On 5/28/25 09:46, Lazar, Lijo wrote:
> On 5/22/2025 4:10 PM, Samuel Zhang wrote:
>> When switching to new GPU index after hibernation and then resume,
>> VRAM offset of each VRAM BO will be changed, and the cached gpu
>> addresses needed to updated.
>>
>> This is to enable pdb0 and switch to use pd
On 5/28/25 10:37, Prike Liang wrote:
> The kernel driver only requires exporting available rings to the mesa
> when the userq is disabled; otherwise, the userq IP mask will be cleaned
> up in the mesa.
Hui? That doesn't sounds correct to me.
That userq is disable in mesa when kernel queues are av
I've long ago stopped trying to fully understand all the locking in
amdkfd, so maybe this is safe for a contrived reason. It's definitely
not how this should be done. Considers this more a request for a
proper patch.
Cc: Felix Kuehling
Cc: amd-gfx@lists.freedesktop.org
Signed-off-by: Simona Vette
On 5/22/2025 4:10 PM, Samuel Zhang wrote:
> add amdgpu_bo_fb_aper_addr() and update the cached GPU addresses to use
> the FB aperture address for SMU and PSP.
>
> 2 reasons for this change:
> 1. when pdb0 is enabled, gpu addr from amdgpu_bo_create_kernel() is GART
> aperture address, it is not
If we hand out cleared blocks to users, they are expected to write
at least some non-zero values somewhere. If we keep the CLEAR bit set on
the block, amdgpu_fill_buffer will assume there is nothing to do and
incorrectly skip clearing the block. Ultimately, the (still dirty) block
will be reused as
Cleared blocks that are handed out to users after allocation cannot be
presumed to remain cleared. Thus, allocators using drm_buddy need to
dirty all blocks on the allocation success path. Provide a helper for
them to use.
Fixes: 96950929eb232 ("drm/buddy: Implement tracking clear page feature")
C
Hi all,
I've stumbled upon this while investigating why AMDGPU seems to fail at
providing cleared VRAM allocations despite being explicitly asked to
with AMDGPU_GEM_CREATE_VRAM_CLEARED[1].
After some code inspection, I believe the problem is actually much worse
than not providing cleared VRAM. AM
On 5/28/25 11:29, Natalie Vock wrote:
> Hi,
>
> On 5/28/25 09:07, Christian König wrote:
>> On 5/27/25 21:43, Natalie Vock wrote:
>>> If we hand out cleared blocks to users, they are expected to write
>>> at least some non-zero values somewhere. If we keep the CLEAR bit set on
>>> the block, amdgp
On 5/28/25 06:19, Alex Deucher wrote:
> We'll want to include the padding in the wptr count
> for resets.
>
> Signed-off-by: Alex Deucher
> ---
> drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c | 2 ++
> 1 file changed, 2 insertions(+)
>
> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c
> b/drivers
On 5/27/25 21:43, Natalie Vock wrote:
> If we hand out cleared blocks to users, they are expected to write
> at least some non-zero values somewhere. If we keep the CLEAR bit set on
> the block, amdgpu_fill_buffer will assume there is nothing to do and
> incorrectly skip clearing the block. Ultimat
On 5/22/2025 4:10 PM, Samuel Zhang wrote:
> IH is not working after switching a new gpu index for the first time.
>
> The msix table in virtual machine is faked. The real msix table will be
> programmed by QEMU when guest enable/disable msix interrupt. But QEMU
> accessing VF msix table (regist
The kernel driver only requires exporting available rings to the mesa
when the userq is disabled; otherwise, the userq IP mask will be cleaned
up in the mesa.
Signed-off-by: Prike Liang
---
drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c | 20 ++--
1 file changed, 10 insertions(+), 10 de
Add client id the for each userq_mgr which is
created per fd to track which fd is for which client
that could be used in debugfs entry to derive
information like vm and mqd.
Signed-off-by: Sunil Khatri
---
drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c | 1 +
drivers/gpu/drm/amd/amdgpu/amdgpu_userq.h
On 5/22/2025 4:10 PM, Samuel Zhang wrote:
> When switching to new GPU index after hibernation and then resume,
> VRAM offset of each VRAM BO will be changed, and the cached gpu
> addresses needed to updated.
>
> This is to enable pdb0 and switch to use pdb0-based virtual gpu
> address by defaul
On 5/22/2025 4:10 PM, Samuel Zhang wrote:
> For SRIOV VM env with XGMI enabled systems, XGMI physical node id may
> change when hibernate and resume with different VF.
