From: "Koenig, Christian" <Christian.Koenig@amd.com>
To: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Cc: "Christoph Hellwig" <hch@infradead.org>,
"Alex Deucher" <alexdeucher@gmail.com>,
"Michel Dänzer" <michel@daenzer.net>,
"Maxime Ripard" <maxime.ripard@bootlin.com>,
"Will Deacon" <will.deacon@arm.com>,
"Linux Kernel Mailing List" <linux-kernel@vger.kernel.org>,
"amd-gfx list" <amd-gfx@lists.freedesktop.org>,
"David Airlie" <airlied@linux.ie>,
"Huang, Ray" <Ray.Huang@amd.com>,
dri-devel <dri-devel@lists.freedesktop.org>,
"Michael Ellerman" <mpe@ellerman.id.au>,
"Zhang, Jerry" <Jerry.Zhang@amd.com>,
"Deucher, Alexander" <Alexander.Deucher@amd.com>,
"Sean Paul" <sean@poorly.run>,
linux-arm-kernel <linux-arm-kernel@lists.infradead.org>
Subject: Re: [RFC PATCH] drm: disable WC optimization for cache coherent devices on non-x86
Date: Thu, 24 Jan 2019 09:45:12 +0000 [thread overview]
Message-ID: <953e5e5f-5d47-d6df-40df-c8c94db5447f@amd.com> (raw)
In-Reply-To: <CAKv+Gu_3LurSN-6w1WijPfpJpc+Pgu0YTnwTfMrXMF3-AKAnJA@mail.gmail.com>
Am 24.01.19 um 10:28 schrieb Ard Biesheuvel:
> On Thu, 24 Jan 2019 at 10:25, Koenig, Christian
> <Christian.Koenig@amd.com> wrote:
>> Am 24.01.19 um 10:13 schrieb Christoph Hellwig:
>>> On Wed, Jan 23, 2019 at 05:52:50PM +0100, Ard Biesheuvel wrote:
>>>> But my concern is that it seems likely that non-cache coherent
>>>> implementations are relying on this hack as well. There must be a
>>>> reason that this hack is only disabled for PowerPC platforms if they
>>>> are cache coherent, for instance, and I suspect that that reason is
>>>> that the hack is the only thing ensuring that the CPU mapping
>>>> attributes match the device ones used for these buffers (the vmap()ed
>>>> ones), whereas the rings and other consistent data structures are
>>>> using the DMA API as intended, and thus getting uncached attributes in
>>>> the correct way.
>>> Dave, who added that commit is on Cc together with just about everyone
>>> involved in the review chain. Based on the previous explanation
>>> that idea what we might want an uncached mapping for some non-coherent
>>> architectures for this to work at all makes sense, but then again
>>> the way to create those mappings is entirely architecture specific,
>>> and also need a cache flushing before creating the mapping to work
>>> properly. So my working theory is that this code never properly
>>> worked on architectures without DMA coherent for PCIe at all, but
>>> I'd love to be corrected by concrete examples including an explanation
>>> of how it actually ends up working.
>> Cache coherency is mandatory for modern GPU operation.
>>
>> Otherwise you can't implement a bunch of the requirements of the
>> userspace APIs.
>>
>> In other words the applications doesn't inform the driver that the GPU
>> or the CPU is accessing data, it just does it and assumes that it works.
>>
> Wonderful!
>
> In that case, do you have any objections to the patch proposed by
> Christoph above?
Yeah, the patch of Christoph actually goes way to far cause we have
reports that this works on a bunch of other architectures.
E.g. X86 64bit, PowerPC (under some conditions) and some MIPS.
The only problematic here actually seems to be ARM, so you should
probably just add an "#ifdef .._ARM return false;".
Regards,
Christian.
next prev parent reply other threads:[~2019-01-24 9:45 UTC|newest]
Thread overview: 40+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-01-21 10:06 [RFC PATCH] drm: disable WC optimization for cache coherent devices on non-x86 Ard Biesheuvel
2019-01-21 10:11 ` Koenig, Christian
2019-01-21 15:07 ` Christoph Hellwig
2019-01-21 15:33 ` Ard Biesheuvel
2019-01-21 15:59 ` Christoph Hellwig
2019-01-21 16:14 ` Ard Biesheuvel
2019-01-21 16:22 ` Christoph Hellwig
2019-01-21 16:30 ` Ard Biesheuvel
2019-01-21 16:35 ` Christoph Hellwig
2019-01-21 16:50 ` Ard Biesheuvel
2019-01-21 17:55 ` Michel Dänzer
2019-01-21 17:59 ` Ard Biesheuvel
2019-01-21 18:04 ` Michel Dänzer
2019-01-21 18:20 ` Ard Biesheuvel
2019-01-21 18:23 ` Michel Dänzer
2019-01-21 18:28 ` Ard Biesheuvel
2019-01-21 19:04 ` Michel Dänzer
2019-01-21 19:18 ` Ard Biesheuvel
2019-01-22 8:38 ` Ard Biesheuvel
2019-01-22 20:56 ` Alex Deucher
2019-01-22 21:07 ` Ard Biesheuvel
2019-01-23 7:15 ` Christoph Hellwig
2019-01-23 9:08 ` Ard Biesheuvel
2019-01-23 16:44 ` Christoph Hellwig
2019-01-23 16:52 ` Ard Biesheuvel
2019-01-24 9:13 ` Christoph Hellwig
2019-01-24 9:25 ` Koenig, Christian
2019-01-24 9:28 ` Ard Biesheuvel
2019-01-24 9:45 ` Koenig, Christian [this message]
2019-01-24 9:59 ` Ard Biesheuvel
2019-01-24 11:23 ` Koenig, Christian
2019-01-24 11:26 ` Ard Biesheuvel
2019-01-24 11:37 ` Koenig, Christian
2019-01-24 11:45 ` Ard Biesheuvel
2019-01-24 13:54 ` Alex Deucher
2019-01-24 13:57 ` Ard Biesheuvel
2019-01-24 14:00 ` Alex Deucher
2019-01-24 16:04 ` Michel Dänzer
2019-01-24 9:31 ` Michel Dänzer
2019-01-24 9:37 ` Ard Biesheuvel
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=953e5e5f-5d47-d6df-40df-c8c94db5447f@amd.com \
--to=christian.koenig@amd.com \
--cc=Alexander.Deucher@amd.com \
--cc=Jerry.Zhang@amd.com \
--cc=Ray.Huang@amd.com \
--cc=airlied@linux.ie \
--cc=alexdeucher@gmail.com \
--cc=amd-gfx@lists.freedesktop.org \
--cc=ard.biesheuvel@linaro.org \
--cc=dri-devel@lists.freedesktop.org \
--cc=hch@infradead.org \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=linux-kernel@vger.kernel.org \
--cc=maxime.ripard@bootlin.com \
--cc=michel@daenzer.net \
--cc=mpe@ellerman.id.au \
--cc=sean@poorly.run \
--cc=will.deacon@arm.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).