_cache_alloc(cachep_c, GFP_KERNEL); -+ if (!fullbitmap_c) -+ goto outbitmapd; -+ -+ fullbitmap_b = kmem_cache_alloc(cachep_b, GFP_KERNEL); -+ if (!fullbitmap_b) -+ goto outbitmapc; -+ -+ ret = ip_set_register_set_type(&ip_set_iptreemap); -+ if (0 > ret) -+ goto outbitmapb; -+ -+ /* Now init our global bitmaps */ -+ memset(fullbitmap_d->bitmap, 0xff, sizeof(fullbitmap_d->bitmap)); -+ -+ for (a = 0; a < 256; a++) -+ fullbitmap_c->tree[a] = fullbitmap_d; -+ -+ for (a = 0; a < 256; a++) -+ fullbitmap_b->tree[a] = fullbitmap_c; -+ memset(fullbitmap_b->dirty, 0, sizeof(fullbitmap_b->dirty)); -+ -+ return 0; -+ -+outbitmapb: -+ kmem_cache_free(cachep_b, fullbitmap_b); -+outbitmapc: -+ kmem_cache_free(cachep_c, fullbitmap_c); -+outbitmapd: -+ kmem_cache_free(cachep_d, fullbitmap_d); -+outd: -+ kmem_cache_destroy(cachep_d); -+outc: -+ kmem_cache_destroy(cachep_c); -+outb: -+ kmem_cache_destroy(cachep_b); -+out: -+ -+ return ret; -+} -+ -+static void __exit ip_set_iptreemap_fini(void) -+{ -+ ip_set_unregister_set_type(&ip_set_iptreemap); -+ kmem_cache_free(cachep_d, fullbitmap_d); -+ kmem_cache_free(cachep_c, fullbitmap_c); -+ kmem_cache_free(cachep_b, fullbitmap_b); -+ kmem_cache_destroy(cachep_d); -+ kmem_cache_destroy(cachep_c); -+ kmem_cache_destroy(cachep_b); -+} -+ -+module_init(ip_set_iptreemap_init); -+module_exit(ip_set_iptreemap_fini); ---- /dev/null -+++ b/net/ipv4/netfilter/ip_set_macipmap.c -@@ -0,0 +1,375 @@ -+/* Copyright (C) 2000-2002 Joakim Axelsson <gozem@linux.nu> -+ * Patrick Schaaf <bof@bof.de> -+ * Martin Josefsson <gandalf@wlug.westbo.se> -+ * Copyright (C) 2003-2004 Jozsef Kadlecsik <kadlec@blackhole.kfki.hu> -+ * -+ * This program is free software; you can redistribute it and/or modify -+ * it under the terms of the GNU General Public License version 2 as -+ * published by the Free Software Foundation. -+ */ -+ -+/* Kernel module implementing an IP set type: the macipmap type */ -+ -+#include <linux/module.h> -+#include <linux/ip.h> -+#include <linux/skbuff.h> -+#include <linux/version.h> -+#include <linux/netfilter_ipv4/ip_tables.h> -+#include <linux/netfilter_ipv4/ip_set.h> -+#include <linux/errno.h> -+#include <asm/uaccess.h> -+#include <asm/bitops.h> -+#include <linux/spinlock.h> -+#include <linux/if_ether.h> -+#include <linux/vmalloc.h> -+ -+#include <linux/netfilter_ipv4/ip_set_malloc.h> -+#include <linux/netfilter_ipv4/ip_set_macipmap.h> -+ -+static int -+testip(struct ip_set *set, const void *data, size_t size, ip_set_ip_t *hash_ip) -+{ -+ struct ip_set_macipmap *map = (struct ip_set_macipmap *) set->data; -+ struct ip_set_macip *table = (struct ip_set_macip *) map->members; -+ struct ip_set_req_macipmap *req = (struct ip_set_req_macipmap *) data; -+ -+ if (size != sizeof(struct ip_set_req_macipmap)) { -+ ip_set_printk("data length wrong (want %zu, have %zu)", -+ sizeof(struct ip_set_req_macipmap), -+ size); -+ return -EINVAL; -+ } -+ -+ if (req->ip < map->first_ip || req->ip > map->last_ip) -+ return -ERANGE; -+ -+ *hash_ip = req->ip; -+ DP("set: %s, ip:%u.%u.%u.%u, %u.%u.%u.%u", -+ set->name, HIPQUAD(req->ip), HIPQUAD(*hash_ip)); -+ if (test_bit(IPSET_MACIP_ISSET, -+ (void *) &table[req->ip - map->first_ip].flags)) { -+ return (memcmp(req->ethernet, -+ &table[req->ip - map->first_ip].ethernet, -+ ETH_ALEN) == 0); -+ } else { -+ return (map->flags & IPSET_MACIP_MATCHUNSET ? 1 : 0); -+ } -+} -+ -+static int -+testip_kernel(struct ip_set *set, -+ const struct sk_buff *skb, -+ ip_set_ip_t *hash_ip, -+ const u_int32_t *flags, -+ unsigned char index) -+{ -+ struct ip_set_macipmap *map = -+ (struct ip_set_macipmap *) set->data; -+ struct ip_set_macip *table = -+ (struct ip_set_macip *) map->members; -+ ip_set_ip_t ip; -+ -+ ip = ntohl(flags[index] & IPSET_SRC -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,22) -+ ? ip_hdr(skb)->saddr -+ : ip_hdr(skb)->daddr); -+#else -+ ? skb->nh.iph->saddr -+ : skb->nh.iph->daddr); -+#endif -+ -+ if (ip < map->first_ip || ip > map->last_ip) -+ return 0; -+ -+ *hash_ip = ip; -+ DP("set: %s, ip:%u.%u.%u.%u, %u.%u.%u.%u", -+ set->name, HIPQUAD(ip), HIPQUAD(*hash_ip)); -+ if (test_bit(IPSET_MACIP_ISSET, -+ (void *) &table[ip - map->first_ip].flags)) { -+ /* Is mac pointer valid? -+ * If so, compare... */ -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,22) -+ return (skb_mac_header(skb) >= skb->head -+ && (skb_mac_header(skb) + ETH_HLEN) <= skb->data -+#else -+ return (skb->mac.raw >= skb->head -+ && (skb->mac.raw + ETH_HLEN) <= skb->data -+#endif -+ && (memcmp(eth_hdr(skb)->h_source, -+ &table[ip - map->first_ip].ethernet, -+ ETH_ALEN) == 0)); -+ } else { -+ return (map->flags & IPSET_MACIP_MATCHUNSET ? 1 : 0); -+ } -+} -+ -+/* returns 0 on success */ -+static inline int -+__addip(struct ip_set *set, -+ ip_set_ip_t ip, unsigned char *ethernet, ip_set_ip_t *hash_ip) -+{ -+ struct ip_set_macipmap *map = -+ (struct ip_set_macipmap *) set->data; -+ struct ip_set_macip *table = -+ (struct ip_set_macip *) map->members; -+ -+ if (ip < map->first_ip || ip > map->last_ip) -+ return -ERANGE; -+ if (test_and_set_bit(IPSET_MACIP_ISSET, -+ (void *) &table[ip - map->first_ip].flags)) -+ return -EEXIST; -+ -+ *hash_ip = ip; -+ DP("%u.%u.%u.%u, %u.%u.%u.%u", HIPQUAD(ip), HIPQUAD(*hash_ip)); -+ memcpy(&table[ip - map->first_ip].ethernet, ethernet, ETH_ALEN); -+ return 0; -+} -+ -+static int -+addip(struct ip_set *set, const void *data, size_t size, -+ ip_set_ip_t *hash_ip) -+{ -+ struct ip_set_req_macipmap *req = -+ (struct ip_set_req_macipmap *) data; -+ -+ if (size != sizeof(struct ip_set_req_macipmap)) { -+ ip_set_printk("data length wrong (want %zu, have %zu)", -+ sizeof(struct ip_set_req_macipmap), -+ size); -+ return -EINVAL; -+ } -+ return __addip(set, req->ip, req->ethernet, hash_ip); -+} -+ -+static int -+addip_kernel(struct ip_set *set, -+ const struct sk_buff *skb, -+ ip_set_ip_t *hash_ip, -+ const u_int32_t *flags, -+ unsigned char index) -+{ -+ ip_set_ip_t ip; -+ -+ ip = ntohl(flags[index] & IPSET_SRC -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,22) -+ ? ip_hdr(skb)->saddr -+ : ip_hdr(skb)->daddr); -+#else -+ ? skb->nh.iph->saddr -+ : skb->nh.iph->daddr); -+#endif -+ -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,22) -+ if (!(skb_mac_header(skb) >= skb->head -+ && (skb_mac_header(skb) + ETH_HLEN) <= skb->data)) -+#else -+ if (!(skb->mac.raw >= skb->head -+ && (skb->mac.raw + ETH_HLEN) <= skb->data)) -+#endif -+ return -EINVAL; -+ -+ return __addip(set, ip, eth_hdr(skb)->h_source, hash_ip); -+} -+ -+static inline int -+__delip(struct ip_set *set, ip_set_ip_t ip, ip_set_ip_t *hash_ip) -+{ -+ struct ip_set_macipmap *map = -+ (struct ip_set_macipmap *) set->data; -+ struct ip_set_macip *table = -+ (struct ip_set_macip *) map->members; -+ -+ if (ip < map->first_ip || ip > map->last_ip) -+ return -ERANGE; -+ if (!test_and_clear_bit(IPSET_MACIP_ISSET, -+ (void *)&table[ip - map->first_ip].flags)) -+ return -EEXIST; -+ -+ *hash_ip = ip; -+ DP("%u.%u.%u.%u, %u.%u.%u.%u", HIPQUAD(ip), HIPQUAD(*hash_ip)); -+ return 0; -+} -+ -+static int -+delip(struct ip_set *set, const void *data, size_t size, -+ ip_set_ip_t *hash_ip) -+{ -+ struct ip_set_req_macipmap *req = -+ (struct ip_set_req_macipmap *) data; -+ -+ if (size != sizeof(struct ip_set_req_macipmap)) { -+ ip_set_printk("data length wrong (want %zu, have %zu)", -+ sizeof(struct ip_set_req_macipmap), -+ size); -+ return -EINVAL; -+ } -+ return __delip(set, req->ip, hash_ip); -+} -+ -+static int -+delip_kernel(struct ip_set *set, -+ const struct sk_buff *skb, -+ ip_set_ip_t *hash_ip, -+ const u_int32_t *flags, -+ unsigned char index) -+{ -+ return __delip(set, -+ ntohl(flags[index] & IPSET_SRC -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,22) -+ ? ip_hdr(skb)->saddr -+ : ip_hdr(skb)->daddr), -+#else -+ ? skb->nh.iph->saddr -+ : skb->nh.iph->daddr), -+#endif -+ hash_ip); -+} -+ -+static inline size_t members_size(ip_set_id_t from, ip_set_id_t to) -+{ -+ return (size_t)((to - from + 1) * sizeof(struct ip_set_macip)); -+} -+ -+static int create(struct ip_set *set, const void *data, size_t size) -+{ -+ int newbytes; -+ struct ip_set_req_macipmap_create *req = -+ (struct ip_set_req_macipmap_create *) data; -+ struct ip_set_macipmap *map; -+ -+ if (size != sizeof(struct ip_set_req_macipmap_create)) { -+ ip_set_printk("data length wrong (want %zu, have %zu)", -+ sizeof(struct ip_set_req_macipmap_create), -+ size); -+ return -EINVAL; -+ } -+ -+ DP("from %u.%u.%u.%u to %u.%u.%u.%u", -+ HIPQUAD(req->from), HIPQUAD(req->to)); -+ -+ if (req->from > req->to) { -+ DP("bad ip range"); -+ return -ENOEXEC; -+ } -+ -+ if (req->to - req->from > MAX_RANGE) { -+ ip_set_printk("range too big (max %d addresses)", -+ MAX_RANGE+1); -+ return -ENOEXEC; -+ } -+ -+ map = kmalloc(sizeof(struct ip_set_macipmap), GFP_KERNEL); -+ if (!map) { -+ DP("out of memory for %d bytes", -+ sizeof(struct ip_set_macipmap)); -+ return -ENOMEM; -+ } -+ map->flags = req->flags; -+ map->first_ip = req->from; -+ map->last_ip = req->to; -+ newbytes = members_size(map->first_ip, map->last_ip); -+ map->members = ip_set_malloc(newbytes); -+ DP("members: %u %p", newbytes, map->members); -+ if (!map->members) { -+ DP("out of memory for %d bytes", newbytes); -+ kfree(map); -+ return -ENOMEM; -+ } -+ memset(map->members, 0, newbytes); -+ -+ set->data = map; -+ return 0; -+} -+ -+static void destroy(struct ip_set *set) -+{ -+ struct ip_set_macipmap *map = -+ (struct ip_set_macipmap *) set->data; -+ -+ ip_set_free(map->members, members_size(map->first_ip, map->last_ip)); -+ kfree(map); -+ -+ set->data = NULL; -+} -+ -+static void flush(struct ip_set *set) -+{ -+ struct ip_set_macipmap *map = -+ (struct ip_set_macipmap *) set->data; -+ memset(map->members, 0, members_size(map->first_ip, map->last_ip)); -+} -+ -+static void list_header(const struct ip_set *set, void *data) -+{ -+ struct ip_set_macipmap *map = -+ (struct ip_set_macipmap *) set->data; -+ struct ip_set_req_macipmap_create *header = -+ (struct ip_set_req_macipmap_create *) data; -+ -+ DP("list_header %x %x %u", map->first_ip, map->last_ip, -+ map->flags); -+ -+ header->from = map->first_ip; -+ header->to = map->last_ip; -+ header->flags = map->flags; -+} -+ -+static int list_members_size(const struct ip_set *set) -+{ -+ struct ip_set_macipmap *map = -+ (struct ip_set_macipmap *) set->data; -+ -+ DP("%u", members_size(map->first_ip, map->last_ip)); -+ return members_size(map->first_ip, map->last_ip); -+} -+ -+static void list_members(const struct ip_set *set, void *data) -+{ -+ struct ip_set_macipmap *map = -+ (struct ip_set_macipmap *) set->data; -+ -+ int bytes = members_size(map->first_ip, map->last_ip); -+ -+ DP("members: %u %p", bytes, map->members); -+ memcpy(data, map->members, bytes); -+} -+ -+static struct ip_set_type ip_set_macipmap = { -+ .typename = SETTYPE_NAME, -+ .features = IPSET_TYPE_IP | IPSET_DATA_SINGLE, -+ .protocol_version = IP_SET_PROTOCOL_VERSION, -+ .create = &create, -+ .destroy = &destroy, -+ .flush = &flush, -+ .reqsize = sizeof(struct ip_set_req_macipmap), -+ .addip = &addip, -+ .addip_kernel = &addip_kernel, -+ .delip = &delip, -+ .delip_kernel = &delip_kernel, -+ .testip = &testip, -+ .testip_kernel = &testip_kernel, -+ .header_size = sizeof(struct ip_set_req_macipmap_create), -+ .list_header = &list_header, -+ .list_members_size = &list_members_size, -+ .list_members = &list_members, -+ .me = THIS_MODULE, -+}; -+ -+MODULE_LICENSE("GPL"); -+MODULE_AUTHOR("Jozsef Kadlecsik <kadlec@blackhole.kfki.hu>"); -+MODULE_DESCRIPTION("macipmap type of IP sets"); -+ -+static int __init ip_set_macipmap_init(void) -+{ -+ init_max_malloc_size(); -+ return ip_set_register_set_type(&ip_set_macipmap); -+} -+ -+static void __exit ip_set_macipmap_fini(void) -+{ -+ /* FIXME: possible race with ip_set_create() */ -+ ip_set_unregister_set_type(&ip_set_macipmap); -+} -+ -+module_init(ip_set_macipmap_init); -+module_exit(ip_set_macipmap_fini); ---- /dev/null -+++ b/net/ipv4/netfilter/ip_set_nethash.c -@@ -0,0 +1,497 @@ -+/* Copyright (C) 2003-2004 Jozsef Kadlecsik <kadlec@blackhole.kfki.hu> -+ * -+ * This program is free software; you can redistribute it and/or modify -+ * it under the terms of the GNU General Public License version 2 as -+ * published by the Free Software Foundation. -+ */ -+ -+/* Kernel module implementing a cidr nethash set */ -+ -+#include <linux/module.h> -+#include <linux/ip.h> -+#include <linux/skbuff.h> -+#include <linux/version.h> -+#include <linux/jhash.h> -+#include <linux/netfilter_ipv4/ip_tables.h> -+#include <linux/netfilter_ipv4/ip_set.h> -+#include <linux/errno.h> -+#include <asm/uaccess.h> -+#include <asm/bitops.h> -+#include <linux/spinlock.h> -+#include <linux/vmalloc.h> -+#include <linux/random.h> -+ -+#include <net/ip.h> -+ -+#include <linux/netfilter_ipv4/ip_set_malloc.h> -+#include <linux/netfilter_ipv4/ip_set_nethash.h> -+ -+static int limit = MAX_RANGE; -+ -+static inline __u32 -+jhash_ip(const struct ip_set_nethash *map, uint16_t i, ip_set_ip_t ip) -+{ -+ return jhash_1word(ip, *(((uint32_t *) map->initval) + i)); -+} -+ -+static inline __u32 -+hash_id_cidr(struct ip_set_nethash *map, -+ ip_set_ip_t ip, -+ unsigned char cidr, -+ ip_set_ip_t *hash_ip) -+{ -+ __u32 id; -+ u_int16_t i; -+ ip_set_ip_t *elem; -+ -+ *hash_ip = pack(ip, cidr); -+ -+ for (i = 0; i < map->probes; i++) { -+ id = jhash_ip(map, i, *hash_ip) % map->hashsize; -+ DP("hash key: %u", id); -+ elem = HARRAY_ELEM(map->members, ip_set_ip_t *, id); -+ if (*elem == *hash_ip) -+ return id; -+ } -+ return UINT_MAX; -+} -+ -+static inline __u32 -+hash_id(struct ip_set *set, ip_set_ip_t ip, ip_set_ip_t *hash_ip) -+{ -+ struct ip_set_nethash *map = (struct ip_set_nethash *) set->data; -+ __u32 id = UINT_MAX; -+ int i; -+ -+ for (i = 0; i < 30 && map->cidr[i]; i++) { -+ id = hash_id_cidr(map, ip, map->cidr[i], hash_ip); -+ if (id != UINT_MAX) -+ break; -+ } -+ return id; -+} -+ -+static inline int -+__testip_cidr(struct ip_set *set, ip_set_ip_t ip, unsigned char cidr, -+ ip_set_ip_t *hash_ip) -+{ -+ struct ip_set_nethash *map = (struct ip_set_nethash *) set->data; -+ -+ return (ip && hash_id_cidr(map, ip, cidr, hash_ip) != UINT_MAX); -+} -+ -+static inline int -+__testip(struct ip_set *set, ip_set_ip_t ip, ip_set_ip_t *hash_ip) -+{ -+ return (ip && hash_id(set, ip, hash_ip) != UINT_MAX); -+} -+ -+static int -+testip(struct ip_set *set, const void *data, size_t size, -+ ip_set_ip_t *hash_ip) -+{ -+ struct ip_set_req_nethash *req = -+ (struct ip_set_req_nethash *) data; -+ -+ if (size != sizeof(struct ip_set_req_nethash)) { -+ ip_set_printk("data length wrong (want %zu, have %zu)", -+ sizeof(struct ip_set_req_nethash), -+ size); -+ return -EINVAL; -+ } -+ return (req->cidr == 32 ? __testip(set, req->ip, hash_ip) -+ : __testip_cidr(set, req->ip, req->cidr, hash_ip)); -+} -+ -+static int -+testip_kernel(struct ip_set *set, -+ const struct sk_buff *skb, -+ ip_set_ip_t *hash_ip, -+ const u_int32_t *flags, -+ unsigned char index) -+{ -+ return __testip(set, -+ ntohl(flags[index] & IPSET_SRC -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,22) -+ ? ip_hdr(skb)->saddr -+ : ip_hdr(skb)->daddr), -+#else -+ ? skb->nh.iph->saddr -+ : skb->nh.iph->daddr), -+#endif -+ hash_ip); -+} -+ -+static inline int -+__addip_base(struct ip_set_nethash *map, ip_set_ip_t ip) -+{ -+ __u32 probe; -+ u_int16_t i; -+ ip_set_ip_t *elem; -+ -+ for (i = 0; i < map->probes; i++) { -+ probe = jhash_ip(map, i, ip) % map->hashsize; -+ elem = HARRAY_ELEM(map->members, ip_set_ip_t *, probe); -+ if (*elem == ip) -+ return -EEXIST; -+ if (!*elem) { -+ *elem = ip; -+ map->elements++; -+ return 0; -+ } -+ } -+ /* Trigger rehashing */ -+ return -EAGAIN; -+} -+ -+static inline int -+__addip(struct ip_set_nethash *map, ip_set_ip_t ip, unsigned char cidr, -+ ip_set_ip_t *hash_ip) -+{ -+ if (!ip || map->elements >= limit) -+ return -ERANGE; -+ -+ *hash_ip = pack(ip, cidr); -+ DP("%u.%u.%u.%u/%u, %u.%u.%u.%u", HIPQUAD(ip), cidr, HIPQUAD(*hash_ip)); -+ -+ return __addip_base(map, *hash_ip); -+} -+ -+static void -+update_cidr_sizes(struct ip_set_nethash *map, unsigned char cidr) -+{ -+ unsigned char next; -+ int i; -+ -+ for (i = 0; i < 30 && map->cidr[i]; i++) { -+ if (map->cidr[i] == cidr) { -+ return; -+ } else if (map->cidr[i] < cidr) { -+ next = map->cidr[i]; -+ map->cidr[i] = cidr; -+ cidr = next; -+ } -+ } -+ if (i < 30) -+ map->cidr[i] = cidr; -+} -+ -+static int -+addip(struct ip_set *set, const void *data, size_t size, -+ ip_set_ip_t *hash_ip) -+{ -+ struct ip_set_req_nethash *req = -+ (struct ip_set_req_nethash *) data; -+ int ret; -+ -+ if (size != sizeof(struct ip_set_req_nethash)) { -+ ip_set_printk("data length wrong (want %zu, have %zu)", -+ sizeof(struct ip_set_req_nethash), -+ size); -+ return -EINVAL; -+ } -+ ret = __addip((struct ip_set_nethash *) set->data, -+ req->ip, req->cidr, hash_ip); -+ -+ if (ret == 0) -+ update_cidr_sizes((struct ip_set_nethash *) set->data, -+ req->cidr); -+ -+ return ret; -+} -+ -+static int -+addip_kernel(struct ip_set *set, -+ const struct sk_buff *skb, -+ ip_set_ip_t *hash_ip, -+ const u_int32_t *flags, -+ unsigned char index) -+{ -+ struct ip_set_nethash *map = (struct ip_set_nethash *) set->data; -+ int ret = -ERANGE; -+ ip_set_ip_t ip = ntohl(flags[index] & IPSET_SRC -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,22) -+ ? ip_hdr(skb)->saddr -+ : ip_hdr(skb)->daddr); -+#else -+ ? skb->nh.iph->saddr -+ : skb->nh.iph->daddr); -+#endif -+ -+ if (map->cidr[0]) -+ ret = __addip(map, ip, map->cidr[0], hash_ip); -+ -+ return ret; -+} -+ -+static int retry(struct ip_set *set) -+{ -+ struct ip_set_nethash *map = (struct ip_set_nethash *) set->data; -+ ip_set_ip_t *elem; -+ void *members; -+ u_int32_t i, hashsize = map->hashsize; -+ int res; -+ struct ip_set_nethash *tmp; -+ -+ if (map->resize == 0) -+ return -ERANGE; -+ -+ again: -+ res = 0; -+ -+ /* Calculate new parameters */ -+ hashsize += (hashsize * map->resize)/100; -+ if (hashsize == map->hashsize) -+ hashsize++; -+ -+ ip_set_printk("rehashing of set %s triggered: " -+ "hashsize grows from %u to %u", -+ set->name, map->hashsize, hashsize); -+ -+ tmp = kmalloc(sizeof(struct ip_set_nethash) -+ + map->probes * sizeof(uint32_t), GFP_ATOMIC); -+ if (!tmp) { -+ DP("out of memory for %d bytes", -+ sizeof(struct ip_set_nethash) -+ + map->probes * sizeof(uint32_t)); -+ return -ENOMEM; -+ } -+ tmp->members = harray_malloc(hashsize, sizeof(ip_set_ip_t), GFP_ATOMIC); -+ if (!tmp->members) { -+ DP("out of memory for %d bytes", hashsize * sizeof(ip_set_ip_t)); -+ kfree(tmp); -+ return -ENOMEM; -+ } -+ tmp->hashsize = hashsize; -+ tmp->elements = 0; -+ tmp->probes = map->probes; -+ tmp->resize = map->resize; -+ memcpy(tmp->initval, map->initval, map->probes * sizeof(uint32_t)); -+ memcpy(tmp->cidr, map->cidr, 30 * sizeof(unsigned char)); -+ -+ write_lock_bh(&set->lock); -+ map = (struct ip_set_nethash *) set->data; /* Play safe */ -+ for (i = 0; i < map->hashsize && res == 0; i++) { -+ elem = HARRAY_ELEM(map->members, ip_set_ip_t *, i); -+ if (*elem) -+ res = __addip_base(tmp, *elem); -+ } -+ if (res) { -+ /* Failure, try again */ -+ write_unlock_bh(&set->lock); -+ harray_free(tmp->members); -+ kfree(tmp); -+ goto again; -+ } -+ -+ /* Success at resizing! */ -+ members = map->members; -+ -+ map->hashsize = tmp->hashsize; -+ map->members = tmp->members; -+ write_unlock_bh(&set->lock); -+ -+ harray_free(members); -+ kfree(tmp); -+ -+ return 0; -+} -+ -+static inline int -+__delip(struct ip_set_nethash *map, ip_set_ip_t ip, unsigned char cidr, -+ ip_set_ip_t *hash_ip) -+{ -+ ip_set_ip_t id, *elem; -+ -+ if (!ip) -+ return -ERANGE; -+ -+ id = hash_id_cidr(map, ip, cidr, hash_ip); -+ if (id == UINT_MAX) -+ return -EEXIST; -+ -+ elem = HARRAY_ELEM(map->members, ip_set_ip_t *, id); -+ *elem = 0; -+ map->elements--; -+ return 0; -+} -+ -+static int -+delip(struct ip_set *set, const void *data, size_t size, -+ ip_set_ip_t *hash_ip) -+{ -+ struct ip_set_req_nethash *req = -+ (struct ip_set_req_nethash *) data; -+ -+ if (size != sizeof(struct ip_set_req_nethash)) { -+ ip_set_printk("data length wrong (want %zu, have %zu)", -+ sizeof(struct ip_set_req_nethash), -+ size); -+ return -EINVAL; -+ } -+ /* TODO: no garbage collection in map->cidr */ -+ return __delip((struct ip_set_nethash *) set->data, -+ req->ip, req->cidr, hash_ip); -+} -+ -+static int -+delip_kernel(struct ip_set *set, -+ const struct sk_buff *skb, -+ ip_set_ip_t *hash_ip, -+ const u_int32_t *flags, -+ unsigned char index) -+{ -+ struct ip_set_nethash *map = (struct ip_set_nethash *) set->data; -+ int ret = -ERANGE; -+ ip_set_ip_t ip = ntohl(flags[index] & IPSET_SRC -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,22) -+ ? ip_hdr(skb)->saddr -+ : ip_hdr(skb)->daddr); -+#else -+ ? skb->nh.iph->saddr -+ : skb->nh.iph->daddr); -+#endif -+ -+ if (map->cidr[0]) -+ ret = __delip(map, ip, map->cidr[0], hash_ip); -+ -+ return ret; -+} -+ -+static int create(struct ip_set *set, const void *data, size_t size) -+{ -+ struct ip_set_req_nethash_create *req = -+ (struct ip_set_req_nethash_create *) data; -+ struct ip_set_nethash *map; -+ uint16_t i; -+ -+ if (size != sizeof(struct ip_set_req_nethash_create)) { -+ ip_set_printk("data length wrong (want %zu, have %zu)", -+ sizeof(struct ip_set_req_nethash_create), -+ size); -+ return -EINVAL; -+ } -+ -+ if (req->hashsize < 1) { -+ ip_set_printk("hashsize too small"); -+ return -ENOEXEC; -+ } -+ if (req->probes < 1) { -+ ip_set_printk("probes too small"); -+ return -ENOEXEC; -+ } -+ -+ map = kmalloc(sizeof(struct ip_set_nethash) -+ + req->probes * sizeof(uint32_t), GFP_KERNEL); -+ if (!map) { -+ DP("out of memory for %d bytes", -+ sizeof(struct ip_set_nethash) -+ + req->probes * sizeof(uint32_t)); -+ return -ENOMEM; -+ } -+ for (i = 0; i < req->probes; i++) -+ get_random_bytes(((uint32_t *) map->initval)+i, 4); -+ map->elements = 0; -+ map->hashsize = req->hashsize; -+ map->probes = req->probes; -+ map->resize = req->resize; -+ memset(map->cidr, 0, 30 * sizeof(unsigned char)); -+ map->members = harray_malloc(map->hashsize, sizeof(ip_set_ip_t), GFP_KERNEL); -+ if (!map->members) { -+ DP("out of memory for %d bytes", map->hashsize * sizeof(ip_set_ip_t)); -+ kfree(map); -+ return -ENOMEM; -+ } -+ -+ set->data = map; -+ return 0; -+} -+ -+static void destroy(struct ip_set *set) -+{ -+ struct ip_set_nethash *map = (struct ip_set_nethash *) set->data; -+ -+ harray_free(map->members); -+ kfree(map); -+ -+ set->data = NULL; -+} -+ -+static void flush(struct ip_set *set) -+{ -+ struct ip_set_nethash *map = (struct ip_set_nethash *) set->data; -+ harray_flush(map->members, map->hashsize, sizeof(ip_set_ip_t)); -+ memset(map->cidr, 0, 30 * sizeof(unsigned char)); -+ map->elements = 0; -+} -+ -+static void list_header(const struct ip_set *set, void *data) -+{ -+ struct ip_set_nethash *map = (struct ip_set_nethash *) set->data; -+ struct ip_set_req_nethash_create *header = -+ (struct ip_set_req_nethash_create *) data; -+ -+ header->hashsize = map->hashsize; -+ header->probes = map->probes; -+ header->resize = map->resize; -+} -+ -+static int list_members_size(const struct ip_set *set) -+{ -+ struct ip_set_nethash *map = (struct ip_set_nethash *) set->data; -+ -+ return (map->hashsize * sizeof(ip_set_ip_t)); -+} -+ -+static void list_members(const struct ip_set *set, void *data) -+{ -+ struct ip_set_nethash *map = (struct ip_set_nethash *) set->data; -+ ip_set_ip_t i, *elem; -+ -+ for (i = 0; i < map->hashsize; i++) { -+ elem = HARRAY_ELEM(map->members, ip_set_ip_t *, i); -+ ((ip_set_ip_t *)data)[i] = *elem; -+ } -+} -+ -+static struct ip_set_type ip_set_nethash = { -+ .typename = SETTYPE_NAME, -+ .features = IPSET_TYPE_IP | IPSET_DATA_SINGLE, -+ .protocol_version = IP_SET_PROTOCOL_VERSION, -+ .create = &create, -+ .destroy = &destroy, -+ .flush = &flush, -+ .reqsize = sizeof(struct ip_set_req_nethash), -+ .addip = &addip, -+ .addip_kernel = &addip_kernel, -+ .retry = &retry, -+ .delip = &delip, -+ .delip_kernel = &delip_kernel, -+ .testip = &testip, -+ .testip_kernel = &testip_kernel, -+ .header_size = sizeof(struct ip_set_req_nethash_create), -+ .list_header = &list_header, -+ .list_members_size = &list_members_size, -+ .list_members = &list_members, -+ .me = THIS_MODULE, -+}; -+ -+MODULE_LICENSE("GPL"); -+MODULE_AUTHOR("Jozsef Kadlecsik <kadlec@blackhole.kfki.hu>"); -+MODULE_DESCRIPTION("nethash type of IP sets"); -+module_param(limit, int, 0600); -+MODULE_PARM_DESC(limit, "maximal number of elements stored in the sets"); -+ -+static int __init ip_set_nethash_init(void) -+{ -+ return ip_set_register_set_type(&ip_set_nethash); -+} -+ -+static void __exit ip_set_nethash_fini(void) -+{ -+ /* FIXME: possible race with ip_set_create() */ -+ ip_set_unregister_set_type(&ip_set_nethash); -+} -+ -+module_init(ip_set_nethash_init); -+module_exit(ip_set_nethash_fini); ---- /dev/null -+++ b/net/ipv4/netfilter/ip_set_portmap.c -@@ -0,0 +1,346 @@ -+/* Copyright (C) 2003-2004 Jozsef Kadlecsik <kadlec@blackhole.kfki.hu> -+ * -+ * This program is free software; you can redistribute it and/or modify -+ * it under the terms of the GNU General Public License version 2 as -+ * published by the Free Software Foundation. -+ */ -+ -+/* Kernel module implementing a port set type as a bitmap */ -+ -+#include <linux/module.h> -+#include <linux/ip.h> -+#include <linux/tcp.h> -+#include <linux/udp.h> -+#include <linux/skbuff.h> -+#include <linux/version.h> -+#include <linux/netfilter_ipv4/ip_tables.h> -+#include <linux/netfilter_ipv4/ip_set.h> -+#include <linux/errno.h> -+#include <asm/uaccess.h> -+#include <asm/bitops.h> -+#include <linux/spinlock.h> -+ -+#include <net/ip.h> -+ -+#include <linux/netfilter_ipv4/ip_set_portmap.h> -+ -+/* We must handle non-linear skbs */ -+static inline ip_set_ip_t -+get_port(const struct sk_buff *skb, u_int32_t flags) -+{ -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,22) -+ struct iphdr *iph = ip_hdr(skb); -+#else -+ struct iphdr *iph = skb->nh.iph; -+#endif -+ u_int16_t offset = ntohs(iph->frag_off) & IP_OFFSET; -+ switch (iph->protocol) { -+ case IPPROTO_TCP: { -+ struct tcphdr tcph; -+ -+ /* See comments at tcp_match in ip_tables.c */ -+ if (offset) -+ return INVALID_PORT; -+ -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,22) -+ if (skb_copy_bits(skb, ip_hdr(skb)->ihl*4, &tcph, sizeof(tcph)) < 0) -+#else -+ if (skb_copy_bits(skb, skb->nh.iph->ihl*4, &tcph, sizeof(tcph)) < 0) -+#endif -+ /* No choice either */ -+ return INVALID_PORT; -+ -+ return ntohs(flags & IPSET_SRC ? -+ tcph.source : tcph.dest); -+ } -+ case IPPROTO_UDP: { -+ struct udphdr udph; -+ -+ if (offset) -+ return INVALID_PORT; -+ -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,22) -+ if (skb_copy_bits(skb, ip_hdr(skb)->ihl*4, &udph, sizeof(udph)) < 0) -+#else -+ if (skb_copy_bits(skb, skb->nh.iph->ihl*4, &udph, sizeof(udph)) < 0) -+#endif -+ /* No choice either */ -+ return INVALID_PORT; -+ -+ return ntohs(flags & IPSET_SRC ? -+ udph.source : udph.dest); -+ } -+ default: -+ return INVALID_PORT; -+ } -+} -+ -+static inline int -+__testport(struct ip_set *set, ip_set_ip_t port, ip_set_ip_t *hash_port) -+{ -+ struct ip_set_portmap *map = (struct ip_set_portmap *) set->data; -+ -+ if (port < map->first_port || port > map->last_port) -+ return -ERANGE; -+ -+ *hash_port = port; -+ DP("set: %s, port:%u, %u", set->name, port, *hash_port); -+ return !!test_bit(port - map->first_port, map->members); -+} -+ -+static int -+testport(struct ip_set *set, const void *data, size_t size, -+ ip_set_ip_t *hash_port) -+{ -+ struct ip_set_req_portmap *req = -+ (struct ip_set_req_portmap *) data; -+ -+ if (size != sizeof(struct ip_set_req_portmap)) { -+ ip_set_printk("data length wrong (want %zu, have %zu)", -+ sizeof(struct ip_set_req_portmap), -+ size); -+ return -EINVAL; -+ } -+ return __testport(set, req->port, hash_port); -+} -+ -+static int -+testport_kernel(struct ip_set *set, -+ const struct sk_buff *skb, -+ ip_set_ip_t *hash_port, -+ const u_int32_t *flags, -+ unsigned char index) -+{ -+ int res; -+ ip_set_ip_t port = get_port(skb, flags[index]); -+ -+ DP("flag %s port %u", flags[index] & IPSET_SRC ? "SRC" : "DST", port); -+ if (port == INVALID_PORT) -+ return 0; -+ -+ res = __testport(set, port, hash_port); -+ -+ return (res < 0 ? 0 : res); -+} -+ -+static inline int -+__addport(struct ip_set *set, ip_set_ip_t port, ip_set_ip_t *hash_port) -+{ -+ struct ip_set_portmap *map = (struct ip_set_portmap *) set->data; -+ -+ if (port < map->first_port || port > map->last_port) -+ return -ERANGE; -+ if (test_and_set_bit(port - map->first_port, map->members)) -+ return -EEXIST; -+ -+ *hash_port = port; -+ DP("port %u", port); -+ return 0; -+} -+ -+static int -+addport(struct ip_set *set, const void *data, size_t size, -+ ip_set_ip_t *hash_port) -+{ -+ struct ip_set_req_portmap *req = -+ (struct ip_set_req_portmap *) data; -+ -+ if (size != sizeof(struct ip_set_req_portmap)) { -+ ip_set_printk("data length wrong (want %zu, have %zu)", -+ sizeof(struct ip_set_req_portmap), -+ size); -+ return -EINVAL; -+ } -+ return __addport(set, req->port, hash_port); -+} -+ -+static int -+addport_kernel(struct ip_set *set, -+ const struct sk_buff *skb, -+ ip_set_ip_t *hash_port, -+ const u_int32_t *flags, -+ unsigned char index) -+{ -+ ip_set_ip_t port = get_port(skb, flags[index]); -+ -+ if (port == INVALID_PORT) -+ return -EINVAL; -+ -+ return __addport(set, port, hash_port); -+} -+ -+static inline int -+__delport(struct ip_set *set, ip_set_ip_t port, ip_set_ip_t *hash_port) -+{ -+ struct ip_set_portmap *map = (struct ip_set_portmap *) set->data; -+ -+ if (port < map->first_port || port > map->last_port) -+ return -ERANGE; -+ if (!test_and_clear_bit(port - map->first_port, map->members)) -+ return -EEXIST; -+ -+ *hash_port = port; -+ DP("port %u", port); -+ return 0; -+} -+ -+static int -+delport(struct ip_set *set, const void *data, size_t size, -+ ip_set_ip_t *hash_port) -+{ -+ struct ip_set_req_portmap *req = -+ (struct ip_set_req_portmap *) data; -+ -+ if (size != sizeof(struct ip_set_req_portmap)) { -+ ip_set_printk("data length wrong (want %zu, have %zu)", -+ sizeof(struct ip_set_req_portmap), -+ size); -+ return -EINVAL; -+ } -+ return __delport(set, req->port, hash_port); -+} -+ -+static int -+delport_kernel(struct ip_set *set, -+ const struct sk_buff *skb, -+ ip_set_ip_t *hash_port, -+ const u_int32_t *flags, -+ unsigned char index) -+{ -+ ip_set_ip_t port = get_port(skb, flags[index]); -+ -+ if (port == INVALID_PORT) -+ return -EINVAL; -+ -+ return __delport(set, port, hash_port); -+} -+ -+static int create(struct ip_set *set, const void *data, size_t size) -+{ -+ int newbytes; -+ struct ip_set_req_portmap_create *req = -+ (struct ip_set_req_portmap_create *) data; -+ struct ip_set_portmap *map; -+ -+ if (size != sizeof(struct ip_set_req_portmap_create)) { -+ ip_set_printk("data length wrong (want %zu, have %zu)", -+ sizeof(struct ip_set_req_portmap_create), -+ size); -+ return -EINVAL; -+ } -+ -+ DP("from %u to %u", req->from, req->to); -+ -+ if (req->from > req->to) { -+ DP("bad port range"); -+ return -ENOEXEC; -+ } -+ -+ if (req->to - req->from > MAX_RANGE) { -+ ip_set_printk("range too big (max %d ports)", -+ MAX_RANGE+1); -+ return -ENOEXEC; -+ } -+ -+ map = kmalloc(sizeof(struct ip_set_portmap), GFP_KERNEL); -+ if (!map) { -+ DP("out of memory for %d bytes", -+ sizeof(struct ip_set_portmap)); -+ return -ENOMEM; -+ } -+ map->first_port = req->from; -+ map->last_port = req->to; -+ newbytes = bitmap_bytes(req->from, req->to); -+ map->members = kmalloc(newbytes, GFP_KERNEL); -+ if (!map->members) { -+ DP("out of memory for %d bytes", newbytes); -+ kfree(map); -+ return -ENOMEM; -+ } -+ memset(map->members, 0, newbytes); -+ -+ set->data = map; -+ return 0; -+} -+ -+static void destroy(struct ip_set *set) -+{ -+ struct ip_set_portmap *map = (struct ip_set_portmap *) set->data; -+ -+ kfree(map->members); -+ kfree(map); -+ -+ set->data = NULL; -+} -+ -+static void flush(struct ip_set *set) -+{ -+ struct ip_set_portmap *map = (struct ip_set_portmap *) set->data; -+ memset(map->members, 0, bitmap_bytes(map->first_port, map->last_port)); -+} -+ -+static void list_header(const struct ip_set *set, void *data) -+{ -+ struct ip_set_portmap *map = (struct ip_set_portmap *) set->data; -+ struct ip_set_req_portmap_create *header = -+ (struct ip_set_req_portmap_create *) data; -+ -+ DP("list_header %u %u", map->first_port, map->last_port); -+ -+ header->from = map->first_port; -+ header->to = map->last_port; -+} -+ -+static int list_members_size(const struct ip_set *set) -+{ -+ struct ip_set_portmap *map = (struct ip_set_portmap *) set->data; -+ -+ return bitmap_bytes(map->first_port, map->last_port); -+} -+ -+static void list_members(const struct ip_set *set, void *data) -+{ -+ struct ip_set_portmap *map = (struct ip_set_portmap *) set->data; -+ int bytes = bitmap_bytes(map->first_port, map->last_port); -+ -+ memcpy(data, map->members, bytes); -+} -+ -+static struct ip_set_type ip_set_portmap = { -+ .typename = SETTYPE_NAME, -+ .features = IPSET_TYPE_PORT | IPSET_DATA_SINGLE, -+ .protocol_version = IP_SET_PROTOCOL_VERSION, -+ .create = &create, -+ .destroy = &destroy, -+ .flush = &flush, -+ .reqsize = sizeof(struct ip_set_req_portmap), -+ .addip = &addport, -+ .addip_kernel = &addport_kernel, -+ .delip = &delport, -+ .delip_kernel = &delport_kernel, -+ .testip = &testport, -+ .testip_kernel = &testport_kernel, -+ .header_size = sizeof(struct ip_set_req_portmap_create), -+ .list_header = &list_header, -+ .list_members_size = &list_members_size, -+ .list_members = &list_members, -+ .me = THIS_MODULE, -+}; -+ -+MODULE_LICENSE("GPL"); -+MODULE_AUTHOR("Jozsef Kadlecsik <kadlec@blackhole.kfki.hu>"); -+MODULE_DESCRIPTION("portmap type of IP sets"); -+ -+static int __init ip_set_portmap_init(void) -+{ -+ return ip_set_register_set_type(&ip_set_portmap); -+} -+ -+static void __exit ip_set_portmap_fini(void) -+{ -+ /* FIXME: possible race with ip_set_create() */ -+ ip_set_unregister_set_type(&ip_set_portmap); -+} -+ -+module_init(ip_set_portmap_init); -+module_exit(ip_set_portmap_fini); ---- /dev/null -+++ b/net/ipv4/netfilter/ipt_set.c -@@ -0,0 +1,160 @@ -+/* Copyright (C) 2000-2002 Joakim Axelsson <gozem@linux.nu> -+ * Patrick Schaaf <bof@bof.de> -+ * Martin Josefsson <gandalf@wlug.westbo.se> -+ * Copyright (C) 2003-2004 Jozsef Kadlecsik <kadlec@blackhole.kfki.hu> -+ * -+ * This program is free software; you can redistribute it and/or modify -+ * it under the terms of the GNU General Public License version 2 as -+ * published by the Free Software Foundation. -+ */ -+ -+/* Kernel module to match an IP set. */ -+ -+#include <linux/module.h> -+#include <linux/ip.h> -+#include <linux/skbuff.h> -+#include <linux/version.h> -+ -+#include <linux/netfilter_ipv4/ip_tables.h> -+#include <linux/netfilter_ipv4/ip_set.h> -+#include <linux/netfilter_ipv4/ipt_set.h> -+ -+static inline int -+match_set(const struct ipt_set_info *info, -+ const struct sk_buff *skb, -+ int inv) -+{ -+ if (ip_set_testip_kernel(info->index, skb, info->flags)) -+ inv = !inv; -+ return inv; -+} -+ -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,23) -+static bool -+#else -+static int -+#endif -+match(const struct sk_buff *skb, -+ const struct net_device *in, -+ const struct net_device *out, -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,17) -+ const struct xt_match *match, -+#endif -+ const void *matchinfo, -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,23) -+ int offset, unsigned int protoff, bool *hotdrop) -+#elif LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,16) -+ int offset, unsigned int protoff, int *hotdrop) -+#else -+ int offset, int *hotdrop) -+#endif -+{ -+ const struct ipt_set_info_match *info = matchinfo; -+ -+ return match_set(&info->match_set, -+ skb, -+ info->match_set.flags[0] & IPSET_MATCH_INV); -+} -+ -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,23) -+bool -+#else -+static int -+#endif -+checkentry(const char *tablename, -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,16) -+ const void *inf, -+#else -+ const struct ipt_ip *ip, -+#endif -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,17) -+ const struct xt_match *match, -+#endif -+ void *matchinfo, -+#if LINUX_VERSION_CODE < KERNEL_VERSION(2,6,19) -+ unsigned int matchsize, -+#endif -+ unsigned int hook_mask) -+{ -+ struct ipt_set_info_match *info = -+ (struct ipt_set_info_match *) matchinfo; -+ ip_set_id_t index; -+ -+#if LINUX_VERSION_CODE < KERNEL_VERSION(2,6,19) -+ if (matchsize != IPT_ALIGN(sizeof(struct ipt_set_info_match))) { -+ ip_set_printk("invalid matchsize %d", matchsize); -+ return 0; -+ } -+#endif -+ -+ index = ip_set_get_byindex(info->match_set.index); -+ -+ if (index == IP_SET_INVALID_ID) { -+ ip_set_printk("Cannot find set indentified by id %u to match", -+ info->match_set.index); -+ return 0; /* error */ -+ } -+ if (info->match_set.flags[IP_SET_MAX_BINDINGS] != 0) { -+ ip_set_printk("That's nasty!"); -+ return 0; /* error */ -+ } -+ -+ return 1; -+} -+ -+static void destroy( -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,17) -+ const struct xt_match *match, -+#endif -+#if LINUX_VERSION_CODE < KERNEL_VERSION(2,6,19) -+ void *matchinfo, unsigned int matchsize) -+#else -+ void *matchinfo) -+#endif -+{ -+ struct ipt_set_info_match *info = matchinfo; -+ -+#if LINUX_VERSION_CODE < KERNEL_VERSION(2,6,19) -+ if (matchsize != IPT_ALIGN(sizeof(struct ipt_set_info_match))) { -+ ip_set_printk("invalid matchsize %d", matchsize); -+ return; -+ } -+#endif -+ ip_set_put(info->match_set.index); -+} -+ -+static struct ipt_match set_match = { -+ .name = "set", -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,21) -+ .family = AF_INET, -+#endif -+ .match = &match, -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,17) -+ .matchsize = sizeof(struct ipt_set_info_match), -+#endif -+ .checkentry = &checkentry, -+ .destroy = &destroy, -+ .me = THIS_MODULE -+}; -+ -+MODULE_LICENSE("GPL"); -+MODULE_AUTHOR("Jozsef Kadlecsik <kadlec@blackhole.kfki.hu>"); -+MODULE_DESCRIPTION("iptables IP set match module"); -+ -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,21) -+#define ipt_register_match xt_register_match -+#define ipt_unregister_match xt_unregister_match -+#endif -+ -+static int __init ipt_ipset_init(void) -+{ -+ return ipt_register_match(&set_match); -+} -+ -+static void __exit ipt_ipset_fini(void) -+{ -+ ipt_unregister_match(&set_match); -+} -+ -+module_init(ipt_ipset_init); -+module_exit(ipt_ipset_fini); ---- /dev/null -+++ b/net/ipv4/netfilter/ipt_SET.c -@@ -0,0 +1,172 @@ -+/* Copyright (C) 2000-2002 Joakim Axelsson <gozem@linux.nu> -+ * Patrick Schaaf <bof@bof.de> -+ * Martin Josefsson <gandalf@wlug.westbo.se> -+ * Copyright (C) 2003-2004 Jozsef Kadlecsik <kadlec@blackhole.kfki.hu> -+ * -+ * This program is free software; you can redistribute it and/or modify -+ * it under the terms of the GNU General Public License version 2 as -+ * published by the Free Software Foundation. -+ */ -+ -+/* ipt_SET.c - netfilter target to manipulate IP sets */ -+ -+#include <linux/types.h> -+#include <linux/ip.h> -+#include <linux/timer.h> -+#include <linux/module.h> -+#include <linux/netfilter.h> -+#include <linux/netdevice.h> -+#include <linux/if.h> -+#include <linux/inetdevice.h> -+#include <linux/version.h> -+#include <net/protocol.h> -+#include <net/checksum.h> -+#include <linux/netfilter_ipv4.h> -+#include <linux/netfilter_ipv4/ip_tables.h> -+#include <linux/netfilter_ipv4/ipt_set.h> -+ -+static unsigned int -+target(struct sk_buff **pskb, -+ const struct net_device *in, -+ const struct net_device *out, -+ unsigned int hooknum, -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,17) -+ const struct xt_target *target, -+#endif -+#if LINUX_VERSION_CODE < KERNEL_VERSION(2,6,19) -+ const void *targinfo, -+ void *userinfo) -+#else -+ const void *targinfo) -+#endif -+{ -+ const struct ipt_set_info_target *info = targinfo; -+ -+ if (info->add_set.index != IP_SET_INVALID_ID) -+ ip_set_addip_kernel(info->add_set.index, -+ *pskb, -+ info->add_set.flags); -+ if (info->del_set.index != IP_SET_INVALID_ID) -+ ip_set_delip_kernel(info->del_set.index, -+ *pskb, -+ info->del_set.flags); -+ -+ return IPT_CONTINUE; -+} -+ -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,23) -+static bool -+#else -+static int -+#endif -+checkentry(const char *tablename, -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,16) -+ const void *e, -+#else -+ const struct ipt_entry *e, -+#endif -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,17) -+ const struct xt_target *target, -+#endif -+ void *targinfo, -+#if LINUX_VERSION_CODE < KERNEL_VERSION(2,6,19) -+ unsigned int targinfosize, -+#endif -+ unsigned int hook_mask) -+{ -+ struct ipt_set_info_target *info = -+ (struct ipt_set_info_target *) targinfo; -+ ip_set_id_t index; -+ -+#if LINUX_VERSION_CODE < KERNEL_VERSION(2,6,19) -+ if (targinfosize != IPT_ALIGN(sizeof(*info))) { -+ DP("bad target info size %u", targinfosize); -+ return 0; -+ } -+#endif -+ -+ if (info->add_set.index != IP_SET_INVALID_ID) { -+ index = ip_set_get_byindex(info->add_set.index); -+ if (index == IP_SET_INVALID_ID) { -+ ip_set_printk("cannot find add_set index %u as target", -+ info->add_set.index); -+ return 0; /* error */ -+ } -+ } -+ -+ if (info->del_set.index != IP_SET_INVALID_ID) { -+ index = ip_set_get_byindex(info->del_set.index); -+ if (index == IP_SET_INVALID_ID) { -+ ip_set_printk("cannot find del_set index %u as target", -+ info->del_set.index); -+ return 0; /* error */ -+ } -+ } -+ if (info->add_set.flags[IP_SET_MAX_BINDINGS] != 0 -+ || info->del_set.flags[IP_SET_MAX_BINDINGS] != 0) { -+ ip_set_printk("That's nasty!"); -+ return 0; /* error */ -+ } -+ -+ return 1; -+} -+ -+static void destroy( -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,17) -+ const struct xt_target *target, -+#endif -+#if LINUX_VERSION_CODE < KERNEL_VERSION(2,6,19) -+ void *targetinfo, unsigned int targetsize) -+#else -+ void *targetinfo) -+#endif -+{ -+ struct ipt_set_info_target *info = targetinfo; -+ -+#if LINUX_VERSION_CODE < KERNEL_VERSION(2,6,19) -+ if (targetsize != IPT_ALIGN(sizeof(struct ipt_set_info_target))) { -+ ip_set_printk("invalid targetsize %d", targetsize); -+ return; -+ } -+#endif -+ if (info->add_set.index != IP_SET_INVALID_ID) -+ ip_set_put(info->add_set.index); -+ if (info->del_set.index != IP_SET_INVALID_ID) -+ ip_set_put(info->del_set.index); -+} -+ -+static struct ipt_target SET_target = { -+ .name = "SET", -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,21) -+ .family = AF_INET, -+#endif -+ .target = target, -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,17) -+ .targetsize = sizeof(struct ipt_set_info_target), -+#endif -+ .checkentry = checkentry, -+ .destroy = destroy, -+ .me = THIS_MODULE -+}; -+ -+MODULE_LICENSE("GPL"); -+MODULE_AUTHOR("Jozsef Kadlecsik <kadlec@blackhole.kfki.hu>"); -+MODULE_DESCRIPTION("iptables IP set target module"); -+ -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,21) -+#define ipt_register_target xt_register_target -+#define ipt_unregister_target xt_unregister_target -+#endif -+ -+static int __init ipt_SET_init(void) -+{ -+ return ipt_register_target(&SET_target); -+} -+ -+static void __exit ipt_SET_fini(void) -+{ -+ ipt_unregister_target(&SET_target); -+} -+ -+module_init(ipt_SET_init); -+module_exit(ipt_SET_fini); ---- a/net/ipv4/netfilter/Kconfig -+++ b/net/ipv4/netfilter/Kconfig -@@ -408,5 +408,122 @@ - Allows altering the ARP packet payload: source and destination - hardware and network addresses. - -+config IP_NF_SET -+ tristate "IP set support" -+ depends on INET && NETFILTER -+ help -+ This option adds IP set support to the kernel. -+ In order to define and use sets, you need the userspace utility -+ ipset(8). -+ -+ To compile it as a module, choose M here. If unsure, say N. -+ -+config IP_NF_SET_MAX -+ int "Maximum number of IP sets" -+ default 256 -+ range 2 65534 -+ depends on IP_NF_SET -+ help -+ You can define here default value of the maximum number -+ of IP sets for the kernel. -+ -+ The value can be overriden by the 'max_sets' module -+ parameter of the 'ip_set' module. -+ -+config IP_NF_SET_HASHSIZE -+ int "Hash size for bindings of IP sets" -+ default 1024 -+ depends on IP_NF_SET -+ help -+ You can define here default value of the hash size for -+ bindings of IP sets. -+ -+ The value can be overriden by the 'hash_size' module -+ parameter of the 'ip_set' module. -+ -+config IP_NF_SET_IPMAP -+ tristate "ipmap set support" -+ depends on IP_NF_SET -+ help -+ This option adds the ipmap set type support. -+ -+ To compile it as a module, choose M here. If unsure, say N. -+ -+config IP_NF_SET_MACIPMAP -+ tristate "macipmap set support" -+ depends on IP_NF_SET -+ help -+ This option adds the macipmap set type support. -+ -+ To compile it as a module, choose M here. If unsure, say N. -+ -+config IP_NF_SET_PORTMAP -+ tristate "portmap set support" -+ depends on IP_NF_SET -+ help -+ This option adds the portmap set type support. -+ -+ To compile it as a module, choose M here. If unsure, say N. -+ -+config IP_NF_SET_IPHASH -+ tristate "iphash set support" -+ depends on IP_NF_SET -+ help -+ This option adds the iphash set type support. -+ -+ To compile it as a module, choose M here. If unsure, say N. -+ -+config IP_NF_SET_NETHASH -+ tristate "nethash set support" -+ depends on IP_NF_SET -+ help -+ This option adds the nethash set type support. -+ -+ To compile it as a module, choose M here. If unsure, say N. -+ -+config IP_NF_SET_IPPORTHASH -+ tristate "ipporthash set support" -+ depends on IP_NF_SET -+ help -+ This option adds the ipporthash set type support. -+ -+ To compile it as a module, choose M here. If unsure, say N. -+ -+config IP_NF_SET_IPTREE -+ tristate "iptree set support" -+ depends on IP_NF_SET -+ help -+ This option adds the iptree set type support. -+ -+ To compile it as a module, choose M here. If unsure, say N. -+ -+config IP_NF_SET_IPTREEMAP -+ tristate "iptreemap set support" -+ depends on IP_NF_SET -+ help -+ This option adds the iptreemap set type support. -+ -+ To compile it as a module, choose M here. If unsure, say N. -+ -+config IP_NF_MATCH_SET -+ tristate "set match support" -+ depends on IP_NF_SET -+ help -+ Set matching matches against given IP sets. -+ You need the ipset utility to create and set up the sets. -+ -+ To compile it as a module, choose M here. If unsure, say N. -+ -+config IP_NF_TARGET_SET -+ tristate "SET target support" -+ depends on IP_NF_SET -+ help -+ The SET target makes possible to add/delete entries -+ in IP sets. -+ You need the ipset utility to create and set up the sets. -+ -+ To compile it as a module, choose M here. If unsure, say N. -+ -+ - endmenu - ---- a/net/ipv4/netfilter/Makefile -+++ b/net/ipv4/netfilter/Makefile -@@ -48,6 +48,7 @@ - obj-$(CONFIG_IP_NF_MATCH_ECN) += ipt_ecn.o - obj-$(CONFIG_IP_NF_MATCH_AH) += ipt_ah.o - obj-$(CONFIG_IP_NF_MATCH_TTL) += ipt_ttl.o -+obj-$(CONFIG_IP_NF_MATCH_SET) += ipt_set.o - obj-$(CONFIG_IP_NF_MATCH_ADDRTYPE) += ipt_addrtype.o - obj-$(CONFIG_IP_NF_MATCH_IPP2P) += ipt_ipp2p.o - -@@ -63,6 +64,18 @@ - obj-$(CONFIG_IP_NF_TARGET_ULOG) += ipt_ULOG.o - obj-$(CONFIG_IP_NF_TARGET_CLUSTERIP) += ipt_CLUSTERIP.o - obj-$(CONFIG_IP_NF_TARGET_TTL) += ipt_TTL.o -+obj-$(CONFIG_IP_NF_TARGET_SET) += ipt_SET.o -+ -+# sets -+obj-$(CONFIG_IP_NF_SET) += ip_set.o -+obj-$(CONFIG_IP_NF_SET_IPMAP) += ip_set_ipmap.o -+obj-$(CONFIG_IP_NF_SET_PORTMAP) += ip_set_portmap.o -+obj-$(CONFIG_IP_NF_SET_MACIPMAP) += ip_set_macipmap.o -+obj-$(CONFIG_IP_NF_SET_IPHASH) += ip_set_iphash.o -+obj-$(CONFIG_IP_NF_SET_NETHASH) += ip_set_nethash.o -+obj-$(CONFIG_IP_NF_SET_IPPORTHASH) += ip_set_ipporthash.o -+obj-$(CONFIG_IP_NF_SET_IPTREE) += ip_set_iptree.o -+obj-$(CONFIG_IP_NF_SET_IPTREEMAP) += ip_set_iptreemap.o - - # generic ARP tables - obj-$(CONFIG_IP_NF_ARPTABLES) += arp_tables.o diff --git a/target/linux/generic-2.6/patches-2.6.22/140-netfilter_time.patch b/target/linux/generic-2.6/patches-2.6.22/140-netfilter_time.patch deleted file mode 100644 index 420ec70..0000000 --- a/target/linux/generic-2.6/patches-2.6.22/140-netfilter_time.patch +++ /dev/null @@ -1,239 +0,0 @@ ---- /dev/null -+++ b/include/linux/netfilter_ipv4/ipt_time.h -@@ -0,0 +1,18 @@ -+#ifndef __ipt_time_h_included__ -+#define __ipt_time_h_included__ -+ -+ -+struct ipt_time_info { -+ u_int8_t days_match; /* 1 bit per day. -SMTWTFS */ -+ u_int16_t time_start; /* 0 < time_start < 23*60+59 = 1439 */ -+ u_int16_t time_stop; /* 0:0 < time_stat < 23:59 */ -+ -+ /* FIXME: Keep this one for userspace iptables binary compability: */ -+ u_int8_t kerneltime; /* ignore skb time (and use kerneltime) or not. */ -+ -+ time_t date_start; -+ time_t date_stop; -+}; -+ -+ -+#endif /* __ipt_time_h_included__ */ ---- /dev/null -+++ b/net/ipv4/netfilter/ipt_time.c -@@ -0,0 +1,180 @@ -+/* -+ This is a module which is used for time matching -+ It is using some modified code from dietlibc (localtime() function) -+ that you can find at http://www.fefe.de/dietlibc/ -+ This file is distributed under the terms of the GNU General Public -+ License (GPL). Copies of the GPL can be obtained from: ftp://prep.ai.mit.edu/pub/gnu/GPL -+ 2001-05-04 Fabrice MARIE <fabrice@netfilter.org> : initial development. -+ 2001-21-05 Fabrice MARIE <fabrice@netfilter.org> : bug fix in the match code, -+ thanks to "Zeng Yu" <zengy@capitel.com.cn> for bug report. -+ 2001-26-09 Fabrice MARIE <fabrice@netfilter.org> : force the match to be in LOCAL_IN or PRE_ROUTING only. -+ 2001-30-11 Fabrice : added the possibility to use the match in FORWARD/OUTPUT with a little hack, -+ added Nguyen Dang Phuoc Dong <dongnd@tlnet.com.vn> patch to support timezones. -+ 2004-05-02 Fabrice : added support for date matching, from an idea of Fabien COELHO. -+*/ -+ -+#include <linux/module.h> -+#include <linux/skbuff.h> -+#include <linux/netfilter_ipv4/ip_tables.h> -+#include <linux/netfilter_ipv4/ipt_time.h> -+#include <linux/time.h> -+ -+MODULE_AUTHOR("Fabrice MARIE <fabrice@netfilter.org>"); -+MODULE_DESCRIPTION("Match arrival timestamp/date"); -+MODULE_LICENSE("GPL"); -+ -+struct tm -+{ -+ int tm_sec; /* Seconds. [0-60] (1 leap second) */ -+ int tm_min; /* Minutes. [0-59] */ -+ int tm_hour; /* Hours. [0-23] */ -+ int tm_mday; /* Day. [1-31] */ -+ int tm_mon; /* Month. [0-11] */ -+ int tm_year; /* Year - 1900. */ -+ int tm_wday; /* Day of week. [0-6] */ -+ int tm_yday; /* Days in year.[0-365] */ -+ int tm_isdst; /* DST. [-1/0/1]*/ -+ -+ long int tm_gmtoff; /* we don't care, we count from GMT */ -+ const char *tm_zone; /* we don't care, we count from GMT */ -+}; -+ -+void -+localtime(const u32 time, struct tm *r); -+ -+static int -+match(const struct sk_buff *skb, -+ const struct net_device *in, -+ const struct net_device *out, -+ const struct xt_match *match, -+ const void *matchinfo, -+ int offset, -+ unsigned int protoff, -+ int *hotdrop) -+{ -+ const struct ipt_time_info *info = matchinfo; /* match info for rule */ -+ struct timeval tv; -+ struct tm currenttime; /* time human readable */ -+ u_int8_t days_of_week[7] = {64, 32, 16, 8, 4, 2, 1}; -+ u_int16_t packet_time; -+ -+ /* We might not have a timestamp, get one */ -+ if (skb->tstamp.tv64 == 0) -+ __net_timestamp((struct sk_buff *)skb); -+ -+ skb_get_timestamp(skb, &tv); -+ /* First we make sure we are in the date start-stop boundaries */ -+ if ((tv.tv_sec < info->date_start) || (tv.tv_sec > info->date_stop)) -+ return 0; /* We are outside the date boundaries */ -+ -+ /* Transform the timestamp of the packet, in a human readable form */ -+ localtime(tv.tv_sec, ¤ttime); -+ -+ /* check if we match this timestamp, we start by the days... */ -+ if ((days_of_week[currenttime.tm_wday] & info->days_match) != days_of_week[currenttime.tm_wday]) -+ return 0; /* the day doesn't match */ -+ -+ /* ... check the time now */ -+ packet_time = (currenttime.tm_hour * 60) + currenttime.tm_min; -+ if ((packet_time < info->time_start) || (packet_time > info->time_stop)) -+ return 0; -+ -+ /* here we match ! */ -+ return 1; -+} -+ -+static int -+checkentry(const char *tablename, -+ const void *ip, -+ const struct xt_match *match, -+ void *matchinfo, -+ unsigned int hook_mask) -+{ -+ struct ipt_time_info *info = matchinfo; /* match info for rule */ -+ -+ /* First, check that we are in the correct hooks */ -+ if (hook_mask -+ & ~((1 << NF_IP_PRE_ROUTING) | (1 << NF_IP_LOCAL_IN) | (1 << NF_IP_FORWARD) | (1 << NF_IP_LOCAL_OUT))) -+ { -+ printk("ipt_time: error, only valid for PRE_ROUTING, LOCAL_IN, FORWARD and OUTPUT)\n"); -+ return 0; -+ } -+ -+ /* Now check the coherence of the data ... */ -+ if ((info->time_start > 1439) || /* 23*60+59 = 1439*/ -+ (info->time_stop > 1439)) -+ { -+ printk(KERN_WARNING "ipt_time: invalid argument\n"); -+ return 0; -+ } -+ -+ return 1; -+} -+ -+static struct ipt_match time_match = { -+ .name = "time", -+ .match = &match, -+ .matchsize = sizeof(struct ipt_time_info), -+ .checkentry = &checkentry, -+ .me = THIS_MODULE -+}; -+ -+static int __init init(void) -+{ -+ printk("ipt_time loading\n"); -+ return xt_register_match(&time_match); -+} -+ -+static void __exit fini(void) -+{ -+ xt_unregister_match(&time_match); -+ printk("ipt_time unloaded\n"); -+} -+ -+module_init(init); -+module_exit(fini); -+ -+ -+/* The part below is borowed and modified from dietlibc */ -+ -+/* seconds per day */ -+#define SPD 24*60*60 -+ -+void -+localtime(const u32 time, struct tm *r) { -+ u32 i, timep; -+ extern struct timezone sys_tz; -+ const unsigned int __spm[12] = -+ { 0, -+ (31), -+ (31+28), -+ (31+28+31), -+ (31+28+31+30), -+ (31+28+31+30+31), -+ (31+28+31+30+31+30), -+ (31+28+31+30+31+30+31), -+ (31+28+31+30+31+30+31+31), -+ (31+28+31+30+31+30+31+31+30), -+ (31+28+31+30+31+30+31+31+30+31), -+ (31+28+31+30+31+30+31+31+30+31+30), -+ }; -+ register u32 work; -+ -+ timep = time - (sys_tz.tz_minuteswest * 60); -+ work=timep%(SPD); -+ r->tm_sec=work%60; work/=60; -+ r->tm_min=work%60; r->tm_hour=work/60; -+ work=timep/(SPD); -+ r->tm_wday=(4+work)%7; -+ for (i=1970; ; ++i) { -+ register time_t k= (!(i%4) && ((i%100) || !(i%400)))?366:365; -+ if (work>k) -+ work-=k; -+ else -+ break; -+ } -+ r->tm_year=i-1900; -+ for (i=11; i && __spm[i]>work; --i) ; -+ r->tm_mon=i; -+ r->tm_mday=work-__spm[i]+1; -+} ---- a/net/ipv4/netfilter/Kconfig -+++ b/net/ipv4/netfilter/Kconfig -@@ -78,6 +78,22 @@ - - To compile it as a module, choose M here. If unsure, say N. - -+ -+config IP_NF_MATCH_TIME -+ tristate 'TIME match support' -+ depends on IP_NF_IPTABLES -+ help -+ This option adds a `time' match, which allows you -+ to match based on the packet arrival time/date -+ (arrival time/date at the machine which netfilter is running on) or -+ departure time/date (for locally generated packets). -+ -+ If you say Y here, try iptables -m time --help for more information. -+ If you want to compile it as a module, say M here and read -+ -+ Documentation/modules.txt. If unsure, say `N'. -+ -+ - config IP_NF_MATCH_RECENT - tristate "recent match support" - depends on IP_NF_IPTABLES ---- a/net/ipv4/netfilter/Makefile -+++ b/net/ipv4/netfilter/Makefile -@@ -44,6 +44,7 @@ - obj-$(CONFIG_IP_NF_MATCH_IPRANGE) += ipt_iprange.o - obj-$(CONFIG_IP_NF_MATCH_OWNER) += ipt_owner.o - obj-$(CONFIG_IP_NF_MATCH_TOS) += ipt_tos.o -+obj-$(CONFIG_IP_NF_MATCH_TIME) += ipt_time.o - obj-$(CONFIG_IP_NF_MATCH_RECENT) += ipt_recent.o - obj-$(CONFIG_IP_NF_MATCH_ECN) += ipt_ecn.o - obj-$(CONFIG_IP_NF_MATCH_AH) += ipt_ah.o diff --git a/target/linux/generic-2.6/patches-2.6.22/150-netfilter_imq.patch b/target/linux/generic-2.6/patches-2.6.22/150-netfilter_imq.patch deleted file mode 100644 index 17b4569..0000000 --- a/target/linux/generic-2.6/patches-2.6.22/150-netfilter_imq.patch +++ /dev/null @@ -1,869 +0,0 @@ ---- /dev/null -+++ b/drivers/net/imq.c -@@ -0,0 +1,402 @@ -+/* -+ * Pseudo-driver for the intermediate queue device. -+ * -+ * This program is free software; you can redistribute it and/or -+ * modify it under the terms of the GNU General Public License -+ * as published by the Free Software Foundation; either version -+ * 2 of the License, or (at your option) any later version. -+ * -+ * Authors: Patrick McHardy, <kaber@trash.net> -+ * -+ * The first version was written by Martin Devera, <devik@cdi.cz> -+ * -+ * Credits: Jan Rafaj <imq2t@cedric.vabo.cz> -+ * - Update patch to 2.4.21 -+ * Sebastian Strollo <sstrollo@nortelnetworks.com> -+ * - Fix "Dead-loop on netdevice imq"-issue -+ * Marcel Sebek <sebek64@post.cz> -+ * - Update to 2.6.2-rc1 -+ * -+ * After some time of inactivity there is a group taking care -+ * of IMQ again: http://www.linuximq.net -+ * -+ * -+ * 2004/06/30 - New version of IMQ patch to kernels <=2.6.7 including -+ * the following changes: -+ * -+ * - Correction of ipv6 support "+"s issue (Hasso Tepper) -+ * - Correction of imq_init_devs() issue that resulted in -+ * kernel OOPS unloading IMQ as module (Norbert Buchmuller) -+ * - Addition of functionality to choose number of IMQ devices -+ * during kernel config (Andre Correa) -+ * - Addition of functionality to choose how IMQ hooks on -+ * PRE and POSTROUTING (after or before NAT) (Andre Correa) -+ * - Cosmetic corrections (Norbert Buchmuller) (Andre Correa) -+ * -+ * -+ * 2005/12/16 - IMQ versions between 2.6.7 and 2.6.13 were -+ * released with almost no problems. 2.6.14-x was released -+ * with some important changes: nfcache was removed; After -+ * some weeks of trouble we figured out that some IMQ fields -+ * in skb were missing in skbuff.c - skb_clone and copy_skb_header. -+ * These functions are correctly patched by this new patch version. -+ * -+ * Thanks for all who helped to figure out all the problems with -+ * 2.6.14.x: Patrick McHardy, Rune Kock, VeNoMouS, Max CtRiX, -+ * Kevin Shanahan, Richard Lucassen, Valery Dachev (hopefully -+ * I didn't forget anybody). I apologize again for my lack of time. -+ * -+ * More info at: http://www.linuximq.net/ (Andre Correa) -+ */ -+ -+#include <linux/module.h> -+#include <linux/kernel.h> -+#include <linux/moduleparam.h> -+#include <linux/skbuff.h> -+#include <linux/netdevice.h> -+#include <linux/rtnetlink.h> -+#include <linux/if_arp.h> -+#include <linux/netfilter.h> -+#include <linux/netfilter_ipv4.h> -+#if defined(CONFIG_IPV6) || defined (CONFIG_IPV6_MODULE) -+ #include <linux/netfilter_ipv6.h> -+#endif -+#include <linux/imq.h> -+#include <net/pkt_sched.h> -+ -+extern int qdisc_restart1(struct net_device *dev); -+ -+static nf_hookfn imq_nf_hook; -+ -+static struct nf_hook_ops imq_ingress_ipv4 = { -+ .hook = imq_nf_hook, -+ .owner = THIS_MODULE, -+ .pf = PF_INET, -+ .hooknum = NF_IP_PRE_ROUTING, -+#if defined(CONFIG_IMQ_BEHAVIOR_BA) || defined(CONFIG_IMQ_BEHAVIOR_BB) -+ .priority = NF_IP_PRI_MANGLE + 1 -+#else -+ .priority = NF_IP_PRI_NAT_DST + 1 -+#endif -+}; -+ -+static struct nf_hook_ops imq_egress_ipv4 = { -+ .hook = imq_nf_hook, -+ .owner = THIS_MODULE, -+ .pf = PF_INET, -+ .hooknum = NF_IP_POST_ROUTING, -+#if defined(CONFIG_IMQ_BEHAVIOR_AA) || defined(CONFIG_IMQ_BEHAVIOR_BA) -+ .priority = NF_IP_PRI_LAST -+#else -+ .priority = NF_IP_PRI_NAT_SRC - 1 -+#endif -+}; -+ -+#if defined(CONFIG_IPV6) || defined (CONFIG_IPV6_MODULE) -+static struct nf_hook_ops imq_ingress_ipv6 = { -+ .hook = imq_nf_hook, -+ .owner = THIS_MODULE, -+ .pf = PF_INET6, -+ .hooknum = NF_IP6_PRE_ROUTING, -+#if defined(CONFIG_IMQ_BEHAVIOR_BA) || defined(CONFIG_IMQ_BEHAVIOR_BB) -+ .priority = NF_IP6_PRI_MANGLE + 1 -+#else -+ .priority = NF_IP6_PRI_NAT_DST + 1 -+#endif -+}; -+ -+static struct nf_hook_ops imq_egress_ipv6 = { -+ .hook = imq_nf_hook, -+ .owner = THIS_MODULE, -+ .pf = PF_INET6, -+ .hooknum = NF_IP6_POST_ROUTING, -+#if defined(CONFIG_IMQ_BEHAVIOR_AA) || defined(CONFIG_IMQ_BEHAVIOR_BA) -+ .priority = NF_IP6_PRI_LAST -+#else -+ .priority = NF_IP6_PRI_NAT_SRC - 1 -+#endif -+}; -+#endif -+ -+#if defined(CONFIG_IMQ_NUM_DEVS) -+static unsigned int numdevs = CONFIG_IMQ_NUM_DEVS; -+#else -+static unsigned int numdevs = 2; -+#endif -+ -+static struct net_device *imq_devs; -+ -+static struct net_device_stats *imq_get_stats(struct net_device *dev) -+{ -+ return (struct net_device_stats *)dev->priv; -+} -+ -+/* called for packets kfree'd in qdiscs at places other than enqueue */ -+static void imq_skb_destructor(struct sk_buff *skb) -+{ -+ struct nf_info *info = skb->nf_info; -+ -+ if (info) { -+ if (info->indev) -+ dev_put(info->indev); -+ if (info->outdev) -+ dev_put(info->outdev); -+ kfree(info); -+ } -+} -+ -+static int imq_dev_xmit(struct sk_buff *skb, struct net_device *dev) -+{ -+ struct net_device_stats *stats = (struct net_device_stats*) dev->priv; -+ -+ stats->tx_bytes += skb->len; -+ stats->tx_packets++; -+ -+ skb->imq_flags = 0; -+ skb->destructor = NULL; -+ -+ dev->trans_start = jiffies; -+ nf_reinject(skb, skb->nf_info, NF_ACCEPT); -+ return 0; -+} -+ -+static int imq_nf_queue(struct sk_buff *skb, struct nf_info *info, unsigned queue_num, void *data) -+{ -+ struct net_device *dev; -+ struct net_device_stats *stats; -+ struct sk_buff *skb2 = NULL; -+ struct Qdisc *q; -+ unsigned int index = skb->imq_flags&IMQ_F_IFMASK; -+ int ret = -1; -+ -+ if (index > numdevs) -+ return -1; -+ -+ dev = imq_devs + index; -+ if (!