>
> Update XGMI info and vram_base_offset on resume for gfx444 SRIOV env.
> Add amdgpu_virt_xgmi_migrate_enabled() as the featu
On 5/27/2025 9:29 PM, Mario Limonciello wrote:
> For supported ASICs DMCU version is exported, but ASICs that support
> DMCUB there is no information exported to sysfs.
>
> Add an attribute for DMCUB.
>
> Signed-off-by: Mario Limonciello
Reviewed-by: Lijo Lazar
Thanks,
Lijo
> ---
> drive
From: Vitaly Prosyak
This patch updates the cleaner shader, which is responsible for
initializing GPU resources such as Local Data Share (LDS), Vector
General Purpose Registers (VGPRs), and Scalar General Purpose Registers
(SGPRs). Changes include adjustments to register clearing and shader
confi
On 5/28/25 06:19, Alex Deucher wrote:
> Re-emit the unprocessed state after resetting the queue.
>
> Signed-off-by: Alex Deucher
> ---
> drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c | 19 ---
> 1 file changed, 12 insertions(+), 7 deletions(-)
>
> diff --git a/drivers/gpu/drm/amd/amdgp
On 5/28/25 06:19, Alex Deucher wrote:
> Re-emit the unprocessed state after resetting the queue.
I don't think we want any of this for compute queues.
Regards,
Christian.
>
> Signed-off-by: Alex Deucher
> ---
> drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c | 13 +++--
> 1 file changed, 11 ins
Hi Alex,
Thanks for your answer!
On Mon, 26 May 2025 at 23:50 Alex Hung wrote:
> Hi Leonardo,
>
> Thank you for this patch, but unfortunately some unit test suites depend
> on the names.
>
> On 5/21/25 07:58, Leonardo Gomes wrote:
> > From: Leonardo da Silva Gomes
> >
> > Adjust the dcn31_apg
On 5/28/2025 2:59 PM, Natalie Vock wrote:
Hi,
On 5/28/25 09:07, Christian König wrote:
On 5/27/25 21:43, Natalie Vock wrote:
If we hand out cleared blocks to users, they are expected to write
at least some non-zero values somewhere. If we keep the CLEAR bit
set on
the block, amdgpu_fill_b
On Mon, May 26, 2025 at 01:27:28PM +0200, Philipp Stanner wrote:
> On Mon, 2025-05-26 at 13:16 +0200, Christian König wrote:
> > On 5/26/25 11:34, Philipp Stanner wrote:
> > > On Mon, 2025-05-26 at 11:25 +0200, Christian König wrote:
> > > > On 5/23/25 16:16, Danilo Krummrich wrote:
> > > > > On Fr
On 2025-05-28 14:14, Paneer Selvam, Arunpravin wrote:
> On 5/28/2025 2:59 PM, Natalie Vock wrote:
>> On 5/28/25 09:07, Christian König wrote:
>>>
>>> But the problem rather seems to be that we sometimes don't clear the
>>> buffers on release for some reason, but still set it as cleared.
>>
>> Yes
On Mon, May 26, 2025 at 02:54:44PM +0200, Pierre-Eric Pelloux-Prayer wrote:
drivers/gpu/drm/xe/xe_sched_job.c| 3 ++-
diff --git a/drivers/gpu/drm/xe/xe_sched_job.c
b/drivers/gpu/drm/xe/xe_sched_job.c
index f0a6ce610948..5921293b25db 100644
--- a/drivers/gpu/drm/xe/xe_sched_job.c
There is a work-in-progress supplement patch. Let's wait and see whether
it fixes the regressions without reverting 46e68dd5066c.
On 5/27/25 11:06, Alex Hung wrote:
This reverts commit 46e68dd5066c9831e9695c1756db017bb9c3762c since it
breaks color enhancement in another OS, indicating these two
From: Christian König
Stopping the scheduler for queue reset is generally a good idea because
it prevents any worker from touching the ring buffer.
But using amdgpu_fence_driver_force_completion() before restarting it was
a really bad idea because it marked fences as failed while the work was
po
[Public]
>From the code, it looks like you want to hold reference to the process to
>ensure that it doesn't get destroyed while sending the fault event to user. If
>that is correct, then your commit message is not reflecting that.
With commit message updated, this patch is Reviewed-by: Harish K
On 5/28/25 14:39, Michel Dänzer wrote:
> On 2025-05-28 14:14, Paneer Selvam, Arunpravin wrote:
>> On 5/28/2025 2:59 PM, Natalie Vock wrote:
>>> On 5/28/25 09:07, Christian König wrote:
But the problem rather seems to be that we sometimes don't clear the
buffers on release for some r
Make the psp_vbflash and psp_vbflash_status available in sysfs.
