For VF, it doesn't have the permission to clear overflow, clear the bit
by reset.
Signed-off-by: Emily Deng
---
drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c | 10 --
drivers/gpu/drm/amd/amdgpu/amdgpu_ih.h | 1 +
drivers/gpu/drm/amd/amdgpu/ih_v6_0.c | 6 +-
drivers/gpu/drm/amd/amdgpu/ve
This patch adds recommended SDMA engines with limited XGMI SDMA engines.
It will help improve overall performance for device to device copies
with this optimization.
Signed-off-by: Shane Xiao
Suggested-by: Jonathan Kim
---
drivers/gpu/drm/amd/amdkfd/kfd_topology.c | 42 ++-
On 4/9/2025 7:16 PM, SRINIVASAN SHANMUGAM wrote:
On 4/9/2025 7:11 PM, SRINIVASAN SHANMUGAM wrote:
On 4/9/2025 6:45 PM, SRINIVASAN SHANMUGAM wrote:
On 4/9/2025 4:15 PM, Christian König wrote:
This reverts commit c2cc3648ba517a6c270500b5447d5a1efdad5936.
Turned out
that this has some negati
Am 09.04.25 um 15:18 schrieb Ce Sun:
> Checking hive is more readable.
>
> The following smatch warning:
> drivers/gpu/drm/amd/amdgpu/amdgpu_device.c:6820 amdgpu_pci_slot_reset()
> warn: iterator used outside loop: 'tmp_adev'
Please also remove the setting of hive and tmp_adev to NULL at the decla
[AMD Official Use Only - AMD Internal Distribution Only]
Reviewed-by: Tao Zhou
> -Original Message-
> From: amd-gfx On Behalf Of Ce Sun
> Sent: Wednesday, April 9, 2025 8:10 PM
> To: amd-gfx@lists.freedesktop.org
> Cc: dan.carpen...@linaro.org; Sun, Ce(Overlord)
> Subject: [PATCH v1 1/
Am 08.04.25 um 10:23 schrieb ZhenGuo Yin:
> Kernel doorbell BOs needs to be freed before ttm_fini.
>
> Fixes: 54c30d2a8def ("drm/amdgpu: create kernel doorbell pages")
> Signed-off-by: ZhenGuo Yin
At least from my point that patch seems to make a lot of sense, so feel free to
add Reviewed-by: Ch
On Wed, Apr 2, 2025 at 5:23 AM jesse.zh...@amd.com wrote:
>
> From: "jesse.zh...@amd.com"
>
> This patch refactors the SDMA v5.0 queue reset and stop logic to improve
> code readability, maintainability, and performance. The key changes include:
>
> 1. **Generalized `sdma_v5_0_gfx_stop` Function*
Acked-by: Alex Deucher
On Wed, Apr 9, 2025 at 1:06 PM wrote:
>
> From: Roman Li
>
> [Why]
> htmldocs build warning: "Function parameter or struct member 'fused_io'
> not described in 'amdgpu_display_manager'".
>
> [How]
> Add missing description.
>
> Fixes: af632d3f59e6 ("drm/amd/display: HDCP
From: Roman Li
[Why]
htmldocs build warning: "Function parameter or struct member 'fused_io'
not described in 'amdgpu_display_manager'".
[How]
Add missing description.
Fixes: af632d3f59e6 ("drm/amd/display: HDCP Locality check using DMUB Fused IO")
Reported-by: Stephen Rothwell
Signed-off-by:
Hi Dave, Sima,
Fixes for 6.15.
The following changes since commit dce8bd9137b88735dd0efc4e2693213d98c15913:
drm/amdgpu/gfx12: fix num_mec (2025-03-26 17:47:18 -0400)
are available in the Git repository at:
https://gitlab.freedesktop.org/agd5f/linux.git
tags/amd-drm-fixes-6.15-2025-04-09
[Public]
Reviewed-by: Roman Li
Please fix your authorship format as "First_name Last_name
" before merge.
Thanks,
Roman
> -Original Message-
> From: chengjya
> Sent: Wednesday, April 9, 2025 5:17 AM
> To: Li, Roman ; amd-gfx@lists.freedesktop.org; Kaszewski,
> Dominik ; Wheeler, Dani
On 4/8/2025 9:30 PM, Alex Deucher wrote:
Switch from a bool to an enum and allow more options
for enforce isolation. There are now 3 modes of operation:
- Disabled (0)
- Enabled (serialization and cleaner shader) (1)
- Enabled in legacy mode (no serialization or cleaner shader) (2)
This provid
Hi all,
I'm currently debugging a Nouveau issue [1] and potentially might want to
add a function that just checks whether a fence is signaled already –
which then would obviously be called dma_fence_is_signaled().
In any case, I think it is reasonable to rename dma_fence_is_signaled()
so that it
On Wed, Apr 9, 2025 at 10:36 AM SRINIVASAN SHANMUGAM
wrote:
>
>
> On 4/8/2025 9:30 PM, Alex Deucher wrote:
> > Switch from a bool to an enum and allow more options
> > for enforce isolation. There are now 3 modes of operation:
> > - Disabled (0)
> > - Enabled (serialization and cleaner shader) (1
Am 09.04.25 um 16:01 schrieb Philipp Stanner:
> On Wed, 2025-04-09 at 15:14 +0200, Christian König wrote:
>> Am 09.04.25 um 14:56 schrieb Philipp Stanner:
>>> On Wed, 2025-04-09 at 14:51 +0200, Philipp Stanner wrote:
On Wed, 2025-04-09 at 14:39 +0200, Boris Brezillon wrote:
> Hi Philipp,
>
Am 09.04.25 um 15:39 schrieb Ce Sun:
> Checking hive is more readable.