(dev->flags & IFF_UP)) { -+ skb->imq_flags = 0; -+ nf_reinject(skb, info, NF_ACCEPT); -+ return 0; -+ } -+ dev->last_rx = jiffies; -+ -+ if (skb->destructor) { -+ skb2 = skb; -+ skb = skb_clone(skb, GFP_ATOMIC); -+ if (!skb) -+ return -1; -+ } -+ skb->nf_info = info; -+ -+ stats = (struct net_device_stats *)dev->priv; -+ stats->rx_bytes+= skb->len; -+ stats->rx_packets++; -+ -+ spin_lock_bh(&dev->queue_lock); -+ q = dev->qdisc; -+ if (q->enqueue) { -+ q->enqueue(skb_get(skb), q); -+ if (skb_shared(skb)) { -+ skb->destructor = imq_skb_destructor; -+ kfree_skb(skb); -+ ret = 0; -+ } -+ } -+ if (spin_is_locked(&dev->_xmit_lock)) -+ netif_schedule(dev); -+ else -+ while (!netif_queue_stopped(dev) && qdisc_restart1(dev) < 0) -+ /* NOTHING */; -+ -+ spin_unlock_bh(&dev->queue_lock); -+ -+ if (skb2) -+ kfree_skb(ret ? skb : skb2); -+ -+ return ret; -+} -+ -+static struct nf_queue_handler nfqh = { -+ .name = "imq", -+ .outfn = imq_nf_queue, -+}; -+ -+static unsigned int imq_nf_hook(unsigned int hook, struct sk_buff **pskb, -+ const struct net_device *indev, -+ const struct net_device *outdev, -+ int (*okfn)(struct sk_buff *)) -+{ -+ if ((*pskb)->imq_flags & IMQ_F_ENQUEUE) -+ return NF_QUEUE; -+ -+ return NF_ACCEPT; -+} -+ -+ -+static int __init imq_init_hooks(void) -+{ -+ int err; -+ -+ err = nf_register_queue_handler(PF_INET, &nfqh); -+ if (err > 0) -+ goto err1; -+ if ((err = nf_register_hook(&imq_ingress_ipv4))) -+ goto err2; -+ if ((err = nf_register_hook(&imq_egress_ipv4))) -+ goto err3; -+#if defined(CONFIG_IPV6) || defined (CONFIG_IPV6_MODULE) -+ if ((err = nf_register_queue_handler(PF_INET6, &nfqh))) -+ goto err4; -+ if ((err = nf_register_hook(&imq_ingress_ipv6))) -+ goto err5; -+ if ((err = nf_register_hook(&imq_egress_ipv6))) -+ goto err6; -+#endif -+ -+ return 0; -+ -+#if defined(CONFIG_IPV6) || defined (CONFIG_IPV6_MODULE) -+err6: -+ nf_unregister_hook(&imq_ingress_ipv6); -+err5: -+ nf_unregister_queue_handler(PF_INET6); -+err4: -+ nf_unregister_hook(&imq_egress_ipv4); -+#endif -+err3: -+ nf_unregister_hook(&imq_ingress_ipv4); -+err2: -+ nf_unregister_queue_handler(PF_INET); -+err1: -+ return err; -+} -+ -+static void __exit imq_unhook(void) -+{ -+ nf_unregister_hook(&imq_ingress_ipv4); -+ nf_unregister_hook(&imq_egress_ipv4); -+ nf_unregister_queue_handler(PF_INET); -+#if defined(CONFIG_IPV6) || defined (CONFIG_IPV6_MODULE) -+ nf_unregister_hook(&imq_ingress_ipv6); -+ nf_unregister_hook(&imq_egress_ipv6); -+ nf_unregister_queue_handler(PF_INET6); -+#endif -+} -+ -+static int __init imq_dev_init(struct net_device *dev) -+{ -+ dev->hard_start_xmit = imq_dev_xmit; -+ dev->type = ARPHRD_VOID; -+ dev->mtu = 1500; -+ dev->tx_queue_len = 30; -+ dev->flags = IFF_NOARP; -+ dev->priv = kmalloc(sizeof(struct net_device_stats), GFP_KERNEL); -+ if (dev->priv == NULL) -+ return -ENOMEM; -+ memset(dev->priv, 0, sizeof(struct net_device_stats)); -+ dev->get_stats = imq_get_stats; -+ -+ return 0; -+} -+ -+static void imq_dev_uninit(struct net_device *dev) -+{ -+ kfree(dev->priv); -+} -+ -+static int __init imq_init_devs(void) -+{ -+ struct net_device *dev; -+ int i,j; -+ j = numdevs; -+ -+ if (!numdevs || numdevs > IMQ_MAX_DEVS) { -+ printk(KERN_ERR "IMQ: numdevs has to be betweed 1 and %u\n", -+ IMQ_MAX_DEVS); -+ return -EINVAL; -+ } -+ -+ imq_devs = kmalloc(sizeof(struct net_device) * numdevs, GFP_KERNEL); -+ if (!imq_devs) -+ return -ENOMEM; -+ memset(imq_devs, 0, sizeof(struct net_device) * numdevs); -+ -+ /* we start counting at zero */ -+ numdevs--; -+ -+ for (i = 0, dev = imq_devs; i <= numdevs; i++, dev++) { -+ SET_MODULE_OWNER(dev); -+ strcpy(dev->name, "imq%d"); -+ dev->init = imq_dev_init; -+ dev->uninit = imq_dev_uninit; -+ -+ if (register_netdev(dev) < 0) -+ goto err_register; -+ } -+ printk(KERN_INFO "IMQ starting with %u devices...\n", j); -+ return 0; -+ -+err_register: -+ for (; i; i--) -+ unregister_netdev(--dev); -+ kfree(imq_devs); -+ return -EIO; -+} -+ -+static void imq_cleanup_devs(void) -+{ -+ int i; -+ struct net_device *dev = imq_devs; -+ -+ for (i = 0; i <= numdevs; i++) -+ unregister_netdev(dev++); -+ -+ kfree(imq_devs); -+} -+ -+static int __init imq_init_module(void) -+{ -+ int err; -+ -+ if ((err = imq_init_devs())) { -+ printk(KERN_ERR "IMQ: Error trying imq_init_devs()\n"); -+ return err; -+ } -+ if ((err = imq_init_hooks())) { -+ printk(KERN_ERR "IMQ: Error trying imq_init_hooks()\n"); -+ imq_cleanup_devs(); -+ return err; -+ } -+ -+ printk(KERN_INFO "IMQ driver loaded successfully.\n"); -+ -+#if defined(CONFIG_IMQ_BEHAVIOR_BA) || defined(CONFIG_IMQ_BEHAVIOR_BB) -+ printk(KERN_INFO "\tHooking IMQ before NAT on PREROUTING.\n"); -+#else -+ printk(KERN_INFO "\tHooking IMQ after NAT on PREROUTING.\n"); -+#endif -+#if defined(CONFIG_IMQ_BEHAVIOR_AB) || defined(CONFIG_IMQ_BEHAVIOR_BB) -+ printk(KERN_INFO "\tHooking IMQ before NAT on POSTROUTING.\n"); -+#else -+ printk(KERN_INFO "\tHooking IMQ after NAT on POSTROUTING.\n"); -+#endif -+ -+ return 0; -+} -+ -+static void __exit imq_cleanup_module(void) -+{ -+ imq_unhook(); -+ imq_cleanup_devs(); -+ printk(KERN_INFO "IMQ driver unloaded successfully.\n"); -+} -+ -+ -+module_init(imq_init_module); -+module_exit(imq_cleanup_module); -+ -+module_param(numdevs, int, 0); -+MODULE_PARM_DESC(numdevs, "number of IMQ devices (how many imq* devices will be created)"); -+MODULE_AUTHOR("http://www.linuximq.net"); -+MODULE_DESCRIPTION("Pseudo-driver for the intermediate queue device. See http://www.linuximq.net/ for more information."); -+MODULE_LICENSE("GPL"); ---- a/drivers/net/Kconfig -+++ b/drivers/net/Kconfig -@@ -96,6 +96,129 @@ - To compile this driver as a module, choose M here: the module - will be called eql. If unsure, say N. - -+config IMQ -+ tristate "IMQ (intermediate queueing device) support" -+ depends on NETDEVICES && NETFILTER -+ ---help--- -+ The IMQ device(s) is used as placeholder for QoS queueing -+ disciplines. Every packet entering/leaving the IP stack can be -+ directed through the IMQ device where it's enqueued/dequeued to the -+ attached qdisc. This allows you to treat network devices as classes -+ and distribute bandwidth among them. Iptables is used to specify -+ through which IMQ device, if any, packets travel. -+ -+ More information at: http://www.linuximq.net/ -+ -+ To compile this driver as a module, choose M here: the module -+ will be called imq. If unsure, say N. -+ -+choice -+ prompt "IMQ behavior (PRE/POSTROUTING)" -+ depends on IMQ -+ default IMQ_BEHAVIOR_BA -+ help -+ -+ This settings defines how IMQ behaves in respect to its -+ hooking in PREROUTING and POSTROUTING. -+ -+ IMQ can work in any of the following ways: -+ -+ PREROUTING | POSTROUTING -+ -----------------|------------------- -+ #1 After NAT | After NAT -+ #2 After NAT | Before NAT -+ #3 Before NAT | After NAT -+ #4 Before NAT | Before NAT -+ -+ The default behavior is to hook before NAT on PREROUTING -+ and after NAT on POSTROUTING (#3). -+ -+ This settings are specially usefull when trying to use IMQ -+ to shape NATed clients. -+ -+ More information can be found at: www.linuximq.net -+ -+ If not sure leave the default settings alone. -+ -+config IMQ_BEHAVIOR_AA -+ bool "IMQ AA" -+ help -+ This settings defines how IMQ behaves in respect to its -+ hooking in PREROUTING and POSTROUTING. -+ -+ Choosing this option will make IMQ hook like this: -+ -+ PREROUTING: After NAT -+ POSTROUTING: After NAT -+ -+ More information can be found at: www.linuximq.net -+ -+ If not sure leave the default settings alone. -+ -+config IMQ_BEHAVIOR_AB -+ bool "IMQ AB" -+ help -+ This settings defines how IMQ behaves in respect to its -+ hooking in PREROUTING and POSTROUTING. -+ -+ Choosing this option will make IMQ hook like this: -+ -+ PREROUTING: After NAT -+ POSTROUTING: Before NAT -+ -+ More information can be found at: www.linuximq.net -+ -+ If not sure leave the default settings alone. -+ -+config IMQ_BEHAVIOR_BA -+ bool "IMQ BA" -+ help -+ This settings defines how IMQ behaves in respect to its -+ hooking in PREROUTING and POSTROUTING. -+ -+ Choosing this option will make IMQ hook like this: -+ -+ PREROUTING: Before NAT -+ POSTROUTING: After NAT -+ -+ More information can be found at: www.linuximq.net -+ -+ If not sure leave the default settings alone. -+ -+config IMQ_BEHAVIOR_BB -+ bool "IMQ BB" -+ help -+ This settings defines how IMQ behaves in respect to its -+ hooking in PREROUTING and POSTROUTING. -+ -+ Choosing this option will make IMQ hook like this: -+ -+ PREROUTING: Before NAT -+ POSTROUTING: Before NAT -+ -+ More information can be found at: www.linuximq.net -+ -+ If not sure leave the default settings alone. -+ -+endchoice -+ -+config IMQ_NUM_DEVS -+ -+ int "Number of IMQ devices" -+ range 2 8 -+ depends on IMQ -+ default "2" -+ help -+ -+ This settings defines how many IMQ devices will be -+ created. -+ -+ The default value is 2. -+ -+ More information can be found at: www.linuximq.net -+ -+ If not sure leave the default settings alone. -+ - config TUN - tristate "Universal TUN/TAP device driver support" - select CRC32 ---- a/drivers/net/Makefile -+++ b/drivers/net/Makefile -@@ -124,6 +124,7 @@ - obj-$(CONFIG_SLHC) += slhc.o - - obj-$(CONFIG_DUMMY) += dummy.o -+obj-$(CONFIG_IMQ) += imq.o - obj-$(CONFIG_IFB) += ifb.o - obj-$(CONFIG_DE600) += de600.o - obj-$(CONFIG_DE620) += de620.o ---- /dev/null -+++ b/include/linux/imq.h -@@ -0,0 +1,9 @@ -+#ifndef _IMQ_H -+#define _IMQ_H -+ -+#define IMQ_MAX_DEVS 16 -+ -+#define IMQ_F_IFMASK 0x7f -+#define IMQ_F_ENQUEUE 0x80 -+ -+#endif /* _IMQ_H */ ---- /dev/null -+++ b/include/linux/netfilter_ipv4/ipt_IMQ.h -@@ -0,0 +1,8 @@ -+#ifndef _IPT_IMQ_H -+#define _IPT_IMQ_H -+ -+struct ipt_imq_info { -+ unsigned int todev; /* target imq device */ -+}; -+ -+#endif /* _IPT_IMQ_H */ ---- /dev/null -+++ b/include/linux/netfilter_ipv6/ip6t_IMQ.h -@@ -0,0 +1,8 @@ -+#ifndef _IP6T_IMQ_H -+#define _IP6T_IMQ_H -+ -+struct ip6t_imq_info { -+ unsigned int todev; /* target imq device */ -+}; -+ -+#endif /* _IP6T_IMQ_H */ ---- a/include/linux/skbuff.h -+++ b/include/linux/skbuff.h -@@ -285,6 +285,10 @@ - struct nf_conntrack *nfct; - struct sk_buff *nfct_reasm; - #endif -+#if defined(CONFIG_IMQ) || defined(CONFIG_IMQ_MODULE) -+ unsigned char imq_flags; -+ struct nf_info *nf_info; -+#endif - #ifdef CONFIG_BRIDGE_NETFILTER - struct nf_bridge_info *nf_bridge; - #endif ---- a/net/core/dev.c -+++ b/net/core/dev.c -@@ -94,6 +94,9 @@ - #include <linux/skbuff.h> - #include <net/sock.h> - #include <linux/rtnetlink.h> -+#if defined(CONFIG_IMQ) || defined(CONFIG_IMQ_MODULE) -+#include <linux/imq.h> -+#endif - #include <linux/proc_fs.h> - #include <linux/seq_file.h> - #include <linux/stat.h> -@@ -1404,6 +1407,9 @@ - { - if (likely(!skb->next)) { - if (!list_empty(&ptype_all)) -+#if defined(CONFIG_IMQ) || defined(CONFIG_IMQ_MODULE) -+ if (!(skb->imq_flags & IMQ_F_ENQUEUE)) -+#endif - dev_queue_xmit_nit(skb, dev); - - if (netif_needs_gso(dev, skb)) { ---- a/net/core/skbuff.c -+++ b/net/core/skbuff.c -@@ -419,6 +419,10 @@ - C(pkt_type); - C(ip_summed); - C(priority); -+#if defined(CONFIG_IMQ) || defined(CONFIG_IMQ_MODULE) -+ C(imq_flags); -+ C(nf_info); -+#endif /*CONFIG_IMQ*/ - #if defined(CONFIG_IP_VS) || defined(CONFIG_IP_VS_MODULE) - C(ipvs_property); - #endif -@@ -485,6 +489,10 @@ - #if defined(CONFIG_IP_VS) || defined(CONFIG_IP_VS_MODULE) - new->ipvs_property = old->ipvs_property; - #endif -+#if defined(CONFIG_IMQ) || defined(CONFIG_IMQ_MODULE) -+ new->imq_flags = old->imq_flags; -+ new->nf_info = old->nf_info; -+#endif /*CONFIG_IMQ*/ - #ifdef CONFIG_NET_SCHED - #ifdef CONFIG_NET_CLS_ACT - new->tc_verd = old->tc_verd; ---- /dev/null -+++ b/net/ipv4/netfilter/ipt_IMQ.c -@@ -0,0 +1,69 @@ -+/* -+ * This target marks packets to be enqueued to an imq device -+ */ -+#include <linux/module.h> -+#include <linux/skbuff.h> -+#include <linux/netfilter_ipv4/ip_tables.h> -+#include <linux/netfilter_ipv4/ipt_IMQ.h> -+#include <linux/imq.h> -+ -+static unsigned int imq_target(struct sk_buff **pskb, -+ const struct net_device *in, -+ const struct net_device *out, -+ unsigned int hooknum, -+ const struct xt_target *target, -+ const void *targinfo) -+{ -+ struct ipt_imq_info *mr = (struct ipt_imq_info*)targinfo; -+ -+ (*pskb)->imq_flags = mr->todev | IMQ_F_ENQUEUE; -+ -+ return XT_CONTINUE; -+} -+ -+static int imq_checkentry(const char *tablename, -+ const void *e, -+ const struct xt_target *target, -+ void *targinfo, -+ unsigned int hook_mask) -+{ -+ struct ipt_imq_info *mr; -+ -+ mr = (struct ipt_imq_info*)targinfo; -+ -+ if (mr->todev > IMQ_MAX_DEVS) { -+ printk(KERN_WARNING -+ "IMQ: invalid device specified, highest is %u\n", -+ IMQ_MAX_DEVS); -+ return 0; -+ } -+ -+ return 1; -+} -+ -+static struct xt_target ipt_imq_reg = { -+ .name = "IMQ", -+ .family = AF_INET, -+ .target = imq_target, -+ .targetsize = sizeof(struct ipt_imq_info), -+ .checkentry = imq_checkentry, -+ .me = THIS_MODULE, -+ .table = "mangle" -+}; -+ -+static int __init init(void) -+{ -+ return xt_register_target(&ipt_imq_reg); -+} -+ -+static void __exit fini(void) -+{ -+ xt_unregister_target(&ipt_imq_reg); -+} -+ -+module_init(init); -+module_exit(fini); -+ -+MODULE_AUTHOR("http://www.linuximq.net"); -+MODULE_DESCRIPTION("Pseudo-driver for the intermediate queue device. See http://www.linuximq.net/ for more information."); -+MODULE_LICENSE("GPL"); ---- a/net/ipv4/netfilter/Kconfig -+++ b/net/ipv4/netfilter/Kconfig -@@ -333,6 +333,17 @@ - - To compile it as a module, choose M here. If unsure, say N. - -+config IP_NF_TARGET_IMQ -+ tristate "IMQ target support" -+ depends on IP_NF_MANGLE -+ help -+ This option adds a `IMQ' target which is used to specify if and -+ to which IMQ device packets should get enqueued/dequeued. -+ -+ For more information visit: http://www.linuximq.net/ -+ -+ To compile it as a module, choose M here. If unsure, say N. -+ - config IP_NF_TARGET_TOS - tristate "TOS target support" - depends on IP_NF_MANGLE ---- a/net/ipv4/netfilter/Makefile -+++ b/net/ipv4/netfilter/Makefile -@@ -57,6 +57,7 @@ - obj-$(CONFIG_IP_NF_TARGET_REJECT) += ipt_REJECT.o - obj-$(CONFIG_IP_NF_TARGET_TOS) += ipt_TOS.o - obj-$(CONFIG_IP_NF_TARGET_ECN) += ipt_ECN.o -+obj-$(CONFIG_IP_NF_TARGET_IMQ) += ipt_IMQ.o - obj-$(CONFIG_IP_NF_TARGET_MASQUERADE) += ipt_MASQUERADE.o - obj-$(CONFIG_IP_NF_TARGET_REDIRECT) += ipt_REDIRECT.o - obj-$(CONFIG_IP_NF_TARGET_NETMAP) += ipt_NETMAP.o ---- /dev/null -+++ b/net/ipv6/netfilter/ip6t_IMQ.c -@@ -0,0 +1,69 @@ -+/* -+ * This target marks packets to be enqueued to an imq device -+ */ -+#include <linux/module.h> -+#include <linux/skbuff.h> -+#include <linux/netfilter_ipv6/ip6_tables.h> -+#include <linux/netfilter_ipv6/ip6t_IMQ.h> -+#include <linux/imq.h> -+ -+static unsigned int imq_target(struct sk_buff **pskb, -+ const struct net_device *in, -+ const struct net_device *out, -+ unsigned int hooknum, -+ const struct xt_target *target, -+ const void *targinfo) -+{ -+ struct ip6t_imq_info *mr = (struct ip6t_imq_info*)targinfo; -+ -+ (*pskb)->imq_flags = mr->todev | IMQ_F_ENQUEUE; -+ -+ return XT_CONTINUE; -+} -+ -+static int imq_checkentry(const char *tablename, -+ const void *entry, -+ const struct xt_target *target, -+ void *targinfo, -+ unsigned int hook_mask) -+{ -+ struct ip6t_imq_info *mr; -+ -+ mr = (struct ip6t_imq_info*)targinfo; -+ -+ if (mr->todev > IMQ_MAX_DEVS) { -+ printk(KERN_WARNING -+ "IMQ: invalid device specified, highest is %u\n", -+ IMQ_MAX_DEVS); -+ return 0; -+ } -+ -+ return 1; -+} -+ -+static struct xt_target ip6t_imq_reg = { -+ .name = "IMQ", -+ .family = AF_INET6, -+ .target = imq_target, -+ .targetsize = sizeof(struct ip6t_imq_info), -+ .table = "mangle", -+ .checkentry = imq_checkentry, -+ .me = THIS_MODULE -+}; -+ -+static int __init init(void) -+{ -+ return xt_register_target(&ip6t_imq_reg); -+} -+ -+static void __exit fini(void) -+{ -+ xt_unregister_target(&ip6t_imq_reg); -+} -+ -+module_init(init); -+module_exit(fini); -+ -+MODULE_AUTHOR("http://www.linuximq.net"); -+MODULE_DESCRIPTION("Pseudo-driver for the intermediate queue device. See http://www.linuximq.net/ for more information."); -+MODULE_LICENSE("GPL"); ---- a/net/ipv6/netfilter/Kconfig -+++ b/net/ipv6/netfilter/Kconfig -@@ -173,6 +173,15 @@ - - To compile it as a module, choose M here. If unsure, say N. - -+config IP6_NF_TARGET_IMQ -+ tristate "IMQ target support" -+ depends on IP6_NF_MANGLE -+ help -+ This option adds a `IMQ' target which is used to specify if and -+ to which imq device packets should get enqueued/dequeued. -+ -+ To compile it as a module, choose M here. If unsure, say N. -+ - config IP6_NF_TARGET_HL - tristate 'HL (hoplimit) target support' - depends on IP6_NF_MANGLE ---- a/net/ipv6/netfilter/Makefile -+++ b/net/ipv6/netfilter/Makefile -@@ -13,6 +13,7 @@ - obj-$(CONFIG_IP6_NF_MATCH_OWNER) += ip6t_owner.o - obj-$(CONFIG_IP6_NF_FILTER) += ip6table_filter.o - obj-$(CONFIG_IP6_NF_MANGLE) += ip6table_mangle.o -+obj-$(CONFIG_IP6_NF_TARGET_IMQ) += ip6t_IMQ.o - obj-$(CONFIG_IP6_NF_TARGET_HL) += ip6t_HL.o - obj-$(CONFIG_IP6_NF_QUEUE) += ip6_queue.o - obj-$(CONFIG_IP6_NF_TARGET_LOG) += ip6t_LOG.o ---- a/net/sched/sch_generic.c -+++ b/net/sched/sch_generic.c -@@ -77,7 +77,6 @@ - - NOTE: Called under dev->queue_lock with locally disabled BH. - */ -- - static inline int qdisc_restart(struct net_device *dev) - { - struct Qdisc *q = dev->qdisc; -@@ -177,6 +176,11 @@ - return q->q.qlen; - } - -+int qdisc_restart1(struct net_device *dev) -+{ -+ return qdisc_restart(dev); -+} -+ - void __qdisc_run(struct net_device *dev) - { - do { -@@ -608,3 +612,4 @@ - EXPORT_SYMBOL(qdisc_reset); - EXPORT_SYMBOL(qdisc_lock_tree); - EXPORT_SYMBOL(qdisc_unlock_tree); -+EXPORT_SYMBOL(qdisc_restart1); diff --git a/target/linux/generic-2.6/patches-2.6.22/160-netfilter_route.patch b/target/linux/generic-2.6/patches-2.6.22/160-netfilter_route.patch deleted file mode 100644 index 109c0d3..0000000 --- a/target/linux/generic-2.6/patches-2.6.22/160-netfilter_route.patch +++ /dev/null @@ -1,948 +0,0 @@ ---- /dev/null -+++ b/include/linux/netfilter_ipv4/ipt_ROUTE.h -@@ -0,0 +1,23 @@ -+/* Header file for iptables ipt_ROUTE target -+ * -+ * (C) 2002 by Cédric de Launois <delaunois@info.ucl.ac.be> -+ * -+ * This software is distributed under GNU GPL v2, 1991 -+ */ -+#ifndef _IPT_ROUTE_H_target -+#define _IPT_ROUTE_H_target -+ -+#define IPT_ROUTE_IFNAMSIZ 16 -+ -+struct ipt_route_target_info { -+ char oif[IPT_ROUTE_IFNAMSIZ]; /* Output Interface Name */ -+ char iif[IPT_ROUTE_IFNAMSIZ]; /* Input Interface Name */ -+ u_int32_t gw; /* IP address of gateway */ -+ u_int8_t flags; -+}; -+ -+/* Values for "flags" field */ -+#define IPT_ROUTE_CONTINUE 0x01 -+#define IPT_ROUTE_TEE 0x02 -+ -+#endif /*_IPT_ROUTE_H_target*/ ---- /dev/null -+++ b/include/linux/netfilter_ipv6/ip6t_ROUTE.h -@@ -0,0 +1,23 @@ -+/* Header file for iptables ip6t_ROUTE target -+ * -+ * (C) 2003 by Cédric de Launois <delaunois@info.ucl.ac.be> -+ * -+ * This software is distributed under GNU GPL v2, 1991 -+ */ -+#ifndef _IPT_ROUTE_H_target -+#define _IPT_ROUTE_H_target -+ -+#define IP6T_ROUTE_IFNAMSIZ 16 -+ -+struct ip6t_route_target_info { -+ char oif[IP6T_ROUTE_IFNAMSIZ]; /* Output Interface Name */ -+ char iif[IP6T_ROUTE_IFNAMSIZ]; /* Input Interface Name */ -+ u_int32_t gw[4]; /* IPv6 address of gateway */ -+ u_int8_t flags; -+}; -+ -+/* Values for "flags" field */ -+#define IP6T_ROUTE_CONTINUE 0x01 -+#define IP6T_ROUTE_TEE 0x02 -+ -+#endif /*_IP6T_ROUTE_H_target*/ ---- /dev/null -+++ b/net/ipv4/netfilter/ipt_ROUTE.c -@@ -0,0 +1,483 @@ -+/* -+ * This implements the ROUTE target, which enables you to setup unusual -+ * routes not supported by the standard kernel routing table. -+ * -+ * Copyright (C) 2002 Cedric de Launois <delaunois@info.ucl.ac.be> -+ * -+ * v 1.11 2004/11/23 -+ * -+ * This software is distributed under GNU GPL v2, 1991 -+ */ -+ -+#include <linux/module.h> -+#include <linux/skbuff.h> -+#include <linux/ip.h> -+#include <linux/netfilter_ipv4/ip_tables.h> -+#include <net/netfilter/nf_conntrack.h> -+#include <linux/netfilter_ipv4/ipt_ROUTE.h> -+#include <linux/netdevice.h> -+#include <linux/route.h> -+#include <linux/version.h> -+#include <linux/if_arp.h> -+#include <net/ip.h> -+#include <net/route.h> -+#include <net/icmp.h> -+#include <net/checksum.h> -+ -+#if 0 -+#define DEBUGP printk -+#else -+#define DEBUGP(format, args...) -+#endif -+ -+MODULE_LICENSE("GPL"); -+MODULE_AUTHOR("Cedric de Launois <delaunois@info.ucl.ac.be>"); -+MODULE_DESCRIPTION("iptables ROUTE target module"); -+ -+/* Try to route the packet according to the routing keys specified in -+ * route_info. Keys are : -+ * - ifindex : -+ * 0 if no oif preferred, -+ * otherwise set to the index of the desired oif -+ * - route_info->gw : -+ * 0 if no gateway specified, -+ * otherwise set to the next host to which the pkt must be routed -+ * If success, skb->dev is the output device to which the packet must -+ * be sent and skb->dst is not NULL -+ * -+ * RETURN: -1 if an error occured -+ * 1 if the packet was succesfully routed to the -+ * destination desired -+ * 0 if the kernel routing table could not route the packet -+ * according to the keys specified -+ */ -+static int route(struct sk_buff *skb, -+ unsigned int ifindex, -+ const struct ipt_route_target_info *route_info) -+{ -+ int err; -+ struct rtable *rt; -+ struct iphdr *iph = ip_hdr(skb); -+ struct flowi fl = { -+ .oif = ifindex, -+ .nl_u = { -+ .ip4_u = { -+ .daddr = iph->daddr, -+ .saddr = 0, -+ .tos = RT_TOS(iph->tos), -+ .scope = RT_SCOPE_UNIVERSE, -+ } -+ } -+ }; -+ -+ /* The destination address may be overloaded by the target */ -+ if (route_info->gw) -+ fl.fl4_dst = route_info->gw; -+ -+ /* Trying to route the packet using the standard routing table. */ -+ if ((err = ip_route_output_key(&rt, &fl))) { -+ if (net_ratelimit()) -+ DEBUGP("ipt_ROUTE: couldn't route pkt (err: %i)",err); -+ return -1; -+ } -+ -+ /* Drop old route. */ -+ dst_release(skb->dst); -+ skb->dst = NULL; -+ -+ /* Success if no oif specified or if the oif correspond to the -+ * one desired */ -+ if (!ifindex || rt->u.dst.dev->ifindex == ifindex) { -+ skb->dst = &rt->u.dst; -+ skb->dev = skb->dst->dev; -+ skb->protocol = htons(ETH_P_IP); -+ return 1; -+ } -+ -+ /* The interface selected by the routing table is not the one -+ * specified by the user. This may happen because the dst address -+ * is one of our own addresses. -+ */ -+ if (net_ratelimit()) -+ DEBUGP("ipt_ROUTE: failed to route as desired gw=%u.%u.%u.%u oif=%i (got oif=%i)\n", -+ NIPQUAD(route_info->gw), ifindex, rt->u.dst.dev->ifindex); -+ -+ return 0; -+} -+ -+ -+/* Stolen from ip_finish_output2 -+ * PRE : skb->dev is set to the device we are leaving by -+ * skb->dst is not NULL -+ * POST: the packet is sent with the link layer header pushed -+ * the packet is destroyed -+ */ -+static void ip_direct_send(struct sk_buff *skb) -+{ -+ struct dst_entry *dst = skb->dst; -+ struct hh_cache *hh = dst->hh; -+ struct net_device *dev = dst->dev; -+ int hh_len = LL_RESERVED_SPACE(dev); -+ -+ /* Be paranoid, rather than too clever. */ -+ if (unlikely(skb_headroom(skb) < hh_len && dev->hard_header)) { -+ struct sk_buff *skb2; -+ -+ skb2 = skb_realloc_headroom(skb, LL_RESERVED_SPACE(dev)); -+ if (skb2 == NULL) { -+ kfree_skb(skb); -+ return; -+ } -+ if (skb->sk) -+ skb_set_owner_w(skb2, skb->sk); -+ kfree_skb(skb); -+ skb = skb2; -+ } -+ -+ if (hh) { -+ int hh_alen; -+ -+ read_lock_bh(&hh->hh_lock); -+ hh_alen = HH_DATA_ALIGN(hh->hh_len); -+ memcpy(skb->data - hh_alen, hh->hh_data, hh_alen); -+ read_unlock_bh(&hh->hh_lock); -+ skb_push(skb, hh->hh_len); -+ hh->hh_output(skb); -+ } else if (dst->neighbour) -+ dst->neighbour->output(skb); -+ else { -+ if (net_ratelimit()) -+ DEBUGP(KERN_DEBUG "ipt_ROUTE: no hdr & no neighbour cache!\n"); -+ kfree_skb(skb); -+ } -+} -+ -+ -+/* PRE : skb->dev is set to the device we are leaving by -+ * POST: - the packet is directly sent to the skb->dev device, without -+ * pushing the link layer header. -+ * - the packet is destroyed -+ */ -+static inline int dev_direct_send(struct sk_buff *skb) -+{ -+ return dev_queue_xmit(skb); -+} -+ -+ -+static unsigned int route_oif(const struct ipt_route_target_info *route_info, -+ struct sk_buff *skb) -+{ -+ unsigned int ifindex = 0; -+ struct net_device *dev_out = NULL; -+ -+ /* The user set the interface name to use. -+ * Getting the current interface index. -+ */ -+ if ((dev_out = dev_get_by_name(route_info->oif))) { -+ ifindex = dev_out->ifindex; -+ } else { -+ /* Unknown interface name : packet dropped */ -+ if (net_ratelimit()) -+ DEBUGP("ipt_ROUTE: oif interface %s not found\n", route_info->oif); -+ return NF_DROP; -+ } -+ -+ /* Trying the standard way of routing packets */ -+ switch (route(skb, ifindex, route_info)) { -+ case 1: -+ dev_put(dev_out); -+ if (route_info->flags & IPT_ROUTE_CONTINUE) -+ return IPT_CONTINUE; -+ -+ ip_direct_send(skb); -+ return NF_STOLEN; -+ -+ case 0: -+ /* Failed to send to oif. Trying the hard way */ -+ if (route_info->flags & IPT_ROUTE_CONTINUE) -+ return NF_DROP; -+ -+ if (net_ratelimit()) -+ DEBUGP("ipt_ROUTE: forcing the use of %i\n", -+ ifindex); -+ -+ /* We have to force the use of an interface. -+ * This interface must be a tunnel interface since -+ * otherwise we can't guess the hw address for -+ * the packet. For a tunnel interface, no hw address -+ * is needed. -+ */ -+ if ((dev_out->type != ARPHRD_TUNNEL) -+ && (dev_out->type != ARPHRD_IPGRE)) { -+ if (net_ratelimit()) -+ DEBUGP("ipt_ROUTE: can't guess the hw addr !\n"); -+ dev_put(dev_out); -+ return NF_DROP; -+ } -+ -+ /* Send the packet. This will also free skb -+ * Do not go through the POST_ROUTING hook because -+ * skb->dst is not set and because it will probably -+ * get confused by the destination IP address. -+ */ -+ skb->dev = dev_out; -+ dev_direct_send(skb); -+ dev_put(dev_out); -+ return NF_STOLEN; -+ -+ default: -+ /* Unexpected error */ -+ dev_put(dev_out); -+ return NF_DROP; -+ } -+} -+ -+ -+static unsigned int route_iif(const struct ipt_route_target_info *route_info, -+ struct sk_buff *skb) -+{ -+ struct net_device *dev_in = NULL; -+ -+ /* Getting the current interface index. */ -+ if (!(dev_in = dev_get_by_name(route_info->iif))) { -+ if (net_ratelimit()) -+ DEBUGP("ipt_ROUTE: iif interface %s not found\n", route_info->iif); -+ return NF_DROP; -+ } -+ -+ skb->dev = dev_in; -+ dst_release(skb->dst); -+ skb->dst = NULL; -+ -+ netif_rx(skb); -+ dev_put(dev_in); -+ return NF_STOLEN; -+} -+ -+ -+static unsigned int route_gw(const struct ipt_route_target_info *route_info, -+ struct sk_buff *skb) -+{ -+ if (route(skb, 0, route_info)!=1) -+ return NF_DROP; -+ -+ if (route_info->flags & IPT_ROUTE_CONTINUE) -+ return IPT_CONTINUE; -+ -+ ip_direct_send(skb); -+ return NF_STOLEN; -+} -+ -+ -+/* To detect and deter routed packet loopback when using the --tee option, -+ * we take a page out of the raw.patch book: on the copied skb, we set up -+ * a fake ->nfct entry, pointing to the local &route_tee_track. We skip -+ * routing packets when we see they already have that ->nfct. -+ */ -+ -+static struct nf_conn route_tee_track; -+ -+static unsigned int ipt_route_target(struct sk_buff **pskb, -+ const struct net_device *in, -+ const struct net_device *out, -+ unsigned int hooknum, -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,17) -+ const struct xt_target *target, -+#endif -+#if LINUX_VERSION_CODE < KERNEL_VERSION(2,6,19) -+ const void *targinfo, -+ void *userinfo) -+#else -+ const void *targinfo) -+#endif -+{ -+ const struct ipt_route_target_info *route_info = targinfo; -+ struct sk_buff *skb = *pskb; -+ unsigned int res; -+ -+ if (skb->nfct == &route_tee_track.ct_general) { -+ /* Loopback - a packet we already routed, is to be -+ * routed another time. Avoid that, now. -+ */ -+ if (net_ratelimit()) -+ DEBUGP(KERN_DEBUG "ipt_ROUTE: loopback - DROP!\n"); -+ return NF_DROP; -+ } -+ -+ /* If we are at PREROUTING or INPUT hook -+ * the TTL isn't decreased by the IP stack -+ */ -+ if (hooknum == NF_IP_PRE_ROUTING || -+ hooknum == NF_IP_LOCAL_IN) { -+ -+ struct iphdr *iph = ip_hdr(skb); -+ -+ if (iph->ttl <= 1) { -+ struct rtable *rt; -+ struct flowi fl = { -+ .oif = 0, -+ .nl_u = { -+ .ip4_u = { -+ .daddr = iph->daddr, -+ .saddr = iph->saddr, -+ .tos = RT_TOS(iph->tos), -+ .scope = ((iph->tos & RTO_ONLINK) ? -+ RT_SCOPE_LINK : -+ RT_SCOPE_UNIVERSE) -+ } -+ } -+ }; -+ -+ if (ip_route_output_key(&rt, &fl)) { -+ return NF_DROP; -+ } -+ -+ if (skb->dev == rt->u.dst.dev) { -+ /* Drop old route. */ -+ dst_release(skb->dst); -+ skb->dst = &rt->u.dst; -+ -+ /* this will traverse normal stack, and -+ * thus call conntrack on the icmp packet */ -+ icmp_send(skb, ICMP_TIME_EXCEEDED, -+ ICMP_EXC_TTL, 0); -+ } -+ -+ return NF_DROP; -+ } -+ -+ /* -+ * If we are at INPUT the checksum must be recalculated since -+ * the length could change as the result of a defragmentation. -+ */ -+ if(hooknum == NF_IP_LOCAL_IN) { -+ iph->ttl = iph->ttl - 1; -+ iph->check = 0; -+ iph->check = ip_fast_csum((unsigned char *)iph, iph->ihl); -+ } else { -+ ip_decrease_ttl(iph); -+ } -+ } -+ -+ if ((route_info->flags & IPT_ROUTE_TEE)) { -+ /* -+ * Copy the *pskb, and route the copy. Will later return -+ * IPT_CONTINUE for the original skb, which should continue -+ * on its way as if nothing happened. The copy should be -+ * independantly delivered to the ROUTE --gw. -+ */ -+ skb = skb_copy(*pskb, GFP_ATOMIC); -+ if (!skb) { -+ if (net_ratelimit()) -+ DEBUGP(KERN_DEBUG "ipt_ROUTE: copy failed!\n"); -+ return IPT_CONTINUE; -+ } -+ } -+ -+ /* Tell conntrack to forget this packet since it may get confused -+ * when a packet is leaving with dst address == our address. -+ * Good idea ? Dunno. Need advice. -+ * -+ * NEW: mark the skb with our &route_tee_track, so we avoid looping -+ * on any already routed packet. -+ */ -+ if (!(route_info->flags & IPT_ROUTE_CONTINUE)) { -+ nf_conntrack_put(skb->nfct); -+ skb->nfct = &route_tee_track.ct_general; -+ skb->nfctinfo = IP_CT_NEW; -+ nf_conntrack_get(skb->nfct); -+ } -+ -+ if (route_info->oif[0] != '\0') { -+ res = route_oif(route_info, skb); -+ } else if (route_info->iif[0] != '\0') { -+ res = route_iif(route_info, skb); -+ } else if (route_info->gw) { -+ res = route_gw(route_info, skb); -+ } else { -+ if (net_ratelimit()) -+ DEBUGP(KERN_DEBUG "ipt_ROUTE: no parameter !\n"); -+ res = IPT_CONTINUE; -+ } -+ -+ if ((route_info->flags & IPT_ROUTE_TEE)) -+ res = IPT_CONTINUE; -+ -+ return res; -+} -+ -+ -+static int ipt_route_checkentry(const char *tablename, -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,16) -+ const void *e, -+#else -+ const struct ipt_ip *ip, -+#endif -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,17) -+ const struct xt_target *target, -+#endif -+ void *targinfo, -+#if LINUX_VERSION_CODE < KERNEL_VERSION(2,6,19) -+ unsigned int targinfosize, -+#endif -+ unsigned int hook_mask) -+{ -+ if (strcmp(tablename, "mangle") != 0) { -+ printk("ipt_ROUTE: bad table `%s', use the `mangle' table.\n", -+ tablename); -+ return 0; -+ } -+ -+ if (hook_mask & ~( (1 << NF_IP_PRE_ROUTING) -+ | (1 << NF_IP_LOCAL_IN) -+ | (1 << NF_IP_FORWARD) -+ | (1 << NF_IP_LOCAL_OUT) -+ | (1 << NF_IP_POST_ROUTING))) { -+ printk("ipt_ROUTE: bad hook\n"); -+ return 0; -+ } -+ -+#if LINUX_VERSION_CODE < KERNEL_VERSION(2,6,19) -+ if (targinfosize != IPT_ALIGN(sizeof(struct ipt_route_target_info))) { -+ printk(KERN_WARNING "ipt_ROUTE: targinfosize %u != %Zu\n", -+ targinfosize, -+ IPT_ALIGN(sizeof(struct ipt_route_target_info))); -+ return 0; -+ } -+#endif -+ -+ return 1; -+} -+ -+ -+static struct ipt_target ipt_route_reg = { -+ .name = "ROUTE", -+ .target = ipt_route_target, -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,17) -+ .targetsize = sizeof(struct ipt_route_target_info), -+#endif -+ .checkentry = ipt_route_checkentry, -+ .me = THIS_MODULE, -+}; -+ -+static int __init init(void) -+{ -+ /* Set up fake conntrack (stolen from raw.patch): -+ - to never be deleted, not in any hashes */ -+ atomic_set(&route_tee_track.ct_general.use, 1); -+ /* - and look it like as a confirmed connection */ -+ set_bit(IPS_CONFIRMED_BIT, &route_tee_track.status); -+ /* Initialize fake conntrack so that NAT will skip it */ -+ route_tee_track.status |= IPS_NAT_DONE_MASK; -+ -+ return xt_register_target(&ipt_route_reg); -+} -+ -+ -+static void __exit fini(void) -+{ -+ xt_unregister_target(&ipt_route_reg); -+} -+ -+module_init(init); -+module_exit(fini); ---- a/net/ipv4/netfilter/Kconfig -+++ b/net/ipv4/netfilter/Kconfig -@@ -552,5 +552,22 @@ - To compile it as a module, choose M here. If unsure, say N. - - -+config IP_NF_TARGET_ROUTE -+ tristate 'ROUTE target support' -+ depends on IP_NF_MANGLE -+ help -+ This option adds a `ROUTE' target, which enables you to setup unusual -+ routes. For example, the ROUTE lets you route a received packet through -+ an interface or towards a host, even if the regular destination of the -+ packet is the router itself. The ROUTE target is also able to change the -+ incoming interface of a packet. -+ -+ The target can be or not a final target. It has to be used inside the -+ mangle table. -+ -+ If you want to compile it as a module, say M here and read -+ Documentation/modules.txt. The module will be called ipt_ROUTE.o. -+ If unsure, say `N'. -+ - endmenu - ---- a/net/ipv4/netfilter/Makefile -+++ b/net/ipv4/netfilter/Makefile -@@ -60,6 +60,7 @@ - obj-$(CONFIG_IP_NF_TARGET_IMQ) += ipt_IMQ.o - obj-$(CONFIG_IP_NF_TARGET_MASQUERADE) += ipt_MASQUERADE.o - obj-$(CONFIG_IP_NF_TARGET_REDIRECT) += ipt_REDIRECT.o -+obj-$(CONFIG_IP_NF_TARGET_ROUTE) += ipt_ROUTE.o - obj-$(CONFIG_IP_NF_TARGET_NETMAP) += ipt_NETMAP.o - obj-$(CONFIG_IP_NF_TARGET_SAME) += ipt_SAME.o - obj-$(CONFIG_IP_NF_TARGET_LOG) += ipt_LOG.o ---- a/net/ipv6/ndisc.c -+++ b/net/ipv6/ndisc.c -@@ -154,6 +154,8 @@ - .gc_thresh3 = 1024, - }; - -+EXPORT_SYMBOL(nd_tbl); -+ - /* ND options */ - struct ndisc_options { - struct nd_opt_hdr *nd_opt_array[__ND_OPT_ARRAY_MAX]; ---- /dev/null -+++ b/net/ipv6/netfilter/ip6t_ROUTE.c -@@ -0,0 +1,330 @@ -+/* -+ * This implements the ROUTE v6 target, which enables you to setup unusual -+ * routes not supported by the standard kernel routing table. -+ * -+ * Copyright (C) 2003 Cedric de Launois <delaunois@info.ucl.ac.be> -+ * -+ * v 1.1 2004/11/23 -+ * -+ * This software is distributed under GNU GPL v2, 1991 -+ */ -+ -+#include <linux/module.h> -+#include <linux/skbuff.h> -+#include <linux/ipv6.h> -+#include <linux/netfilter_ipv6/ip6_tables.h> -+#include <linux/netfilter_ipv6/ip6t_ROUTE.h> -+#include <linux/netdevice.h> -+#include <linux/version.h> -+#include <net/ipv6.h> -+#include <net/ndisc.h> -+#include <net/ip6_route.h> -+#include <linux/icmpv6.h> -+ -+#if 1 -+#define DEBUGP printk -+#else -+#define DEBUGP(format, args...) -+#endif -+ -+#define NIP6(addr) \ -+ ntohs((addr).s6_addr16[0]), \ -+ ntohs((addr).s6_addr16[1]), \ -+ ntohs((addr).s6_addr16[2]), \ -+ ntohs((addr).s6_addr16[3]), \ -+ ntohs((addr).s6_addr16[4]), \ -+ ntohs((addr).s6_addr16[5]), \ -+ ntohs((addr).s6_addr16[6]), \ -+ ntohs((addr).s6_addr16[7]) -+ -+/* Route the packet according to the routing keys specified in -+ * route_info. Keys are : -+ * - ifindex : -+ * 0 if no oif preferred, -+ * otherwise set to the index of the desired oif -+ * - route_info->gw : -+ * 0 if no gateway specified, -+ * otherwise set to the next host to which the pkt must be routed -+ * If success, skb->dev is the output device to which the packet must -+ * be sent and skb->dst is not NULL -+ * -+ * RETURN: 1 if the packet was succesfully routed to the -+ * destination desired -+ * 0 if the kernel routing table could not route the packet -+ * according to the keys specified -+ */ -+static int -+route6(struct sk_buff *skb, -+ unsigned int ifindex, -+ const struct ip6t_route_target_info *route_info) -+{ -+ struct rt6_info *rt = NULL; -+ struct ipv6hdr *ipv6h = ipv6_hdr(skb); -+ struct in6_addr *gw = (struct in6_addr*)&route_info->gw; -+ -+ DEBUGP("ip6t_ROUTE: called with: "); -+ DEBUGP("DST=%04x:%04x:%04x:%04x:%04x:%04x:%04x:%04x ", NIP6(ipv6h->daddr)); -+ DEBUGP("GATEWAY=%04x:%04x:%04x:%04x:%04x:%04x:%04x:%04x ", NIP6(*gw)); -+ DEBUGP("OUT=%s\n", route_info->oif); -+ -+ if (ipv6_addr_any(gw)) -+ rt = rt6_lookup(&ipv6h->daddr, &ipv6h->saddr, ifindex, 1); -+ else -+ rt = rt6_lookup(gw, &ipv6h->saddr, ifindex, 1); -+ -+ if (!rt) -+ goto no_route; -+ -+ DEBUGP("ip6t_ROUTE: routing gives: "); -+ DEBUGP("DST=%04x:%04x:%04x:%04x:%04x:%04x:%04x:%04x ", NIP6(rt->rt6i_dst.addr)); -+ DEBUGP("GATEWAY=%04x:%04x:%04x:%04x:%04x:%04x:%04x:%04x ", NIP6(rt->rt6i_gateway)); -+ DEBUGP("OUT=%s\n", rt->rt6i_dev->name); -+ -+ if (ifindex && rt->rt6i_dev->ifindex!=ifindex) -+ goto wrong_route; -+ -+ if (!rt->rt6i_nexthop) { -+ DEBUGP("ip6t_ROUTE: discovering neighbour\n"); -+ rt->rt6i_nexthop = ndisc_get_neigh(rt->rt6i_dev, &rt->rt6i_dst.addr); -+ } -+ -+ /* Drop old route. */ -+ dst_release(skb->dst); -+ skb->dst = &rt->u.dst; -+ skb->dev = rt->rt6i_dev; -+ return 1; -+ -+ wrong_route: -+ dst_release(&rt->u.dst); -+ no_route: -+ if (!net_ratelimit()) -+ return 0; -+ -+ printk("ip6t_ROUTE: no explicit route found "); -+ if (ifindex) -+ printk("via interface %s ", route_info->oif); -+ if (!ipv6_addr_any(gw)) -+ printk("via gateway %04x:%04x:%04x:%04x:%04x:%04x:%04x:%04x", NIP6(*gw)); -+ printk("\n"); -+ return 0; -+} -+ -+ -+/* Stolen from ip6_output_finish -+ * PRE : skb->dev is set to the device we are leaving by -+ * skb->dst is not NULL -+ * POST: the packet is sent with the link layer header pushed -+ * the packet is destroyed -+ */ -+static void ip_direct_send(struct sk_buff *skb) -+{ -+ struct dst_entry *dst = skb->dst; -+ struct hh_cache *hh = dst->hh; -+ -+ if (hh) { -+ read_lock_bh(&hh->hh_lock); -+ memcpy(skb->data - 16, hh->hh_data, 16); -+ read_unlock_bh(&hh->hh_lock); -+ skb_push(skb, hh->hh_len); -+ hh->hh_output(skb); -+ } else if (dst->neighbour) -+ dst->neighbour->output(skb); -+ else { -+ if (net_ratelimit()) -+ DEBUGP(KERN_DEBUG "ip6t_ROUTE: no hdr & no neighbour cache!\n"); -+ kfree_skb(skb); -+ } -+} -+ -+ -+static unsigned int -+route6_oif(const struct ip6t_route_target_info *route_info, -+ struct sk_buff *skb) -+{ -+ unsigned int ifindex = 0; -+ struct net_device *dev_out = NULL; -+ -+ /* The user set the interface name to use. -+ * Getting the current interface index. -+ */ -+ if ((dev_out = dev_get_by_name(route_info->oif))) { -+ ifindex = dev_out->ifindex; -+ } else { -+ /* Unknown interface name : packet dropped */ -+ if (net_ratelimit()) -+ DEBUGP("ip6t_ROUTE: oif interface %s not found\n", route_info->oif); -+ -+ if (route_info->flags & IP6T_ROUTE_CONTINUE) -+ return IP6T_CONTINUE; -+ else -+ return NF_DROP; -+ } -+ -+ /* Trying the standard way of routing packets */ -+ if (route6(skb, ifindex, route_info)) { -+ dev_put(dev_out); -+ if (route_info->flags & IP6T_ROUTE_CONTINUE) -+ return IP6T_CONTINUE; -+ -+ ip_direct_send(skb); -+ return NF_STOLEN; -+ } else -+ return NF_DROP; -+} -+ -+ -+static unsigned int -+route6_gw(const struct ip6t_route_target_info *route_info, -+ struct sk_buff *skb) -+{ -+ if (route6(skb, 0, route_info)) { -+ if (route_info->flags & IP6T_ROUTE_CONTINUE) -+ return IP6T_CONTINUE; -+ -+ ip_direct_send(skb); -+ return NF_STOLEN; -+ } else -+ return NF_DROP; -+} -+ -+ -+static unsigned int -+ip6t_route_target(struct sk_buff **pskb, -+ const struct net_device *in, -+ const struct net_device *out, -+ unsigned int hooknum, -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,17) -+ const struct xt_target *target, -+#endif -+#if LINUX_VERSION_CODE < KERNEL_VERSION(2,6,19) -+ const void *targinfo, -+ void *userinfo) -+#else -+ const void *targinfo) -+#endif -+{ -+ const struct ip6t_route_target_info *route_info = targinfo; -+ struct sk_buff *skb = *pskb; -+ struct in6_addr *gw = (struct in6_addr*)&route_info->gw; -+ unsigned int res; -+ -+ if (route_info->flags & IP6T_ROUTE_CONTINUE) -+ goto do_it; -+ -+ /* If we are at PREROUTING or INPUT hook -+ * the TTL isn't decreased by the IP stack -+ */ -+ if (hooknum == NF_IP6_PRE_ROUTING || -+ hooknum == NF_IP6_LOCAL_IN) { -+ -+ struct ipv6hdr *ipv6h = ipv6_hdr(skb); -+ -+ if (ipv6h->hop_limit <= 1) { -+ /* Force OUTPUT device used as source address */ -+ skb->dev = skb->dst->dev; -+ -+ icmpv6_send(skb, ICMPV6_TIME_EXCEED, -+ ICMPV6_EXC_HOPLIMIT, 0, skb->dev); -+ -+ return NF_DROP; -+ } -+ -+ ipv6h->hop_limit--; -+ } -+ -+ if ((route_info->flags & IP6T_ROUTE_TEE)) { -+ /* -+ * Copy the *pskb, and route the copy. Will later return -+ * IP6T_CONTINUE for the original skb, which should continue -+ * on its way as if nothing happened. The copy should be -+ * independantly delivered to the ROUTE --gw. -+ */ -+ skb = skb_copy(*pskb, GFP_ATOMIC); -+ if (!skb) { -+ if (net_ratelimit()) -+ DEBUGP(KERN_DEBUG "ip6t_ROUTE: copy failed!\n"); -+ return IP6T_CONTINUE; -+ } -+ } -+ -+do_it: -+ if (route_info->oif[0]) { -+ res = route6_oif(route_info, skb); -+ } else if (!ipv6_addr_any(gw)) { -+ res = route6_gw(route_info, skb); -+ } else { -+ if (net_ratelimit()) -+ DEBUGP(KERN_DEBUG "ip6t_ROUTE: no parameter !\n"); -+ res = IP6T_CONTINUE; -+ } -+ -+ if ((route_info->flags & IP6T_ROUTE_TEE)) -+ res = IP6T_CONTINUE; -+ -+ return res; -+} -+ -+ -+static int -+ip6t_route_checkentry(const char *tablename, -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,16) -+ const void *entry, -+#else -+ const struct ip6t_entry *entry -+#endif -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,17) -+ const struct xt_target *target, -+#endif -+ void *targinfo, -+#if LINUX_VERSION_CODE < KERNEL_VERSION(2,6,19) -+ unsigned int targinfosize, -+#endif -+ unsigned int hook_mask) -+{ -+ if (strcmp(tablename, "mangle") != 0) { -+ printk("ip6t_ROUTE: can only be called from \"mangle\" table.\n"); -+ return 0; -+ } -+ -+#if LINUX_VERSION_CODE < KERNEL_VERSION(2,6,19) -+ if (targinfosize != IP6T_ALIGN(sizeof(struct ip6t_route_target_info))) { -+ printk(KERN_WARNING "ip6t_ROUTE: targinfosize %u != %Zu\n", -+ targinfosize, -+ IP6T_ALIGN(sizeof(struct ip6t_route_target_info))); -+ return 0; -+ } -+#endif -+ -+ return 1; -+} -+ -+ -+static struct ip6t_target ip6t_route_reg = { -+ .name = "ROUTE", -+ .target = ip6t_route_target, -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,17) -+ .targetsize = sizeof(struct ip6t_route_target_info), -+#endif -+ .checkentry = ip6t_route_checkentry, -+ .me = THIS_MODULE -+}; -+ -+ -+static int __init init(void) -+{ -+ printk(KERN_DEBUG "registering ipv6 ROUTE target\n"); -+ if (xt_register_target(&ip6t_route_reg)) -+ return -EINVAL; -+ -+ return 0; -+} -+ -+ -+static void __exit fini(void) -+{ -+ xt_unregister_target(&ip6t_route_reg); -+} -+ -+module_init(init); -+module_exit(fini); -+MODULE_LICENSE("GPL"); ---- a/net/ipv6/netfilter/Kconfig -+++ b/net/ipv6/netfilter/Kconfig -@@ -209,5 +209,18 @@ - If you want to compile it as a module, say M here and read - <file:Documentation/kbuild/modules.txt>. If unsure, say `N'. - -+config IP6_NF_TARGET_ROUTE -+ tristate 'ROUTE target support' -+ depends on IP6_NF_MANGLE -+ help -+ This option adds a `ROUTE' target, which enables you to setup unusual -+ routes. The ROUTE target is also able to change the incoming interface -+ of a packet. -+ -+ The target can be or not a final target. It has to be used inside the -+ mangle table. -+ -+ Not working as a module. -+ - endmenu - ---- a/net/ipv6/netfilter/Makefile -+++ b/net/ipv6/netfilter/Makefile -@@ -20,6 +20,7 @@ - obj-$(CONFIG_IP6_NF_RAW) += ip6table_raw.o - obj-$(CONFIG_IP6_NF_MATCH_HL) += ip6t_hl.o - obj-$(CONFIG_IP6_NF_TARGET_REJECT) += ip6t_REJECT.o -+obj-$(CONFIG_IP6_NF_TARGET_ROUTE) += ip6t_ROUTE.o - obj-$(CONFIG_IP6_NF_MATCH_MH) += ip6t_mh.o - - # objects for l3 independent conntrack diff --git a/target/linux/generic-2.6/patches-2.6.22/170-netfilter_chaostables_0.8.patch b/target/linux/generic-2.6/patches-2.6.22/170-netfilter_chaostables_0.8.patch deleted file mode 100644 index e3c15cd..0000000 --- a/target/linux/generic-2.6/patches-2.6.22/170-netfilter_chaostables_0.8.patch +++ /dev/null @@ -1,844 +0,0 @@ ---- /dev/null -+++ b/include/linux/netfilter/oot_conntrack.h -@@ -0,0 +1,5 @@ -+#if defined(CONFIG_IP_NF_CONNTRACK) || defined(CONFIG_IP_NF_CONNTRACK_MODULE) -+# include <linux/netfilter_ipv4/ip_conntrack.h> -+#else /* linux-2.6.20+ */ -+# include <net/netfilter/nf_nat_rule.h> -+#endif ---- /dev/null -+++ b/include/linux/netfilter/oot_trans.h -@@ -0,0 +1,14 @@ -+/* Out of tree workarounds */ -+#include <linux/version.h> -+#if LINUX_VERSION_CODE <= KERNEL_VERSION(2, 6, 18) -+# define HAVE_MATCHINFOSIZE 1 -+# define HAVE_TARGUSERINFO 1 -+# define HAVE_TARGINFOSIZE 1 -+#endif -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2, 6, 20) -+# define nfmark mark -+#endif -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2, 6, 21) -+# define tcp_v4_check(tcph, tcph_sz, s, d, csp) \ -+ tcp_v4_check((tcph_sz), (s), (d), (csp)) -+#endif ---- /dev/null -+++ b/include/linux/netfilter/xt_CHAOS.h -@@ -0,0 +1,14 @@ -+#ifndef _LINUX_NETFILTER_XT_CHAOS_H -+#define _LINUX_NETFILTER_XT_CHAOS_H 1 -+ -+enum xt_chaos_target_variant { -+ XTCHAOS_NORMAL, -+ XTCHAOS_TARPIT, -+ XTCHAOS_DELUDE, -+}; -+ -+struct xt_chaos_target_info { -+ uint8_t variant; -+}; -+ -+#endif /* _LINUX_NETFILTER_XT_CHAOS_H */ ---- /dev/null -+++ b/include/linux/netfilter/xt_portscan.h -@@ -0,0 +1,8 @@ -+#ifndef _LINUX_NETFILTER_XT_PORTSCAN_H -+#define _LINUX_NETFILTER_XT_PORTSCAN_H 1 -+ -+struct xt_portscan_match_info { -+ uint8_t match_stealth, match_syn, match_cn, match_gr; -+}; -+ -+#endif /* _LINUX_NETFILTER_XT_PORTSCAN_H */ ---- /dev/null -+++ b/net/netfilter/find_match.c -@@ -0,0 +1,39 @@ -+/* -+ xt_request_find_match -+ by Jan Engelhardt <jengelh [at] gmx de>, 2006 - 2007 -+ -+ Based upon linux-2.6.18.5/net/netfilter/x_tables.c: -+ Copyright (C) 2006-2006 Harald Welte <laforge@netfilter.org> -+ This program is free software; you can redistribute it and/or modify -+ it under the terms of the GNU General Public License version 2 as -+ published by the Free Software Foundation. -+*/ -+#include <linux/err.h> -+#include <linux/netfilter_arp.h> -+#include <linux/socket.h> -+#include <linux/netfilter/x_tables.h> -+ -+/* -+ * Yeah this code is sub-optimal, but the function is missing in -+ * mainline so far. -jengelh -+ */ -+static struct xt_match *xt_request_find_match_lo(int af, const char *name, -+ u8 revision) -+{ -+ static const char *const xt_prefix[] = { -+ [AF_INET] = "ip", -+ [AF_INET6] = "ip6", -+ [NF_ARP] = "arp", -+ }; -+ struct xt_match *match; -+ -+ match = try_then_request_module(xt_find_match(af, name, revision), -+ "%st_%s", xt_prefix[af], name); -+ if (IS_ERR(match) || match == NULL) -+ return NULL; -+ -+ return match; -+} -+ -+/* In case it goes into mainline, let this out-of-tree package compile */ -+#define xt_request_find_match xt_request_find_match_lo ---- a/net/netfilter/Kconfig -+++ b/net/netfilter/Kconfig -@@ -255,6 +255,14 @@ - - # alphabetically ordered list of targets - -+config NETFILTER_XT_TARGET_CHAOS -+ tristate '"CHAOS" target support' -+ depends on NETFILTER_XTABLES -+ help -+ This option adds a `CHAOS' target. -+ -+ To compile it as a module, choose M here. If unsure, say N. -+ - config NETFILTER_XT_TARGET_CLASSIFY - tristate '"CLASSIFY" target support' - depends on NETFILTER_XTABLES -@@ -282,6 +290,14 @@ - <file:Documentation/kbuild/modules.txt>. The module will be called - ipt_CONNMARK.ko. If unsure, say `N'. - -+config NETFILTER_XT_TARGET_DELUDE -+ tristate '"DELUDE" target support' -+ depends on NETFILTER_XTABLES -+ help -+ This option adds a `DELUDE' target. -+ -+ To compile it as a module, choose M here. If unsure, say N. -+ - config NETFILTER_XT_TARGET_DSCP - tristate '"DSCP" target support' - depends on NETFILTER_XTABLES -@@ -526,6 +542,14 @@ - - To compile it as a module, choose M here. If unsure, say N. - -+config NETFILTER_XT_MATCH_PORTSCAN -+ tristate '"portscan" match support' -+ depends on NETFILTER_XTABLES -+ help -+ This option adds a 'portscan' match support. -+ -+ To compile it as a module, choose M here. If unsure, say N. -+ - config NETFILTER_XT_MATCH_MULTIPORT - tristate "Multiple port match support" - depends on NETFILTER_XTABLES ---- a/net/netfilter/Makefile -+++ b/net/netfilter/Makefile -@@ -47,6 +47,8 @@ - obj-$(CONFIG_NETFILTER_XT_TARGET_SECMARK) += xt_SECMARK.o - obj-$(CONFIG_NETFILTER_XT_TARGET_TCPMSS) += xt_TCPMSS.o - obj-$(CONFIG_NETFILTER_XT_TARGET_CONNSECMARK) += xt_CONNSECMARK.o -+obj-$(CONFIG_NETFILTER_XT_TARGET_CHAOS) += xt_CHAOS.o -+obj-$(CONFIG_NETFILTER_XT_TARGET_DELUDE) += xt_DELUDE.o - - # matches - obj-$(CONFIG_NETFILTER_XT_MATCH_COMMENT) += xt_comment.o -@@ -74,3 +76,4 @@ - obj-$(CONFIG_NETFILTER_XT_MATCH_TCPMSS) += xt_tcpmss.o - obj-$(CONFIG_NETFILTER_XT_MATCH_PHYSDEV) += xt_physdev.o - obj-$(CONFIG_NETFILTER_XT_MATCH_HASHLIMIT) += xt_hashlimit.o -+obj-$(CONFIG_NETFILTER_XT_MATCH_PORTSCAN) += xt_portscan.o ---- /dev/null -+++ b/net/netfilter/xt_CHAOS.c -@@ -0,0 +1,200 @@ -+/* -+ * CHAOS target for netfilter -+ * Copyright © CC Computer Consultants GmbH, 2006 - 2007 -+ * Contact: Jan Engelhardt <jengelh@computergmbh.de> -+ * -+ * This program is free software; you can redistribute it and/or modify -+ * it under the terms of the GNU General Public License; either version -+ * 2 or 3 as published by the Free Software Foundation. -+ */ -+#include <linux/icmp.h> -+#include <linux/in.h> -+#include <linux/ip.h> -+#include <linux/module.h> -+#include <linux/skbuff.h> -+#include <linux/stat.h> -+#include <linux/netfilter/x_tables.h> -+#include <linux/netfilter/xt_tcpudp.h> -+#include <linux/netfilter_ipv4/ipt_REJECT.h> -+#include <net/ip.h> -+#if defined(_LOCAL) -+# include "xt_CHAOS.h" -+# include "find_match.c" -+#elif defined(CONFIG_NETFILTER_XT_TARGET_CHAOS) || \ -+ defined(CONFIG_NETFILTER_XT_TARGET_CHAOS_MODULE) -+# include <linux/netfilter/xt_CHAOS.h> -+# include "find_match.c" -+#else -+# include "xt_CHAOS.h" -+# include "find_match.c" -+#endif -+#define PFX KBUILD_MODNAME ": " -+ -+/* Module parameters */ -+static unsigned int reject_percentage = ~0U * .01; -+static unsigned int delude_percentage = ~0U * .0101; -+module_param(reject_percentage, uint, S_IRUGO | S_IWUSR); -+module_param(delude_percentage, uint, S_IRUGO | S_IWUSR); -+ -+/* References to other matches/targets */ -+static struct xt_match *xm_tcp; -+static struct xt_target *xt_delude, *xt_reject, *xt_tarpit; -+ -+static int have_delude, have_tarpit; -+ -+/* Static data for other matches/targets */ -+static const struct ipt_reject_info reject_params = { -+ .with = ICMP_HOST_UNREACH, -+}; -+ -+static const struct xt_tcp tcp_params = { -+ .spts = {0, ~0}, -+ .dpts = {0, ~0}, -+}; -+ -+/* CHAOS functions */ -+static void xt_chaos_total(const struct xt_chaos_target_info *info, -+ struct sk_buff **pskb, const struct net_device *in, -+ const struct net_device *out, unsigned int hooknum) -+{ -+ const struct iphdr *iph = ip_hdr(*pskb); -+ const int protoff = 4 * iph->ihl; -+ const int offset = ntohs(iph->frag_off) & IP_OFFSET; -+ const struct xt_target *destiny; -+ int hotdrop = 0, ret; -+ -+ ret = xm_tcp->match(*pskb, in, out, xm_tcp, &tcp_params, -+ offset, protoff, &hotdrop); -+ if (!ret || hotdrop || (unsigned int)net_random() > delude_percentage) -+ return; -+ -+ destiny = (info->variant == XTCHAOS_TARPIT) ? xt_tarpit : xt_delude; -+ destiny->target(pskb, in, out, hooknum, destiny, NULL); -+ return; -+} -+ -+static unsigned int chaos_tg(struct sk_buff **pskb, -+ const struct net_device *in, const struct net_device *out, -+ unsigned int hooknum, const struct xt_target *target, const void *targinfo) -+{ -+ /* -+ * Equivalent to: -+ * -A chaos -m statistic --mode random --probability \ -+ * $reject_percentage -j REJECT --reject-with host-unreach; -+ * -A chaos -p tcp -m statistic --mode random --probability \ -+ * $delude_percentage -j DELUDE; -+ * -A chaos -j DROP; -+ */ -+ const struct xt_chaos_target_info *info = targinfo; -+ const struct iphdr *iph = ip_hdr(*pskb); -+ -+ if ((unsigned int)net_random() <= reject_percentage) -+ return xt_reject->target(pskb, in, out, hooknum, target, -+ &reject_params); -+ -+ /* TARPIT/DELUDE may not be called from the OUTPUT chain */ -+ if (iph->protocol == IPPROTO_TCP && -+ info->variant != XTCHAOS_NORMAL && hooknum != NF_IP_LOCAL_OUT) -+ xt_chaos_total(info, pskb, in, out, hooknum); -+ -+ return NF_DROP; -+} -+ -+static int chaos_tg_check(const char *tablename, const void *entry, -+ const struct xt_target *target, void *targinfo, unsigned int hook_mask) -+{ -+ const struct xt_chaos_target_info *info = targinfo; -+ -+ if (info->variant == XTCHAOS_DELUDE && !have_delude) { -+ printk(KERN_WARNING PFX "Error: Cannot use --delude when " -+ "DELUDE module not available\n"); -+ return false; -+ } -+ if (info->variant == XTCHAOS_TARPIT && !have_tarpit) { -+ printk(KERN_WARNING PFX "Error: Cannot use --tarpit when " -+ "TARPIT module not available\n"); -+ return false; -+ } -+ -+ return true; -+} -+ -+static struct xt_target chaos_tg_reg = { -+ .name = "CHAOS", -+ .family = AF_INET, -+ .table = "filter", -+ .hooks = (1 << NF_IP_LOCAL_IN) | (1 << NF_IP_FORWARD) | -+ (1 << NF_IP_LOCAL_OUT), -+ .checkentry = chaos_tg_check, -+ .target = chaos_tg, -+ .targetsize = sizeof(struct xt_chaos_target_info), -+ .me = THIS_MODULE, -+}; -+ -+static int __init chaos_tg_init(void) -+{ -+ int ret = -EINVAL; -+ -+ xm_tcp = xt_request_find_match(AF_INET, "tcp", 0); -+ if (xm_tcp == NULL) { -+ printk(KERN_WARNING PFX "Error: Could not find or load " -+ "\"tcp\" match\n"); -+ return -EINVAL; -+ } -+ -+ xt_reject = xt_request_find_target(AF_INET, "REJECT", 0); -+ if (xt_reject == NULL) { -+ printk(KERN_WARNING PFX "Error: Could not find or load " -+ "\"REJECT\" target\n"); -+ goto out2; -+ } -+ -+ xt_tarpit = xt_request_find_target(AF_INET, "TARPIT", 0); -+ have_tarpit = xt_tarpit != NULL; -+ if (!have_tarpit) -+ printk(KERN_WARNING PFX "Warning: Could not find or load " -+ "\"TARPIT\" target\n"); -+ -+ xt_delude = xt_request_find_target(AF_INET, "DELUDE", 0); -+ have_delude = xt_delude != NULL; -+ if (!have_delude) -+ printk(KERN_WARNING PFX "Warning: Could not find or load " -+ "\"DELUDE\" target\n"); -+ -+ if ((ret = xt_register_target(&chaos_tg_reg)) != 0) { -+ printk(KERN_WARNING PFX "xt_register_target returned " -+ "error %d\n", ret); -+ goto out3; -+ } -+ -+ return 0; -+ -+ out3: -+ if (have_delude) -+ module_put(xt_delude->me); -+ if (have_tarpit) -+ module_put(xt_tarpit->me); -+ module_put(xt_reject->me); -+ out2: -+ module_put(xm_tcp->me); -+ return ret; -+} -+ -+static void __exit chaos_tg_exit(void) -+{ -+ xt_unregister_target(&chaos_tg_reg); -+ module_put(xm_tcp->me); -+ module_put(xt_reject->me); -+ if (have_delude) -+ module_put(xt_delude->me); -+ if (have_tarpit) -+ module_put(xt_tarpit->me); -+ return; -+} -+ -+module_init(chaos_tg_init); -+module_exit(chaos_tg_exit); -+MODULE_AUTHOR("Jan Engelhardt <jengelh@computergmbh.de>"); -+MODULE_DESCRIPTION("netfilter \"CHAOS\" target"); -+MODULE_LICENSE("GPL"); -+MODULE_ALIAS("ipt_CHAOS"); ---- /dev/null -+++ b/net/netfilter/xt_DELUDE.c -@@ -0,0 +1,197 @@ -+/* -+ * DELUDE target -+ * Copyright © CC Computer Consultants GmbH, 2007 -+ * Contact: Jan Engelhardt <jengelh@computergmbh.de> -+ * -+ * Based upon linux-2.6.18.5/net/ipv4/netfilter/ipt_REJECT.c: -+ * (C) 1999-2001 Paul `Rusty' Russell -+ * (C) 2002-2004 Netfilter Core Team <coreteam@netfilter.org> -+ * -+ * xt_DELUDE acts like REJECT, but does reply with SYN-ACK on SYN. -+ * -+ * This program is free software; you can redistribute it and/or modify -+ * it under the terms of the GNU General Public License version 2 as -+ * published by the Free Software Foundation. -+ */ -+#include <linux/module.h> -+#include <linux/skbuff.h> -+#include <linux/ip.h> -+#include <linux/tcp.h> -+#include <linux/netfilter/x_tables.h> -+#ifdef CONFIG_BRIDGE_NETFILTER -+# include <linux/netfilter_bridge.h> -+#endif -+#include <net/tcp.h> -+#define PFX KBUILD_MODNAME ": " -+ -+static void delude_send_reset(struct sk_buff *oldskb, unsigned int hook) -+{ -+ struct tcphdr _otcph, *oth, *tcph; -+ unsigned int addr_type; -+ struct sk_buff *nskb; -+ u_int16_t tmp_port; -+ u_int32_t tmp_addr; -+ struct iphdr *niph; -+ bool needs_ack; -+ -+ /* IP header checks: fragment. */ -+ if (ip_hdr(oldskb)->frag_off & htons(IP_OFFSET)) -+ return; -+ -+ oth = skb_header_pointer(oldskb, ip_hdrlen(oldskb), -+ sizeof(_otcph), &_otcph); -+ if (oth == NULL) -+ return; -+ -+ /* No RST for RST. */ -+ if (oth->rst) -+ return; -+ -+ /* Check checksum */ -+ if (nf_ip_checksum(oldskb, hook, ip_hdrlen(oldskb), IPPROTO_TCP)) -+ return; -+ -+ /* We need a linear, writeable skb. We also need to expand -+ headroom in case hh_len of incoming interface < hh_len of -+ outgoing interface */ -+ nskb = skb_copy_expand(oldskb, LL_MAX_HEADER, skb_tailroom(oldskb), -+ GFP_ATOMIC); -+ if (!nskb) -+ return; -+ -+ /* This packet will not be the same as the other: clear nf fields */ -+ nf_reset(nskb); -+ nskb->mark = 0; -+ skb_init_secmark(nskb); -+ -+ skb_shinfo(nskb)->gso_size = 0; -+ skb_shinfo(nskb)->gso_segs = 0; -+ skb_shinfo(nskb)->gso_type = 0; -+ -+ tcph = (struct tcphdr *)(skb_network_header(nskb) + ip_hdrlen(nskb)); -+ -+ /* Swap source and dest */ -+ niph = ip_hdr(nskb); -+ tmp_addr = niph->saddr; -+ niph->saddr = niph->daddr; -+ niph->daddr = tmp_addr; -+ tmp_port = tcph->source; -+ tcph->source = tcph->dest; -+ tcph->dest = tmp_port; -+ -+ /* Truncate to length (no data) */ -+ tcph->doff = sizeof(struct tcphdr) / 4; -+ skb_trim(nskb, ip_hdrlen(nskb) + sizeof(struct tcphdr)); -+ niph->tot_len = htons(nskb->len); -+ -+ if (oth->syn && !oth->ack && !oth->rst && !oth->fin) { -+ /* DELUDE essential part */ -+ tcph->ack_seq = htonl(ntohl(oth->seq) + oth->syn + oth->fin + -+ oldskb->len - ip_hdrlen(oldskb) - -+ (oth->doff << 2)); -+ tcph->seq = false; -+ tcph->ack = true; -+ } else { -+ if (!tcph->ack) { -+ needs_ack = true; -+ tcph->ack_seq = htonl(ntohl(oth->seq) + oth->syn + -+ oth->fin + oldskb->len - -+ ip_hdrlen(oldskb) - (oth->doff<<2)); -+ tcph->seq = false; -+ } else { -+ needs_ack = false; -+ tcph->seq = oth->ack_seq; -+ tcph->ack_seq = false; -+ } -+ -+ /* Reset flags */ -+ ((u_int8_t *)tcph)[13] = 0; -+ tcph->rst = true; -+ tcph->ack = needs_ack; -+ } -+ -+ tcph->window = 0; -+ tcph->urg_ptr = 0; -+ -+ /* Adjust TCP checksum */ -+ tcph->check = 0; -+ tcph->check = tcp_v4_check(sizeof(struct tcphdr), niph->saddr, -+ niph->daddr, csum_partial((char *)tcph, -+ sizeof(struct tcphdr), 0)); -+ -+ /* Set DF, id = 0 */ -+ niph->frag_off = htons(IP_DF); -+ niph->id = 0; -+ -+ addr_type = RTN_UNSPEC; -+#ifdef CONFIG_BRIDGE_NETFILTER -+ if (hook != NF_IP_FORWARD || (nskb->nf_bridge != NULL && -+ nskb->nf_bridge->mask & BRNF_BRIDGED)) -+#else -+ if (hook != NF_IP_FORWARD) -+#endif -+ addr_type = RTN_LOCAL; -+ -+ if (ip_route_me_harder(&nskb, addr_type)) -+ goto free_nskb; -+ -+ nskb->ip_summed = CHECKSUM_NONE; -+ -+ /* Adjust IP TTL */ -+ niph->ttl = dst_metric(nskb->dst, RTAX_HOPLIMIT); -+ -+ /* Adjust IP checksum */ -+ niph->check = 0; -+ niph->check = ip_fast_csum(skb_network_header(nskb), niph->ihl); -+ -+ /* "Never happens" */ -+ if (nskb->len > dst_mtu(nskb->dst)) -+ goto free_nskb; -+ -+ nf_ct_attach(nskb, oldskb); -+ -+ NF_HOOK(PF_INET, NF_IP_LOCAL_OUT, nskb, NULL, nskb->dst->dev, -+ dst_output); -+ return; -+ -+ free_nskb: -+ kfree_skb(nskb); -+} -+ -+static unsigned int delude_tg(struct sk_buff **pskb, -+ const struct net_device *in, const struct net_device *out, -+ unsigned int hooknum, const struct xt_target *target, const void *targinfo) -+{ -+ /* WARNING: This code causes reentry within iptables. -+ This means that the iptables jump stack is now crap. We -+ must return an absolute verdict. --RR */ -+ delude_send_reset(*pskb, hooknum); -+ return NF_DROP; -+} -+ -+static struct xt_target delude_tg_reg = { -+ .name = "DELUDE", -+ .family = AF_INET, -+ .table = "filter", -+ .hooks = (1 << NF_IP_LOCAL_IN) | (1 << NF_IP_FORWARD), -+ .target = delude_tg, -+ .proto = IPPROTO_TCP, -+ .me = THIS_MODULE, -+}; -+ -+static int __init delude_tg_init(void) -+{ -+ return xt_register_target(&delude_tg_reg); -+} -+ -+static void __exit delude_tg_exit(void) -+{ -+ xt_unregister_target(&delude_tg_reg); -+} -+ -+module_init(delude_tg_init); -+module_exit(delude_tg_exit); -+MODULE_AUTHOR("Jan Engelhardt <jengelh@computergmbh.de>"); -+MODULE_DESCRIPTION("netfilter \"DELUDE\" target"); -+MODULE_LICENSE("GPL"); -+MODULE_ALIAS("ipt_DELUDE"); ---- /dev/null -+++ b/net/netfilter/xt_portscan.c -@@ -0,0 +1,269 @@ -+/* -+ * portscan match for netfilter -+ * Copyright © CC Computer Consultants GmbH, 2006 - 2007 -+ * Contact: Jan Engelhardt <jengelh@computergmbh.de> -+ * -+ * This program is free software; you can redistribute it and/or modify -+ * it under the terms of the GNU General Public License; either version -+ * 2 or 3 as published by the Free Software Foundation. -+ */ -+#include <linux/in.h> -+#include <linux/ip.h> -+#include <linux/module.h> -+#include <linux/moduleparam.h> -+#include <linux/skbuff.h> -+#include <linux/stat.h> -+#include <linux/tcp.h> -+#include <linux/types.h> -+#include <linux/version.h> -+#include <linux/netfilter/x_tables.h> -+#include <linux/netfilter/xt_tcpudp.h> -+#include <net/netfilter/nf_nat_rule.h> -+#if defined(_LOCAL) -+# include "xt_portscan.h" -+#elif defined(CONFIG_NETFILTER_XT_MATCH_PORTSCAN) || \ -+ defined(CONFIG_NETFILTER_XT_MATCH_PORTSCAN_MODULE) -+# include <linux/netfilter/xt_portscan.h> -+#else -+# include "xt_portscan.h" -+#endif -+#define PFX KBUILD_MODNAME ": " -+ -+enum { -+ TCP_FLAGS_ALL3 = TCP_FLAG_FIN | TCP_FLAG_RST | TCP_FLAG_SYN, -+ TCP_FLAGS_ALL4 = TCP_FLAGS_ALL3 | TCP_FLAG_ACK, -+ TCP_FLAGS_ALL6 = TCP_FLAGS_ALL4 | TCP_FLAG_PSH | TCP_FLAG_URG, -+}; -+ -+/* Module parameters */ -+static unsigned int -+ connmark_mask = ~0, -+ packet_mask = ~0, -+ mark_seen = 0x9, -+ mark_synrcv = 0x1, -+ mark_closed = 0x2, -+ mark_synscan = 0x3, -+ mark_estab1 = 0x4, -+ mark_estab2 = 0x5, -+ mark_cnscan = 0x6, -+ mark_grscan = 0x7, -+ mark_valid = 0x8; -+ -+module_param(connmark_mask, uint, S_IRUGO | S_IWUSR); -+module_param(packet_mask, uint, S_IRUGO | S_IWUSR); -+module_param(mark_seen, uint, S_IRUGO | S_IWUSR); -+module_param(mark_synrcv, uint, S_IRUGO | S_IWUSR); -+module_param(mark_closed, uint, S_IRUGO | S_IWUSR); -+module_param(mark_synscan, uint, S_IRUGO | S_IWUSR); -+module_param(mark_estab1, uint, S_IRUGO | S_IWUSR); -+module_param(mark_estab2, uint, S_IRUGO | S_IWUSR); -+module_param(mark_cnscan, uint, S_IRUGO | S_IWUSR); -+module_param(mark_grscan, uint, S_IRUGO | S_IWUSR); -+module_param(mark_valid, uint, S_IRUGO | S_IWUSR); -+MODULE_PARM_DESC(connmark_mask, "only set specified bits in connection mark"); -+MODULE_PARM_DESC(packet_mask, "only set specified bits in packet mark"); -+MODULE_PARM_DESC(mark_seen, "nfmark value for packet-seen state"); -+MODULE_PARM_DESC(mark_synrcv, "connmark value for SYN Received state"); -+MODULE_PARM_DESC(mark_closed, "connmark value for closed state"); -+MODULE_PARM_DESC(mark_synscan, "connmark value for SYN Scan state"); -+MODULE_PARM_DESC(mark_estab1, "connmark value for Established-1 state"); -+MODULE_PARM_DESC(mark_estab2, "connmark value for Established-2 state"); -+MODULE_PARM_DESC(mark_cnscan, "connmark value for Connect Scan state"); -+MODULE_PARM_DESC(mark_grscan, "connmark value for Grab Scan state"); -+MODULE_PARM_DESC(mark_valid, "connmark value for Valid state"); -+ -+/* TCP flag functions */ -+static inline bool tflg_ack4(const struct tcphdr *th) -+{ -+ return (tcp_flag_word(th) & TCP_FLAGS_ALL4) == TCP_FLAG_ACK; -+} -+ -+static inline bool tflg_ack6(const struct tcphdr *th) -+{ -+ return (tcp_flag_word(th) & TCP_FLAGS_ALL6) == TCP_FLAG_ACK; -+} -+ -+static inline bool tflg_fin(const struct tcphdr *th) -+{ -+ return (tcp_flag_word(th) & TCP_FLAGS_ALL3) == TCP_FLAG_FIN; -+} -+ -+static inline bool tflg_rst(const struct tcphdr *th) -+{ -+ return (tcp_flag_word(th) & TCP_FLAGS_ALL3) == TCP_FLAG_RST; -+} -+ -+static inline bool tflg_rstack(const struct tcphdr *th) -+{ -+ return (tcp_flag_word(th) & TCP_FLAGS_ALL4) == -+ (TCP_FLAG_ACK | TCP_FLAG_RST); -+} -+ -+static inline bool tflg_syn(const struct tcphdr *th) -+{ -+ return (tcp_flag_word(th) & TCP_FLAGS_ALL4) == TCP_FLAG_SYN; -+} -+ -+static inline bool tflg_synack(const struct tcphdr *th) -+{ -+ return (tcp_flag_word(th) & TCP_FLAGS_ALL4) == -+ (TCP_FLAG_SYN | TCP_FLAG_ACK); -+} -+ -+/* portscan functions */ -+static inline bool portscan_mt_stealth(const struct tcphdr *th) -+{ -+ /* -+ * "Connection refused" replies to our own probes must not be matched. -+ */ -+ if (tflg_rstack(th)) -+ return false; -+ -+ if (tflg_rst(th) && printk_ratelimit()) { -+ printk(KERN_WARNING PFX "Warning: Pure RST received\n"); -+ return false; -+ } -+ -+ /* -+ * -p tcp ! --syn -m conntrack --ctstate INVALID: Looking for non-start -+ * packets that are not associated with any connection -- this will -+ * match most scan types (NULL, XMAS, FIN) and ridiculous flag -+ * combinations (SYN-RST, SYN-FIN, SYN-FIN-RST, FIN-RST, etc.). -+ */ -+ return !tflg_syn(th); -+} -+ -+static inline unsigned int portscan_mt_full(int mark, -+ enum ip_conntrack_info ctstate, bool loopback, const struct tcphdr *tcph, -+ unsigned int payload_len) -+{ -+ if (mark == mark_estab2) { -+ /* -+ * -m connmark --mark $ESTAB2 -+ */ -+ if (tflg_ack4(tcph) && payload_len == 0) -+ return mark; /* keep mark */ -+ else if (tflg_rst(tcph) || tflg_fin(tcph)) -+ return mark_grscan; -+ else -+ return mark_valid; -+ } else if (mark == mark_estab1) { -+ /* -+ * -m connmark --mark $ESTAB1 -+ */ -+ if (tflg_rst(tcph) || tflg_fin(tcph)) -+ return mark_cnscan; -+ else if (!loopback && tflg_ack4(tcph) && payload_len == 0) -+ return mark_estab2; -+ else -+ return mark_valid; -+ } else if (mark == mark_synrcv) { -+ /* -+ * -m connmark --mark $SYN -+ */ -+ if (loopback && tflg_synack(tcph)) -+ return mark; /* keep mark */ -+ else if (loopback && tflg_rstack(tcph)) -+ return mark_closed; -+ else if (tflg_ack6(tcph)) -+ return mark_estab1; -+ else -+ return mark_synscan; -+ } else if (ctstate == IP_CT_NEW && tflg_syn(tcph)) { -+ /* -+ * -p tcp --syn --ctstate NEW -+ */ -+ return mark_synrcv; -+ } -+ return mark; -+} -+ -+static int portscan_mt(const struct sk_buff *skb, -+ const struct net_device *in, const struct net_device *out, -+ const struct xt_match *match, const void *matchinfo, int offset, -+ unsigned int protoff, int *hotdrop) -+{ -+ const struct xt_portscan_match_info *info = matchinfo; -+ enum ip_conntrack_info ctstate; -+ const struct tcphdr *tcph; -+ struct nf_conn *ctdata; -+ struct tcphdr tcph_buf; -+ -+ tcph = skb_header_pointer(skb, protoff, sizeof(tcph_buf), &tcph_buf); -+ if (tcph == NULL) -+ return false; -+ -+ /* Check for invalid packets: -m conntrack --ctstate INVALID */ -+ if ((ctdata = nf_ct_get(skb, &ctstate)) == NULL) { -+ if (info->match_stealth) -+ return portscan_mt_stealth(tcph); -+ /* -+ * If @ctdata is NULL, we cannot match the other scan -+ * types, return. -+ */ -+ return false; -+ } -+ -+ /* -+ * If -m portscan was previously applied to this packet, the rules we -+ * simulate must not be run through again. And for speedup, do not call -+ * it either when the connection is already VALID. -+ */ -+ if ((ctdata->mark & connmark_mask) == mark_valid || -+ (skb->mark & packet_mask) != mark_seen) { -+ unsigned int n; -+ -+ n = portscan_mt_full(ctdata->mark & connmark_mask, ctstate, -+ in == &loopback_dev, tcph, -+ skb->len - protoff - 4 * tcph->doff); -+ -+ ctdata->mark = (ctdata->mark & ~connmark_mask) | n; -+ ((struct sk_buff *)skb)->mark = -+ (skb->mark & ~packet_mask) ^ mark_seen; -+ } -+ -+ return (info->match_syn && ctdata->mark == mark_synscan) || -+ (info->match_cn && ctdata->mark == mark_cnscan) || -+ (info->match_gr && ctdata->mark == mark_grscan); -+} -+ -+static int portscan_mt_check(const char *tablename, const void *entry, -+ const struct xt_match *match, void *matchinfo, unsigned int hook_mask) -+{ -+ const struct xt_portscan_match_info *info = matchinfo; -+ -+ if ((info->match_stealth & ~1) || (info->match_syn & ~1) || -+ (info->match_cn & ~1) || (info->match_gr & ~1)) { -+ printk(KERN_WARNING PFX "Invalid flags\n"); -+ return false; -+ } -+ return true; -+} -+ -+static struct xt_match portscan_mt_reg __read_mostly = { -+ .name = "portscan", -+ .family = AF_INET, -+ .match = portscan_mt, -+ .checkentry = portscan_mt_check, -+ .matchsize = sizeof(struct xt_portscan_match_info), -+ .proto = IPPROTO_TCP, -+ .me = THIS_MODULE, -+}; -+ -+static int __init portscan_mt_init(void) -+{ -+ return xt_register_match(&portscan_mt_reg); -+} -+ -+static void __exit portscan_mt_exit(void) -+{ -+ xt_unregister_match(&portscan_mt_reg); -+ return; -+} -+ -+module_init(portscan_mt_init); -+module_exit(portscan_mt_exit); -+MODULE_AUTHOR("Jan Engelhardt <jengelh@computergmbh.de>"); -+MODULE_DESCRIPTION("netfilter \"portscan\" match"); -+MODULE_LICENSE("GPL"); -+MODULE_ALIAS("ipt_portscan"); ---- a/drivers/char/random.c -+++ b/drivers/char/random.c -@@ -1564,6 +1564,8 @@ - return seq; - } - -+EXPORT_SYMBOL(secure_tcp_sequence_number); -+ - /* Generate secure starting point for ephemeral IPV4 transport port search */ - u32 secure_ipv4_port_ephemeral(__be32 saddr, __be32 daddr, __be16 dport) - { diff --git a/target/linux/generic-2.6/patches-2.6.22/171-netfilter_tarpit.patch b/target/linux/generic-2.6/patches-2.6.22/171-netfilter_tarpit.patch deleted file mode 100644 index 2261efa..0000000 --- a/target/linux/generic-2.6/patches-2.6.22/171-netfilter_tarpit.patch +++ /dev/null @@ -1,319 +0,0 @@ ---- a/net/netfilter/Kconfig -+++ b/net/netfilter/Kconfig -@@ -379,6 +379,23 @@ - - To compile it as a module, choose M here. If unsure, say N. - -+config NETFILTER_XT_TARGET_TARPIT -+ tristate '"TARPIT" target support' -+ depends on NETFILTER_XTABLES -+ ---help--- -+ Adds a TARPIT target to iptables, which captures and holds -+ incoming TCP connections using no local per-connection resources. -+ Connections are accepted, but immediately switched to the persist -+ state (0 byte window), in which the remote side stops sending data -+ and asks to continue every 60-240 seconds. Attempts to close the -+ connection are ignored, forcing the remote side to time out the -+ connection in 12-24 minutes. -+ -+ This offers similar functionality to LaBrea -+ <http://www.hackbusters.net/LaBrea/>, but does not require dedicated -+ hardware or IPs. Any TCP port that you would normally DROP or REJECT -+ can instead become a tarpit. -+ - config NETFILTER_XT_TARGET_TCPMSS - tristate '"TCPMSS" target support' - depends on NETFILTER_XTABLES && (IPV6 || IPV6=n) ---- a/net/netfilter/Makefile -+++ b/net/netfilter/Makefile -@@ -45,6 +45,7 @@ - obj-$(CONFIG_NETFILTER_XT_TARGET_NFLOG) += xt_NFLOG.o - obj-$(CONFIG_NETFILTER_XT_TARGET_NOTRACK) += xt_NOTRACK.o - obj-$(CONFIG_NETFILTER_XT_TARGET_SECMARK) += xt_SECMARK.o -+obj-$(CONFIG_NETFILTER_XT_TARGET_TARPIT) += xt_TARPIT.o - obj-$(CONFIG_NETFILTER_XT_TARGET_TCPMSS) += xt_TCPMSS.o - obj-$(CONFIG_NETFILTER_XT_TARGET_CONNSECMARK) += xt_CONNSECMARK.o - obj-$(CONFIG_NETFILTER_XT_TARGET_CHAOS) += xt_CHAOS.o ---- /dev/null -+++ b/net/netfilter/xt_TARPIT.c -@@ -0,0 +1,280 @@ -+/* -+ * Kernel module to capture and hold incoming TCP connections using -+ * no local per-connection resources. -+ * -+ * Based on ipt_REJECT.c and offering functionality similar to -+ * LaBrea <http://www.hackbusters.net/LaBrea/>. -+ * -+ * Copyright (c) 2002 Aaron Hopkins <tools@die.net> -+ * -+ * This program is free software; you can redistribute it and/or modify -+ * it under the terms of the GNU General Public License as published by -+ * the Free Software Foundation; either version 2 of the License, or -+ * (at your option) any later version. -+ * -+ * This program is distributed in the hope that it will be useful, -+ * but WITHOUT ANY WARRANTY; without even the implied warranty of -+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the -+ * GNU General Public License for more details. -+ * -+ * You should have received a copy of the GNU General Public License -+ * along with this program; if not, write to the Free Software -+ * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. -+ * -+ * Goal: -+ * - Allow incoming TCP connections to be established. -+ * - Passing data should result in the connection being switched to the -+ * persist state (0 byte window), in which the remote side stops sending -+ * data and asks to continue every 60 seconds. -+ * - Attempts to shut down the connection should be ignored completely, so -+ * the remote side ends up having to time it out. -+ * -+ * This means: -+ * - Reply to TCP SYN,!ACK,!RST,!FIN with SYN-ACK, window 5 bytes -+ * - Reply to TCP SYN,ACK,!RST,!FIN with RST to prevent spoofing -+ * - Reply to TCP !SYN,!RST,!FIN with ACK, window 0 bytes, rate-limited -+ */ -+ -+#include <linux/version.h> -+#include <linux/module.h> -+#include <linux/skbuff.h> -+#include <linux/ip.h> -+#include <net/ip.h> -+#include <net/tcp.h> -+#include <net/icmp.h> -+struct in_device; -+#include <net/route.h> -+#include <linux/random.h> -+#include <linux/netfilter_ipv4/ip_tables.h> -+ -+#if 0 -+#define DEBUGP printk -+#else -+#define DEBUGP(format, args...) -+#endif -+ -+/* Stolen from ip_finish_output2 */ -+static int ip_direct_send(struct sk_buff *skb) -+{ -+ struct dst_entry *dst = skb->dst; -+ -+ if (dst->hh != NULL) -+ return neigh_hh_output(dst->hh, skb); -+ else if (dst->neighbour != NULL) -+ return dst->neighbour->output(skb); -+ -+ if (net_ratelimit()) -+ printk(KERN_DEBUG "TARPIT ip_direct_send: no header cache and no neighbor!\n"); -+ -+ kfree_skb(skb); -+ return -EINVAL; -+} -+ -+ -+/* Send reply */ -+static void tarpit_tcp(const struct sk_buff *oskb, struct rtable *ort, -+ unsigned int local) -+{ -+ struct sk_buff *nskb; -+ struct rtable *nrt; -+ struct tcphdr *otcph, *ntcph; -+ struct flowi fl = {}; -+ unsigned int otcplen; -+ u_int16_t tmp; -+ -+ const struct iphdr *oiph = ip_hdr(oskb); -+ struct iphdr *niph; -+ -+ /* A truncated TCP header is not going to be useful */ -+ if (oskb->len < ip_hdrlen(oskb) + sizeof(struct tcphdr)) -+ return; -+ -+ otcph = (void *)oiph + ip_hdrlen(oskb); -+ otcplen = oskb->len - ip_hdrlen(oskb); -+ -+ /* No replies for RST or FIN */ -+ if (otcph->rst || otcph->fin) -+ return; -+ -+ /* No reply to !SYN,!ACK. Rate-limit replies to !SYN,ACKs */ -+ if (!otcph->syn && (!otcph->ack || !xrlim_allow(&ort->u.dst, 1*HZ))) -+ return; -+ -+ /* Check checksum. */ -+ if (tcp_v4_check(otcplen, oiph->saddr, oiph->daddr, -+ csum_partial((char *)otcph, otcplen, 0)) != 0) -+ return; -+ -+ /* -+ * Copy skb (even if skb is about to be dropped, we cannot just -+ * clone it because there may be other things, such as tcpdump, -+ * interested in it) -+ */ -+ nskb = skb_copy(oskb, GFP_ATOMIC); -+ if (nskb == NULL) -+ return; -+ -+ niph = ip_hdr(nskb); -+ -+ /* This packet will not be the same as the other: clear nf fields */ -+ nf_conntrack_put(nskb->nfct); -+ nskb->nfct = NULL; -+#ifdef CONFIG_NETFILTER_DEBUG -+ nskb->nf_debug = 0; -+#endif -+ -+ ntcph = (void *)niph + ip_hdrlen(nskb); -+ -+ /* Truncate to length (no data) */ -+ ntcph->doff = sizeof(struct tcphdr)/4; -+ skb_trim(nskb, ip_hdrlen(nskb) + sizeof(struct tcphdr)); -+ niph->tot_len = htons(nskb->len); -+ -+ /* Swap source and dest */ -+ niph->daddr = xchg(&niph->saddr, niph->daddr); -+ tmp = ntcph->source; -+ ntcph->source = ntcph->dest; -+ ntcph->dest = tmp; -+ -+ /* Use supplied sequence number or make a new one */ -+ ntcph->seq = otcph->ack ? otcph->ack_seq -+ : htonl(secure_tcp_sequence_number(niph->saddr, -+ niph->daddr, -+ ntcph->source, -+ ntcph->dest)); -+ -+ /* Our SYN-ACKs must have a >0 window */ -+ ntcph->window = (otcph->syn && !otcph->ack) ? htons(5) : 0; -+ -+ ntcph->urg_ptr = 0; -+ -+ /* Reset flags */ -+ ((u_int8_t *)ntcph)[13] = 0; -+ -+ if (otcph->syn && otcph->ack) { -+ ntcph->rst = 1; -+ ntcph->ack_seq = 0; -+ } else { -+ ntcph->syn = otcph->syn; -+ ntcph->ack = 1; -+ ntcph->ack_seq = htonl(ntohl(otcph->seq) + otcph->syn); -+ } -+ -+ /* Adjust TCP checksum */ -+ ntcph->check = 0; -+ ntcph->check = tcp_v4_check(sizeof(struct tcphdr), -+ niph->saddr, -+ niph->daddr, -+ csum_partial((char *)ntcph, -+ sizeof(struct tcphdr), 0)); -+ -+ fl.nl_u.ip4_u.daddr = niph->daddr; -+ fl.nl_u.ip4_u.saddr = local ? niph->saddr : 0; -+ fl.nl_u.ip4_u.tos = RT_TOS(niph->tos) | RTO_CONN; -+ fl.oif = 0; -+ -+ if (ip_route_output_key(&nrt, &fl)) -+ goto free_nskb; -+ -+ dst_release(nskb->dst); -+ nskb->dst = &nrt->u.dst; -+ -+ /* Adjust IP TTL */ -+ niph->ttl = dst_metric(nskb->dst, RTAX_HOPLIMIT); -+ -+ /* Set DF, id = 0 */ -+ niph->frag_off = htons(IP_DF); -+ niph->id = 0; -+ -+ /* Adjust IP checksum */ -+ niph->check = 0; -+ niph->check = ip_fast_csum((unsigned char *)niph, niph->ihl); -+ -+ /* "Never happens" */ -+ if (nskb->len > dst_mtu(nskb->dst)) -+ goto free_nskb; -+ -+ ip_direct_send(nskb); -+ return; -+ -+ free_nskb: -+ kfree_skb(nskb); -+} -+ -+static unsigned int xt_tarpit_target(struct sk_buff **pskb, -+ const struct net_device *in, -+ const struct net_device *out, -+ unsigned int hooknum, -+ const struct xt_target *target, -+ const void *targinfo) -+{ -+ const struct sk_buff *skb = *pskb; -+ const struct iphdr *iph = ip_hdr(skb); -+ struct rtable *rt = (void *)skb->dst; -+ -+ /* Do we have an input route cache entry? */ -+ if (rt == NULL) -+ return NF_DROP; -+ -+ /* No replies to physical multicast/broadcast */ -+ if (skb->pkt_type != PACKET_HOST && skb->pkt_type != PACKET_OTHERHOST) -+ return NF_DROP; -+ -+ /* Now check at the protocol level */ -+ if (rt->rt_flags & (RTCF_BROADCAST | RTCF_MULTICAST)) -+ return NF_DROP; -+ -+ /* -+ * Our naive response construction does not deal with IP -+ * options, and probably should not try. -+ */ -+ if (iph->ihl * 4 != sizeof(struct iphdr)) -+ return NF_DROP; -+ -+ /* We are not interested in fragments */ -+ if (iph->frag_off & htons(IP_OFFSET)) -+ return NF_DROP; -+ -+ tarpit_tcp(skb, rt, hooknum == NF_IP_LOCAL_IN); -+ return NF_DROP; -+} -+ -+static int xt_tarpit_check(const char *tablename, const void *entry, -+ const struct xt_target *target, void *targinfo, -+ unsigned int hook_mask) -+{ -+ bool invalid; -+ -+ if (strcmp(tablename, "raw") == 0 && hook_mask == NF_IP_PRE_ROUTING) -+ return true; -+ if (strcmp(tablename, "filter") != 0) -+ return false; -+ invalid = hook_mask & ~((1 << NF_IP_LOCAL_IN) | (1 << NF_IP_FORWARD)); -+ return !invalid; -+} -+ -+static struct xt_target xt_tarpit_reg = { -+ .name = "TARPIT", -+ .family = AF_INET, -+ .proto = IPPROTO_TCP, -+ .target = xt_tarpit_target, -+ .checkentry = xt_tarpit_check, -+ .me = THIS_MODULE, -+}; -+ -+static int __init xt_tarpit_init(void) -+{ -+ return xt_register_target(&xt_tarpit_reg); -+} -+ -+static void __exit xt_tarpit_exit(void) -+{ -+ xt_unregister_target(&xt_tarpit_reg); -+} -+ -+module_init(xt_tarpit_init); -+module_exit(xt_tarpit_exit); -+MODULE_DESCRIPTION("netfilter xt_TARPIT target module"); -+MODULE_AUTHOR("Jan Engelhardt <jengelh@gmx.de>"); -+MODULE_LICENSE("GPL"); -+MODULE_ALIAS("ipt_TARPIT"); diff --git a/target/linux/generic-2.6/patches-2.6.22/180-netfilter_depends.patch b/target/linux/generic-2.6/patches-2.6.22/180-netfilter_depends.patch deleted file mode 100644 index 68e0251..0000000 --- a/target/linux/generic-2.6/patches-2.6.22/180-netfilter_depends.patch +++ /dev/null @@ -1,20 +0,0 @@ ---- a/net/netfilter/Kconfig -+++ b/net/netfilter/Kconfig -@@ -133,7 +133,7 @@ - - config NF_CONNTRACK_H323 - tristate "H.323 protocol support (EXPERIMENTAL)" -- depends on EXPERIMENTAL && NF_CONNTRACK && (IPV6 || IPV6=n) -+ depends on EXPERIMENTAL && NF_CONNTRACK - help - H.323 is a VoIP signalling protocol from ITU-T. As one of the most - important VoIP protocols, it is widely used by voice hardware and -@@ -398,7 +398,7 @@ - - config NETFILTER_XT_TARGET_TCPMSS - tristate '"TCPMSS" target support' -- depends on NETFILTER_XTABLES && (IPV6 || IPV6=n) -+ depends on NETFILTER_XTABLES - ---help--- - This option adds a `TCPMSS' target, which allows you to alter the - MSS value of TCP SYN packets, to control the maximum size for that diff --git a/target/linux/generic-2.6/patches-2.6.22/190-netfilter_rtsp.patch b/target/linux/generic-2.6/patches-2.6.22/190-netfilter_rtsp.patch deleted file mode 100644 index b54754f..0000000 --- a/target/linux/generic-2.6/patches-2.6.22/190-netfilter_rtsp.patch +++ /dev/null @@ -1,1364 +0,0 @@ ---- /dev/null -+++ b/include/linux/netfilter/nf_conntrack_rtsp.h -@@ -0,0 +1,63 @@ -+/* -+ * RTSP extension for IP connection tracking. -+ * (C) 2003 by Tom Marshall <tmarshall at real.com> -+ * based on ip_conntrack_irc.h -+ * -+ * This program is free software; you can redistribute it and/or -+ * modify it under the terms of the GNU General Public License -+ * as published by the Free Software Foundation; either version -+ * 2 of the License, or (at your option) any later version. -+ */ -+#ifndef _IP_CONNTRACK_RTSP_H -+#define _IP_CONNTRACK_RTSP_H -+ -+//#define IP_NF_RTSP_DEBUG 1 -+#define IP_NF_RTSP_VERSION "0.6.21" -+ -+#ifdef __KERNEL__ -+/* port block types */ -+typedef enum { -+ pb_single, /* client_port=x */ -+ pb_range, /* client_port=x-y */ -+ pb_discon /* client_port=x/y (rtspbis) */ -+} portblock_t; -+ -+/* We record seq number and length of rtsp headers here, all in host order. */ -+ -+/* -+ * This structure is per expected connection. It is a member of struct -+ * ip_conntrack_expect. The TCP SEQ for the conntrack expect is stored -+ * there and we are expected to only store the length of the data which -+ * needs replaced. If a packet contains multiple RTSP messages, we create -+ * one expected connection per message. -+ * -+ * We use these variables to mark the entire header block. This may seem -+ * like overkill, but the nature of RTSP requires it. A header may appear -+ * multiple times in a message. We must treat two Transport headers the -+ * same as one Transport header with two entries. -+ */ -+struct ip_ct_rtsp_expect -+{ -+ u_int32_t len; /* length of header block */ -+ portblock_t pbtype; /* Type of port block that was requested */ -+ u_int16_t loport; /* Port that was requested, low or first */ -+ u_int16_t hiport; /* Port that was requested, high or second */ -+#if 0 -+ uint method; /* RTSP method */ -+ uint cseq; /* CSeq from request */ -+#endif -+}; -+ -+extern unsigned int (*nf_nat_rtsp_hook)(struct sk_buff **pskb, -+ enum ip_conntrack_info ctinfo, -+ unsigned int matchoff, unsigned int matchlen, -+ struct ip_ct_rtsp_expect *prtspexp, -+ struct nf_conntrack_expect *exp); -+ -+extern void (*nf_nat_rtsp_hook_expectfn)(struct nf_conn *ct, struct nf_conntrack_expect *exp); -+ -+#define RTSP_PORT 554 -+ -+#endif /* __KERNEL__ */ -+ -+#endif /* _IP_CONNTRACK_RTSP_H */ ---- /dev/null -+++ b/include/linux/netfilter_helpers.h -@@ -0,0 +1,133 @@ -+/* -+ * Helpers for netfiler modules. This file provides implementations for basic -+ * functions such as strncasecmp(), etc. -+ * -+ * gcc will warn for defined but unused functions, so we only include the -+ * functions requested. The following macros are used: -+ * NF_NEED_STRNCASECMP nf_strncasecmp() -+ * NF_NEED_STRTOU16 nf_strtou16() -+ * NF_NEED_STRTOU32 nf_strtou32() -+ */ -+#ifndef _NETFILTER_HELPERS_H -+#define _NETFILTER_HELPERS_H -+ -+/* Only include these functions for kernel code. */ -+#ifdef __KERNEL__ -+ -+#include <linux/ctype.h> -+#define iseol(c) ( (c) == '\r' || (c) == '\n' ) -+ -+/* -+ * The standard strncasecmp() -+ */ -+#ifdef NF_NEED_STRNCASECMP -+static int -+nf_strncasecmp(const char* s1, const char* s2, u_int32_t len) -+{ -+ if (s1 == NULL || s2 == NULL) -+ { -+ if (s1 == NULL && s2 == NULL) -+ { -+ return 0; -+ } -+ return (s1 == NULL) ? -1 : 1; -+ } -+ while (len > 0 && tolower(*s1) == tolower(*s2)) -+ { -+ len--; -+ s1++; -+ s2++; -+ } -+ return ( (len == 0) ? 0 : (tolower(*s1) - tolower(*s2)) ); -+} -+#endif /* NF_NEED_STRNCASECMP */ -+ -+/* -+ * Parse a string containing a 16-bit unsigned integer. -+ * Returns the number of chars used, or zero if no number is found. -+ */ -+#ifdef NF_NEED_STRTOU16 -+static int -+nf_strtou16(const char* pbuf, u_int16_t* pval) -+{ -+ int n = 0; -+ -+ *pval = 0; -+ while (isdigit(pbuf[n])) -+ { -+ *pval = (*pval * 10) + (pbuf[n] - '0'); -+ n++; -+ } -+ -+ return n; -+} -+#endif /* NF_NEED_STRTOU16 */ -+ -+/* -+ * Parse a string containing a 32-bit unsigned integer. -+ * Returns the number of chars used, or zero if no number is found. -+ */ -+#ifdef NF_NEED_STRTOU32 -+static int -+nf_strtou32(const char* pbuf, u_int32_t* pval) -+{ -+ int n = 0; -+ -+ *pval = 0; -+ while (pbuf[n] >= '0' && pbuf[n] <= '9') -+ { -+ *pval = (*pval * 10) + (pbuf[n] - '0'); -+ n++; -+ } -+ -+ return n; -+} -+#endif /* NF_NEED_STRTOU32 */ -+ -+/* -+ * Given a buffer and length, advance to the next line and mark the current -+ * line. -+ */ -+#ifdef NF_NEED_NEXTLINE -+static int -+nf_nextline(char* p, uint len, uint* poff, uint* plineoff, uint* plinelen) -+{ -+ uint off = *poff; -+ uint physlen = 0; -+ -+ if (off >= len) -+ { -+ return 0; -+ } -+ -+ while (p[off] != '\n') -+ { -+ if (len-off <= 1) -+ { -+ return 0; -+ } -+ -+ physlen++; -+ off++; -+ } -+ -+ /* if we saw a crlf, physlen needs adjusted */ -+ if (physlen > 0 && p[off] == '\n' && p[off-1] == '\r') -+ { -+ physlen--; -+ } -+ -+ /* advance past the newline */ -+ off++; -+ -+ *plineoff = *poff; -+ *plinelen = physlen; -+ *poff = off; -+ -+ return 1; -+} -+#endif /* NF_NEED_NEXTLINE */ -+ -+#endif /* __KERNEL__ */ -+ -+#endif /* _NETFILTER_HELPERS_H */ ---- /dev/null -+++ b/include/linux/netfilter_mime.h -@@ -0,0 +1,89 @@ -+/* -+ * MIME functions for netfilter modules. This file provides implementations -+ * for basic MIME parsing. MIME headers are used in many protocols, such as -+ * HTTP, RTSP, SIP, etc. -+ * -+ * gcc will warn for defined but unused functions, so we only include the -+ * functions requested. The following macros are used: -+ * NF_NEED_MIME_NEXTLINE nf_mime_nextline() -+ */ -+#ifndef _NETFILTER_MIME_H -+#define _NETFILTER_MIME_H -+ -+/* Only include these functions for kernel code. */ -+#ifdef __KERNEL__ -+ -+#include <linux/ctype.h> -+ -+/* -+ * Given a buffer and length, advance to the next line and mark the current -+ * line. If the current line is empty, *plinelen will be set to zero. If -+ * not, it will be set to the actual line length (including CRLF). -+ * -+ * 'line' in this context means logical line (includes LWS continuations). -+ * Returns 1 on success, 0 on failure. -+ */ -+#ifdef NF_NEED_MIME_NEXTLINE -+static int -+nf_mime_nextline(char* p, uint len, uint* poff, uint* plineoff, uint* plinelen) -+{ -+ uint off = *poff; -+ uint physlen = 0; -+ int is_first_line = 1; -+ -+ if (off >= len) -+ { -+ return 0; -+ } -+ -+ do -+ { -+ while (p[off] != '\n') -+ { -+ if (len-off <= 1) -+ { -+ return 0; -+ } -+ -+ physlen++; -+ off++; -+ } -+ -+ /* if we saw a crlf, physlen needs adjusted */ -+ if (physlen > 0 && p[off] == '\n' && p[off-1] == '\r') -+ { -+ physlen--; -+ } -+ -+ /* advance past the newline */ -+ off++; -+ -+ /* check for an empty line */ -+ if (physlen == 0) -+ { -+ break; -+ } -+ -+ /* check for colon on the first physical line */ -+ if (is_first_line) -+ { -+ is_first_line = 0; -+ if (memchr(p+(*poff), ':', physlen) == NULL) -+ { -+ return 0; -+ } -+ } -+ } -+ while (p[off] == ' ' || p[off] == '\t'); -+ -+ *plineoff = *poff; -+ *plinelen = (physlen == 0) ? 0 : (off - *poff); -+ *poff = off; -+ -+ return 1; -+} -+#endif /* NF_NEED_MIME_NEXTLINE */ -+ -+#endif /* __KERNEL__ */ -+ -+#endif /* _NETFILTER_MIME_H */ ---- a/net/ipv4/netfilter/Makefile -+++ b/net/ipv4/netfilter/Makefile -@@ -23,6 +23,7 @@ - obj-$(CONFIG_NF_NAT_FTP) += nf_nat_ftp.o - obj-$(CONFIG_NF_NAT_H323) += nf_nat_h323.o - obj-$(CONFIG_NF_NAT_IRC) += nf_nat_irc.o -+obj-$(CONFIG_NF_NAT_RTSP) += nf_nat_rtsp.o - obj-$(CONFIG_NF_NAT_PPTP) += nf_nat_pptp.o - obj-$(CONFIG_NF_NAT_SIP) += nf_nat_sip.o - obj-$(CONFIG_NF_NAT_SNMP_BASIC) += nf_nat_snmp_basic.o ---- a/net/netfilter/Kconfig -+++ b/net/netfilter/Kconfig -@@ -239,6 +239,16 @@ - - To compile it as a module, choose M here. If unsure, say N. - -+config NF_CONNTRACK_RTSP -+ tristate "RTSP protocol support" -+ depends on NF_CONNTRACK -+ help -+ Support the RTSP protocol. This allows UDP transports to be setup -+ properly, including RTP and RDT. -+ -+ If you want to compile it as a module, say 'M' here and read -+ Documentation/modules.txt. If unsure, say 'Y'. -+ - config NF_CT_NETLINK - tristate 'Connection tracking netlink interface (EXPERIMENTAL)' - depends on EXPERIMENTAL && NF_CONNTRACK && NETFILTER_NETLINK ---- a/net/netfilter/Makefile -+++ b/net/netfilter/Makefile -@@ -32,6 +32,7 @@ - obj-$(CONFIG_NF_CONNTRACK_SANE) += nf_conntrack_sane.o - obj-$(CONFIG_NF_CONNTRACK_SIP) += nf_conntrack_sip.o - obj-$(CONFIG_NF_CONNTRACK_TFTP) += nf_conntrack_tftp.o -+obj-$(CONFIG_NF_CONNTRACK_RTSP) += nf_conntrack_rtsp.o - - # generic X tables - obj-$(CONFIG_NETFILTER_XTABLES) += x_tables.o xt_tcpudp.o ---- a/net/ipv4/netfilter/Kconfig -+++ b/net/ipv4/netfilter/Kconfig -@@ -296,6 +296,11 @@ - depends on IP_NF_IPTABLES && NF_CONNTRACK && NF_NAT - default NF_NAT && NF_CONNTRACK_IRC - -+config NF_NAT_RTSP -+ tristate -+ depends on IP_NF_IPTABLES && NF_CONNTRACK && NF_NAT -+ default NF_NAT && NF_CONNTRACK_RTSP -+ - config NF_NAT_TFTP - tristate - depends on IP_NF_IPTABLES && NF_CONNTRACK && NF_NAT ---- /dev/null -+++ b/net/netfilter/nf_conntrack_rtsp.c -@@ -0,0 +1,515 @@ -+/* -+ * RTSP extension for IP connection tracking -+ * (C) 2003 by Tom Marshall <tmarshall at real.com> -+ * based on ip_conntrack_irc.c -+ * -+ * This program is free software; you can redistribute it and/or -+ * modify it under the terms of the GNU General Public License -+ * as published by the Free Software Foundation; either version -+ * 2 of the License, or (at your option) any later version. -+ * -+ * Module load syntax: -+ * insmod nf_conntrack_rtsp.o ports=port1,port2,...port<MAX_PORTS> -+ * max_outstanding=n setup_timeout=secs -+ * -+ * If no ports are specified, the default will be port 554. -+ * -+ * With max_outstanding you can define the maximum number of not yet -+ * answered SETUP requests per RTSP session (default 8). -+ * With setup_timeout you can specify how long the system waits for -+ * an expected data channel (default 300 seconds). -+ * -+ * 2005-02-13: Harald Welte <laforge at netfilter.org> -+ * - port to 2.6 -+ * - update to recent post-2.6.11 api changes -+ * 2006-09-14: Steven Van Acker <deepstar at singularity.be> -+ * - removed calls to NAT code from conntrack helper: NAT no longer needed to use rtsp-conntrack -+ * 2007-04-18: Michael Guntsche <mike at it-loops.com> -+ * - Port to new NF API -+ */ -+ -+#include <linux/module.h> -+#include <linux/netfilter.h> -+#include <linux/ip.h> -+#include <linux/inet.h> -+#include <net/tcp.h> -+ -+#include <net/netfilter/nf_conntrack.h> -+#include <net/netfilter/nf_conntrack_expect.h> -+#include <net/netfilter/nf_conntrack_helper.h> -+#include <linux/netfilter/nf_conntrack_rtsp.h> -+ -+#define NF_NEED_STRNCASECMP -+#define NF_NEED_STRTOU16 -+#define NF_NEED_STRTOU32 -+#define NF_NEED_NEXTLINE -+#include <linux/netfilter_helpers.h> -+#define NF_NEED_MIME_NEXTLINE -+#include <linux/netfilter_mime.h> -+ -+#include <linux/ctype.h> -+#define MAX_SIMUL_SETUP 8 /* XXX: use max_outstanding */ -+#define INFOP(fmt, args...) printk(KERN_INFO "%s: %s: " fmt, __FILE__, __FUNCTION__ , ## args) -+#if 0 -+#define DEBUGP(fmt, args...) printk(KERN_DEBUG "%s: %s: " fmt, __FILE__, __FUNCTION__ , ## args) -+#else -+#define DEBUGP(fmt, args...) -+#endif -+ -+#define MAX_PORTS 8 -+static int ports[MAX_PORTS]; -+static int num_ports = 0; -+static int max_outstanding = 8; -+static unsigned int setup_timeout = 300; -+ -+MODULE_AUTHOR("Tom Marshall <tmarshall at real.com>"); -+MODULE_DESCRIPTION("RTSP connection tracking module"); -+MODULE_LICENSE("GPL"); -+module_param_array(ports, int, &num_ports, 0400); -+MODULE_PARM_DESC(ports, "port numbers of RTSP servers"); -+module_param(max_outstanding, int, 0400); -+MODULE_PARM_DESC(max_outstanding, "max number of outstanding SETUP requests per RTSP session"); -+module_param(setup_timeout, int, 0400); -+MODULE_PARM_DESC(setup_timeout, "timeout on for unestablished data channels"); -+ -+static char *rtsp_buffer; -+static DEFINE_SPINLOCK(rtsp_buffer_lock); -+ -+unsigned int (*nf_nat_rtsp_hook)(struct sk_buff **pskb, -+ enum ip_conntrack_info ctinfo, -+ unsigned int matchoff, unsigned int matchlen,struct ip_ct_rtsp_expect* prtspexp, -+ struct nf_conntrack_expect *exp); -+void (*nf_nat_rtsp_hook_expectfn)(struct nf_conn *ct, struct nf_conntrack_expect *exp); -+ -+EXPORT_SYMBOL_GPL(nf_nat_rtsp_hook); -+ -+/* -+ * Max mappings we will allow for one RTSP connection (for RTP, the number -+ * of allocated ports is twice this value). Note that SMIL burns a lot of -+ * ports so keep this reasonably high. If this is too low, you will see a -+ * lot of "no free client map entries" messages. -+ */ -+#define MAX_PORT_MAPS 16 -+ -+/*** default port list was here in the masq code: 554, 3030, 4040 ***/ -+ -+#define SKIP_WSPACE(ptr,len,off) while(off < len && isspace(*(ptr+off))) { off++; } -+ -+/* -+ * Parse an RTSP packet. -+ * -+ * Returns zero if parsing failed. -+ * -+ * Parameters: -+ * IN ptcp tcp data pointer -+ * IN tcplen tcp data len -+ * IN/OUT ptcpoff points to current tcp offset -+ * OUT phdrsoff set to offset of rtsp headers -+ * OUT phdrslen set to length of rtsp headers -+ * OUT pcseqoff set to offset of CSeq header -+ * OUT pcseqlen set to length of CSeq header -+ */ -+static int -+rtsp_parse_message(char* ptcp, uint tcplen, uint* ptcpoff, -+ uint* phdrsoff, uint* phdrslen, -+ uint* pcseqoff, uint* pcseqlen, -+ uint* transoff, uint* translen) -+{ -+ uint entitylen = 0; -+ uint lineoff; -+ uint linelen; -+ -+ if (!nf_nextline(ptcp, tcplen, ptcpoff, &lineoff, &linelen)) -+ return 0; -+ -+ *phdrsoff = *ptcpoff; -+ while (nf_mime_nextline(ptcp, tcplen, ptcpoff, &lineoff, &linelen)) { -+ if (linelen == 0) { -+ if (entitylen > 0) -+ *ptcpoff += min(entitylen, tcplen - *ptcpoff); -+ break; -+ } -+ if (lineoff+linelen > tcplen) { -+ INFOP("!! overrun !!\n"); -+ break; -+ } -+ -+ if (nf_strncasecmp(ptcp+lineoff, "CSeq:", 5) == 0) { -+ *pcseqoff = lineoff; -+ *pcseqlen = linelen; -+ } -+ -+ if (nf_strncasecmp(ptcp+lineoff, "Transport:", 10) == 0) { -+ *transoff = lineoff; -+ *translen = linelen; -+ } -+ -+ if (nf_strncasecmp(ptcp+lineoff, "Content-Length:", 15) == 0) { -+ uint off = lineoff+15; -+ SKIP_WSPACE(ptcp+lineoff, linelen, off); -+ nf_strtou32(ptcp+off, &entitylen); -+ } -+ } -+ *phdrslen = (*ptcpoff) - (*phdrsoff); -+ -+ return 1; -+} -+ -+/* -+ * Find lo/hi client ports (if any) in transport header -+ * In: -+ * ptcp, tcplen = packet -+ * tranoff, tranlen = buffer to search -+ * -+ * Out: -+ * pport_lo, pport_hi = lo/hi ports (host endian) -+ * -+ * Returns nonzero if any client ports found -+ * -+ * Note: it is valid (and expected) for the client to request multiple -+ * transports, so we need to parse the entire line. -+ */ -+static int -+rtsp_parse_transport(char* ptran, uint tranlen, -+ struct ip_ct_rtsp_expect* prtspexp) -+{ -+ int rc = 0; -+ uint off = 0; -+ -+ if (tranlen < 10 || !iseol(ptran[tranlen-1]) || -+ nf_strncasecmp(ptran, "Transport:", 10) != 0) { -+ INFOP("sanity check failed\n"); -+ return 0; -+ } -+ -+ DEBUGP("tran='%.*s'\n", (int)tranlen, ptran); -+ off += 10; -+ SKIP_WSPACE(ptran, tranlen, off); -+ -+ /* Transport: tran;field;field=val,tran;field;field=val,... */ -+ while (off < tranlen) { -+ const char* pparamend; -+ uint nextparamoff; -+ -+ pparamend = memchr(ptran+off, ',', tranlen-off); -+ pparamend = (pparamend == NULL) ? ptran+tranlen : pparamend+1; -+ nextparamoff = pparamend-ptran; -+ -+ while (off < nextparamoff) { -+ const char* pfieldend; -+ uint nextfieldoff; -+ -+ pfieldend = memchr(ptran+off, ';', nextparamoff-off); -+ nextfieldoff = (pfieldend == NULL) ? nextparamoff : pfieldend-ptran+1; -+ -+ if (strncmp(ptran+off, "client_port=", 12) == 0) { -+ u_int16_t port; -+ uint numlen; -+ -+ off += 12; -+ numlen = nf_strtou16(ptran+off, &port); -+ off += numlen; -+ if (prtspexp->loport != 0 && prtspexp->loport != port) -+ DEBUGP("multiple ports found, port %hu ignored\n", port); -+ else { -+ DEBUGP("lo port found : %hu\n", port); -+ prtspexp->loport = prtspexp->hiport = port; -+ if (ptran[off] == '-') { -+ off++; -+ numlen = nf_strtou16(ptran+off, &port); -+ off += numlen; -+ prtspexp->pbtype = pb_range; -+ prtspexp->hiport = port; -+ -+ // If we have a range, assume rtp: -+ // loport must be even, hiport must be loport+1 -+ if ((prtspexp->loport & 0x0001) != 0 || -+ prtspexp->hiport != prtspexp->loport+1) { -+ DEBUGP("incorrect range: %hu-%hu, correcting\n", -+ prtspexp->loport, prtspexp->hiport); -+ prtspexp->loport &= 0xfffe; -+ prtspexp->hiport = prtspexp->loport+1; -+ } -+ } else if (ptran[off] == '/') { -+ off++; -+ numlen = nf_strtou16(ptran+off, &port); -+ off += numlen; -+ prtspexp->pbtype = pb_discon; -+ prtspexp->hiport = port; -+ } -+ rc = 1; -+ } -+ } -+ -+ /* -+ * Note we don't look for the destination parameter here. -+ * If we are using NAT, the NAT module will handle it. If not, -+ * and the client is sending packets elsewhere, the expectation -+ * will quietly time out. -+ */ -+ -+ off = nextfieldoff; -+ } -+ -+ off = nextparamoff; -+ } -+ -+ return rc; -+} -+ -+void expected(struct nf_conn *ct, struct nf_conntrack_expect *exp) -+{ -+ if(nf_nat_rtsp_hook_expectfn) { -+ nf_nat_rtsp_hook_expectfn(ct,exp); -+ } -+} -+ -+/*** conntrack functions ***/ -+ -+/* outbound packet: client->server */ -+ -+static inline int -+help_out(struct sk_buff **pskb, unsigned char *rb_ptr, unsigned int datalen, -+ struct nf_conn *ct, enum ip_conntrack_info ctinfo) -+{ -+ struct ip_ct_rtsp_expect expinfo; -+ -+ int dir = CTINFO2DIR(ctinfo); /* = IP_CT_DIR_ORIGINAL */ -+ //struct tcphdr* tcph = (void*)iph + iph->ihl * 4; -+ //uint tcplen = pktlen - iph->ihl * 4; -+ char* pdata = rb_ptr; -+ //uint datalen = tcplen - tcph->doff * 4; -+ uint dataoff = 0; -+ int ret = NF_ACCEPT; -+ -+ struct nf_conntrack_expect *exp; -+ -+ __be16 be_loport; -+ -+ memset(&expinfo, 0, sizeof(expinfo)); -+ -+ while (dataoff < datalen) { -+ uint cmdoff = dataoff; -+ uint hdrsoff = 0; -+ uint hdrslen = 0; -+ uint cseqoff = 0; -+ uint cseqlen = 0; -+ uint transoff = 0; -+ uint translen = 0; -+ uint off; -+ -+ if (!rtsp_parse_message(pdata, datalen, &dataoff, -+ &hdrsoff, &hdrslen, -+ &cseqoff, &cseqlen, -+ &transoff, &translen)) -+ break; /* not a valid message */ -+ -+ if (strncmp(pdata+cmdoff, "SETUP ", 6) != 0) -+ continue; /* not a SETUP message */ -+ DEBUGP("found a setup message\n"); -+ -+ off = 0; -+ if(translen) { -+ rtsp_parse_transport(pdata+transoff, translen, &expinfo); -+ } -+ -+ if (expinfo.loport == 0) { -+ DEBUGP("no udp transports found\n"); -+ continue; /* no udp transports found */ -+ } -+ -+ DEBUGP("udp transport found, ports=(%d,%hu,%hu)\n", -+ (int)expinfo.pbtype, expinfo.loport, expinfo.hiport); -+ -+ exp = nf_conntrack_expect_alloc(ct); -+ if (!exp) { -+ ret = NF_DROP; -+ goto out; -+ } -+ -+ be_loport = htons(expinfo.loport); -+ -+ nf_conntrack_expect_init(exp, ct->tuplehash[!dir].tuple.src.l3num, -+ &ct->tuplehash[!dir].tuple.src.u3, &ct->tuplehash[!dir].tuple.dst.u3, -+ IPPROTO_UDP, NULL, &be_loport); -+ -+ exp->master = ct; -+ -+ exp->expectfn = expected; -+ exp->flags = 0; -+ -+ if (expinfo.pbtype == pb_range) { -+ DEBUGP("Changing expectation mask to handle multiple ports\n"); -+ exp->mask.dst.u.udp.port = 0xfffe; -+ } -+ -+ DEBUGP("expect_related %u.%u.%u.%u:%u-%u.%u.%u.%u:%u\n", -+ NIPQUAD(exp->tuple.src.u3.ip), -+ ntohs(exp->tuple.src.u.udp.port), -+ NIPQUAD(exp->tuple.dst.u3.ip), -+ ntohs(exp->tuple.dst.u.udp.port)); -+ -+ if (nf_nat_rtsp_hook) -+ /* pass the request off to the nat helper */ -+ ret = nf_nat_rtsp_hook(pskb, ctinfo, hdrsoff, hdrslen, &expinfo, exp); -+ else if (nf_conntrack_expect_related(exp) != 0) { -+ INFOP("nf_conntrack_expect_related failed\n"); -+ ret = NF_DROP; -+ } -+ nf_conntrack_expect_put(exp); -+ goto out; -+ } -+out: -+ -+ return ret; -+} -+ -+ -+static inline int -+help_in(struct sk_buff **pskb, size_t pktlen, -+ struct nf_conn* ct, enum ip_conntrack_info ctinfo) -+{ -+ return NF_ACCEPT; -+} -+ -+static int help(struct sk_buff **pskb, unsigned int protoff, -+ struct nf_conn *ct, enum ip_conntrack_info ctinfo) -+{ -+ struct tcphdr _tcph, *th; -+ unsigned int dataoff, datalen; -+ char *rb_ptr; -+ int ret = NF_DROP; -+ -+ /* Until there's been traffic both ways, don't look in packets. */ -+ if (ctinfo != IP_CT_ESTABLISHED && -+ ctinfo != IP_CT_ESTABLISHED + IP_CT_IS_REPLY) { -+ DEBUGP("conntrackinfo = %u\n", ctinfo); -+ return NF_ACCEPT; -+ } -+ -+ /* Not whole TCP header? */ -+ th = skb_header_pointer(*pskb,protoff, sizeof(_tcph), &_tcph); -+ -+ if (!th) -+ return NF_ACCEPT; -+ -+ /* No data ? */ -+ dataoff = protoff + th->doff*4; -+ datalen = (*pskb)->len - dataoff; -+ if (dataoff >= (*pskb)->len) -+ return NF_ACCEPT; -+ -+ spin_lock_bh(&rtsp_buffer_lock); -+ rb_ptr = skb_header_pointer(*pskb, dataoff, -+ (*pskb)->len - dataoff, rtsp_buffer); -+ BUG_ON(rb_ptr == NULL); -+ -+#if 0 -+ /* Checksum invalid? Ignore. */ -+ /* FIXME: Source route IP option packets --RR */ -+ if (tcp_v4_check(tcph, tcplen, iph->saddr, iph->daddr, -+ csum_partial((char*)tcph, tcplen, 0))) -+ { -+ DEBUGP("bad csum: %p %u %u.%u.%u.%u %u.%u.%u.%u\n", -+ tcph, tcplen, NIPQUAD(iph->saddr), NIPQUAD(iph->daddr)); -+ return NF_ACCEPT; -+ } -+#endif -+ -+ switch (CTINFO2DIR(ctinfo)) { -+ case IP_CT_DIR_ORIGINAL: -+ ret = help_out(pskb, rb_ptr, datalen, ct, ctinfo); -+ break; -+ case IP_CT_DIR_REPLY: -+ DEBUGP("IP_CT_DIR_REPLY\n"); -+ /* inbound packet: server->client */ -+ ret = NF_ACCEPT; -+ break; -+ } -+ -+ spin_unlock_bh(&rtsp_buffer_lock); -+ -+ return ret; -+} -+ -+static struct nf_conntrack_helper rtsp_helpers[MAX_PORTS]; -+static char rtsp_names[MAX_PORTS][10]; -+ -+/* This function is intentionally _NOT_ defined as __exit */ -+static void -+fini(void) -+{ -+ int i; -+ for (i = 0; i < num_ports; i++) { -+ DEBUGP("unregistering port %d\n", ports[i]); -+ nf_conntrack_helper_unregister(&rtsp_helpers[i]); -+ } -+ kfree(rtsp_buffer); -+} -+ -+static int __init -+init(void) -+{ -+ int i, ret; -+ struct nf_conntrack_helper *hlpr; -+ char *tmpname; -+ -+ printk("nf_conntrack_rtsp v" IP_NF_RTSP_VERSION " loading\n"); -+ -+ if (max_outstanding < 1) { -+ printk("nf_conntrack_rtsp: max_outstanding must be a positive integer\n"); -+ return -EBUSY; -+ } -+ if (setup_timeout < 0) { -+ printk("nf_conntrack_rtsp: setup_timeout must be a positive integer\n"); -+ return -EBUSY; -+ } -+ -+ rtsp_buffer = kmalloc(65536, GFP_KERNEL); -+ if (!rtsp_buffer) -+ return -ENOMEM; -+ -+ /* If no port given, default to standard rtsp port */ -+ if (ports[0] == 0) { -+ ports[0] = RTSP_PORT; -+ } -+ -+ for (i = 0; (i < MAX_PORTS) && ports[i]; i++) { -+ hlpr = &rtsp_helpers[i]; -+ memset(hlpr, 0, sizeof(struct nf_conntrack_helper)); -+ hlpr->tuple.src.u.tcp.port = htons(ports[i]); -+ hlpr->tuple.dst.protonum = IPPROTO_TCP; -+ hlpr->mask.src.u.tcp.port = 0xFFFF; -+ hlpr->mask.dst.protonum = 0xFF; -+ hlpr->max_expected = max_outstanding; -+ hlpr->timeout = setup_timeout; -+ hlpr->me = THIS_MODULE; -+ hlpr->help = help; -+ -+ tmpname = &rtsp_names[i][0]; -+ if (ports[i] == RTSP_PORT) { -+ sprintf(tmpname, "rtsp"); -+ } else { -+ sprintf(tmpname, "rtsp-%d", i); -+ } -+ hlpr->name = tmpname; -+ -+ DEBUGP("port #%d: %d\n", i, ports[i]); -+ -+ ret = nf_conntrack_helper_register(hlpr); -+ -+ if (ret) { -+ printk("nf_conntrack_rtsp: ERROR registering port %d\n", ports[i]); -+ fini(); -+ return -EBUSY; -+ } -+ num_ports++; -+ } -+ return 0; -+} -+ -+module_init(init); -+module_exit(fini); -+ -+EXPORT_SYMBOL(nf_nat_rtsp_hook_expectfn); -+ ---- /dev/null -+++ b/net/ipv4/netfilter/nf_nat_rtsp.c -@@ -0,0 +1,496 @@ -+/* -+ * RTSP extension for TCP NAT alteration -+ * (C) 2003 by Tom Marshall <tmarshall at real.com> -+ * based on ip_nat_irc.c -+ * -+ * This program is free software; you can redistribute it and/or -+ * modify it under the terms of the GNU General Public License -+ * as published by the Free Software Foundation; either version -+ * 2 of the License, or (at your option) any later version. -+ * -+ * Module load syntax: -+ * insmod nf_nat_rtsp.o ports=port1,port2,...port<MAX_PORTS> -+ * stunaddr=<address> -+ * destaction=[auto|strip|none] -+ * -+ * If no ports are specified, the default will be port 554 only. -+ * -+ * stunaddr specifies the address used to detect that a client is using STUN. -+ * If this address is seen in the destination parameter, it is assumed that -+ * the client has already punched a UDP hole in the firewall, so we don't -+ * mangle the client_port. If none is specified, it is autodetected. It -+ * only needs to be set if you have multiple levels of NAT. It should be -+ * set to the external address that the STUN clients detect. Note that in -+ * this case, it will not be possible for clients to use UDP with servers -+ * between the NATs. -+ * -+ * If no destaction is specified, auto is used. -+ * destaction=auto: strip destination parameter if it is not stunaddr. -+ * destaction=strip: always strip destination parameter (not recommended). -+ * destaction=none: do not touch destination parameter (not recommended). -+ */ -+ -+#include <linux/module.h> -+#include <net/tcp.h> -+#include <net/netfilter/nf_nat_helper.h> -+#include <net/netfilter/nf_nat_rule.h> -+#include <linux/netfilter/nf_conntrack_rtsp.h> -+#include <net/netfilter/nf_conntrack_expect.h> -+ -+#include <linux/inet.h> -+#include <linux/ctype.h> -+#define NF_NEED_STRNCASECMP -+#define NF_NEED_STRTOU16 -+#include <linux/netfilter_helpers.h> -+#define NF_NEED_MIME_NEXTLINE -+#include <linux/netfilter_mime.h> -+ -+#define INFOP(fmt, args...) printk(KERN_INFO "%s: %s: " fmt, __FILE__, __FUNCTION__ , ## args) -+#if 0 -+#define DEBUGP(fmt, args...) printk(KERN_DEBUG "%s: %s: " fmt, __FILE__, __FUNCTION__ , ## args) -+#else -+#define DEBUGP(fmt, args...) -+#endif -+ -+#define MAX_PORTS 8 -+#define DSTACT_AUTO 0 -+#define DSTACT_STRIP 1 -+#define DSTACT_NONE 2 -+ -+static char* stunaddr = NULL; -+static char* destaction = NULL; -+ -+static u_int32_t extip = 0; -+static int dstact = 0; -+ -+MODULE_AUTHOR("Tom Marshall <tmarshall at real.com>"); -+MODULE_DESCRIPTION("RTSP network address translation module"); -+MODULE_LICENSE("GPL"); -+module_param(stunaddr, charp, 0644); -+MODULE_PARM_DESC(stunaddr, "Address for detecting STUN"); -+module_param(destaction, charp, 0644); -+MODULE_PARM_DESC(destaction, "Action for destination parameter (auto/strip/none)"); -+ -+#define SKIP_WSPACE(ptr,len,off) while(off < len && isspace(*(ptr+off))) { off++; } -+ -+/*** helper functions ***/ -+ -+static void -+get_skb_tcpdata(struct sk_buff* skb, char** pptcpdata, uint* ptcpdatalen) -+{ -+ struct iphdr* iph = ip_hdr(skb); -+ struct tcphdr* tcph = (void *)iph + ip_hdrlen(skb); -+ -+ *pptcpdata = (char*)tcph + tcph->doff*4; -+ *ptcpdatalen = ((char*)skb_transport_header(skb) + skb->len) - *pptcpdata; -+} -+ -+/*** nat functions ***/ -+ -+/* -+ * Mangle the "Transport:" header: -+ * - Replace all occurences of "client_port=<spec>" -+ * - Handle destination parameter -+ * -+ * In: -+ * ct, ctinfo = conntrack context -+ * pskb = packet -+ * tranoff = Transport header offset from TCP data -+ * tranlen = Transport header length (incl. CRLF) -+ * rport_lo = replacement low port (host endian) -+ * rport_hi = replacement high port (host endian) -+ * -+ * Returns packet size difference. -+ * -+ * Assumes that a complete transport header is present, ending with CR or LF -+ */ -+static int -+rtsp_mangle_tran(enum ip_conntrack_info ctinfo, -+ struct nf_conntrack_expect* exp, -+ struct ip_ct_rtsp_expect* prtspexp, -+ struct sk_buff** pskb, uint tranoff, uint tranlen) -+{ -+ char* ptcp; -+ uint tcplen; -+ char* ptran; -+ char rbuf1[16]; /* Replacement buffer (one port) */ -+ uint rbuf1len; /* Replacement len (one port) */ -+ char rbufa[16]; /* Replacement buffer (all ports) */ -+ uint rbufalen; /* Replacement len (all ports) */ -+ u_int32_t newip; -+ u_int16_t loport, hiport; -+ uint off = 0; -+ uint diff; /* Number of bytes we removed */ -+ -+ struct nf_conn *ct = exp->master; -+ struct nf_conntrack_tuple *t; -+ -+ char szextaddr[15+1]; -+ uint extaddrlen; -+ int is_stun; -+ -+ get_skb_tcpdata(*pskb, &ptcp, &tcplen); -+ ptran = ptcp+tranoff; -+ -+ if (tranoff+tranlen > tcplen || tcplen-tranoff < tranlen || -+ tranlen < 10 || !iseol(ptran[tranlen-1]) || -+ nf_strncasecmp(ptran, "Transport:", 10) != 0) -+ { -+ INFOP("sanity check failed\n"); -+ return 0; -+ } -+ off += 10; -+ SKIP_WSPACE(ptcp+tranoff, tranlen, off); -+ -+ newip = ct->tuplehash[IP_CT_DIR_REPLY].tuple.dst.u3.ip; -+ t = &exp->tuple; -+ t->dst.u3.ip = newip; -+ -+ extaddrlen = extip ? sprintf(szextaddr, "%u.%u.%u.%u", NIPQUAD(extip)) -+ : sprintf(szextaddr, "%u.%u.%u.%u", NIPQUAD(newip)); -+ DEBUGP("stunaddr=%s (%s)\n", szextaddr, (extip?"forced":"auto")); -+ -+ rbuf1len = rbufalen = 0; -+ switch (prtspexp->pbtype) -+ { -+ case pb_single: -+ for (loport = prtspexp->loport; loport != 0; loport++) /* XXX: improper wrap? */ -+ { -+ t->dst.u.udp.port = htons(loport); -+ if (nf_conntrack_expect_related(exp) == 0) -+ { -+ DEBUGP("using port %hu\n", loport); -+ break; -+ } -+ } -+ if (loport != 0) -+ { -+ rbuf1len = sprintf(rbuf1, "%hu", loport); -+ rbufalen = sprintf(rbufa, "%hu", loport); -+ } -+ break; -+ case pb_range: -+ for (loport = prtspexp->loport; loport != 0; loport += 2) /* XXX: improper wrap? */ -+ { -+ t->dst.u.udp.port = htons(loport); -+ if (nf_conntrack_expect_related(exp) == 0) -+ { -+ hiport = loport + ~exp->mask.dst.u.udp.port; -+ DEBUGP("using ports %hu-%hu\n", loport, hiport); -+ break; -+ } -+ } -+ if (loport != 0) -+ { -+ rbuf1len = sprintf(rbuf1, "%hu", loport); -+ rbufalen = sprintf(rbufa, "%hu-%hu", loport, loport+1); -+ } -+ break; -+ case pb_discon: -+ for (loport = prtspexp->loport; loport != 0; loport++) /* XXX: improper wrap? */ -+ { -+ t->dst.u.udp.port = htons(loport); -+ if (nf_conntrack_expect_related(exp) == 0) -+ { -+ DEBUGP("using port %hu (1 of 2)\n", loport); -+ break; -+ } -+ } -+ for (hiport = prtspexp->hiport; hiport != 0; hiport++) /* XXX: improper wrap? */ -+ { -+ t->dst.u.udp.port = htons(hiport); -+ if (nf_conntrack_expect_related(exp) == 0) -+ { -+ DEBUGP("using port %hu (2 of 2)\n", hiport); -+ break; -+ } -+ } -+ if (loport != 0 && hiport != 0) -+ { -+ rbuf1len = sprintf(rbuf1, "%hu", loport); -+ if (hiport == loport+1) -+ { -+ rbufalen = sprintf(rbufa, "%hu-%hu", loport, hiport); -+ } -+ else -+ { -+ rbufalen = sprintf(rbufa, "%hu/%hu", loport, hiport); -+ } -+ } -+ break; -+ } -+ -+ if (rbuf1len == 0) -+ { -+ return 0; /* cannot get replacement port(s) */ -+ } -+ -+ /* Transport: tran;field;field=val,tran;field;field=val,... */ -+ while (off < tranlen) -+ { -+ uint saveoff; -+ const char* pparamend; -+ uint nextparamoff; -+ -+ pparamend = memchr(ptran+off, ',', tranlen-off); -+ pparamend = (pparamend == NULL) ? ptran+tranlen : pparamend+1; -+ nextparamoff = pparamend-ptcp; -+ -+ /* -+ * We pass over each param twice. On the first pass, we look for a -+ * destination= field. It is handled by the security policy. If it -+ * is present, allowed, and equal to our external address, we assume -+ * that STUN is being used and we leave the client_port= field alone. -+ */ -+ is_stun = 0; -+ saveoff = off; -+ while (off < nextparamoff) -+ { -+ const char* pfieldend; -+ uint nextfieldoff; -+ -+ pfieldend = memchr(ptran+off, ';', nextparamoff-off); -+ nextfieldoff = (pfieldend == NULL) ? nextparamoff : pfieldend-ptran+1; -+ -+ if (dstact != DSTACT_NONE && strncmp(ptran+off, "destination=", 12) == 0) -+ { -+ if (strncmp(ptran+off+12, szextaddr, extaddrlen) == 0) -+ { -+ is_stun = 1; -+ } -+ if (dstact == DSTACT_STRIP || (dstact == DSTACT_AUTO && !is_stun)) -+ { -+ diff = nextfieldoff-off; -+ if (!nf_nat_mangle_tcp_packet(pskb, ct, ctinfo, -+ off, diff, NULL, 0)) -+ { -+ /* mangle failed, all we can do is bail */ -+ nf_conntrack_unexpect_related(exp); -+ return 0; -+ } -+ get_skb_tcpdata(*pskb, &ptcp, &tcplen); -+ ptran = ptcp+tranoff; -+ tranlen -= diff; -+ nextparamoff -= diff; -+ nextfieldoff -= diff; -+ } -+ } -+ -+ off = nextfieldoff; -+ } -+ if (is_stun) -+ { -+ continue; -+ } -+ off = saveoff; -+ while (off < nextparamoff) -+ { -+ const char* pfieldend; -+ uint nextfieldoff; -+ -+ pfieldend = memchr(ptran+off, ';', nextparamoff-off); -+ nextfieldoff = (pfieldend == NULL) ? nextparamoff : pfieldend-ptran+1; -+ -+ if (strncmp(ptran+off, "client_port=", 12) == 0) -+ { -+ u_int16_t port; -+ uint numlen; -+ uint origoff; -+ uint origlen; -+ char* rbuf = rbuf1; -+ uint rbuflen = rbuf1len; -+ -+ off += 12; -+ origoff = (ptran-ptcp)+off; -+ origlen = 0; -+ numlen = nf_strtou16(ptran+off, &port); -+ off += numlen; -+ origlen += numlen; -+ if (port != prtspexp->loport) -+ { -+ DEBUGP("multiple ports found, port %hu ignored\n", port); -+ } -+ else -+ { -+ if (ptran[off] == '-' || ptran[off] == '/') -+ { -+ off++; -+ origlen++; -+ numlen = nf_strtou16(ptran+off, &port); -+ off += numlen; -+ origlen += numlen; -+ rbuf = rbufa; -+ rbuflen = rbufalen; -+ } -+ -+ /* -+ * note we cannot just memcpy() if the sizes are the same. -+ * the mangle function does skb resizing, checks for a -+ * cloned skb, and updates the checksums. -+ * -+ * parameter 4 below is offset from start of tcp data. -+ */ -+ diff = origlen-rbuflen; -+ if (!nf_nat_mangle_tcp_packet(pskb, ct, ctinfo, -+ origoff, origlen, rbuf, rbuflen)) -+ { -+ /* mangle failed, all we can do is bail */ -+ nf_conntrack_unexpect_related(exp); -+ return 0; -+ } -+ get_skb_tcpdata(*pskb, &ptcp, &tcplen); -+ ptran = ptcp+tranoff; -+ tranlen -= diff; -+ nextparamoff -= diff; -+ nextfieldoff -= diff; -+ } -+ } -+ -+ off = nextfieldoff; -+ } -+ -+ off = nextparamoff; -+ } -+ -+ return 1; -+} -+ -+static uint -+help_out(struct sk_buff **pskb, enum ip_conntrack_info ctinfo, -+ unsigned int matchoff, unsigned int matchlen, struct ip_ct_rtsp_expect* prtspexp, -+ struct nf_conntrack_expect* exp) -+{ -+ char* ptcp; -+ uint tcplen; -+ uint hdrsoff; -+ uint hdrslen; -+ uint lineoff; -+ uint linelen; -+ uint off; -+ -+ //struct iphdr* iph = (struct iphdr*)(*pskb)->nh.iph; -+ //struct tcphdr* tcph = (struct tcphdr*)((void*)iph + iph->ihl*4); -+ -+ get_skb_tcpdata(*pskb, &ptcp, &tcplen); -+ hdrsoff = matchoff;//exp->seq - ntohl(tcph->seq); -+ hdrslen = matchlen; -+ off = hdrsoff; -+ DEBUGP("NAT rtsp help_out\n"); -+ -+ while (nf_mime_nextline(ptcp, hdrsoff+hdrslen, &off, &lineoff, &linelen)) -+ { -+ if (linelen == 0) -+ { -+ break; -+ } -+ if (off > hdrsoff+hdrslen) -+ { -+ INFOP("!! overrun !!"); -+ break; -+ } -+ DEBUGP("hdr: len=%u, %.*s", linelen, (int)linelen, ptcp+lineoff); -+ -+ if (nf_strncasecmp(ptcp+lineoff, "Transport:", 10) == 0) -+ { -+ uint oldtcplen = tcplen; -+ DEBUGP("hdr: Transport\n"); -+ if (!rtsp_mangle_tran(ctinfo, exp, prtspexp, pskb, lineoff, linelen)) -+ { -+ DEBUGP("hdr: Transport mangle failed"); -+ break; -+ } -+ get_skb_tcpdata(*pskb, &ptcp, &tcplen); -+ hdrslen -= (oldtcplen-tcplen); -+ off -= (oldtcplen-tcplen); -+ lineoff -= (oldtcplen-tcplen); -+ linelen -= (oldtcplen-tcplen); -+ DEBUGP("rep: len=%u, %.*s", linelen, (int)linelen, ptcp+lineoff); -+ } -+ } -+ -+ return NF_ACCEPT; -+} -+ -+static unsigned int -+help(struct sk_buff **pskb, enum ip_conntrack_info ctinfo, -+ unsigned int matchoff, unsigned int matchlen, struct ip_ct_rtsp_expect* prtspexp, -+ struct nf_conntrack_expect* exp) -+{ -+ int dir = CTINFO2DIR(ctinfo); -+ int rc = NF_ACCEPT; -+ -+ switch (dir) -+ { -+ case IP_CT_DIR_ORIGINAL: -+ rc = help_out(pskb, ctinfo, matchoff, matchlen, prtspexp, exp); -+ break; -+ case IP_CT_DIR_REPLY: -+ DEBUGP("unmangle ! %u\n", ctinfo); -+ /* XXX: unmangle */ -+ rc = NF_ACCEPT; -+ break; -+ } -+ //UNLOCK_BH(&ip_rtsp_lock); -+ -+ return rc; -+} -+ -+static void expected(struct nf_conn* ct, struct nf_conntrack_expect *exp) -+{ -+ struct nf_nat_multi_range_compat mr; -+ u_int32_t newdstip, newsrcip, newip; -+ -+ struct nf_conn *master = ct->master; -+ -+ newdstip = master->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.u3.ip; -+ newsrcip = ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.u3.ip; -+ //FIXME (how to port that ?) -+ //code from 2.4 : newip = (HOOK2MANIP(hooknum) == IP_NAT_MANIP_SRC) ? newsrcip : newdstip; -+ newip = newdstip; -+ -+ DEBUGP("newsrcip=%u.%u.%u.%u, newdstip=%u.%u.%u.%u, newip=%u.%u.%u.%u\n", -+ NIPQUAD(newsrcip), NIPQUAD(newdstip), NIPQUAD(newip)); -+ -+ mr.rangesize = 1; -+ // We don't want to manip the per-protocol, just the IPs. -+ mr.range[0].flags = IP_NAT_RANGE_MAP_IPS; -+ mr.range[0].min_ip = mr.range[0].max_ip = newip; -+ -+ nf_nat_setup_info(ct, &mr.range[0], NF_IP_PRE_ROUTING); -+} -+ -+ -+static void __exit fini(void) -+{ -+ nf_nat_rtsp_hook = NULL; -+ nf_nat_rtsp_hook_expectfn = NULL; -+ synchronize_net(); -+} -+ -+static int __init init(void) -+{ -+ printk("nf_nat_rtsp v" IP_NF_RTSP_VERSION " loading\n"); -+ -+ BUG_ON(nf_nat_rtsp_hook); -+ nf_nat_rtsp_hook = help; -+ nf_nat_rtsp_hook_expectfn = &expected; -+ -+ if (stunaddr != NULL) -+ extip = in_aton(stunaddr); -+ -+ if (destaction != NULL) { -+ if (strcmp(destaction, "auto") == 0) -+ dstact = DSTACT_AUTO; -+ -+ if (strcmp(destaction, "strip") == 0) -+ dstact = DSTACT_STRIP; -+ -+ if (strcmp(destaction, "none") == 0) -+ dstact = DSTACT_NONE; -+ } -+ -+ return 0; -+} -+ -+module_init(init); -+module_exit(fini); diff --git a/target/linux/generic-2.6/patches-2.6.22/200-sched_esfq.patch b/target/linux/generic-2.6/patches-2.6.22/200-sched_esfq.patch deleted file mode 100644 index 005f7ed..0000000 --- a/target/linux/generic-2.6/patches-2.6.22/200-sched_esfq.patch +++ /dev/null @@ -1,789 +0,0 @@ ---- a/include/linux/pkt_sched.h -+++ b/include/linux/pkt_sched.h -@@ -146,8 +146,40 @@ - * - * The only reason for this is efficiency, it is possible - * to change these parameters in compile time. -+ * -+ * If you need to play with these values use esfq instead. - */ - -+/* ESFQ section */ -+ -+enum -+{ -+ /* traditional */ -+ TCA_SFQ_HASH_CLASSIC, -+ TCA_SFQ_HASH_DST, -+ TCA_SFQ_HASH_SRC, -+ TCA_SFQ_HASH_FWMARK, -+ /* direct */ -+ TCA_SFQ_HASH_DSTDIR, -+ TCA_SFQ_HASH_SRCDIR, -+ TCA_SFQ_HASH_FWMARKDIR, -+ /* conntrack */ -+ TCA_SFQ_HASH_CTORIGDST, -+ TCA_SFQ_HASH_CTORIGSRC, -+ TCA_SFQ_HASH_CTREPLDST, -+ TCA_SFQ_HASH_CTREPLSRC, -+}; -+ -+struct tc_esfq_qopt -+{ -+ unsigned quantum; /* Bytes per round allocated to flow */ -+ int perturb_period; /* Period of hash perturbation */ -+ __u32 limit; /* Maximal packets in queue */ -+ unsigned divisor; /* Hash divisor */ -+ unsigned flows; /* Maximal number of flows */ -+ unsigned hash_kind; /* Hash function to use for flow identification */ -+}; -+ - /* RED section */ - - enum ---- a/net/sched/Kconfig -+++ b/net/sched/Kconfig -@@ -133,6 +133,26 @@ - To compile this code as a module, choose M here: the - module will be called sch_sfq. - -+config NET_SCH_ESFQ -+ tristate "Enhanced Stochastic Fairness Queueing (ESFQ)" -+ ---help--- -+ Say Y here if you want to use the Enhanced Stochastic Fairness -+ Queueing (ESFQ) packet scheduling algorithm for some of your network -+ devices or as a leaf discipline for a classful qdisc such as HTB or -+ CBQ (see the top of <file:net/sched/sch_esfq.c> for details and -+ references to the SFQ algorithm). -+ -+ This is an enchanced SFQ version which allows you to control some -+ hardcoded values in the SFQ scheduler. -+ -+ ESFQ also adds control of the hash function used to identify packet -+ flows. The original SFQ discipline hashes by connection; ESFQ add -+ several other hashing methods, such as by src IP or by dst IP, which -+ can be more fair to users in some networking situations. -+ -+ To compile this code as a module, choose M here: the -+ module will be called sch_esfq. -+ - config NET_SCH_TEQL - tristate "True Link Equalizer (TEQL)" - ---help--- ---- a/net/sched/Makefile -+++ b/net/sched/Makefile -@@ -23,6 +23,7 @@ - obj-$(CONFIG_NET_SCH_INGRESS) += sch_ingress.o - obj-$(CONFIG_NET_SCH_DSMARK) += sch_dsmark.o - obj-$(CONFIG_NET_SCH_SFQ) += sch_sfq.o -+obj-$(CONFIG_NET_SCH_ESFQ) += sch_esfq.o - obj-$(CONFIG_NET_SCH_TBF) += sch_tbf.o - obj-$(CONFIG_NET_SCH_TEQL) += sch_teql.o - obj-$(CONFIG_NET_SCH_PRIO) += sch_prio.o ---- /dev/null -+++ b/net/sched/sch_esfq.c -@@ -0,0 +1,704 @@ -+/* -+ * net/sched/sch_esfq.c Extended Stochastic Fairness Queueing discipline. -+ * -+ * This program is free software; you can redistribute it and/or -+ * modify it under the terms of the GNU General Public License -+ * as published by the Free Software Foundation; either version -+ * 2 of the License, or (at your option) any later version. -+ * -+ * Authors: Alexey Kuznetsov, <kuznet@ms2.inr.ac.ru> -+ * -+ * Changes: Alexander Atanasov, <alex@ssi.bg> -+ * Added dynamic depth,limit,divisor,hash_kind options. -+ * Added dst and src hashes. -+ * -+ * Alexander Clouter, <alex@digriz.org.uk> -+ * Ported ESFQ to Linux 2.6. -+ * -+ * Corey Hickey, <bugfood-c@fatooh.org> -+ * Maintenance of the Linux 2.6 port. -+ * Added fwmark hash (thanks to Robert Kurjata). -+ * Added direct hashing for src, dst, and fwmark. -+ * Added usage of jhash. -+ * -+ */ -+ -+#include <linux/module.h> -+#include <asm/uaccess.h> -+#include <asm/system.h> -+#include <linux/bitops.h> -+#include <linux/types.h> -+#include <linux/kernel.h> -+#include <linux/jiffies.h> -+#include <linux/string.h> -+#include <linux/mm.h> -+#include <linux/socket.h> -+#include <linux/sockios.h> -+#include <linux/in.h> -+#include <linux/errno.h> -+#include <linux/interrupt.h> -+#include <linux/if_ether.h> -+#include <linux/inet.h> -+#include <linux/netdevice.h> -+#include <linux/etherdevice.h> -+#include <linux/notifier.h> -+#include <linux/init.h> -+#include <net/ip.h> -+#include <linux/ipv6.h> -+#include <net/route.h> -+#include <linux/skbuff.h> -+#include <net/sock.h> -+#include <net/pkt_sched.h> -+#include <linux/jhash.h> -+ -+#ifdef CONFIG_NF_CONNTRACK_ENABLED -+#include <net/netfilter/nf_conntrack.h> -+#endif -+ -+/* Stochastic Fairness Queuing algorithm. -+ For more comments look at sch_sfq.c. -+ The difference is that you can change limit, depth, -+ hash table size and choose alternate hash types. -+ -+ classic: same as in sch_sfq.c -+ dst: destination IP address -+ src: source IP address -+ fwmark: netfilter mark value -+ dst_direct: -+ src_direct: -+ fwmark_direct: direct hashing of the above sources -+ ctorigdst: original destination IP address -+ ctorigsrc: original source IP address -+ ctrepldst: reply destination IP address -+ ctreplsrc: reply source IP -+ -+*/ -+ -+ -+/* This type should contain at least SFQ_DEPTH*2 values */ -+typedef unsigned int esfq_index; -+ -+struct esfq_head -+{ -+ esfq_index next; -+ esfq_index prev; -+}; -+ -+struct esfq_sched_data -+{ -+/* Parameters */ -+ int perturb_period; -+ unsigned quantum; /* Allotment per round: MUST BE >= MTU */ -+ int limit; -+ unsigned depth; -+ unsigned hash_divisor; -+ unsigned hash_kind; -+/* Variables */ -+ struct timer_list perturb_timer; -+ int perturbation; -+ esfq_index tail; /* Index of current slot in round */ -+ esfq_index max_depth; /* Maximal depth */ -+ -+ esfq_index *ht; /* Hash table */ -+ esfq_index *next; /* Active slots link */ -+ short *allot; /* Current allotment per slot */ -+ unsigned short *hash; /* Hash value indexed by slots */ -+ struct sk_buff_head *qs; /* Slot queue */ -+ struct esfq_head *dep; /* Linked list of slots, indexed by depth */ -+ unsigned dyn_min; /* For dynamic divisor adjustment; minimum value seen */ -+ unsigned dyn_max; /* maximum value seen */ -+ unsigned dyn_range; /* saved range */ -+}; -+ -+/* This contains the info we will hash. */ -+struct esfq_packet_info -+{ -+ u32 proto; /* protocol or port */ -+ u32 src; /* source from packet header */ -+ u32 dst; /* destination from packet header */ -+ u32 ctorigsrc; /* original source from conntrack */ -+ u32 ctorigdst; /* original destination from conntrack */ -+ u32 ctreplsrc; /* reply source from conntrack */ -+ u32 ctrepldst; /* reply destination from conntrack */ -+ u32 mark; /* netfilter mark (fwmark) */ -+}; -+ -+/* Hash input values directly into the "nearest" slot, taking into account the -+ * range of input values seen. This is most useful when the hash table is at -+ * least as large as the range of possible values. -+ * Note: this functionality was added before the change to using jhash, and may -+ * no longer be useful. */ -+static __inline__ unsigned esfq_hash_direct(struct esfq_sched_data *q, u32 h) -+{ -+ /* adjust minimum and maximum */ -+ if (h < q->dyn_min || h > q->dyn_max) { -+ q->dyn_min = h < q->dyn_min ? h : q->dyn_min; -+ q->dyn_max = h > q->dyn_max ? h : q->dyn_max; -+ -+ /* find new range */ -+ if ((q->dyn_range = q->dyn_max - q->dyn_min) >= q->hash_divisor) -+ printk(KERN_WARNING "ESFQ: (direct hash) Input range %u is larger than hash " -+ "table. See ESFQ README for details.