Signed-off-by: Shiwu Zhang
---
drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
index dd37264f1ec9..c2f
On Wed, May 28, 2025 at 8:45 AM Simona Vetter wrote:
>
> On Mon, May 26, 2025 at 01:27:28PM +0200, Philipp Stanner wrote:
> > On Mon, 2025-05-26 at 13:16 +0200, Christian König wrote:
> > > On 5/26/25 11:34, Philipp Stanner wrote:
> > > > On Mon, 2025-05-26 at 11:25 +0200, Christian König wrote:
>
Hi,
On 5/28/25 09:07, Christian König wrote:
On 5/27/25 21:43, Natalie Vock wrote:
If we hand out cleared blocks to users, they are expected to write
at least some non-zero values somewhere. If we keep the CLEAR bit set on
the block, amdgpu_fill_buffer will assume there is nothing to do and
inc
On 2025-05-27 21:55, Lang Yu wrote:
> For simulation C models that don't run CP FW where adev->mes.sched_version
> is not populated correctly. This causes NULL dereference in
> amdgpu_amdkfd_free_gtt_mem(dev->adev, (void **)&pqn->q->wptr_bo_gart)
> and warning on unpinned BO in amdgpu_bo_gpu_offset
On Wed, May 28, 2025 at 9:48 AM Christian König
wrote:
>
> On 5/28/25 15:38, Alex Deucher wrote:
> > On Wed, May 28, 2025 at 7:40 AM Christian König
> > wrote:
> >>
> >> On 5/28/25 06:19, Alex Deucher wrote:
> >>> Re-emit the unprocessed state after resetting the queue.
> >>>
> >>> Signed-off-by:
On Wed, May 28, 2025 at 09:29:30AM -0400, Alex Deucher wrote:
> On Wed, May 28, 2025 at 8:45 AM Simona Vetter wrote:
> > I do occasionally find it useful as a record of different approaches
> > considered, which sometimes people fail to adequately cover in their
> > commit messages. Also useful in
On Wed, May 28, 2025 at 04:39:01PM +0200, Danilo Krummrich wrote:
> On Wed, May 28, 2025 at 09:29:30AM -0400, Alex Deucher wrote:
> > On Wed, May 28, 2025 at 8:45 AM Simona Vetter
> > wrote:
> > > I do occasionally find it useful as a record of different approaches
> > > considered, which sometim
Hi,
Le 28/05/2025 à 21:07, Lucas De Marchi a écrit :
On Mon, May 26, 2025 at 02:54:44PM +0200, Pierre-Eric Pelloux-Prayer wrote:
drivers/gpu/drm/xe/xe_sched_job.c | 3 ++-
diff --git a/drivers/gpu/drm/xe/xe_sched_job.c
b/drivers/gpu/drm/xe/xe_sched_job.c
index f0a6ce610948..5921
On 2025-05-28 13:19, James Zhu wrote:
upages is assigned under cpages = 0, so it isn't really used in this function.
Signed-off-by: James Zhu
Reviewed-by: Philip.Yang
---
drivers/gpu/drm/amd/amdkfd/kfd_migrate.c | 1 -
1 file changed, 1 deletion(-)
diff --git a/drivers/gpu/drm/amd/amdkf
On 2025-05-28 13:19, James Zhu wrote:
to get migration pages. When migrating pages from system to vram,
needn't check bit MIGRATE_PFN_VALID, since the system page could
be allocated, but not be accessed.
I think the corner case is vram_pages becomes negative value when
migrating prange from
On 5/28/2025 1:19 AM, Deng, Emily wrote:
[AMD Official Use Only - AMD Internal Distribution Only]
*From:*amd-gfx *On Behalf Of
*Deng, Emily
*Sent:* Monday, May 26, 2025 9:51 AM
*To:* Chen, Xiaogang ;
amd-gfx@lists.freedesktop.org
*Subject:* RE: [PATCH v2] drm/ttm: Should to return the ev
to get migration pages. dst bit MIGRATE_PFN_VALID and src
bit MIGRATE_PFN_MIGRATE should always be set when success.
-v2 use dst to check MIGRATE_PFN_VALID bit(suggested-by philip)
Signed-off-by: James Zhu
---
drivers/gpu/drm/amd/amdkfd/kfd_migrate.c | 22 ++
1 file changed,
On 2025-05-28 14:24, Kasiviswanathan, Harish wrote:
[Public]
From the code, it looks like you want to hold reference to the process to
ensure that it doesn't get destroyed while sending the fault event to user. If
that is correct, then your commit message is not reflecting that.
