On 1/31/2025 1:22 PM, Sundararaju, Sathishkumar wrote:
> Hi Lijo,
>
> Gentle ping, reg_dump_fini() is now moved to sw_fini() in this version
> as suggested.
>
Ah, didn't know this was waiting on me. I was not particular about it,
thought Leo's R-B was good enough.
Series is -
Acked-b
Hi,
The initial goal of this series was to improve the drm and amdgpu
trace events to be able to expose more of the inner workings of
the scheduler and drivers to developers via tools.
Then, the series evolved to become focused only on gpu_scheduler.
The changes around vblank events will be part
This will be used in a later commit to trace the drm client_id in
some of the gpu_scheduler trace events.
This requires changing all the users of drm_sched_job_init to
add an extra parameter.
The newly added drm_client_id field in the drm_sched_fence is a bit
of a duplicate of the owner one. One
Split the code on a per instance basis. This will allow
us to use the per instance functions in the future to
handle more things per instance.
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c | 352
1 file changed, 174 insertions(+), 178 deletions
Split the code on a per instance basis. This will allow
us to use the per instance functions in the future to
handle more things per instance.
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c | 417
1 file changed, 206 insertions(+), 211 deletions
Split the code on a per instance basis. This will allow
us to use the per instance functions in the future to
handle more things per instance.
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c | 436 +-
1 file changed, 215 insertions(+), 221 deletions
These series reworks how instances are handled
for VCN to better split the state between instances.
This allows for per instance powergating and opens
the path to further splitting VCN IP blocks
per instance in the future if that is desirable. Per
instance powergating is needed for per engine rese
No need for an IP specific version.
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c | 13 +
1 file changed, 1 insertion(+), 12 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c
b/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c
index ecbf0dd0b6603..5409a877
Rework the code as a vcn instance callback.
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c | 36 ++---
1 file changed, 20 insertions(+), 16 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
i
Split the code on a per instance basis. This will allow
us to use the per instance functions in the future to
handle more things per instance.
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/vcn_v4_0_5.c | 438
1 file changed, 216 insertions(+), 222 deletions
No need for an IP specific version.
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c | 13 +
1 file changed, 1 insertion(+), 12 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c
b/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c
index fdb954c7377eb..f3d41ddd
This is per instance so add a new function pointer for it.
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h | 2 ++
1 file changed, 2 insertions(+)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h
b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h
index 8d34e3814ab19..508fa02
Pass the vcn instance structure to these functions rather
than adev and the instance number.
TODO: clean up the function internals to use the vinst state
directly rather than accessing it indirectly via adev->vcn.inst[].
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c |
Pass the vcn instance structure to these functions rather
than adev and the instance number.
TODO: clean up the function internals to use the vinst state
directly rather than accessing it indirectly via adev->vcn.inst[].
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c | 79
Rework the code as a vcn instance callback.
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c | 25 +
1 file changed, 13 insertions(+), 12 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c
b/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c
i
Pass the instance to the helpers.
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c | 437
drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h | 12 +-
drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c | 16 +-
drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c | 16 +-
drivers/
Use the vcn instance power gating callbacks rather than
the IP powergating callback. This limits power gating to
only the instance in use rather than all of the instances.
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c | 27 -
1 file changed, 13
Pass the vcn instance structure to these functions rather
than adev and the instance number.
TODO: clean up the function internals to use the vinst state
directly rather than accessing it indirectly via adev->vcn.inst[].
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/vcn_v4_0_5.c |
Split the code on a per instance basis. This will allow
us to use the per instance functions in the future to
handle more things per instance.
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/vcn_v5_0_0.c | 386
1 file changed, 190 insertions(+), 196 deletions
Pass the vcn instance structure to these functions rather
than adev and the instance number.
TODO: clean up the function internals to use the vinst state
directly rather than accessing it indirectly via adev->vcn.inst[].