\n", q->dyn_range); -+ } -+ -+ /* hash input values into slot numbers */ -+ if (q->dyn_min == q->dyn_max) -+ return 0; /* only one value seen; avoid division by 0 */ -+ else -+ return (h - q->dyn_min) * (q->hash_divisor - 1) / q->dyn_range; -+} -+ -+static __inline__ unsigned esfq_jhash_1word(struct esfq_sched_data *q,u32 a) -+{ -+ return jhash_1word(a, q->perturbation) & (q->hash_divisor-1); -+} -+ -+static __inline__ unsigned esfq_jhash_2words(struct esfq_sched_data *q, u32 a, u32 b) -+{ -+ return jhash_2words(a, b, q->perturbation) & (q->hash_divisor-1); -+} -+ -+static __inline__ unsigned esfq_jhash_3words(struct esfq_sched_data *q, u32 a, u32 b, u32 c) -+{ -+ return jhash_3words(a, b, c, q->perturbation) & (q->hash_divisor-1); -+} -+ -+ -+static unsigned esfq_hash(struct esfq_sched_data *q, struct sk_buff *skb) -+{ -+ struct esfq_packet_info info; -+#ifdef CONFIG_NF_CONNTRACK_ENABLED -+ enum ip_conntrack_info ctinfo; -+ struct nf_conn *ct = nf_ct_get(skb, &ctinfo); -+#endif -+ -+ switch (skb->protocol) { -+ case __constant_htons(ETH_P_IP): -+ { -+ struct iphdr *iph = ip_hdr(skb); -+ info.dst = iph->daddr; -+ info.src = iph->saddr; -+ if (!(iph->frag_off&htons(IP_MF|IP_OFFSET)) && -+ (iph->protocol == IPPROTO_TCP || -+ iph->protocol == IPPROTO_UDP || -+ iph->protocol == IPPROTO_SCTP || -+ iph->protocol == IPPROTO_DCCP || -+ iph->protocol == IPPROTO_ESP)) -+ info.proto = *(((u32*)iph) + iph->ihl); -+ else -+ info.proto = iph->protocol; -+ break; -+ } -+ case __constant_htons(ETH_P_IPV6): -+ { -+ struct ipv6hdr *iph = ipv6_hdr(skb); -+ /* Hash ipv6 addresses into a u32. This isn't ideal, -+ * but the code is simple. */ -+ info.dst = jhash2(iph->daddr.s6_addr32, 4, q->perturbation); -+ info.src = jhash2(iph->saddr.s6_addr32, 4, q->perturbation); -+ if (iph->nexthdr == IPPROTO_TCP || -+ iph->nexthdr == IPPROTO_UDP || -+ iph->nexthdr == IPPROTO_SCTP || -+ iph->nexthdr == IPPROTO_DCCP || -+ iph->nexthdr == IPPROTO_ESP) -+ info.proto = *(u32*)&iph[1]; -+ else -+ info.proto = iph->nexthdr; -+ break; -+ } -+ default: -+ info.dst = (u32)(unsigned long)skb->dst; -+ info.src = (u32)(unsigned long)skb->sk; -+ info.proto = skb->protocol; -+ } -+ -+ info.mark = skb->mark; -+ -+#ifdef CONFIG_NF_CONNTRACK_ENABLED -+ /* defaults if there is no conntrack info */ -+ info.ctorigsrc = info.src; -+ info.ctorigdst = info.dst; -+ info.ctreplsrc = info.dst; -+ info.ctrepldst = info.src; -+ /* collect conntrack info */ -+ if (ct && ct != &nf_conntrack_untracked) { -+ if (skb->protocol == __constant_htons(ETH_P_IP)) { -+ info.ctorigsrc = ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.u3.ip; -+ info.ctorigdst = ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.dst.u3.ip; -+ info.ctreplsrc = ct->tuplehash[IP_CT_DIR_REPLY].tuple.src.u3.ip; -+ info.ctrepldst = ct->tuplehash[IP_CT_DIR_REPLY].tuple.dst.u3.ip; -+ } -+ else if (skb->protocol == __constant_htons(ETH_P_IPV6)) { -+ /* Again, hash ipv6 addresses into a single u32. */ -+ info.ctorigsrc = jhash2(ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.u3.ip6, 4, q->perturbation); -+ info.ctorigdst = jhash2(ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.dst.u3.ip6, 4, q->perturbation); -+ info.ctreplsrc = jhash2(ct->tuplehash[IP_CT_DIR_REPLY].tuple.src.u3.ip6, 4, q->perturbation); -+ info.ctrepldst = jhash2(ct->tuplehash[IP_CT_DIR_REPLY].tuple.dst.u3.ip6, 4, q->perturbation); -+ } -+ -+ } -+#endif -+ -+ switch(q->hash_kind) -+ { -+ case TCA_SFQ_HASH_CLASSIC: -+ return esfq_jhash_3words(q, info.dst, info.src, info.proto); -+ case TCA_SFQ_HASH_DST: -+ return esfq_jhash_1word(q, info.dst); -+ case TCA_SFQ_HASH_DSTDIR: -+ return esfq_hash_direct(q, ntohl(info.dst)); -+ case TCA_SFQ_HASH_SRC: -+ return esfq_jhash_1word(q, info.src); -+ case TCA_SFQ_HASH_SRCDIR: -+ return esfq_hash_direct(q, ntohl(info.src)); -+ case TCA_SFQ_HASH_FWMARK: -+ return esfq_jhash_1word(q, info.mark); -+ case TCA_SFQ_HASH_FWMARKDIR: -+ return esfq_hash_direct(q, info.mark); -+#ifdef CONFIG_NF_CONNTRACK_ENABLED -+ case TCA_SFQ_HASH_CTORIGDST: -+ return esfq_jhash_1word(q, info.ctorigdst); -+ case TCA_SFQ_HASH_CTORIGSRC: -+ return esfq_jhash_1word(q, info.ctorigsrc); -+ case TCA_SFQ_HASH_CTREPLDST: -+ return esfq_jhash_1word(q, info.ctrepldst); -+ case TCA_SFQ_HASH_CTREPLSRC: -+ return esfq_jhash_1word(q, info.ctreplsrc); -+#endif -+ default: -+ if (net_ratelimit()) -+ printk(KERN_WARNING "ESFQ: Unknown hash method. Falling back to classic.\n"); -+ } -+ return esfq_jhash_3words(q, info.dst, info.src, info.proto); -+} -+ -+static inline void esfq_link(struct esfq_sched_data *q, esfq_index x) -+{ -+ esfq_index p, n; -+ int d = q->qs[x].qlen + q->depth; -+ -+ p = d; -+ n = q->dep[d].next; -+ q->dep[x].next = n; -+ q->dep[x].prev = p; -+ q->dep[p].next = q->dep[n].prev = x; -+} -+ -+static inline void esfq_dec(struct esfq_sched_data *q, esfq_index x) -+{ -+ esfq_index p, n; -+ -+ n = q->dep[x].next; -+ p = q->dep[x].prev; -+ q->dep[p].next = n; -+ q->dep[n].prev = p; -+ -+ if (n == p && q->max_depth == q->qs[x].qlen + 1) -+ q->max_depth--; -+ -+ esfq_link(q, x); -+} -+ -+static inline void esfq_inc(struct esfq_sched_data *q, esfq_index x) -+{ -+ esfq_index p, n; -+ int d; -+ -+ n = q->dep[x].next; -+ p = q->dep[x].prev; -+ q->dep[p].next = n; -+ q->dep[n].prev = p; -+ d = q->qs[x].qlen; -+ if (q->max_depth < d) -+ q->max_depth = d; -+ -+ esfq_link(q, x); -+} -+ -+static unsigned int esfq_drop(struct Qdisc *sch) -+{ -+ struct esfq_sched_data *q = qdisc_priv(sch); -+ esfq_index d = q->max_depth; -+ struct sk_buff *skb; -+ unsigned int len; -+ -+ /* Queue is full! Find the longest slot and -+ drop a packet from it */ -+ -+ if (d > 1) { -+ esfq_index x = q->dep[d+q->depth].next; -+ skb = q->qs[x].prev; -+ len = skb->len; -+ __skb_unlink(skb, &q->qs[x]); -+ kfree_skb(skb); -+ esfq_dec(q, x); -+ sch->q.qlen--; -+ sch->qstats.drops++; -+ sch->qstats.backlog -= len; -+ return len; -+ } -+ -+ if (d == 1) { -+ /* It is difficult to believe, but ALL THE SLOTS HAVE LENGTH 1. */ -+ d = q->next[q->tail]; -+ q->next[q->tail] = q->next[d]; -+ q->allot[q->next[d]] += q->quantum; -+ skb = q->qs[d].prev; -+ len = skb->len; -+ __skb_unlink(skb, &q->qs[d]); -+ kfree_skb(skb); -+ esfq_dec(q, d); -+ sch->q.qlen--; -+ q->ht[q->hash[d]] = q->depth; -+ sch->qstats.drops++; -+ sch->qstats.backlog -= len; -+ return len; -+ } -+ -+ return 0; -+} -+ -+static int -+esfq_enqueue(struct sk_buff *skb, struct Qdisc* sch) -+{ -+ struct esfq_sched_data *q = qdisc_priv(sch); -+ unsigned hash = esfq_hash(q, skb); -+ unsigned depth = q->depth; -+ esfq_index x; -+ -+ x = q->ht[hash]; -+ if (x == depth) { -+ q->ht[hash] = x = q->dep[depth].next; -+ q->hash[x] = hash; -+ } -+ sch->qstats.backlog += skb->len; -+ __skb_queue_tail(&q->qs[x], skb); -+ esfq_inc(q, x); -+ if (q->qs[x].qlen == 1) { /* The flow is new */ -+ if (q->tail == depth) { /* It is the first flow */ -+ q->tail = x; -+ q->next[x] = x; -+ q->allot[x] = q->quantum; -+ } else { -+ q->next[x] = q->next[q->tail]; -+ q->next[q->tail] = x; -+ q->tail = x; -+ } -+ } -+ if (++sch->q.qlen < q->limit-1) { -+ sch->bstats.bytes += skb->len; -+ sch->bstats.packets++; -+ return 0; -+ } -+ -+ esfq_drop(sch); -+ return NET_XMIT_CN; -+} -+ -+static int -+esfq_requeue(struct sk_buff *skb, struct Qdisc* sch) -+{ -+ struct esfq_sched_data *q = qdisc_priv(sch); -+ unsigned hash = esfq_hash(q, skb); -+ unsigned depth = q->depth; -+ esfq_index x; -+ -+ x = q->ht[hash]; -+ if (x == depth) { -+ q->ht[hash] = x = q->dep[depth].next; -+ q->hash[x] = hash; -+ } -+ sch->qstats.backlog += skb->len; -+ __skb_queue_head(&q->qs[x], skb); -+ esfq_inc(q, x); -+ if (q->qs[x].qlen == 1) { /* The flow is new */ -+ if (q->tail == depth) { /* It is the first flow */ -+ q->tail = x; -+ q->next[x] = x; -+ q->allot[x] = q->quantum; -+ } else { -+ q->next[x] = q->next[q->tail]; -+ q->next[q->tail] = x; -+ q->tail = x; -+ } -+ } -+ if (++sch->q.qlen < q->limit - 1) { -+ sch->qstats.requeues++; -+ return 0; -+ } -+ -+ sch->qstats.drops++; -+ esfq_drop(sch); -+ return NET_XMIT_CN; -+} -+ -+ -+ -+ -+static struct sk_buff * -+esfq_dequeue(struct Qdisc* sch) -+{ -+ struct esfq_sched_data *q = qdisc_priv(sch); -+ struct sk_buff *skb; -+ unsigned depth = q->depth; -+ esfq_index a, old_a; -+ -+ /* No active slots */ -+ if (q->tail == depth) -+ return NULL; -+ -+ a = old_a = q->next[q->tail]; -+ -+ /* Grab packet */ -+ skb = __skb_dequeue(&q->qs[a]); -+ esfq_dec(q, a); -+ sch->q.qlen--; -+ sch->qstats.backlog -= skb->len; -+ -+ /* Is the slot empty? */ -+ if (q->qs[a].qlen == 0) { -+ q->ht[q->hash[a]] = depth; -+ a = q->next[a]; -+ if (a == old_a) { -+ q->tail = depth; -+ return skb; -+ } -+ q->next[q->tail] = a; -+ q->allot[a] += q->quantum; -+ } else if ((q->allot[a] -= skb->len) <= 0) { -+ q->tail = a; -+ a = q->next[a]; -+ q->allot[a] += q->quantum; -+ } -+ -+ return skb; -+} -+ -+static void -+esfq_reset(struct Qdisc* sch) -+{ -+ struct sk_buff *skb; -+ -+ while ((skb = esfq_dequeue(sch)) != NULL) -+ kfree_skb(skb); -+} -+ -+static void esfq_perturbation(unsigned long arg) -+{ -+ struct Qdisc *sch = (struct Qdisc*)arg; -+ struct esfq_sched_data *q = qdisc_priv(sch); -+ -+ q->perturbation = net_random()&0x1F; -+ -+ if (q->perturb_period) { -+ q->perturb_timer.expires = jiffies + q->perturb_period; -+ add_timer(&q->perturb_timer); -+ } -+} -+ -+static int esfq_change(struct Qdisc *sch, struct rtattr *opt) -+{ -+ struct esfq_sched_data *q = qdisc_priv(sch); -+ struct tc_esfq_qopt *ctl = RTA_DATA(opt); -+ int old_perturb = q->perturb_period; -+ -+ if (opt->rta_len < RTA_LENGTH(sizeof(*ctl))) -+ return -EINVAL; -+ -+ sch_tree_lock(sch); -+ q->quantum = ctl->quantum ? : psched_mtu(sch->dev); -+ q->perturb_period = ctl->perturb_period*HZ; -+// q->hash_divisor = ctl->divisor; -+// q->tail = q->limit = q->depth = ctl->flows; -+ -+ if (ctl->limit) -+ q->limit = min_t(u32, ctl->limit, q->depth); -+ -+ if (ctl->hash_kind) { -+ q->hash_kind = ctl->hash_kind; -+ if (q->hash_kind != TCA_SFQ_HASH_CLASSIC) -+ q->perturb_period = 0; -+ } -+ -+ // is sch_tree_lock enough to do this ? -+ while (sch->q.qlen >= q->limit-1) -+ esfq_drop(sch); -+ -+ if (old_perturb) -+ del_timer(&q->perturb_timer); -+ if (q->perturb_period) { -+ q->perturb_timer.expires = jiffies + q->perturb_period; -+ add_timer(&q->perturb_timer); -+ } else { -+ q->perturbation = 0; -+ } -+ sch_tree_unlock(sch); -+ return 0; -+} -+ -+static int esfq_init(struct Qdisc *sch, struct rtattr *opt) -+{ -+ struct esfq_sched_data *q = qdisc_priv(sch); -+ struct tc_esfq_qopt *ctl; -+ esfq_index p = ~0U/2; -+ int i; -+ -+ if (opt && opt->rta_len < RTA_LENGTH(sizeof(*ctl))) -+ return -EINVAL; -+ -+ init_timer(&q->perturb_timer); -+ q->perturb_timer.data = (unsigned long)sch; -+ q->perturb_timer.function = esfq_perturbation; -+ q->perturbation = 0; -+ q->hash_kind = TCA_SFQ_HASH_CLASSIC; -+ q->max_depth = 0; -+ q->dyn_min = ~0U; /* maximum value for this type */ -+ q->dyn_max = 0; /* dyn_min/dyn_max will be set properly upon first packet */ -+ if (opt == NULL) { -+ q->quantum = psched_mtu(sch->dev); -+ q->perturb_period = 0; -+ q->hash_divisor = 1024; -+ q->tail = q->limit = q->depth = 128; -+ -+ } else { -+ ctl = RTA_DATA(opt); -+ q->quantum = ctl->quantum ? : psched_mtu(sch->dev); -+ q->perturb_period = ctl->perturb_period*HZ; -+ q->hash_divisor = ctl->divisor ? : 1024; -+ q->tail = q->limit = q->depth = ctl->flows ? : 128; -+ -+ if ( q->depth > p - 1 ) -+ return -EINVAL; -+ -+ if (ctl->limit) -+ q->limit = min_t(u32, ctl->limit, q->depth); -+ -+ if (ctl->hash_kind) { -+ q->hash_kind = ctl->hash_kind; -+ } -+ -+ if (q->perturb_period) { -+ q->perturb_timer.expires = jiffies + q->perturb_period; -+ add_timer(&q->perturb_timer); -+ } -+ } -+ -+ q->ht = kmalloc(q->hash_divisor*sizeof(esfq_index), GFP_KERNEL); -+ if (!q->ht) -+ goto err_case; -+ -+ q->dep = kmalloc((1+q->depth*2)*sizeof(struct esfq_head), GFP_KERNEL); -+ if (!q->dep) -+ goto err_case; -+ q->next = kmalloc(q->depth*sizeof(esfq_index), GFP_KERNEL); -+ if (!q->next) -+ goto err_case; -+ -+ q->allot = kmalloc(q->depth*sizeof(short), GFP_KERNEL); -+ if (!q->allot) -+ goto err_case; -+ q->hash = kmalloc(q->depth*sizeof(unsigned short), GFP_KERNEL); -+ if (!q->hash) -+ goto err_case; -+ q->qs = kmalloc(q->depth*sizeof(struct sk_buff_head), GFP_KERNEL); -+ if (!q->qs) -+ goto err_case; -+ -+ for (i=0; i< q->hash_divisor; i++) -+ q->ht[i] = q->depth; -+ for (i=0; i<q->depth; i++) { -+ skb_queue_head_init(&q->qs[i]); -+ q->dep[i+q->depth].next = i+q->depth; -+ q->dep[i+q->depth].prev = i+q->depth; -+ } -+ -+ for (i=0; i<q->depth; i++) -+ esfq_link(q, i); -+ return 0; -+err_case: -+ del_timer(&q->perturb_timer); -+ if (q->ht) -+ kfree(q->ht); -+ if (q->dep) -+ kfree(q->dep); -+ if (q->next) -+ kfree(q->next); -+ if (q->allot) -+ kfree(q->allot); -+ if (q->hash) -+ kfree(q->hash); -+ if (q->qs) -+ kfree(q->qs); -+ return -ENOBUFS; -+} -+ -+static void esfq_destroy(struct Qdisc *sch) -+{ -+ struct esfq_sched_data *q = qdisc_priv(sch); -+ del_timer(&q->perturb_timer); -+ if(q->ht) -+ kfree(q->ht); -+ if(q->dep) -+ kfree(q->dep); -+ if(q->next) -+ kfree(q->next); -+ if(q->allot) -+ kfree(q->allot); -+ if(q->hash) -+ kfree(q->hash); -+ if(q->qs) -+ kfree(q->qs); -+} -+ -+static int esfq_dump(struct Qdisc *sch, struct sk_buff *skb) -+{ -+ struct esfq_sched_data *q = qdisc_priv(sch); -+ unsigned char *b = skb->tail; -+ struct tc_esfq_qopt opt; -+ -+ opt.quantum = q->quantum; -+ opt.perturb_period = q->perturb_period/HZ; -+ -+ opt.limit = q->limit; -+ opt.divisor = q->hash_divisor; -+ opt.flows = q->depth; -+ opt.hash_kind = q->hash_kind; -+ -+ RTA_PUT(skb, TCA_OPTIONS, sizeof(opt), &opt); -+ -+ return skb->len; -+ -+rtattr_failure: -+ skb_trim(skb, b - skb->data); -+ return -1; -+} -+ -+static struct Qdisc_ops esfq_qdisc_ops = -+{ -+ .next = NULL, -+ .cl_ops = NULL, -+ .id = "esfq", -+ .priv_size = sizeof(struct esfq_sched_data), -+ .enqueue = esfq_enqueue, -+ .dequeue = esfq_dequeue, -+ .requeue = esfq_requeue, -+ .drop = esfq_drop, -+ .init = esfq_init, -+ .reset = esfq_reset, -+ .destroy = esfq_destroy, -+ .change = NULL, /* esfq_change - needs more work */ -+ .dump = esfq_dump, -+ .owner = THIS_MODULE, -+}; -+ -+static int __init esfq_module_init(void) -+{ -+ return register_qdisc(&esfq_qdisc_ops); -+} -+static void __exit esfq_module_exit(void) -+{ -+ unregister_qdisc(&esfq_qdisc_ops); -+} -+module_init(esfq_module_init) -+module_exit(esfq_module_exit) -+MODULE_LICENSE("GPL"); diff --git a/target/linux/generic-2.6/patches-2.6.22/201-multiple_default_gateways.patch b/target/linux/generic-2.6/patches-2.6.22/201-multiple_default_gateways.patch deleted file mode 100644 index 47d53b7..0000000 --- a/target/linux/generic-2.6/patches-2.6.22/201-multiple_default_gateways.patch +++ /dev/null @@ -1,336 +0,0 @@ ---- a/include/linux/rtnetlink.h -+++ b/include/linux/rtnetlink.h -@@ -293,6 +293,8 @@ - #define RTNH_F_DEAD 1 /* Nexthop is dead (used by multipath) */ - #define RTNH_F_PERVASIVE 2 /* Do recursive gateway lookup */ - #define RTNH_F_ONLINK 4 /* Gateway is forced on link */ -+#define RTNH_F_SUSPECT 8 /* We don't know the real state */ -+#define RTNH_F_BADSTATE (RTNH_F_DEAD | RTNH_F_SUSPECT) - - /* Macros to handle hexthops */ - ---- a/include/net/flow.h -+++ b/include/net/flow.h -@@ -19,6 +19,8 @@ - struct { - __be32 daddr; - __be32 saddr; -+ __u32 lsrc; -+ __u32 gw; - __u8 tos; - __u8 scope; - } ip4_u; -@@ -43,6 +45,8 @@ - #define fl6_flowlabel nl_u.ip6_u.flowlabel - #define fl4_dst nl_u.ip4_u.daddr - #define fl4_src nl_u.ip4_u.saddr -+#define fl4_lsrc nl_u.ip4_u.lsrc -+#define fl4_gw nl_u.ip4_u.gw - #define fl4_tos nl_u.ip4_u.tos - #define fl4_scope nl_u.ip4_u.scope - ---- a/net/ipv4/route.c -+++ b/net/ipv4/route.c -@@ -1208,6 +1208,7 @@ - - /* Gateway is different ... */ - rt->rt_gateway = new_gw; -+ if (rt->fl.fl4_gw) rt->fl.fl4_gw = new_gw; - - /* Redirect received -> path was valid */ - dst_confirm(&rth->u.dst); -@@ -1643,6 +1644,7 @@ - rth->fl.fl4_tos = tos; - rth->fl.mark = skb->mark; - rth->fl.fl4_src = saddr; -+ rth->fl.fl4_lsrc = 0; - rth->rt_src = saddr; - #ifdef CONFIG_NET_CLS_ROUTE - rth->u.dst.tclassid = itag; -@@ -1653,6 +1655,7 @@ - dev_hold(rth->u.dst.dev); - rth->idev = in_dev_get(rth->u.dst.dev); - rth->fl.oif = 0; -+ rth->fl.fl4_gw = 0; - rth->rt_gateway = daddr; - rth->rt_spec_dst= spec_dst; - rth->rt_type = RTN_MULTICAST; -@@ -1716,7 +1719,7 @@ - static inline int __mkroute_input(struct sk_buff *skb, - struct fib_result* res, - struct in_device *in_dev, -- __be32 daddr, __be32 saddr, u32 tos, -+ __be32 daddr, __be32 saddr, u32 tos, u32 lsrc, - struct rtable **result) - { - -@@ -1751,6 +1754,7 @@ - flags |= RTCF_DIRECTSRC; - - if (out_dev == in_dev && err && !(flags & (RTCF_NAT | RTCF_MASQ)) && -+ !lsrc && - (IN_DEV_SHARED_MEDIA(out_dev) || - inet_addr_onlink(out_dev, saddr, FIB_RES_GW(*res)))) - flags |= RTCF_DOREDIRECT; -@@ -1788,6 +1792,7 @@ - rth->fl.mark = skb->mark; - rth->fl.fl4_src = saddr; - rth->rt_src = saddr; -+ rth->fl.fl4_lsrc = lsrc; - rth->rt_gateway = daddr; - rth->rt_iif = - rth->fl.iif = in_dev->dev->ifindex; -@@ -1795,6 +1800,7 @@ - dev_hold(rth->u.dst.dev); - rth->idev = in_dev_get(rth->u.dst.dev); - rth->fl.oif = 0; -+ rth->fl.fl4_gw = 0; - rth->rt_spec_dst= spec_dst; - - rth->u.dst.input = ip_forward; -@@ -1816,19 +1822,21 @@ - struct fib_result* res, - const struct flowi *fl, - struct in_device *in_dev, -- __be32 daddr, __be32 saddr, u32 tos) -+ __be32 daddr, __be32 saddr, u32 tos, -+ u32 lsrc) - { - struct rtable* rth = NULL; - int err; - unsigned hash; - -+ fib_select_default(fl, res); - #ifdef CONFIG_IP_ROUTE_MULTIPATH -- if (res->fi && res->fi->fib_nhs > 1 && fl->oif == 0) -+ if (res->fi && res->fi->fib_nhs > 1) - fib_select_multipath(fl, res); - #endif - - /* create a routing cache entry */ -- err = __mkroute_input(skb, res, in_dev, daddr, saddr, tos, &rth); -+ err = __mkroute_input(skb, res, in_dev, daddr, saddr, tos, lsrc, &rth); - if (err) - return err; - -@@ -1841,7 +1849,8 @@ - struct fib_result* res, - const struct flowi *fl, - struct in_device *in_dev, -- __be32 daddr, __be32 saddr, u32 tos) -+ __be32 daddr, __be32 saddr, u32 tos, -+ u32 lsrc) - { - #ifdef CONFIG_IP_ROUTE_MULTIPATH_CACHED - struct rtable* rth = NULL, *rtres; -@@ -1857,7 +1866,7 @@ - /* distinguish between multipath and singlepath */ - if (hopcount < 2) - return ip_mkroute_input_def(skb, res, fl, in_dev, daddr, -- saddr, tos); -+ saddr, tos, 0); - - /* add all alternatives to the routing cache */ - for (hop = 0; hop < hopcount; hop++) { -@@ -1869,7 +1878,7 @@ - - /* create a routing cache entry */ - err = __mkroute_input(skb, res, in_dev, daddr, saddr, tos, -- &rth); -+ 0, &rth); - if (err) - return err; - -@@ -1889,7 +1898,7 @@ - skb->dst = &rtres->u.dst; - return err; - #else /* CONFIG_IP_ROUTE_MULTIPATH_CACHED */ -- return ip_mkroute_input_def(skb, res, fl, in_dev, daddr, saddr, tos); -+ return ip_mkroute_input_def(skb, res, fl, in_dev, daddr, saddr, tos, lsrc); - #endif /* CONFIG_IP_ROUTE_MULTIPATH_CACHED */ - } - -@@ -1905,18 +1914,18 @@ - */ - - static int ip_route_input_slow(struct sk_buff *skb, __be32 daddr, __be32 saddr, -- u8 tos, struct net_device *dev) -+ u8 tos, struct net_device *dev, u32 lsrc) - { - struct fib_result res; - struct in_device *in_dev = in_dev_get(dev); - struct flowi fl = { .nl_u = { .ip4_u = - { .daddr = daddr, -- .saddr = saddr, -+ .saddr = lsrc ? : saddr, - .tos = tos, - .scope = RT_SCOPE_UNIVERSE, - } }, - .mark = skb->mark, -- .iif = dev->ifindex }; -+ .iif = lsrc? loopback_dev.ifindex : dev->ifindex }; - unsigned flags = 0; - u32 itag = 0; - struct rtable * rth; -@@ -1949,6 +1958,12 @@ - if (BADCLASS(daddr) || ZERONET(daddr) || LOOPBACK(daddr)) - goto martian_destination; - -+ if (lsrc) { -+ if (MULTICAST(lsrc) || BADCLASS(lsrc) || -+ ZERONET(lsrc) || LOOPBACK(lsrc)) -+ goto e_inval; -+ } -+ - /* - * Now we are ready to route packet. - */ -@@ -1958,6 +1973,10 @@ - goto no_route; - } - free_res = 1; -+ if (lsrc && res.type != RTN_UNICAST && res.type != RTN_NAT) -+ goto e_inval; -+ fl.iif = dev->ifindex; -+ fl.fl4_src = saddr; - - RT_CACHE_STAT_INC(in_slow_tot); - -@@ -1982,7 +2001,7 @@ - if (res.type != RTN_UNICAST) - goto martian_destination; - -- err = ip_mkroute_input(skb, &res, &fl, in_dev, daddr, saddr, tos); -+ err = ip_mkroute_input(skb, &res, &fl, in_dev, daddr, saddr, tos, lsrc); - if (err == -ENOBUFS) - goto e_nobufs; - if (err == -EINVAL) -@@ -1997,6 +2016,8 @@ - brd_input: - if (skb->protocol != htons(ETH_P_IP)) - goto e_inval; -+ if (lsrc) -+ goto e_inval; - - if (ZERONET(saddr)) - spec_dst = inet_select_addr(dev, 0, RT_SCOPE_LINK); -@@ -2037,6 +2058,7 @@ - rth->u.dst.dev = &loopback_dev; - dev_hold(rth->u.dst.dev); - rth->idev = in_dev_get(rth->u.dst.dev); -+ rth->fl.fl4_gw = 0; - rth->rt_gateway = daddr; - rth->rt_spec_dst= spec_dst; - rth->u.dst.input= ip_local_deliver; -@@ -2086,8 +2108,9 @@ - goto e_inval; - } - --int ip_route_input(struct sk_buff *skb, __be32 daddr, __be32 saddr, -- u8 tos, struct net_device *dev) -+static inline int -+ip_route_input_cached(struct sk_buff *skb, __be32 daddr, __be32 saddr, -+ u8 tos, struct net_device *dev, u32 lsrc) - { - struct rtable * rth; - unsigned hash; -@@ -2102,6 +2125,7 @@ - if (rth->fl.fl4_dst == daddr && - rth->fl.fl4_src == saddr && - rth->fl.iif == iif && -+ rth->fl.fl4_lsrc == lsrc && - rth->fl.oif == 0 && - rth->fl.mark == skb->mark && - rth->fl.fl4_tos == tos) { -@@ -2148,7 +2172,19 @@ - rcu_read_unlock(); - return -EINVAL; - } -- return ip_route_input_slow(skb, daddr, saddr, tos, dev); -+ return ip_route_input_slow(skb, daddr, saddr, tos, dev, lsrc); -+} -+ -+int ip_route_input(struct sk_buff *skb, u32 daddr, u32 saddr, -+ u8 tos, struct net_device *dev) -+{ -+ return ip_route_input_cached(skb, daddr, saddr, tos, dev, 0); -+} -+ -+int ip_route_input_lookup(struct sk_buff *skb, u32 daddr, u32 saddr, -+ u8 tos, struct net_device *dev, u32 lsrc) -+{ -+ return ip_route_input_cached(skb, daddr, saddr, tos, dev, lsrc); - } - - static inline int __mkroute_output(struct rtable **result, -@@ -2227,6 +2263,7 @@ - rth->fl.fl4_tos = tos; - rth->fl.fl4_src = oldflp->fl4_src; - rth->fl.oif = oldflp->oif; -+ rth->fl.fl4_gw = oldflp->fl4_gw; - rth->fl.mark = oldflp->mark; - rth->rt_dst = fl->fl4_dst; - rth->rt_src = fl->fl4_src; -@@ -2367,6 +2404,7 @@ - struct flowi fl = { .nl_u = { .ip4_u = - { .daddr = oldflp->fl4_dst, - .saddr = oldflp->fl4_src, -+ .gw = oldflp->fl4_gw, - .tos = tos & IPTOS_RT_MASK, - .scope = ((tos & RTO_ONLINK) ? - RT_SCOPE_LINK : -@@ -2470,6 +2508,7 @@ - dev_out = &loopback_dev; - dev_hold(dev_out); - fl.oif = loopback_dev.ifindex; -+ fl.fl4_gw = 0; - res.type = RTN_LOCAL; - flags |= RTCF_LOCAL; - goto make_route; -@@ -2477,7 +2516,7 @@ - - if (fib_lookup(&fl, &res)) { - res.fi = NULL; -- if (oldflp->oif) { -+ if (oldflp->oif && dev_out->flags & IFF_UP) { - /* Apparently, routing tables are wrong. Assume, - that the destination is on link. - -@@ -2517,6 +2556,7 @@ - dev_out = &loopback_dev; - dev_hold(dev_out); - fl.oif = dev_out->ifindex; -+ fl.fl4_gw = 0; - if (res.fi) - fib_info_put(res.fi); - res.fi = NULL; -@@ -2524,13 +2564,12 @@ - goto make_route; - } - -+ if (res.type == RTN_UNICAST) -+ fib_select_default(&fl, &res); - #ifdef CONFIG_IP_ROUTE_MULTIPATH -- if (res.fi->fib_nhs > 1 && fl.oif == 0) -+ if (res.fi->fib_nhs > 1) - fib_select_multipath(&fl, &res); -- else - #endif -- if (!res.prefixlen && res.type == RTN_UNICAST && !fl.oif) -- fib_select_default(&fl, &res); - - if (!fl.fl4_src) - fl.fl4_src = FIB_RES_PREFSRC(res); -@@ -2567,6 +2606,7 @@ - rth->fl.fl4_src == flp->fl4_src && - rth->fl.iif == 0 && - rth->fl.oif == flp->oif && -+ rth->fl.fl4_gw == flp->fl4_gw && - rth->fl.mark == flp->mark && - !((rth->fl.fl4_tos ^ flp->fl4_tos) & - (IPTOS_RT_MASK | RTO_ONLINK))) { -@@ -3267,3 +3307,4 @@ - EXPORT_SYMBOL(__ip_select_ident); - EXPORT_SYMBOL(ip_route_input); - EXPORT_SYMBOL(ip_route_output_key); -+EXPORT_SYMBOL(ip_route_input_lookup); diff --git a/target/linux/generic-2.6/patches-2.6.22/202-mips-freestanding.patch b/target/linux/generic-2.6/patches-2.6.22/202-mips-freestanding.patch deleted file mode 100644 index 2ae8bc5..0000000 --- a/target/linux/generic-2.6/patches-2.6.22/202-mips-freestanding.patch +++ /dev/null @@ -1,12 +0,0 @@ ---- a/arch/mips/Makefile -+++ b/arch/mips/Makefile -@@ -589,6 +589,9 @@ - core-$(CONFIG_TOSHIBA_RBTX4938) += arch/mips/tx4938/common/ - load-$(CONFIG_TOSHIBA_RBTX4938) += 0xffffffff80100000 - -+# temporary until string.h is fixed -+cflags-y += -ffreestanding -+ - cflags-y += -Iinclude/asm-mips/mach-generic - drivers-$(CONFIG_PCI) += arch/mips/pci/ - diff --git a/target/linux/generic-2.6/patches-2.6.22/204-jffs2_eofdetect.patch b/target/linux/generic-2.6/patches-2.6.22/204-jffs2_eofdetect.patch deleted file mode 100644 index 5053401..0000000 --- a/target/linux/generic-2.6/patches-2.6.22/204-jffs2_eofdetect.patch +++ /dev/null @@ -1,56 +0,0 @@ ---- a/fs/jffs2/build.c -+++ b/fs/jffs2/build.c -@@ -105,6 +105,17 @@ - dbg_fsbuild("scanned flash completely\n"); - jffs2_dbg_dump_block_lists_nolock(c); - -+ if (c->flags & (1 << 7)) { -+ printk("%s(): unlocking the mtd device... ", __func__); -+ if (c->mtd->unlock) -+ c->mtd->unlock(c->mtd, 0, c->mtd->size); -+ printk("done.\n"); -+ -+ printk("%s(): erasing all blocks after the end marker... ", __func__); -+ jffs2_erase_pending_blocks(c, -1); -+ printk("done.\n"); -+ } -+ - dbg_fsbuild("pass 1 starting\n"); - c->flags |= JFFS2_SB_FLAG_BUILDING; - /* Now scan the directory tree, increasing nlink according to every dirent found. */ ---- a/fs/jffs2/scan.c -+++ b/fs/jffs2/scan.c -@@ -142,9 +142,12 @@ - - /* reset summary info for next eraseblock scan */ - jffs2_sum_reset_collected(s); -- -- ret = jffs2_scan_eraseblock(c, jeb, buf_size?flashbuf:(flashbuf+jeb->offset), -- buf_size, s); -+ -+ if (c->flags & (1 << 7)) -+ ret = BLK_STATE_ALLFF; -+ else -+ ret = jffs2_scan_eraseblock(c, jeb, buf_size?flashbuf:(flashbuf+jeb->offset), -+ buf_size, s); - - if (ret < 0) - goto out; -@@ -545,6 +548,17 @@ - return err; - } - -+ if ((buf[0] == 0xde) && -+ (buf[1] == 0xad) && -+ (buf[2] == 0xc0) && -+ (buf[3] == 0xde)) { -+ /* end of filesystem. erase everything after this point */ -+ printk("%s(): End of filesystem marker found at 0x%x\n", __func__, jeb->offset); -+ c->flags |= (1 << 7); -+ -+ return BLK_STATE_ALLFF; -+ } -+ - /* We temporarily use 'ofs' as a pointer into the buffer/jeb */ - ofs = 0; - diff --git a/target/linux/generic-2.6/patches-2.6.22/207-powerpc_asm_segment_h.patch b/target/linux/generic-2.6/patches-2.6.22/207-powerpc_asm_segment_h.patch deleted file mode 100644 index 1272e82..0000000 --- a/target/linux/generic-2.6/patches-2.6.22/207-powerpc_asm_segment_h.patch +++ /dev/null @@ -1,9 +0,0 @@ ---- /dev/null -+++ b/include/asm-powerpc/segment.h -@@ -0,0 +1,6 @@ -+#ifndef _ASM_SEGMENT_H -+#define _ASM_SEGMENT_H -+ -+/* Only here because we have some old header files that expect it.. */ -+ -+#endif /* _ASM_SEGMENT_H */ diff --git a/target/linux/generic-2.6/patches-2.6.22/208-rtl8110sb_fix.patch b/target/linux/generic-2.6/patches-2.6.22/208-rtl8110sb_fix.patch deleted file mode 100644 index a5a1e4c..0000000 --- a/target/linux/generic-2.6/patches-2.6.22/208-rtl8110sb_fix.patch +++ /dev/null @@ -1,24 +0,0 @@ ---- a/drivers/net/r8169.c -+++ b/drivers/net/r8169.c -@@ -494,7 +494,7 @@ - #endif - - static const u16 rtl8169_intr_mask = -- SYSErr | LinkChg | RxOverflow | RxFIFOOver | TxErr | TxOK | RxErr | RxOK; -+ LinkChg | RxOverflow | RxFIFOOver | TxErr | TxOK | RxErr | RxOK; - static const u16 rtl8169_napi_event = - RxOK | RxOverflow | RxFIFOOver | TxOK | TxErr; - static const unsigned int rtl8169_rx_config = -@@ -2640,10 +2640,12 @@ - if (!(status & rtl8169_intr_mask)) - break; - -+#if 0 - if (unlikely(status & SYSErr)) { - rtl8169_pcierr_interrupt(dev); - break; - } -+#endif - - if (status & LinkChg) - rtl8169_check_link_status(dev, tp, ioaddr); diff --git a/target/linux/generic-2.6/patches-2.6.22/209-mini_fo.patch b/target/linux/generic-2.6/patches-2.6.22/209-mini_fo.patch deleted file mode 100644 index a124936..0000000 --- a/target/linux/generic-2.6/patches-2.6.22/209-mini_fo.patch +++ /dev/null @@ -1,7779 +0,0 @@ ---- a/fs/Kconfig -+++ b/fs/Kconfig -@@ -461,6 +461,9 @@ - This option will enlarge your kernel, but it allows debugging of - ocfs2 filesystem issues. - -+config MINI_FO -+ tristate "Mini fanout overlay filesystem" -+ - config MINIX_FS - tristate "Minix fs support" - help ---- a/fs/Makefile -+++ b/fs/Makefile -@@ -76,6 +76,7 @@ - obj-$(CONFIG_RAMFS) += ramfs/ - obj-$(CONFIG_HUGETLBFS) += hugetlbfs/ - obj-$(CONFIG_CODA_FS) += coda/ -+obj-$(CONFIG_MINI_FO) += mini_fo/ - obj-$(CONFIG_MINIX_FS) += minix/ - obj-$(CONFIG_FAT_FS) += fat/ - obj-$(CONFIG_MSDOS_FS) += msdos/ ---- /dev/null -+++ b/fs/mini_fo/aux.c -@@ -0,0 +1,580 @@ -+/* -+ * Copyright (c) 1997-2003 Erez Zadok -+ * Copyright (c) 2001-2003 Stony Brook University -+ * -+ * For specific licensing information, see the COPYING file distributed with -+ * this package, or get one from ftp://ftp.filesystems.org/pub/fist/COPYING. -+ * -+ * This Copyright notice must be kept intact and distributed with all -+ * fistgen sources INCLUDING sources generated by fistgen. -+ */ -+/* -+ * Copyright (C) 2004, 2005 Markus Klotzbuecher <mk@creamnet.de> -+ * -+ * This program is free software; you can redistribute it and/or -+ * modify it under the terms of the GNU General Public License -+ * as published by the Free Software Foundation; either version -+ * 2 of the License, or (at your option) any later version. -+ */ -+/* -+ * $Id$ -+ */ -+ -+#ifdef HAVE_CONFIG_H -+# include <config.h> -+#endif -+ -+#include "fist.h" -+#include "mini_fo.h" -+ -+/* check if file exists in storage */ -+int exists_in_storage(dentry_t *dentry) -+{ -+ check_mini_fo_dentry(dentry); -+ if(dtost(dentry) == MODIFIED || dtost(dentry) == CREATED || dtost(dentry) == DEL_REWRITTEN) -+ return 1; -+ return 0; -+} -+ -+/* check if dentry is in an existing state */ -+int is_mini_fo_existant(dentry_t *dentry) -+{ -+ check_mini_fo_dentry(dentry); -+ -+ if(dtost(dentry) == DELETED || dtost(dentry) == NON_EXISTANT) -+ return 0; -+ else -+ return 1; -+} -+ -+/* -+ * This function will create a negative storage dentry for -+ * dentry, what is required for many create like options. -+ * It will create the storage structure if necessary. -+ */ -+int get_neg_sto_dentry(dentry_t *dentry) -+{ -+ int err = 0; -+ unsigned int len; -+ const unsigned char *name; -+ -+ if(!dentry || -+ !dtopd(dentry) || -+ !(dtost(dentry) == UNMODIFIED || -+ dtost(dentry) == NON_EXISTANT || -+ dtost(dentry) == DELETED)) { -+ printk(KERN_CRIT "mini_fo: get_neg_sto_dentry: invalid dentry passed.\n"); -+ err = -1; -+ goto out; -+ } -+ /* Have we got a neg. dentry already? */ -+ if(dtohd2(dentry)) { -+ err = 0; -+ goto out; -+ } -+ if(dtost(dentry->d_parent) == UNMODIFIED) { -+ /* build sto struct */ -+ err = build_sto_structure(dentry->d_parent->d_parent, dentry->d_parent); -+ if(err || -+ dtost(dentry->d_parent) != MODIFIED) { -+ printk(KERN_CRIT "mini_fo: get_neg_sto_dentry: ERROR building sto structure.\n"); -+ err = -1; -+ goto out; -+ } -+ } -+ -+ len = dentry->d_name.len; -+ name = dentry->d_name.name; -+ -+ dtohd2(dentry) = -+ lookup_one_len(name, dtohd2(dentry->d_parent), len); -+ -+ out: -+ return err; -+} -+ -+int check_mini_fo_dentry(dentry_t *dentry) -+{ -+ ASSERT(dentry != NULL); -+ ASSERT(dtopd(dentry) != NULL); -+ ASSERT((dtohd(dentry) != NULL) || (dtohd2(dentry) != NULL)); -+ -+/* if(dtost(dentry) == MODIFIED) { */ -+/* ASSERT(dentry->d_inode != NULL); */ -+/* ASSERT(dtohd(dentry) != NULL); */ -+/* ASSERT(dtohd(dentry)->d_inode != NULL); */ -+/* ASSERT(dtohd2(dentry) != NULL); */ -+/* ASSERT(dtohd2(dentry)->d_inode != NULL); */ -+/* } */ -+/* else if(dtost(dentry) == UNMODIFIED) { */ -+/* ASSERT(dentry->d_inode != NULL); */ -+/* ASSERT( */ -+/* } */ -+ return 0; -+} -+ -+int check_mini_fo_file(file_t *file) -+{ -+ ASSERT(file != NULL); -+ ASSERT(ftopd(file) != NULL); -+ ASSERT(file->f_dentry != NULL); -+ -+ /* violent checking, check depending of state and type -+ * if(S_ISDIR(file->f_dentry->d_inode->i_mode)) {} -+ */ -+ ASSERT((ftohf(file) != NULL) || (ftohf2(file) != NULL)); -+ return 0; -+} -+ -+int check_mini_fo_inode(inode_t *inode) -+{ -+ ASSERT(inode != NULL); -+ ASSERT(itopd(inode) != NULL); -+ ASSERT((itohi(inode) != NULL) || (itohi2(inode) != NULL)); -+ return 0; -+} -+ -+/* -+ * will walk a base path as provided by get_mini_fo_bpath and return -+ * the (hopefully ;-) ) positive dentry of the renamed base dir. -+ * -+ * This does some work of path_init. -+ */ -+dentry_t *bpath_walk(super_block_t *sb, char *bpath) -+{ -+ int err; -+ struct nameidata nd; -+ -+ /* be paranoid */ -+ if(!bpath || bpath[0] != '/') { -+ printk(KERN_CRIT "mini_fo: bpath_walk: Invalid string.\n"); -+ return NULL; -+ } -+ if(!sb || !stopd(sb)) { -+ printk(KERN_CRIT "mini_fo: bpath_walk: Invalid sb.\n"); -+ return NULL; -+ } -+ -+ /* setup nd as path_init does */ -+ nd.last_type = LAST_ROOT; /* if there are only slashes... */ -+ nd.flags = LOOKUP_FOLLOW; -+ /* fix this: how do I reach this lock? -+ * read_lock(¤t->fs->lock); */ -+ nd.mnt = mntget(stopd(sb)->hidden_mnt); -+ nd.dentry = dget(stopd(sb)->base_dir_dentry); -+ /* read_unlock(¤t->fs->lock); */ -+ -+ err = path_walk(bpath+1, &nd); -+ -+ /* validate */ -+ if (err || !nd.dentry || !nd.dentry->d_inode) { -+ printk(KERN_CRIT "mini_fo: bpath_walk: path_walk failed.\n"); -+ return NULL; -+ } -+ return nd.dentry; -+} -+ -+ -+/* returns the full path of the basefile incl. its name */ -+int get_mini_fo_bpath(dentry_t *dentry, char **bpath, int *bpath_len) -+{ -+ char *buf_walker; -+ int len = 0; -+ dentry_t *sky_walker; -+ -+ if(!dentry || !dtohd(dentry)) { -+ printk(KERN_CRIT "mini_fo: get_mini_fo_bpath: invalid dentry passed.\n"); -+ return -1; -+ } -+ sky_walker = dtohd(dentry); -+ -+ do { -+ len += sky_walker->d_name.len + 1 ; /* 1 for '/' */ -+ sky_walker = sky_walker->d_parent; -+ } while(sky_walker != stopd(dentry->d_inode->i_sb)->base_dir_dentry); -+ -+ /* 1 to oil the loop */ -+ *bpath = (char*) kmalloc(len + 1, GFP_KERNEL); -+ if(!*bpath) { -+ printk(KERN_CRIT "mini_fo: get_mini_fo_bpath: out of mem.\n"); -+ return -1; -+ } -+ buf_walker = *bpath+len; /* put it on last char */ -+ *buf_walker = '\n'; -+ sky_walker = dtohd(dentry); -+ -+ do { -+ buf_walker -= sky_walker->d_name.len; -+ strncpy(buf_walker, -+ sky_walker->d_name.name, -+ sky_walker->d_name.len); -+ *(--buf_walker) = '/'; -+ sky_walker = sky_walker->d_parent; -+ } while(sky_walker != stopd(dentry->d_inode->i_sb)->base_dir_dentry); -+ -+ /* bpath_len doesn't count newline! */ -+ *bpath_len = len; -+ return 0; -+} -+ -+int mini_fo_cp_cont(dentry_t *tgt_dentry, struct vfsmount *tgt_mnt, -+ dentry_t *src_dentry, struct vfsmount *src_mnt) -+{ -+ void *buf; -+ mm_segment_t old_fs; -+ file_t *tgt_file; -+ file_t *src_file; -+ int bytes, len, tmp, err; -+ err = 0; -+ -+ if(!(tgt_dentry->d_inode && src_dentry->d_inode)) { -+ printk(KERN_CRIT "mini_fo_cp_cont: ERROR, neg. dentry passed.\n"); -+ err = -EINVAL; -+ goto out; -+ } -+ -+ dget(tgt_dentry); -+ dget(src_dentry); -+ mntget(tgt_mnt); -+ mntget(src_mnt); -+ -+ /* open file write only */ -+ tgt_file = dentry_open(tgt_dentry, tgt_mnt, 0x1); -+ if(!tgt_file || IS_ERR(tgt_file)) { -+ printk(KERN_CRIT "mini_fo_cp_cont: ERROR opening target file.\n"); -+ err = PTR_ERR(tgt_file); -+ goto out_err; -+ } -+ -+ /* open file read only */ -+ src_file = dentry_open(src_dentry, src_mnt, 0x0); -+ if(!src_file || IS_ERR(src_file)) { -+ printk(KERN_CRIT "mini_fo_cp_cont: ERROR opening source file.\n"); -+ err = PTR_ERR(src_file); -+ -+ /* close target file */ -+ fput(tgt_file); -+ goto out_err; -+ } -+ -+ /* check if the filesystem(s) support read respective write */ -+ if(!src_file->f_op->read || !tgt_file->f_op->write) { -+ printk(KERN_CRIT "mini_fo_cp_cont: ERROR, no fs read or write support.\n"); -+ err = -EPERM; -+ goto out_close; -+ } -+ -+ /* allocate a page for transfering the data */ -+ buf = (void *) __get_free_page(GFP_KERNEL); -+ if(!buf) { -+ printk(KERN_CRIT "mini_fo_cp_cont: ERROR, out of kernel mem.\n"); -+ goto out_err; -+ } -+ -+ tgt_file->f_pos = 0; -+ src_file->f_pos = 0; -+ -+ old_fs = get_fs(); -+ set_fs(KERNEL_DS); -+ -+ /* Doing this I assume that a read operation will return a full -+ * buffer while there is still data to read, and a less than -+ * full buffer when all data has been read. -+ */ -+ bytes = len = PAGE_SIZE; -+ while(bytes == len) { -+ bytes = src_file->f_op->read(src_file, buf, len, -+ &src_file->f_pos); -+ tmp = tgt_file->f_op->write(tgt_file, buf, bytes, -+ &tgt_file->f_pos); -+ if(tmp != bytes) { -+ printk(KERN_CRIT "mini_fo_cp_cont: ERROR writing.\n"); -+ goto out_close_unset; -+ } -+ } -+ -+ free_page((unsigned long) buf); -+ set_fs(old_fs); -+ fput(tgt_file); -+ fput(src_file); -+ goto out; -+ -+ out_close_unset: -+ free_page((unsigned long) buf); -+ set_fs(old_fs); -+ -+ out_close: -+ fput(tgt_file); -+ fput(src_file); -+ -+ out_err: -+ dput(tgt_dentry); -+ dput(src_dentry); -+ -+ /* mk: not sure if this need to be done */ -+ mntput(tgt_mnt); -+ mntput(src_mnt); -+ -+ out: -+ return err; -+} -+ -+/* mk: -+ * ndl (no-duplicate list) stuff -+ * This is used in mini_fo_readdir, to save the storage directory contents -+ * and later when reading base, match them against the list in order -+ * to avoid duplicates. -+ */ -+ -+/* add a file specified by name and len to the ndl -+ * Return values: 0 on success, <0 on failure. -+ */ -+int ndl_add_entry(struct readdir_data *rd, const char *name, int len) -+{ -+ struct ndl_entry *tmp_entry; -+ -+ tmp_entry = (struct ndl_entry *) -+ kmalloc(sizeof(struct ndl_entry), GFP_KERNEL); -+ if(!tmp_entry) { -+ printk(KERN_CRIT "mini_fo: ndl_add_entry: out of mem.\n"); -+ return -ENOMEM; -+ } -+ tmp_entry->name = (char*) kmalloc(len, GFP_KERNEL); -+ if(!tmp_entry->name) { -+ printk(KERN_CRIT "mini_fo: ndl_add_entry: out of mem.\n"); -+ return -ENOMEM; -+ } -+ strncpy(tmp_entry->name, name, len); -+ tmp_entry->len = len; -+ -+ list_add(&tmp_entry->list, &rd->ndl_list); -+ rd->ndl_size++; -+ return 0; -+} -+ -+/* delete all list entries and free memory */ -+void ndl_put_list(struct readdir_data *rd) -+{ -+ struct list_head *tmp; -+ struct ndl_entry *tmp_entry; -+ -+ if(rd->ndl_size <= 0) -+ return; -+ while(!list_empty(&rd->ndl_list)) { -+ tmp = rd->ndl_list.next; -+ list_del(tmp); -+ tmp_entry = list_entry(tmp, struct ndl_entry, list); -+ kfree(tmp_entry->name); -+ kfree(tmp_entry); -+ } -+ rd->ndl_size = 0; -+} -+ -+/* Check if a file specified by name and len is in the ndl -+ * Return value: 0 if not in list, 1 if file is found in ndl. -+ */ -+int ndl_check_entry(struct readdir_data *rd, const char *name, int len) -+{ -+ struct list_head *tmp; -+ struct ndl_entry *tmp_entry; -+ -+ if(rd->ndl_size <= 0) -+ return 0; -+ -+ list_for_each(tmp, &rd->ndl_list) { -+ tmp_entry = list_entry(tmp, struct ndl_entry, list); -+ if(tmp_entry->len != len) -+ continue; -+ if(!strncmp(tmp_entry->name, name, len)) -+ return 1; -+ } -+ return 0; -+} -+ -+/* mk: -+ * Recursive function to create corresponding directorys in the storage fs. -+ * The function will build the storage directorys up to dentry. -+ */ -+int build_sto_structure(dentry_t *dir, dentry_t *dentry) -+{ -+ int err; -+ dentry_t *hidden_sto_dentry; -+ dentry_t *hidden_sto_dir_dentry; -+ -+ if(dentry->d_parent != dir) { -+ printk(KERN_CRIT "mini_fo: build_sto_structure: invalid parameter or meta data corruption [1].\n"); -+ return 1; -+ } -+ -+ if(dtost(dir) != MODIFIED) { -+ err = build_sto_structure(dir->d_parent, dentry->d_parent); -+ if(err) -+ return err; -+ } -+ -+ /* ok, coming back again. */ -+ check_mini_fo_dentry(dentry); -+ hidden_sto_dentry = dtohd2(dentry); -+ -+ if(!hidden_sto_dentry) { -+ /* -+ * This is the case after creating the first -+ * hidden_sto_dentry. -+ * After one negative storage_dentry, all pointers to -+ * hidden_storage dentries are set to NULL. We need to -+ * create the negative dentry before we create the storage -+ * file. -+ */ -+ unsigned int len; -+ const unsigned char *name; -+ len = dtohd(dentry)->d_name.len; -+ name = dtohd(dentry)->d_name.name; -+ hidden_sto_dentry = lookup_one_len(name, dtohd2(dir), len); -+ dtohd2(dentry) = hidden_sto_dentry; -+ } -+ -+ /* was: hidden_sto_dir_dentry = lock_parent(hidden_sto_dentry); */ -+ hidden_sto_dir_dentry = dget(hidden_sto_dentry->d_parent); -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,16) -+ mutex_lock(&hidden_sto_dir_dentry->d_inode->i_mutex); -+#else -+ down(&hidden_sto_dir_dentry->d_inode->i_sem); -+#endif -+ /* lets be safe */ -+ if(dtohd2(dir) != hidden_sto_dir_dentry) { -+ printk(KERN_CRIT "mini_fo: build_sto_structure: invalid parameter or meta data corruption [2].\n"); -+ return 1; -+ } -+ -+ /* check for errors in lock_parent */ -+ err = PTR_ERR(hidden_sto_dir_dentry); -+ if(IS_ERR(hidden_sto_dir_dentry)) { -+ printk(KERN_CRIT "mini_fo: build_sto_structure: lock_parent failed.\n"); -+ return err; -+ } -+ -+ err = vfs_mkdir(hidden_sto_dir_dentry->d_inode, -+ hidden_sto_dentry, -+ dir->d_inode->i_mode); -+ -+ if(err) { -+ printk(KERN_CRIT "mini_fo: build_sto_structure: failed to create storage dir [1].\n"); -+ /* was: unlock_dir(dir); */ -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,16) -+ mutex_unlock(&dir->d_inode->i_mutex); -+#else -+ up(&dir->d_inode->i_sem); -+#endif -+ dput(dir); -+ return err; -+ } -+ -+ /* everything ok! */ -+ if(!dtohd2(dentry)->d_inode) { -+ printk(KERN_CRIT "mini_fo: build_sto_structure: failed to create storage dir [2].\n"); -+ /* was: unlock_dir(dir); */ -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,16) -+ mutex_unlock(&dir->d_inode->i_mutex); -+#else -+ up(&dir->d_inode->i_sem); -+#endif -+ dput(dir); -+ return 1; -+ } -+ -+ /* interpose the new inode and set new state */ -+ itohi2(dentry->d_inode) = igrab(dtohd2(dentry)->d_inode); -+ dtopd(dentry)->state = MODIFIED; -+ -+ /* initalize the wol list */ -+ itopd(dentry->d_inode)->deleted_list_size = -1; -+ itopd(dentry->d_inode)->renamed_list_size = -1; -+ meta_build_lists(dentry); -+ -+ fist_copy_attr_all(dentry->d_inode, itohi2(dentry->d_inode)); -+ fist_copy_attr_timesizes(dir->d_inode, -+ hidden_sto_dir_dentry->d_inode); -+ dir->d_inode->i_nlink++; -+ /* was: unlock_dir(hidden_sto_dir_dentry); */ -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,16) -+ mutex_unlock(&hidden_sto_dir_dentry->d_inode->i_mutex); -+#else -+ up(&hidden_sto_dir_dentry->d_inode->i_sem); -+#endif -+ dput(hidden_sto_dir_dentry); -+ return 0; -+} -+ -+ -+#if 0 /* unused */ -+ -+/* -+ * Read "len" bytes from "filename" into "buf". -+ * "buf" is in kernel space. -+ */ -+int -+mini_fo_read_file(const char *filename, void *buf, int len) -+{ -+ file_t *filp; -+ mm_segment_t oldfs; -+ int bytes; -+ /* Chroot? Maybe NULL isn't right here */ -+ filp = filp_open(filename, O_RDONLY, 0); -+ if (!filp || IS_ERR(filp)) { -+ printk("mini_fo_read_file err %d\n", (int) PTR_ERR(filp)); -+ return -1; /* or do something else */ -+ } -+ -+ if (!filp->f_op->read) -+ return -2; /* file(system) doesn't allow reads */ -+ -+ /* now read len bytes from offset 0 */ -+ filp->f_pos = 0; /* start offset */ -+ oldfs = get_fs(); -+ set_fs(KERNEL_DS); -+ bytes = filp->f_op->read(filp, buf, len, &filp->f_pos); -+ set_fs(oldfs); -+ -+ /* close the file */ -+ fput(filp); -+ -+ return bytes; -+} -+ -+ -+ -+/* -+ * Write "len" bytes from "buf" to "filename" -+ * "buf" is in kernel space. -+ */ -+int -+mini_fo_write_file(const char *filename, void *buf, int len) -+{ -+ file_t *filp; -+ mm_segment_t oldfs; -+ int bytes; -+ /* Chroot? Maybe NULL isn't right here */ -+ filp = filp_open(filename, O_RDWR|O_CREAT, 0640); -+ if (!filp || IS_ERR(filp)) { -+ printk("mini_fo_write_file err %d\n", (int) PTR_ERR(filp)); -+ return -1; /* or do something else */ -+ } -+ -+ if (!filp->f_op->write) -+ return -2; /* file(system) doesn't allow writes */ -+ -+ /* now write len bytes from offset 0 */ -+ filp->f_pos = 0; /* start offset */ -+ oldfs = get_fs(); -+ set_fs(KERNEL_DS); -+ bytes = filp->f_op->write(filp, buf, len, &filp->f_pos); -+ set_fs(oldfs); -+ -+ /* close the file */ -+ fput(filp); -+ -+ return bytes; -+} -+ -+#endif /* unused */ -+ ---- /dev/null -+++ b/fs/mini_fo/ChangeLog -@@ -0,0 +1,281 @@ -+2006-01-24 Markus Klotzbuecher <mk@mary.denx.de> -+ -+ * Add tons of ugly ifdefs to Ed L. Cashin's mutex patch to -+ retain backwards compatibility. -+ -+2006-01-24 Ed L. Cashin <ecashin@coraid.com> -+ -+ * Support for the new mutex infrastructure -+ (7892f2f48d165a34b0b8130c8a195dfd807b8cb6) -+ -+2005-10-15 Markus Klotzbuecher <mk@localhost.localdomain> -+ -+ * Bugfix for a serious memory leak in mini_fo_follow_link. -+ -+2005-09-21 Markus Klotzbuecher <mk@mary> -+ -+ * new release 0.6.1 -+ -+ * fix of a compiler warning due to changes in 2.6.13 -+ -+2005-09-21 Klaus Wenninger <klaus.wenninger@siemens.com> -+ -+ * file.c: readdir: fix for a bug that caused directory entries -+ to show up twice when using storage filesystems such as -+ minixfs or pramfs. -+ -+2005-06-30 Eric Lammerts <eric@lammerts.org> -+ -+ * fix for an oops when overwriting a binary thats beeing -+ executed. -+ -+2005-06-09 <mk@mary> -+ -+ * Renamed overlay to mini_fo-overlay. -+ -+ * Added mini_fo-merge script to allow merging of storage and base -+ after making modifications. -+ -+2005-05-22 root <mk@mary> -+ -+ * Added overlay script that allows to easily mount mini_fo ontop -+ of a given base directory -+ -+2005-05-10 <mk@mary> -+ -+ * inode.c: xattr functions return -EOPNOSUPP instead of -+ -ENOSUPP, what confuses "ls -l" -+ -+ * Changed license from LGPL to GPL. -+ -+2005-05-08 root <mk@mary> -+ -+ * Makefile: clean it up and added make install and make -+ uninstall. -+ -+2005-05-06 <mk@mary> -+ -+ * merged devel branch back to main. [v0-6-0-pre3] -+ -+ * removed unused files print.c and fist_ioctl. [devel-0-0-18] -+ -+ * ioctl: removed fist_ioctl stuff, that is not needed for -+ now. -+ -+2005-05-03 <mk@mary> -+ -+ * file.c: simplified mini_fo_open and mini_fo_setattr using -+ new state changing functions. [devel-0-0-17] -+ -+ * inode.c: Fixed getattr state bug (see below) in 2.4 function -+ mini_fo_inode revalidate. -+ -+ * inode.c: found an other bug in mini_fo_getattr. States are not -+ reliable in this function, as a file can be opened, unlinked and -+ the getattr function called. This results in a deleted dentry -+ with an inode. Fix is to ignore states and simply use the inode -+ available. -+ -+2005-04-29 <mk@mary> -+ -+ * file.c: Bugfix and cleanup in fasync and fsync. [devel-0-0-16] -+ -+ * file.c: do not use mini_fo_lock so the generic version is -+ used (I guess). -+ -+ * inode.c: getattr, never call getattr on lower files, as this -+ will cause the inum to change. -+ -+ * inode.c: rename_reg_file renamed to rename_nondir, as it -+ doesn't matter as long it't not a dir. Removed all -+ rename_dev_file etc. -+ -+ * tagged as devel-0-0-15 -+ -+ * inode.c: added support for chosing support for extended -+ attrs at compile time by XATTR define in mini_fo.h . -+ -+ * inode.c: fixed mini_fo_getattr to use mini_fo inode and not -+ lower again, what avoids inode number changes that confused -+ rm again. This is the proper solution. -+ -+2005-04-24 <mk@mary> -+ -+ * all files: updated Copyright notive to 2005. [devel-0-0-14] -+ -+ * inode.c: fixed mini_fo_getattr to not change the inode -+ number, even if lower files change. -+ -+ * super.c: fixed a bug that caused deleted base file to show -+ up suddenly after some time, or after creating a special -+ file. The problem was that after some time or after special -+ file creating sync_sb_inodes is called by the vfs, that -+ called our mini_fo_put_inode. There was (wrongly) called -+ __meta_put_lists, that nuked the lists, although the inode -+ was going to continue its life. Moving __meta_put_lists to -+ mini_fo_clear_inode, where an inode is really destroyed, -+ solved the problem. -+ -+ -+2005-04-23 <mk@mary> -+ -+ * state.c, aux.c: more cleaning up and -+ simplifications. [devel-0-0-13] -+ -+ * inode.c: implemented mini_fo_getattr, that was required for -+ 2.6 because inode_revalidate has been remove there, and the -+ old "du" bug returned. -+ -+ -+2005-04-20 <mk@mary> -+ -+ * aux.c: get_neg_sto_dentry(): allow to be called for dentries -+ in state UNMODIFIED, NON_EXISTANT _and_ DELETED. -+ -+2005-04-19 <mk@mary> -+ -+ * Fixed a bug under 2.6 that caused files deleted via mini_fo -+ not to be deleted properly and therefore the fs filled up -+ untill no memory was left. [devel-0-0-12] -+ -+ * Added basic hard link support. This means that creating -+ hardlinks will work, but existing ones will be treated as -+ individual files. [devel-0-0-11] -+ -+2005-04-17 <mk@mary> -+ -+ * Bugfixes -+ -+2005-04-13 root <mk@mary> -+ -+ * Added file state.c for the state transition -+ functions. Doesn't work very well yet, though... -+ -+2005-04-12 <mk@mary> -+ -+ * Porting to 2.6 started, which is easier than expected, also -+ due to Olivier previous work. -+ -+2005-04-08 <mk@mary> -+ -+ * Fixed the bug that caused du to return invalid sizes of -+ directory trees. The problem was that -+ mini_fo_inode_revalidate didn't always copy the attributes -+ from the base inode properly. -+ -+2005-04-01 Markus Klotzbuecher <mk@chasey> -+ -+ * Merged devel branch back to main trunk and updated the -+ RELEASE notes. This will be 0-6-0-pre1. -+ -+2005-03-31 Markus Klotzbuecher <mk@chasey> -+ -+ * Fixed some bugs in rename_reg_file, that only showed up in -+ the kernel compile test. Kernel compiles cleanly ontop of -+ mini_fo, now also make mrproper etc. work. Seems pretty stable. -+ -+2005-03-28 Markus Klotzbuecher <mk@chasey> -+ -+ * Many, many directory renaming bugfixes and a lot of other -+ cleanup. Dir renaming seems to work relatively stable. -+ -+2005-03-22 Markus Klotzbuecher <mk@chasey> -+ -+ * Finished implementing lightweight directory renaming. Some -+ basic testing indicates it works fine. -+ Next is to implement testcases for the testsuite and confirm -+ everything is really working ok. -+ -+2005-03-18 Markus Klotzbuecher <mk@chasey> -+ -+ * Finished implementing meta.c stuff required for directory -+ renaming. -+ -+2005-03-17 Markus Klotzbuecher <mk@chasey> -+ -+ * Fixed all compile warnings + an extremly old bug that -+ somehow crept in while reworking the wol stuff to the META -+ system. Turning on -Werror again... :-) -+ -+ * Fixed some bugs in the new rename_reg_file function. -+ -+ * Rewrote mini_fo rename and split it into several -+ subfunctions, that handle the different types -+ seperately. Rewrote the regular file function aswell, as it -+ was implemented somewhat inefficient. -+ -+2005-03-16 Markus Klotzbuecher <mk@chasey> -+ -+ * Implemented new META subsystem, removed old WOL stuff in favor -+ if it. -+ -+ * After some basic testing everything seems ok... -+ -+2005-03-11 Markus Klotzbuecher <mk@chasey> -+ -+ * Renaming a non regular file caused trouble because I always -+ tried to copy the contents. Now I only do this for regular -+ files. mini_fo_rename still isn't implemented properly, renaming -+ of device files, symlinks etc. results in a empty regular file -+ instead of the proper type. -+ -+ * Directory renaming suddenly works! What a surprise! I guess -+ this is because renaming is implemented as making a copy and -+ removing the original. Still this might not work -+ everywhere... -+ -+2005-03-09 Markus Klotzbuecher <mk@chasey> -+ -+ * Bugfix, when a mini_fo directory that exists in storage -+ (state: MODIFIED, CREATED and DEL_REWRITTEN) is deleted, a -+ possibly existing WOL file contained in it needs to be -+ deleted too. -+ -+ * Starting cleanup: defined state names in order to get rid of -+ the state numbers. -+ -+2005-03-08 Markus Klotzbuecher <mk@chasey> -+ -+ * Makefile fix, fist_ioctl was built against wrong sources if ARCH=um -+ -+ * Fixed a bug in dentry.c, mini_fo_d_hash. In state 4 = -+ DEL_REWRITTEN the hash was calculated from the base dentry, -+ which was wrong and and caused assertions in -+ __mini_fo_hidden_dentry to fail. -+ -+2005-02-21 <mk@mary> -+ -+ * Implemented directory deleting (inode.c) -+ -+ * main.c: made mini_fo_parse_options a little more robust. -+ -+2004-12-22 <mk@mary> -+ -+ * Makefile cleanup and uml stuff, removed unneccessary files -+ -+ * Created a new and hopefully more informative README -+ -+ * CHANGELOG: created a new CHANGELOG and added old entries reversely -+ -+ -+2004-10-24 Gleb Natapov <gleb@nbase.co.il> -+ -+ * Fix: owner and group where not correctly copied from base to -+ storage. -+ -+ -+2004-10-05 Gleb Natapov <gleb@nbase.co.il> -+ -+ * Implementation of fsync, fasync and lock mini_fo functions. -+ -+ -+2004-09-29 Bob Lee <bob@pantasys.com> -+ -+ * Fix of a serious pointer bug -+ -+ -+2004-09-28 Gleb Natapov <gleb@nbase.co.il> -+ -+ * Implementation of mini_fo_mknod and mini_fo_rename, support -+ for device files. -+ ---- /dev/null -+++ b/fs/mini_fo/dentry.c -@@ -0,0 +1,244 @@ -+/* -+ * Copyright (c) 1997-2003 Erez Zadok -+ * Copyright (c) 2001-2003 Stony Brook University -+ * -+ * For specific licensing information, see the COPYING file distributed with -+ * this package, or get one from ftp://ftp.filesystems.org/pub/fist/COPYING. -+ * -+ * This Copyright notice must be kept intact and distributed with all -+ * fistgen sources INCLUDING sources generated by fistgen. -+ */ -+/* -+ * Copyright (C) 2004, 2005 Markus Klotzbuecher <mk@creamnet.de> -+ * -+ * This program is free software; you can redistribute it and/or -+ * modify it under the terms of the GNU General Public License -+ * as published by the Free Software Foundation; either version -+ * 2 of the License, or (at your option) any later version. -+ */ -+ -+/* -+ * $Id$ -+ */ -+ -+#ifdef HAVE_CONFIG_H -+# include <config.h> -+#endif -+ -+#include "fist.h" -+#include "mini_fo.h" -+ -+/* -+ * THIS IS A BOOLEAN FUNCTION: returns 1 if valid, 0 otherwise. -+ */ -+STATIC int -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,0) -+mini_fo_d_revalidate(dentry_t *dentry, struct nameidata *nd) -+#else -+mini_fo_d_revalidate(dentry_t *dentry, int flags) -+#endif -+{ -+ int err1 = 1; /* valid = 1, invalid = 0 */ -+ int err2 = 1; -+ dentry_t *hidden_dentry; -+ dentry_t *hidden_sto_dentry; -+ -+ -+ check_mini_fo_dentry(dentry); -+ -+ hidden_dentry = dtohd(dentry); -+ hidden_sto_dentry = dtohd2(dentry); -+ -+ if(hidden_dentry && -+ hidden_dentry->d_op && -+ hidden_dentry->d_op->d_revalidate) { -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,0) -+ err1 = hidden_dentry->d_op->d_revalidate(hidden_dentry, nd); -+#else -+ err1 = hidden_dentry->d_op->d_revalidate(hidden_dentry, flags); -+#endif -+ } -+ if(hidden_sto_dentry && -+ hidden_sto_dentry->d_op && -+ hidden_sto_dentry->d_op->d_revalidate) { -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,0) -+ err2 = hidden_sto_dentry->d_op->d_revalidate(hidden_sto_dentry, -+ nd); -+#else -+ err2 = hidden_sto_dentry->d_op->d_revalidate(hidden_sto_dentry, -+ flags); -+#endif -+ } -+ -+ /* mk: if one of the lower level dentries are valid, -+ * the mini_fo dentry is too. -+ */ -+ return (err1 || err2); -+} -+ -+ -+STATIC int -+mini_fo_d_hash(dentry_t *dentry, qstr_t *name) -+{ -+ int err = 0; -+ dentry_t *hidden_dentry; -+ dentry_t *hidden_sto_dentry; -+ -+ /* hidden_dentry = mini_fo_hidden_dentry(dentry); -+ * hidden_sto_dentry = mini_fo_hidden_sto_dentry(dentry); */ -+ -+ /* state 1, 3, 4, 5: build the hash for the storage dentry */ -+ if((dtopd(dentry)->state == MODIFIED) || -+ (dtopd(dentry)->state == CREATED) || -+ (dtopd(dentry)->state == DEL_REWRITTEN) || -+ (dtopd(dentry)->state == DELETED)) { -+ hidden_sto_dentry = dtohd2(dentry); -+ if(hidden_sto_dentry && -+ hidden_sto_dentry->d_op && -+ hidden_sto_dentry->d_op->d_hash) { -+ err = hidden_sto_dentry->d_op->d_hash(hidden_sto_dentry, name); -+ } -+ goto out; -+ } -+ /* state 2: build the hash for the base dentry */ -+ if(dtopd(dentry)->state == UNMODIFIED) { -+ hidden_dentry = dtohd(dentry); -+ if(hidden_dentry && -+ hidden_dentry->d_op && -+ hidden_dentry->d_op->d_hash) { -+ err = hidden_dentry->d_op->d_hash(hidden_dentry, name); -+ } -+ goto out; -+ } -+ /* state 6: build hash for the dentry that exists */ -+ if(dtopd(dentry)->state == NON_EXISTANT) { -+ hidden_sto_dentry = dtohd2(dentry); -+ if(hidden_sto_dentry && -+ hidden_sto_dentry->d_op && -+ hidden_sto_dentry->d_op->d_hash) { -+ err = hidden_sto_dentry->d_op->d_hash(hidden_sto_dentry, name); -+ goto out; -+ } -+ hidden_dentry = dtohd(dentry); -+ if(hidden_dentry && -+ hidden_dentry->d_op && -+ hidden_dentry->d_op->d_hash) { -+ err = hidden_dentry->d_op->d_hash(hidden_dentry, name); -+ goto out; -+ } -+ } -+ -+ printk(KERN_CRIT "mini_fo: d_hash: invalid state detected.