With commi
[Public]
> -Original Message-
> From: Alex Deucher
> Sent: Wednesday, May 28, 2025 9:11 PM
> To: Liang, Prike
> Cc: amd-gfx@lists.freedesktop.org; Deucher, Alexander
> ; Koenig, Christian
> Subject: Re: [PATCH] drm/amdgpu: only export available rings to mesa for
> enabling
> kq|uq
>
>
[Public]
> -Original Message-
> From: amd-gfx On Behalf Of Liang,
> Prike
> Sent: Thursday, May 29, 2025 9:48 AM
> To: Koenig, Christian ;
> amd-gfx@lists.freedesktop.org
> Cc: Deucher, Alexander
> Subject: RE: [PATCH] drm/amdgpu: only export available rings to mesa for
> enabling
> k
[Public]
> From: Koenig, Christian
> Sent: Wednesday, May 28, 2025 7:21 PM
> To: Liang, Prike ; amd-gfx@lists.freedesktop.org
> Cc: Deucher, Alexander
> Subject: Re: [PATCH] drm/amdgpu: only export available rings to mesa for
> enabling
> kq|uq
>
> On 5/28/25 10:37, Prike Liang wrote:
> > The k
[AMD Official Use Only - AMD Internal Distribution Only]
From: Chen, Xiaogang
Sent: Thursday, May 29, 2025 5:15 AM
To: Deng, Emily ; Zhang, Owen(SRDC)
Cc: amd-gfx@lists.freedesktop.org
Subject: Re: [PATCH v2] drm/ttm: Should to return the evict error
On 5/28/2025 1:19 AM, Deng, Emily wrote:
On 5/28/25 14:30, Simona Vetter wrote:
>> Yup, I've seen that a few times. I think we, the DRM community, should
>> stop that. It's just not useful and makes the commit messages larger,
>> both for the human reader while scrolling, as for the hard drive
>> regarding storage size
>
> I do occasiona
On Wed, May 28, 2025 at 5:30 AM Srinivasan Shanmugam
wrote:
>
> From: Vitaly Prosyak
>
> This patch updates the cleaner shader, which is responsible for
> initializing GPU resources such as Local Data Share (LDS), Vector
> General Purpose Registers (VGPRs), and Scalar General Purpose Registers
>
On 5/28/25 15:38, Alex Deucher wrote:
> On Wed, May 28, 2025 at 7:40 AM Christian König
> wrote:
>>
>> On 5/28/25 06:19, Alex Deucher wrote:
>>> Re-emit the unprocessed state after resetting the queue.
>>>
>>> Signed-off-by: Alex Deucher
>>> ---
>>> drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c | 19 ++
On Wed, May 28, 2025 at 7:45 AM Christian König
wrote:
>
> On 5/28/25 06:19, Alex Deucher wrote:
> > We'll want to include the padding in the wptr count
> > for resets.
> >
> > Signed-off-by: Alex Deucher
> > ---
> > drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c | 2 ++
> > 1 file changed, 2 insertions
On 5/28/25 15:41, Alex Deucher wrote:
> On Wed, May 28, 2025 at 7:45 AM Christian König
> wrote:
>>
>> On 5/28/25 06:19, Alex Deucher wrote:
>>> We'll want to include the padding in the wptr count
>>> for resets.
>>>
>>> Signed-off-by: Alex Deucher
>>> ---
>>> drivers/gpu/drm/amd/amdgpu/amdgpu_i
On Wed, May 28, 2025 at 7:50 AM Christian König
wrote:
>
> On 5/28/25 06:19, Alex Deucher wrote:
> > Re-emit the unprocessed state after resetting the queue.
>
> I don't think we want any of this for compute queues.
Why not? This allows us to do per job resets and can be trivially
extended to al
On Wed, May 28, 2025 at 4:38 AM Prike Liang wrote:
>
> The kernel driver only requires exporting available rings to the mesa
> when the userq is disabled; otherwise, the userq IP mask will be cleaned
> up in the mesa.
The logic should work correctly as is. There are three possible states:
1. KQ
On Wed, May 28, 2025 at 7:40 AM Christian König
wrote:
>
> On 5/28/25 06:19, Alex Deucher wrote:
> > Re-emit the unprocessed state after resetting the queue.
