diff options
| author | Manfred Spraul <manfred@colorfullife.com> | 2005-12-24 14:19:24 +0100 |
|---|---|---|
| committer | Linus Torvalds <torvalds@g5.osdl.org> | 2005-12-24 20:53:32 +0100 |
| commit | 1836098f97d22c81652aeeec64d1819dc2177bdb (patch) | |
| tree | 1bd940fb161aea432af117543e936ca9b677c985 /drivers | |
| parent | Merge master.kernel.org:/pub/scm/linux/kernel/git/davem/sparc-2.6 (diff) | |
| download | linux-1836098f97d22c81652aeeec64d1819dc2177bdb.tar.xz linux-1836098f97d22c81652aeeec64d1819dc2177bdb.zip | |
[PATCH] forcedeth: fix random memory scribbling bug
Two critical bugs were found in forcedeth 0.47:
- TSO doesn't work.
- pci_map_single() for the rx buffers is called with size==0. This bug
is critical, it causes random memory corruptions on systems with an
iommu.
Below is a minimal fix for both bugs, for 2.6.15.
TSO will be fixed properly in the next version. Tested on x86-64.
Signed-Off-By: Manfred Spraul <manfred@colorfullife.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Diffstat (limited to 'drivers')
| -rw-r--r-- | drivers/net/forcedeth.c | 15 |
1 files changed, 8 insertions, 7 deletions
diff --git a/drivers/net/forcedeth.c b/drivers/net/forcedeth.c index 525624fc03b4..c39344adecce 100644 --- a/drivers/net/forcedeth.c +++ b/drivers/net/forcedeth.c @@ -10,7 +10,7 @@ * trademarks of NVIDIA Corporation in the United States and other * countries. * - * Copyright (C) 2003,4 Manfred Spraul + * Copyright (C) 2003,4,5 Manfred Spraul * Copyright (C) 2004 Andrew de Quincey (wol support) * Copyright (C) 2004 Carl-Daniel Hailfinger (invalid MAC handling, insane * IRQ rate fixes, bigendian fixes, cleanups, verification) @@ -100,6 +100,7 @@ * 0.45: 18 Sep 2005: Remove nv_stop/start_rx from every link check * 0.46: 20 Oct 2005: Add irq optimization modes. * 0.47: 26 Oct 2005: Add phyaddr 0 in phy scan. + * 0.48: 24 Dec 2005: Disable TSO, bugfix for pci_map_single * * Known bugs: * We suspect that on some hardware no TX done interrupts are generated. @@ -111,7 +112,7 @@ * DEV_NEED_TIMERIRQ will not harm you on sane hardware, only generating a few * superfluous timer interrupts from the nic. */ -#define FORCEDETH_VERSION "0.47" +#define FORCEDETH_VERSION "0.48" #define DRV_NAME "forcedeth" #include <linux/module.h> @@ -871,8 +872,8 @@ static int nv_alloc_rx(struct net_device *dev) } else { skb = np->rx_skbuff[nr]; } - np->rx_dma[nr] = pci_map_single(np->pci_dev, skb->data, skb->len, - PCI_DMA_FROMDEVICE); + np->rx_dma[nr] = pci_map_single(np->pci_dev, skb->data, + skb->end-skb->data, PCI_DMA_FROMDEVICE); if (np->desc_ver == DESC_VER_1 || np->desc_ver == DESC_VER_2) { np->rx_ring.orig[nr].PacketBuffer = cpu_to_le32(np->rx_dma[nr]); wmb(); @@ -999,7 +1000,7 @@ static void nv_drain_rx(struct net_device *dev) wmb(); if (np->rx_skbuff[i]) { pci_unmap_single(np->pci_dev, np->rx_dma[i], - np->rx_skbuff[i]->len, + np->rx_skbuff[i]->end-np->rx_skbuff[i]->data, PCI_DMA_FROMDEVICE); dev_kfree_skb(np->rx_skbuff[i]); np->rx_skbuff[i] = NULL; @@ -1334,7 +1335,7 @@ static void nv_rx_process(struct net_device *dev) * the performance. */ pci_unmap_single(np->pci_dev, np->rx_dma[i], - np->rx_skbuff[i]->len, + np->rx_skbuff[i]->end-np->rx_skbuff[i]->data, PCI_DMA_FROMDEVICE); { @@ -2455,7 +2456,7 @@ static int __devinit nv_probe(struct pci_dev *pci_dev, const struct pci_device_i np->txrxctl_bits |= NVREG_TXRXCTL_RXCHECK; dev->features |= NETIF_F_HW_CSUM | NETIF_F_SG; #ifdef NETIF_F_TSO - dev->features |= NETIF_F_TSO; + /* disabled dev->features |= NETIF_F_TSO; */ #endif } |