From: Al Farleigh <AWFour@charter.net>
To: Christoph Hellwig <hch@infradead.org>
Cc: Tom Lendacky <thomas.lendacky@amd.com>, iommu@lists.linux-foundation.org
Subject: Re: x86-64 kernel dma issue; bisected
Date: Thu, 18 Jul 2019 16:26:12 -0400 [thread overview]
Message-ID: <93c60ba2-44d3-25ce-b174-137a9eed7831@charter.net> (raw)
In-Reply-To: <20190718093129.GA25126@infradead.org>
On 7/18/19 5:31 AM, Christoph Hellwig wrote:
> On Tue, Jul 16, 2019 at 10:43:33PM -0400, Al Farleigh wrote:
>> re: the dma-direct code commit below
>>
>> I have bisected the kernel to isolate a PCI board problem on my AMD x86-64
>> ASROCK system. The board worked at (Fedora kernel) 4.18.16 but stopped when
>> moving to (Fedora kernel) 5.0. I then used (git/torvalds/linux) 4.20-rc4 or
>> so to locate the fault via bisect.
>>
>> I now have two kernels, good/bad, that straddle the commit.
>>
>> I was asked to try v5.2 just in case it was fixed; I compiled it and the
>> fault appears to still be present.
>>
>> Simply, mpeg video does not stream from board; no errors, but no video.
>>
>> My work is documented at
>> https://bugs.kde.org/show_bug.cgi?id=408004
>
> I've looked through that log but I'm still not sure what driver
> you are actually using for video capture. Am I right in that your
> are using the saa7134 driver based on the dmesg?
>
> That driver uses a 32-bit DMA mask, so we should be bounce buffering.
>
> We recently had a bug with the AMD memory encryption and the encrypted
> bit, are you using that? The "Freeing unused decrypted memory:" message
> somehow suggests that but I can't find where that would come from by a
> simple grep. If your are using AMD memory encryption or are not sure
> please try the patch below:
>
Hello; I installed the patch on v5.2, verified by eye it got applied,
recompiled, and found no change; that is, the video stream doesn't display.
I also went back and verified my known good v4.20-rc4 kernel still
worked, and my known bad v4.20-rc4 stayed bad. No changes.
I will note that the video has once or twice but extremely rarely had
started on the bad kernels; maybe a timing or race, or somehow being set
up from a previous boot; I try to power off between tests to make sure
system starts from a known, uncontaminated environment.
Regards, Al Farleigh
> ---
>>From 9087c37584fb7d8315877bb55f85e4268cc0b4f4 Mon Sep 17 00:00:00 2001
> From: Tom Lendacky <thomas.lendacky@amd.com>
> Date: Wed, 10 Jul 2019 19:01:19 +0000
> Subject: dma-direct: Force unencrypted DMA under SME for certain DMA masks
>
_______________________________________________
iommu mailing list
iommu@lists.linux-foundation.org
https://lists.linuxfoundation.org/mailman/listinfo/iommu
next prev parent reply other threads:[~2019-07-18 20:26 UTC|newest]
Thread overview: 6+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-07-17 2:43 x86-64 kernel dma issue; bisected Al Farleigh
2019-07-18 9:31 ` Christoph Hellwig
2019-07-18 15:45 ` Lendacky, Thomas
2019-07-18 16:59 ` Al Farleigh
2019-07-18 20:26 ` Al Farleigh [this message]
2019-07-19 13:41 ` Al Farleigh
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=93c60ba2-44d3-25ce-b174-137a9eed7831@charter.net \
--to=awfour@charter.net \
--cc=hch@infradead.org \
--cc=iommu@lists.linux-foundation.org \
--cc=thomas.lendacky@amd.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).