Re: [PATCH RESEND] atl1: fix 4G memory corruption bug

Re: [PATCH RESEND] atl1: fix 4G memory corruption bug

[Jeff Garzik]

Alexey Dobriyan wrote:
> From: Alexey Dobriyan <adobriyan@gmail.com>
> 
> When using 4+ GB RAM and SWIOTLB is active, the driver corrupts
> memory by writing an skb after the relevant DMA page has been
> unmapped.  Although this doesn't happen when *not* using bounce
> buffers, clearing the pointer to the DMA page after unmapping
> it fixes the problem.
> 
> http://marc.info/?t=120861317000005&r=2&w=2
> 
> 
> Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
> Signed-off-by: Jay Cliburn <jacliburn@bellsouth.net>
> ---
> 
>  drivers/net/atlx/atl1.c |    1 +
>  1 files changed, 1 insertions(+), 0 deletions(-)

applied

[PATCH RESEND] atl1: fix 4G memory corruption bug

[Alexey Dobriyan]

From: Alexey Dobriyan <adobriyan@gmail.com>

When using 4+ GB RAM and SWIOTLB is active, the driver corrupts
memory by writing an skb after the relevant DMA page has been
unmapped.  Although this doesn't happen when *not* using bounce
buffers, clearing the pointer to the DMA page after unmapping
it fixes the problem.

http://marc.info/?t=120861317000005&r=2&w=2


Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: Jay Cliburn <jacliburn@bellsouth.net>
---

 drivers/net/atlx/atl1.c |    1 +
 1 files changed, 1 insertions(+), 0 deletions(-)

--- a/drivers/net/atlx/atl1.c
+++ b/drivers/net/atlx/atl1.c
@@ -2027,6 +2027,7 @@ rrd_ok:
 		/* Good Receive */
 		pci_unmap_page(adapter->pdev, buffer_info->dma,
 			       buffer_info->length, PCI_DMA_FROMDEVICE);
+		buffer_info->dma = 0;
 		skb = buffer_info->skb;
 		length = le16_to_cpu(rrd->xsz.xsum_sz.pkt_size);
 

Re: [PATCH RESEND] atl1: fix 4G memory corruption bug

[Jay Cliburn]

On Thu, 22 May 2008 17:20:30 -0400
Jeff Garzik <jeff@garzik.org> wrote:

> Alexey Dobriyan wrote:
> > From: Alexey Dobriyan <adobriyan@gmail.com>
> > 
> > When using 4+ GB RAM and SWIOTLB is active, the driver corrupts
> > memory by writing an skb after the relevant DMA page has been
> > unmapped.  Although this doesn't happen when *not* using bounce
> > buffers, clearing the pointer to the DMA page after unmapping
> > it fixes the problem.
> > 
> > http://marc.info/?t=120861317000005&r=2&w=2
> > 
> > 
> > Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
> > Signed-off-by: Jay Cliburn <jacliburn@bellsouth.net>
> > ---
> > 
> >  drivers/net/atlx/atl1.c |    1 +
> >  1 files changed, 1 insertions(+), 0 deletions(-)
> 
> applied

Hi Jeff,

Where did you apply this patch?  I don't exclude the possibility that I
goofed when I fetched them, but I can't find it in any of the following
branches of your netdev tree.

	master
	net-2.6
	net-next
	upstream
	upstream-davem
	upstream-next
	upstream-next-davem

Thanks,
Jay