\n"); -+ -+ out: -+ return err; -+} -+ -+ -+STATIC int -+mini_fo_d_compare(dentry_t *dentry, qstr_t *a, qstr_t *b) -+{ -+ int err; -+ dentry_t *hidden_dentry=NULL; -+ -+ /* hidden_dentry = mini_fo_hidden_dentry(dentry); */ -+ if(dtohd2(dentry)) -+ hidden_dentry = dtohd2(dentry); -+ else if(dtohd(dentry)) -+ hidden_dentry = dtohd(dentry); -+ -+ if (hidden_dentry && hidden_dentry->d_op && hidden_dentry->d_op->d_compare) { -+ err = hidden_dentry->d_op->d_compare(hidden_dentry, a, b); -+ } else { -+ err = ((a->len != b->len) || memcmp(a->name, b->name, b->len)); -+ } -+ -+ return err; -+} -+ -+ -+int -+mini_fo_d_delete(dentry_t *dentry) -+{ -+ dentry_t *hidden_dentry; -+ dentry_t *hidden_sto_dentry; -+ int err = 0; -+ -+ /* this could be a negative dentry, so check first */ -+ if (!dtopd(dentry)) { -+ printk(KERN_CRIT "mini_fo_d_delete: negative dentry passed.\n"); -+ goto out; -+ } -+ hidden_dentry = dtohd(dentry); -+ hidden_sto_dentry = dtohd2(dentry); -+ -+ if(hidden_dentry) { -+ if(hidden_dentry->d_op && -+ hidden_dentry->d_op->d_delete) { -+ err = hidden_dentry->d_op->d_delete(hidden_dentry); -+ } -+ } -+ if(hidden_sto_dentry) { -+ if(hidden_sto_dentry->d_op && -+ hidden_sto_dentry->d_op->d_delete) { -+ err = hidden_sto_dentry->d_op->d_delete(hidden_sto_dentry); -+ } -+ } -+ -+ out: -+ return err; -+} -+ -+ -+void -+mini_fo_d_release(dentry_t *dentry) -+{ -+ dentry_t *hidden_dentry; -+ dentry_t *hidden_sto_dentry; -+ -+ -+ /* this could be a negative dentry, so check first */ -+ if (!dtopd(dentry)) { -+ printk(KERN_CRIT "mini_fo_d_release: no private data.\n"); -+ goto out; -+ } -+ hidden_dentry = dtohd(dentry); -+ hidden_sto_dentry = dtohd2(dentry); -+ -+ if(hidden_dentry) { -+ /* decrement hidden dentry's counter and free its inode */ -+ dput(hidden_dentry); -+ } -+ if(hidden_sto_dentry) { -+ /* decrement hidden dentry's counter and free its inode */ -+ dput(hidden_sto_dentry); -+ } -+ -+ /* free private data (mini_fo_dentry_info) here */ -+ kfree(dtopd(dentry)); -+ __dtopd(dentry) = NULL; /* just to be safe */ -+ out: -+ return; -+} -+ -+ -+/* -+ * we don't really need mini_fo_d_iput, because dentry_iput will call iput() if -+ * mini_fo_d_iput is not defined. We left this implemented for ease of -+ * tracing/debugging. -+ */ -+void -+mini_fo_d_iput(dentry_t *dentry, inode_t *inode) -+{ -+ iput(inode); -+} -+ -+ -+struct dentry_operations mini_fo_dops = { -+ d_revalidate: mini_fo_d_revalidate, -+ d_hash: mini_fo_d_hash, -+ d_compare: mini_fo_d_compare, -+ d_release: mini_fo_d_release, -+ d_delete: mini_fo_d_delete, -+ d_iput: mini_fo_d_iput, -+}; ---- /dev/null -+++ b/fs/mini_fo/file.c -@@ -0,0 +1,713 @@ -+/* -+ * Copyright (c) 1997-2003 Erez Zadok -+ * Copyright (c) 2001-2003 Stony Brook University -+ * -+ * For specific licensing information, see the COPYING file distributed with -+ * this package, or get one from ftp://ftp.filesystems.org/pub/fist/COPYING. -+ * -+ * This Copyright notice must be kept intact and distributed with all -+ * fistgen sources INCLUDING sources generated by fistgen. -+ */ -+/* -+ * Copyright (C) 2004, 2005 Markus Klotzbuecher <mk@creamnet.de> -+ * -+ * This program is free software; you can redistribute it and/or -+ * modify it under the terms of the GNU General Public License -+ * as published by the Free Software Foundation; either version -+ * 2 of the License, or (at your option) any later version. -+ */ -+ -+/* -+ * $Id$ -+ */ -+ -+#ifdef HAVE_CONFIG_H -+# include <config.h> -+#endif -+ -+#include "fist.h" -+#include "mini_fo.h" -+#define ROUND_UP(x) (((x)+sizeof(long)-1) & ~(sizeof(long)-1)) -+ -+/******************* -+ * File Operations * -+ *******************/ -+ -+STATIC loff_t -+mini_fo_llseek(file_t *file, loff_t offset, int origin) -+{ -+ loff_t err; -+ file_t *hidden_file = NULL; -+ -+ if(S_ISDIR(file->f_dentry->d_inode->i_mode)) { -+ /* Check if trying to llseek from a directory */ -+ err = -EISDIR; -+ goto out; -+ } -+ if (ftopd(file) != NULL) { -+ if(ftohf2(file)) { -+ hidden_file = ftohf2(file); -+ } else { -+ hidden_file = ftohf(file); -+ } -+ } -+ -+ /* always set hidden position to this one */ -+ hidden_file->f_pos = file->f_pos; -+ -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,0) -+ memcpy(&(hidden_file->f_ra), -+ &(file->f_ra), -+ sizeof(struct file_ra_state)); -+#else -+ if (file->f_reada) { /* update readahead information if needed */ -+ hidden_file->f_reada = file->f_reada; -+ hidden_file->f_ramax = file->f_ramax; -+ hidden_file->f_raend = file->f_raend; -+ hidden_file->f_ralen = file->f_ralen; -+ hidden_file->f_rawin = file->f_rawin; -+ } -+#endif -+ if (hidden_file->f_op && hidden_file->f_op->llseek) -+ err = hidden_file->f_op->llseek(hidden_file, offset, origin); -+ else -+ err = generic_file_llseek(hidden_file, offset, origin); -+ -+ if (err < 0) -+ goto out; -+ -+ if (err != file->f_pos) { -+ file->f_pos = err; -+ // ION maybe this? -+ // file->f_pos = hidden_file->f_pos; -+#if LINUX_VERSION_CODE < KERNEL_VERSION(2,6,0) -+ file->f_reada = 0; -+#endif -+ file->f_version++; -+ } -+ -+ out: -+ return err; -+} -+ -+ -+/* mk: fanout capable */ -+STATIC ssize_t -+mini_fo_read(file_t *file, char *buf, size_t count, loff_t *ppos) -+{ -+ int err = -EINVAL; -+ file_t *hidden_file = NULL; -+ loff_t pos = *ppos; -+ -+ if(S_ISDIR(file->f_dentry->d_inode->i_mode)) { -+ /* Check if trying to read from a directory */ -+ /* printk(KERN_CRIT "mini_fo_read: ERROR: trying to read data from a directory.\n"); */ -+ err = -EISDIR; -+ goto out; -+ } -+ -+ if (ftopd(file) != NULL) { -+ if(ftohf2(file)) { -+ hidden_file = ftohf2(file); -+ } else { -+ hidden_file = ftohf(file); -+ } -+ } -+ -+ if (!hidden_file->f_op || !hidden_file->f_op->read) -+ goto out; -+ -+ err = hidden_file->f_op->read(hidden_file, buf, count, &pos); -+ *ppos = pos; -+ -+ if (err >= 0) { -+ /* atime should also be updated for reads of size zero or more */ -+ fist_copy_attr_atime(file->f_dentry->d_inode, -+ hidden_file->f_dentry->d_inode); -+ } -+ -+#if LINUX_VERSION_CODE < KERNEL_VERSION(2,6,0) -+ /* -+ * MAJOR HACK -+ * because pread() does not have any way to tell us that it is -+ * our caller, then we don't know for sure if we have to update -+ * the file positions. This hack relies on read() having passed us -+ * the "real" pointer of its struct file's f_pos field. -+ */ -+ if (ppos == &file->f_pos) -+ hidden_file->f_pos = *ppos = pos; -+ if (hidden_file->f_reada) { /* update readahead information if needed */ -+ file->f_reada = hidden_file->f_reada; -+ file->f_ramax = hidden_file->f_ramax; -+ file->f_raend = hidden_file->f_raend; -+ file->f_ralen = hidden_file->f_ralen; -+ file->f_rawin = hidden_file->f_rawin; -+ } -+#else -+ memcpy(&(file->f_ra),&(hidden_file->f_ra),sizeof(struct file_ra_state)); -+#endif -+ -+ out: -+ return err; -+} -+ -+ -+/* this mini_fo_write() does not modify data pages! */ -+STATIC ssize_t -+mini_fo_write(file_t *file, const char *buf, size_t count, loff_t *ppos) -+{ -+ int err = -EINVAL; -+ file_t *hidden_file = NULL; -+ inode_t *inode; -+ inode_t *hidden_inode; -+ loff_t pos = *ppos; -+ -+ /* mk: fan out: */ -+ if (ftopd(file) != NULL) { -+ if(ftohf2(file)) { -+ hidden_file = ftohf2(file); -+ } else { -+ /* This is bad! We have no storage file to write to. This -+ * should never happen because if a file is opened for -+ * writing, a copy should have been made earlier. -+ */ -+ printk(KERN_CRIT "mini_fo: write : ERROR, no storage file to write.\n"); -+ err = -EINVAL; -+ goto out; -+ } -+ } -+ -+ inode = file->f_dentry->d_inode; -+ hidden_inode = itohi2(inode); -+ if(!hidden_inode) { -+ printk(KERN_CRIT "mini_fo: write: no sto inode found, not good.\n"); -+ goto out; -+ } -+ -+ if (!hidden_file->f_op || !hidden_file->f_op->write) -+ goto out; -+ -+ /* adjust for append -- seek to the end of the file */ -+ if (file->f_flags & O_APPEND) -+ pos = inode->i_size; -+ -+ err = hidden_file->f_op->write(hidden_file, buf, count, &pos); -+ -+ /* -+ * copy ctime and mtime from lower layer attributes -+ * atime is unchanged for both layers -+ */ -+ if (err >= 0) -+ fist_copy_attr_times(inode, hidden_inode); -+ -+ *ppos = pos; -+#if LINUX_VERSION_CODE < KERNEL_VERSION(2,6,0) -+ /* -+ * XXX: MAJOR HACK -+ * -+ * because pwrite() does not have any way to tell us that it is -+ * our caller, then we don't know for sure if we have to update -+ * the file positions. This hack relies on write() having passed us -+ * the "real" pointer of its struct file's f_pos field. -+ */ -+ if (ppos == &file->f_pos) -+ hidden_file->f_pos = *ppos = pos; -+#endif -+ /* update this inode's size */ -+ if (pos > inode->i_size) -+ inode->i_size = pos; -+ -+ out: -+ return err; -+} -+ -+/* Global variable to hold a file_t pointer. -+ * This serves to allow mini_fo_filldir function to know which file is -+ * beeing read, which is required for two reasons: -+ * -+ * - be able to call wol functions in order to avoid listing deleted -+ * base files. -+ * - if we're reading a directory which is in state 1, we need to -+ * maintain a list (in mini_fo_filldir) of which files allready -+ * have been copied to userspace,to detect files existing in base -+ * and storage and not list them twice. -+ */ -+filldir_t mini_fo_filldir_orig; -+file_t *mini_fo_filldir_file; -+ -+/* mainly copied from fs/readdir.c */ -+STATIC int -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,18) -+mini_fo_filldir(void * __buf, const char * name, int namlen, loff_t offset, -+ u64 ino, unsigned int d_type) -+#else -+mini_fo_filldir(void * __buf, const char * name, int namlen, loff_t offset, -+ ino_t ino, unsigned int d_type) -+#endif -+{ -+ struct getdents_callback * buf = (struct getdents_callback *) __buf; -+ file_t* file = mini_fo_filldir_file; -+ -+ /* In theses states we filter meta files in storage (WOL) */ -+ if(file && (dtopd(file->f_dentry)->state == MODIFIED || -+ dtopd(file->f_dentry)->state == CREATED || -+ dtopd(file->f_dentry)->state == DEL_REWRITTEN)) { -+ -+ int tmp = strlen(META_FILENAME); -+ if(tmp == namlen) { -+ if(!strncmp(name, META_FILENAME, namlen)) -+ return 0; -+ } -+ } -+ -+ /* check if we are merging the contents of storage and base */ -+ if(file && dtopd(file->f_dentry)->state == MODIFIED) { -+ /* check if we are still reading storage contents, if -+ * yes, we just save the name of the file for duplicate -+ * checking later. */ -+ -+ if(!ftopd(file)->rd.sto_done) { -+ /* put file into ndl list */ -+ if(ndl_add_entry(&ftopd(file)->rd, name, namlen)) -+ printk(KERN_CRIT "mini_fo_filldir: Error adding to ndl.\n"); -+ } else { -+ /* check if file has been deleted */ -+ if(meta_check_d_entry(file->f_dentry, name, namlen)) -+ return 0; -+ -+ /* do duplicate checking */ -+ if(ndl_check_entry(&ftopd(file)->rd, name, namlen)) -+ return 0; -+ } -+ } -+ -+ return mini_fo_filldir_orig(buf, name, namlen, offset, ino, d_type); -+} -+ -+ -+STATIC int -+mini_fo_readdir(file_t *file, void *dirent, filldir_t filldir) -+{ -+ int err = 0;/* mk: ??? -ENOTDIR; */ -+ file_t *hidden_file = NULL; -+ file_t *hidden_sto_file = NULL; -+ inode_t *inode; -+ struct getdents_callback *buf; -+ int oldcount; -+ -+#if defined(FIST_FILTER_NAME) || defined(FIST_FILTER_SCA) -+ struct mini_fo_getdents_callback buf; -+#endif /* FIST_FILTER_NAME || FIST_FILTER_SCA */ -+ -+ buf = (struct getdents_callback *) dirent; -+ oldcount = buf->count; -+ inode = file->f_dentry->d_inode; -+ mini_fo_filldir_file = file; -+ mini_fo_filldir_orig = filldir; -+ -+ ftopd(file)->rd.sto_done = 0; -+ do { -+ if (ftopd(file) != NULL) { -+ if(ftohf2(file)) { -+ hidden_sto_file = ftohf2(file); -+ err = vfs_readdir(hidden_sto_file, mini_fo_filldir, dirent); -+ file->f_pos = hidden_sto_file->f_pos; -+ if (err > 0) -+ fist_copy_attr_atime(inode, hidden_sto_file->f_dentry->d_inode); -+ /* not finshed yet, we'll be called again */ -+ if (buf->count != oldcount) -+ break; -+ } -+ -+ ftopd(file)->rd.sto_done = 1; -+ -+ if(ftohf(file)) { -+ hidden_file = ftohf(file); -+ err = vfs_readdir(hidden_file, mini_fo_filldir, dirent); -+ file->f_pos = hidden_file->f_pos; -+ if (err > 0) -+ fist_copy_attr_atime(inode, hidden_file->f_dentry->d_inode); -+ } -+ -+ } -+ } while (0); -+ -+ /* mk: -+ * we need to check if all the directory data has been copied to userspace, -+ * or if we will be called again by userspace to complete the operation. -+ */ -+ if(buf->count == oldcount) { -+ ndl_put_list(&ftopd(file)->rd); -+ } -+ -+ /* unset this, safe */ -+ mini_fo_filldir_file = NULL; -+ return err; -+} -+ -+ -+STATIC unsigned int -+mini_fo_poll(file_t *file, poll_table *wait) -+{ -+ unsigned int mask = DEFAULT_POLLMASK; -+ file_t *hidden_file = NULL; -+ -+ if (ftopd(file) != NULL) { -+ if(ftohf2(file)) { -+ hidden_file = ftohf2(file); -+ } else { -+ hidden_file = ftohf(file); -+ } -+ } -+ -+ if (!hidden_file->f_op || !hidden_file->f_op->poll) -+ goto out; -+ -+ mask = hidden_file->f_op->poll(hidden_file, wait); -+ -+ out: -+ return mask; -+} -+ -+/* FIST-LITE special version of mmap */ -+STATIC int -+mini_fo_mmap(file_t *file, vm_area_t *vma) -+{ -+ int err = 0; -+ file_t *hidden_file = NULL; -+ -+ /* fanout capability */ -+ if (ftopd(file) != NULL) { -+ if(ftohf2(file)) { -+ hidden_file = ftohf2(file); -+ } else { -+ hidden_file = ftohf(file); -+ } -+ } -+ -+ ASSERT(hidden_file != NULL); -+ ASSERT(hidden_file->f_op != NULL); -+ ASSERT(hidden_file->f_op->mmap != NULL); -+ -+ vma->vm_file = hidden_file; -+ err = hidden_file->f_op->mmap(hidden_file, vma); -+ get_file(hidden_file); /* make sure it doesn't get freed on us */ -+ fput(file); /* no need to keep extra ref on ours */ -+ -+ return err; -+} -+ -+ -+ -+STATIC int -+mini_fo_open(inode_t *inode, file_t *file) -+{ -+ int err = 0; -+ int hidden_flags; -+ file_t *hidden_file = NULL; -+ dentry_t *hidden_dentry = NULL; -+ -+ /* fanout stuff */ -+ file_t *hidden_sto_file = NULL; -+ dentry_t *hidden_sto_dentry = NULL; -+ -+ __ftopd(file) = -+ kmalloc(sizeof(struct mini_fo_file_info), GFP_KERNEL); -+ if (!ftopd(file)) { -+ err = -ENOMEM; -+ goto out; -+ } -+ -+ /* init the readdir_helper structure */ -+ INIT_LIST_HEAD(&ftopd(file)->rd.ndl_list); -+ ftopd(file)->rd.ndl_size = 0; -+ -+ /* In certain paths this could stay uninitalized and cause trouble */ -+ ftohf(file) = NULL; -+ ftohf2(file) = NULL; -+ hidden_flags = file->f_flags; -+ -+ /* create storage files? */ -+ if(dtost(file->f_dentry) == UNMODIFIED) { -+ if(!IS_WRITE_FLAG(file->f_flags)) { -+ hidden_dentry = dtohd(file->f_dentry); -+ dget(hidden_dentry); -+ /* dentry_open will decrement mnt refcnt if err. -+ * otherwise fput() will do an mntput() for us upon file close. */ -+ mntget(stopd(inode->i_sb)->hidden_mnt); -+ hidden_file = dentry_open(hidden_dentry, -+ stopd(inode->i_sb)->hidden_mnt, -+ hidden_flags); -+ if (IS_ERR(hidden_file)) { -+ err = PTR_ERR(hidden_file); -+ dput(hidden_dentry); -+ goto out; -+ } -+ ftohf(file) = hidden_file; /* link two files */ -+ goto out; -+ } -+ else { -+ if(S_ISDIR(file->f_dentry->d_inode->i_mode)) { -+ err = dir_unmod_to_mod(file->f_dentry); -+ } else -+ err = nondir_unmod_to_mod(file->f_dentry, 1); -+ -+ if (err) { -+ printk("mini_fo_open: ERROR creating storage file.\n"); -+ goto out; -+ } -+ } -+ } -+ hidden_sto_dentry = dtohd2(file->f_dentry); -+ dget(hidden_sto_dentry); -+ -+ if(dtopd(file->f_dentry)->state == MODIFIED) { -+ /* Directorys are special, interpose on both lower level files */ -+ if(S_ISDIR(itohi(inode)->i_mode)) { -+ /* check for invalid file types of lower level files */ -+ if(!(S_ISDIR(itohi(inode)->i_mode) && S_ISDIR(itohi2(inode)->i_mode))) { -+ printk(KERN_CRIT "mini_fo_open: meta data corruption detected.\n"); -+ dput(hidden_sto_dentry); -+ err = -EINVAL; -+ goto out; -+ } -+ -+ /* lower level directorys are ok, open the base file */ -+ hidden_dentry = dtohd(file->f_dentry); -+ dget(hidden_dentry); -+ -+ mntget(stopd(inode->i_sb)->hidden_mnt); -+ hidden_file = dentry_open(hidden_dentry, -+ stopd(inode->i_sb)->hidden_mnt, -+ hidden_flags); -+ if (IS_ERR(hidden_file)) { -+ err = PTR_ERR(hidden_file); -+ dput(hidden_dentry); -+ dput(hidden_sto_dentry); -+ goto out; -+ } -+ ftohf(file) = hidden_file; /* link the two files */ -+ } -+ } -+ -+ if(!exists_in_storage(file->f_dentry)) { -+ printk(KERN_CRIT "mini_fo_open: invalid file state detected.\n"); -+ err = -EINVAL; -+ dput(hidden_sto_dentry); -+ -+ /* If the base file has been opened, we need to close it here */ -+ if(ftohf(file)) { -+ if (hidden_file->f_op && hidden_file->f_op->flush) -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,18) -+ hidden_file->f_op->flush(hidden_file, NULL); -+#else -+ hidden_file->f_op->flush(hidden_file); -+#endif -+ dput(hidden_dentry); -+ } -+ goto out; -+ } -+ -+ /* ok, now we can safely open the storage file */ -+ mntget(stopd(inode->i_sb)->hidden_mnt2); -+ hidden_sto_file = dentry_open(hidden_sto_dentry, -+ stopd(inode->i_sb)->hidden_mnt2, -+ hidden_flags); -+ -+ /* dentry_open dputs the dentry if it fails */ -+ if (IS_ERR(hidden_sto_file)) { -+ err = PTR_ERR(hidden_sto_file); -+ /* close base file if open */ -+ if(ftohf(file)) { -+ if (hidden_file->f_op && hidden_file->f_op->flush) -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,18) -+ hidden_file->f_op->flush(hidden_file, NULL); -+#else -+ hidden_file->f_op->flush(hidden_file); -+#endif -+ dput(hidden_dentry); -+ } -+ goto out; -+ } -+ ftohf2(file) = hidden_sto_file; /* link storage file */ -+ -+ out: -+ if (err < 0 && ftopd(file)) { -+ kfree(ftopd(file)); -+ } -+ return err; -+} -+ -+STATIC int -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,18) -+mini_fo_flush(file_t *file, fl_owner_t id) -+#else -+mini_fo_flush(file_t *file) -+#endif -+{ -+ int err1 = 0; /* assume ok (see open.c:close_fp) */ -+ int err2 = 0; -+ file_t *hidden_file = NULL; -+ -+ check_mini_fo_file(file); -+ -+ /* mk: we don't do any state checking here, as its not worth the time. -+ * Just flush the lower level files if they exist. -+ */ -+ if(ftopd(file) != NULL) { -+ if(ftohf(file) != NULL) { -+ hidden_file = ftohf(file); -+ if (hidden_file->f_op && hidden_file->f_op->flush) -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,18) -+ err1 = hidden_file->f_op->flush(hidden_file, id); -+#else -+ err1 = hidden_file->f_op->flush(hidden_file); -+#endif -+ } -+ if(ftohf2(file) != NULL) { -+ hidden_file = ftohf2(file); -+ if (hidden_file->f_op && hidden_file->f_op->flush) -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,18) -+ err2 = hidden_file->f_op->flush(hidden_file, id); -+#else -+ err2 = hidden_file->f_op->flush(hidden_file); -+#endif -+ } -+ } -+ return (err1 | err2); -+} -+ -+ -+STATIC int -+mini_fo_release(inode_t *inode, file_t *file) -+{ -+ int err = 0; -+ file_t *hidden_file = NULL; -+ -+ if (ftopd(file) != NULL) { -+ if(ftohf(file)) { -+ hidden_file = ftohf(file); -+ fput(hidden_file); -+ } -+ if(ftohf2(file)) { -+ hidden_file = ftohf2(file); -+ fput(hidden_file); -+ } -+ kfree(ftopd(file)); -+ } -+ return err; -+} -+ -+STATIC int -+mini_fo_fsync(file_t *file, dentry_t *dentry, int datasync) -+{ -+ int err1 = 0; -+ int err2 = 0; -+ file_t *hidden_file = NULL; -+ dentry_t *hidden_dentry; -+ -+ check_mini_fo_file(file); -+ -+ if ((hidden_file = ftohf(file)) != NULL) { -+ hidden_dentry = dtohd(dentry); -+ if (hidden_file->f_op && hidden_file->f_op->fsync) { -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,16) -+ mutex_lock(&hidden_dentry->d_inode->i_mutex); -+#else -+ down(&hidden_dentry->d_inode->i_sem); -+#endif -+ err1 = hidden_file->f_op->fsync(hidden_file, hidden_dentry, datasync); -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,16) -+ mutex_unlock(&hidden_dentry->d_inode->i_mutex); -+#else -+ up(&hidden_dentry->d_inode->i_sem); -+#endif -+ } -+ } -+ -+ if ((hidden_file = ftohf2(file)) != NULL) { -+ hidden_dentry = dtohd2(dentry); -+ if (hidden_file->f_op && hidden_file->f_op->fsync) { -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,16) -+ mutex_lock(&hidden_dentry->d_inode->i_mutex); -+#else -+ down(&hidden_dentry->d_inode->i_sem); -+#endif -+ err2 = hidden_file->f_op->fsync(hidden_file, hidden_dentry, datasync); -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,16) -+ mutex_unlock(&hidden_dentry->d_inode->i_mutex); -+#else -+ up(&hidden_dentry->d_inode->i_sem); -+#endif -+ } -+ } -+ else -+ goto err; -+ -+err: -+ return (err1 || err2); -+} -+ -+ -+STATIC int -+mini_fo_fasync(int fd, file_t *file, int flag) -+{ -+ int err1 = 0; -+ int err2 = 0; -+ -+ file_t *hidden_file = NULL; -+ -+ check_mini_fo_file(file); -+ -+ if((hidden_file = ftohf(file)) != NULL) { -+ err1 = hidden_file->f_op->fasync(fd, hidden_file, flag); -+ } -+ if((hidden_file = ftohf2(file)) != NULL) { -+ err2 = hidden_file->f_op->fasync(fd, hidden_file, flag); -+ } -+ -+ return (err1 || err2); -+} -+ -+ -+ -+struct file_operations mini_fo_dir_fops = -+ { -+ read: generic_read_dir, -+ write: mini_fo_write, -+ readdir: mini_fo_readdir, -+ poll: mini_fo_poll, -+ /* ioctl: mini_fo_ioctl, */ -+ mmap: mini_fo_mmap, -+ open: mini_fo_open, -+ flush: mini_fo_flush, -+ release: mini_fo_release, -+ fsync: mini_fo_fsync, -+ fasync: mini_fo_fasync, -+ /* not needed lock: mini_fo_lock, */ -+ /* not needed: readv */ -+ /* not needed: writev */ -+ /* not implemented: sendpage */ -+ /* not implemented: get_unmapped_area */ -+ }; -+ -+struct file_operations mini_fo_main_fops = -+ { -+ llseek: mini_fo_llseek, -+ read: mini_fo_read, -+ write: mini_fo_write, -+ readdir: mini_fo_readdir, -+ poll: mini_fo_poll, -+ /* ioctl: mini_fo_ioctl, */ -+ mmap: mini_fo_mmap, -+ open: mini_fo_open, -+ flush: mini_fo_flush, -+ release: mini_fo_release, -+ fsync: mini_fo_fsync, -+ fasync: mini_fo_fasync, -+ /* not needed: lock: mini_fo_lock, */ -+ /* not needed: readv */ -+ /* not needed: writev */ -+ /* not implemented: sendpage */ -+ /* not implemented: get_unmapped_area */ -+ }; ---- /dev/null -+++ b/fs/mini_fo/fist.h -@@ -0,0 +1,252 @@ -+/* -+ * Copyright (c) 1997-2003 Erez Zadok -+ * Copyright (c) 2001-2003 Stony Brook University -+ * -+ * For specific licensing information, see the COPYING file distributed with -+ * this package, or get one from ftp://ftp.filesystems.org/pub/fist/COPYING. -+ * -+ * This Copyright notice must be kept intact and distributed with all -+ * fistgen sources INCLUDING sources generated by fistgen. -+ */ -+/* -+ * Copyright (C) 2004, 2005 Markus Klotzbuecher <mk@creamnet.de> -+ * -+ * This program is free software; you can redistribute it and/or -+ * modify it under the terms of the GNU General Public License -+ * as published by the Free Software Foundation; either version -+ * 2 of the License, or (at your option) any later version. -+ */ -+ -+ -+/* -+ * $Id$ -+ */ -+ -+#ifndef __FIST_H_ -+#define __FIST_H_ -+ -+/* -+ * KERNEL ONLY CODE: -+ */ -+#ifdef __KERNEL__ -+#include <linux/version.h> -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,19) -+#include <linux/autoconf.h> -+#else -+#include <linux/config.h> -+#endif -+#if LINUX_VERSION_CODE < KERNEL_VERSION(2,6,0) -+#ifdef CONFIG_MODVERSIONS -+# define MODVERSIONS -+# include <linux/modversions.h> -+#endif /* CONFIG_MODVERSIONS */ -+#endif /* KERNEL_VERSION < 2.6.0 */ -+#include <linux/sched.h> -+#include <linux/kernel.h> -+#include <linux/mm.h> -+#include <linux/string.h> -+#include <linux/stat.h> -+#include <linux/errno.h> -+#include <linux/wait.h> -+#include <linux/limits.h> -+#if LINUX_VERSION_CODE < KERNEL_VERSION(2,6,0) -+#include <linux/locks.h> -+#else -+#include <linux/buffer_head.h> -+#include <linux/pagemap.h> -+#include <linux/namei.h> -+#include <linux/module.h> -+#include <linux/mount.h> -+#include <linux/page-flags.h> -+#include <linux/writeback.h> -+#include <linux/statfs.h> -+#endif -+#include <linux/smp.h> -+#include <linux/smp_lock.h> -+#include <linux/file.h> -+#include <linux/slab.h> -+#include <linux/vmalloc.h> -+#include <linux/poll.h> -+#include <linux/list.h> -+#include <linux/init.h> -+ -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,4,20) -+#include <linux/xattr.h> -+#endif -+ -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,0) -+#include <linux/security.h> -+#endif -+ -+#include <linux/swap.h> -+ -+#include <asm/system.h> -+/* #include <asm/segment.h> */ -+#include <asm/mman.h> -+#include <linux/seq_file.h> -+ -+/* -+ * MACROS: -+ */ -+ -+/* those mapped to ATTR_* were copied from linux/fs.h */ -+#define FA_MODE ATTR_MODE -+#define FA_UID ATTR_UID -+#define FA_GID ATTR_GID -+#define FA_SIZE ATTR_SIZE -+#define FA_ATIME ATTR_ATIME -+#define FA_MTIME ATTR_MTIME -+#define FA_CTIME ATTR_CTIME -+#define FA_ATIME_SET ATTR_ATIME_SET -+#define FA_MTIME_SET ATTR_MTIME_SET -+#define FA_FORCE ATTR_FORCE -+#define FA_ATTR_FLAGS ATTR_ATTR_FLAG -+ -+/* must be greater than all other ATTR_* flags! */ -+#define FA_NLINK 2048 -+#define FA_BLKSIZE 4096 -+#define FA_BLOCKS 8192 -+#define FA_TIMES (FA_ATIME|FA_MTIME|FA_CTIME) -+#define FA_ALL 0 -+ -+/* macros to manage changes between kernels */ -+#define INODE_DATA(i) (&(i)->i_data) -+ -+#define MIN(x,y) ((x < y) ? (x) : (y)) -+#define MAX(x,y) ((x > y) ? (x) : (y)) -+#define MAXPATHLEN PATH_MAX -+ -+#if LINUX_VERSION_CODE < KERNEL_VERSION(2,4,5) -+# define lookup_one_len(a,b,c) lookup_one(a,b) -+#endif /* LINUX_VERSION_CODE < KERNEL_VERSION(2,4,5) */ -+ -+#if LINUX_VERSION_CODE < KERNEL_VERSION(2,4,8) -+# define generic_file_llseek default_llseek -+#endif /* LINUX_VERSION_CODE < KERNEL_VERSION(2,4,8) */ -+ -+#ifndef SEEK_SET -+# define SEEK_SET 0 -+#endif /* not SEEK_SET */ -+ -+#ifndef SEEK_CUR -+# define SEEK_CUR 1 -+#endif /* not SEEK_CUR */ -+ -+#ifndef SEEK_END -+# define SEEK_END 2 -+#endif /* not SEEK_END */ -+ -+#ifndef DEFAULT_POLLMASK -+# define DEFAULT_POLLMASK (POLLIN | POLLOUT | POLLRDNORM | POLLWRNORM) -+#endif /* not DEFAULT_POLLMASK */ -+ -+/* XXX: fix this so fistgen generates kfree() code directly */ -+#define kfree_s(a,b) kfree(a) -+ -+/* -+ * TYPEDEFS: -+ */ -+typedef struct dentry dentry_t; -+typedef struct file file_t; -+typedef struct inode inode_t; -+typedef inode_t vnode_t; -+typedef struct page page_t; -+typedef struct qstr qstr_t; -+typedef struct super_block super_block_t; -+typedef super_block_t vfs_t; -+typedef struct vm_area_struct vm_area_t; -+ -+ -+/* -+ * EXTERNALS: -+ */ -+ -+#define FPPF(str,page) printk("PPF %s 0x%x/%d: Lck:%d Err:%d Ref:%d Upd:%d Other::%d:%d:%d:%d:\n", \ -+ str, \ -+ (int) page, \ -+ (int) page->index, \ -+ (PageLocked(page) ? 1 : 0), \ -+ (PageError(page) ? 1 : 0), \ -+ (PageReferenced(page) ? 1 : 0), \ -+ (Page_Uptodate(page) ? 1 : 0), \ -+ (PageDecrAfter(page) ? 1 : 0), \ -+ (PageSlab(page) ? 1 : 0), \ -+ (PageSwapCache(page) ? 1 : 0), \ -+ (PageReserved(page) ? 1 : 0) \ -+ ) -+#define EZKDBG printk("EZK %s:%d:%s\n",__FILE__,__LINE__,__FUNCTION__) -+#if 0 -+# define EZKDBG1 printk("EZK %s:%d\n",__FILE__,__LINE__) -+#else -+# define EZKDBG1 -+#endif -+ -+extern int fist_get_debug_value(void); -+extern int fist_set_debug_value(int val); -+#if 0 /* mini_fo doesn't need these */ -+extern void fist_dprint_internal(int level, char *str,...); -+extern void fist_print_dentry(char *str, const dentry_t *dentry); -+extern void fist_print_inode(char *str, const inode_t *inode); -+extern void fist_print_file(char *str, const file_t *file); -+extern void fist_print_buffer_flags(char *str, struct buffer_head *buffer); -+extern void fist_print_page_flags(char *str, page_t *page); -+extern void fist_print_page_bytes(char *str, page_t *page); -+extern void fist_print_pte_flags(char *str, const page_t *page); -+extern void fist_checkinode(inode_t *inode, char *msg); -+extern void fist_print_sb(char *str, const super_block_t *sb); -+ -+/* §$% by mk: special debug functions */ -+extern void fist_mk_print_dentry(char *str, const dentry_t *dentry); -+extern void fist_mk_print_inode(char *str, const inode_t *inode); -+ -+extern char *add_indent(void); -+extern char *del_indent(void); -+#endif/* mini_fo doesn't need these */ -+ -+ -+#define STATIC -+#define ASSERT(EX) \ -+do { \ -+ if (!(EX)) { \ -+ printk(KERN_CRIT "ASSERTION FAILED: %s at %s:%d (%s)\n", #EX, \ -+ __FILE__, __LINE__, __FUNCTION__); \ -+ (*((char *)0))=0; \ -+ } \ -+} while (0) -+/* same ASSERT, but tell me who was the caller of the function */ -+#define ASSERT2(EX) \ -+do { \ -+ if (!(EX)) { \ -+ printk(KERN_CRIT "ASSERTION FAILED (caller): %s at %s:%d (%s)\n", #EX, \ -+ file, line, func); \ -+ (*((char *)0))=0; \ -+ } \ -+} while (0) -+ -+#if 0 /* mini_fo doesn't need these */ -+#define dprintk(format, args...) printk(KERN_DEBUG format, ##args) -+#define fist_dprint(level, str, args...) fist_dprint_internal(level, KERN_DEBUG str, ## args) -+#define print_entry_location() fist_dprint(4, "%sIN: %s %s:%d\n", add_indent(), __FUNCTION__, __FILE__, __LINE__) -+#define print_exit_location() fist_dprint(4, "%s OUT: %s %s:%d\n", del_indent(), __FUNCTION__, __FILE__, __LINE__) -+#define print_exit_status(status) fist_dprint(4, "%s OUT: %s %s:%d, STATUS: %d\n", del_indent(), __FUNCTION__, __FILE__, __LINE__, status) -+#define print_exit_pointer(status) \ -+do { \ -+ if (IS_ERR(status)) \ -+ fist_dprint(4, "%s OUT: %s %s:%d, RESULT: %ld\n", del_indent(), __FUNCTION__, __FILE__, __LINE__, PTR_ERR(status)); \ -+ else \ -+ fist_dprint(4, "%s OUT: %s %s:%d, RESULT: 0x%x\n", del_indent(), __FUNCTION__, __FILE__, __LINE__, PTR_ERR(status)); \ -+} while (0) -+#endif/* mini_fo doesn't need these */ -+ -+#endif /* __KERNEL__ */ -+ -+ -+/* -+ * DEFINITIONS FOR USER AND KERNEL CODE: -+ * (Note: ioctl numbers 1--9 are reserved for fistgen, the rest -+ * are auto-generated automatically based on the user's .fist file.) -+ */ -+# define FIST_IOCTL_GET_DEBUG_VALUE _IOR(0x15, 1, int) -+# define FIST_IOCTL_SET_DEBUG_VALUE _IOW(0x15, 2, int) -+ -+#endif /* not __FIST_H_ */ ---- /dev/null -+++ b/fs/mini_fo/inode.c -@@ -0,0 +1,1564 @@ -+/* -+ * Copyright (c) 1997-2003 Erez Zadok -+ * Copyright (c) 2001-2003 Stony Brook University -+ * -+ * For specific licensing information, see the COPYING file distributed with -+ * this package, or get one from ftp://ftp.filesystems.org/pub/fist/COPYING. -+ * -+ * This Copyright notice must be kept intact and distributed with all -+ * fistgen sources INCLUDING sources generated by fistgen. -+ */ -+/* -+ * Copyright (C) 2004, 2005 Markus Klotzbuecher <mk@creamnet.de> -+ * -+ * This program is free software; you can redistribute it and/or -+ * modify it under the terms of the GNU General Public License -+ * as published by the Free Software Foundation; either version -+ * 2 of the License, or (at your option) any later version. -+ */ -+ -+/* -+ * $Id$ -+ */ -+ -+#ifdef HAVE_CONFIG_H -+# include <config.h> -+#endif -+ -+#include "fist.h" -+#include "mini_fo.h" -+ -+STATIC int -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,0) -+mini_fo_create(inode_t *dir, dentry_t *dentry, int mode, struct nameidata *nd) -+#else -+mini_fo_create(inode_t *dir, dentry_t *dentry, int mode) -+#endif -+{ -+ int err = 0; -+ -+ check_mini_fo_dentry(dentry); -+ -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,0) -+ err = create_sto_reg_file(dentry, mode, nd); -+#else -+ err = create_sto_reg_file(dentry, mode); -+#endif -+ check_mini_fo_dentry(dentry); -+ return err; -+} -+ -+ -+STATIC dentry_t * -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,0) -+mini_fo_lookup(inode_t *dir, dentry_t *dentry, struct nameidata* nd) -+#else -+mini_fo_lookup(inode_t *dir, dentry_t *dentry) -+#endif -+{ -+ int err = 0; -+ dentry_t *hidden_dir_dentry; -+ dentry_t *hidden_dentry = NULL; -+ -+ dentry_t *hidden_sto_dir_dentry; -+ dentry_t *hidden_sto_dentry = NULL; -+ -+ /* whiteout flag */ -+ int del_flag = 0; -+ char *bpath = NULL; -+ -+ const char *name; -+ unsigned int namelen; -+ -+ /* Don't allow lookups of META-files */ -+ namelen = strlen(META_FILENAME); -+ if(namelen == dentry->d_name.len) { -+ if(!strncmp(dentry->d_name.name, META_FILENAME, namelen)) { -+ err = -ENOENT; -+ goto out; -+ } -+ } -+ -+ hidden_dir_dentry = dtohd(dentry->d_parent); -+ hidden_sto_dir_dentry = dtohd2(dentry->d_parent); -+ -+ name = dentry->d_name.name; -+ namelen = dentry->d_name.len; -+ -+ /* must initialize dentry operations */ -+ dentry->d_op = &mini_fo_dops; -+ -+ /* setup the del_flag */ -+ del_flag = __meta_check_d_entry(dir, name, namelen); -+ bpath = __meta_check_r_entry(dir, name, namelen); -+ -+ /* perform the lookups of base and storage files: -+ * -+ * This caused some serious trouble, as a lookup_one_len passing -+ * a negative dentry oopses. Solution is to only do the lookup -+ * if the dentry is positive, else we set it to NULL -+ * More trouble, who said a *_dir_dentry can't be NULL? -+ */ -+ if(bpath) { -+ /* Cross-Interposing (C), yeah! */ -+ hidden_dentry = bpath_walk(dir->i_sb, bpath); -+ if(!hidden_dentry || !hidden_dentry->d_inode) { -+ printk(KERN_CRIT "mini_fo_lookup: bpath_walk failed.\n"); -+ err= -EINVAL; -+ goto out; -+ } -+ -+ /* this can be set up safely without fear of spaghetti -+ * interposing as it is only used for copying times */ -+ hidden_dir_dentry = hidden_dentry->d_parent; -+ kfree(bpath); -+ } -+ else if(hidden_dir_dentry && hidden_dir_dentry->d_inode) -+ hidden_dentry = -+ lookup_one_len(name, hidden_dir_dentry, namelen); -+ else -+ hidden_dentry = NULL; -+ -+ if(hidden_sto_dir_dentry && hidden_sto_dir_dentry->d_inode) -+ hidden_sto_dentry = -+ lookup_one_len(name, hidden_sto_dir_dentry, namelen); -+ else -+ hidden_sto_dentry = NULL; -+ -+ /* catch error in lookup */ -+ if (IS_ERR(hidden_dentry) || IS_ERR(hidden_sto_dentry)) { -+ /* mk: we need to call dput on the dentry, whose -+ * lookup_one_len operation failed, in order to avoid -+ * unmount trouble. -+ */ -+ if(IS_ERR(hidden_dentry)) { -+ printk(KERN_CRIT "mini_fo_lookup: ERR from base dentry, lookup failed.\n"); -+ err = PTR_ERR(hidden_dentry); -+ } else { -+ dput(hidden_dentry); -+ } -+ if(IS_ERR(hidden_sto_dentry)) { -+ printk(KERN_CRIT "mini_fo_lookup: ERR from storage dentry, lookup failed.\n"); -+ err = PTR_ERR(hidden_sto_dentry); -+ } else { -+ dput(hidden_sto_dentry); -+ } -+ goto out; -+ } -+ -+ /* allocate dentry private data */ -+ __dtopd(dentry) = (struct mini_fo_dentry_info *) -+ kmalloc(sizeof(struct mini_fo_dentry_info), GFP_KERNEL); -+ -+ if (!dtopd(dentry)) { -+ err = -ENOMEM; -+ goto out_dput; -+ } -+ -+ /* check for different states of the mini_fo file to be looked up. */ -+ -+ /* state 1, file has been modified */ -+ if(hidden_dentry && hidden_sto_dentry && -+ hidden_dentry->d_inode && hidden_sto_dentry->d_inode && !del_flag) { -+ -+ /* update parent directory's atime */ -+ fist_copy_attr_atime(dir, hidden_sto_dir_dentry->d_inode); -+ -+ dtopd(dentry)->state = MODIFIED; -+ dtohd(dentry) = hidden_dentry; -+ dtohd2(dentry) = hidden_sto_dentry; -+ -+ err = mini_fo_tri_interpose(hidden_dentry, -+ hidden_sto_dentry, -+ dentry, dir->i_sb, 1); -+ if (err) { -+ printk(KERN_CRIT "mini_fo_lookup: error interposing (state1).\n"); -+ goto out_free; -+ } -+ goto out; -+ } -+ /* state 2, file is unmodified */ -+ if(hidden_dentry && hidden_dentry->d_inode && !del_flag) { -+ -+ fist_copy_attr_atime(dir, hidden_dir_dentry->d_inode); -+ -+ dtopd(dentry)->state = UNMODIFIED; -+ dtohd(dentry) = hidden_dentry; -+ dtohd2(dentry) = hidden_sto_dentry; /* could be negative */ -+ -+ err = mini_fo_tri_interpose(hidden_dentry, -+ hidden_sto_dentry, -+ dentry, dir->i_sb, 1); -+ if (err) { -+ printk(KERN_CRIT "mini_fo_lookup: error interposing (state2).\n"); -+ goto out_free; -+ } -+ goto out; -+ } -+ /* state 3, file has been newly created */ -+ if(hidden_sto_dentry && hidden_sto_dentry->d_inode && !del_flag) { -+ -+ fist_copy_attr_atime(dir, hidden_sto_dir_dentry->d_inode); -+ dtopd(dentry)->state = CREATED; -+ dtohd(dentry) = hidden_dentry; /* could be negative */ -+ dtohd2(dentry) = hidden_sto_dentry; -+ -+ err = mini_fo_tri_interpose(hidden_dentry, -+ hidden_sto_dentry, -+ dentry, dir->i_sb, 1); -+ if (err) { -+ printk(KERN_CRIT "mini_fo_lookup: error interposing (state3).\n"); -+ goto out_free; -+ } -+ goto out; -+ } -+ -+ /* state 4, file has deleted and created again. */ -+ if(hidden_dentry && hidden_sto_dentry && -+ hidden_dentry->d_inode && -+ hidden_sto_dentry->d_inode && del_flag) { -+ -+ fist_copy_attr_atime(dir, hidden_sto_dir_dentry->d_inode); -+ dtopd(dentry)->state = DEL_REWRITTEN; -+ dtohd(dentry) = NULL; -+ dtohd2(dentry) = hidden_sto_dentry; -+ -+ err = mini_fo_tri_interpose(NULL, -+ hidden_sto_dentry, -+ dentry, dir->i_sb, 1); -+ if (err) { -+ printk(KERN_CRIT "mini_fo_lookup: error interposing (state4).\n"); -+ goto out_free; -+ } -+ /* We will never need this dentry again, as the file has been -+ * deleted from base */ -+ dput(hidden_dentry); -+ goto out; -+ } -+ /* state 5, file has been deleted in base */ -+ if(hidden_dentry && hidden_sto_dentry && -+ hidden_dentry->d_inode && -+ !hidden_sto_dentry->d_inode && del_flag) { -+ -+ /* check which parents atime we need for updating */ -+ if(hidden_sto_dir_dentry->d_inode) -+ fist_copy_attr_atime(dir, -+ hidden_sto_dir_dentry->d_inode); -+ else -+ fist_copy_attr_atime(dir, -+ hidden_dir_dentry->d_inode); -+ -+ dtopd(dentry)->state = DELETED; -+ dtohd(dentry) = NULL; -+ dtohd2(dentry) = hidden_sto_dentry; -+ -+ /* add negative dentry to dcache to speed up lookups */ -+ d_add(dentry, NULL); -+ dput(hidden_dentry); -+ goto out; -+ } -+ /* state 6, file does not exist */ -+ if(((hidden_dentry && !hidden_dentry->d_inode) || -+ (hidden_sto_dentry && !hidden_sto_dentry->d_inode)) && !del_flag) -+ { -+ /* check which parents atime we need for updating */ -+ if(hidden_sto_dir_dentry && hidden_sto_dir_dentry->d_inode) -+ fist_copy_attr_atime(dir, hidden_sto_dir_dentry->d_inode); -+ else -+ fist_copy_attr_atime(dir, hidden_dir_dentry->d_inode); -+ -+ dtopd(dentry)->state = NON_EXISTANT; -+ dtohd(dentry) = hidden_dentry; -+ dtohd2(dentry) = hidden_sto_dentry; -+ d_add(dentry, NULL); -+ goto out; -+ } -+ -+ /* if we get to here, were in an invalid state. bad. */ -+ printk(KERN_CRIT "mini_fo_lookup: ERROR, meta data corruption detected.\n"); -+ -+ /* end state checking */ -+ out_free: -+ d_drop(dentry); /* so that our bad dentry will get destroyed */ -+ kfree(dtopd(dentry)); -+ __dtopd(dentry) = NULL; /* be safe */ -+ -+ out_dput: -+ if(hidden_dentry) -+ dput(hidden_dentry); -+ if(hidden_sto_dentry) -+ dput(hidden_sto_dentry); /* drops usage count and marks for release */ -+ -+ out: -+ /* initalize wol if file exists and is directory */ -+ if(dentry->d_inode) { -+ if(S_ISDIR(dentry->d_inode->i_mode)) { -+ itopd(dentry->d_inode)->deleted_list_size = -1; -+ itopd(dentry->d_inode)->renamed_list_size = -1; -+ meta_build_lists(dentry); -+ } -+ } -+ return ERR_PTR(err); -+} -+ -+ -+STATIC int -+mini_fo_link(dentry_t *old_dentry, inode_t *dir, dentry_t *new_dentry) -+{ -+ int err; -+ dentry_t *hidden_old_dentry; -+ dentry_t *hidden_new_dentry; -+ dentry_t *hidden_dir_dentry; -+ -+ -+ check_mini_fo_dentry(old_dentry); -+ check_mini_fo_dentry(new_dentry); -+ check_mini_fo_inode(dir); -+ -+ /* no links to directorys and existing targets target allowed */ -+ if(S_ISDIR(old_dentry->d_inode->i_mode) || -+ is_mini_fo_existant(new_dentry)) { -+ err = -EPERM; -+ goto out; -+ } -+ -+ /* bring it directly from unmod to del_rew */ -+ if(dtost(old_dentry) == UNMODIFIED) { -+ err = nondir_unmod_to_mod(old_dentry, 1); -+ if(err) { -+ err = -EINVAL; -+ goto out; -+ } -+ err = meta_add_d_entry(old_dentry->d_parent, -+ old_dentry->d_name.name, -+ old_dentry->d_name.len); -+ if(err) { -+ err = -EINVAL; -+ goto out; -+ } -+ dput(dtohd(old_dentry)); -+ dtohd(old_dentry) = NULL; -+ dtost(old_dentry) = DEL_REWRITTEN; -+ } -+ -+ err = get_neg_sto_dentry(new_dentry); -+ if(err) { -+ err = -EINVAL; -+ goto out; -+ } -+ -+ hidden_old_dentry = dtohd2(old_dentry); -+ hidden_new_dentry = dtohd2(new_dentry); -+ -+ dget(hidden_old_dentry); -+ dget(hidden_new_dentry); -+ -+ /* was: hidden_dir_dentry = lock_parent(hidden_new_dentry); */ -+ hidden_dir_dentry = dget(hidden_new_dentry->d_parent); -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,16) -+ mutex_lock(&hidden_dir_dentry->d_inode->i_mutex); -+#else -+ down(&hidden_dir_dentry->d_inode->i_sem); -+#endif -+ -+ err = vfs_link(hidden_old_dentry, -+ hidden_dir_dentry->d_inode, -+ hidden_new_dentry); -+ if (err || !hidden_new_dentry->d_inode) -+ goto out_lock; -+ -+ dtost(new_dentry) = CREATED; -+ err = mini_fo_tri_interpose(NULL, hidden_new_dentry, new_dentry, dir->i_sb, 0); -+ if (err) -+ goto out_lock; -+ -+ fist_copy_attr_timesizes(dir, hidden_new_dentry->d_inode); -+ /* propagate number of hard-links */ -+ old_dentry->d_inode->i_nlink = itohi2(old_dentry->d_inode)->i_nlink; -+ -+ out_lock: -+ /* was: unlock_dir(hidden_dir_dentry); */ -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,16) -+ mutex_unlock(&hidden_dir_dentry->d_inode->i_mutex); -+#else -+ up(&hidden_dir_dentry->d_inode->i_sem); -+#endif -+ dput(hidden_dir_dentry); -+ -+ dput(hidden_new_dentry); -+ dput(hidden_old_dentry); -+ if (!new_dentry->d_inode) -+ d_drop(new_dentry); -+ -+ out: -+ return err; -+} -+ -+ -+STATIC int -+mini_fo_unlink(inode_t *dir, dentry_t *dentry) -+{ -+ int err = 0; -+ -+ dget(dentry); -+ if(dtopd(dentry)->state == MODIFIED) { -+ err = nondir_mod_to_del(dentry); -+ goto out; -+ } -+ else if(dtopd(dentry)->state == UNMODIFIED) { -+ err = nondir_unmod_to_del(dentry); -+ goto out; -+ } -+ else if(dtopd(dentry)->state == CREATED) { -+ err = nondir_creat_to_del(dentry); -+ goto out; -+ } -+ else if(dtopd(dentry)->state == DEL_REWRITTEN) { -+ err = nondir_del_rew_to_del(dentry); -+ goto out; -+ } -+ -+ printk(KERN_CRIT "mini_fo_unlink: ERROR, invalid state detected.\n"); -+ -+ out: -+ fist_copy_attr_times(dir, itohi2(dentry->d_parent->d_inode)); -+ -+ if(!err) { -+ /* is this causing my pain? d_delete(dentry); */ -+ d_drop(dentry); -+ } -+ -+ dput(dentry); -+ return err; -+} -+ -+ -+STATIC int -+mini_fo_symlink(inode_t *dir, dentry_t *dentry, const char *symname) -+{ -+ int err=0; -+ dentry_t *hidden_sto_dentry; -+ dentry_t *hidden_sto_dir_dentry; -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,0) -+ umode_t mode; -+#endif -+ -+ /* Fail if the symlink file exists */ -+ if(!(dtost(dentry) == DELETED || -+ dtost(dentry) == NON_EXISTANT)) { -+ err = -EEXIST; -+ goto out; -+ } -+ -+ err = get_neg_sto_dentry(dentry); -+ if(err) { -+ err = -EINVAL; -+ goto out; -+ } -+ hidden_sto_dentry = dtohd2(dentry); -+ -+ dget(hidden_sto_dentry); -+ /* was: hidden_sto_dir_dentry = lock_parent(hidden_sto_dentry); */ -+ hidden_sto_dir_dentry = dget(hidden_sto_dentry->d_parent); -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,16) -+ mutex_lock(&hidden_sto_dir_dentry->d_inode->i_mutex); -+#else -+ down(&hidden_sto_dir_dentry->d_inode->i_sem); -+#endif -+ -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,0) -+ mode = S_IALLUGO; -+ err = vfs_symlink(hidden_sto_dir_dentry->d_inode, -+ hidden_sto_dentry, symname, mode); -+#else -+ err = vfs_symlink(hidden_sto_dir_dentry->d_inode, -+ hidden_sto_dentry, -+ symname); -+#endif -+ if (err || !hidden_sto_dentry->d_inode) -+ goto out_lock; -+ -+ if(dtost(dentry) == DELETED) { -+ dtost(dentry) = DEL_REWRITTEN; -+ err = mini_fo_tri_interpose(NULL, hidden_sto_dentry, dentry, dir->i_sb, 0); -+ if(err) -+ goto out_lock; -+ } else if(dtost(dentry) == NON_EXISTANT) { -+ dtost(dentry) = CREATED; -+ err = mini_fo_tri_interpose(dtohd(dentry), hidden_sto_dentry, dentry, dir->i_sb, 0); -+ if(err) -+ goto out_lock; -+ } -+ fist_copy_attr_timesizes(dir, hidden_sto_dir_dentry->d_inode); -+ -+ out_lock: -+ /* was: unlock_dir(hidden_sto_dir_dentry); */ -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,16) -+ mutex_unlock(&hidden_sto_dir_dentry->d_inode->i_mutex); -+#else -+ up(&hidden_sto_dir_dentry->d_inode->i_sem); -+#endif -+ dput(hidden_sto_dir_dentry); -+ -+ dput(hidden_sto_dentry); -+ if (!dentry->d_inode) -+ d_drop(dentry); -+ out: -+ return err; -+} -+ -+STATIC int -+mini_fo_mkdir(inode_t *dir, dentry_t *dentry, int mode) -+{ -+ int err; -+ -+ err = create_sto_dir(dentry, mode); -+ -+ check_mini_fo_dentry(dentry); -+ -+ return err; -+} -+ -+ -+STATIC int -+mini_fo_rmdir(inode_t *dir, dentry_t *dentry) -+{ -+ int err = 0; -+ -+ dentry_t *hidden_sto_dentry; -+ dentry_t *hidden_sto_dir_dentry; -+ dentry_t *meta_dentry; -+ inode_t *hidden_sto_dir = NULL; -+ -+ check_mini_fo_dentry(dentry); -+ check_mini_fo_inode(dir); -+ -+ dget(dentry); -+ if(dtopd(dentry)->state == MODIFIED) { -+ /* XXX: disabled, because it does not bother to check files on -+ * the original filesystem - just a hack, but better than simply -+ * removing it without testing */ -+ err = -EINVAL; -+ goto out; -+ -+ hidden_sto_dir = itohi2(dir); -+ hidden_sto_dentry = dtohd2(dentry); -+ -+ /* was:hidden_sto_dir_dentry = lock_parent(hidden_sto_dentry); */ -+ hidden_sto_dir_dentry = dget(hidden_sto_dentry->d_parent); -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,16) -+ mutex_lock(&hidden_sto_dir_dentry->d_inode->i_mutex); -+#else -+ down(&hidden_sto_dir_dentry->d_inode->i_sem); -+#endif -+ -+ /* Delete an old WOL file contained in the storage dir */ -+ meta_dentry = lookup_one_len(META_FILENAME, -+ hidden_sto_dentry, -+ strlen(META_FILENAME)); -+ if(meta_dentry->d_inode) { -+ err = vfs_unlink(hidden_sto_dentry->d_inode, meta_dentry); -+ dput(meta_dentry); -+ if(!err) -+ d_delete(meta_dentry); -+ } -+ -+ err = vfs_rmdir(hidden_sto_dir, hidden_sto_dentry); -+ dput(hidden_sto_dentry); -+ if(!err) -+ d_delete(hidden_sto_dentry); -+ -+ /* propagate number of hard-links */ -+ dentry->d_inode->i_nlink = itohi2(dentry->d_inode)->i_nlink; -+ -+ dput(dtohd(dentry)); -+ -+ dtohd(dentry) = NULL; -+ dtopd(dentry)->state = DELETED; -+ -+ /* carefull with R files */ -+ if( __meta_is_r_entry(dir, -+ dentry->d_name.name, -+ dentry->d_name.len) == 1) { -+ err = meta_remove_r_entry(dentry->d_parent, -+ dentry->d_name.name, -+ dentry->d_name.len); -+ if(err) { -+ printk(KERN_CRIT "mini_fo: rmdir: meta_remove_r_entry failed.\n"); -+ goto out; -+ } -+ } -+ else { -+ /* ok, add deleted file to META */ -+ meta_add_d_entry(dentry->d_parent, -+ dentry->d_name.name, -+ dentry->d_name.len); -+ } -+ /* was: unlock_dir(hidden_sto_dir_dentry); */ -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,16) -+ mutex_unlock(&hidden_sto_dir_dentry->d_inode->i_mutex); -+#else -+ up(&hidden_sto_dir_dentry->d_inode->i_sem); -+#endif -+ dput(hidden_sto_dir_dentry); -+ goto out; -+ } -+ else if(dtopd(dentry)->state == UNMODIFIED) { -+ /* XXX: simply adding it to the delete list here is fscking dangerous! -+ * as a temporary hack, i will disable rmdir on unmodified directories -+ * for now. -+ */ -+ err = -EINVAL; -+ goto out; -+ -+ err = get_neg_sto_dentry(dentry); -+ if(err) { -+ err = -EINVAL; -+ goto out; -+ } -+ -+ /* dput base dentry, this will relase the inode and free the -+ * dentry, as we will never need it again. */ -+ dput(dtohd(dentry)); -+ dtohd(dentry) = NULL; -+ dtopd(dentry)->state = DELETED; -+ -+ /* add deleted file to META-file */ -+ meta_add_d_entry(dentry->d_parent, -+ dentry->d_name.name, -+ dentry->d_name.len); -+ goto out; -+ } -+ else if(dtopd(dentry)->state == CREATED) { -+ hidden_sto_dir = itohi2(dir); -+ hidden_sto_dentry = dtohd2(dentry); -+ -+ /* was: hidden_sto_dir_dentry = lock_parent(hidden_sto_dentry);*/ -+ hidden_sto_dir_dentry = dget(hidden_sto_dentry->d_parent); -+ -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,16) -+ mutex_lock(&hidden_sto_dir_dentry->d_inode->i_mutex); -+#else -+ down(&hidden_sto_dir_dentry->d_inode->i_sem); -+#endif -+ -+ /* Delete an old WOL file contained in the storage dir */ -+ meta_dentry = lookup_one_len(META_FILENAME, -+ hidden_sto_dentry, -+ strlen(META_FILENAME)); -+ if(meta_dentry->d_inode) { -+ /* is this necessary? dget(meta_dentry); */ -+ err = vfs_unlink(hidden_sto_dentry->d_inode, -+ meta_dentry); -+ dput(meta_dentry); -+ if(!err) -+ d_delete(meta_dentry); -+ } -+ -+ err = vfs_rmdir(hidden_sto_dir, hidden_sto_dentry); -+ dput(hidden_sto_dentry); -+ if(!err) -+ d_delete(hidden_sto_dentry); -+ -+ /* propagate number of hard-links */ -+ dentry->d_inode->i_nlink = itohi2(dentry->d_inode)->i_nlink; -+ dtopd(dentry)->state = NON_EXISTANT; -+ -+ /* was: unlock_dir(hidden_sto_dir_dentry); */ -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,16) -+ mutex_unlock(&hidden_sto_dir_dentry->d_inode->i_mutex); -+#else -+ up(&hidden_sto_dir_dentry->d_inode->i_sem); -+#endif -+ dput(hidden_sto_dir_dentry); -+ -+ goto out; -+ } -+ else if(dtopd(dentry)->state == DEL_REWRITTEN) { -+ hidden_sto_dir = itohi2(dir); -+ hidden_sto_dentry = dtohd2(dentry); -+ -+ /* was: hidden_sto_dir_dentry = lock_parent(hidden_sto_dentry);*/ -+ hidden_sto_dir_dentry = dget(hidden_sto_dentry->d_parent); -+ -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,16) -+ mutex_lock(&hidden_sto_dir_dentry->d_inode->i_mutex); -+#else -+ down(&hidden_sto_dir_dentry->d_inode->i_sem); -+#endif -+ -+ /* Delete an old WOL file contained in the storage dir */ -+ meta_dentry = lookup_one_len(META_FILENAME, -+ hidden_sto_dentry, -+ strlen(META_FILENAME)); -+ if(meta_dentry->d_inode) { -+ /* is this necessary? dget(meta_dentry); */ -+ err = vfs_unlink(hidden_sto_dentry->d_inode, -+ meta_dentry); -+ dput(meta_dentry); -+ if(!err) -+ d_delete(meta_dentry); -+ } -+ -+ err = vfs_rmdir(hidden_sto_dir, hidden_sto_dentry); -+ dput(hidden_sto_dentry); -+ if(!err) -+ d_delete(hidden_sto_dentry); -+ -+ /* propagate number of hard-links */ -+ dentry->d_inode->i_nlink = itohi2(dentry->d_inode)->i_nlink; -+ dtopd(dentry)->state = DELETED; -+ /* was: unlock_dir(hidden_sto_dir_dentry); */ -+ -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,16) -+ mutex_unlock(&hidden_sto_dir_dentry->d_inode->i_mutex); -+#else -+ up(&hidden_sto_dir_dentry->d_inode->i_sem); -+#endif -+ dput(hidden_sto_dir_dentry); -+ goto out; -+ } -+ -+ printk(KERN_CRIT "mini_fo_rmdir: ERROR, invalid state detected.\n"); -+ -+ out: -+ if(!err) { -+ d_drop(dentry); -+ } -+ -+ fist_copy_attr_times(dir, itohi2(dentry->d_parent->d_inode)); -+ dput(dentry); -+ -+ return err; -+} -+ -+ -+STATIC int -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,0) -+mini_fo_mknod(inode_t *dir, dentry_t *dentry, int mode, dev_t dev) -+#else -+mini_fo_mknod(inode_t *dir, dentry_t *dentry, int mode, int dev) -+#endif -+{ -+ int err = 0; -+ -+ check_mini_fo_dentry(dentry); -+ -+ err = create_sto_nod(dentry, mode, dev); -+ if(err) { -+ printk(KERN_CRIT "mini_fo_mknod: creating sto nod failed.\n"); -+ err = -EINVAL; -+ } -+ -+ check_mini_fo_dentry(dentry); -+ return err; -+} -+ -+ -+STATIC int -+mini_fo_rename(inode_t *old_dir, dentry_t *old_dentry, -+ inode_t *new_dir, dentry_t *new_dentry) -+{ -+ /* dispatch */ -+ if(S_ISDIR(old_dentry->d_inode->i_mode)) -+ return rename_directory(old_dir, old_dentry, new_dir, new_dentry); -+ return rename_nondir(old_dir, old_dentry, new_dir, new_dentry); -+ -+} -+ -+int rename_directory(inode_t *old_dir, dentry_t *old_dentry, -+ inode_t *new_dir, dentry_t *new_dentry) -+{ -+ int err, bpath_len; -+ char *bpath; -+ -+ dentry_t *hidden_old_dentry; -+ dentry_t *hidden_new_dentry; -+ dentry_t *hidden_old_dir_dentry; -+ dentry_t *hidden_new_dir_dentry; -+ -+ err = 0; -+ bpath = NULL; -+ bpath_len = 0; -+ -+ /* this is a test, chuck out if it works */ -+ if(!(dtopd(new_dentry)->state == DELETED || -+ dtopd(new_dentry)->state == NON_EXISTANT)) { -+ printk(KERN_CRIT "mini_fo: rename_directory: \ -+ uh, ah, new_dentry not negative.\n"); -+ /* return -1; */ -+ } -+ -+ /* state = UNMODIFIED */ -+ if(dtopd(old_dentry)->state == UNMODIFIED) { -+ err = dir_unmod_to_mod(old_dentry); -+ if (err) -+ goto out; -+ } -+ -+ /* state = MODIFIED */ -+ if(dtopd(old_dentry)->state == MODIFIED) { -+ bpath = meta_check_r_entry(old_dentry->d_parent, -+ old_dentry->d_name.name, -+ old_dentry->d_name.len); -+ if(bpath) { -+ err = meta_remove_r_entry(old_dentry->d_parent, -+ old_dentry->d_name.name, -+ old_dentry->d_name.len); -+ if(err) { -+ printk(KERN_CRIT "mini_fo: rename_directory:\ -+ meta_remove_r_entry \ -+ failed.\n"); -+ goto out; -+ } -+ err = meta_add_r_entry(new_dentry->d_parent, -+ bpath, -+ strlen(bpath), -+ new_dentry->d_name.name, -+ new_dentry->d_name.len); -+ kfree(bpath); -+ } -+ else {/* wol it */ -+ err = meta_add_d_entry(old_dentry->d_parent, -+ old_dentry->d_name.name, -+ old_dentry->d_name.len); -+ if (err) -+ goto out; -+ /* put it on rename list */ -+ err = get_mini_fo_bpath(old_dentry, -+ &bpath, -+ &bpath_len); -+ if (err) -+ goto out; -+ err = meta_add_r_entry(new_dentry->d_parent, -+ bpath, bpath_len, -+ new_dentry->d_name.name, -+ new_dentry->d_name.len); -+ if (err) -+ goto out; -+ } -+ /* no state change, MODIFIED stays MODIFIED */ -+ } -+ /* state = CREATED */ -+ if(dtopd(old_dentry)->state == CREATED || -+ dtopd(old_dentry)->state == DEL_REWRITTEN) { -+ if(dtohd(old_dentry)) -+ dput(dtohd(old_dentry)); -+ -+ if(dtopd(new_dentry)->state == DELETED) { -+ dtopd(old_dentry)->state = DEL_REWRITTEN; -+ dtohd(old_dentry) = NULL; -+ } -+ else if(dtopd(new_dentry)->state == NON_EXISTANT) { -+ dtopd(old_dentry)->state = CREATED; -+ /* steal new dentry's neg. base dentry */ -+ dtohd(old_dentry) = dtohd(new_dentry); -+ dtohd(new_dentry) = NULL; -+ } -+ } -+ if(dtopd(new_dentry)->state == UNMODIFIED || -+ dtopd(new_dentry)->state == NON_EXISTANT) { -+ err = get_neg_sto_dentry(new_dentry); -+ if(err) -+ goto out; -+ } -+ -+ /* now move sto file */ -+ hidden_old_dentry = dtohd2(old_dentry); -+ hidden_new_dentry = dtohd2(new_dentry); -+ -+ dget(hidden_old_dentry); -+ dget(hidden_new_dentry); -+ -+ hidden_old_dir_dentry = dget(hidden_old_dentry->d_parent); -+ hidden_new_dir_dentry = dget(hidden_new_dentry->d_parent); -+ double_lock(hidden_old_dir_dentry, hidden_new_dir_dentry); -+ -+ err = vfs_rename(hidden_old_dir_dentry->d_inode, hidden_old_dentry, -+ hidden_new_dir_dentry->d_inode, hidden_new_dentry); -+ if(err) -+ goto out_lock; -+ -+ fist_copy_attr_all(new_dir, hidden_new_dir_dentry->d_inode); -+ if (new_dir != old_dir) -+ fist_copy_attr_all(old_dir, -+ hidden_old_dir_dentry->d_inode); -+ -+ out_lock: -+ /* double_unlock will dput the new/old parent dentries -+ * whose refcnts were incremented via get_parent above. */ -+ double_unlock(hidden_old_dir_dentry, hidden_new_dir_dentry); -+ dput(hidden_new_dentry); -+ dput(hidden_old_dentry); -+ -+ out: -+ return err; -+} -+ -+int rename_nondir(inode_t *old_dir, dentry_t *old_dentry, -+ inode_t *new_dir, dentry_t *new_dentry) -+{ -+ int err=0; -+ -+ check_mini_fo_dentry(old_dentry); -+ check_mini_fo_dentry(new_dentry); -+ check_mini_fo_inode(old_dir); -+ check_mini_fo_inode(new_dir); -+ -+ /* state: UNMODIFIED */ -+ if(dtost(old_dentry) == UNMODIFIED) { -+ err = nondir_unmod_to_mod(old_dentry, 1); -+ if(err) { -+ err = -EINVAL; -+ goto out; -+ } -+ } -+ -+ /* the easy states */ -+ if(exists_in_storage(old_dentry)) { -+ -+ dentry_t *hidden_old_dentry; -+ dentry_t *hidden_new_dentry; -+ dentry_t *hidden_old_dir_dentry; -+ dentry_t *hidden_new_dir_dentry; -+ -+ /* if old file is MODIFIED, add it to the deleted_list */ -+ if(dtopd(old_dentry)->state == MODIFIED) { -+ meta_add_d_entry(old_dentry->d_parent, -+ old_dentry->d_name.name, -+ old_dentry->d_name.len); -+ -+ dput(dtohd(old_dentry)); -+ } -+ /* if old file is CREATED, we only release the base dentry */ -+ if(dtopd(old_dentry)->state == CREATED) { -+ if(dtohd(old_dentry)) -+ dput(dtohd(old_dentry)); -+ } -+ -+ /* now setup the new states (depends on new_dentry state) */ -+ /* new dentry state = MODIFIED */ -+ if(dtopd(new_dentry)->state == MODIFIED) { -+ meta_add_d_entry(new_dentry->d_parent, -+ new_dentry->d_name.name, -+ new_dentry->d_name.len); -+ -+ /* new dentry will be d_put'ed later by the vfs -+ * so don't do it here -+ * dput(dtohd(new_dentry)); -+ */ -+ dtohd(old_dentry) = NULL; -+ dtopd(old_dentry)->state = DEL_REWRITTEN; -+ } -+ /* new dentry state = UNMODIFIED */ -+ else if(dtopd(new_dentry)->state == UNMODIFIED) { -+ if(get_neg_sto_dentry(new_dentry)) -+ return -EINVAL; -+ -+ meta_add_d_entry(new_dentry->d_parent, -+ new_dentry->d_name.name, -+ new_dentry->d_name.len); -+ -+ /* is this right??? */ -+ /*dput(dtohd(new_dentry));*/ -+ dtohd(old_dentry) = NULL; -+ dtopd(old_dentry)->state = DEL_REWRITTEN; -+ } -+ /* new dentry state = CREATED */ -+ else if(dtopd(new_dentry)->state == CREATED) { -+ /* we keep the neg. base dentry (if exists) */ -+ dtohd(old_dentry) = dtohd(new_dentry); -+ /* ...and set it to Null, or we'll get -+ * dcache.c:345 if it gets dput twice... */ -+ dtohd(new_dentry) = NULL; -+ dtopd(old_dentry)->state = CREATED; -+ } -+ /* new dentry state = NON_EXISTANT */ -+ else if(dtopd(new_dentry)->state == NON_EXISTANT) { -+ if(get_neg_sto_dentry(new_dentry)) -+ return -EINVAL; -+ -+ /* we keep the neg. base dentry (if exists) */ -+ dtohd(old_dentry) = dtohd(new_dentry); -+ /* ...and set it to Null, or we'll get -+ * Dr. dcache.c:345 if it gets dput twice... */ -+ dtohd(new_dentry) = NULL; -+ dtopd(old_dentry)->state = CREATED; -+ } -+ /* new dentry state = DEL_REWRITTEN or DELETED */ -+ else if(dtopd(new_dentry)->state == DEL_REWRITTEN || -+ dtopd(new_dentry)->state == DELETED) { -+ dtohd(old_dentry) = NULL; -+ dtopd(old_dentry)->state = DEL_REWRITTEN; -+ } -+ else { /* not possible, uhh, ahh */ -+ printk(KERN_CRIT -+ "mini_fo: rename_reg_file: invalid state detected [1].\n"); -+ return -1; -+ } -+ -+ /* now we definitely have a sto file */ -+ hidden_old_dentry = dtohd2(old_dentry); -+ hidden_new_dentry = dtohd2(new_dentry); -+ -+ dget(hidden_old_dentry); -+ dget(hidden_new_dentry); -+ -+ hidden_old_dir_dentry = dget(hidden_old_dentry->d_parent); -+ hidden_new_dir_dentry = dget(hidden_new_dentry->d_parent); -+ double_lock(hidden_old_dir_dentry, hidden_new_dir_dentry); -+ -+ err = vfs_rename(hidden_old_dir_dentry->d_inode, -+ hidden_old_dentry, -+ hidden_new_dir_dentry->d_inode, -+ hidden_new_dentry); -+ if(err) -+ goto out_lock; -+ -+ fist_copy_attr_all(new_dir, hidden_new_dir_dentry->d_inode); -+ if (new_dir != old_dir) -+ fist_copy_attr_all(old_dir, hidden_old_dir_dentry->d_inode); -+ -+ out_lock: -+ /* double_unlock will dput the new/old parent dentries -+ * whose refcnts were incremented via get_parent above. -+ */ -+ double_unlock(hidden_old_dir_dentry, hidden_new_dir_dentry); -+ dput(hidden_new_dentry); -+ dput(hidden_old_dentry); -+ out: -+ return err; -+ } -+ else { /* invalid state */ -+ printk(KERN_CRIT "mini_fo: rename_reg_file: ERROR: invalid state detected [2].\n"); -+ return -1; -+ } -+} -+ -+ -+STATIC int -+mini_fo_readlink(dentry_t *dentry, char *buf, int bufsiz) -+{ -+ int err=0; -+ dentry_t *hidden_dentry = NULL; -+ -+ if(dtohd2(dentry) && dtohd2(dentry)->d_inode) { -+ hidden_dentry = dtohd2(dentry); -+ } else if(dtohd(dentry) && dtohd(dentry)->d_inode) { -+ hidden_dentry = dtohd(dentry); -+ } else { -+ goto out; -+ } -+ -+ if (!hidden_dentry->d_inode->i_op || -+ !hidden_dentry->d_inode->i_op->readlink) { -+ err = -EINVAL; goto out; -+ } -+ -+ err = hidden_dentry->d_inode->i_op->readlink(hidden_dentry, -+ buf, -+ bufsiz); -+ if (err > 0) -+ fist_copy_attr_atime(dentry->d_inode, hidden_dentry->d_inode); -+ -+ out: -+ return err; -+} -+ -+ -+#if LINUX_VERSION_CODE < KERNEL_VERSION(2,6,13) -+static int mini_fo_follow_link(dentry_t *dentry, struct nameidata *nd) -+#else -+static void* mini_fo_follow_link(dentry_t *dentry, struct nameidata *nd) -+#endif -+{ -+ char *buf; -+ int len = PAGE_SIZE, err; -+ mm_segment_t old_fs; -+ -+ /* in 2.6 this is freed by mini_fo_put_link called by __do_follow_link */ -+ buf = kmalloc(len, GFP_KERNEL); -+ if (!buf) { -+ err = -ENOMEM; -+ goto out; -+ } -+ -+ /* read the symlink, and then we will follow it */ -+ old_fs = get_fs(); -+ set_fs(KERNEL_DS); -+ err = dentry->d_inode->i_op->readlink(dentry, buf, len); -+ set_fs(old_fs); -+ if (err < 0) { -+ kfree(buf); -+ buf = NULL; -+ goto out; -+ } -+ buf[err] = 0; -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,0) -+ nd_set_link(nd, buf); -+ err = 0; -+#else -+ err = vfs_follow_link(nd, buf); -+#endif -+ -+ out: -+#if LINUX_VERSION_CODE < KERNEL_VERSION(2,6,0) -+ kfree(buf); -+#endif -+ -+#if LINUX_VERSION_CODE < KERNEL_VERSION(2,6,13) -+ return err; -+#else -+ return ERR_PTR(err); -+#endif -+} -+ -+STATIC -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,0) -+#if LINUX_VERSION_CODE < KERNEL_VERSION(2,6,13) -+void mini_fo_put_link(struct dentry *dentry, struct nameidata *nd) -+#else -+void mini_fo_put_link(struct dentry *dentry, struct nameidata *nd, void *cookie) -+#endif -+{ -+ char *link; -+ link = nd_get_link(nd); -+ kfree(link); -+} -+#endif -+ -+STATIC int -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,0) -+mini_fo_permission(inode_t *inode, int mask, struct nameidata *nd) -+#else -+mini_fo_permission(inode_t *inode, int mask) -+#endif -+{ -+ inode_t *hidden_inode; -+ int mode; -+ int err; -+ -+ if(itohi2(inode)) { -+ hidden_inode = itohi2(inode); -+ } else { -+ hidden_inode = itohi(inode); -+ } -+ mode = inode->i_mode; -+ -+ /* not really needed, as permission handles everything: -+ * err = vfs_permission(inode, mask); -+ * if (err) -+ * goto out; -+ */ -+ -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,0) -+ err = permission(hidden_inode, mask, nd); -+#else -+ err = permission(hidden_inode, mask); -+#endif -+ -+ /* out: */ -+ return err; -+} -+ -+#if LINUX_VERSION_CODE < KERNEL_VERSION(2,6,0) -+STATIC int -+mini_fo_inode_revalidate(dentry_t *dentry) -+{ -+ int err = 0; -+ dentry_t *hidden_dentry; -+ inode_t *hidden_inode; -+ -+ ASSERT(dentry->d_inode); -+ ASSERT(itopd(dentry->d_inode)); -+ -+ if(itohi2(dentry->d_inode)) { -+ hidden_dentry = dtohd2(dentry); -+ hidden_inode = hidden_dentry->d_inode; -+ } else if(itohi(dentry->d_inode)) { -+ hidden_dentry = dtohd(dentry); -+ hidden_inode = hidden_dentry->d_inode; -+ } else { -+ printk(KERN_CRIT "mini_fo_inode_revalidate: ERROR, invalid state detected.