diff options
| author | Hauke Mehrtens <hauke@hauke-m.de> | 2014-04-08 19:50:17 +0000 |
|---|---|---|
| committer | Hauke Mehrtens <hauke@hauke-m.de> | 2014-04-08 19:50:17 +0000 |
| commit | 64f54741e3912a4cd8c57a47c100273ad995931b (patch) | |
| tree | e9eb66be132a65811639674ef021e9237952cfc0 /target/linux/generic/patches-3.12 | |
| parent | 4abcae7a1cd1092e8fd3ef5aee4c97108917a4ad (diff) | |
| download | mtk-20170518-64f54741e3912a4cd8c57a47c100273ad995931b.zip mtk-20170518-64f54741e3912a4cd8c57a47c100273ad995931b.tar.gz mtk-20170518-64f54741e3912a4cd8c57a47c100273ad995931b.tar.bz2 | |
kernel: bgmac: rework patch checking packet length
This bgmac patch was an attempt to fix/workaround bug reported in
https://dev.openwrt.org/ticket/7198 noticed on WNR3500L.
Patch assumed length reported by the hardware was 0 and was trying to
read it until getting a different value. This was actually the opposite.
Lenghts were some invalid & huge values that resulted in skb_over_panic.
For example:
skbuff: skb_over_panic: text:83b21074 len:57222 (...)
skbuff: skb_over_panic: text:87af1024 len:43226 (...)
skbuff: skb_over_panic: text:87af5024 len:8739 (...)
So instead of that not-working patch checking for 0, write a new one
checking for huge values. In case something like that happens, dump
hardware state and drop the packet.
Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
SVN-Revision: 40424
Diffstat (limited to 'target/linux/generic/patches-3.12')
| -rw-r--r-- | target/linux/generic/patches-3.12/775-bgmac-check-length-of-received-frame.patch | 39 |
1 files changed, 39 insertions, 0 deletions
diff --git a/target/linux/generic/patches-3.12/775-bgmac-check-length-of-received-frame.patch b/target/linux/generic/patches-3.12/775-bgmac-check-length-of-received-frame.patch new file mode 100644 index 0000000..c3d63b6 --- /dev/null +++ b/target/linux/generic/patches-3.12/775-bgmac-check-length-of-received-frame.patch @@ -0,0 +1,39 @@ +From 2d12a9abf3f81de5b51852e3cfcba8cedac82642 Mon Sep 17 00:00:00 2001 +From: Hauke Mehrtens <hauke@hauke-m.de> +Date: Fri, 6 Dec 2013 01:14:52 +0100 +Subject: [PATCH] bgmac: check length of received frame + +--- + drivers/net/ethernet/broadcom/bgmac.c | 9 ++++++++- + 1 file changed, 8 insertions(+), 1 deletion(-) + +--- a/drivers/net/ethernet/broadcom/bgmac.c ++++ b/drivers/net/ethernet/broadcom/bgmac.c +@@ -363,6 +363,27 @@ static int bgmac_dma_rx_read(struct bgma + dma_addr_t old_dma_addr = slot->dma_addr; + int err; + ++ if (len > BGMAC_RX_MAX_FRAME_SIZE) { ++ struct bgmac_dma_desc *dma_desc = ring->cpu_base + ring->start; ++ ++ bgmac_err(bgmac, "Hardware reported invalid packet length %d for slot %d!\n", len, ring->start); ++ bgmac_err(bgmac, "flags: 0x%04X\n", flags); ++ bgmac_err(bgmac, "ctl0: 0x%08X\tctl1: 0x%08X\n", le32_to_cpu(dma_desc->ctl0), le32_to_cpu(dma_desc->ctl1)); ++ ++ bgmac_err(bgmac, " BGMAC_DMA_RX_CTL: 0x%08X\n", bgmac_read(bgmac, ring->mmio_base + BGMAC_DMA_RX_CTL)); ++ bgmac_err(bgmac, " BGMAC_DMA_RX_INDEX: 0x%08X\n", bgmac_read(bgmac, ring->mmio_base + BGMAC_DMA_RX_INDEX)); ++ bgmac_err(bgmac, "BGMAC_DMA_RX_RINGLO: 0x%08X\n", bgmac_read(bgmac, ring->mmio_base + BGMAC_DMA_RX_RINGLO)); ++ bgmac_err(bgmac, "BGMAC_DMA_RX_RINGHI: 0x%08X\n", bgmac_read(bgmac, ring->mmio_base + BGMAC_DMA_RX_RINGHI)); ++ bgmac_err(bgmac, "BGMAC_DMA_RX_STATUS: 0x%08X\n", bgmac_read(bgmac, ring->mmio_base + BGMAC_DMA_RX_STATUS)); ++ bgmac_err(bgmac, " BGMAC_DMA_RX_ERROR: 0x%08X\n", bgmac_read(bgmac, ring->mmio_base + BGMAC_DMA_RX_ERROR)); ++ ++ dma_sync_single_for_device(dma_dev, ++ slot->dma_addr, ++ BGMAC_RX_BUF_SIZE, ++ DMA_FROM_DEVICE); ++ break; ++ } ++ + /* Check for poison and drop or pass the packet */ + if (len == 0xdead && flags == 0xbeef) { + bgmac_err(bgmac, "Found poisoned packet at slot %d, DMA issue!\n", |