[AMD Official Use Only - Internal Distribution Only]
This workaround is to fix the s3 issue with video playback on raven1 before.
Changfeng, can you have a quick test whether we don't need it right now?
Thanks,
Ray
-Original Message-
From: Zhu, Changfeng
Sent: Wednesday, April 15, 2
[AMD Official Use Only - Internal Distribution Only]
This actually introduced at very early stage we enabled GFXOFF for the first
time on Raven platform. At that time gfxoff can't work with Video play back
(this is general issue across all OSes). So we disabled gfxoff when there is
workload on
[AMD Public Use]
Some good advice on getting ioctls right:
https://www.kernel.org/doc/html/v5.4-preprc-cpu/ioctl/botching-up-ioctls.html
Alex
From: amd-gfx on behalf of Felix
Kuehling
Sent: Tuesday, April 14, 2020 10:40 PM
To: Lin, Amber ; amd-gfx@lists.freede
Hi Amber,
I understand that different processes can get the same FD. My statement
about FD being unique is relative to one process.
The main problem with the global client ID is, that it allows process A
to change the event mask of process B just by specifying process B's
client ID. That can lead
[AMD Official Use Only - Internal Distribution Only]
+Ray
BR,
Changfeng.
-Original Message-
From: Zhu, James
Sent: Tuesday, April 14, 2020 11:00 PM
To: Alex Deucher ; Zhu, James ;
Zhang, Hawking
Cc: amd-gfx list ; Zhu, Changfeng
Subject: Re: [PATCH] drm/amdgpu/vcn: fix gfxoff issue
[AMD Official Use Only - Internal Distribution Only]
Hi Felix,
That was my assumption too that each registration will get different file
descriptor, but it turns out not. When I started two process and both register
gpu0 and gpu1, they both got fd=15. If I have process A register gpu0+gpu1, and
Hey everyone,
This patch broke variable refresh rate in games (all that I've tried so
far... Project CARS 2, DiRT Rally 2.0, Assetto Corsa Competizione) as
well as a simple freesync tester application.
FreeSync tester I've been using: https://github.com/Nixola/VRRTest
I'm not at all familiar wit
Hi Amber,
Some general remarks about the multi-client support. You added a global
client id that's separate from the file descriptor. That's problematic
for two reasons:
1. A process could change a different process' event mask
2. The FD should already be unique per process, no need to invent
When the compute is malfunctioning or performance drops, the system admin
will use SMI (System Management Interface) tool to monitor/diagnostic what
went wrong. This patch provides an event watch interface for the user
space to register devices and subscribe events they are interested. After
regist
[AMD Official Use Only - Internal Distribution Only]
If we're passing the test on the revert, then the only thing that's different
is we're not invalidating HDP and doing a copy to host anymore in
amdgpu_device_vram_access since the function is still called in ttm
access_memory with BAR.
Also
I wouldn't call it premature. Revert is a usual practice when there is a
serious regression that isn't fully understood or root-caused. As far as
I can tell, the problem has been reproduced on multiple systems,
different GPUs, and clearly regressed to Christian's commit. I think
that justifies reve
On Tue, Apr 14, 2020 at 9:05 AM Bernard Zhao wrote:
>
> On some processors, the / operate will call the compiler`s div lib,
> which is low efficient, We can replace the / operation with shift,
> so that we can replace the call of the division library with one
> shift assembly instruction.
>
> Sign
On Tue, Apr 14, 2020 at 4:29 PM Kenny Ho wrote:
>
> On Tue, Apr 14, 2020 at 10:04 AM Daniel Vetter wrote:
> >
> > This has _nothing_ to do with Intel (I think over the past 25 years or
> > so intel has implemented all 4 versions of gpu splitting that I
> > listed, but not entirely sure).
> >
> >
+Hawking
Hi Hawking,
can we drop this WA?
Thanks!
James
On 2020-04-14 10:52 a.m., James Zhu wrote:
+Rex
This is introduce by below patch.
commit 3fded222f4bf7f4c56ef4854872a39a4de08f7a8
Author: Rex Zhu
Date: Fri Jul 27 17:00:02 2018 +0800
drm/amdgpu: Disable gfx off if VCN is busy
+Rex
This is introduce by below patch.
commit 3fded222f4bf7f4c56ef4854872a39a4de08f7a8
Author: Rex Zhu
Date: Fri Jul 27 17:00:02 2018 +0800
drm/amdgpu: Disable gfx off if VCN is busy
this patch is a workaround for the gpu hang
at video begin/end time if gfx off is enabled.
[AMD Official Use Only - Internal Distribution Only]
I think it's premature to push this revert.
With more testing, I'm getting failures from different tests or sometimes none
at all on my machine.