\n"); -+ err = -ENOENT; -+ goto out; -+ } -+ if (hidden_inode && hidden_inode->i_op && hidden_inode->i_op->revalidate){ -+ err = hidden_inode->i_op->revalidate(hidden_dentry); -+ if (err) -+ goto out; -+ } -+ fist_copy_attr_all(dentry->d_inode, hidden_inode); -+ out: -+ return err; -+} -+#endif -+ -+STATIC int -+mini_fo_setattr(dentry_t *dentry, struct iattr *ia) -+{ -+ int err = 0; -+ -+ check_mini_fo_dentry(dentry); -+ -+ if(!is_mini_fo_existant(dentry)) { -+ printk(KERN_CRIT "mini_fo_setattr: ERROR, invalid state detected [1].\n"); -+ goto out; -+ } -+ -+ if(dtost(dentry) == UNMODIFIED) { -+ if(!IS_COPY_FLAG(ia->ia_valid)) -+ goto out; /* we ignore these changes to base */ -+ -+ if(S_ISDIR(dentry->d_inode->i_mode)) { -+ err = dir_unmod_to_mod(dentry); -+ } else { -+ /* we copy contents if file is not beeing truncated */ -+ if(S_ISREG(dentry->d_inode->i_mode) && -+ !(ia->ia_size == 0 && (ia->ia_valid & ATTR_SIZE))) { -+ err = nondir_unmod_to_mod(dentry, 1); -+ } else -+ err = nondir_unmod_to_mod(dentry, 0); -+ } -+ if(err) { -+ err = -EINVAL; -+ printk(KERN_CRIT "mini_fo_setattr: ERROR changing states.\n"); -+ goto out; -+ } -+ } -+ if(!exists_in_storage(dentry)) { -+ printk(KERN_CRIT "mini_fo_setattr: ERROR, invalid state detected [2].\n"); -+ err = -EINVAL; -+ goto out; -+ } -+ ASSERT(dentry->d_inode); -+ ASSERT(dtohd2(dentry)); -+ ASSERT(itopd(dentry->d_inode)); -+ ASSERT(itohi2(dentry->d_inode)); -+ -+ err = notify_change(dtohd2(dentry), ia); -+ fist_copy_attr_all(dentry->d_inode, itohi2(dentry->d_inode)); -+ out: -+ return err; -+} -+ -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,0) -+STATIC int -+mini_fo_getattr(struct vfsmount *mnt, struct dentry *dentry, struct kstat *stat) -+{ -+ int err = 0; -+ dentry_t *hidden_dentry; -+ -+ ASSERT(dentry->d_inode); -+ ASSERT(itopd(dentry->d_inode)); -+ -+ if(itohi2(dentry->d_inode)) { -+ hidden_dentry = dtohd2(dentry); -+ } else if(itohi(dentry->d_inode)) { -+ hidden_dentry = dtohd(dentry); -+ } else { -+ printk(KERN_CRIT "mini_fo_getattr: ERROR, invalid state detected.\n"); -+ err = -ENOENT; -+ goto out; -+ } -+ fist_copy_attr_all(dentry->d_inode, hidden_dentry->d_inode); -+ -+ ASSERT(hidden_dentry); -+ ASSERT(hidden_dentry->d_inode); -+ ASSERT(hidden_dentry->d_inode->i_op); -+ -+ generic_fillattr(dentry->d_inode, stat); -+ if (!stat->blksize) { -+ struct super_block *s = hidden_dentry->d_inode->i_sb; -+ unsigned blocks; -+ blocks = (stat->size+s->s_blocksize-1) >> s->s_blocksize_bits; -+ stat->blocks = (s->s_blocksize / 512) * blocks; -+ stat->blksize = s->s_blocksize; -+ } -+ out: -+ return err; -+} -+#endif -+ -+#if defined(XATTR) && (LINUX_VERSION_CODE >= KERNEL_VERSION(2,4,20)) -+#if 0 /* no xattr_alloc() and xattr_free() */ -+/* This is lifted from fs/xattr.c */ -+static void * -+xattr_alloc(size_t size, size_t limit) -+{ -+ void *ptr; -+ -+ if (size > limit) -+ return ERR_PTR(-E2BIG); -+ -+ if (!size) /* size request, no buffer is needed */ -+ return NULL; -+ else if (size <= PAGE_SIZE) -+ ptr = kmalloc((unsigned long) size, GFP_KERNEL); -+ else -+ ptr = vmalloc((unsigned long) size); -+ if (!ptr) -+ return ERR_PTR(-ENOMEM); -+ return ptr; -+} -+ -+static void -+xattr_free(void *ptr, size_t size) -+{ -+ if (!size) /* size request, no buffer was needed */ -+ return; -+ else if (size <= PAGE_SIZE) -+ kfree(ptr); -+ else -+ vfree(ptr); -+} -+#endif /* no xattr_alloc() and xattr_free() */ -+ -+/* BKL held by caller. -+ * dentry->d_inode->i_sem down -+ */ -+STATIC int -+mini_fo_getxattr(struct dentry *dentry, const char *name, void *value, size_t size) { -+ struct dentry *hidden_dentry = NULL; -+ int err = -EOPNOTSUPP; -+ /* Define these anyway so we don't need as much ifdef'ed code. */ -+ char *encoded_name = NULL; -+ char *encoded_value = NULL; -+ -+ check_mini_fo_dentry(dentry); -+ -+ if(exists_in_storage(dentry)) -+ hidden_dentry = dtohd2(dentry); -+ else -+ hidden_dentry = dtohd(dentry); -+ -+ ASSERT(hidden_dentry); -+ ASSERT(hidden_dentry->d_inode); -+ ASSERT(hidden_dentry->d_inode->i_op); -+ -+ if (hidden_dentry->d_inode->i_op->getxattr) { -+ encoded_name = (char *)name; -+ encoded_value = (char *)value; -+ -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,16) -+ mutex_lock(&hidden_dentry->d_inode->i_mutex); -+#else -+ down(&hidden_dentry->d_inode->i_sem); -+#endif -+ /* lock_kernel() already done by caller. */ -+ err = hidden_dentry->d_inode->i_op->getxattr(hidden_dentry, encoded_name, encoded_value, size); -+ /* unlock_kernel() will be done by caller. */ -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,16) -+ mutex_lock(&hidden_dentry->d_inode->i_mutex); -+#else -+ up(&hidden_dentry->d_inode->i_sem); -+#endif -+ } -+ return err; -+} -+ -+/* BKL held by caller. -+ * dentry->d_inode->i_sem down -+ */ -+STATIC int -+#if ((LINUX_VERSION_CODE >= KERNEL_VERSION(2,4,21) \ -+ && LINUX_VERSION_CODE <= KERNEL_VERSION(2,4,23)) \ -+ || LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,0)) -+mini_fo_setxattr(struct dentry *dentry, const char *name, -+ const void *value, size_t size, int flags) -+#else -+mini_fo_setxattr(struct dentry *dentry, const char *name, -+ void *value, size_t size, int flags) -+#endif -+ -+{ -+ struct dentry *hidden_dentry = NULL; -+ int err = -EOPNOTSUPP; -+ -+ /* Define these anyway, so we don't have as much ifdef'ed code. */ -+ char *encoded_value = NULL; -+ char *encoded_name = NULL; -+ -+ check_mini_fo_dentry(dentry); -+ -+ if(exists_in_storage(dentry)) -+ hidden_dentry = dtohd2(dentry); -+ else -+ hidden_dentry = dtohd(dentry); -+ -+ ASSERT(hidden_dentry); -+ ASSERT(hidden_dentry->d_inode); -+ ASSERT(hidden_dentry->d_inode->i_op); -+ -+ if (hidden_dentry->d_inode->i_op->setxattr) { -+ encoded_name = (char *)name; -+ encoded_value = (char *)value; -+ -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,16) -+ mutex_lock(&hidden_dentry->d_inode->i_mutex); -+#else -+ down(&hidden_dentry->d_inode->i_sem); -+#endif -+ /* lock_kernel() already done by caller. */ -+ err = hidden_dentry->d_inode->i_op->setxattr(hidden_dentry, encoded_name, encoded_value, size, flags); -+ /* unlock_kernel() will be done by caller. */ -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,16) -+ mutex_unlock(&hidden_dentry->d_inode->i_mutex); -+#else -+ up(&hidden_dentry->d_inode->i_sem); -+#endif -+ } -+ return err; -+} -+ -+/* BKL held by caller. -+ * dentry->d_inode->i_sem down -+ */ -+STATIC int -+mini_fo_removexattr(struct dentry *dentry, const char *name) { -+ struct dentry *hidden_dentry = NULL; -+ int err = -EOPNOTSUPP; -+ char *encoded_name; -+ -+ check_mini_fo_dentry(dentry); -+ -+ if(exists_in_storage(dentry)) -+ hidden_dentry = dtohd2(dentry); -+ else -+ hidden_dentry = dtohd(dentry); -+ -+ ASSERT(hidden_dentry); -+ ASSERT(hidden_dentry->d_inode); -+ ASSERT(hidden_dentry->d_inode->i_op); -+ -+ if (hidden_dentry->d_inode->i_op->removexattr) { -+ encoded_name = (char *)name; -+ -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,16) -+ mutex_lock(&hidden_dentry->d_inode->i_mutex); -+#else -+ down(&hidden_dentry->d_inode->i_sem); -+#endif -+ /* lock_kernel() already done by caller. */ -+ err = hidden_dentry->d_inode->i_op->removexattr(hidden_dentry, encoded_name); -+ /* unlock_kernel() will be done by caller. */ -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,16) -+ mutex_unlock(&hidden_dentry->d_inode->i_mutex); -+#else -+ up(&hidden_dentry->d_inode->i_sem); -+#endif -+ } -+ return err; -+} -+ -+/* BKL held by caller. -+ * dentry->d_inode->i_sem down -+ */ -+STATIC int -+mini_fo_listxattr(struct dentry *dentry, char *list, size_t size) { -+ struct dentry *hidden_dentry = NULL; -+ int err = -EOPNOTSUPP; -+ char *encoded_list = NULL; -+ -+ check_mini_fo_dentry(dentry); -+ -+ if(exists_in_storage(dentry)) -+ hidden_dentry = dtohd2(dentry); -+ else -+ hidden_dentry = dtohd(dentry); -+ -+ ASSERT(hidden_dentry); -+ ASSERT(hidden_dentry->d_inode); -+ ASSERT(hidden_dentry->d_inode->i_op); -+ -+ if (hidden_dentry->d_inode->i_op->listxattr) { -+ encoded_list = list; -+ -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,16) -+ mutex_lock(&hidden_dentry->d_inode->i_mutex); -+#else -+ down(&hidden_dentry->d_inode->i_sem); -+#endif -+ /* lock_kernel() already done by caller. */ -+ err = hidden_dentry->d_inode->i_op->listxattr(hidden_dentry, encoded_list, size); -+ /* unlock_kernel() will be done by caller. */ -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,16) -+ mutex_unlock(&hidden_dentry->d_inode->i_mutex); -+#else -+ up(&hidden_dentry->d_inode->i_sem); -+#endif -+ } -+ return err; -+} -+# endif /* defined(XATTR) && (LINUX_VERSION_CODE >= KERNEL_VERSION(2,4,20)) */ -+ -+struct inode_operations mini_fo_symlink_iops = -+ { -+ readlink: mini_fo_readlink, -+ follow_link: mini_fo_follow_link, -+ /* mk: permission: mini_fo_permission, */ -+#if LINUX_VERSION_CODE < KERNEL_VERSION(2,6,0) -+ revalidate: mini_fo_inode_revalidate, -+#endif -+ setattr: mini_fo_setattr, -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,0) -+ getattr: mini_fo_getattr, -+ put_link: mini_fo_put_link, -+#endif -+ -+#if defined(XATTR) && (LINUX_VERSION_CODE >= KERNEL_VERSION(2,4,20)) -+ setxattr: mini_fo_setxattr, -+ getxattr: mini_fo_getxattr, -+ listxattr: mini_fo_listxattr, -+ removexattr: mini_fo_removexattr -+# endif /* defined(XATTR) && (LINUX_VERSION_CODE >= KERNEL_VERSION(2,4,20)) */ -+ }; -+ -+struct inode_operations mini_fo_dir_iops = -+ { -+ create: mini_fo_create, -+ lookup: mini_fo_lookup, -+ link: mini_fo_link, -+ unlink: mini_fo_unlink, -+ symlink: mini_fo_symlink, -+ mkdir: mini_fo_mkdir, -+ rmdir: mini_fo_rmdir, -+ mknod: mini_fo_mknod, -+ rename: mini_fo_rename, -+ /* no readlink/follow_link for non-symlinks */ -+ // off because we have setattr -+ // truncate: mini_fo_truncate, -+ /* mk:permission: mini_fo_permission, */ -+#if LINUX_VERSION_CODE < KERNEL_VERSION(2,6,0) -+ revalidate: mini_fo_inode_revalidate, -+#endif -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,0) -+ getattr: mini_fo_getattr, -+#endif -+ setattr: mini_fo_setattr, -+#if defined(XATTR) && (LINUX_VERSION_CODE >= KERNEL_VERSION(2,4,20)) -+ setxattr: mini_fo_setxattr, -+ getxattr: mini_fo_getxattr, -+ listxattr: mini_fo_listxattr, -+ removexattr: mini_fo_removexattr -+# endif /* XATTR && LINUX_VERSION_CODE >= KERNEL_VERSION(2,4,20) */ -+ }; -+ -+struct inode_operations mini_fo_main_iops = -+ { -+ /* permission: mini_fo_permission, */ -+#if LINUX_VERSION_CODE < KERNEL_VERSION(2,6,0) -+ revalidate: mini_fo_inode_revalidate, -+#endif -+ setattr: mini_fo_setattr, -+ -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,0) -+ getattr: mini_fo_getattr, -+#endif -+#if defined(XATTR) && (LINUX_VERSION_CODE >= KERNEL_VERSION(2,4,20)) -+ setxattr: mini_fo_setxattr, -+ getxattr: mini_fo_getxattr, -+ listxattr: mini_fo_listxattr, -+ removexattr: mini_fo_removexattr -+# endif /* XATTR && LINUX_VERSION_CODE >= KERNEL_VERSION(2,4,20) */ -+ }; ---- /dev/null -+++ b/fs/mini_fo/main.c -@@ -0,0 +1,423 @@ -+/* -+ * Copyright (c) 1997-2003 Erez Zadok -+ * Copyright (c) 2001-2003 Stony Brook University -+ * -+ * For specific licensing information, see the COPYING file distributed with -+ * this package, or get one from ftp://ftp.filesystems.org/pub/fist/COPYING. -+ * -+ * This Copyright notice must be kept intact and distributed with all -+ * fistgen sources INCLUDING sources generated by fistgen. -+ */ -+/* -+ * Copyright (C) 2004, 2005 Markus Klotzbuecher <mk@creamnet.de> -+ * -+ * This program is free software; you can redistribute it and/or -+ * modify it under the terms of the GNU General Public License -+ * as published by the Free Software Foundation; either version -+ * 2 of the License, or (at your option) any later version. -+ */ -+ -+/* -+ * $Id$ -+ */ -+ -+#ifdef HAVE_CONFIG_H -+# include <config.h> -+#endif -+ -+#include "fist.h" -+#include "mini_fo.h" -+#include <linux/module.h> -+ -+/* This definition must only appear after we include <linux/module.h> */ -+#ifndef MODULE_LICENSE -+# define MODULE_LICENSE(bison) -+#endif /* not MODULE_LICENSE */ -+ -+/* -+ * This is the mini_fo tri interpose function, which extends the -+ * functionality of the regular interpose by interposing a higher -+ * level inode on top of two lower level ones: the base filesystem -+ * inode and the storage filesystem inode. -+ * -+ * sb we pass is mini_fo's super_block -+ */ -+int -+mini_fo_tri_interpose(dentry_t *hidden_dentry, -+ dentry_t *hidden_sto_dentry, -+ dentry_t *dentry, super_block_t *sb, int flag) -+{ -+ inode_t *hidden_inode = NULL; -+ inode_t *hidden_sto_inode = NULL; /* store corresponding storage inode */ -+ int err = 0; -+ inode_t *inode; -+ -+ /* Pointer to hidden_sto_inode if exists, else to hidden_inode. -+ * This is used to copy the attributes of the correct inode. */ -+ inode_t *master_inode; -+ -+ if(hidden_dentry) -+ hidden_inode = hidden_dentry->d_inode; -+ if(hidden_sto_dentry) -+ hidden_sto_inode = hidden_sto_dentry->d_inode; -+ -+ ASSERT(dentry->d_inode == NULL); -+ -+ /* mk: One of the inodes associated with the dentrys is likely to -+ * be NULL, so carefull: -+ */ -+ ASSERT((hidden_inode != NULL) || (hidden_sto_inode != NULL)); -+ -+ if(hidden_sto_inode) -+ master_inode = hidden_sto_inode; -+ else -+ master_inode = hidden_inode; -+ -+ /* -+ * We allocate our new inode below, by calling iget. -+ * iget will call our read_inode which will initialize some -+ * of the new inode's fields -+ */ -+ -+ /* -+ * original: inode = iget(sb, hidden_inode->i_ino); -+ */ -+ inode = iget(sb, iunique(sb, 25)); -+ if (!inode) { -+ err = -EACCES; /* should be impossible??? */ -+ goto out; -+ } -+ -+ /* -+ * interpose the inode if not already interposed -+ * this is possible if the inode is being reused -+ * XXX: what happens if we get_empty_inode() but there's another already? -+ * for now, ASSERT() that this can't happen; fix later. -+ */ -+ if (itohi(inode) != NULL) { -+ printk(KERN_CRIT "mini_fo_tri_interpose: itohi(inode) != NULL.\n"); -+ } -+ if (itohi2(inode) != NULL) { -+ printk(KERN_CRIT "mini_fo_tri_interpose: itohi2(inode) != NULL.\n"); -+ } -+ -+ /* mk: Carefull, igrab can't handle NULL inodes (ok, why should it?), so -+ * we need to check here: -+ */ -+ if(hidden_inode) -+ itohi(inode) = igrab(hidden_inode); -+ else -+ itohi(inode) = NULL; -+ -+ if(hidden_sto_inode) -+ itohi2(inode) = igrab(hidden_sto_inode); -+ else -+ itohi2(inode) = NULL; -+ -+ -+ /* Use different set of inode ops for symlinks & directories*/ -+ if (S_ISLNK(master_inode->i_mode)) -+ inode->i_op = &mini_fo_symlink_iops; -+ else if (S_ISDIR(master_inode->i_mode)) -+ inode->i_op = &mini_fo_dir_iops; -+ -+ /* Use different set of file ops for directories */ -+ if (S_ISDIR(master_inode->i_mode)) -+ inode->i_fop = &mini_fo_dir_fops; -+ -+ /* properly initialize special inodes */ -+ if (S_ISBLK(master_inode->i_mode) || S_ISCHR(master_inode->i_mode) || -+ S_ISFIFO(master_inode->i_mode) || S_ISSOCK(master_inode->i_mode)) { -+ init_special_inode(inode, master_inode->i_mode, master_inode->i_rdev); -+ } -+ -+ /* Fix our inode's address operations to that of the lower inode */ -+ if (inode->i_mapping->a_ops != master_inode->i_mapping->a_ops) { -+ inode->i_mapping->a_ops = master_inode->i_mapping->a_ops; -+ } -+ -+ /* only (our) lookup wants to do a d_add */ -+ if (flag) -+ d_add(dentry, inode); -+ else -+ d_instantiate(dentry, inode); -+ -+ ASSERT(dtopd(dentry) != NULL); -+ -+ /* all well, copy inode attributes */ -+ fist_copy_attr_all(inode, master_inode); -+ -+ out: -+ return err; -+} -+ -+/* parse mount options "base=" and "sto=" */ -+dentry_t * -+mini_fo_parse_options(super_block_t *sb, char *options) -+{ -+ dentry_t *hidden_root = ERR_PTR(-EINVAL); -+ dentry_t *hidden_root2 = ERR_PTR(-EINVAL); -+ struct nameidata nd, nd2; -+ char *name, *tmp, *end; -+ int err = 0; -+ -+ /* We don't want to go off the end of our arguments later on. */ -+ for (end = options; *end; end++); -+ -+ while (options < end) { -+ tmp = options; -+ while (*tmp && *tmp != ',') -+ tmp++; -+ *tmp = '\0'; -+ if (!strncmp("base=", options, 5)) { -+ name = options + 5; -+ printk(KERN_INFO "mini_fo: using base directory: %s\n", name); -+ -+#if LINUX_VERSION_CODE < KERNEL_VERSION(2,6,0) -+ if (path_init(name, LOOKUP_FOLLOW, &nd)) -+ err = path_walk(name, &nd); -+#else -+ err = path_lookup(name, LOOKUP_FOLLOW, &nd); -+#endif -+ if (err) { -+ printk(KERN_CRIT "mini_fo: error accessing hidden directory '%s'\n", name); -+ hidden_root = ERR_PTR(err); -+ goto out; -+ } -+ hidden_root = nd.dentry; -+ stopd(sb)->base_dir_dentry = nd.dentry; -+ stopd(sb)->hidden_mnt = nd.mnt; -+ -+ } else if(!strncmp("sto=", options, 4)) { -+ /* parse the storage dir */ -+ name = options + 4; -+ printk(KERN_INFO "mini_fo: using storage directory: %s\n", name); -+#if LINUX_VERSION_CODE < KERNEL_VERSION(2,6,0) -+ if(path_init(name, LOOKUP_FOLLOW, &nd2)) -+ err = path_walk(name, &nd2); -+#else -+ err = path_lookup(name, LOOKUP_FOLLOW, &nd2); -+#endif -+ if(err) { -+ printk(KERN_CRIT "mini_fo: error accessing hidden storage directory '%s'\n", name); -+ -+ hidden_root2 = ERR_PTR(err); -+ goto out; -+ } -+ hidden_root2 = nd2.dentry; -+ stopd(sb)->storage_dir_dentry = nd2.dentry; -+ stopd(sb)->hidden_mnt2 = nd2.mnt; -+ stohs2(sb) = hidden_root2->d_sb; -+ -+ /* validate storage dir, this is done in -+ * mini_fo_read_super for the base directory. -+ */ -+ if (IS_ERR(hidden_root2)) { -+ printk(KERN_WARNING "mini_fo_parse_options: storage dentry lookup failed (err = %ld)\n", PTR_ERR(hidden_root2)); -+ goto out; -+ } -+ if (!hidden_root2->d_inode) { -+ printk(KERN_WARNING "mini_fo_parse_options: no storage dir to interpose on.\n"); -+ goto out; -+ } -+ stohs2(sb) = hidden_root2->d_sb; -+ } else { -+ printk(KERN_WARNING "mini_fo: unrecognized option '%s'\n", options); -+ hidden_root = ERR_PTR(-EINVAL); -+ goto out; -+ } -+ options = tmp + 1; -+ } -+ -+ out: -+ if(IS_ERR(hidden_root2)) -+ return hidden_root2; -+ return hidden_root; -+} -+ -+ -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,0) -+static int -+#else -+super_block_t * -+#endif -+mini_fo_read_super(super_block_t *sb, void *raw_data, int silent) -+{ -+ dentry_t *hidden_root; -+ int err = 0; -+ -+ if (!raw_data) { -+ printk(KERN_WARNING "mini_fo_read_super: missing argument\n"); -+ err = -EINVAL; -+ goto out; -+ } -+ /* -+ * Allocate superblock private data -+ */ -+ __stopd(sb) = kmalloc(sizeof(struct mini_fo_sb_info), GFP_KERNEL); -+ if (!stopd(sb)) { -+ printk(KERN_WARNING "%s: out of memory\n", __FUNCTION__); -+ err = -ENOMEM; -+ goto out; -+ } -+ stohs(sb) = NULL; -+ -+ hidden_root = mini_fo_parse_options(sb, raw_data); -+ if (IS_ERR(hidden_root)) { -+ printk(KERN_WARNING "mini_fo_read_super: lookup_dentry failed (err = %ld)\n", PTR_ERR(hidden_root)); -+ err = PTR_ERR(hidden_root); -+ goto out_free; -+ } -+ if (!hidden_root->d_inode) { -+ printk(KERN_WARNING "mini_fo_read_super: no directory to interpose on\n"); -+ goto out_free; -+ } -+ stohs(sb) = hidden_root->d_sb; -+ -+ /* -+ * Linux 2.4.2-ac3 and beyond has code in -+ * mm/filemap.c:generic_file_write() that requires sb->s_maxbytes -+ * to be populated. If not set, all write()s under that sb will -+ * return 0. -+ * -+ * Linux 2.4.4+ automatically sets s_maxbytes to MAX_NON_LFS; -+ * the filesystem should override it only if it supports LFS. -+ */ -+ /* non-SCA code is good to go with LFS */ -+ sb->s_maxbytes = hidden_root->d_sb->s_maxbytes; -+ -+ sb->s_op = &mini_fo_sops; -+ /* -+ * we can't use d_alloc_root if we want to use -+ * our own interpose function unchanged, -+ * so we simply replicate *most* of the code in d_alloc_root here -+ */ -+#if LINUX_VERSION_CODE < KERNEL_VERSION(2,6,0) -+ sb->s_root = d_alloc(NULL, &(const struct qstr) { "/", 1, 0 }); -+#else -+ sb->s_root = d_alloc(NULL, &(const struct qstr){hash: 0, name: "/", len : 1}); -+#endif -+ if (IS_ERR(sb->s_root)) { -+ printk(KERN_WARNING "mini_fo_read_super: d_alloc failed\n"); -+ err = -ENOMEM; -+ goto out_dput; -+ } -+ -+ sb->s_root->d_op = &mini_fo_dops; -+ sb->s_root->d_sb = sb; -+ sb->s_root->d_parent = sb->s_root; -+ -+ /* link the upper and lower dentries */ -+ __dtopd(sb->s_root) = (struct mini_fo_dentry_info *) -+ kmalloc(sizeof(struct mini_fo_dentry_info), GFP_KERNEL); -+ if (!dtopd(sb->s_root)) { -+ err = -ENOMEM; -+ goto out_dput2; -+ } -+ dtopd(sb->s_root)->state = MODIFIED; -+ dtohd(sb->s_root) = hidden_root; -+ -+ /* fanout relevant, interpose on storage root dentry too */ -+ dtohd2(sb->s_root) = stopd(sb)->storage_dir_dentry; -+ -+ /* ...and call tri-interpose to interpose root dir inodes -+ * if (mini_fo_interpose(hidden_root, sb->s_root, sb, 0)) -+ */ -+ if(mini_fo_tri_interpose(hidden_root, dtohd2(sb->s_root), sb->s_root, sb, 0)) -+ goto out_dput2; -+ -+ /* initalize the wol list */ -+ itopd(sb->s_root->d_inode)->deleted_list_size = -1; -+ itopd(sb->s_root->d_inode)->renamed_list_size = -1; -+ meta_build_lists(sb->s_root); -+ -+ goto out; -+ -+ out_dput2: -+ dput(sb->s_root); -+ out_dput: -+ dput(hidden_root); -+ dput(dtohd2(sb->s_root)); /* release the hidden_sto_dentry too */ -+ out_free: -+ kfree(stopd(sb)); -+ __stopd(sb) = NULL; -+ out: -+ -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,0) -+ return err; -+#else -+ if (err) { -+ return ERR_PTR(err); -+ } else { -+ return sb; -+ } -+#endif -+} -+ -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,0) -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,18) -+static int mini_fo_get_sb(struct file_system_type *fs_type, -+ int flags, const char *dev_name, -+ void *raw_data, struct vfsmount *mnt) -+{ -+ return get_sb_nodev(fs_type, flags, raw_data, mini_fo_read_super, mnt); -+} -+#else -+static struct super_block *mini_fo_get_sb(struct file_system_type *fs_type, -+ int flags, const char *dev_name, -+ void *raw_data) -+{ -+ return get_sb_nodev(fs_type, flags, raw_data, mini_fo_read_super); -+} -+#endif -+ -+void mini_fo_kill_block_super(struct super_block *sb) -+{ -+ generic_shutdown_super(sb); -+ /* -+ * XXX: BUG: Halcrow: Things get unstable sometime after this point: -+ * lib/rwsem-spinlock.c:127: spin_is_locked on uninitialized -+ * fs/fs-writeback.c:402: spin_lock(fs/super.c:a0381828) already -+ * locked by fs/fs-writeback.c/402 -+ * -+ * Apparently, someone's not releasing a lock on sb_lock... -+ */ -+} -+ -+static struct file_system_type mini_fo_fs_type = { -+ .owner = THIS_MODULE, -+ .name = "mini_fo", -+ .get_sb = mini_fo_get_sb, -+ .kill_sb = mini_fo_kill_block_super, -+ .fs_flags = 0, -+}; -+ -+ -+#else -+static DECLARE_FSTYPE(mini_fo_fs_type, "mini_fo", mini_fo_read_super, 0); -+#endif -+ -+static int __init init_mini_fo_fs(void) -+{ -+ printk("Registering mini_fo version $Id$\n"); -+ return register_filesystem(&mini_fo_fs_type); -+} -+static void __exit exit_mini_fo_fs(void) -+{ -+ printk("Unregistering mini_fo version $Id$\n"); -+ unregister_filesystem(&mini_fo_fs_type); -+} -+ -+#if LINUX_VERSION_CODE < KERNEL_VERSION(2,6,0) -+EXPORT_NO_SYMBOLS; -+#endif -+ -+MODULE_AUTHOR("Erez Zadok <ezk@cs.sunysb.edu>"); -+MODULE_DESCRIPTION("FiST-generated mini_fo filesystem"); -+MODULE_LICENSE("GPL"); -+ -+/* MODULE_PARM(fist_debug_var, "i"); */ -+/* MODULE_PARM_DESC(fist_debug_var, "Debug level"); */ -+ -+module_init(init_mini_fo_fs) -+module_exit(exit_mini_fo_fs) ---- /dev/null -+++ b/fs/mini_fo/Makefile -@@ -0,0 +1,17 @@ -+# -+# Makefile for mini_fo 2.4 and 2.6 Linux kernels -+# -+# Copyright (C) 2004, 2005 Markus Klotzbuecher <mk@creamnet.de> -+# -+# This program is free software; you can redistribute it and/or -+# modify it under the terms of the GNU General Public License -+# as published by the Free Software Foundation; either version -+# 2 of the License, or (at your option) any later version. -+# -+ -+obj-$(CONFIG_MINI_FO) := mini_fo.o -+mini_fo-objs := meta.o dentry.o file.o inode.o main.o super.o state.o aux.o -+ -+# dependencies -+${mini_fo-objs}: mini_fo.h fist.h -+ ---- /dev/null -+++ b/fs/mini_fo/meta.c -@@ -0,0 +1,1000 @@ -+/* -+ * Copyright (C) 2004, 2005 Markus Klotzbuecher <mk@creamnet.de> -+ * -+ * This program is free software; you can redistribute it and/or -+ * modify it under the terms of the GNU General Public License -+ * as published by the Free Software Foundation; either version -+ * 2 of the License, or (at your option) any later version. -+ */ -+ -+#ifdef HAVE_CONFIG_H -+# include <config.h> -+#endif /* HAVE_CONFIG_H */ -+#include "fist.h" -+#include "mini_fo.h" -+ -+int meta_build_lists(dentry_t *dentry) -+{ -+ struct mini_fo_inode_info *inode_info; -+ -+ dentry_t *meta_dentry = 0; -+ file_t *meta_file = 0; -+ mm_segment_t old_fs; -+ void *buf; -+ -+ int bytes, len; -+ struct vfsmount *meta_mnt; -+ char *entry; -+ -+ inode_info = itopd(dentry->d_inode); -+ if(!(inode_info->deleted_list_size == -1 && -+ inode_info->renamed_list_size == -1)) { -+ printk(KERN_CRIT "mini_fo: meta_build_lists: \ -+ Error, list(s) not virgin.\n"); -+ return -1; -+ } -+ -+ /* init our meta lists */ -+ INIT_LIST_HEAD(&inode_info->deleted_list); -+ inode_info->deleted_list_size = 0; -+ -+ INIT_LIST_HEAD(&inode_info->renamed_list); -+ inode_info->renamed_list_size = 0; -+ -+ /* might there be a META-file? */ -+ if(dtohd2(dentry) && dtohd2(dentry)->d_inode) { -+ meta_dentry = lookup_one_len(META_FILENAME, -+ dtohd2(dentry), -+ strlen(META_FILENAME)); -+ if(!meta_dentry->d_inode) { -+ dput(meta_dentry); -+ goto out_ok; -+ } -+ /* $%& err, is this correct? */ -+ meta_mnt = stopd(dentry->d_inode->i_sb)->hidden_mnt2; -+ mntget(meta_mnt); -+ -+ -+ /* open META-file for reading */ -+ meta_file = dentry_open(meta_dentry, meta_mnt, 0x0); -+ if(!meta_file || IS_ERR(meta_file)) { -+ printk(KERN_CRIT "mini_fo: meta_build_lists: \ -+ ERROR opening META file.\n"); -+ goto out_err; -+ } -+ -+ /* check if fs supports reading */ -+ if(!meta_file->f_op->read) { -+ printk(KERN_CRIT "mini_fo: meta_build_lists: \ -+ ERROR, fs does not support reading.\n"); -+ goto out_err_close; -+ } -+ -+ /* allocate a page for transfering the data */ -+ buf = (void *) __get_free_page(GFP_KERNEL); -+ if(!buf) { -+ printk(KERN_CRIT "mini_fo: meta_build_lists: \ -+ ERROR, out of mem.\n"); -+ goto out_err_close; -+ } -+ meta_file->f_pos = 0; -+ old_fs = get_fs(); -+ set_fs(KERNEL_DS); -+ do { -+ char *c; -+ bytes = meta_file->f_op->read(meta_file, buf, PAGE_SIZE, &meta_file->f_pos); -+ if(bytes == PAGE_SIZE) { -+ /* trim a cut off filename and adjust f_pos to get it next time */ -+ for(c = (char*) buf+PAGE_SIZE; -+ *c != '\n'; -+ c--, bytes--, meta_file->f_pos--); -+ } -+ entry = (char *) buf; -+ while(entry < (char *) buf+bytes) { -+ -+ char *old_path; -+ char *dir_name; -+ int old_len, new_len; -+ -+ /* len without '\n'*/ -+ len = (int) (strchr(entry, '\n') - entry); -+ switch (*entry) { -+ case 'D': -+ /* format: "D filename" */ -+ meta_list_add_d_entry(dentry, -+ entry+2, -+ len-2); -+ break; -+ case 'R': -+ /* format: "R path/xy/dir newDir" */ -+ old_path = entry+2; -+ dir_name = strchr(old_path, ' ') + 1; -+ old_len = dir_name - old_path - 1; -+ new_len = ((int) entry) + len - ((int ) dir_name); -+ meta_list_add_r_entry(dentry, -+ old_path, -+ old_len, -+ dir_name, -+ new_len); -+ break; -+ default: -+ /* unknown entry type detected */ -+ break; -+ } -+ entry += len+1; -+ } -+ -+ } while(meta_file->f_pos < meta_dentry->d_inode->i_size); -+ -+ free_page((unsigned long) buf); -+ set_fs(old_fs); -+ fput(meta_file); -+ } -+ goto out_ok; -+ -+ out_err_close: -+ fput(meta_file); -+ out_err: -+ mntput(meta_mnt); -+ dput(meta_dentry); -+ return -1; -+ out_ok: -+ return 1; /* check this!!! inode_info->wol_size; */ -+} -+ -+/* cleanups up all lists and free's the mem by dentry */ -+int meta_put_lists(dentry_t *dentry) -+{ -+ if(!dentry || !dentry->d_inode) { -+ printk("mini_fo: meta_put_lists: invalid dentry passed.\n"); -+ return -1; -+ } -+ return __meta_put_lists(dentry->d_inode); -+} -+ -+/* cleanups up all lists and free's the mem by inode */ -+int __meta_put_lists(inode_t *inode) -+{ -+ int err = 0; -+ if(!inode || !itopd(inode)) { -+ printk("mini_fo: __meta_put_lists: invalid inode passed.\n"); -+ return -1; -+ } -+ err = __meta_put_d_list(inode); -+ err |= __meta_put_r_list(inode); -+ return err; -+} -+ -+int meta_sync_lists(dentry_t *dentry) -+{ -+ int err = 0; -+ if(!dentry || !dentry->d_inode) { -+ printk("mini_fo: meta_sync_lists: \ -+ invalid dentry passed.\n"); -+ return -1; -+ } -+ err = meta_sync_d_list(dentry, 0); -+ err |= meta_sync_r_list(dentry, 1); -+ return err; -+} -+ -+ -+/* remove all D entries from the renamed list and free the mem */ -+int __meta_put_d_list(inode_t *inode) -+{ -+ struct list_head *tmp; -+ struct deleted_entry *del_entry; -+ struct mini_fo_inode_info *inode_info; -+ -+ if(!inode || !itopd(inode)) { -+ printk(KERN_CRIT "mini_fo: __meta_put_d_list: \ -+ invalid inode passed.\n"); -+ return -1; -+ } -+ inode_info = itopd(inode); -+ -+ /* nuke the DELETED-list */ -+ if(inode_info->deleted_list_size <= 0) -+ return 0; -+ -+ while(!list_empty(&inode_info->deleted_list)) { -+ tmp = inode_info->deleted_list.next; -+ list_del(tmp); -+ del_entry = list_entry(tmp, struct deleted_entry, list); -+ kfree(del_entry->name); -+ kfree(del_entry); -+ } -+ inode_info->deleted_list_size = 0; -+ -+ return 0; -+} -+ -+/* remove all R entries from the renamed list and free the mem */ -+int __meta_put_r_list(inode_t *inode) -+{ -+ struct list_head *tmp; -+ struct renamed_entry *ren_entry; -+ struct mini_fo_inode_info *inode_info; -+ -+ if(!inode || !itopd(inode)) { -+ printk(KERN_CRIT "mini_fo: meta_put_r_list: invalid inode.\n"); -+ return -1; -+ } -+ inode_info = itopd(inode); -+ -+ /* nuke the RENAMED-list */ -+ if(inode_info->renamed_list_size <= 0) -+ return 0; -+ -+ while(!list_empty(&inode_info->renamed_list)) { -+ tmp = inode_info->renamed_list.next; -+ list_del(tmp); -+ ren_entry = list_entry(tmp, struct renamed_entry, list); -+ kfree(ren_entry->new_name); -+ kfree(ren_entry->old_name); -+ kfree(ren_entry); -+ } -+ inode_info->renamed_list_size = 0; -+ -+ return 0; -+} -+ -+int meta_add_d_entry(dentry_t *dentry, const char *name, int len) -+{ -+ int err = 0; -+ err = meta_list_add_d_entry(dentry, name, len); -+ err |= meta_write_d_entry(dentry,name,len); -+ return err; -+} -+ -+/* add a D entry to the deleted list */ -+int meta_list_add_d_entry(dentry_t *dentry, const char *name, int len) -+{ -+ struct deleted_entry *del_entry; -+ struct mini_fo_inode_info *inode_info; -+ -+ if(!dentry || !dentry->d_inode) { -+ printk(KERN_CRIT "mini_fo: meta_list_add_d_entry: \ -+ invalid dentry passed.\n"); -+ return -1; -+ } -+ inode_info = itopd(dentry->d_inode); -+ -+ if(inode_info->deleted_list_size < 0) -+ return -1; -+ -+ del_entry = (struct deleted_entry *) -+ kmalloc(sizeof(struct deleted_entry), GFP_KERNEL); -+ del_entry->name = (char*) kmalloc(len, GFP_KERNEL); -+ if(!del_entry || !del_entry->name) { -+ printk(KERN_CRIT "mini_fo: meta_list_add_d_entry: \ -+ out of mem.\n"); -+ kfree(del_entry->name); -+ kfree(del_entry); -+ return -ENOMEM; -+ } -+ -+ strncpy(del_entry->name, name, len); -+ del_entry->len = len; -+ -+ list_add(&del_entry->list, &inode_info->deleted_list); -+ inode_info->deleted_list_size++; -+ return 0; -+} -+ -+int meta_add_r_entry(dentry_t *dentry, -+ const char *old_name, int old_len, -+ const char *new_name, int new_len) -+{ -+ int err = 0; -+ err = meta_list_add_r_entry(dentry, -+ old_name, old_len, -+ new_name, new_len); -+ err |= meta_write_r_entry(dentry, -+ old_name, old_len, -+ new_name, new_len); -+ return err; -+} -+ -+/* add a R entry to the renamed list */ -+int meta_list_add_r_entry(dentry_t *dentry, -+ const char *old_name, int old_len, -+ const char *new_name, int new_len) -+{ -+ struct renamed_entry *ren_entry; -+ struct mini_fo_inode_info *inode_info; -+ -+ if(!dentry || !dentry->d_inode) { -+ printk(KERN_CRIT "mini_fo: meta_list_add_r_entry: \ -+ invalid dentry passed.\n"); -+ return -1; -+ } -+ inode_info = itopd(dentry->d_inode); -+ -+ if(inode_info->renamed_list_size < 0) -+ return -1; -+ -+ ren_entry = (struct renamed_entry *) -+ kmalloc(sizeof(struct renamed_entry), GFP_KERNEL); -+ ren_entry->old_name = (char*) kmalloc(old_len, GFP_KERNEL); -+ ren_entry->new_name = (char*) kmalloc(new_len, GFP_KERNEL); -+ -+ if(!ren_entry || !ren_entry->old_name || !ren_entry->new_name) { -+ printk(KERN_CRIT "mini_fo: meta_list_add_r_entry: \ -+ out of mem.\n"); -+ kfree(ren_entry->new_name); -+ kfree(ren_entry->old_name); -+ kfree(ren_entry); -+ return -ENOMEM; -+ } -+ -+ strncpy(ren_entry->old_name, old_name, old_len); -+ ren_entry->old_len = old_len; -+ strncpy(ren_entry->new_name, new_name, new_len); -+ ren_entry->new_len = new_len; -+ -+ list_add(&ren_entry->list, &inode_info->renamed_list); -+ inode_info->renamed_list_size++; -+ return 0; -+} -+ -+ -+int meta_remove_r_entry(dentry_t *dentry, const char *name, int len) -+{ -+ int err = 0; -+ if(!dentry || !dentry->d_inode) { -+ printk(KERN_CRIT -+ "mini_fo: meta_remove_r_entry: \ -+ invalid dentry passed.\n"); -+ return -1; -+ } -+ -+ err = meta_list_remove_r_entry(dentry, name, len); -+ err |= meta_sync_lists(dentry); -+ return err; -+} -+ -+int meta_list_remove_r_entry(dentry_t *dentry, const char *name, int len) -+{ -+ if(!dentry || !dentry->d_inode) { -+ printk(KERN_CRIT -+ "mini_fo: meta_list_remove_r_entry: \ -+ invalid dentry passed.\n"); -+ return -1; -+ } -+ return __meta_list_remove_r_entry(dentry->d_inode, name, len); -+} -+ -+int __meta_list_remove_r_entry(inode_t *inode, const char *name, int len) -+{ -+ struct list_head *tmp; -+ struct renamed_entry *ren_entry; -+ struct mini_fo_inode_info *inode_info; -+ -+ if(!inode || !itopd(inode)) -+ printk(KERN_CRIT -+ "mini_fo: __meta_list_remove_r_entry: \ -+ invalid inode passed.\n"); -+ inode_info = itopd(inode); -+ -+ if(inode_info->renamed_list_size < 0) -+ return -1; -+ if(inode_info->renamed_list_size == 0) -+ return 1; -+ -+ list_for_each(tmp, &inode_info->renamed_list) { -+ ren_entry = list_entry(tmp, struct renamed_entry, list); -+ if(ren_entry->new_len != len) -+ continue; -+ -+ if(!strncmp(ren_entry->new_name, name, len)) { -+ list_del(tmp); -+ kfree(ren_entry->new_name); -+ kfree(ren_entry->old_name); -+ kfree(ren_entry); -+ inode_info->renamed_list_size--; -+ return 0; -+ } -+ } -+ return 1; -+} -+ -+ -+/* append a single D entry to the meta file */ -+int meta_write_d_entry(dentry_t *dentry, const char *name, int len) -+{ -+ dentry_t *meta_dentry = 0; -+ file_t *meta_file = 0; -+ mm_segment_t old_fs; -+ -+ int bytes, err; -+ struct vfsmount *meta_mnt = 0; -+ char *buf; -+ -+ err = 0; -+ -+ if(itopd(dentry->d_inode)->deleted_list_size < 0) { -+ err = -1; -+ goto out; -+ } -+ -+ if(dtopd(dentry)->state == UNMODIFIED) { -+ err = build_sto_structure(dentry->d_parent, dentry); -+ if(err) { -+ printk(KERN_CRIT "mini_fo: meta_write_d_entry: \ -+ build_sto_structure failed.\n"); -+ goto out; -+ } -+ } -+ meta_dentry = lookup_one_len(META_FILENAME, -+ dtohd2(dentry), strlen (META_FILENAME)); -+ -+ /* We need to create a META-file */ -+ if(!meta_dentry->d_inode) { -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,0) -+ vfs_create(dtohd2(dentry)->d_inode, -+ meta_dentry, -+ S_IRUSR | S_IWUSR, -+ NULL); -+#else -+ vfs_create(dtohd2(dentry)->d_inode, -+ meta_dentry, -+ S_IRUSR | S_IWUSR); -+#endif -+ } -+ /* open META-file for writing */ -+ meta_file = dentry_open(meta_dentry, meta_mnt, 0x1); -+ if(!meta_file || IS_ERR(meta_file)) { -+ printk(KERN_CRIT "mini_fo: meta_write_d_entry: \ -+ ERROR opening meta file.\n"); -+ mntput(meta_mnt); /* $%& is this necessary? */ -+ dput(meta_dentry); -+ err = -1; -+ goto out; -+ } -+ -+ /* check if fs supports writing */ -+ if(!meta_file->f_op->write) { -+ printk(KERN_CRIT "mini_fo: meta_write_d_entry: \ -+ ERROR, fs does not support writing.\n"); -+ goto out_err_close; -+ } -+ -+ meta_file->f_pos = meta_dentry->d_inode->i_size; /* append */ -+ old_fs = get_fs(); -+ set_fs(KERNEL_DS); -+ -+ /* size: len for name, 1 for \n and 2 for "D " */ -+ buf = (char *) kmalloc(len+3, GFP_KERNEL); -+ if (!buf) { -+ printk(KERN_CRIT "mini_fo: meta_write_d_entry: \ -+ out of mem.\n"); -+ return -ENOMEM; -+ } -+ -+ buf[0] = 'D'; -+ buf[1] = ' '; -+ strncpy(buf+2, name, len); -+ buf[len+2] = '\n'; -+ bytes = meta_file->f_op->write(meta_file, buf, len+3, -+ &meta_file->f_pos); -+ if(bytes != len+3) { -+ printk(KERN_CRIT "mini_fo: meta_write_d_entry: \ -+ ERROR writing.\n"); -+ err = -1; -+ } -+ kfree(buf); -+ set_fs(old_fs); -+ -+ out_err_close: -+ fput(meta_file); -+ out: -+ return err; -+} -+ -+/* append a single R entry to the meta file */ -+int meta_write_r_entry(dentry_t *dentry, -+ const char *old_name, int old_len, -+ const char *new_name, int new_len) -+{ -+ dentry_t *meta_dentry = 0; -+ file_t *meta_file = 0; -+ mm_segment_t old_fs; -+ -+ int bytes, err, buf_len; -+ struct vfsmount *meta_mnt = 0; -+ char *buf; -+ -+ -+ err = 0; -+ -+ if(itopd(dentry->d_inode)->renamed_list_size < 0) { -+ err = -1; -+ goto out; -+ } -+ -+ /* build the storage structure? */ -+ if(dtopd(dentry)->state == UNMODIFIED) { -+ err = build_sto_structure(dentry->d_parent, dentry); -+ if(err) { -+ printk(KERN_CRIT "mini_fo: meta_write_r_entry: \ -+ build_sto_structure failed.\n"); -+ goto out; -+ } -+ } -+ meta_dentry = lookup_one_len(META_FILENAME, -+ dtohd2(dentry), -+ strlen (META_FILENAME)); -+ if(!meta_dentry->d_inode) { -+ /* We need to create a META-file */ -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,0) -+ vfs_create(dtohd2(dentry)->d_inode, -+ meta_dentry, S_IRUSR | S_IWUSR, NULL); -+#else -+ vfs_create(dtohd2(dentry)->d_inode, -+ meta_dentry, S_IRUSR | S_IWUSR); -+#endif -+ } -+ /* open META-file for writing */ -+ meta_file = dentry_open(meta_dentry, meta_mnt, 0x1); -+ if(!meta_file || IS_ERR(meta_file)) { -+ printk(KERN_CRIT "mini_fo: meta_write_r_entry: \ -+ ERROR opening meta file.\n"); -+ mntput(meta_mnt); -+ dput(meta_dentry); -+ err = -1; -+ goto out; -+ } -+ -+ /* check if fs supports writing */ -+ if(!meta_file->f_op->write) { -+ printk(KERN_CRIT "mini_fo: meta_write_r_entry: \ -+ ERROR, fs does not support writing.\n"); -+ goto out_err_close; -+ } -+ -+ meta_file->f_pos = meta_dentry->d_inode->i_size; /* append */ -+ old_fs = get_fs(); -+ set_fs(KERNEL_DS); -+ -+ /* size: 2 for "R ", old_len+new_len for names, 1 blank+1 \n */ -+ buf_len = old_len + new_len + 4; -+ buf = (char *) kmalloc(buf_len, GFP_KERNEL); -+ if (!buf) { -+ printk(KERN_CRIT "mini_fo: meta_write_r_entry: out of mem.\n"); -+ return -ENOMEM; -+ } -+ -+ buf[0] = 'R'; -+ buf[1] = ' '; -+ strncpy(buf + 2, old_name, old_len); -+ buf[old_len + 2] = ' '; -+ strncpy(buf + old_len + 3, new_name, new_len); -+ buf[buf_len -1] = '\n'; -+ bytes = meta_file->f_op->write(meta_file, buf, buf_len, &meta_file->f_pos); -+ if(bytes != buf_len) { -+ printk(KERN_CRIT "mini_fo: meta_write_r_entry: ERROR writing.\n"); -+ err = -1; -+ } -+ -+ kfree(buf); -+ set_fs(old_fs); -+ -+ out_err_close: -+ fput(meta_file); -+ out: -+ return err; -+} -+ -+/* sync D list to disk, append data if app_flag is 1 */ -+/* check the meta_mnt, which seems not to be used (properly) */ -+ -+int meta_sync_d_list(dentry_t *dentry, int app_flag) -+{ -+ dentry_t *meta_dentry; -+ file_t *meta_file; -+ mm_segment_t old_fs; -+ -+ int bytes, err; -+ struct vfsmount *meta_mnt; -+ char *buf; -+ -+ struct list_head *tmp; -+ struct deleted_entry *del_entry; -+ struct mini_fo_inode_info *inode_info; -+ -+ err = 0; -+ meta_file=0; -+ meta_mnt=0; -+ -+ if(!dentry || !dentry->d_inode) { -+ printk(KERN_CRIT "mini_fo: meta_sync_d_list: \ -+ invalid inode passed.\n"); -+ err = -1; -+ goto out; -+ } -+ inode_info = itopd(dentry->d_inode); -+ -+ if(inode_info->deleted_list_size < 0) { -+ err = -1; -+ goto out; -+ } -+ -+ /* ok, there is something to sync */ -+ -+ /* build the storage structure? */ -+ if(!dtohd2(dentry) && !itohi2(dentry->d_inode)) { -+ err = build_sto_structure(dentry->d_parent, dentry); -+ if(err) { -+ printk(KERN_CRIT "mini_fo: meta_sync_d_list: \ -+ build_sto_structure failed.\n"); -+ goto out; -+ } -+ } -+ meta_dentry = lookup_one_len(META_FILENAME, -+ dtohd2(dentry), -+ strlen(META_FILENAME)); -+ if(!meta_dentry->d_inode) { -+ /* We need to create a META-file */ -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,0) -+ vfs_create(dtohd2(dentry)->d_inode, -+ meta_dentry, S_IRUSR | S_IWUSR, NULL); -+#else -+ vfs_create(dtohd2(dentry)->d_inode, -+ meta_dentry, S_IRUSR | S_IWUSR); -+#endif -+ app_flag = 0; -+ } -+ /* need we truncate the meta file? */ -+ if(!app_flag) { -+ struct iattr newattrs; -+ newattrs.ia_size = 0; -+ newattrs.ia_valid = ATTR_SIZE | ATTR_CTIME; -+ -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,16) -+ mutex_lock(&meta_dentry->d_inode->i_mutex); -+#else -+ down(&meta_dentry->d_inode->i_sem); -+#endif -+ err = notify_change(meta_dentry, &newattrs); -+ -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,16) -+ mutex_unlock(&meta_dentry->d_inode->i_mutex); -+#else -+ up(&meta_dentry->d_inode->i_sem); -+#endif -+ -+ if(err || meta_dentry->d_inode->i_size != 0) { -+ printk(KERN_CRIT "mini_fo: meta_sync_d_list: \ -+ ERROR truncating meta file.\n"); -+ goto out_err_close; -+ } -+ } -+ -+ /* open META-file for writing */ -+ meta_file = dentry_open(meta_dentry, meta_mnt, 0x1); -+ if(!meta_file || IS_ERR(meta_file)) { -+ printk(KERN_CRIT "mini_fo: meta_sync_d_list: \ -+ ERROR opening meta file.\n"); -+ /* we don't mntget so we dont't mntput (for now) -+ * mntput(meta_mnt); -+ */ -+ dput(meta_dentry); -+ err = -1; -+ goto out; -+ } -+ -+ /* check if fs supports writing */ -+ if(!meta_file->f_op->write) { -+ printk(KERN_CRIT "mini_fo: meta_sync_d_list: \ -+ ERROR, fs does not support writing.\n"); -+ goto out_err_close; -+ } -+ -+ meta_file->f_pos = meta_dentry->d_inode->i_size; /* append */ -+ old_fs = get_fs(); -+ set_fs(KERNEL_DS); -+ -+ /* here we go... */ -+ list_for_each(tmp, &inode_info->deleted_list) { -+ del_entry = list_entry(tmp, struct deleted_entry, list); -+ -+ /* size: len for name, 1 for \n and 2 for "D " */ -+ buf = (char *) kmalloc(del_entry->len+3, GFP_KERNEL); -+ if (!buf) { -+ printk(KERN_CRIT "mini_fo: meta_sync_d_list: \ -+ out of mem.\n"); -+ return -ENOMEM; -+ } -+ -+ buf[0] = 'D'; -+ buf[1] = ' '; -+ strncpy(buf+2, del_entry->name, del_entry->len); -+ buf[del_entry->len+2] = '\n'; -+ bytes = meta_file->f_op->write(meta_file, buf, -+ del_entry->len+3, -+ &meta_file->f_pos); -+ if(bytes != del_entry->len+3) { -+ printk(KERN_CRIT "mini_fo: meta_sync_d_list: \ -+ ERROR writing.\n"); -+ err |= -1; -+ } -+ kfree(buf); -+ } -+ set_fs(old_fs); -+ -+ out_err_close: -+ fput(meta_file); -+ out: -+ return err; -+ -+} -+ -+int meta_sync_r_list(dentry_t *dentry, int app_flag) -+{ -+ dentry_t *meta_dentry; -+ file_t *meta_file; -+ mm_segment_t old_fs; -+ -+ int bytes, err, buf_len; -+ struct vfsmount *meta_mnt; -+ char *buf; -+ -+ struct list_head *tmp; -+ struct renamed_entry *ren_entry; -+ struct mini_fo_inode_info *inode_info; -+ -+ err = 0; -+ meta_file=0; -+ meta_mnt=0; -+ -+ if(!dentry || !dentry->d_inode) { -+ printk(KERN_CRIT "mini_fo: meta_sync_r_list: \ -+ invalid dentry passed.\n"); -+ err = -1; -+ goto out; -+ } -+ inode_info = itopd(dentry->d_inode); -+ -+ if(inode_info->deleted_list_size < 0) { -+ err = -1; -+ goto out; -+ } -+ -+ /* ok, there is something to sync */ -+ -+ /* build the storage structure? */ -+ if(!dtohd2(dentry) && !itohi2(dentry->d_inode)) { -+ err = build_sto_structure(dentry->d_parent, dentry); -+ if(err) { -+ printk(KERN_CRIT "mini_fo: meta_sync_r_list: \ -+ build_sto_structure failed.\n"); -+ goto out; -+ } -+ } -+ meta_dentry = lookup_one_len(META_FILENAME, -+ dtohd2(dentry), -+ strlen(META_FILENAME)); -+ if(!meta_dentry->d_inode) { -+ /* We need to create a META-file */ -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,0) -+ vfs_create(dtohd2(dentry)->d_inode, -+ meta_dentry, S_IRUSR | S_IWUSR, NULL); -+#else -+ vfs_create(dtohd2(dentry)->d_inode, -+ meta_dentry, S_IRUSR | S_IWUSR); -+#endif -+ app_flag = 0; -+ } -+ /* need we truncate the meta file? */ -+ if(!app_flag) { -+ struct iattr newattrs; -+ newattrs.ia_size = 0; -+ newattrs.ia_valid = ATTR_SIZE | ATTR_CTIME; -+ -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,16) -+ mutex_lock(&meta_dentry->d_inode->i_mutex); -+#else -+ down(&meta_dentry->d_inode->i_sem); -+#endif -+ err = notify_change(meta_dentry, &newattrs); -+ -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,16) -+ mutex_unlock(&meta_dentry->d_inode->i_mutex); -+#else -+ up(&meta_dentry->d_inode->i_sem); -+#endif -+ if(err || meta_dentry->d_inode->i_size != 0) { -+ printk(KERN_CRIT "mini_fo: meta_sync_r_list: \ -+ ERROR truncating meta file.\n"); -+ goto out_err_close; -+ } -+ } -+ -+ /* open META-file for writing */ -+ meta_file = dentry_open(meta_dentry, meta_mnt, 0x1); -+ if(!meta_file || IS_ERR(meta_file)) { -+ printk(KERN_CRIT "mini_fo: meta_sync_r_list: \ -+ ERROR opening meta file.\n"); -+ /* we don't mntget so we dont't mntput (for now) -+ * mntput(meta_mnt); -+ */ -+ dput(meta_dentry); -+ err = -1; -+ goto out; -+ } -+ -+ /* check if fs supports writing */ -+ if(!meta_file->f_op->write) { -+ printk(KERN_CRIT "mini_fo: meta_sync_r_list: \ -+ ERROR, fs does not support writing.\n"); -+ goto out_err_close; -+ } -+ -+ meta_file->f_pos = meta_dentry->d_inode->i_size; /* append */ -+ old_fs = get_fs(); -+ set_fs(KERNEL_DS); -+ -+ /* here we go... */ -+ list_for_each(tmp, &inode_info->renamed_list) { -+ ren_entry = list_entry(tmp, struct renamed_entry, list); -+ /* size: -+ * 2 for "R ", old_len+new_len for names, 1 blank+1 \n */ -+ buf_len = ren_entry->old_len + ren_entry->new_len + 4; -+ buf = (char *) kmalloc(buf_len, GFP_KERNEL); -+ if (!buf) { -+ printk(KERN_CRIT "mini_fo: meta_sync_r_list: \ -+ out of mem.\n"); -+ return -ENOMEM; -+ } -+ buf[0] = 'R'; -+ buf[1] = ' '; -+ strncpy(buf + 2, ren_entry->old_name, ren_entry->old_len); -+ buf[ren_entry->old_len + 2] = ' '; -+ strncpy(buf + ren_entry->old_len + 3, -+ ren_entry->new_name, ren_entry->new_len); -+ buf[buf_len - 1] = '\n'; -+ bytes = meta_file->f_op->write(meta_file, buf, -+ buf_len, &meta_file->f_pos); -+ if(bytes != buf_len) { -+ printk(KERN_CRIT "mini_fo: meta_sync_r_list: \ -+ ERROR writing.\n"); -+ err |= -1; -+ } -+ kfree(buf); -+ } -+ set_fs(old_fs); -+ -+ out_err_close: -+ fput(meta_file); -+ out: -+ return err; -+} -+ -+int meta_check_d_entry(dentry_t *dentry, const char *name, int len) -+{ -+ if(!dentry || !dentry->d_inode) -+ printk(KERN_CRIT "mini_fo: meta_check_d_dentry: \ -+ invalid dentry passed.\n"); -+ return __meta_check_d_entry(dentry->d_inode, name, len); -+} -+ -+int __meta_check_d_entry(inode_t *inode, const char *name, int len) -+{ -+ struct list_head *tmp; -+ struct deleted_entry *del_entry; -+ struct mini_fo_inode_info *inode_info; -+ -+ if(!inode || !itopd(inode)) -+ printk(KERN_CRIT "mini_fo: __meta_check_d_dentry: \ -+ invalid inode passed.\n"); -+ -+ inode_info = itopd(inode); -+ -+ if(inode_info->deleted_list_size <= 0) -+ return 0; -+ -+ list_for_each(tmp, &inode_info->deleted_list) { -+ del_entry = list_entry(tmp, struct deleted_entry, list); -+ if(del_entry->len != len) -+ continue; -+ -+ if(!strncmp(del_entry->name, name, len)) -+ return 1; -+ } -+ return 0; -+} -+ -+/* -+ * check if file has been renamed and return path to orig. base dir. -+ * Implements no error return values so far, what of course sucks. -+ * String is null terminated.' -+ */ -+char* meta_check_r_entry(dentry_t *dentry, const char *name, int len) -+{ -+ if(!dentry || !dentry->d_inode) { -+ printk(KERN_CRIT "mini_fo: meta_check_r_dentry: \ -+ invalid dentry passed.\n"); -+ return NULL; -+ } -+ return __meta_check_r_entry(dentry->d_inode, name, len); -+} -+ -+char* __meta_check_r_entry(inode_t *inode, const char *name, int len) -+{ -+ struct list_head *tmp; -+ struct renamed_entry *ren_entry; -+ struct mini_fo_inode_info *inode_info; -+ char *old_path; -+ -+ if(!inode || !itopd(inode)) { -+ printk(KERN_CRIT "mini_fo: meta_check_r_dentry: \ -+ invalid inode passed.\n"); -+ return NULL; -+ } -+ inode_info = itopd(inode); -+ -+ if(inode_info->renamed_list_size <= 0) -+ return NULL; -+ -+ list_for_each(tmp, &inode_info->renamed_list) { -+ ren_entry = list_entry(tmp, struct renamed_entry, list); -+ if(ren_entry->new_len != len) -+ continue; -+ -+ if(!strncmp(ren_entry->new_name, name, len)) { -+ old_path = (char *) -+ kmalloc(ren_entry->old_len+1, GFP_KERNEL); -+ strncpy(old_path, -+ ren_entry->old_name, -+ ren_entry->old_len); -+ old_path[ren_entry->old_len]='\0'; -+ return old_path; -+ } -+ } -+ return NULL; -+} -+ -+/* -+ * This version only checks if entry exists and return: -+ * 1 if exists, -+ * 0 if not, -+ * -1 if error. -+ */ -+int meta_is_r_entry(dentry_t *dentry, const char *name, int len) -+{ -+ if(!dentry || !dentry->d_inode) { -+ printk(KERN_CRIT "mini_fo: meta_check_r_dentry [2]: \ -+ invalid dentry passed.\n"); -+ return -1; -+ } -+ return __meta_is_r_entry(dentry->d_inode, name, len); -+} -+ -+int __meta_is_r_entry(inode_t *inode, const char *name, int len) -+{ -+ struct list_head *tmp; -+ struct renamed_entry *ren_entry; -+ struct mini_fo_inode_info *inode_info; -+ -+ if(!inode || !itopd(inode)) { -+ printk(KERN_CRIT "mini_fo: meta_check_r_dentry [2]: \ -+ invalid inode passed.\n"); -+ return -1; -+ } -+ inode_info = itopd(inode); -+ -+ if(inode_info->renamed_list_size <= 0) -+ return -1; -+ -+ list_for_each(tmp, &inode_info->renamed_list) { -+ ren_entry = list_entry(tmp, struct renamed_entry, list); -+ if(ren_entry->new_len != len) -+ continue; -+ -+ if(!strncmp(ren_entry->new_name, name, len)) -+ return 1; -+ } -+ return 0; -+} -+ ---- /dev/null -+++ b/fs/mini_fo/mini_fo.h -@@ -0,0 +1,510 @@ -+/* -+ * Copyright (c) 1997-2003 Erez Zadok -+ * Copyright (c) 2001-2003 Stony Brook University -+ * -+ * For specific licensing information, see the COPYING file distributed with -+ * this package, or get one from ftp://ftp.filesystems.org/pub/fist/COPYING. -+ * -+ * This Copyright notice must be kept intact and distributed with all -+ * fistgen sources INCLUDING sources generated by fistgen. -+ */ -+/* -+ * Copyright (C) 2004, 2005 Markus Klotzbuecher <mk@creamnet.de> -+ * -+ * This program is free software; you can redistribute it and/or -+ * modify it under the terms of the GNU General Public License -+ * as published by the Free Software Foundation; either version -+ * 2 of the License, or (at your option) any later version. -+ */ -+ -+/* -+ * $Id$ -+ */ -+ -+#ifndef __MINI_FO_H_ -+#define __MINI_FO_H_ -+ -+#ifdef __KERNEL__ -+ -+/* META stuff */ -+#define META_FILENAME "META_dAfFgHE39ktF3HD2sr" -+ -+/* use xattrs? */ -+#define XATTR -+ -+/* File attributes that when changed, result in a file beeing copied to storage */ -+#define COPY_FLAGS ATTR_MODE | ATTR_UID | ATTR_GID | ATTR_SIZE -+ -+/* -+ * mini_fo filestates -+ */ -+#define MODIFIED 1 -+#define UNMODIFIED 2 -+#define CREATED 3 -+#define DEL_REWRITTEN 4 -+#define DELETED 5 -+#define NON_EXISTANT 6 -+ -+/* fist file systems superblock magic */ -+# define MINI_FO_SUPER_MAGIC 0xf15f -+ -+/* -+ * STRUCTURES: -+ */ -+ -+/* mini_fo inode data in memory */ -+struct mini_fo_inode_info { -+ inode_t *wii_inode; -+ inode_t *wii_inode2; /* pointer to storage inode */ -+ -+ /* META-data lists */ -+ /* deleted list, ex wol */ -+ struct list_head deleted_list; -+ int deleted_list_size; -+ -+ /* renamed list */ -+ struct list_head renamed_list; -+ int renamed_list_size; -+ -+ /* add other lists here ... */ -+}; -+ -+/* mini_fo dentry data in memory */ -+struct mini_fo_dentry_info { -+ dentry_t *wdi_dentry; -+ dentry_t *wdi_dentry2; /* pointer to storage dentry */ -+ unsigned int state; /* state of the mini_fo dentry */ -+}; -+ -+ -+/* mini_fo super-block data in memory */ -+struct mini_fo_sb_info { -+ super_block_t *wsi_sb, *wsi_sb2; /* mk: might point to the same sb */ -+ struct vfsmount *hidden_mnt, *hidden_mnt2; -+ dentry_t *base_dir_dentry; -+ dentry_t *storage_dir_dentry; -+ ; -+}; -+ -+/* readdir_data, readdir helper struct */ -+struct readdir_data { -+ struct list_head ndl_list; /* linked list head ptr */ -+ int ndl_size; /* list size */ -+ int sto_done; /* flag to show that the storage dir entries have -+ * all been read an now follow base entries */ -+}; -+ -+/* file private data. */ -+struct mini_fo_file_info { -+ struct file *wfi_file; -+ struct file *wfi_file2; /* pointer to storage file */ -+ struct readdir_data rd; -+}; -+ -+/* struct ndl_entry */ -+struct ndl_entry { -+ struct list_head list; -+ char *name; -+ int len; -+}; -+ -+/******************************** -+ * META-data structures -+ ********************************/ -+ -+/* deleted entry */ -+struct deleted_entry { -+ struct list_head list; -+ char *name; -+ int len; -+}; -+ -+/* renamed entry */ -+struct renamed_entry { -+ struct list_head list; -+ char *old_name; /* old directory with full path */ -+ int old_len; /* length of above string */ -+ char *new_name; /* new directory name */ -+ int new_len; /* length of above string */ -+}; -+ -+/* attr_change entry */ -+struct attr_change_entry { -+ struct list_head list; -+ char *name; -+ int len; -+}; -+ -+/* link entry */ -+struct link_entry { -+ struct list_head list; -+ int links_moved; -+ int inum_base; -+ int inum_sto; -+ char *weird_name; -+ int weird_name_len; -+}; -+ -+ -+/* Some other stuff required for mini_fo_filldir64, copied from -+ * fs/readdir.c -+ */ -+ -+#define ROUND_UP64(x) (((x)+sizeof(u64)-1) & ~(sizeof(u64)-1)) -+#define NAME_OFFSET(de) ((int) ((de)->d_name - (char *) (de))) -+ -+ -+struct linux_dirent64 { -+ u64 d_ino; -+ s64 d_off; -+ unsigned short d_reclen; -+ unsigned char d_type; -+ char d_name[0]; -+}; -+ -+ -+struct getdents_callback64 { -+ struct linux_dirent64 * current_dir; -+ struct linux_dirent64 * previous; -+ int count; -+ int error; -+}; -+ -+struct linux_dirent { -+ unsigned long d_ino; -+ unsigned long d_off; -+ unsigned short d_reclen; -+ char d_name[1]; -+}; -+ -+struct getdents_callback { -+ struct linux_dirent * current_dir; -+ struct linux_dirent * previous; -+ int count; -+ int error; -+}; -+ -+ -+/* -+ * MACROS: -+ */ -+ -+/* file TO private_data */ -+# define ftopd(file) ((struct mini_fo_file_info *)((file)->private_data)) -+# define __ftopd(file) ((file)->private_data) -+/* file TO hidden_file */ -+# define ftohf(file) ((ftopd(file))->wfi_file) -+# define ftohf2(file) ((ftopd(file))->wfi_file2) -+ -+/* inode TO private_data */ -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,18) -+# define itopd(ino) ((struct mini_fo_inode_info *)(ino)->i_private) -+# define __itopd(ino) ((ino)->i_private) -+#else -+# define itopd(ino) ((struct mini_fo_inode_info *)(ino)->u.generic_ip) -+# define __itopd(ino) ((ino)->u.generic_ip) -+#endif -+/* inode TO hidden_inode */ -+# define itohi(ino) (itopd(ino)->wii_inode) -+# define itohi2(ino) (itopd(ino)->wii_inode2) -+ -+/* superblock TO private_data */ -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,0) -+# define stopd(super) ((struct mini_fo_sb_info *)(super)->s_fs_info) -+# define __stopd(super) ((super)->s_fs_info) -+#else -+# define stopd(super) ((struct mini_fo_sb_info *)(super)->u.generic_sbp) -+# define __stopd(super) ((super)->u.generic_sbp) -+#endif -+ -+/* unused? # define vfs2priv stopd */ -+/* superblock TO hidden_superblock */ -+ -+# define stohs(super) (stopd(super)->wsi_sb) -+# define stohs2(super) (stopd(super)->wsi_sb2) -+ -+/* dentry TO private_data */ -+# define dtopd(dentry) ((struct mini_fo_dentry_info *)(dentry)->d_fsdata) -+# define __dtopd(dentry) ((dentry)->d_fsdata) -+/* dentry TO hidden_dentry */ -+# define dtohd(dent) (dtopd(dent)->wdi_dentry) -+# define dtohd2(dent) (dtopd(dent)->wdi_dentry2) -+ -+/* dentry to state */ -+# define dtost(dent) (dtopd(dent)->state) -+# define sbt(sb) ((sb)->s_type->name) -+ -+#define IS_WRITE_FLAG(flag) (flag & (O_RDWR | O_WRONLY | O_APPEND)) -+#define IS_COPY_FLAG(flag) (flag & (COPY_FLAGS)) -+ -+/* macros to simplify non-SCA code */ -+# define MALLOC_PAGE_POINTERS(hidden_pages, num_hidden_pages) -+# define MALLOC_PAGEDATA_POINTERS(hidden_pages_data, num_hidden_pages) -+# define FREE_PAGE_POINTERS(hidden_pages, num) -+# define FREE_PAGEDATA_POINTERS(hidden_pages_data, num) -+# define FOR_EACH_PAGE -+# define CURRENT_HIDDEN_PAGE hidden_page -+# define CURRENT_HIDDEN_PAGEDATA hidden_page_data -+# define CURRENT_HIDDEN_PAGEINDEX page->index -+ -+/* -+ * EXTERNALS: -+ */ -+extern struct file_operations mini_fo_main_fops; -+extern struct file_operations mini_fo_dir_fops; -+extern struct inode_operations mini_fo_main_iops; -+extern struct inode_operations mini_fo_dir_iops; -+extern struct inode_operations mini_fo_symlink_iops; -+extern struct super_operations mini_fo_sops; -+extern struct dentry_operations mini_fo_dops; -+extern struct vm_operations_struct mini_fo_shared_vmops; -+extern struct vm_operations_struct mini_fo_private_vmops; -+extern struct address_space_operations mini_fo_aops; -+ -+#if 0 /* unused by mini_fo */ -+extern int mini_fo_interpose(dentry_t *hidden_dentry, dentry_t *this_dentry, super_block_t *sb, int flag); -+#if defined(FIST_FILTER_DATA) || defined(FIST_FILTER_SCA) -+extern page_t *mini_fo_get1page(file_t *file, int index); -+extern int mini_fo_fill_zeros(file_t *file, page_t *page, unsigned from); -+# endif /* FIST_FILTER_DATA || FIST_FILTER_SCA */ -+ -+ -+# define mini_fo_hidden_dentry(d) __mini_fo_hidden_dentry(__FILE__,__FUNCTION__,__LINE__,(d)) -+# define mini_fo_hidden_sto_dentry(d) __mini_fo_hidden_sto_dentry(__FILE__,__FUNCTION__,__LINE__,(d)) -+ -+extern dentry_t *__mini_fo_hidden_dentry(char *file, char *func, int line, dentry_t *this_dentry); -+extern dentry_t *__mini_fo_hidden_sto_dentry(char *file, char *func, int line, dentry_t *this_dentry); -+ -+extern int mini_fo_read_file(const char *filename, void *buf, int len); -+extern int mini_fo_write_file(const char *filename, void *buf, int len); -+extern dentry_t *fist_lookup(dentry_t *dir, const char *name, vnode_t **out, uid_t uid, gid_t gid); -+#endif /* unused by mini_fo */ -+ -+/* state transition functions */ -+extern int nondir_unmod_to_mod(dentry_t *dentry, int cp_flag); -+extern int nondir_del_rew_to_del(dentry_t *dentry); -+extern int nondir_creat_to_del(dentry_t *dentry); -+extern int nondir_mod_to_del(dentry_t *dentry); -+extern int nondir_unmod_to_del(dentry_t *dentry); -+ -+extern int dir_unmod_to_mod(dentry_t *dentry); -+ -+/* rename specials */ -+extern int rename_directory(inode_t *old_dir, dentry_t *old_dentry, inode_t *new_dir, dentry_t *new_dentry); -+extern int rename_nondir(inode_t *old_dir, dentry_t *old_dentry, inode_t *new_dir, dentry_t *new_dentry); -+ -+/* misc stuff */ -+extern int mini_fo_tri_interpose(dentry_t *hidden_dentry, -+ dentry_t *hidden_sto_dentry, -+ dentry_t *dentry, -+ super_block_t *sb, int flag); -+ -+extern int mini_fo_cp_cont(dentry_t *tgt_dentry, struct vfsmount *tgt_mnt, -+ dentry_t *src_dentry, struct vfsmount *src_mnt); -+ -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,0) -+extern int mini_fo_create(inode_t *dir, dentry_t *dentry, int mode, struct nameidata *nd); -+ -+extern int create_sto_nod(dentry_t *dentry, int mode, dev_t dev); -+extern int create_sto_reg_file(dentry_t *dentry, int mode, struct nameidata *nd); -+#else -+extern int mini_fo_create(inode_t *dir, dentry_t *dentry, int mode); -+ -+extern int create_sto_nod(dentry_t *dentry, int mode, int dev); -+extern int create_sto_reg_file(dentry_t *dentry, int mode); -+#endif -+ -+extern int create_sto_dir(dentry_t *dentry, int mode); -+ -+extern int exists_in_storage(dentry_t *dentry); -+extern int is_mini_fo_existant(dentry_t *dentry); -+extern int get_neg_sto_dentry(dentry_t *dentry); -+extern int build_sto_structure(dentry_t *dir, dentry_t *dentry); -+extern int get_mini_fo_bpath(dentry_t *dentry, char **bpath, int *bpath_len); -+extern dentry_t *bpath_walk(super_block_t *sb, char *bpath); -+extern int bpath_put(dentry_t *dentry); -+ -+/* check_mini_fo types functions */ -+extern int check_mini_fo_dentry(dentry_t *dentry); -+extern int check_mini_fo_file(file_t *file); -+extern int check_mini_fo_inode(inode_t *inode); -+ -+/* General meta functions, can be called from outside of meta.c */ -+extern int meta_build_lists(dentry_t *dentry); -+extern int meta_put_lists(dentry_t *dentry); -+extern int __meta_put_lists(inode_t *inode); -+ -+extern int meta_add_d_entry(dentry_t *dentry, const char *name, int len); -+extern int meta_add_r_entry(dentry_t *dentry, -+ const char *old_name, int old_len, -+ const char *new_name, int new_len); -+ -+extern int meta_remove_r_entry(dentry_t *dentry, const char *name, int len); -+ -+extern int meta_check_d_entry(dentry_t *dentry, const char *name, int len); -+extern int __meta_check_d_entry(inode_t *inode, const char *name, int len); -+ -+extern char* meta_check_r_entry(dentry_t *dentry, const char *name, int len); -+extern char* __meta_check_r_entry(inode_t *inode, const char *name, int len); -+extern int meta_is_r_entry(dentry_t *dentry, const char *name, int len); -+extern int __meta_is_r_entry(inode_t *inode, const char *name, int len); -+ -+/* Specific meta functions, should be called only inside meta.c */ -+extern int __meta_put_d_list(inode_t *inode); -+extern int __meta_put_r_list(inode_t *inode); -+ -+extern int meta_list_add_d_entry(dentry_t *dentry, -+ const char *name, int len); -+extern int meta_list_add_r_entry(dentry_t *dentry, -+ const char *old_name, int old_len, -+ const char *new_name, int new_len); -+ -+extern int meta_list_remove_r_entry(dentry_t *dentry, -+ const char *name, int len); -+ -+extern int __meta_list_remove_r_entry(inode_t *inode, -+ const char *name, int len); -+ -+extern int meta_write_d_entry(dentry_t *dentry, const char *name, int len); -+extern int meta_write_r_entry(dentry_t *dentry, -+ const char *old_name, int old_len, -+ const char *new_name, int new_len); -+ -+extern int meta_sync_lists(dentry_t *dentry); -+extern int meta_sync_d_list(dentry_t *dentry, int app_flag); -+extern int meta_sync_r_list(dentry_t *dentry, int app_flag); -+ -+/* ndl stuff */ -+extern int ndl_add_entry(struct readdir_data *rd, const char *name, int len); -+extern void ndl_put_list(struct readdir_data *rd); -+extern int ndl_check_entry(struct readdir_data *rd, -+ const char *name, int len); -+ -+ -+# define copy_inode_size(dst, src) \ -+ dst->i_size = src->i_size; \ -+ dst->i_blocks = src->i_blocks; -+ -+static inline void -+fist_copy_attr_atime(inode_t *dest, const inode_t *src) -+{ -+ ASSERT(dest != NULL); -+ ASSERT(src != NULL); -+ dest->i_atime = src->i_atime; -+} -+static inline void -+fist_copy_attr_times(inode_t *dest, const inode_t *src) -+{ -+ ASSERT(dest != NULL); -+ ASSERT(src != NULL); -+ dest->i_atime = src->i_atime; -+ dest->i_mtime = src->i_mtime; -+ dest->i_ctime = src->i_ctime; -+} -+static inline void -+fist_copy_attr_timesizes(inode_t *dest, const inode_t *src) -+{ -+ ASSERT(dest != NULL); -+ ASSERT(src != NULL); -+ dest->i_atime = src->i_atime; -+ dest->i_mtime = src->i_mtime; -+ dest->i_ctime = src->i_ctime; -+ copy_inode_size(dest, src); -+} -+static inline void -+fist_copy_attr_all(inode_t *dest, const inode_t *src) -+{ -+ ASSERT(dest != NULL); -+ ASSERT(src != NULL); -+ dest->i_mode = src->i_mode; -+ dest->i_nlink = src->i_nlink; -+ dest->i_uid = src->i_uid; -+ dest->i_gid = src->i_gid; -+ dest->i_rdev = src->i_rdev; -+ dest->i_atime = src->i_atime; -+ dest->i_mtime = src->i_mtime; -+ dest->i_ctime = src->i_ctime; -+#if LINUX_VERSION_CODE < KERNEL_VERSION(2,6,19) -+ dest->i_blksize = src->i_blksize; -+#endif -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,4,12) -+ dest->i_blkbits = src->i_blkbits; -+# endif /* linux 2.4.12 and newer */ -+ copy_inode_size(dest, src); -+#if LINUX_VERSION_CODE < KERNEL_VERSION(2,6,0) -+ dest->i_attr_flags = src->i_attr_flags; -+#else -+ dest->i_flags = src->i_flags; -+#endif -+} -+ -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,0) -+/* copied from linux/fs.h */ -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,16) -+static inline void double_lock(struct dentry *d1, struct dentry *d2) -+{ -+ struct mutex *m1 = &d1->d_inode->i_mutex; -+ struct mutex *m2 = &d2->d_inode->i_mutex; -+ if (m1 != m2) { -+ if ((unsigned long) m1 < (unsigned long) m2) { -+ struct mutex *tmp = m2; -+ m2 = m1; m1 = tmp; -+ } -+ mutex_lock(m1); -+ } -+ mutex_lock(m2); -+} -+ -+static inline void double_unlock(struct dentry *d1, struct dentry *d2) -+{ -+ struct mutex *m1 = &d1->d_inode->i_mutex; -+ struct mutex *m2 = &d2->d_inode->i_mutex; -+ mutex_unlock(m1); -+ if (m1 != m2) -+ mutex_unlock(m2); -+ dput(d1); -+ dput(d2); -+} -+ -+#else -+static inline void double_down(struct semaphore *s1, struct semaphore *s2) -+{ -+ if (s1 != s2) { -+ if ((unsigned long) s1 < (unsigned long) s2) { -+ struct semaphore *tmp = s2; -+ s2 = s1; s1 = tmp; -+ } -+ down(s1); -+ } -+ down(s2); -+} -+ -+static inline void double_up(struct semaphore *s1, struct semaphore *s2) -+{ -+ up(s1); -+ if (s1 != s2) -+ up(s2); -+} -+ -+static inline void double_lock(struct dentry *d1, struct dentry *d2) -+{ -+ double_down(&d1->d_inode->i_sem, &d2->d_inode->i_sem); -+} -+ -+static inline void double_unlock(struct dentry *d1, struct dentry *d2) -+{ -+ double_up(&d1->d_inode->i_sem,&d2->d_inode->i_sem); -+ dput(d1); -+ dput(d2); -+} -+#endif /* if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,16) */ -+#endif /* if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,0) */ -+#endif /* __KERNEL__ */ -+ -+/* -+ * Definitions for user and kernel code -+ */ -+ -+/* ioctls */ -+ -+#endif /* not __MINI_FO_H_ */ ---- /dev/null -+++ b/fs/mini_fo/mini_fo-merge -@@ -0,0 +1,180 @@ -+#!