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c | 79
Move more per instance data into the per instance structure.
v2: index instances directly on vcn1.0 and 2.0 to make
it clear that they only support a single instance (Lijo)
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c | 2 +-
drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
Split the code on a per instance basis. This will allow
us to use the per instance functions in the future to
handle more things per instance.
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c | 436 +-
1 file changed, 213 insertions(+), 223 deletions
Store it per instance so we can track it per instance.
v2: index instances directly on vcn1.0 and 2.0 to make
it clear that they only support a single instance (Lijo)
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h | 2 +-
drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c |
Rework the code as a vcn instance callback.
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c | 37 +++
1 file changed, 20 insertions(+), 17 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c
b/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c
index
Split the code on a per instance basis. This will allow
us to use the per instance functions in the future to
handle more things per instance.
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c | 841 +-
1 file changed, 411 insertions(+), 430 deletions
Pass the vcn instance structure to these functions rather
than adev and the instance number.
TODO: clean up the function internals to use the vinst state
directly rather than accessing it indirectly via adev->vcn.inst[].
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/vcn_v5_0_0.c |
Rework the code as a vcn instance callback.
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c | 38 +++
1 file changed, 21 insertions(+), 17 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c
b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c
index
Have a separate work handler for each VCN instance. This
paves the way for per instance VCN power gating at runtime.
v2: index instances directly on vcn1.0 and 2.0 to make
it clear that they only support a single instance (Lijo)
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/amdgpu_
No need for an IP specific version.
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c | 19 +--
1 file changed, 1 insertion(+), 18 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c
b/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c
index 353aab5565702..2f
If each instance uses the same fw image, only store one
copy in the driver.
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c | 30 +
drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h | 4 +++-
drivers/gpu/drm/amd/amdgpu/vcn_v4_0_5.c | 3 +++
3 files changed
No need for an IP specific version.
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c | 27 +
1 file changed, 1 insertion(+), 26 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c
b/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c
index 5ef26
Rework the code as a vcn instance callback.
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/vcn_v5_0_0.c | 33 ++---
1 file changed, 18 insertions(+), 15 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_0.c
b/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_0.c
i
Pass the vcn instance structure to these functions rather
than adev and the instance number.
TODO: clean up the function internals to use the vinst state
directly rather than accessing it indirectly via adev->vcn.inst[].
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c | 85
Rework the code as a vcn instance callback.
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/vcn_v4_0_5.c | 33 ++---
1 file changed, 18 insertions(+), 15 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_5.c
b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_5.c
i
Pass the vcn instance structure to these functions rather
than adev and the instance number.
TODO: clean up the function internals to use the vinst state
directly rather than accessing it indirectly via adev->vcn.inst[].
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c |
It's common for all VCN variants.
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c | 23 +++
drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h | 3 +++
2 files changed, 26 insertions(+)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
b/drivers/gpu/drm/amd
Pass the vcn instance structure to these functions rather
than adev and the instance number.
TODO: clean up the function internals to use the vinst state
directly rather than accessing it indirectly via adev->vcn.inst[].
v2: index instances directly on vcn1.0 and 2.0 to make
it clear that they on
Change it to take a vcn instance rather than adev to align
with the vcn instance changes.
TODO: clean up the function internals to use the vinst state
directly rather than accessing it indirectly via adev->vcn.inst[].
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c | 4
No need for an IP specific version.
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/vcn_v5_0_0.c | 27 +
1 file changed, 1 insertion(+), 26 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_0.c
b/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_0.c
index 95881
No need for an IP specific version.
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/vcn_v4_0_5.c | 27 +
1 file changed, 1 insertion(+), 26 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_5.c
b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_5.c
index 0ac41
No need for an IP specific version.
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c | 27 +
1 file changed, 1 insertion(+), 26 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
index 219e9
No need for an IP specific version.
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c | 27 +--
1 file changed, 1 insertion(+), 26 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c
b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c
index 790b53360
Pass the vcn instance structure to these functions rather
than adev and the instance number.
TODO: clean up the function internals to use the vinst state
directly rather than accessing it indirectly via adev->vcn.inst[].
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c | 11
To properly handle multiple GPUs.