> >
> > Signed-off-by: Alex Deucher
> > ---
> > drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c | 19 ---
> > 1 file changed, 12 in
On Wed, May 28, 2025 at 9:42 AM Christian König
wrote:
>
> On 5/28/25 15:41, Alex Deucher wrote:
> > On Wed, May 28, 2025 at 7:45 AM Christian König
> > wrote:
> >>
> >> On 5/28/25 06:19, Alex Deucher wrote:
> >>> We'll want to include the padding in the wptr count
> >>> for resets.
> >>>
> >>> S
On Wed, May 28, 2025 at 9:57 AM Alex Deucher wrote:
>
> On Wed, May 28, 2025 at 9:48 AM Christian König
> wrote:
> >
> > On 5/28/25 15:38, Alex Deucher wrote:
> > > On Wed, May 28, 2025 at 7:40 AM Christian König
> > > wrote:
> > >>
> > >> On 5/28/25 06:19, Alex Deucher wrote:
> > >>> Re-emit th
upages is assigned under cpages = 0, so it isn't really used in this function.
Signed-off-by: James Zhu
---
drivers/gpu/drm/amd/amdkfd/kfd_migrate.c | 1 -
1 file changed, 1 deletion(-)
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
index 79251
to get migration pages. When migrating pages from system to vram,
needn't check bit MIGRATE_PFN_VALID, since the system page could
be allocated, but not be accessed.
Signed-off-by: James Zhu
---
drivers/gpu/drm/amd/amdkfd/kfd_migrate.c | 25
1 file changed, 13 insertions
It doesn't work reliably and we have soft recover and
full adapter reset so drop this.
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c | 71 ---
1 file changed, 71 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
b/drivers/gpu/drm/am
Re-emit the unprocessed state after resetting the queue.
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c | 25 ++---
1 file changed, 18 insertions(+), 7 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
b/drivers/gpu/drm/amd/amdgpu/gfx_
Enable legacy enforce isolation (just serialize kernel
GC submissions). This way we can reset a ring and
only affect the the process currently using that ring.
This mirrors what windows does.
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c | 9 +
1 file changed, 9
This set improves per queue reset support for GC10+.
When we reset the queue, the queue is lost so we need
to re-emit the unprocessed state from subsequent submissions.
To that end, in order to make sure we actually restore
unprocessed state, we need to enable legacy enforce isolation
so that we ca
Re-emit the unprocessed state after resetting the queue.
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c | 33 +-
1 file changed, 16 insertions(+), 17 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
b/drivers/gpu/drm/amd/amdgpu/gf
We need to know the wptr and sequence number associated
with a job so that we can re-emit the unprocessed state
after a ring reset. Pre-allocate storage space for
the ring buffer contents and add a helper to save off
the unprocessed state so that it can be re-emitted
after the queue is reset.
Sig
Re-emit the unprocessed state after resetting the queue.
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/gfx_v12_0.c | 18 --
1 file changed, 16 insertions(+), 2 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v12_0.c
b/drivers/gpu/drm/amd/amdgpu/gfx_v12_0.c
Re-emit the unprocessed state after resetting the queue.
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c | 33 --
1 file changed, 15 insertions(+), 18 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
b/drivers/gpu/drm/amd/amdgpu/gf
Enable legacy enforce isolation (just serialize kernel
GC submissions). This way we can reset a ring and
only affect the the process currently using that ring.
This mirrors what windows does.
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c | 9 +
1 file changed, 9
It doesn't work reliably and we have soft recover and
full adapter reset so drop this.
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c | 46 ---
1 file changed, 46 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
b/drivers/gpu/drm/am
Re-emit the unprocessed state after resetting the queue.
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c | 19 +--
1 file changed, 17 insertions(+), 2 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
i
It doesn't work reliably and we have soft recover and
full adapter reset so drop this.
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c | 71 ---
1 file changed, 71 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c
b/drivers/gpu/drm/am
Re-emit the unprocessed state after resetting the queue.
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c | 21 ++---
1 file changed, 18 insertions(+), 3 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
b/drivers/gpu/drm/amd/amdgpu/gfx_v9
Re-emit the unprocessed state after resetting the queue.
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c | 18 --
1 file changed, 16 insertions(+), 2 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
Enable legacy enforce isolation (just serialize kernel
GC submissions). This way we can reset a ring and
only affect the the process currently using that ring.
This mirrors what windows does.
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/gfx_v12_0.c | 9 +
1 file changed, 9
Re-emit the unprocessed state after resetting the queue.
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/gfx_v12_0.c | 33 +-
1 file changed, 16 insertions(+), 17 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v12_0.c
b/drivers/gpu/drm/amd/amdgpu/gf
73 matches
Mail list logo