/bin/bash -+# -+# Copyright (C) 2005 Markus Klotzbuecher <mk@creamnet.de> -+# This program is free software; you can redistribute it and/or -+# modify it under the terms of the GNU General Public License -+# as published by the Free Software Foundation; either version -+# 2 of the License, or (at your option) any later version. -+# -+ -+BASE= -+STO= -+HELP= -+DRYRUN= -+VERBOSE= -+TMP="/tmp/" -+META_NAME="META_dAfFgHE39ktF3HD2sr" -+SKIP_DEL_LIST="skip-delete-list.mini_fo-merge" -+ -+COMMAND= -+exec_command() -+{ -+ if [ x$DRYRUN == "xset" ]; then -+ echo " would run: $COMMAND" -+ elif ! [ x$DRYRUN == "xset" ]; then -+ if [ x$VERBOSE == "xset" ]; then -+ echo " running: $COMMAND" -+ fi -+ eval $COMMAND -+ fi -+} -+ -+usage() -+{ -+cat <<EOF -+ -+USAGE: $0 -b <base dir> -s <storage dir> -+Version 0.1 -+ -+This script merges the contents of a mini_fo storage file system back -+to the base file system. -+ -+!!! Warning: This will modify the base filesystem and can destroy data -+ if used wrongly. -+ -+Options: -+ -b <base dir> -+ the directory of the base file system. -+ -+ -s <storage dir> -+ the directory of the storage file system. -+ -+ -d dry run, will not change anything and print the commands that -+ would be executed. -+ -+ -t tmp dir for storing temporary file. default: $TMP -+ -+ -v show what operations are performed. -+ -+ -h displays this message. -+ -+EOF -+} -+ -+# parse parameters -+while getopts hdvt:b:s: OPTS -+ do -+ case $OPTS in -+ h) HELP="set";; -+ d) DRYRUN="set";; -+ v) VERBOSE="set";; -+ b) BASE="$OPTARG";; -+ s) STO="$OPTARG";; -+ t) TMP="$OPTARG";; -+ ?) usage -+ exit 1;; -+ esac -+done -+ -+if [ "x$HELP" == "xset" ]; then -+ usage -+ exit -1 -+fi -+ -+if ! [ -d "$BASE" ] || ! [ -d "$STO" ]; then -+ echo -e "$0:\n Error, -s and/or -b argument missing. type $0 -h for help." -+ exit -1; -+fi -+ -+# get full paths -+pushd $STO; STO=`pwd`; popd -+pushd $BASE; BASE=`pwd`; popd -+TMP=${TMP%/} -+ -+ -+cat<<EOF -+############################################################################### -+# mini_fo-merge -+# -+# base dir: $BASE -+# storage dir: $STO -+# meta filename: $META_NAME -+# dry run: $DRYRUN -+# verbose: $VERBOSE -+# tmp files: $TMP -+############################################################################### -+ -+EOF -+ -+rm $TMP/$SKIP_DEL_LIST -+ -+# first process all renamed dirs -+echo "Merging renamed directories..." -+pushd $STO &> /dev/null -+find . -name $META_NAME -type f -print0 | xargs -0 -e grep -e '^R ' | tr -s ':R' ' ' | while read ENTRY; do -+ echo "entry: $ENTRY" -+ META_FILE=`echo $ENTRY | cut -d ' ' -f 1` -+ OLD_B_DIR=`echo $ENTRY | cut -d ' ' -f 2 | sed -e 's/\///'` -+ NEW_NAME=`echo $ENTRY | cut -d ' ' -f 3` -+ NEW_B_DIR=`echo $META_FILE | sed -e "s/$META_NAME/$NEW_NAME/" | sed -e 's/^\.\///'` -+ echo "META_FILE: $META_FILE" -+ echo "OLD_B_DIR: $OLD_B_DIR" -+ echo "NEW_NAME: $NEW_NAME" -+ echo "NEW_B_DIR: $NEW_B_DIR" -+ -+ pushd $BASE &> /dev/null -+ # remove an existing dir in storage -+ COMMAND="rm -rf $NEW_B_DIR"; exec_command -+ COMMAND="cp -R $OLD_B_DIR $NEW_B_DIR"; exec_command -+ echo "" -+ popd &> /dev/null -+ -+ # remember this dir to exclude it from deleting later -+ echo $NEW_B_DIR >> $TMP/$SKIP_DEL_LIST -+done -+ -+# delete all whiteouted files from base -+echo -e "\nDeleting whiteout'ed files from base file system..." -+find . -name $META_NAME -type f -print0 | xargs -0 -e grep -e '^D ' | sed -e 's/:D//' | while read ENTRY; do -+ META_FILE=`echo $ENTRY | cut -d ' ' -f 1` -+ DEL_NAME=`echo $ENTRY | cut -d ' ' -f 2` -+ DEL_FILE=`echo $META_FILE | sed -e "s/$META_NAME/$DEL_NAME/" | sed -e 's/^\.\///'` -+ grep -x $DEL_FILE $TMP/$SKIP_DEL_LIST &> /dev/null -+ if [ $? -ne 0 ]; then -+ pushd $BASE &> /dev/null -+ COMMAND="rm -rf $DEL_FILE"; exec_command -+ popd &> /dev/null -+ else -+ echo " excluding: $DEL_FILE as in skip-del-list." -+ fi -+done -+ -+# create all dirs and update permissions -+echo -e "\nSetting up directory structures in base file system..." -+find . -type d | sed -e 's/^\.\///' | while read DIR; do -+ PERMS=`stat -c %a $DIR` -+ DIR_UID=`stat -c %u $DIR` -+ DIR_GID=`stat -c %g $DIR` -+ pushd $BASE &> /dev/null -+ if ! [ -d $DIR ]; then -+ COMMAND="mkdir -p $DIR"; exec_command -+ fi -+ COMMAND="chmod $PERMS $DIR"; exec_command -+ COMMAND="chown $DIR_UID:$DIR_GID $DIR"; exec_command -+ popd &> /dev/null -+done -+ -+# merge all non-directory files -+echo -e "\nMerging all non-directory files...." -+for i in b c p f l s; do -+ find . -type $i | sed -e 's/^\.\///' | grep -v "$META_NAME" | while read FILE; do -+ pushd $BASE #&> /dev/null -+ COMMAND="cp -df $STO/$FILE $BASE/$FILE"; exec_command -+ popd &> /dev/null -+ done -+done -+popd &> /dev/null -+ -+#rm $TMP/$SKIP_DEL_LIST -+ -+echo "Done!" ---- /dev/null -+++ b/fs/mini_fo/mini_fo-overlay -@@ -0,0 +1,130 @@ -+#!/bin/bash -+# -+# Copyright (C) 2005 Markus Klotzbuecher <mk@creamnet.de> -+# This program is free software; you can redistribute it and/or -+# modify it under the terms of the GNU General Public License -+# as published by the Free Software Foundation; either version -+# 2 of the License, or (at your option) any later version. -+# -+ -+HELP= -+SUFF= -+MNTP= -+MNT_DIR="/mnt" -+STO= -+STO_DIR="/tmp" -+BASE= -+ -+usage() -+{ -+cat <<EOF -+ -+Usage: $0 [-s suffix] [-d sto_dir_dir] [-m mount point] base_dir -+Version 0.1 -+ -+This script overlays the given base directory using the mini_fo file -+system. If only the base directory base_dir is given, $0 -+will use a storage directory called "sto-<base_dir_name>" in $STO_DIR, -+and mount point "mini_fo-<base_dir_dir>" in $MNT_DIR. -+ -+Options: -+ -s <suffix> -+ add given suffix to storage directory and the mount -+ point. This is usefull for overlaying one base directory -+ several times and avoiding conflicts with storage directory -+ names and mount points. -+ -+ -d <sto_dir_dir> -+ change the directory in which the storage directory will be -+ created (default is currently "$STO_DIR". -+ -+ -m <mount point> -+ use an alternative directory to create the mini_fo -+ mountpoint (default is currently "$MNT_DIR". -+ -+ -h displays this message. -+ -+EOF -+exit 1; -+} -+ -+while getopts hm:s:d: OPTS -+ do -+ case $OPTS in -+ s) SUFF="$OPTARG";; -+ d) STO_DIR="$OPTARG";; -+ m) MNT_DIR="$OPTARG";; -+ h) HELP="set";; -+ ?) usage -+ exit 1;; -+ esac -+done -+shift $(($OPTIND - 1)) -+ -+BASE="$1" -+ -+if [ "x$HELP" == "xset" ]; then -+ usage -+ exit -1 -+fi -+ -+# fix suffix -+if [ "x$SUFF" != "x" ]; then -+ SUFF="-$SUFF" -+fi -+ -+# kill trailing slashes -+MNT_DIR=${MNT_DIR%/} -+STO_DIR=${STO_DIR%/} -+BASE=${BASE%/} -+ -+ -+if ! [ -d "$BASE" ]; then -+ echo "invalid base dir $BASE, run $0 -h for help." -+ exit -1 -+fi -+ -+# check opts -+if ! [ -d "$MNT_DIR" ]; then -+ echo "invalid mount dir $MNT_DIR, run $0 -h for help." -+ exit -1 -+fi -+ -+if ! [ -d "$STO_DIR" ]; then -+ echo "invalid sto_dir_dir $STO_DIR, run $0 -h for help." -+ exit -1 -+fi -+ -+MNTP="$MNT_DIR/mini_fo-`basename $BASE`$SUFF" -+STO="$STO_DIR/sto-`basename $BASE`$SUFF" -+ -+# create the mount point if it doesn't exist -+mkdir -p $MNTP -+if [ $? -ne 0 ]; then -+ echo "Error, failed to create mount point $MNTP" -+fi -+ -+mkdir -p $STO -+if [ $? -ne 0 ]; then -+ echo "Error, failed to create storage dir $STO" -+fi -+ -+# check if fs is already mounted -+mount | grep mini_fo | grep $MNTP &> /dev/null -+if [ $? -eq 0 ]; then -+ echo "Error, existing mini_fo mount at $MNTP." -+ exit -1 -+fi -+ -+mount | grep mini_fo | grep $STO &> /dev/null -+if [ $? -eq 0 ]; then -+ echo "Error, $STO seems to be used already." -+ exit -1 -+fi -+ -+# mount -+mount -t mini_fo -o base=$BASE,sto=$STO $BASE $MNTP -+ -+if [ $? -ne 0 ]; then -+ echo "Error, mounting failed, maybe no permisson to mount?" -+fi ---- /dev/null -+++ b/fs/mini_fo/mmap.c -@@ -0,0 +1,637 @@ -+/* -+ * Copyright (c) 1997-2003 Erez Zadok -+ * Copyright (c) 2001-2003 Stony Brook University -+ * -+ * For specific licensing information, see the COPYING file distributed with -+ * this package, or get one from ftp://ftp.filesystems.org/pub/fist/COPYING. -+ * -+ * This Copyright notice must be kept intact and distributed with all -+ * fistgen sources INCLUDING sources generated by fistgen. -+ */ -+/* -+ * Copyright (C) 2004, 2005 Markus Klotzbuecher <mk@creamnet.de> -+ * -+ * This program is free software; you can redistribute it and/or -+ * modify it under the terms of the GNU General Public License -+ * as published by the Free Software Foundation; either version -+ * 2 of the License, or (at your option) any later version. -+ */ -+ -+/* -+ * $Id$ -+ */ -+ -+#ifdef HAVE_CONFIG_H -+# include <config.h> -+#endif /* HAVE_CONFIG_H */ -+ -+#include "fist.h" -+#include "mini_fo.h" -+ -+ -+#ifdef FIST_COUNT_WRITES -+/* for counting writes in the middle vs. regular writes */ -+unsigned long count_writes = 0, count_writes_middle = 0; -+#endif /* FIST_COUNT_WRITES */ -+ -+/* forward declaration of commit write and prepare write */ -+STATIC int mini_fo_commit_write(file_t *file, page_t *page, unsigned from, unsigned to); -+STATIC int mini_fo_prepare_write(file_t *file, page_t *page, unsigned from, unsigned to); -+ -+ -+/* -+ * Function for handling creation of holes when lseek-ing past the -+ * end of the file and then writing some data. -+ */ -+int -+mini_fo_fill_zeros(file_t* file, page_t *page, unsigned from) -+{ -+ int err = 0; -+ dentry_t *dentry = file->f_dentry; -+ inode_t *inode = dentry->d_inode; -+ page_t *tmp_page; -+ int index; -+ -+ print_entry_location(); -+ -+ for (index = inode->i_size >> PAGE_CACHE_SHIFT; index < page->index; index++) { -+ tmp_page = mini_fo_get1page(file, index); -+ if (IS_ERR(tmp_page)) { -+ err = PTR_ERR(tmp_page); -+ goto out; -+ } -+ -+ /* -+ * zero out rest of the contents of the page between the appropriate -+ * offsets. -+ */ -+ memset((char*)page_address(tmp_page) + (inode->i_size & ~PAGE_CACHE_MASK), 0, PAGE_CACHE_SIZE - (inode->i_size & ~PAGE_CACHE_MASK)); -+ -+ if (! (err = mini_fo_prepare_write(file, tmp_page, 0, PAGE_CACHE_SIZE))) -+ err = mini_fo_commit_write(file, tmp_page, 0, PAGE_CACHE_SIZE); -+ -+ page_cache_release(tmp_page); -+ if (err < 0) -+ goto out; -+ if (current->need_resched) -+ schedule(); -+ } -+ -+ /* zero out appropriate parts of last page */ -+ -+ /* -+ * if the encoding type is block, then adjust the 'from' (where the -+ * zeroing will start) offset appropriately -+ */ -+ from = from & (~(FIST_ENCODING_BLOCKSIZE - 1)); -+ -+ if ((from - (inode->i_size & ~PAGE_CACHE_MASK)) > 0) { -+ -+ memset((char*)page_address(page) + (inode->i_size & ~PAGE_CACHE_MASK), 0, from - (inode->i_size & ~PAGE_CACHE_MASK)); -+ if (! (err = mini_fo_prepare_write(file, page, 0, PAGE_CACHE_SIZE))) -+ err = mini_fo_commit_write(file, page, 0, PAGE_CACHE_SIZE); -+ -+ if (err < 0) -+ goto out; -+ if (current->need_resched) -+ schedule(); -+ } -+ -+ out: -+ print_exit_status(err); -+ return err; -+} -+ -+ -+ -+STATIC int -+mini_fo_writepage(page_t *page) -+{ -+ int err = -EIO; -+ inode_t *inode; -+ inode_t *hidden_inode; -+ page_t *hidden_page; -+ char *kaddr, *hidden_kaddr; -+ -+ print_entry_location(); -+ -+ inode = page->mapping->host; -+ hidden_inode = itohi(inode); -+ -+ /* -+ * writepage is called when shared mmap'ed files need to write -+ * their pages, while prepare/commit_write are called from the -+ * non-paged write() interface. (However, in 2.3 the two interfaces -+ * share the same cache, while in 2.2 they didn't.) -+ * -+ * So we pretty much have to duplicate much of what commit_write does. -+ */ -+ -+ /* find lower page (returns a locked page) */ -+ hidden_page = grab_cache_page(hidden_inode->i_mapping, page->index); -+ if (!hidden_page) -+ goto out; -+ -+ /* get page address, and encode it */ -+ kaddr = (char *) kmap(page); -+ hidden_kaddr = (char*) kmap(hidden_page); -+ mini_fo_encode_block(kaddr, hidden_kaddr, PAGE_CACHE_SIZE, inode, inode->i_sb, page->index); -+ /* if encode_block could fail, then return error */ -+ kunmap(page); -+ kunmap(hidden_page); -+ -+ /* call lower writepage (expects locked page) */ -+ err = hidden_inode->i_mapping->a_ops->writepage(hidden_page); -+ -+ /* -+ * update mtime and ctime of lower level file system -+ * mini_fo' mtime and ctime are updated by generic_file_write -+ */ -+ hidden_inode->i_mtime = hidden_inode->i_ctime = CURRENT_TIME; -+ -+#if LINUX_VERSION_CODE < KERNEL_VERSION(2,4,1) -+ UnlockPage(hidden_page); /* b/c grab_cache_page locked it */ -+# endif /* kernel older than 2.4.1 */ -+ page_cache_release(hidden_page); /* b/c grab_cache_page increased refcnt */ -+ -+ if (err) -+ ClearPageUptodate(page); -+ else -+ SetPageUptodate(page); -+ out: -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,4,1) -+ UnlockPage(page); -+# endif /* kernel 2.4.1 and newer */ -+ print_exit_status(err); -+ return err; -+} -+ -+ -+/* -+ * get one page from cache or lower f/s, return error otherwise. -+ * returns unlocked, up-to-date page (if ok), with increased refcnt. -+ */ -+page_t * -+mini_fo_get1page(file_t *file, int index) -+{ -+ page_t *page; -+ dentry_t *dentry; -+ inode_t *inode; -+ struct address_space *mapping; -+ int err; -+ -+ print_entry_location(); -+ -+ dentry = file->f_dentry; /* CPW: Moved below print_entry_location */ -+ inode = dentry->d_inode; -+ mapping = inode->i_mapping; -+ -+ fist_dprint(8, "%s: read page index %d pid %d\n", __FUNCTION__, index, current->pid); -+ if (index < 0) { -+ printk("%s BUG: index=%d\n", __FUNCTION__, index); -+ page = ERR_PTR(-EIO); -+ goto out; -+ } -+ page = read_cache_page(mapping, -+ index, -+ (filler_t *) mapping->a_ops->readpage, -+ (void *) file); -+ if (IS_ERR(page)) -+ goto out; -+ wait_on_page(page); -+ if (!Page_Uptodate(page)) { -+ lock_page(page); -+ err = mapping->a_ops->readpage(file, page); -+ if (err) { -+ page = ERR_PTR(err); -+ goto out; -+ } -+ wait_on_page(page); -+ if (!Page_Uptodate(page)) { -+ page = ERR_PTR(-EIO); -+ goto out; -+ } -+ } -+ -+ out: -+ print_exit_pointer(page); -+ return page; -+} -+ -+ -+/* -+ * get one page from cache or lower f/s, return error otherwise. -+ * similar to get1page, but doesn't guarantee that it will return -+ * an unlocked page. -+ */ -+page_t * -+mini_fo_get1page_cached(file_t *file, int index) -+{ -+ page_t *page; -+ dentry_t *dentry; -+ inode_t *inode; -+ struct address_space *mapping; -+ int err; -+ -+ print_entry_location(); -+ -+ dentry = file->f_dentry; /* CPW: Moved below print_entry_location */ -+ inode = dentry->d_inode; -+ mapping = inode->i_mapping; -+ -+ fist_dprint(8, "%s: read page index %d pid %d\n", __FUNCTION__, index, current->pid); -+ if (index < 0) { -+ printk("%s BUG: index=%d\n", __FUNCTION__, index); -+ page = ERR_PTR(-EIO); -+ goto out; -+ } -+ page = read_cache_page(mapping, -+ index, -+ (filler_t *) mapping->a_ops->readpage, -+ (void *) file); -+ if (IS_ERR(page)) -+ goto out; -+ -+ out: -+ print_exit_pointer(page); -+ return page; -+} -+ -+ -+/* -+ * readpage is called from generic_page_read and the fault handler. -+ * If your file system uses generic_page_read for the read op, it -+ * must implement readpage. -+ * -+ * Readpage expects a locked page, and must unlock it. -+ */ -+STATIC int -+mini_fo_do_readpage(file_t *file, page_t *page) -+{ -+ int err = -EIO; -+ dentry_t *dentry; -+ file_t *hidden_file = NULL; -+ dentry_t *hidden_dentry; -+ inode_t *inode; -+ inode_t *hidden_inode; -+ char *page_data; -+ page_t *hidden_page; -+ char *hidden_page_data; -+ int real_size; -+ -+ print_entry_location(); -+ -+ dentry = file->f_dentry; /* CPW: Moved below print_entry_location */ -+ if (ftopd(file) != NULL) -+ hidden_file = ftohf(file); -+ hidden_dentry = dtohd(dentry); -+ inode = dentry->d_inode; -+ hidden_inode = itohi(inode); -+ -+ fist_dprint(7, "%s: requesting page %d from file %s\n", __FUNCTION__, page->index, dentry->d_name.name); -+ -+ MALLOC_PAGE_POINTERS(hidden_pages, num_hidden_pages); -+ MALLOC_PAGEDATA_POINTERS(hidden_pages_data, num_hidden_pages); -+ FOR_EACH_PAGE -+ CURRENT_HIDDEN_PAGE = NULL; -+ -+ /* find lower page (returns a locked page) */ -+ FOR_EACH_PAGE { -+ fist_dprint(8, "%s: Current page index = %d\n", __FUNCTION__, CURRENT_HIDDEN_PAGEINDEX); -+ CURRENT_HIDDEN_PAGE = read_cache_page(hidden_inode->i_mapping, -+ CURRENT_HIDDEN_PAGEINDEX, -+ (filler_t *) hidden_inode->i_mapping->a_ops->readpage, -+ (void *) hidden_file); -+ if (IS_ERR(CURRENT_HIDDEN_PAGE)) { -+ err = PTR_ERR(CURRENT_HIDDEN_PAGE); -+ CURRENT_HIDDEN_PAGE = NULL; -+ goto out_release; -+ } -+ } -+ -+ /* -+ * wait for the page data to show up -+ * (signaled by readpage as unlocking the page) -+ */ -+ FOR_EACH_PAGE { -+ wait_on_page(CURRENT_HIDDEN_PAGE); -+ if (!Page_Uptodate(CURRENT_HIDDEN_PAGE)) { -+ /* -+ * call readpage() again if we returned from wait_on_page with a -+ * page that's not up-to-date; that can happen when a partial -+ * page has a few buffers which are ok, but not the whole -+ * page. -+ */ -+ lock_page(CURRENT_HIDDEN_PAGE); -+ err = hidden_inode->i_mapping->a_ops->readpage(hidden_file, -+ CURRENT_HIDDEN_PAGE); -+ if (err) { -+ CURRENT_HIDDEN_PAGE = NULL; -+ goto out_release; -+ } -+ wait_on_page(CURRENT_HIDDEN_PAGE); -+ if (!Page_Uptodate(CURRENT_HIDDEN_PAGE)) { -+ err = -EIO; -+ goto out_release; -+ } -+ } -+ } -+ -+ /* map pages, get their addresses */ -+ page_data = (char *) kmap(page); -+ FOR_EACH_PAGE -+ CURRENT_HIDDEN_PAGEDATA = (char *) kmap(CURRENT_HIDDEN_PAGE); -+ -+ /* if decode_block could fail, then return error */ -+ err = 0; -+ real_size = hidden_inode->i_size - (page->index << PAGE_CACHE_SHIFT); -+ if (real_size <= 0) -+ memset(page_data, 0, PAGE_CACHE_SIZE); -+ else if (real_size < PAGE_CACHE_SIZE) { -+ mini_fo_decode_block(hidden_page_data, page_data, real_size, inode, inode->i_sb, page->index); -+ memset(page_data + real_size, 0, PAGE_CACHE_SIZE - real_size); -+ } else -+ mini_fo_decode_block(hidden_page_data, page_data, PAGE_CACHE_SIZE, inode, inode->i_sb, page->index); -+ -+ FOR_EACH_PAGE -+ kunmap(CURRENT_HIDDEN_PAGE); -+ kunmap(page); -+ -+ out_release: -+ FOR_EACH_PAGE -+ if (CURRENT_HIDDEN_PAGE) -+ page_cache_release(CURRENT_HIDDEN_PAGE); /* undo read_cache_page */ -+ -+ FREE_PAGE_POINTERS(hidden_pages, num_hidden_pages); -+ FREE_PAGEDATA_POINTERS(hidden_pages_data, num_hidden_pages); -+ -+ out: -+ if (err == 0) -+ SetPageUptodate(page); -+ else -+ ClearPageUptodate(page); -+ -+ print_exit_status(err); -+ return err; -+} -+ -+ -+STATIC int -+mini_fo_readpage(file_t *file, page_t *page) -+{ -+ int err; -+ print_entry_location(); -+ -+ err = mini_fo_do_readpage(file, page); -+ -+ /* -+ * we have to unlock our page, b/c we _might_ have gotten a locked page. -+ * but we no longer have to wakeup on our page here, b/c UnlockPage does -+ * it -+ */ -+ UnlockPage(page); -+ -+ print_exit_status(err); -+ return err; -+} -+ -+ -+STATIC int -+mini_fo_prepare_write(file_t *file, page_t *page, unsigned from, unsigned to) -+{ -+ int err = 0; -+ -+ print_entry_location(); -+ -+ /* -+ * we call kmap(page) only here, and do the kunmap -+ * and the actual downcalls, including unlockpage and uncache -+ * in commit_write. -+ */ -+ kmap(page); -+ -+ /* fast path for whole page writes */ -+ if (from == 0 && to == PAGE_CACHE_SIZE) -+ goto out; -+ /* read the page to "revalidate" our data */ -+ /* call the helper function which doesn't unlock the page */ -+ if (!Page_Uptodate(page)) -+ err = mini_fo_do_readpage(file, page); -+ -+ out: -+ print_exit_status(err); -+ return err; -+} -+ -+ -+ -+STATIC int -+mini_fo_commit_write(file_t *file, page_t *page, unsigned from, unsigned to) -+{ -+ int err = -ENOMEM; -+ inode_t *inode; -+ inode_t *hidden_inode; -+ page_t *hidden_page; -+ file_t *hidden_file = NULL; -+ loff_t pos; -+ unsigned bytes = to - from; -+ unsigned hidden_from, hidden_to, hidden_bytes; -+ -+ print_entry_location(); -+ -+ inode = page->mapping->host; /* CPW: Moved below print_entry_location */ -+ hidden_inode = itohi(inode); -+ -+ ASSERT(file != NULL); -+ /* -+ * here we have a kmapped page, with data from the user copied -+ * into it. we need to encode_block it, and then call the lower -+ * commit_write. We also need to simulate same behavior of -+ * generic_file_write, and call prepare_write on the lower f/s first. -+ */ -+#ifdef FIST_COUNT_WRITES -+ count_writes++; -+# endif /* FIST_COUNT_WRITES */ -+ -+ /* this is append and/or extend -- we can't have holes so fill them in */ -+ if (page->index > (hidden_inode->i_size >> PAGE_CACHE_SHIFT)) { -+ page_t *tmp_page; -+ int index; -+ for (index = hidden_inode->i_size >> PAGE_CACHE_SHIFT; index < page->index; index++) { -+ tmp_page = mini_fo_get1page(file, index); -+ if (IS_ERR(tmp_page)) { -+ err = PTR_ERR(tmp_page); -+ goto out; -+ } -+ /* zero out the contents of the page at the appropriate offsets */ -+ memset((char*)page_address(tmp_page) + (inode->i_size & ~PAGE_CACHE_MASK), 0, PAGE_CACHE_SIZE - (inode->i_size & ~PAGE_CACHE_MASK)); -+ if (!(err = mini_fo_prepare_write(file, tmp_page, 0, PAGE_CACHE_SIZE))) -+ err = mini_fo_commit_write(file, tmp_page, 0, PAGE_CACHE_SIZE); -+ page_cache_release(tmp_page); -+ if (err < 0) -+ goto out; -+ if (current->need_resched) -+ schedule(); -+ } -+ } -+ -+ if (ftopd(file) != NULL) -+ hidden_file = ftohf(file); -+ -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,16) -+ mutex_lock(&hidden_inode->i_mutex); -+#else -+ down(&hidden_inode->i_sem); -+#endif -+ /* find lower page (returns a locked page) */ -+ hidden_page = grab_cache_page(hidden_inode->i_mapping, page->index); -+ if (!hidden_page) -+ goto out; -+ -+#if FIST_ENCODING_BLOCKSIZE > 1 -+# error encoding_blocksize greater than 1 is not yet supported -+# endif /* FIST_ENCODING_BLOCKSIZE > 1 */ -+ -+ hidden_from = from & (~(FIST_ENCODING_BLOCKSIZE - 1)); -+ hidden_to = ((to + FIST_ENCODING_BLOCKSIZE - 1) & (~(FIST_ENCODING_BLOCKSIZE - 1))); -+ if ((page->index << PAGE_CACHE_SHIFT) + to > hidden_inode->i_size) { -+ -+ /* -+ * if this call to commit_write had introduced holes and the code -+ * for handling holes was invoked, then the beginning of this page -+ * must be zeroed out -+ * zero out bytes from 'size_of_file%pagesize' to 'from'. -+ */ -+ if ((hidden_from - (inode->i_size & ~PAGE_CACHE_MASK)) > 0) -+ memset((char*)page_address(page) + (inode->i_size & ~PAGE_CACHE_MASK), 0, hidden_from - (inode->i_size & ~PAGE_CACHE_MASK)); -+ -+ } -+ hidden_bytes = hidden_to - hidden_from; -+ -+ /* call lower prepare_write */ -+ err = -EINVAL; -+ if (hidden_inode->i_mapping && -+ hidden_inode->i_mapping->a_ops && -+ hidden_inode->i_mapping->a_ops->prepare_write) -+ err = hidden_inode->i_mapping->a_ops->prepare_write(hidden_file, -+ hidden_page, -+ hidden_from, -+ hidden_to); -+ if (err) -+ /* don't leave locked pages behind, esp. on an ENOSPC */ -+ goto out_unlock; -+ -+ fist_dprint(8, "%s: encoding %d bytes\n", __FUNCTION__, hidden_bytes); -+ mini_fo_encode_block((char *) page_address(page) + hidden_from, (char*) page_address(hidden_page) + hidden_from, hidden_bytes, inode, inode->i_sb, page->index); -+ /* if encode_block could fail, then goto unlock and return error */ -+ -+ /* call lower commit_write */ -+ err = hidden_inode->i_mapping->a_ops->commit_write(hidden_file, -+ hidden_page, -+ hidden_from, -+ hidden_to); -+ -+ if (err < 0) -+ goto out_unlock; -+ -+ err = bytes; /* convert error to no. of bytes */ -+ -+ inode->i_blocks = hidden_inode->i_blocks; -+ /* we may have to update i_size */ -+ pos = (page->index << PAGE_CACHE_SHIFT) + to; -+ if (pos > inode->i_size) -+ inode->i_size = pos; -+ -+ /* -+ * update mtime and ctime of lower level file system -+ * mini_fo' mtime and ctime are updated by generic_file_write -+ */ -+ hidden_inode->i_mtime = hidden_inode->i_ctime = CURRENT_TIME; -+ -+ mark_inode_dirty_sync(inode); -+ -+ out_unlock: -+ UnlockPage(hidden_page); -+ page_cache_release(hidden_page); -+ kunmap(page); /* kmap was done in prepare_write */ -+ out: -+ /* we must set our page as up-to-date */ -+ if (err < 0) -+ ClearPageUptodate(page); -+ else -+ SetPageUptodate(page); -+ -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,16) -+ mutex_unlock(&hidden_inode->i_mutex); -+#else -+ up(&hidden_inode->i_sem); -+#endif -+ print_exit_status(err); -+ return err; /* assume all is ok */ -+} -+ -+ -+STATIC int -+mini_fo_bmap(struct address_space *mapping, long block) -+{ -+ int err = 0; -+ inode_t *inode; -+ inode_t *hidden_inode; -+ -+ print_entry_location(); -+ -+ inode = (inode_t *) mapping->host; -+ hidden_inode = itohi(inode); -+ -+ if (hidden_inode->i_mapping->a_ops->bmap) -+ err = hidden_inode->i_mapping->a_ops->bmap(hidden_inode->i_mapping, block); -+ print_exit_location(); -+ return err; -+} -+ -+ -+/* -+ * This function is copied verbatim from mm/filemap.c. -+ * XXX: It should be simply moved to some header file instead -- bug Al about it! -+ */ -+static inline int sync_page(struct page *page) -+{ -+ struct address_space *mapping = page->mapping; -+ -+ if (mapping && mapping->a_ops && mapping->a_ops->sync_page) -+ return mapping->a_ops->sync_page(page); -+ return 0; -+} -+ -+ -+/* -+ * XXX: we may not need this function if not FIST_FILTER_DATA. -+ * FIXME: for FIST_FILTER_SCA, get all lower pages and sync them each. -+ */ -+STATIC int -+mini_fo_sync_page(page_t *page) -+{ -+ int err = 0; -+ inode_t *inode; -+ inode_t *hidden_inode; -+ page_t *hidden_page; -+ -+ print_entry_location(); -+ -+ inode = page->mapping->host; /* CPW: Moved below print_entry_location */ -+ hidden_inode = itohi(inode); -+ -+ /* find lower page (returns a locked page) */ -+ hidden_page = grab_cache_page(hidden_inode->i_mapping, page->index); -+ if (!hidden_page) -+ goto out; -+ -+ err = sync_page(hidden_page); -+ -+ UnlockPage(hidden_page); /* b/c grab_cache_page locked it */ -+ page_cache_release(hidden_page); /* b/c grab_cache_page increased refcnt */ -+ -+ out: -+ print_exit_status(err); -+ return err; -+} ---- /dev/null -+++ b/fs/mini_fo/README -@@ -0,0 +1,163 @@ -+README for the mini_fo overlay file system -+========================================= -+ -+ -+WHAT IS MINI_FO? -+---------------- -+ -+mini_fo is a virtual kernel file system that can make read-only -+file systems writable. This is done by redirecting modifying operations -+to a writeable location called "storage directory", and leaving the -+original data in the "base directory" untouched. When reading, the -+file system merges the modifed and original data so that only the -+newest versions will appear. This occurs transparently to the user, -+who can access the data like on any other read-write file system. -+ -+Base and storage directories may be located on the same or on -+different partitions and may be of different file system types. While -+the storage directory obviously needs to be writable, the base may or -+may not be writable, what doesn't matter as it will no be modified -+anyway. -+ -+ -+WHAT IS GOOD FOR? -+----------------- -+ -+The primary purpose of the mini_fo file system is to allow easy -+software updates to embedded systems, that often store their root -+file system in a read-only flash file system, but there are many -+more as for example sandboxing, or for allowing live-cds to -+permanently store information. -+ -+ -+BUILDING -+-------- -+This should be simple. Adjust the Makefile to point to the correct -+kernel headers you want to build the module for. Then: -+ -+ # make -+ -+should build "mini_fo.o" for a 2.4 kernel or "mini_fo.ko" for a 2.6 -+kernel. -+ -+If you are building the module for you current kernel, you can install -+the module (as root): -+ -+ # make install -+ -+or uninstall with -+ -+ # make uninstall -+ -+ -+USING THE FILE SYSTEM -+-------------------- -+ -+the general mount syntax is: -+ -+ mount -t mini_fo -o base=<base directory>,sto=<storage directory>\ -+ <base directory> <mount point> -+ -+Example: -+ -+You have mounted a cdrom to /mnt/cdrom and want to modifiy some files -+on it: -+ -+load the module (as root) -+ -+ # insmod mini_fo.o for a 2.4 kernel or -+ -+ # insmod mini_fo.ko for a 2.6 kernel -+ -+ -+create a storage dir in tmp and a mountpoint for mini_fo: -+ -+ # mkdir /tmp/sto -+ # mkdir /mnt/mini_fo -+ -+and mount the mini_fo file system: -+ -+ # mount -t mini_fo -o base=/mnt/cdrom,sto=/tmp/sto /mnt/cdrom /mnt/mini_fo -+ -+ -+Now the data stored on the cd can be accessed via the mini_fo -+mountpoint just like any read-write file system, files can be modified -+and deleted, new ones can be created and so on. When done unmount the -+file system: -+ -+ # unmount /mnt/mini_fo -+ -+Note that if the file system is mounted again using the same storage -+file system, of course it will appear in the modified state again. If -+you remount it using an new empty storage directory, it will be -+unmodified. Therefore by executing: -+ -+ # cd /tmp/sto -+ # rm -rf * -+ -+you can nuke all the changes you made to the original file system. But -+ remember NEVER do this while the mini_fo file system is mounted! -+ -+ -+Alternatively you can use the mini_fo-overlay bash script, that -+simplifies managing mini_fo mounts. See TOOLS Section. -+ -+ -+TOOLS -+----- -+ -+mini_fo-merge (experimental): -+ -+This is a bash script that will merge changes contained in the storage -+directory back to the base directory. This allows mini_fo to function -+as a cache file system by overlaying a slow (network, ...) file system -+and using a fast (ramdisk, ...) as storage. When done, changes can be -+merged back to the (slow) base with mini_fo-merge. See "mini_fo-merge -+-h" for details. -+ -+It can be usefull for merging changes back after a successfull test -+(patches, software updates...) -+ -+ -+mini_fo-overlay: -+ -+This bash script simplifies managing one or more mini_fo mounts. For -+overlaying a directory called "basedir1", you can just call: -+ -+ # mini_fo-overlay basedir1 -+ -+This will mount mini_fo with "basedir1" as base, "/tmp/sto-basedir1/" -+as storage to "/mnt/mini_fo-basedir1/". It has more options though, -+type "mini_fo-overlay -h" for details. -+ -+ -+DOCUMENTATION, REPORTING BUGS, GETTING HELP -+------------------------------------------- -+ -+Please visit the mini_fo project page at: -+ -+http://www.denx.de/twiki/bin/view/Know/MiniFOHome -+ -+ -+WARNINGS -+-------- -+ -+Never modify the base or the storage directorys while the mini_fo -+file system is mounted, or you might crash you system. Simply accessing -+and reading should not cause any trouble. -+ -+Exporting a mini_fo mount point via NFS has not been tested, and may -+or may not work. -+ -+Check the RELEASE_NOTES for details on bugs and features. -+ -+ -+ -+Copyright (C) 2004, 2005 Markus Klotzbuecher <mk@creamnet.de> -+ -+This program is free software; you can redistribute it and/or -+modify it under the terms of the GNU General Public License -+as published by the Free Software Foundation; either version -+2 of the License, or (at your option) any later version. -+ -+ ---- /dev/null -+++ b/fs/mini_fo/RELEASE_NOTES -@@ -0,0 +1,111 @@ -+Release: mini_fo-0.6.1 (v0-6-1) -+Date: 21.09.2005 -+ -+ -+Changes: -+-------- -+v0-6-1: -+ -+- bugfixes (see ChangeLog) -+ -+- two helper scripts "mini_fo_merge" and "mini_fo_overlay" (see -+ README for details). -+ -+v0-6-0: -+ -+- Support for 2.4 and 2.6 (see Makefile) -+ -+- Partial hard link support (creating works as expected, but already -+ existing links in the base file system will be treated as if they -+ were individual files). -+ -+- Various bugfixes and cleanups. -+ -+ -+v0-6-0-pre1: -+ -+- This is mini_fo-0-6-0-pre1! This release is a complete rewrite of -+ many vital mini_fo parts such as the old whiteout list code which -+ has been replaced by the new META subsystem. -+ -+- Light weight directory renaming implemented. This means if a -+ directory is renamed via the mini_fo filesystem this will no longer -+ result in a complete copy in storage, instead only one empty -+ directory will be created. All base filed contained in the original -+ directory stay there until modified. -+ -+- Special files (creating, renaming, deleting etc.) now working. -+ -+- Many bugfixes and cleanup, mini_fo is now a lot more stable. -+ -+ -+v0-5-10: -+ -+- Final release of the 0-5-* versions. Next will be a complete rewrite -+ of many features. This release contains several bugfixes related to -+ directory renaming. -+ -+ -+v0-5-10-pre6: -+ -+- Lots of cleanup and several bugfixes related to directory deleting -+ -+- Directory renaming suddenly works, what is most likely due to the -+ fact tha that "mv" is smart: if the classic rename doesn't work it -+ will assume that source and target file are on different fs and will -+ copy the directory and try to remove the source directory. Until -+ directory removing wasn't implemented, it would fail to do this and -+ rollback. -+ So, directory renaming works for now, but it doesn't yet do what you -+ would expect from a overlay fs, so use with care. -+ -+ -+v0-5-10-pre5: -+ -+- implemented directory deleting -+- made parsing of mount options more stable -+- New format of mount options! (See README) -+- I can't reproduce the unknown panic with 2.4.25 anymore, so I'll -+ happily assume it never existed! -+ -+ -+Implemented features: -+--------------------- -+ -+- creating hard links (see BUGS on already existing hard links) -+- lightweight directory renaming -+- renaming device files, pipes, sockets, etc. -+- creating, renaming, deleting of special files -+- deleting directorys -+- general directory reading (simple "ls" ) -+- creating files in existing directorys -+- creating directorys -+- renaming files. -+- reading and writing files (involves opening) -+- appending to files (creates copy in storage) -+- deleting files -+- llseek works too, what allows editors to work -+- persistency (a deleted file stay deleted over remounts) -+- use of symbolic links -+- creating of device files -+ -+ -+Not (yet) implemented features: -+------------------------------- -+ -+- full hard link support. -+ -+ -+ -+BUGS: -+----- -+ -+Hard links in the base file system will be treated as individual -+files, not as links to one inode. -+ -+The main problem with hard links isn't allowing to create them, but -+their pure existence. If you modify a base hard link, the changes made -+will only show up on this link, the other link will remain in the -+original state. I hope to fix this someday. Please note that this does -+not effect the special hard links '.' and '..', that are handled -+seperately by the lower fs. ---- /dev/null -+++ b/fs/mini_fo/state.c -@@ -0,0 +1,620 @@ -+/* -+ * Copyright (C) 2005 Markus Klotzbuecher <mk@creamnet.de> -+ * -+ * This program is free software; you can redistribute it and/or -+ * modify it under the terms of the GNU General Public License -+ * as published by the Free Software Foundation; either version -+ * 2 of the License, or (at your option) any later version. -+ */ -+ -+#ifdef HAVE_CONFIG_H -+# include <config.h> -+#endif /* HAVE_CONFIG_H */ -+ -+#include "fist.h" -+#include "mini_fo.h" -+ -+ -+/* create the storage file, setup new states */ -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,0) -+int create_sto_reg_file(dentry_t *dentry, int mode, struct nameidata *nd) -+#else -+int create_sto_reg_file(dentry_t *dentry, int mode) -+#endif -+{ -+ int err = 0; -+ inode_t *dir; -+ dentry_t *hidden_sto_dentry; -+ dentry_t *hidden_sto_dir_dentry; -+ -+ if(exists_in_storage(dentry)) { -+ printk(KERN_CRIT "mini_fo: create_sto_file: wrong type or state.\n"); -+ err = -EINVAL; -+ goto out; -+ } -+ err = get_neg_sto_dentry(dentry); -+ -+ if (err) { -+ printk(KERN_CRIT "mini_fo: create_sto_file: ERROR getting neg. sto dentry.\n"); -+ goto out; -+ } -+ -+ dir = dentry->d_parent->d_inode; -+ hidden_sto_dentry = dtohd2(dentry); -+ -+ /* lock parent */ -+ hidden_sto_dir_dentry = dget(hidden_sto_dentry->d_parent); -+ -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,16) -+ mutex_lock(&hidden_sto_dir_dentry->d_inode->i_mutex); -+#else -+ down(&hidden_sto_dir_dentry->d_inode->i_sem); -+#endif -+ -+ err = PTR_ERR(hidden_sto_dir_dentry); -+ if (IS_ERR(hidden_sto_dir_dentry)) -+ goto out; -+ -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,0) -+ err = vfs_create(hidden_sto_dir_dentry->d_inode, -+ hidden_sto_dentry, -+ mode, nd); -+#else -+ err = vfs_create(hidden_sto_dir_dentry->d_inode, -+ hidden_sto_dentry, -+ mode); -+#endif -+ if(err) { -+ printk(KERN_CRIT "mini_fo: create_sto_file: ERROR creating sto file.\n"); -+ goto out_lock; -+ } -+ -+ if(!dtohd2(dentry)->d_inode) { -+ printk(KERN_CRIT "mini_fo: create_sto_file: ERROR creating sto file [2].\n"); -+ err = -EINVAL; -+ goto out_lock; -+ } -+ -+ /* interpose the new inode */ -+ if(dtost(dentry) == DELETED) { -+ dtost(dentry) = DEL_REWRITTEN; -+ err = mini_fo_tri_interpose(NULL, hidden_sto_dentry, dentry, dir->i_sb, 0); -+ if(err) -+ goto out_lock; -+ } -+ else if(dtost(dentry) == NON_EXISTANT) { -+ dtost(dentry) = CREATED; -+ err = mini_fo_tri_interpose(dtohd(dentry), hidden_sto_dentry, dentry, dir->i_sb, 0); -+ if(err) -+ goto out_lock; -+ } -+ else if(dtost(dentry) == UNMODIFIED) { -+ dtost(dentry) = MODIFIED; -+ /* interpose on new inode */ -+ if(itohi2(dentry->d_inode) != NULL) { -+ printk(KERN_CRIT "mini_fo: create_sto_file: invalid inode detected.\n"); -+ err = -EINVAL; -+ goto out_lock; -+ } -+ itohi2(dentry->d_inode) = igrab(dtohd2(dentry)->d_inode); -+ } -+ fist_copy_attr_timesizes(dentry->d_parent->d_inode, -+ hidden_sto_dir_dentry->d_inode); -+ -+ out_lock: -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,16) -+ mutex_unlock(&hidden_sto_dir_dentry->d_inode->i_mutex); -+#else -+ up(&hidden_sto_dir_dentry->d_inode->i_sem); -+#endif -+ dput(hidden_sto_dir_dentry); -+ out: -+ return err; -+} -+ -+/* create the sto dir, setup states */ -+int create_sto_dir(dentry_t *dentry, int mode) -+{ -+ int err = 0; -+ inode_t *dir; -+ dentry_t *hidden_sto_dentry; -+ dentry_t *hidden_sto_dir_dentry; -+ -+ /* had to take the "!S_ISDIR(mode))" check out, because it failed */ -+ if(exists_in_storage(dentry)) { -+ printk(KERN_CRIT "mini_fo: create_sto_dir: wrong type or state.\\ -+n"); -+ err = -EINVAL; -+ goto out; -+ } -+ -+ err = get_neg_sto_dentry(dentry); -+ if(err) { -+ err = -EINVAL; -+ goto out; -+ } -+ -+ dir = dentry->d_parent->d_inode; -+ hidden_sto_dentry = dtohd2(dentry); -+ -+ /* was: hidden_sto_dir_dentry = lock_parent(hidden_sto_dentry); */ -+ hidden_sto_dir_dentry = dget(hidden_sto_dentry->d_parent); -+ -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,16) -+ mutex_lock(&hidden_sto_dir_dentry->d_inode->i_mutex); -+#else -+ down(&hidden_sto_dir_dentry->d_inode->i_sem); -+#endif -+ -+ err = PTR_ERR(hidden_sto_dir_dentry); -+ if (IS_ERR(hidden_sto_dir_dentry)) -+ goto out; -+ -+ err = vfs_mkdir(hidden_sto_dir_dentry->d_inode, -+ hidden_sto_dentry, -+ mode); -+ if(err) { -+ printk(KERN_CRIT "mini_fo: create_sto_dir: ERROR creating sto dir.\n"); -+ goto out_lock; -+ } -+ -+ if(!dtohd2(dentry)->d_inode) { -+ printk(KERN_CRIT "mini_fo: create_sto_dir: ERROR creating sto dir [2].\n"); -+ err = -EINVAL; -+ goto out_lock; -+ } -+ -+ /* interpose the new inode */ -+ if(dtost(dentry) == DELETED) { -+ dtost(dentry) = DEL_REWRITTEN; -+ err = mini_fo_tri_interpose(NULL, hidden_sto_dentry, dentry, dir->i_sb, 0); -+ if(err) -+ goto out_lock; -+ } -+ else if(dtopd(dentry)->state == NON_EXISTANT) { -+ dtopd(dentry)->state = CREATED; -+ err = mini_fo_tri_interpose(dtohd(dentry), hidden_sto_dentry, dentry, dir->i_sb, 0); -+ if(err) -+ goto out_lock; -+ } -+ else if(dtopd(dentry)->state == UNMODIFIED) { -+ dtopd(dentry)->state = MODIFIED; -+ /* interpose on new inode */ -+ if(itohi2(dentry->d_inode) != NULL) { -+ printk(KERN_CRIT "mini_fo: create_sto_dir: ERROR, invalid inode detected.\n"); -+ err = -EINVAL; -+ goto out_lock; -+ } -+ itohi2(dentry->d_inode) = igrab(dtohd2(dentry)->d_inode); -+ } -+ -+ fist_copy_attr_timesizes(dir, hidden_sto_dir_dentry->d_inode); -+ -+ /* initalize the wol list */ -+ itopd(dentry->d_inode)->deleted_list_size = -1; -+ itopd(dentry->d_inode)->renamed_list_size = -1; -+ meta_build_lists(dentry); -+ -+ -+ out_lock: -+ /* was: unlock_dir(hidden_sto_dir_dentry); */ -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,16) -+ mutex_unlock(&hidden_sto_dir_dentry->d_inode->i_mutex); -+#else -+ up(&hidden_sto_dir_dentry->d_inode->i_sem); -+#endif -+ dput(hidden_sto_dir_dentry); -+ out: -+ return err; -+} -+ -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,0) -+int create_sto_nod(dentry_t *dentry, int mode, dev_t dev) -+#else -+int create_sto_nod(dentry_t *dentry, int mode, int dev) -+#endif -+{ -+ int err = 0; -+ inode_t *dir; -+ dentry_t *hidden_sto_dentry; -+ dentry_t *hidden_sto_dir_dentry; -+ -+ if(exists_in_storage(dentry)) { -+ err = -EEXIST; -+ goto out; -+ } -+ err = get_neg_sto_dentry(dentry); -+ -+ if (err) { -+ printk(KERN_CRIT "mini_fo: create_sto_nod: ERROR getting neg. sto dentry.\n"); -+ goto out; -+ } -+ -+ dir = dentry->d_parent->d_inode; -+ hidden_sto_dentry = dtohd2(dentry); -+ -+ /* lock parent */ -+ hidden_sto_dir_dentry = dget(hidden_sto_dentry->d_parent); -+ -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,16) -+ mutex_lock(&hidden_sto_dir_dentry->d_inode->i_mutex); -+#else -+ down(&hidden_sto_dir_dentry->d_inode->i_sem); -+#endif -+ -+ err = PTR_ERR(hidden_sto_dir_dentry); -+ if (IS_ERR(hidden_sto_dir_dentry)) -+ goto out; -+ -+ err = vfs_mknod(hidden_sto_dir_dentry->d_inode, hidden_sto_dentry, mode, dev); -+ if(err) -+ goto out_lock; -+ -+ if(!dtohd2(dentry)->d_inode) { -+ printk(KERN_CRIT "mini_fo: create_sto_nod: creating storage inode failed [1].\n"); -+ err = -EINVAL; /* return something indicating failure */ -+ goto out_lock; -+ } -+ -+ /* interpose the new inode */ -+ if(dtost(dentry) == DELETED) { -+ dtost(dentry) = DEL_REWRITTEN; -+ err = mini_fo_tri_interpose(NULL, hidden_sto_dentry, dentry, dir->i_sb, 0); -+ if(err) -+ goto out_lock; -+ } -+ else if(dtost(dentry) == NON_EXISTANT) { -+ dtost(dentry) = CREATED; -+ err = mini_fo_tri_interpose(dtohd(dentry), hidden_sto_dentry, dentry, dir->i_sb, 0); -+ if(err) -+ goto out_lock; -+ } -+ else if(dtost(dentry) == UNMODIFIED) { -+ dtost(dentry) = MODIFIED; -+ /* interpose on new inode */ -+ if(itohi2(dentry->d_inode) != NULL) { -+ printk(KERN_CRIT "mini_fo: create_sto_nod: error, invalid inode detected.\n"); -+ err = -EINVAL; -+ goto out_lock; -+ } -+ itohi2(dentry->d_inode) = igrab(dtohd2(dentry)->d_inode); -+ } -+ -+ fist_copy_attr_timesizes(dir, hidden_sto_dir_dentry->d_inode); -+ -+ out_lock: -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,16) -+ mutex_unlock(&hidden_sto_dir_dentry->d_inode->i_mutex); -+#else -+ up(&hidden_sto_dir_dentry->d_inode->i_sem); -+#endif -+ dput(hidden_sto_dir_dentry); -+ out: -+ return err; -+} -+ -+ -+/* unimplemented (and possibly not usefull): -+ -+ nondir-del_to_del_rew -+ nondir-non_exist_to_creat -+ -+ dir-unmod_to_del -+ dir-mod_to_del -+ dir-creat_to_del -+ dir-del_rew_to_del -+ dir-del_to_del_rew -+ dir-non_exist_to_creat -+*/ -+ -+ -+/* bring a file of any type from state UNMODIFIED to MODIFIED */ -+int nondir_unmod_to_mod(dentry_t *dentry, int cp_flag) -+{ -+ int err = 0; -+ struct vfsmount *tgt_mnt; -+ struct vfsmount *src_mnt; -+ dentry_t *tgt_dentry; -+ dentry_t *src_dentry; -+ dentry_t *hidden_sto_dentry; -+ dentry_t *hidden_sto_dir_dentry; -+ -+ check_mini_fo_dentry(dentry); -+ -+ if((dtost(dentry) != UNMODIFIED) || -+ S_ISDIR(dentry->d_inode->i_mode)) { -+ printk(KERN_CRIT "mini_fo: nondir_unmod_to_mod: \ -+ wrong type or state.\n"); -+ err = -1; -+ goto out; -+ } -+ err = get_neg_sto_dentry(dentry); -+ -+ if (err) { -+ printk(KERN_CRIT "mini_fo: nondir_unmod_to_mod: \ -+ ERROR getting neg. sto dentry.\n"); -+ goto out; -+ } -+ -+ /* create sto file */ -+ hidden_sto_dentry = dtohd2(dentry); -+ -+ /* lock parent */ -+ hidden_sto_dir_dentry = dget(hidden_sto_dentry->d_parent); -+ -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,16) -+ mutex_lock(&hidden_sto_dir_dentry->d_inode->i_mutex); -+#else -+ down(&hidden_sto_dir_dentry->d_inode->i_sem); -+#endif -+ -+ err = PTR_ERR(hidden_sto_dir_dentry); -+ if (IS_ERR(hidden_sto_dir_dentry)) -+ goto out; -+ -+ /* handle different types of nondirs */ -+ if(S_ISCHR(dentry->d_inode->i_mode) || -+ S_ISBLK(dentry->d_inode->i_mode)) { -+ err = vfs_mknod(hidden_sto_dir_dentry->d_inode, -+ hidden_sto_dentry, -+ dtohd(dentry)->d_inode->i_mode, -+ dtohd(dentry)->d_inode->i_rdev); -+ } -+ -+ else if(S_ISREG(dentry->d_inode->i_mode)) { -+ -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,0) -+ err = vfs_create(hidden_sto_dir_dentry->d_inode, -+ hidden_sto_dentry, -+ dtohd(dentry)->d_inode->i_mode, NULL); -+#else -+ err = vfs_create(hidden_sto_dir_dentry->d_inode, -+ hidden_sto_dentry, -+ dtohd(dentry)->d_inode->i_mode); -+#endif -+ } -+ if(err) { -+ printk(KERN_CRIT "mini_fo: nondir_unmod_to_mod: \ -+ ERROR creating sto file.\n"); -+ goto out_lock; -+ } -+ -+ /* interpose on new inode */ -+ if(itohi2(dentry->d_inode) != NULL) { -+ printk(KERN_CRIT "mini_fo: nondir_unmod_to_mod: \ -+ ERROR, invalid inode detected.\n"); -+ err = -EINVAL; -+ goto out_lock; -+ } -+ -+ itohi2(dentry->d_inode) = igrab(dtohd2(dentry)->d_inode); -+ -+ fist_copy_attr_timesizes(dentry->d_parent->d_inode, -+ hidden_sto_dir_dentry->d_inode); -+ dtost(dentry) = MODIFIED; -+ -+ /* copy contents if regular file and cp_flag = 1 */ -+ if((cp_flag == 1) && S_ISREG(dentry->d_inode->i_mode)) { -+ -+ /* unlock first */ -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,16) -+ mutex_unlock(&hidden_sto_dir_dentry->d_inode->i_mutex); -+#else -+ up(&hidden_sto_dir_dentry->d_inode->i_sem); -+#endif -+ -+ dput(hidden_sto_dir_dentry); -+ -+ tgt_dentry = dtohd2(dentry); -+ tgt_mnt = stopd(dentry->d_inode->i_sb)->hidden_mnt2; -+ src_dentry = dtohd(dentry); -+ src_mnt = stopd(dentry->d_inode->i_sb)->hidden_mnt; -+ -+ err = mini_fo_cp_cont(tgt_dentry, tgt_mnt, -+ src_dentry, src_mnt); -+ if(err) { -+ printk(KERN_CRIT "mini_fo: nondir_unmod_to_mod: \ -+ ERROR copying contents.\n"); -+ } -+ goto out; -+ } -+ -+ out_lock: -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,16) -+ mutex_unlock(&hidden_sto_dir_dentry->d_inode->i_mutex); -+#else -+ up(&hidden_sto_dir_dentry->d_inode->i_sem); -+#endif -+ dput(hidden_sto_dir_dentry); -+ out: -+ return err; -+} -+ -+/* this function is currently identical to nondir_creat_to_del */ -+int nondir_del_rew_to_del(dentry_t *dentry) -+{ -+ return nondir_creat_to_del(dentry); -+} -+ -+int nondir_creat_to_del(dentry_t *dentry) -+{ -+ int err = 0; -+ -+ inode_t *hidden_sto_dir_inode; -+ dentry_t *hidden_sto_dir_dentry; -+ dentry_t *hidden_sto_dentry; -+ -+ check_mini_fo_dentry(dentry); -+ -+ /* for now this function serves for both state DEL_REWRITTEN and -+ * CREATED */ -+ if(!(dtost(dentry) == CREATED || (dtost(dentry) == DEL_REWRITTEN)) || -+ S_ISDIR(dentry->d_inode->i_mode)) { -+ printk(KERN_CRIT "mini_fo: nondir_mod_to_del/del_rew_to_del: \ -+ wrong type or state.\n"); -+ err = -1; -+ goto out; -+ } -+ -+ hidden_sto_dir_inode = itohi2(dentry->d_parent->d_inode); -+ hidden_sto_dentry = dtohd2(dentry); -+ -+ /* was: hidden_sto_dir_dentry = lock_parent(hidden_sto_dentry);*/ -+ hidden_sto_dir_dentry = dget(hidden_sto_dentry->d_parent); -+ -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,16) -+ mutex_lock(&hidden_sto_dir_dentry->d_inode->i_mutex); -+#else -+ down(&hidden_sto_dir_dentry->d_inode->i_sem); -+#endif -+ -+ /* avoid destroying the hidden inode if the file is in use */ -+ dget(hidden_sto_dentry); -+ err = vfs_unlink(hidden_sto_dir_inode, hidden_sto_dentry); -+ dput(hidden_sto_dentry); -+ if(!err) -+ d_delete(hidden_sto_dentry); -+ -+ /* propagate number of hard-links */ -+ dentry->d_inode->i_nlink = itohi2(dentry->d_inode)->i_nlink; -+ -+ dtost(dentry) = NON_EXISTANT; -+ -+ /* was: unlock_dir(hidden_sto_dir_dentry); */ -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,16) -+ mutex_unlock(&hidden_sto_dir_dentry->d_inode->i_mutex); -+#else -+ up(&hidden_sto_dir_dentry->d_inode->i_sem); -+#endif -+ dput(hidden_sto_dir_dentry); -+ -+ out: -+ return err; -+} -+ -+int nondir_mod_to_del(dentry_t *dentry) -+{ -+ int err; -+ dentry_t *hidden_sto_dentry; -+ inode_t *hidden_sto_dir_inode; -+ dentry_t *hidden_sto_dir_dentry; -+ -+ check_mini_fo_dentry(dentry); -+ -+ if(dtost(dentry) != MODIFIED || -+ S_ISDIR(dentry->d_inode->i_mode)) { -+ printk(KERN_CRIT "mini_fo: nondir_mod_to_del: \ -+ wrong type or state.\n"); -+ err = -1; -+ goto out; -+ } -+ -+ hidden_sto_dir_inode = itohi2(dentry->d_parent->d_inode); -+ hidden_sto_dentry = dtohd2(dentry); -+ -+ /* was hidden_sto_dir_dentry = lock_parent(hidden_sto_dentry); */ -+ hidden_sto_dir_dentry = dget(hidden_sto_dentry->d_parent); -+ -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,16) -+ mutex_lock(&hidden_sto_dir_dentry->d_inode->i_mutex); -+#else -+ down(&hidden_sto_dir_dentry->d_inode->i_sem); -+#endif -+ -+ /* avoid destroying the hidden inode if the file is in use */ -+ dget(hidden_sto_dentry); -+ err = vfs_unlink(hidden_sto_dir_inode, hidden_sto_dentry); -+ dput(hidden_sto_dentry); -+ if(!err) -+ d_delete(hidden_sto_dentry); -+ -+ /* propagate number of hard-links */ -+ dentry->d_inode->i_nlink = itohi2(dentry->d_inode)->i_nlink; -+ -+ /* dput base dentry, this will relase the inode and free the -+ * dentry, as we will never need it again. */ -+ dput(dtohd(dentry)); -+ dtohd(dentry) = NULL; -+ dtost(dentry) = DELETED; -+ -+ /* add deleted file to META-file */ -+ meta_add_d_entry(dentry->d_parent, -+ dentry->d_name.name, -+ dentry->d_name.len); -+ -+ /* was: unlock_dir(hidden_sto_dir_dentry); */ -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,16) -+ mutex_unlock(&hidden_sto_dir_dentry->d_inode->i_mutex); -+#else -+ up(&hidden_sto_dir_dentry->d_inode->i_sem); -+#endif -+ dput(hidden_sto_dir_dentry); -+ -+ out: -+ return err; -+} -+ -+int nondir_unmod_to_del(dentry_t *dentry) -+{ -+ int err = 0; -+ -+ check_mini_fo_dentry(dentry); -+ -+ if(dtost(dentry) != UNMODIFIED || -+ S_ISDIR(dentry->d_inode->i_mode)) { -+ printk(KERN_CRIT "mini_fo: nondir_unmod_to_del: \ -+ wrong type or state.\n"); -+ err = -1; -+ goto out; -+ } -+ -+ /* next we have to get a negative dentry for the storage file */ -+ err = get_neg_sto_dentry(dentry); -+ -+ if(err) -+ goto out; -+ -+ /* add deleted file to META lists */ -+ err = meta_add_d_entry(dentry->d_parent, -+ dentry->d_name.name, -+ dentry->d_name.len); -+ -+ if(err) -+ goto out; -+ -+ /* dput base dentry, this will relase the inode and free the -+ * dentry, as we will never need it again. */ -+ dput(dtohd(dentry)); -+ dtohd(dentry) = NULL; -+ dtost(dentry) = DELETED; -+ -+ out: -+ return err; -+} -+ -+/* bring a dir from state UNMODIFIED to MODIFIED */ -+int dir_unmod_to_mod(dentry_t *dentry) -+{ -+ int err; -+ -+ check_mini_fo_dentry(dentry); -+ -+ if(dtost(dentry) != UNMODIFIED || -+ !S_ISDIR(dentry->d_inode->i_mode)) { -+ printk(KERN_CRIT "mini_fo: dir_unmod_to_mod: \ -+ wrong type or state.\n"); -+ err = -1; -+ goto out; -+ } -+ -+ /* this creates our dir incl. sto. structure */ -+ err = build_sto_structure(dentry->d_parent, dentry); -+ if(err) { -+ printk(KERN_CRIT "mini_fo: dir_unmod_to_mod: \ -+ build_sto_structure failed.\n"); -+ goto out; -+ } -+ out: -+ return err; -+} -+ ---- /dev/null -+++ b/fs/mini_fo/super.c -@@ -0,0 +1,281 @@ -+/* -+ * Copyright (c) 1997-2003 Erez Zadok -+ * Copyright (c) 2001-2003 Stony Brook University -+ * -+ * For specific licensing information, see the COPYING file distributed with -+ * this package, or get one from ftp://ftp.filesystems.org/pub/fist/COPYING. -+ * -+ * This Copyright notice must be kept intact and distributed with all -+ * fistgen sources INCLUDING sources generated by fistgen. -+ */ -+/* -+ * Copyright (C) 2004, 2005 Markus Klotzbuecher <mk@creamnet.de> -+ * -+ * This program is free software; you can redistribute it and/or -+ * modify it under the terms of the GNU General Public License -+ * as published by the Free Software Foundation; either version -+ * 2 of the License, or (at your option) any later version. -+ */ -+ -+/* -+ * $Id$ -+ */ -+ -+#ifdef HAVE_CONFIG_H -+# include <config.h> -+#endif -+ -+#include "fist.h" -+#include "mini_fo.h" -+ -+ -+STATIC void -+mini_fo_read_inode(inode_t *inode) -+{ -+ static struct address_space_operations mini_fo_empty_aops; -+ -+ __itopd(inode) = kmalloc(sizeof(struct mini_fo_inode_info), GFP_KERNEL); -+ if (!itopd(inode)) { -+ printk("<0>%s:%s:%d: No kernel memory!\n", __FILE__, __FUNCTION__, __LINE__); -+ ASSERT(NULL); -+ } -+ itohi(inode) = NULL; -+ itohi2(inode) = NULL; -+ -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,0) -+ inode->i_version++; -+#else -+ inode->i_version = ++event; /* increment inode version */ -+#endif -+ inode->i_op = &mini_fo_main_iops; -+ inode->i_fop = &mini_fo_main_fops; -+#if 0 -+ /* -+ * XXX: To export a file system via NFS, it has to have the -+ * FS_REQUIRES_DEV flag, so turn it on. But should we inherit it from -+ * the lower file system, or can we allow our file system to be exported -+ * even if the lower one cannot be natively exported. -+ */ -+ inode->i_sb->s_type->fs_flags |= FS_REQUIRES_DEV; -+ /* -+ * OK, the above was a hack, which is now turned off because it may -+ * cause a panic/oops on some systems. The correct way to export a -+ * "nodev" filesystem is via using nfs-utils > 1.0 and the "fsid=" export -+ * parameter, which requires 2.4.20 or later. -+ */ -+#endif -+ /* I don't think ->a_ops is ever allowed to be NULL */ -+ inode->i_mapping->a_ops = &mini_fo_empty_aops; -+} -+ -+ -+#if defined(FIST_DEBUG) || defined(FIST_FILTER_SCA) -+/* -+ * No need to call write_inode() on the lower inode, as it -+ * will have been marked 'dirty' anyway. But we might need -+ * to write some of our own stuff to disk. -+ */ -+STATIC void -+mini_fo_write_inode(inode_t *inode, int sync) -+{ -+ print_entry_location(); -+ print_exit_location(); -+} -+#endif /* defined(FIST_DEBUG) || defined(FIST_FILTER_SCA) */ -+ -+ -+STATIC void -+mini_fo_put_inode(inode_t *inode) -+{ -+ /* -+ * This is really funky stuff: -+ * Basically, if i_count == 1, iput will then decrement it and this inode will be destroyed. -+ * It is currently holding a reference to the hidden inode. -+ * Therefore, it needs to release that reference by calling iput on the hidden inode. -+ * iput() _will_ do it for us (by calling our clear_inode), but _only_ if i_nlink == 0. -+ * The problem is, NFS keeps i_nlink == 1 for silly_rename'd files. -+ * So we must for our i_nlink to 0 here to trick iput() into calling our clear_inode. -+ */ -+ if (atomic_read(&inode->i_count) == 1) -+ inode->i_nlink = 0; -+} -+ -+ -+#if defined(FIST_DEBUG) || defined(FIST_FILTER_SCA) -+/* -+ * we now define delete_inode, because there are two VFS paths that may -+ * destroy an inode: one of them calls clear inode before doing everything -+ * else that's needed, and the other is fine. This way we truncate the inode -+ * size (and its pages) and then clear our own inode, which will do an iput -+ * on our and the lower inode. -+ */ -+STATIC void -+mini_fo_delete_inode(inode_t *inode) -+{ -+ print_entry_location(); -+ -+ fist_checkinode(inode, "mini_fo_delete_inode IN"); -+ inode->i_size = 0; /* every f/s seems to do that */ -+ clear_inode(inode); -+ -+ print_exit_location(); -+} -+#endif /* defined(FIST_DEBUG) || defined(FIST_FILTER_SCA) */ -+ -+ -+/* final actions when unmounting a file system */ -+STATIC void -+mini_fo_put_super(super_block_t *sb) -+{ -+ if (stopd(sb)) { -+ mntput(stopd(sb)->hidden_mnt); -+ mntput(stopd(sb)->hidden_mnt2); -+ -+ /* mk: no! dput(stopd(sb)->base_dir_dentry); -+ dput(stopd(sb)->storage_dir_dentry); */ -+ -+ kfree(stopd(sb)); -+ __stopd(sb) = NULL; -+ } -+} -+ -+ -+#ifdef NOT_NEEDED -+/* -+ * This is called in do_umount before put_super. -+ * The superblock lock is not held yet. -+ * We probably do not need to define this or call write_super -+ * on the hidden_sb, because sync_supers() will get to hidden_sb -+ * sooner or later. But it is also called from file_fsync()... -+ */ -+STATIC void -+mini_fo_write_super(super_block_t *sb) -+{ -+ return; -+} -+#endif /* NOT_NEEDED */ -+ -+ -+STATIC int -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,18) -+mini_fo_statfs(struct dentry *d, struct kstatfs *buf) -+#elif LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,0) -+mini_fo_statfs(super_block_t *sb, struct kstatfs *buf) -+#else -+mini_fo_statfs(super_block_t *sb, struct statfs *buf) -+#endif -+{ -+ int err = 0; -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,18) -+ struct dentry *hidden_d; -+ -+ hidden_d = dtohd(d); -+ err = vfs_statfs(hidden_d, buf); -+#else -+ super_block_t *hidden_sb; -+ -+ hidden_sb = stohs(sb); -+ err = vfs_statfs(hidden_sb, buf); -+#endif -+ -+ return err; -+} -+ -+ -+/* -+ * XXX: not implemented. This is not allowed yet. -+ * Should we call this on the hidden_sb? Probably not. -+ */ -+STATIC int -+mini_fo_remount_fs(super_block_t *sb, int *flags, char *data) -+{ -+ //printk(KERN_CRIT "mini_fo_remount_fs: WARNING, this function is umimplemented.