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
index edbcb11c382a2..8d8b39e6d197a
No need for an IP specific version.
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c | 21 ++---
1 file changed, 2 insertions(+), 19 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c
b/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c
index f5889d1bd0d4c.
Rework the code as a vcn instance callback.
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c | 39 ++-
1 file changed, 20 insertions(+), 19 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c
b/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c
index
Rework the code as a vcn instance callback.
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c | 32 ++-
1 file changed, 22 insertions(+), 10 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c
b/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c
index
Rework the code as a vcn instance callback.
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c | 35 ++-
1 file changed, 24 insertions(+), 11 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c
b/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c
index
Am 31.01.25 um 17:55 schrieb Sundararaju, Sathishkumar:
On 1/31/2025 10:19 PM, Christian König wrote:
Am 31.01.25 um 17:40 schrieb Sundararaju, Sathishkumar:
Hi Christian,
On 1/31/2025 9:56 PM, Christian König wrote:
Am 31.01.25 um 17:23 schrieb Sathishkumar S:
Add ring reset function callb
On Fri, Jan 31, 2025 at 11:32 AM Sathishkumar S
wrote:
>
> Add ring reset function callback for JPEG4_0_3 to
> recover from job timeouts without a full gpu reset.
>
> Signed-off-by: Sathishkumar S
> ---
> drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c | 60 ++--
> 1 file changed, 5
I can't see patch #1 in my inbox for some reason, but I already know
what it does from your repository.
Feel free to add Reviewed-by: Christian König
to the entire series.
Regards,
Christian.
Am 31.01.25 um 12:02 schrieb Pierre-Eric Pelloux-Prayer:
Hi,
The initial goal of this series was
Add helper functions to handle per-instance and per-core
initialization and deinitialization in JPEG4_0_3.
Signed-off-by: Sathishkumar S
---
drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c | 190 ---
1 file changed, 98 insertions(+), 92 deletions(-)
diff --git a/drivers/gpu/drm/amd
Hi Alex,
On 1/31/2025 10:42 PM, Alex Deucher wrote:
On Fri, Jan 31, 2025 at 11:32 AM Sathishkumar S
wrote:
Add ring reset function callback for JPEG4_0_3 to
recover from job timeouts without a full gpu reset.
Signed-off-by: Sathishkumar S
---
drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c | 60
This patch (in combination with the accompanying CRIU patch)
allows the amdgpu CRIU plugin to support dmabuf IPC.
It includes
- A new amdgpu ioctl (amdgpu_criu_op_ioctl), which has similar
options to kfd_ioctl_criu, and accompanying structs.
- New "is_retry" field in amdkfd CRIU ioctls, to
We are working on extending support for CRIU checkpoint/restore in the amdgpu
driver to support new use cases and ROCm applications increasingly using
render node ioctls for memory management. In the longer term this may also
allow checkpoint/restore of graphics application. With this patch series
Same as amdgpu_gfx_off_ctrl(), but without the delay
for gfxoff disallow.
Suggested-by: Błażej Szczygieł
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c | 53 +++--
drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h | 1 +
2 files changed, 41 insertions(+), 13
Bump the driver version for RV/PCO compute stability fix
so mesa can use this check to enable compute queues on
RV/PCO.
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_
Use amdgpu_gfx_off_ctrl_immediate() when powergating.
There's no need for the delay in gfx off allow. The
powergating is dynamically disabled/enabled as for
RV/PCO on compute queues and allowing gfx off again as
soon the job is submitted improves power savings.
Suggested-by: Błażej Szczygieł
Lin
When mesa started using compute queues more often
we started seeing additional hangs with compute queues.
Disabling gfxoff seems to mitigate that. Manually
control gfxoff and gfx pg with command submissions to avoid
any issues related to gfxoff. KFD already does the same
thing for these chips.
v
Hi Christian,
On 1/31/2025 9:56 PM, Christian König wrote:
Am 31.01.25 um 17:23 schrieb Sathishkumar S:
Add ring reset function callback for JPEG4_0_3 to
recover from job timeouts without a full gpu reset.