Kent, let's continue the discussion on the original thread.
Thanks,
Jon
From: Koenig, Christia
That's exactly my concern as well.
This looks a bit like the test creates erroneous data somehow, but there
doesn't seems to be a RAS check in the MM data path.
And now that we use the BAR path it goes up in flames.
I just don't see how we can create erroneous data in a test case?
Christian.
[AMD Public Use]
If this causes an issue, any access to vram via the BAR could cause an issue.
Alex
From: amd-gfx on behalf of Russell,
Kent
Sent: Tuesday, April 14, 2020 10:19 AM
To: Koenig, Christian ; amd-gfx@lists.freedesktop.org
Cc: Kuehling, Felix ; Kim
Reviewed-by: Andrey Grodzovsky
Andrey
On 4/14/20 10:22 AM, Kent Russell wrote:
From: Christian König
We are racing to initialize sched->thread here, just always check the
current thread.
Signed-off-by: Christian Koenig
Reviewed-by: Kent Russell
---
drivers/gpu/drm/scheduler/sched_main.c
On Tue, Apr 14, 2020 at 10:04 AM Daniel Vetter wrote:
>
> This has _nothing_ to do with Intel (I think over the past 25 years or
> so intel has implemented all 4 versions of gpu splitting that I
> listed, but not entirely sure).
>
> So again pls less tribal fighting, more collaboration. If you can
On Tue, Apr 14, 2020 at 8:05 AM James Zhu wrote:
>
> Turn off gfxoff control when vcn is gated.
>
> Signed-off-by: James Zhu
> ---
> drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c | 8 +---
> 1 file changed, 5 insertions(+), 3 deletions(-)
>
> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
>
From: Christian König
We are racing to initialize sched->thread here, just always check the
current thread.
Signed-off-by: Christian Koenig
Reviewed-by: Kent Russell
---
drivers/gpu/drm/scheduler/sched_main.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/
[AMD Official Use Only - Internal Distribution Only]
On VG20 or MI100, as soon as we run the subtest, we get the dmesg output below,
and then the kernel ends up hanging. I don't know enough about the test itself
to know why this is occurring, but Jon Kim and Felix were discussing it on a
separa
[AMD Official Use Only - Internal Distribution Only]
Tested-by: changzhu
BR,
Changfeng.
-Original Message-
From: Zhu, James
Sent: Tuesday, April 14, 2020 8:05 PM
To: amd-gfx@lists.freedesktop.org
Cc: Zhu, James ; Zhu, Changfeng
Subject: [PATCH] drm/amdgpu/vcn: fix gfxoff issue
Turn
On Tue, Apr 14, 2020 at 3:50 PM Kenny Ho wrote:
>
> Hi Daniel,
>
> I appreciate many of your review so far and I much prefer keeping
> things technical but that is very difficult to do when I get Intel
> developers calling my implementation "most AMD-specific solution
> possible" and objecting to
Am 13.04.20 um 20:20 schrieb Kent Russell:
This reverts commit c12b84d6e0d70f1185e6daddfd12afb671791b6e.
The original patch causes a RAS event and subsequent kernel hard-hang
when running the KFDMemoryTest.PtraceAccessInvisibleVram on VG20 and
Arcturus
dmesg output at hang time:
[drm] RAS event
Hi Daniel,
I appreciate many of your review so far and I much prefer keeping
things technical but that is very difficult to do when I get Intel
developers calling my implementation "most AMD-specific solution
possible" and objecting to an implementation because their hardware
cannot support it. C
On Tue, Apr 14, 2020 at 3:14 PM Kenny Ho wrote:
>
> Ok. I was hoping you can clarify the contradiction between the
> existance of the spec below and your "not something any other gpu can
> reasonably support" statement. I mean, OneAPI is Intel's spec and
> doesn't that at least make SubDevice su
Ok. I was hoping you can clarify the contradiction between the
existance of the spec below and your "not something any other gpu can
reasonably support" statement. I mean, OneAPI is Intel's spec and
doesn't that at least make SubDevice support "reasonable" for one more
vendor?
Partisanship aside
On Tue, Apr 14, 2020 at 4:21 AM Greg KH wrote:
>
> On Mon, Apr 13, 2020 at 01:48:58PM -0400, Alex Deucher wrote:
> > On Mon, Apr 13, 2020 at 1:47 PM Paul Menzel wrote:
> > >
> > > Dear Prike, dear Alex, dear Linux folks,
> > >
> > >
> > > Am 13.04.20 um 10:44 schrieb Paul Menzel:
> > >
> > > > A
On Mon, Apr 13, 2020 at 01:48:58PM -0400, Alex Deucher wrote:
> On Mon, Apr 13, 2020 at 1:47 PM Paul Menzel wrote:
> >
> > Dear Prike, dear Alex, dear Linux folks,
> >
> >
> > Am 13.04.20 um 10:44 schrieb Paul Menzel:
> >
> > > A regression between causes a system with the AMD board MSI B350M MORT
On Tue, Apr 14, 2020 at 09:00:13AM +0200, Christoph Hellwig wrote:
> On Mon, Apr 13, 2020 at 08:04:10PM -0400, Yan Zhao wrote:
> > > I can't think of another way for a kernel thread to have a mm indeed.