\n"); -+ return -ENOSYS; -+} -+ -+ -+/* -+ * Called by iput() when the inode reference count reached zero -+ * and the inode is not hashed anywhere. Used to clear anything -+ * that needs to be, before the inode is completely destroyed and put -+ * on the inode free list. -+ */ -+STATIC void -+mini_fo_clear_inode(inode_t *inode) -+{ -+ /* -+ * Decrement a reference to a hidden_inode, which was incremented -+ * by our read_inode when it was created initially. -+ */ -+ -+ /* release the wol_list */ -+ if(S_ISDIR(inode->i_mode)) { -+ __meta_put_lists(inode); -+ } -+ -+ /* mk: fan out fun */ -+ if(itohi(inode)) -+ iput(itohi(inode)); -+ if(itohi2(inode)) -+ iput(itohi2(inode)); -+ -+ // XXX: why this assertion fails? -+ // because it doesn't like us -+ // ASSERT((inode->i_state & I_DIRTY) == 0); -+ kfree(itopd(inode)); -+ __itopd(inode) = NULL; -+} -+ -+ -+/* -+ * Called in do_umount() if the MNT_FORCE flag was used and this -+ * function is defined. See comment in linux/fs/super.c:do_umount(). -+ * Used only in nfs, to kill any pending RPC tasks, so that subsequent -+ * code can actually succeed and won't leave tasks that need handling. -+ * -+ * PS. I wonder if this is somehow useful to undo damage that was -+ * left in the kernel after a user level file server (such as amd) -+ * dies. -+ */ -+STATIC void -+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,18) -+mini_fo_umount_begin(struct vfsmount *mnt, int flags) -+{ -+ struct vfsmount *hidden_mnt; -+ -+ hidden_mnt = stopd(mnt->mnt_sb)->hidden_mnt; -+ -+ if (hidden_mnt->mnt_sb->s_op->umount_begin) -+ hidden_mnt->mnt_sb->s_op->umount_begin(hidden_mnt, flags); -+ -+} -+#else -+mini_fo_umount_begin(super_block_t *sb) -+{ -+ super_block_t *hidden_sb; -+ -+ hidden_sb = stohs(sb); -+ -+ if (hidden_sb->s_op->umount_begin) -+ hidden_sb->s_op->umount_begin(hidden_sb); -+ -+} -+#endif -+ -+ -+struct super_operations mini_fo_sops = -+{ -+ read_inode: mini_fo_read_inode, -+#if defined(FIST_DEBUG) || defined(FIST_FILTER_SCA) -+ write_inode: mini_fo_write_inode, -+#endif /* defined(FIST_DEBUG) || defined(FIST_FILTER_SCA) */ -+ put_inode: mini_fo_put_inode, -+#if defined(FIST_DEBUG) || defined(FIST_FILTER_SCA) -+ delete_inode: mini_fo_delete_inode, -+#endif /* defined(FIST_DEBUG) || defined(FIST_FILTER_SCA) */ -+ put_super: mini_fo_put_super, -+ statfs: mini_fo_statfs, -+ remount_fs: mini_fo_remount_fs, -+ clear_inode: mini_fo_clear_inode, -+ umount_begin: mini_fo_umount_begin, -+}; diff --git a/target/linux/generic-2.6/patches-2.6.22/210-mac80211_include_wireless_dev.patch b/target/linux/generic-2.6/patches-2.6.22/210-mac80211_include_wireless_dev.patch deleted file mode 100644 index dd2d1c5..0000000 --- a/target/linux/generic-2.6/patches-2.6.22/210-mac80211_include_wireless_dev.patch +++ /dev/null @@ -1,806 +0,0 @@ ---- a/include/linux/ieee80211.h -+++ b/include/linux/ieee80211.h -@@ -106,6 +106,75 @@ - } __attribute__ ((packed)); - - -+struct ieee80211_ht_capability { -+ __le16 capabilities_info; -+ u8 mac_ht_params_info; -+ u8 supported_mcs_set[16]; -+ __le16 extended_ht_capability_info; -+ __le32 tx_BF_capability_info; -+ u8 antenna_selection_info; -+}__attribute__ ((packed)); -+ -+struct ieee80211_ht_additional_info { -+ u8 control_chan; -+ u8 ht_param; -+ __le16 operation_mode; -+ __le16 stbc_param; -+ u8 basic_set[16]; -+}__attribute__ ((packed)); -+ -+ -+#define IEEE80211_TSINFO_TYPE(a) ((a.byte1 & 0x01) >> 0) -+#define IEEE80211_TSINFO_TSID(a) ((a.byte1 & 0x1E) >> 1) -+#define IEEE80211_TSINFO_DIR(a) ((a.byte1 & 0x60) >> 5) -+#define IEEE80211_TSINFO_POLICY(a) ((a.byte1 & 0x80) >> 7 + \ -+ (a.byte2 & 0x01) << 1) -+#define IEEE80211_TSINFO_AGG(a) ((a.byte2 & 0x02) >> 1) -+#define IEEE80211_TSINFO_APSD(a) ((a.byte2 & 0x04) >> 2) -+#define IEEE80211_TSINFO_UP(a) ((a.byte2 & 0x38) >> 3) -+#define IEEE80211_TSINFO_ACK(a) ((a.byte2 & 0xC0) >> 6) -+#define IEEE80211_TSINFO_SCHEDULE(a) ((a.byte3 & 0x01) >> 0) -+ -+#define IEEE80211_SET_TSINFO_TYPE(i, d) (i.byte1 |= (d << 0) & 0x01) -+#define IEEE80211_SET_TSINFO_TSID(i, d) (i.byte1 |= (d << 1) & 0x1E) -+#define IEEE80211_SET_TSINFO_DIR(i, d) (i.byte1 |= (d << 5) & 0x60) -+#define IEEE80211_SET_TSINFO_POLICY(i, d) \ -+do { \ -+ i.byte1 |= (d & 0x01) << 7; \ -+ i.byte2 |= (d & 0x02) >> 1; \ -+} while(0) -+#define IEEE80211_SET_TSINFO_AGG(i, d) (i.byte2 |= (d << 1) & 0x02) -+#define IEEE80211_SET_TSINFO_APSD(i, d) (i.byte2 |= (d << 2) & 0x04) -+#define IEEE80211_SET_TSINFO_UP(i, d) (i.byte2 |= (d << 3) & 0x38) -+#define IEEE80211_SET_TSINFO_ACK(i, d) (i.byte2 |= (d << 6) & 0xC0) -+#define IEEE80211_SET_TSINFO_SCHEDULE(i, d) (i.byte3 |= (d << 0) & 0x01) -+ -+struct ieee80211_ts_info { -+ u8 byte1; -+ u8 byte2; -+ u8 byte3; -+} __attribute__ ((packed)); -+ -+struct ieee80211_elem_tspec { -+ struct ieee80211_ts_info ts_info; -+ __le16 nominal_msdu_size; -+ __le16 max_msdu_size; -+ __le32 min_service_interval; -+ __le32 max_service_interval; -+ __le32 inactivity_interval; -+ __le32 suspension_interval; -+ __le32 service_start_time; -+ __le32 min_data_rate; -+ __le32 mean_data_rate; -+ __le32 peak_data_rate; -+ __le32 burst_size; -+ __le32 delay_bound; -+ __le32 min_phy_rate; -+ __le16 surplus_band_allow; -+ __le16 medium_time; -+} __attribute__ ((packed)); -+ -+ - struct ieee80211_mgmt { - __le16 frame_control; - __le16 duration; -@@ -173,9 +242,51 @@ - struct { - u8 action_code; - u8 dialog_token; -+ u8 variable[0]; -+ } __attribute__ ((packed)) addts_req; -+ struct { -+ u8 action_code; -+ u8 dialog_token; -+ __le16 status_code; -+ u8 variable[0]; -+ } __attribute__ ((packed)) addts_resp; -+ struct { -+ u8 action_code; -+ struct ieee80211_ts_info ts_info; -+ __le16 reason_code; -+ } __attribute__ ((packed)) delts; -+ struct { -+ u8 action_code; -+ u8 dialog_token; - u8 status_code; - u8 variable[0]; - } __attribute__ ((packed)) wme_action; -+ struct { -+ u8 action_code; -+ u8 dest[6]; -+ u8 src[6]; -+ __le16 capab_info; -+ __le16 timeout; -+ /* Followed by Supported Rates and -+ * Extended Supported Rates */ -+ u8 variable[0]; -+ } __attribute__ ((packed)) dls_req; -+ struct { -+ u8 action_code; -+ __le16 status_code; -+ u8 dest[6]; -+ u8 src[6]; -+ /* Followed by Capability Information, -+ * Supported Rates and Extended -+ * Supported Rates */ -+ u8 variable[0]; -+ } __attribute__ ((packed)) dls_resp; -+ struct { -+ u8 action_code; -+ u8 dest[6]; -+ u8 src[6]; -+ __le16 reason_code; -+ } __attribute__ ((packed)) dls_teardown; - struct{ - u8 action_code; - u8 element_id; -@@ -184,6 +295,25 @@ - u8 new_chan; - u8 switch_count; - } __attribute__((packed)) chan_switch; -+ struct{ -+ u8 action_code; -+ u8 dialog_token; -+ __le16 capab; -+ __le16 timeout; -+ __le16 start_seq_num; -+ } __attribute__((packed)) addba_req; -+ struct{ -+ u8 action_code; -+ u8 dialog_token; -+ __le16 status; -+ __le16 capab; -+ __le16 timeout; -+ } __attribute__((packed)) addba_resp; -+ struct{ -+ u8 action_code; -+ __le16 params; -+ __le16 reason_code; -+ }__attribute__((packed)) delba; - } u; - } __attribute__ ((packed)) action; - } u; -@@ -259,6 +389,18 @@ - WLAN_STATUS_UNSUPP_RSN_VERSION = 44, - WLAN_STATUS_INVALID_RSN_IE_CAP = 45, - WLAN_STATUS_CIPHER_SUITE_REJECTED = 46, -+ /* 802.11e */ -+ WLAN_STATUS_UNSPECIFIED_QOS = 32, -+ WLAN_STATUS_ASSOC_DENIED_NOBANDWIDTH = 33, -+ WLAN_STATUS_ASSOC_DENIED_LOWACK = 34, -+ WLAN_STATUS_ASSOC_DENIED_UNSUPP_QOS = 35, -+ WLAN_STATUS_REQUEST_DECLINED = 37, -+ WLAN_STATUS_INVALID_QOS_PARAM = 38, -+ WLAN_STATUS_CHANGE_TSPEC = 39, -+ WLAN_STATUS_WAIT_TS_DELAY = 47, -+ WLAN_STATUS_NO_DIRECT_LINK = 48, -+ WLAN_STATUS_STA_NOT_PRESENT = 49, -+ WLAN_STATUS_STA_NOT_QSTA = 50, - }; - - -@@ -289,9 +431,50 @@ - WLAN_REASON_INVALID_RSN_IE_CAP = 22, - WLAN_REASON_IEEE8021X_FAILED = 23, - WLAN_REASON_CIPHER_SUITE_REJECTED = 24, -+ /* 802.11e */ -+ WLAN_REASON_DISASSOC_UNSPECIFIED_QOS = 32, -+ WLAN_REASON_DISASSOC_QAP_NO_BANDWIDTH = 33, -+ WLAN_REASON_DISASSOC_LOW_ACK = 34, -+ WLAN_REASON_DISASSOC_QAP_EXCEED_TXOP = 35, -+ WLAN_REASON_QSTA_LEAVE_QBSS = 36, -+ WLAN_REASON_QSTA_NOT_USE = 37, -+ WLAN_REASON_QSTA_REQUIRE_SETUP = 38, -+ WLAN_REASON_QSTA_TIMEOUT = 39, -+ WLAN_REASON_QSTA_CIPHER_NOT_SUPP = 45, - }; - - -+/* Category Code */ -+enum ieee80211_category { -+ WLAN_CATEGORY_SPECTRUM_MGMT = 0, -+ WLAN_CATEGORY_QOS = 1, -+ WLAN_CATEGORY_DLS = 2, -+ WLAN_CATEGORY_BACK = 3, -+ WLAN_CATEGORY_WMM = 17, -+}; -+ -+/* QoS Action Code */ -+enum ieee80211_qos_actioncode { -+ WLAN_ACTION_QOS_ADDTS_REQ = 0, -+ WLAN_ACTION_QOS_ADDTS_RESP = 1, -+ WLAN_ACTION_QOS_DELTS = 2, -+ WLAN_ACTION_QOS_SCHEDULE = 3, -+}; -+ -+/* DLS Action Code */ -+enum ieee80211_dls_actioncode { -+ WLAN_ACTION_DLS_REQ = 0, -+ WLAN_ACTION_DLS_RESP = 1, -+ WLAN_ACTION_DLS_TEARDOWN = 2, -+}; -+ -+/* BACK Action Code */ -+enum ieee80211_back_actioncode { -+ WLAN_ACTION_ADDBA_REQ = 0, -+ WLAN_ACTION_ADDBA_RESP = 1, -+ WLAN_ACTION_DELBA = 2, -+}; -+ - /* Information Element IDs */ - enum ieee80211_eid { - WLAN_EID_SSID = 0, -@@ -307,6 +490,15 @@ - WLAN_EID_HP_PARAMS = 8, - WLAN_EID_HP_TABLE = 9, - WLAN_EID_REQUEST = 10, -+ /* 802.11e */ -+ WLAN_EID_QBSS_LOAD = 11, -+ WLAN_EID_EDCA_PARAM_SET = 12, -+ WLAN_EID_TSPEC = 13, -+ WLAN_EID_TCLAS = 14, -+ WLAN_EID_SCHEDULE = 15, -+ WLAN_EID_TS_DELAY = 43, -+ WLAN_EID_TCLAS_PROCESSING = 44, -+ WLAN_EID_QOS_CAPA = 46, - /* 802.11h */ - WLAN_EID_PWR_CONSTRAINT = 32, - WLAN_EID_PWR_CAPABILITY = 33, -@@ -321,6 +513,9 @@ - /* 802.11g */ - WLAN_EID_ERP_INFO = 42, - WLAN_EID_EXT_SUPP_RATES = 50, -+ /* 802.11n */ -+ WLAN_EID_HT_CAPABILITY = 45, -+ WLAN_EID_HT_EXTRA_INFO = 61, - /* 802.11i */ - WLAN_EID_RSN = 48, - WLAN_EID_WPA = 221, -@@ -329,6 +524,9 @@ - WLAN_EID_QOS_PARAMETER = 222 - }; - -+/* 80211n */ -+#define IEEE80211_QOS_CONTROL_A_MSDU_PRESENT 0x0080 -+ - /* cipher suite selectors */ - #define WLAN_CIPHER_SUITE_USE_GROUP 0x000FAC00 - #define WLAN_CIPHER_SUITE_WEP40 0x000FAC01 -@@ -339,4 +537,37 @@ - - #define WLAN_MAX_KEY_LEN 32 - -+enum ieee80211_tsinfo_direction { -+ WLAN_TSINFO_UPLINK = 0, -+ WLAN_TSINFO_DOWNLINK = 1, -+ WLAN_TSINFO_DIRECTLINK = 2, -+ WLAN_TSINFO_BIDIRECTIONAL = 3, -+}; -+ -+enum ieee80211_tsinfo_access { -+ WLAN_TSINFO_EDCA = 1, -+ WLAN_TSINFO_HCCA = 2, -+ WLAN_TSINFO_HEMM = 3, -+}; -+ -+enum ieee80211_tsinfo_psb { -+ WLAN_TSINFO_PSB_LEGACY = 0, -+ WLAN_TSINFO_PSB_APSD = 1, -+}; -+ -+ -+/* WI-FI Alliance OUI Type and Subtype */ -+enum wifi_oui_type { -+ WIFI_OUI_TYPE_WPA = 1, -+ WIFI_OUI_TYPE_WMM = 2, -+ WIFI_OUI_TYPE_WSC = 4, -+ WIFI_OUI_TYPE_PSD = 6, -+}; -+ -+enum wifi_oui_stype_wmm { -+ WIFI_OUI_STYPE_WMM_INFO = 0, -+ WIFI_OUI_STYPE_WMM_PARAM = 1, -+ WIFI_OUI_STYPE_WMM_TSPEC = 2, -+}; -+ - #endif /* IEEE80211_H */ ---- a/include/linux/nl80211.h -+++ b/include/linux/nl80211.h -@@ -7,6 +7,217 @@ - */ - - /** -+ * enum nl80211_commands - supported nl80211 commands -+ * @NL80211_CMD_UNSPEC: unspecified command to catch errors -+ * @NL80211_CMD_RENAME_WIPHY: rename a wiphy, needs -+ * %NL80211_ATTR_WIPHY and %NL80211_ATTR_WIPHY_NAME -+ * @NL80211_CMD_WIPHY_NEWNAME: rename notification -+ * @NL80211_CMD_GET_CMDLIST: TO BE DEFINED PROPERLY. currently the code makes -+ * it depend on the wiphy only but it really should depend on the -+ * interface type too.... -+ * @NL80211_CMD_NEW_CMDLIST: command list result -+ * @NL80211_CMD_ADD_VIRTUAL_INTERFACE: create a virtual interface for the -+ * wiphy identified by an %NL80211_ATTR_WIPHY attribute with the given -+ * %NL80211_ATTR_IFTYPE and %NL80211_ATTR_IFNAME. -+ * @NL80211_CMD_DEL_VIRTUAL_INTERFACE: destroy a virtual interface identified -+ * by %NL80211_ATTR_IFINDEX. -+ * @NL80211_CMD_CHANGE_VIRTUAL_INTERFACE: change type of virtual interface to -+ * the type given by %NL80211_ATTR_IFTYPE, the interface is identified by -+ * %NL80211_ATTR_IFINDEX. -+ * @NL80211_CMD_GET_WIPHYS: request a list of all wiphys present in the system -+ * @NL80211_CMD_NEW_WIPHYS: returned list of all wiphys -+ * @NL80211_CMD_GET_INTERFACES: request a list of all interfaces belonging to -+ * the wiphy identified by %NL80211_ATTR_WIPHY -+ * @NL80211_CMD_NEW_INTERFACES: result for %NL80211_CMD_GET_INTERFACES -+ * @NL80211_CMD_INITIATE_SCAN: initiate a scan with the passed parameters. THe -+ * parameters may contain %NL80211_ATTR_FLAG_SCAN_ACTIVE, -+ * %NL80211_ATTR_PHYMODE and a list of channels in an -+ * %NL80211_ATTR_CHANNEL_LIST attribute (an array of nested attributes) -+ * containing %NL80211_ATTR_CHANNEL, %NL80211_ATTR_PHYMODE, and possibly -+ * %NL80211_ATTR_FLAG_SCAN_ACTIVE. The outer %NL80211_ATTR_FLAG_SCAN_ACTIVE -+ * is ignored when a channel list is present. -+ * @NL80211_CMD_SCAN_RESULT: scan result, contains an array in -+ * %NL80211_ATTR_BSS_LIST. -+ * @NL80211_CMD_ASSOCIATE: associate with the given parameters -+ * (%NL80211_ATTR_SSID is mandatory, %NL80211_ATTR_TIMEOUT_TU, -+ * %NL80211_ATTR_BSSID, %NL80211_ATTR_CHANNEL, %NL80211_ATTR_PHYMODE, -+ * and %NL80211_ATTR_IE may be given) -+ * @NL80211_CMD_ADD_KEY: add a key with given %NL80211_ATTR_KEY_DATA, -+ * %NL80211_ATTR_KEY_ID, %NL80211_ATTR_KEY_TYPE, %NL80211_ATTR_MAC and -+ * %NL80211_ATTR_KEY_CIPHER attributes. -+ * @NL80211_CMD_DEL_KEY: delete a key identified by %NL80211_ATTR_KEY_ID, -+ * %NL80211_ATTR_KEY_TYPE and %NL80211_ATTR_MAC or all keys. -+ * @__NL80211_CMD_AFTER_LAST: internal use -+ */ -+enum nl80211_commands { -+/* don't change the order or add anything inbetween, this is ABI! */ -+ NL80211_CMD_UNSPEC, -+ /* %input: wiphy, wiphy_name */ -+ NL80211_CMD_RENAME_WIPHY, -+ NL80211_CMD_WIPHY_NEWNAME, -+ /* %input: wiphy|ifindex */ -+ NL80211_CMD_GET_CMDLIST, -+ NL80211_CMD_NEW_CMDLIST, -+ /* %input: wiphy, ifname, {iftype} */ -+ NL80211_CMD_ADD_VIRTUAL_INTERFACE, -+ /* %input: wiphy, ifindex */ -+ NL80211_CMD_DEL_VIRTUAL_INTERFACE, -+ /* %input: ifindex, iftype */ -+ NL80211_CMD_CHANGE_VIRTUAL_INTERFACE, -+ /* %input: */ -+ NL80211_CMD_GET_WIPHYS, -+ NL80211_CMD_NEW_WIPHYS, -+ /* %input: wiphy */ -+ NL80211_CMD_GET_INTERFACES, -+ NL80211_CMD_NEW_INTERFACES, -+ NL80211_CMD_INITIATE_SCAN, -+ NL80211_CMD_SCAN_RESULT, -+ NL80211_CMD_GET_ASSOCIATION, -+ NL80211_CMD_ASSOCIATION_CHANGED, -+ NL80211_CMD_ASSOCIATE, -+ NL80211_CMD_DISASSOCIATE, -+ NL80211_CMD_DEAUTH, -+ NL80211_CMD_GET_AUTH_LIST, -+ NL80211_CMD_NEW_AUTH_LIST, -+ NL80211_CMD_AUTHENTICATION_CHANGED, -+ NL80211_CMD_AP_SET_BEACON, -+ NL80211_CMD_AP_ADD_STA, -+ NL80211_CMD_AP_UPDATE_STA, -+ NL80211_CMD_AP_GET_STA_INFO, -+ NL80211_CMD_AP_SET_RATESETS, -+ NL80211_CMD_ADD_KEY, -+ NL80211_CMD_DEL_KEY, -+ -+ /* add commands here */ -+ -+ /* used to define NL80211_CMD_MAX below */ -+ __NL80211_CMD_AFTER_LAST -+}; -+#define NL80211_CMD_MAX (__NL80211_CMD_AFTER_LAST - 1) -+ -+ -+/** -+ * enum nl80211_attrs - nl80211 netlink attributes -+ * @NL80211_ATTR_UNSPEC: unspecified attribute to catch errors -+ * @NL80211_ATTR_IFINDEX: network interface index of the device to operate on -+ * @NL80211_ATTR_IFNAME: network interface name -+ * @NL80211_ATTR_WIPHY: index of wiphy to operate on, cf. -+ * /sys/class/ieee80211/<phyname>/index -+ * @NL80211_ATTR_WIPHY_NAME: wiphy name (used for renaming) -+ * @NL80211_ATTR_CMDS: list of u8's identifying commands a device supports -+ * @NL80211_ATTR_IFTYPE: type of virtual interface, see &enum nl80211_iftype -+ * @NL80211_ATTR_INTERFACE_LIST: interface array, nested netlink attribute -+ * @NL80211_ATTR_WIPHY_LIST: wiphy array, nested netlink attribute -+ * @NL80211_ATTR_BSSID: BSSID (must be 6 bytes) -+ * @NL80211_ATTR_SSID: SSID (1-32 bytes) -+ * @NL80211_ATTR_CHANNEL: channel number -+ * @NL80211_ATTR_PHYMODE: PHY mode, see &enum nl80211_phymode -+ * @NL80211_ATTR_CHANNEL_LIST: netlink nested attribute array containing scan -+ * parameters for channels -+ * @NL80211_ATTR_BSS_LIST: nested attribute containing an array -+ * @NL80211_ATTR_BSSTYPE: BSS type, see &enum nl80211_bsstype -+ * @NL80211_ATTR_BEACON_PERIOD: beacon period -+ * @NL80211_ATTR_DTIM_PERIOD: DTIM period -+ * @NL80211_ATTR_TIMESTAMP: 64-bit timestamp of received beacon/probe response -+ * @NL80211_ATTR_IE: information element(s), maximum length %NL80211_MAX_IE_LEN -+ * @NL80211_ATTR_AUTH_ALGORITHM: authentication algorithm -+ * @NL80211_ATTR_TIMEOUT_TU: timeout in TU (TO BE USED) -+ * @NL80211_ATTR_REASON_CODE: 802.11 reason code -+ * @NL80211_ATTR_ASSOCIATION_ID: association ID (u16, 1-2007) -+ * @NL80211_ATTR_DEAUTHENTICATED: TO BE USED -+ * @NL80211_ATTR_RX_SENSITIVITY: receiver sensitivity in dBm -+ * @NL80211_ATTR_TRANSMIT_POWER: transmit power in mW -+ * @NL80211_ATTR_FRAG_THRESHOLD: fragmentation threshold (bytes) -+ * @NL80211_ATTR_FLAG_SCAN_ACTIVE: netlink flag indiciating active scan -+ * @NL80211_ATTR_KEY_DATA: temporal key data -+ * @NL80211_ATTR_KEY_ID: key ID (u8, 0-3) -+ * @NL80211_ATTR_KEY_TYPE: key type (see &enum nl80211_keytype) -+ * @NL80211_ATTR_MAC: MAC address -+ * @NL80211_ATTR_KEY_CIPHER: key cipher suite (u32) -+ * @__NL80211_ATTR_AFTER_LAST: internal use -+ */ -+enum nl80211_attrs { -+/* don't change the order or add anything inbetween, this is ABI! */ -+ NL80211_ATTR_UNSPEC, -+ /* %type: u32 */ -+ NL80211_ATTR_IFINDEX, -+ /* %type: nulstring */ -+ NL80211_ATTR_IFNAME, -+ /* %type: u32 */ -+ NL80211_ATTR_WIPHY, -+ /* %type: nulstring */ -+ NL80211_ATTR_WIPHY_NAME, -+ NL80211_ATTR_CMDS, -+ /* %type: u32 */ -+ NL80211_ATTR_IFTYPE, -+ NL80211_ATTR_INTERFACE_LIST, -+ NL80211_ATTR_WIPHY_LIST, -+ NL80211_ATTR_BSSID, -+ NL80211_ATTR_SSID, -+ NL80211_ATTR_CHANNEL, -+ NL80211_ATTR_PHYMODE, -+ NL80211_ATTR_CHANNEL_LIST, -+ NL80211_ATTR_BSS_LIST, -+ NL80211_ATTR_BSSTYPE, -+ NL80211_ATTR_BEACON_PERIOD, -+ NL80211_ATTR_DTIM_PERIOD, -+ NL80211_ATTR_TIMESTAMP, -+ NL80211_ATTR_IE, -+ NL80211_ATTR_AUTH_ALGORITHM, -+ NL80211_ATTR_TIMEOUT_TU, -+ NL80211_ATTR_REASON_CODE, -+ NL80211_ATTR_ASSOCIATION_ID, -+ NL80211_ATTR_DEAUTHENTICATED, -+ NL80211_ATTR_RX_SENSITIVITY, -+ NL80211_ATTR_TRANSMIT_POWER, -+ NL80211_ATTR_FRAG_THRESHOLD, -+ NL80211_ATTR_FLAG_SCAN_ACTIVE, -+ -+ NL80211_ATTR_KEY_DATA, -+ NL80211_ATTR_KEY_ID, -+ NL80211_ATTR_KEY_TYPE, -+ NL80211_ATTR_MAC, -+ NL80211_ATTR_KEY_CIPHER, -+ -+ NL80211_ATTR_BEACON_HEAD, -+ NL80211_ATTR_BEACON_TAIL, -+ -+ /* add attributes here, update the policy in nl80211.c */ -+ -+ /* used to define NL80211_ATTR_MAX below */ -+ __NL80211_ATTR_AFTER_LAST, -+}; -+#define NL80211_ATTR_MAX (__NL80211_ATTR_AFTER_LAST - 1) -+ -+/** -+ * enum nl80211_multicast_groups - multicast groups for nl80211 -+ * @NL80211_GROUP_CONFIG: members of this group are notified of -+ * configuration changes -+ */ -+enum nl80211_multicast_groups { -+ /* be notified of configuration changes like wiphy renames */ -+ NL80211_GROUP_CONFIG, -+ -+ /* add groups here */ -+ -+ /* keep last */ -+ __NL80211_GROUP_AFTER_LAST -+}; -+#define NL80211_GROUP_MAX (__NL80211_GROUP_AFTER_LAST - 1) -+ -+/* -+ * maximum length of IE(s) passed in an NL80211_ATTR_IE. -+ * this is an arbitrary limit, 774 means three full-length -+ * IEs would fit... increase if necessary */ -+#define NL80211_MAX_IE_LEN 774 -+ -+/* -+ * maximum number of items in an ATTR_CHANNEL_LIST, -+ * just to avoid too large allocations -+ */ -+#define NL80211_MAX_CHANNEL_LIST_ITEM 200 -+ -+/** - * enum nl80211_iftype - (virtual) interface types - * @NL80211_IFTYPE_UNSPECIFIED: unspecified type, driver decides - * @NL80211_IFTYPE_ADHOC: independent BSS member -@@ -35,4 +246,56 @@ - }; - #define NL80211_IFTYPE_MAX (__NL80211_IFTYPE_AFTER_LAST - 1) - -+/** -+ * enum nl80211_phymode - PHY modes -+ * @NL80211_PHYMODE_A: 5 GHz PHY -+ * @NL80211_PHYMODE_B: 2.4 GHz PHY (B mode) -+ * @NL80211_PHYMODE_G: 2.4 GHz PHY (G, compatible with B) -+ * @__NL80211_PHYMODE_AFTER_LAST: internal use -+ * -+ * These values are used for %NL80211_ATTR_PHYMODE. -+ */ -+enum nl80211_phymode { -+ NL80211_PHYMODE_A, -+ NL80211_PHYMODE_B, -+ NL80211_PHYMODE_G, -+ -+ /* keep last */ -+ __NL80211_PHYMODE_AFTER_LAST -+}; -+#define NL80211_PHYMODE_MAX (__NL80211_PHYMODE_AFTER_LAST - 1) -+ -+/** -+ * enum nl80211_bsstype - BSS types -+ * @NL80211_BSSTYPE_INFRASTRUCTURE: infrastructure BSS -+ * @NL80211_BSSTYPE_INDEPENDENT: independent BSS (ad-hoc network) -+ * @__NL80211_BSSTYPE_AFTER_LAST: internal use -+ * -+ * These values are used for %NL80211_ATTR_BSSTYPE. -+ */ -+enum nl80211_bsstype { -+ NL80211_BSSTYPE_INFRASTRUCTURE, -+ NL80211_BSSTYPE_INDEPENDENT, -+ -+ /* keep last */ -+ __NL80211_BSSTYPE_AFTER_LAST -+}; -+#define NL80211_BSSTYPE_MAX (__NL80211_BSSTYPE_AFTER_LAST - 1) -+ -+/** -+ * enum nl80211_keytype - key types -+ * @NL80211_KEYTYPE_GROUP: group key -+ * @NL80211_KEYTYPE_PAIRWISE: pairwise key -+ * @NL80211_KEYTYPE_PEER: peer key -+ */ -+enum nl80211_keytype { -+ NL80211_KEYTYPE_GROUP, -+ NL80211_KEYTYPE_PAIRWISE, -+ NL80211_KEYTYPE_PEER, -+ -+ /* keep last */ -+ __NL80211_KEYTYPE_AFTER_LAST -+}; -+#define NL80211_KEYTYPE_MAX (__NL80211_KEYTYPE_AFTER_LAST - 1) -+ - #endif /* __LINUX_NL80211_H */ ---- a/include/net/cfg80211.h -+++ b/include/net/cfg80211.h -@@ -3,6 +3,7 @@ - - #include <linux/netlink.h> - #include <linux/skbuff.h> -+#include <linux/nl80211.h> - #include <net/genetlink.h> - - /* -@@ -11,6 +12,69 @@ - * Copyright 2006 Johannes Berg <johannes@sipsolutions.net> - */ - -+/** -+ * struct scan_channel - describes a single channel to scan -+ * @phymode: PHY mode for this channel -+ * @channel: channel number (1-14, ...) -+ * @active: scan actively or passively on this channel -+ */ -+struct scan_channel { -+ enum nl80211_phymode phymode; -+ u32 channel; -+ int active; -+}; -+ -+/** -+ * struct scan_params - describes scan parameters -+ * @n_channels: number of items in @channels array or -1 to indicate all -+ * channels should be scanned (in that case @channels will be %NULL) -+ * @active: when n_channels is -1 this determines active/passive scanning. -+ * @phymode: when n_channels is -1 this determines PHY mode to scan. It is -+ * not possible to scan different PHY modes in one request w/o giving -+ * a channel list. -+ * @channels: array containing @n_channels &struct scan_channel items -+ */ -+struct scan_params { -+ int n_channels; -+ int active; -+ enum nl80211_phymode phymode; -+ struct scan_channel *channels; -+}; -+ -+/** -+ * struct association_params - describes association parameters -+ * @valid: this member contains flags which items are valid -+ * @bssid: the BSSID of the BSS to associate [%ASSOC_PARAMS_BSSID] -+ * @timeout: timeout (in TU) [%ASSOC_PARAMS_TIMEOUT] -+ * @ie: information element(s) to include in the association frames [%ASSOC_PARAMS_IE] -+ * @ie_len: length of the information element(s) -+ * @ssid: the SSID, always valid. -+ * @ssid_len: length of the SSID -+ */ -+struct association_params { -+ u8 *bssid; -+ u32 timeout; -+ u8 *ie; -+ int ie_len; -+ u8 *ssid; -+ int ssid_len; -+ -+ unsigned int valid; -+}; -+#define ASSOC_PARAMS_TIMEOUT (1<<0) -+ -+/** -+ * struct key_params - key information -+ */ -+struct key_params { -+ u8 *key; -+ int key_len; -+ int key_id; -+ u32 key_type; -+ u8 *macaddress; -+ u32 cipher; -+}; -+ - /* from net/wireless.h */ - struct wiphy; - -@@ -30,11 +94,62 @@ - * @add_virtual_intf: create a new virtual interface with the given name - * - * @del_virtual_intf: remove the virtual interface determined by ifindex. -+ * -+ * @change_virtual_intf: change type of virtual interface -+ * -+ * @associate: associate with given parameters -+ * -+ * @disassociate: disassociate from current AP -+ * -+ * @deauth: deauth from current AP -+ * -+ * @initiate_scan: scan with the given information (see &struct scan_params above) -+ * -+ * @get_association: get BSSID of the BSS that the device is currently -+ * associated to and return 1, or return 0 if not -+ * associated (or a negative error code) -+ * @get_auth_list: get list of BSSIDs of all BSSs the device has -+ * authenticated with, must call next_bssid for each, -+ * next_bssid returns non-zero on error, the given data -+ * is to be passed to that callback -+ * @add_key: add a key using &struct key_params -+ * @del_key: delete a key using info from &struct key_params - */ - struct cfg80211_ops { - int (*add_virtual_intf)(struct wiphy *wiphy, char *name, -- unsigned int type); -+ enum nl80211_iftype type); - int (*del_virtual_intf)(struct wiphy *wiphy, int ifindex); -+ int (*change_virtual_intf)(struct wiphy *wiphy, int ifindex, -+ enum nl80211_iftype type); -+ -+ int (*associate)(struct wiphy *wiphy, struct net_device *dev, -+ struct association_params *params); -+ int (*disassociate)(struct wiphy *wiphy, struct net_device *dev); -+ int (*deauth)(struct wiphy *wiphy, struct net_device *dev); -+ -+ -+ int (*initiate_scan)(struct wiphy *wiphy, struct net_device *dev, -+ struct scan_params *params); -+ -+ -+ int (*get_association)(struct wiphy *wiphy, struct net_device *dev, -+ u8 *bssid); -+ -+ int (*get_auth_list)(struct wiphy *wiphy, struct net_device *dev, -+ void *data, -+ int (*next_bssid)(void *data, u8 *bssid)); -+ -+ int (*add_key)(struct wiphy *wiphy, struct net_device *dev, -+ struct key_params *params); -+ int (*del_key)(struct wiphy *wiphy, struct net_device *dev, -+ struct key_params *params); - }; - -+ -+/* helper functions specific to nl80211 */ -+extern void *nl80211hdr_put(struct sk_buff *skb, u32 pid, -+ u32 seq, int flags, u8 cmd); -+extern void *nl80211msg_new(struct sk_buff **skb, u32 pid, -+ u32 seq, int flags, u8 cmd); -+ - #endif /* __NET_CFG80211_H */ ---- a/include/net/iw_handler.h -+++ b/include/net/iw_handler.h -@@ -431,7 +431,13 @@ - * Those may be called only within the kernel. - */ - --/* functions that may be called by driver modules */ -+/* First : function strictly used inside the kernel */ -+ -+/* Handle /proc/net/wireless, called in net/code/dev.c */ -+extern int dev_get_wireless_info(char * buffer, char **start, off_t offset, -+ int length); -+ -+/* Second : functions that may be called by driver modules */ - - /* Send a single event to user space */ - extern void wireless_send_event(struct net_device * dev, ---- a/include/net/mac80211.h -+++ b/include/net/mac80211.h -@@ -300,7 +300,6 @@ - /* Following five fields are used for IEEE 802.11H */ - unsigned int radar_detect; - unsigned int spect_mgmt; -- /* All following fields are currently unused. */ - unsigned int quiet_duration; /* duration of quiet period */ - unsigned int quiet_offset; /* how far into the beacon is the quiet - * period */ -@@ -514,6 +513,9 @@ - * per-packet RC4 key with each TX frame when doing hwcrypto */ - #define IEEE80211_HW_TKIP_REQ_PHASE2_KEY (1<<14) - -+ /* The device capable of supporting 11n */ -+#define IEEE80211_HW_SUPPORT_HT_MODE (1<<15) -+ - u32 flags; /* hardware flags defined above */ - - /* Set to the size of a needed device specific skb headroom for TX skbs. */ -@@ -641,8 +643,7 @@ - * used if the wlan hardware or low-level driver implements PAE. - * 80211.o module will anyway filter frames based on authorization - * state, so this function pointer can be NULL if low-level driver does -- * not require event notification about port state changes. -- * Currently unused. */ -+ * not require event notification about port state changes. */ - int (*set_port_auth)(struct ieee80211_hw *hw, u8 *addr, - int authorized); - -@@ -694,8 +695,9 @@ - /* Get statistics of the current TX queue status. This is used to get - * number of currently queued packets (queue length), maximum queue - * size (limit), and total number of packets sent using each TX queue -- * (count). -- * Currently unused. */ -+ * (count). This information is used for WMM to find out which TX -+ * queues have room for more packets and by hostapd to provide -+ * statistics about the current queueing state to external programs. */ - int (*get_tx_stats)(struct ieee80211_hw *hw, - struct ieee80211_tx_queue_stats *stats); - -@@ -705,12 +707,25 @@ - * Must be atomic. */ - u64 (*get_tsf)(struct ieee80211_hw *hw); - -+ /* Call low level driver with 11n Block Ack action */ -+ int (*handle_ba_action)(struct ieee80211_hw *hw, -+ struct ieee80211_mgmt *mgmt); -+ - /* Reset the TSF timer and allow firmware/hardware to synchronize with - * other STAs in the IBSS. This is only used in IBSS mode. This - * function is optional if the firmware/hardware takes full care of - * TSF synchronization. */ - void (*reset_tsf)(struct ieee80211_hw *hw); - -+ /* Configure ht parameters. */ -+ int (*conf_ht)(struct ieee80211_hw *hw, -+ struct ieee80211_ht_capability *ht_cap_param, -+ struct ieee80211_ht_additional_info *ht_extra_param); -+ -+ /* Get ht capabilities from the device */ -+ void (*get_ht_capab)(struct ieee80211_hw *hw, -+ struct ieee80211_ht_capability *ht_cap_param); -+ - /* Setup beacon data for IBSS beacons. Unlike access point (Master), - * IBSS uses a fixed beacon frame which is configured using this - * function. This handler is required only for IBSS mode. */ diff --git a/target/linux/generic-2.6/patches-2.6.22/213-kobject_uevent.patch b/target/linux/generic-2.6/patches-2.6.22/213-kobject_uevent.patch deleted file mode 100644 index 5a3e237..0000000 --- a/target/linux/generic-2.6/patches-2.6.22/213-kobject_uevent.patch +++ /dev/null @@ -1,37 +0,0 @@ ---- a/lib/kobject_uevent.c -+++ b/lib/kobject_uevent.c -@@ -30,9 +30,22 @@ - char uevent_helper[UEVENT_HELPER_PATH_LEN] = "/sbin/hotplug"; - static DEFINE_SPINLOCK(sequence_lock); - #if defined(CONFIG_NET) --static struct sock *uevent_sock; -+struct sock *uevent_sock = NULL; -+EXPORT_SYMBOL_GPL(uevent_sock); - #endif - -+u64 uevent_next_seqnum(void) -+{ -+ u64 seq; -+ -+ spin_lock(&sequence_lock); -+ seq = ++uevent_seqnum; -+ spin_unlock(&sequence_lock); -+ -+ return seq; -+} -+EXPORT_SYMBOL_GPL(uevent_next_seqnum); -+ - static char *action_to_string(enum kobject_action action) - { - switch (action) { -@@ -169,9 +182,7 @@ - } - - /* we will send an event, request a new sequence number */ -- spin_lock(&sequence_lock); -- seq = ++uevent_seqnum; -- spin_unlock(&sequence_lock); -+ seq = uevent_next_seqnum(); - sprintf(seq_buff, "SEQNUM=%llu", (unsigned long long)seq); - - #if defined(CONFIG_NET) diff --git a/target/linux/generic-2.6/patches-2.6.22/400-ledtrig_morse.patch b/target/linux/generic-2.6/patches-2.6.22/400-ledtrig_morse.patch deleted file mode 100644 index 144094a..0000000 --- a/target/linux/generic-2.6/patches-2.6.22/400-ledtrig_morse.patch +++ /dev/null @@ -1,19 +0,0 @@ ---- a/drivers/leds/Kconfig -+++ b/drivers/leds/Kconfig -@@ -128,5 +128,9 @@ - load average. - If unsure, say Y. - -+config LEDS_TRIGGER_MORSE -+ tristate "LED Morse Trigger" -+ depends on LEDS_TRIGGERS -+ - endmenu - ---- a/drivers/leds/Makefile -+++ b/drivers/leds/Makefile -@@ -21,3 +21,4 @@ - obj-$(CONFIG_LEDS_TRIGGER_TIMER) += ledtrig-timer.o - obj-$(CONFIG_LEDS_TRIGGER_IDE_DISK) += ledtrig-ide-disk.o - obj-$(CONFIG_LEDS_TRIGGER_HEARTBEAT) += ledtrig-heartbeat.o -+obj-$(CONFIG_LEDS_TRIGGER_MORSE) += ledtrig-morse.o diff --git a/target/linux/generic-2.6/patches-2.6.22/410-gpio_buttons.patch b/target/linux/generic-2.6/patches-2.6.22/410-gpio_buttons.patch deleted file mode 100644 index 31c311a..0000000 --- a/target/linux/generic-2.6/patches-2.6.22/410-gpio_buttons.patch +++ /dev/null @@ -1,30 +0,0 @@ ---- a/drivers/input/misc/Kconfig -+++ b/drivers/input/misc/Kconfig -@@ -178,4 +178,20 @@ - Say Y here if you want to support the built-in real time clock - of the HP SDC controller. - -+config INPUT_GPIO_BUTTONS -+ tristate "Polled GPIO buttons interface" -+ depends on GENERIC_GPIO -+ select INPUT_POLLDEV -+ help -+ This driver implements support for buttons connected -+ to GPIO pins of various CPUs (and some other chips). -+ -+ Say Y here if your device has buttons connected -+ directly to such GPIO pins. Your board-specific -+ setup logic must also provide a platform device, -+ with configuration data saying which GPIOs are used. -+ -+ To compile this driver as a module, choose M here: the -+ module will be called gpio-buttons. -+ - endif ---- a/drivers/input/misc/Makefile -+++ b/drivers/input/misc/Makefile -@@ -18,3 +18,4 @@ - obj-$(CONFIG_INPUT_YEALINK) += yealink.o - obj-$(CONFIG_HP_SDC_RTC) += hp_sdc_rtc.o - obj-$(CONFIG_INPUT_UINPUT) += uinput.o -+obj-$(CONFIG_INPUT_GPIO_BUTTONS) += gpio_buttons.o diff --git a/target/linux/generic-2.6/patches-2.6.22/510-yaffs_support.patch b/target/linux/generic-2.6/patches-2.6.22/510-yaffs_support.patch deleted file mode 100644 index a48dfbb..0000000 --- a/target/linux/generic-2.6/patches-2.6.22/510-yaffs_support.patch +++ /dev/null @@ -1,17 +0,0 @@ ---- a/fs/Kconfig -+++ b/fs/Kconfig -@@ -419,6 +419,7 @@ - - source "fs/xfs/Kconfig" - source "fs/gfs2/Kconfig" -+source "fs/yaffs2/Kconfig" - - config OCFS2_FS - tristate "OCFS2 file system support" ---- a/fs/Makefile -+++ b/fs/Makefile -@@ -120,3 +120,4 @@ - obj-$(CONFIG_DEBUG_FS) += debugfs/ - obj-$(CONFIG_OCFS2_FS) += ocfs2/ - obj-$(CONFIG_GFS2_FS) += gfs2/ -+obj-$(CONFIG_YAFFS_FS) += yaffs2/ diff --git a/target/linux/generic-2.6/patches-2.6.22/511-yaffs_reduce_compiler_warnings.patch b/target/linux/generic-2.6/patches-2.6.22/511-yaffs_reduce_compiler_warnings.patch deleted file mode 100644 index e3f0263..0000000 --- a/target/linux/generic-2.6/patches-2.6.22/511-yaffs_reduce_compiler_warnings.patch +++ /dev/null @@ -1,80 +0,0 @@ ---- a/fs/yaffs2/yaffs_fs.c -+++ b/fs/yaffs2/yaffs_fs.c -@@ -965,7 +965,7 @@ - f->f_version = inode->i_version; - } - -- list_for_each(i, &obj->variant.directoryVariant.children) { -+ list_for_each(i, (struct list_head *)&obj->variant.directoryVariant.children) { - curoffs++; - if (curoffs >= offset) { - l = list_entry(i, yaffs_Object, siblings); -@@ -1269,7 +1269,7 @@ - - if (target && - target->variantType == YAFFS_OBJECT_TYPE_DIRECTORY && -- !list_empty(&target->variant.directoryVariant.children)) { -+ !list_empty((struct list_head *)&target->variant.directoryVariant.children)) { - - T(YAFFS_TRACE_OS, (KERN_DEBUG "target is non-empty dir\n")); - -@@ -1503,7 +1503,7 @@ - yaffs_GrossUnlock(dev); - - /* we assume this is protected by lock_kernel() in mount/umount */ -- list_del(&dev->devList); -+ list_del((struct list_head *)&dev->devList); - - if(dev->spareBuffer){ - YFREE(dev->spareBuffer); -@@ -1847,7 +1847,7 @@ - dev->skipCheckpointWrite = options.skip_checkpoint_write; - - /* we assume this is protected by lock_kernel() in mount/umount */ -- list_add_tail(&dev->devList, &yaffs_dev_list); -+ list_add_tail((struct list_head *)&dev->devList, &yaffs_dev_list); - - init_MUTEX(&dev->grossLock); - ---- a/fs/yaffs2/yaffs_mtdif1.c -+++ b/fs/yaffs2/yaffs_mtdif1.c -@@ -323,7 +323,7 @@ - * Always returns YAFFS_OK. - */ - int nandmtd1_QueryNANDBlock(struct yaffs_DeviceStruct *dev, int blockNo, -- yaffs_BlockState * pState, int *pSequenceNumber) -+ yaffs_BlockState * pState, __u32 *pSequenceNumber) - { - struct mtd_info * mtd = dev->genericDevice; - int chunkNo = blockNo * dev->nChunksPerBlock; ---- a/fs/yaffs2/yaffs_mtdif1.h -+++ b/fs/yaffs2/yaffs_mtdif1.h -@@ -23,6 +23,6 @@ - int nandmtd1_MarkNANDBlockBad(struct yaffs_DeviceStruct *dev, int blockNo); - - int nandmtd1_QueryNANDBlock(struct yaffs_DeviceStruct *dev, int blockNo, -- yaffs_BlockState * state, int *sequenceNumber); -+ yaffs_BlockState * state, __u32 *sequenceNumber); - - #endif ---- a/fs/yaffs2/yaffs_mtdif2.c -+++ b/fs/yaffs2/yaffs_mtdif2.c -@@ -188,7 +188,7 @@ - } - - int nandmtd2_QueryNANDBlock(struct yaffs_DeviceStruct *dev, int blockNo, -- yaffs_BlockState * state, int *sequenceNumber) -+ yaffs_BlockState * state, __u32 *sequenceNumber) - { - struct mtd_info *mtd = (struct mtd_info *)(dev->genericDevice); - int retval; ---- a/fs/yaffs2/yaffs_mtdif2.h -+++ b/fs/yaffs2/yaffs_mtdif2.h -@@ -24,6 +24,6 @@ - __u8 * data, yaffs_ExtendedTags * tags); - int nandmtd2_MarkNANDBlockBad(struct yaffs_DeviceStruct *dev, int blockNo); - int nandmtd2_QueryNANDBlock(struct yaffs_DeviceStruct *dev, int blockNo, -- yaffs_BlockState * state, int *sequenceNumber); -+ yaffs_BlockState * state, __u32 *sequenceNumber); - - #endif diff --git a/target/linux/generic-2.6/patches-2.6.22/600-eeprom_93cx6.patch b/target/linux/generic-2.6/patches-2.6.22/600-eeprom_93cx6.patch deleted file mode 100644 index 205ede2..0000000 --- a/target/linux/generic-2.6/patches-2.6.22/600-eeprom_93cx6.patch +++ /dev/null @@ -1,348 +0,0 @@ -From: Ivo van Doorn <ivdoorn@gmail.com> -Date: Fri, 11 May 2007 19:59:40 +0000 (-0400) -Subject: [PATCH] Add 93cx6 eeprom library -X-Git-Tag: v2.6.23-rc1~1201^2~74 -X-Git-Url: http://git.kernel.org/?p=linux%2Fkernel%2Fgit%2Ftorvalds%2Flinux-2.6.git;a=commitdiff_plain;h=9467d64b0e88763914c01f71ddf591b166c4f526 - -[PATCH] Add 93cx6 eeprom library - -This patch adds a library for reading from 93cx6 eeproms. - -Signed-off-by: Michael Wu <flamingice@sourmilk.net> -Signed-off-by: John W. Linville <linville@tuxdriver.com> ---- - ---- a/drivers/misc/Kconfig -+++ b/drivers/misc/Kconfig -@@ -34,6 +34,11 @@ - If you choose to build module, its name will be phantom. If unsure, - say N here. - -+config EEPROM_93CX6 -+ tristate "EEPROM 93CX6 support" -+ ---help--- -+ This is a driver for the EEPROM chipsets 93c46 and 93c66. -+ The driver supports both read as well as write commands. - - If unsure, say N. - -@@ -187,5 +192,4 @@ - - If you are not sure, say Y here. - -- - endmenu ---- a/drivers/misc/Makefile -+++ b/drivers/misc/Makefile -@@ -14,3 +14,4 @@ - obj-$(CONFIG_SGI_IOC4) += ioc4.o - obj-$(CONFIG_SONY_LAPTOP) += sony-laptop.o - obj-$(CONFIG_THINKPAD_ACPI) += thinkpad_acpi.o -+obj-$(CONFIG_EEPROM_93CX6) += eeprom_93cx6.o ---- /dev/null -+++ b/drivers/misc/eeprom_93cx6.c -@@ -0,0 +1,229 @@ -+/* -+ Copyright (C) 2004 - 2006 rt2x00 SourceForge Project -+ <http://rt2x00.serialmonkey.com> -+ -+ This program is free software; you can redistribute it and/or modify -+ it under the terms of the GNU General Public License as published by -+ the Free Software Foundation; either version 2 of the License, or -+ (at your option) any later version. -+ -+ This program is distributed in the hope that it will be useful, -+ but WITHOUT ANY WARRANTY; without even the implied warranty of -+ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the -+ GNU General Public License for more details. -+ -+ You should have received a copy of the GNU General Public License -+ along with this program; if not, write to the -+ Free Software Foundation, Inc., -+ 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. -+ */ -+ -+/* -+ Module: eeprom_93cx6 -+ Abstract: EEPROM reader routines for 93cx6 chipsets. -+ Supported chipsets: 93c46 & 93c66. -+ */ -+ -+#include <linux/kernel.h> -+#include <linux/module.h> -+#include <linux/version.h> -+#include <linux/delay.h> -+#include <linux/eeprom_93cx6.h> -+ -+MODULE_AUTHOR("http://rt2x00.serialmonkey.com"); -+MODULE_VERSION("1.0"); -+MODULE_DESCRIPTION("EEPROM 93cx6 chip driver"); -+MODULE_LICENSE("GPL"); -+ -+static inline void eeprom_93cx6_pulse_high(struct eeprom_93cx6 *eeprom) -+{ -+ eeprom->reg_data_clock = 1; -+ eeprom->register_write(eeprom); -+ udelay(1); -+} -+ -+static inline void eeprom_93cx6_pulse_low(struct eeprom_93cx6 *eeprom) -+{ -+ eeprom->reg_data_clock = 0; -+ eeprom->register_write(eeprom); -+ udelay(1); -+} -+ -+static void eeprom_93cx6_startup(struct eeprom_93cx6 *eeprom) -+{ -+ /* -+ * Clear all flags, and enable chip select. -+ */ -+ eeprom->register_read(eeprom); -+ eeprom->reg_data_in = 0; -+ eeprom->reg_data_out = 0; -+ eeprom->reg_data_clock = 0; -+ eeprom->reg_chip_select = 1; -+ eeprom->register_write(eeprom); -+ -+ /* -+ * kick a pulse. -+ */ -+ eeprom_93cx6_pulse_high(eeprom); -+ eeprom_93cx6_pulse_low(eeprom); -+} -+ -+static void eeprom_93cx6_cleanup(struct eeprom_93cx6 *eeprom) -+{ -+ /* -+ * Clear chip_select and data_in flags. -+ */ -+ eeprom->register_read(eeprom); -+ eeprom->reg_data_in = 0; -+ eeprom->reg_chip_select = 0; -+ eeprom->register_write(eeprom); -+ -+ /* -+ * kick a pulse. -+ */ -+ eeprom_93cx6_pulse_high(eeprom); -+ eeprom_93cx6_pulse_low(eeprom); -+} -+ -+static void eeprom_93cx6_write_bits(struct eeprom_93cx6 *eeprom, -+ const u16 data, const u16 count) -+{ -+ unsigned int i; -+ -+ eeprom->register_read(eeprom); -+ -+ /* -+ * Clear data flags. -+ */ -+ eeprom->reg_data_in = 0; -+ eeprom->reg_data_out = 0; -+ -+ /* -+ * Start writing all bits. -+ */ -+ for (i = count; i > 0; i--) { -+ /* -+ * Check if this bit needs to be set. -+ */ -+ eeprom->reg_data_in = !!(data & (1 << (i - 1))); -+ -+ /* -+ * Write the bit to the eeprom register. -+ */ -+ eeprom->register_write(eeprom); -+ -+ /* -+ * Kick a pulse. -+ */ -+ eeprom_93cx6_pulse_high(eeprom); -+ eeprom_93cx6_pulse_low(eeprom); -+ } -+ -+ eeprom->reg_data_in = 0; -+ eeprom->register_write(eeprom); -+} -+ -+static void eeprom_93cx6_read_bits(struct eeprom_93cx6 *eeprom, -+ u16 *data, const u16 count) -+{ -+ unsigned int i; -+ u16 buf = 0; -+ -+ eeprom->register_read(eeprom); -+ -+ /* -+ * Clear data flags. -+ */ -+ eeprom->reg_data_in = 0; -+ eeprom->reg_data_out = 0; -+ -+ /* -+ * Start reading all bits. -+ */ -+ for (i = count; i > 0; i--) { -+ eeprom_93cx6_pulse_high(eeprom); -+ -+ eeprom->register_read(eeprom); -+ -+ /* -+ * Clear data_in flag. -+ */ -+ eeprom->reg_data_in = 0; -+ -+ /* -+ * Read if the bit has been set. -+ */ -+ if (eeprom->reg_data_out) -+ buf |= (1 << (i - 1)); -+ -+ eeprom_93cx6_pulse_low(eeprom); -+ } -+ -+ *data = buf; -+} -+ -+/** -+ * eeprom_93cx6_read - Read multiple words from eeprom -+ * @eeprom: Pointer to eeprom structure -+ * @word: Word index from where we should start reading -+ * @data: target pointer where the information will have to be stored -+ * -+ * This function will read the eeprom data as host-endian word -+ * into the given data pointer. -+ */ -+void eeprom_93cx6_read(struct eeprom_93cx6 *eeprom, const u8 word, -+ u16 *data) -+{ -+ u16 command; -+ -+ /* -+ * Initialize the eeprom register -+ */ -+ eeprom_93cx6_startup(eeprom); -+ -+ /* -+ * Select the read opcode and the word to be read. -+ */ -+ command = (PCI_EEPROM_READ_OPCODE << eeprom->width) | word; -+ eeprom_93cx6_write_bits(eeprom, command, -+ PCI_EEPROM_WIDTH_OPCODE + eeprom->width); -+ -+ /* -+ * Read the requested 16 bits. -+ */ -+ eeprom_93cx6_read_bits(eeprom, data, 16); -+ -+ /* -+ * Cleanup eeprom register. -+ */ -+ eeprom_93cx6_cleanup(eeprom); -+} -+EXPORT_SYMBOL_GPL(eeprom_93cx6_read); -+ -+/** -+ * eeprom_93cx6_multiread - Read multiple words from eeprom -+ * @eeprom: Pointer to eeprom structure -+ * @word: Word index from where we should start reading -+ * @data: target pointer where the information will have to be stored -+ * @words: Number of words that should be read. -+ * -+ * This function will read all requested words from the eeprom, -+ * this is done by calling eeprom_93cx6_read() multiple times. -+ * But with the additional change that while the eeprom_93cx6_read -+ * will return host ordered bytes, this method will return little -+ * endian words. -+ */ -+void eeprom_93cx6_multiread(struct eeprom_93cx6 *eeprom, const u8 word, -+ __le16 *data, const u16 words) -+{ -+ unsigned int i; -+ u16 tmp; -+ -+ for (i = 0; i < words; i++) { -+ tmp = 0; -+ eeprom_93cx6_read(eeprom, word + i, &tmp); -+ data[i] = cpu_to_le16(tmp); -+ } -+} -+EXPORT_SYMBOL_GPL(eeprom_93cx6_multiread); -+ ---- /dev/null -+++ b/include/linux/eeprom_93cx6.h -@@ -0,0 +1,72 @@ -+/* -+ Copyright (C) 2004 - 2006 rt2x00 SourceForge Project -+ <http://rt2x00.serialmonkey.com> -+ -+ This program is free software; you can redistribute it and/or modify -+ it under the terms of the GNU General Public License as published by -+ the Free Software Foundation; either version 2 of the License, or -+ (at your option) any later version. -+ -+ This program is distributed in the hope that it will be useful, -+ but WITHOUT ANY WARRANTY; without even the implied warranty of -+ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the -+ GNU General Public License for more details. -+ -+ You should have received a copy of the GNU General Public License -+ along with this program; if not, write to the -+ Free Software Foundation, Inc., -+ 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. -+ */ -+ -+/* -+ Module: eeprom_93cx6 -+ Abstract: EEPROM reader datastructures for 93cx6 chipsets. -+ Supported chipsets: 93c46 & 93c66. -+ */ -+ -+/* -+ * EEPROM operation defines. -+ */ -+#define PCI_EEPROM_WIDTH_93C46 6 -+#define PCI_EEPROM_WIDTH_93C66 8 -+#define PCI_EEPROM_WIDTH_OPCODE 3 -+#define PCI_EEPROM_WRITE_OPCODE 0x05 -+#define PCI_EEPROM_READ_OPCODE 0x06 -+#define PCI_EEPROM_EWDS_OPCODE 0x10 -+#define PCI_EEPROM_EWEN_OPCODE 0x13 -+ -+/** -+ * struct eeprom_93cx6 - control structure for setting the commands -+ * for reading the eeprom data. -+ * @data: private pointer for the driver. -+ * @register_read(struct eeprom_93cx6 *eeprom): handler to -+ * read the eeprom register, this function should set all reg_* fields. -+ * @register_write(struct eeprom_93cx6 *eeprom): handler to -+ * write to the eeprom register by using all reg_* fields. -+ * @width: eeprom width, should be one of the PCI_EEPROM_WIDTH_* defines -+ * @reg_data_in: register field to indicate data input -+ * @reg_data_out: register field to indicate data output -+ * @reg_data_clock: register field to set the data clock -+ * @reg_chip_select: register field to set the chip select -+ * -+ * This structure is used for the communication between the driver -+ * and the eeprom_93cx6 handlers for reading the eeprom. -+ */ -+struct eeprom_93cx6 { -+ void *data; -+ -+ void (*register_read)(struct eeprom_93cx6 *eeprom); -+ void (*register_write)(struct eeprom_93cx6 *eeprom); -+ -+ int width; -+ -+ char reg_data_in; -+ char reg_data_out; -+ char reg_data_clock; -+ char reg_chip_select; -+}; -+ -+extern void eeprom_93cx6_read(struct eeprom_93cx6 *eeprom, -+ const u8 word, u16 *data); -+extern void eeprom_93cx6_multiread(struct eeprom_93cx6 *eeprom, -+ const u8 word, __le16 *data, const u16 words); diff --git a/target/linux/generic-2.6/patches-2.6.22/601-eeprom_93cx6_fixes.patch b/target/linux/generic-2.6/patches-2.6.22/601-eeprom_93cx6_fixes.patch deleted file mode 100644 index 71c169f..0000000 --- a/target/linux/generic-2.6/patches-2.6.22/601-eeprom_93cx6_fixes.patch +++ /dev/null @@ -1,31 +0,0 @@ ---- a/drivers/misc/eeprom_93cx6.c -+++ b/drivers/misc/eeprom_93cx6.c -@@ -39,14 +39,26 @@ - { - eeprom->reg_data_clock = 1; - eeprom->register_write(eeprom); -- udelay(1); -+ -+ /* -+ * Add a short delay for the pulse to work. -+ * According to the specifications the "maximum minimum" -+ * time should be 450ns. -+ */ -+ ndelay(450); - } - - static inline void eeprom_93cx6_pulse_low(struct eeprom_93cx6 *eeprom) - { - eeprom->reg_data_clock = 0; - eeprom->register_write(eeprom); -- udelay(1); -+ -+ /* -+ * Add a short delay for the pulse to work. -+ * According to the specifications the "maximum minimum" -+ * time should be 450ns. -+ */ -+ ndelay(450); - } - - static void eeprom_93cx6_startup(struct eeprom_93cx6 *eeprom) diff --git a/target/linux/generic-2.6/patches-2.6.22/801-usb_serial_endpoint_size.patch b/target/linux/generic-2.6/patches-2.6.22/801-usb_serial_endpoint_size.patch deleted file mode 100644 index bbca936..0000000 --- a/target/linux/generic-2.6/patches-2.6.22/801-usb_serial_endpoint_size.patch +++ /dev/null @@ -1,25 +0,0 @@ ---- a/drivers/usb/serial/usb-serial.c -+++ b/drivers/usb/serial/usb-serial.c -@@ -56,6 +56,7 @@ - drivers depend on it. - */ - -+static ushort maxSize = 0; - static int debug; - static struct usb_serial *serial_table[SERIAL_TTY_MINORS]; /* initially all NULL */ - static spinlock_t table_lock; -@@ -864,7 +865,7 @@ - dev_err(&interface->dev, "No free urbs available\n"); - goto probe_error; - } -- buffer_size = le16_to_cpu(endpoint->wMaxPacketSize); -+ buffer_size = (endpoint->wMaxPacketSize > maxSize) ? endpoint->wMaxPacketSize : maxSize; - port->bulk_in_size = buffer_size; - port->bulk_in_endpointAddress = endpoint->bEndpointAddress; - port->bulk_in_buffer = kmalloc (buffer_size, GFP_KERNEL); -@@ -1245,3 +1246,5 @@ - - module_param(debug, bool, S_IRUGO | S_IWUSR); - MODULE_PARM_DESC(debug, "Debug enabled or not"); -+module_param(maxSize, ushort,0); -+MODULE_PARM_DESC(maxSize,"User specified USB endpoint size"); diff --git a/target/linux/generic-2.6/patches-2.6.22/900-headers_type_and_time.patch b/target/linux/generic-2.6/patches-2.6.22/900-headers_type_and_time.patch deleted file mode 100644 index 465f788..0000000 --- a/target/linux/generic-2.6/patches-2.6.22/900-headers_type_and_time.patch +++ /dev/null @@ -1,46 +0,0 @@ ---- a/include/linux/time.h -+++ b/include/linux/time.h -@@ -1,6 +1,10 @@ - #ifndef _LINUX_TIME_H - #define _LINUX_TIME_H - -+#ifndef __KERNEL__ -+#include <time.h> -+#else -+ - #include <linux/types.h> - - #ifdef __KERNEL__ -@@ -225,4 +229,6 @@ - */ - #define TIMER_ABSTIME 0x01 - -+#endif /* __KERNEL__ DEBIAN */ -+ - #endif ---- a/include/linux/types.h -+++ b/include/linux/types.h -@@ -1,6 +1,14 @@ - #ifndef _LINUX_TYPES_H - #define _LINUX_TYPES_H - -+/* Debian: Use userland types instead. */ -+#ifndef __KERNEL__ -+# include <sys/types.h> -+/* For other kernel headers. */ -+# include <linux/posix_types.h> -+# include <asm/types.h> -+#else -+ - #ifdef __KERNEL__ - - #define BITS_TO_LONGS(bits) \ -@@ -162,6 +170,8 @@ - - #endif /* __KERNEL_STRICT_NAMES */ - -+#endif /* __KERNEL__ DEBIAN */ -+ - /* - * Below are truly Linux-specific types that should never collide with - * any application/library that wants linux/types.h. diff --git a/target/linux/generic-2.6/patches-2.6.22/902-darwin_scripts_include.patch b/target/linux/generic-2.6/patches-2.6.22/902-darwin_scripts_include.patch deleted file mode 100644 index 301c0d8..0000000 --- a/target/linux/generic-2.6/patches-2.6.22/902-darwin_scripts_include.patch +++ /dev/null @@ -1,102 +0,0 @@ ---- a/scripts/genksyms/parse.c_shipped -+++ b/scripts/genksyms/parse.c_shipped -@@ -144,7 +144,9 @@ - - - #include <assert.h> -+#ifndef __APPLE__ - #include <malloc.h> -+#endif - #include "genksyms.h" - - static int is_typedef; ---- a/scripts/genksyms/parse.y -+++ b/scripts/genksyms/parse.y -@@ -24,7 +24,9 @@ - %{ - - #include <assert.h> -+#ifndef __APPLE__ - #include <malloc.h> -+#endif - #include "genksyms.h" - - static int is_typedef; ---- a/scripts/kallsyms.c -+++ b/scripts/kallsyms.c -@@ -30,6 +30,35 @@ - #include <stdlib.h> - #include <string.h> - #include <ctype.h> -+#ifdef __APPLE__ -+/* Darwin has no memmem implementation, this one is ripped of the uClibc-0.9.28 source */ -+void *memmem (const void *haystack, size_t haystack_len, -+ const void *needle, size_t needle_len) -+{ -+ const char *begin; -+ const char *const last_possible -+ = (const char *) haystack + haystack_len - needle_len; -+ -+ if (needle_len == 0) -+ /* The first occurrence of the empty string is deemed to occur at -+ the beginning of the string. */ -+ return (void *) haystack; -+ -+ /* Sanity check, otherwise the loop might search through the whole -+ memory. */ -+ if (__builtin_expect (haystack_len < needle_len, 0)) -+ return NULL; -+ -+ for (begin = (const char *) haystack; begin <= last_possible; ++begin) -+ if (begin[0] == ((const char *) needle)[0] && -+ !memcmp ((const void *) &begin[1], -+ (const void *) ((const char *) needle + 1), -+ needle_len - 1)) -+ return (void *) begin; -+ -+ return NULL; -+} -+#endif - - #define KSYM_NAME_LEN 127 - ---- a/scripts/kconfig/Makefile -+++ b/scripts/kconfig/Makefile -@@ -87,6 +87,9 @@ - # we really need to do so. (Do not call gcc as part of make mrproper) - HOST_EXTRACFLAGS = $(shell $(CONFIG_SHELL) $(check-lxdialog) -ccflags) - HOST_LOADLIBES = $(shell $(CONFIG_SHELL) $(check-lxdialog) -ldflags $(HOSTCC)) -+ifeq ($(shell uname -s),Darwin) -+HOST_LOADLIBES += -lncurses -+endif - - HOST_EXTRACFLAGS += -DLOCALE - ---- a/scripts/mod/mk_elfconfig.c -+++ b/scripts/mod/mk_elfconfig.c -@@ -1,7 +1,11 @@ - #include <stdio.h> - #include <stdlib.h> - #include <string.h> -+#ifndef __APPLE__ - #include <elf.h> -+#else -+#include "../../../../../tools/sstrip/include/elf.h" -+#endif - - int - main(int argc, char **argv) ---- a/scripts/mod/modpost.h -+++ b/scripts/mod/modpost.h -@@ -7,7 +7,11 @@ - #include <sys/mman.h> - #include <fcntl.h> - #include <unistd.h> -+#ifndef __APPLE__ - #include <elf.h> -+#else -+#include "../../../../../tools/sstrip/include/elf.h" -+#endif - - #include "elfconfig.h" - diff --git a/target/linux/generic-2.6/patches-2.6.22/903-stddef_include.patch b/target/linux/generic-2.6/patches-2.6.22/903-stddef_include.patch deleted file mode 100644 index c0c01c0..0000000 --- a/target/linux/generic-2.6/patches-2.6.22/903-stddef_include.patch +++ /dev/null @@ -1,17 +0,0 @@ ---- a/include/linux/stddef.h -+++ b/include/linux/stddef.h -@@ -16,6 +16,7 @@ - false = 0, - true = 1 - }; -+#endif /* __KERNEL__ */ - - #undef offsetof - #ifdef __compiler_offsetof -@@ -23,6 +24,5 @@ - #else - #define offsetof(TYPE, MEMBER) ((size_t) &((TYPE *)0)->MEMBER) - #endif --#endif /* __KERNEL__ */ - - #endif diff --git a/target/linux/generic-2.6/patches-2.6.22/904-ls_time_locale.patch b/target/linux/generic-2.6/patches-2.6.22/904-ls_time_locale.patch deleted file mode 100644 index 401b151..0000000 --- a/target/linux/generic-2.6/patches-2.6.22/904-ls_time_locale.patch +++ /dev/null @@ -1,20 +0,0 @@ ---- a/scripts/gen_initramfs_list.sh -+++ b/scripts/gen_initramfs_list.sh -@@ -125,7 +125,7 @@ - str="${ftype} ${name} ${location} ${str}" - ;; - "nod") -- local dev=`LC_ALL=C ls -l "${location}"` -+ local dev=`LC_ALL=C ls -l --time-style=locale "${location}"` - local maj=`field 5 ${dev}` - local min=`field 6 ${dev}` - maj=${maj%,} -@@ -135,7 +135,7 @@ - str="${ftype} ${name} ${str} ${dev} ${maj} ${min}" - ;; - "slink") -- local target=`field 11 $(LC_ALL=C ls -l "${location}")` -+ local target=`field 11 $(LC_ALL=C ls -l --time-style=locale "${location}")` - str="${ftype} ${name} ${target} ${str}" - ;; - *) diff --git a/target/linux/generic-2.6/patches-2.6.22/999-icplus.patch b/target/linux/generic-2.6/patches-2.6.22/999-icplus.patch deleted file mode 100644 index d38df05..0000000 --- a/target/linux/generic-2.6/patches-2.6.22/999-icplus.patch +++ /dev/null @@ -1,175 +0,0 @@ -From: Michael Barkowski <michael.barkowski@freescale.com> -Date: Fri, 11 May 2007 23:24:51 +0000 (-0500) -Subject: phylib: add the ICPlus IP175C PHY driver -X-Git-Tag: v2.6.23-rc1~1201^2~58 -X-Git-Url: http://git.kernel.org/?p=linux%2Fkernel%2Fgit%2Ftorvalds%2Flinux-2.6.git;a=commitdiff_plain;h=0cefeebaf3da39d768bffcf62460fe2088e824ef - -phylib: add the ICPlus IP175C PHY driver - -The ICPlus IP175C sports a 100Mbit/s 4-port switch in addition -to a dedicated 100Mbit/s WAN port. - -Signed-off-by: Michael Barkowski <michael.barkowski@freescale.com> -Signed-off-by: Kim Phillips <kim.phillips@freescale.com> -Signed-off-by: Jeff Garzik <jeff@garzik.org> ---- - ---- a/drivers/net/phy/Kconfig -+++ b/drivers/net/phy/Kconfig -@@ -55,6 +55,11 @@ - ---help--- - Currently supports the BCM5411, BCM5421 and BCM5461 PHYs. - -+config ICPLUS_PHY -+ tristate "Drivers for ICPlus PHYs" -+ ---help--- -+ Currently supports the IP175C PHY. -+ - config FIXED_PHY - tristate "Drivers for PHY emulation on fixed speed/link" - ---help--- ---- a/drivers/net/phy/Makefile -+++ b/drivers/net/phy/Makefile -@@ -11,4 +11,5 @@ - obj-$(CONFIG_SMSC_PHY) += smsc.o - obj-$(CONFIG_VITESSE_PHY) += vitesse.o - obj-$(CONFIG_BROADCOM_PHY) += broadcom.o -+obj-$(CONFIG_ICPLUS_PHY) += icplus.o - obj-$(CONFIG_FIXED_PHY) += fixed.o ---- /dev/null -+++ b/drivers/net/phy/icplus.c -@@ -0,0 +1,134 @@ -+/* -+ * Driver for ICPlus PHYs -+ * -+ * Copyright (c) 2007 Freescale Semiconductor, Inc. -+ * -+ * This program is free software; you can redistribute it and/or modify it -+ * under the terms of the GNU General Public License as published by the -+ * Free Software Foundation; either version 2 of the License, or (at your -+ * option) any later version. -+ * -+ */ -+#include <linux/kernel.h> -+#include <linux/string.h> -+#include <linux/errno.h> -+#include <linux/unistd.h> -+#include <linux/slab.h> -+#include <linux/interrupt.h> -+#include <linux/init.h> -+#include <linux/delay.h> -+#include <linux/netdevice.h> -+#include <linux/etherdevice.h> -+#include <linux/skbuff.h> -+#include <linux/spinlock.h> -+#include <linux/mm.h> -+#include <linux/module.h> -+#include <linux/mii.h> -+#include <linux/ethtool.h> -+#include <linux/phy.h> -+ -+#include <asm/io.h> -+#include <asm/irq.h> -+#include <asm/uaccess.h> -+ -+MODULE_DESCRIPTION("ICPlus IP175C PHY driver"); -+MODULE_AUTHOR("Michael Barkowski"); -+MODULE_LICENSE("GPL"); -+ -+static int ip175c_config_init(struct phy_device *phydev) -+{ -+ int err, i; -+ static int full_reset_performed = 0; -+ -+ if (full_reset_performed == 0) { -+ -+ /* master reset */ -+ err = phydev->bus->write(phydev->bus, 30, 0, 0x175c); -+ if (err < 0) -+ return err; -+ -+ /* ensure no bus delays overlap reset period */ -+ err = phydev->bus->read(phydev->bus, 30, 0); -+ -+ /* data sheet specifies reset period is 2 msec */ -+ mdelay(2); -+ -+ /* enable IP175C mode */ -+ err = phydev->bus->write(phydev->bus, 29, 31, 0x175c); -+ if (err < 0) -+ return err; -+ -+ /* Set MII0 speed and duplex (in PHY mode) */ -+ err = phydev->bus->write(phydev->bus, 29, 22, 0x420); -+ if (err < 0) -+ return err; -+ -+ /* reset switch ports */ -+ for (i = 0; i < 5; i++) { -+ err = phydev->bus->write(phydev->bus, i, -+ MII_BMCR, BMCR_RESET); -+ if (err < 0) -+ return err; -+ } -+ -+ for (i = 0; i < 5; i++) -+ err = phydev->bus->read(phydev->bus, i, MII_BMCR); -+ -+ mdelay(2); -+ -+ full_reset_performed = 1; -+ } -+ -+ if (phydev->addr != 4) { -+ phydev->state = PHY_RUNNING; -+ phydev->speed = SPEED_100; -+ phydev->duplex = DUPLEX_FULL; -+ phydev->link = 1; -+ netif_carrier_on(phydev->attached_dev); -+ } -+ -+ return 0; -+} -+ -+static int ip175c_read_status(struct phy_device *phydev) -+{ -+ if (phydev->addr == 4) /* WAN port */ -+ genphy_read_status(phydev); -+ else -+ /* Don't need to read status for switch ports */ -+ phydev->irq = PHY_IGNORE_INTERRUPT; -+ -+ return 0; -+} -+ -+static int ip175c_config_aneg(struct phy_device *phydev) -+{ -+ if (phydev->addr == 4) /* WAN port */ -+ genphy_config_aneg(phydev); -+ -+ return 0; -+} -+ -+static struct phy_driver ip175c_driver = { -+ .phy_id = 0x02430d80, -+ .name = "ICPlus IP175C", -+ .phy_id_mask = 0x0ffffff0, -+ .features = PHY_BASIC_FEATURES, -+ .config_init = &ip175c_config_init, -+ .config_aneg = &ip175c_config_aneg, -+ .read_status = &ip175c_read_status, -+ .driver = { .owner = THIS_MODULE,}, -+}; -+ -+static int __init ip175c_init(void) -+{ -+ return phy_driver_register(&ip175c_driver); -+} -+ -+static void __exit ip175c_exit(void) -+{ -+ phy_driver_unregister(&ip175c_driver); -+} -+ -+module_init(ip175c_init); -+module_exit(ip175c_exit); |