Signed-off-by: Sathishkumar S
---
drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c | 60
On 2025-01-30 15:51, Alex Deucher
wrote:
If the user has configured a large carveout on a small APU,
only use GTT for VRAM allocations if GTT is larger than
VRAM.
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 6 --
1 file ch
Am 31.01.25 um 17:40 schrieb Sundararaju, Sathishkumar:
Hi Christian,
On 1/31/2025 9:56 PM, Christian König wrote:
Am 31.01.25 um 17:23 schrieb Sathishkumar S:
Add ring reset function callback for JPEG4_0_3 to
recover from job timeouts without a full gpu reset.
Signed-off-by: Sathishkumar S
From: Xiaogang Chen
When register a vm range at svm the added vm range may be split into multiple
subranges and/or existing pranges got spitted. The new pranges need validated
and mapped. This patch changes error handling for pranges that fail updating:
1: free prange resources and remove it fro
On 1/31/2025 10:19 PM, Christian König wrote:
Am 31.01.25 um 17:40 schrieb Sundararaju, Sathishkumar:
Hi Christian,
On 1/31/2025 9:56 PM, Christian König wrote:
Am 31.01.25 um 17:23 schrieb Sathishkumar S:
Add ring reset function callback for JPEG4_0_3 to
recover from job timeouts withou
On Fri, Jan 31, 2025 at 8:02 AM Lijo Lazar wrote:
>
> JPEG 5.0.1 supports upto 10 rings, however PMFW support for SMU v13.0.6
> variants is now limited to 8 per instance. Limit to 8 temporarily to
> avoid out of bounds access.
>
> Signed-off-by: Lijo Lazar
Acked-by: Alex Deucher
> ---
> drive
Add ring reset function callback for JPEG4_0_3 to
recover from job timeouts without a full gpu reset.
Signed-off-by: Sathishkumar S
---
drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c | 60 ++--
1 file changed, 57 insertions(+), 3 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgp
Add ring reset function callback for JPEG4_0_0 to
recover from job timeouts without a full gpu reset.
Signed-off-by: Sathishkumar S
---
drivers/gpu/drm/amd/amdgpu/jpeg_v4_0.c | 22 +++---
1 file changed, 19 insertions(+), 3 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/jp
Add helper functions to handle per-instance initialization
and deinitialization in JPEG2_5_0.
Signed-off-by: Sathishkumar S
---
drivers/gpu/drm/amd/amdgpu/jpeg_v2_5.c | 102 +
1 file changed, 55 insertions(+), 47 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_
Add ring reset function callback for JPEG3_0_0 to
recover from job timeouts without a full gpu reset.
Signed-off-by: Sathishkumar S
---
drivers/gpu/drm/amd/amdgpu/jpeg_v3_0.c | 22 ++
1 file changed, 22 insertions(+)
diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v3_0.c
b/dri
Add ring reset function callback for JPEG2_5_0 to
recover from job timeouts without a full gpu reset.
Signed-off-by: Sathishkumar S
---
drivers/gpu/drm/amd/amdgpu/jpeg_v2_5.c | 23 +++
1 file changed, 23 insertions(+)
diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v2_5.c
b/dr
This patch-series adds jpeg ring reset callback function
to recover from job timeouts without a full gpu reset.
How is it tested:-
A bad job on jpeg causes timeout and recovers alongside
an unaffected mpv vaapi-decode(vcn) in playback loop.
Sathishkumar S (7):
drm/amdgpu: Per-instance init fu
Am 31.01.25 um 17:23 schrieb Sathishkumar S:
Add ring reset function callback for JPEG4_0_3 to
recover from job timeouts without a full gpu reset.
Signed-off-by: Sathishkumar S
---
drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c | 60 ++--
1 file changed, 57 insertions(+), 3 del
Add ring reset function callback for JPEG2_0_0 to
recover from job timeouts without a full gpu reset.