>
> The following smatch warning:
> drivers/gpu/drm/amd/amdgpu/amdgpu_device.c:6820 amdgpu_pci_slot_reset()
> warn: iterator used outside loop: 'tmp_adev'
>
> Fixes: 8ba904f54148 ("drm/amdgpu: Multi-GPU DPC recovery support")
I
Am 09.04.25 um 12:28 schrieb Philipp Stanner:
> On Fri, 2025-03-21 at 16:58 +0100, Christian König wrote:
>> Sometimes drivers need to be able to submit multiple jobs which
>> depend on
>> each other to different schedulers at the same time, but using
>> drm_sched_job_add_dependency() can't fail an
On 4/9/2025 7:09 PM, Ce Sun wrote:
> Checking hive is more readable.
>
> The following smatch warning:
> drivers/gpu/drm/amd/amdgpu/amdgpu_device.c:6820 amdgpu_pci_slot_reset()
> warn: iterator used outside loop: 'tmp_adev'
>
> Fixes: 8ba904f54148 ("drm/amdgpu: Multi-GPU DPC recovery support")
On 4/9/2025 7:11 PM, SRINIVASAN SHANMUGAM wrote:
On 4/9/2025 6:45 PM, SRINIVASAN SHANMUGAM wrote:
On 4/9/2025 4:15 PM, Christian König wrote:
This reverts commit c2cc3648ba517a6c270500b5447d5a1efdad5936. Turned
out
that this has some negative consequences for some workloads. Instead
check
On 4/9/2025 6:45 PM, SRINIVASAN SHANMUGAM wrote:
On 4/9/2025 4:15 PM, Christian König wrote:
This reverts commit c2cc3648ba517a6c270500b5447d5a1efdad5936. Turned out
that this has some negative consequences for some workloads. Instead
check
if the cleaner shader should run directly.
While
Checking hive is more readable.
The following smatch warning:
drivers/gpu/drm/amd/amdgpu/amdgpu_device.c:6820 amdgpu_pci_slot_reset()
warn: iterator used outside loop: 'tmp_adev'
Fixes: 8ba904f54148 ("drm/amdgpu: Multi-GPU DPC recovery support")
Reported-by: Dan Carpenter
Signed-off-by: Ce Sun
On 4/9/2025 4:15 PM, Christian König wrote:
This reverts commit c2cc3648ba517a6c270500b5447d5a1efdad5936. Turned out
that this has some negative consequences for some workloads. Instead check
if the cleaner shader should run directly.
While at it remove amdgpu_vm_need_pipeline_sync(), we also
Checking hive is more readable.
The following smatch warning:
drivers/gpu/drm/amd/amdgpu/amdgpu_device.c:6820 amdgpu_pci_slot_reset()
warn: iterator used outside loop: 'tmp_adev'
Fixes: 8ba904f54148 ("drm/amdgpu: Multi-GPU DPC recovery support")
Reported-by: Dan Carpenter
Signed-off-by: Ce Sun
[AMD Official Use Only - AMD Internal Distribution Only]
Hi Philipp,
I feel like the problem has two parts. The documentation does not make explicit
that DMA_FENCE_FLAG_SIGNALED_BIT is "caching" the hardware state when a fence
is backed by hardware, so what dma_fence_is_signaled here is doing i
dma_fence_is_signaled()'s name strongly reads as if this function were
intended for checking whether a fence is already signaled. Also the
boolean it returns hints at that.
The function's behavior, however, is more complex: it can check with a
driver callback whether the hardware's sequence number
Fixes smatch warning:
drivers/gpu/drm/amd/amdgpu/amdgpu_device.c:6820 amdgpu_pci_slot_reset()
warn: iterator used outside loop: 'tmp_adev'
Fixes: 8ba904f54148 ("drm/amdgpu: Multi-GPU DPC recovery support")
Signed-off-by: Ce Sun
---
drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 2 +-
1 file chan
The documentation of the return value of dma_fence_check_and_signal()
and dma_fence_check_and_signal_locked() reads as if the returned boolean
only describes whether dma_fence_signal() (or similar) has been called
before this function call already. That's not the case, since
dma_fence_ops.signaled(
This reverts commit c2cc3648ba517a6c270500b5447d5a1efdad5936. Turned out
that this has some negative consequences for some workloads. Instead check
if the cleaner shader should run directly.
While at it remove amdgpu_vm_need_pipeline_sync(), we also check again
if the VMID has seen a GPU reset sin
[AMD Official Use Only - AMD Internal Distribution Only]
Hi Dan Carpenter,
Thank you for your review, there is indeed a problem with NULL Pointers, I will
correct this problem immediately
Regards,
Sun,Ce
From: Dan Carpenter
Sent: Wednesday, April 9, 2025 4:
The drm_err function expects a struct drm_device * pointer, so fix it.
Fixes: af632d3f59e6 ("drm/amd/display: HDCP Locality check using DMUB Fused IO")
Signed-off-by: chengjya
---
drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git
Hi,
I've rebased the series on top of drm-next, applied the minor tweaks suggested by Tvrtko on v8 and
the R-b tags. The result can be found on gitlab.fdo:
https://gitlab.freedesktop.org/pepp/linux/-/commits/improve_gpu_scheduler_trace_v9
I believe it's ready to be merged, unless I've missed
On Tue, 8 Apr 2025 at 09:07, Fedor Pchelkin wrote:
>
> > Linus comment is about kvmalloc(), but the code here is using
> > kvmalloc_array() which as far as I know is explicitly made to safely
> > allocate arrays with parameters provided by userspace.
No.
ABSOLUTELY NOTHING CAN ALLOCATE ARRAYS WI
32 matches
Mail list logo
