netdev
[Top] [All Lists]

Re: Mystery packet killing tg3

To: "David S. Miller" <davem@xxxxxxxxxxxxx>
Subject: Re: Mystery packet killing tg3
From: Andi Kleen <ak@xxxxxx>
Date: Wed, 04 May 2005 20:30:00 +0200
Cc: jgarzik@xxxxxxxxx, netdev@xxxxxxxxxxx
In-reply-to: <20050502200251.38271b61.davem@xxxxxxxxxxxxx> (David S. Miller's message of "Mon, 2 May 2005 20:02:51 -0700")
References: <20050502162405.65dfb4a9@xxxxxxxxxxxxxxxxxxxxx> <20050502200251.38271b61.davem@xxxxxxxxxxxxx>
Sender: netdev-bounce@xxxxxxxxxxx
User-agent: Gnus/5.110002 (No Gnus v0.2) Emacs/21.3 (gnu/linux)
"David S. Miller" <davem@xxxxxxxxxxxxx> writes:
>
> This usually means that there is some DMA corruption.
> For example, some bug in the x86_64 IOMMU code or similar
> causes a bogus DMA address to be fed to the tg3 or even
> worse a DMA mapping is unmapped before tg3 is actually
> done with it.

IOMMU code on x86-64 should be never active unless Stephen
used IOMMU_DEBUG or iommu=force.  THat is because the tg3
is a 64bit capable device and should always use bypass.

> Since AMD promised me an Opteron system last year, but never
> made good on that promise, I've never been able to work on
> fixing this bug myself. :-/

You could just buy one ,-) Lowend A64s boards+cpus are *really* cheap
these days. And it will likely be much quieter than a noisy
original AMD box.  And the IOMMU works on that too if you force
it (normally it is never active because they dont have enough
memory to need remapping) 

-Andi

<Prev in Thread] Current Thread [Next in Thread>