Signed-off-by: Sathishkumar S
---
drivers/gpu/drm/amd/amdgpu/jpeg_v2_0.c | 22 ++
1 file changed, 22 insertions(+)
diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v2_0.c
b/dri
On 1/31/2025 10:35 PM, Christian König wrote:
Am 31.01.25 um 17:55 schrieb Sundararaju, Sathishkumar:
On 1/31/2025 10:19 PM, Christian König wrote:
Am 31.01.25 um 17:40 schrieb Sundararaju, Sathishkumar:
Hi Christian,
On 1/31/2025 9:56 PM, Christian König wrote:
Am 31.01.25 um 17:23 sch
New features require the new fields defines
Signed-off-by: Shaoyun Liu
---
drivers/gpu/drm/amd/include/mes_v11_api_def.h | 46 ++-
drivers/gpu/drm/amd/include/mes_v12_api_def.h | 34 +-
2 files changed, 78 insertions(+), 2 deletions(-)
diff --git a/drivers/gpu/drm/am
[AMD Official Use Only - AMD Internal Distribution Only]
Sorry, please ignore this one . sent wrong patch.
Regards
Shaoyun.liu
-Original Message-
From: Liu, Shaoyun
Sent: Friday, January 31, 2025 10:35 PM
To: amd-gfx@lists.freedesktop.org
Cc: Liu, Shaoyun
Subject: [PATCH] drm/amd : Upd
1. MES fence_value will be updated in fence_addr if API success, otherwise
upper 32 bit will be used to indicate error code. In any case, MES will
trigger an EOP interrupt with 0xb1 as context id in the interrupt cookie
2. Add RRMT option support which will be used for remote die register acc
If a device supports runtime pm, then pm_runtime_get_if_active returns 0
if a device is not active and 1 if already active. However, if a device
doesn't support runtime pm, the API returns -EINVAL. A device not
supporting runtime pm implies it's not affected by runtime pm and it's
active. Hence no
JPEG 5.0.1 supports upto 10 rings, however PMFW support for SMU v13.0.6
variants is now limited to 8 per instance. Limit to 8 temporarily to
avoid out of bounds access.
Signed-off-by: Lijo Lazar
---
drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c | 8 +---
1 file changed, 5 insertions(+
On Fri, Jan 31, 2025 at 12:13 PM Philip Yang wrote:
>
>
> On 2025-01-30 15:51, Alex Deucher wrote:
>
> If the user has configured a large carveout on a small APU,
> only use GTT for VRAM allocations if GTT is larger than
> VRAM.
>
> Signed-off-by: Alex Deucher
> ---
> drivers/gpu/drm/amd/amdgpu/
If the user has configured a large carveout on a small APU,
only use GTT for VRAM allocations if GTT is larger than
VRAM.
v2: fix reversed check (Philip)
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 6 --
1 file changed, 4 insertions(+), 2 deletions(-)
diff --g
On big and small APUs we send KFD VRAM allocations to GTT
since the carve out is either non-existent or relatively
small. However, if someone sets the carve out size to be
relatively large, we may end up using GTT rather than VRAM.
No change of logic with this patch, but it allows the
driver to d
FRAME_WARN), $(frame_warn_limit)),y)
+frame_warn_flag := -Wframe-larger-than=$(frame_warn_limit)
+endif
endif
subdir-ccflags-y += -I$(FULL_AMD_DISPLAY_PATH)/dc/dml2
---
base-commit: 7f2b5237e313e39008a85b33ca94ab503a8fdff9
change-id: 20250131-amdgpu-respect-config_frame_warn-739a9b24496e
Be
Hi Alex, I missed to see your other comment below, answered it inline.
On 1/31/2025 11:11 PM, Sundararaju, Sathishkumar wrote:
Hi Alex,
On 1/31/2025 10:42 PM, Alex Deucher wrote:
On Fri, Jan 31, 2025 at 11:32 AM Sathishkumar S
wrote:
Add ring reset function callback for JPEG4_0_3 to
recover
84 matches
Mail list logo