> > for example, before calling to vfio_dma_rw(), a kernel thread has already
> > called use_mm(
On some processors, the / operate will call the compiler`s div lib,
which is low efficient, We can replace the / operation with shift,
so that we can replace the call of the division library with one
shift assembly instruction.
Signed-off-by: Bernard Zhao
---
drivers/gpu/drm/amd/amdgpu/gmc_v6_0.
On Tue, Apr 14, 2020 at 2:47 PM Kenny Ho wrote:
> On Tue, Apr 14, 2020 at 8:20 AM Daniel Vetter wrote:
> > My understanding from talking with a few other folks is that
> > the cpumask-style CU-weight thing is not something any other gpu can
> > reasonably support (and we have about 6+ of those in
Hi Daniel,
On Tue, Apr 14, 2020 at 8:20 AM Daniel Vetter wrote:
> My understanding from talking with a few other folks is that
> the cpumask-style CU-weight thing is not something any other gpu can
> reasonably support (and we have about 6+ of those in-tree)
How does Intel plan to support the Su
On Mon, Apr 13, 2020 at 03:11:36PM -0400, Tejun Heo wrote:
> Hello, Kenny.
>
> On Tue, Mar 24, 2020 at 02:49:27PM -0400, Kenny Ho wrote:
> > Can you elaborate more on what are the missing pieces?
>
> Sorry about the long delay, but I think we've been going in circles for quite
> a while now. Let'
Turn off gfxoff control when vcn is gated.
Signed-off-by: James Zhu
---
drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c | 8 +---
1 file changed, 5 insertions(+), 3 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
index dab34f6..aa9a7a5 10
[AMD Official Use Only - Internal Distribution Only]
Reviewed-by: Hawking Zhang
Regards,
Hawking
From: Clements, John
Sent: Tuesday, April 14, 2020 15:54
To: amd-gfx@lists.freedesktop.org; Zhang, Hawking
Subject: [PATCH] drm/amdgpu: cache smu fw version info
[AMD Official Use Only - Internal
Reviewed-by: Evan Quan
From: amd-gfx On Behalf Of Clements,
John
Sent: Tuesday, April 14, 2020 3:54 PM
To: amd-gfx@lists.freedesktop.org; Zhang, Hawking
Subject: [PATCH] drm/amdgpu: cache smu fw version info
[AMD Official Use Only - Internal Distribution Only]
Submitting patch to save smu f
Hi John,
Please limit this to RAS triggered gpu reset only.
if (adev->in_gpu_reset) {
As for non-RAS triggered gpu reset, smu fw reloading is not needed.
Regards,
Evan
From: amd-gfx On Behalf Of Clements,
John
Sent: Tuesday, April 14, 2020 3:06 PM
To: amd-gfx@lists.freedesktop.org; Zhang, Hawki
[AMD Official Use Only - Internal Distribution Only]
Submitting patch to save smu fw version in local smu context to avoid multiple
erroneous submissions to smu when requesting smu version info
Thank you,
John Clements
0001-drm-amdgpu-cache-smu-fw-version-info.patch
Description: 0001-drm-amdgp
[AMD Official Use Only - Internal Distribution Only]
Series is
Reviewed-by: Hawking Zhang
Regards,
Hawking
From: Clements, John
Sent: Tuesday, April 14, 2020 15:06
To: amd-gfx@lists.freedesktop.org; Zhang, Hawking
Subject: [PATCH] series to refactor psp np fw loading
[AMD Official Use Only
On Mon, Apr 13, 2020 at 08:04:10PM -0400, Yan Zhao wrote:
> > I can't think of another way for a kernel thread to have a mm indeed.
> for example, before calling to vfio_dma_rw(), a kernel thread has already
> called use_mm(), then its current->mm is not null, and it has flag
> PF_KTHREAD.
> in thi
[AMD Official Use Only - Internal Distribution Only]
Submitting patch series to refactor psp np fw loading sequence and set MP1
state to unload in preparation for SMU FW loading during a GPU reset
Thank you,
John Clements
0001-drm-amdgpu-update-psp-fw-loading-sequence.patch
Description: 0001-d
44 matches
Mail list logo