linux-kernel - Re: [PATCH 0/16] DMA-API debugging facility v2

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20090206160812.GH27684@sequoia.sous-sol.org>
Date:	Fri, 6 Feb 2009 08:08:12 -0800
From:	Chris Wright <chrisw@...s-sol.org>
To:	David Woodhouse <dwmw2@...radead.org>
Cc:	Chris Wright <chrisw@...s-sol.org>,
	Joerg Roedel <joerg.roedel@....com>,
	fujita.tomonori@....ntt.co.jp, netdev@...r.kernel.org,
	iommu@...ts.linux-foundation.org, mingo@...hat.com,
	linux-kernel@...r.kernel.org
Subject: Re: [PATCH 0/16] DMA-API debugging facility v2

* David Woodhouse (dwmw2@...radead.org) wrote:
> What machine did you get that on?

That's a T400 (I'd expect it's same issue as X200).

> Yeah, I saw one of those. If could be a driver bug, of course -- it
> could be unmapping a range before it's actually finished with it. But
> that's unlikely.

One thing I noticed is many of the faults are the page prior to a
coherent range 0x8000 in length, which seems to correspond to tfd buffer.
No obvious off-by-one error anywhere, and not all faults fit that pattern,
so w/out more iwlagn driver knowledge hard to say if that's meaningful
or just mapping coincidence.

> An alternative explanation... The DMA is aborted¹, and the device
> interrupts us to tell us about it at the _same_ time that the IOMMU
> interrupts us to tell us about the fault. We process the device
> interrupt first, unmap that buffer. And then we process the IOMMU
> interrupt... and the buffer is already gone from the list.

I'd have expected the iommu fault to be delivered first, but hey...

> It might be interesting to make this code also remember and print the
> last range that was unmapped, as well as the currently-mapped ranges.

That's what I was thinking too.  Almost need a flight recorder mode to
see if the range was ever mapped/unmapped.

thanks,
-chris
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/