linux-stable.git/drivers/gpu/drm/amd/pm/amdgpu_pm.c, branch linux-6.14.y Linux kernel stable tree drm/amd/pm: add unique_id for gfx12 2025-03-18T20:26:53+00:00 Harish Kasiviswanathan Harish.Kasiviswanathan@amd.com 2025-03-11T18:15:18+00:00 19b53f96856b5316ee1fd6ca485af0889e001677 Expose unique_id for gfx12 Signed-off-by: Harish Kasiviswanathan <Harish.Kasiviswanathan@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> (cherry picked from commit 16fbc18cb07470cd33fb5f37ad181b51583e6dc0) Cc: stable@vger.kernel.org # 6.12.x 
Expose unique_id for gfx12

Signed-off-by: Harish Kasiviswanathan <Harish.Kasiviswanathan@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit 16fbc18cb07470cd33fb5f37ad181b51583e6dc0)
Cc: stable@vger.kernel.org # 6.12.x
drm/amd/pm: fix and simplify workload handling 2024-12-02T23:36:15+00:00 Alex Deucher alexander.deucher@amd.com 2024-11-16T13:20:59+00:00 1443dd3c67f6d1a8bd1f810e598e2f0c6f19205c smu->workload_mask is IP specific and should not be messed with in the common code. The mask bits vary across SMU versions. Move all handling of smu->workload_mask in to the backends and simplify the code. Store the user's preference in smu->power_profile_mode which will be reflected in sysfs. For internal driver profile switches for KFD or VCN, just update the workload mask so that the user's preference is retained. Remove all of the extra now unused workload related elements in the smu structure. v2: use refcounts for workload profiles v3: rework based on feedback from Lijo v4: fix the refcount on failure, drop backend mask v5: rework custom handling v6: handle failure cleanup with custom profile v7: Update documentation Reviewed-by: Lijo Lazar <lijo.lazar@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Cc: Kenneth Feng <kenneth.feng@amd.com> Cc: Lijo Lazar <lijo.lazar@amd.com> Cc: stable@vger.kernel.org # 6.11.x 
smu->workload_mask is IP specific and should not be messed with in
the common code. The mask bits vary across SMU versions.

Move all handling of smu->workload_mask in to the backends and
simplify the code.  Store the user's preference in smu->power_profile_mode
which will be reflected in sysfs.  For internal driver profile
switches for KFD or VCN, just update the workload mask so that the
user's preference is retained.  Remove all of the extra now unused
workload related elements in the smu structure.

v2: use refcounts for workload profiles
v3: rework based on feedback from Lijo
v4: fix the refcount on failure, drop backend mask
v5: rework custom handling
v6: handle failure cleanup with custom profile
v7: Update documentation

Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: Kenneth Feng <kenneth.feng@amd.com>
Cc: Lijo Lazar <lijo.lazar@amd.com>
Cc: stable@vger.kernel.org # 6.11.x
drm/amd/pm: add zero RPM stop temperature OD setting support for SMU13 2024-11-05T15:33:47+00:00 Wolfgang Müller wolf@oriole.systems 2024-10-29T11:17:53+00:00 6bfe777e9267ee6d1c4712b52bb5d32e59508a3d Together with the feature to enable or disable zero RPM in the last commit, it also makes sense to expose the OD setting determining under which temperature the fan should stop if zero RPM is enabled. Reviewed-by: Kenneth Feng <kenneth.feng@amd.com> Signed-off-by: Wolfgang Müller <wolf@oriole.systems> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 
Together with the feature to enable or disable zero RPM in the last
commit, it also makes sense to expose the OD setting determining under
which temperature the fan should stop if zero RPM is enabled.

Reviewed-by: Kenneth Feng <kenneth.feng@amd.com>
Signed-off-by: Wolfgang Müller <wolf@oriole.systems>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
drm/amd/pm: add zero RPM OD setting support for SMU13 2024-11-05T15:33:30+00:00 Wolfgang Müller wolf@oriole.systems 2024-10-29T11:17:52+00:00 cfffd980bf21b5a84fd364861d482d5a2ec21c49 Whilst we have support for setting fan curves there is no support for disabling the zero RPM feature. Since the relevant bits are already present in the OverDriveTable, hook them up to a sysctl setting so users can influence this behaviour. Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/3489 Reviewed-by: Kenneth Feng <kenneth.feng@amd.com> Signed-off-by: Wolfgang Müller <wolf@oriole.systems> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 
Whilst we have support for setting fan curves there is no support for
disabling the zero RPM feature. Since the relevant bits are already
present in the OverDriveTable, hook them up to a sysctl setting so users
can influence this behaviour.

Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/3489
Reviewed-by: Kenneth Feng <kenneth.feng@amd.com>
Signed-off-by: Wolfgang Müller <wolf@oriole.systems>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
drm/amd/pm: use pm_runtime_get_if_active for debugfs getters 2024-10-07T18:08:45+00:00 Pierre-Eric Pelloux-Prayer pierre-eric.pelloux-prayer@amd.com 2024-09-16T14:49:15+00:00 6e796cb4a972b385aac369718156aa71220c3c4f Don't wake up the GPU for reading pm values. Instead, take a runtime powermanagement ref when trying to read it if and only if the GPU is already awake. This avoids spurious wake ups (eg: from applets). We use pm_runtime_get_if_in_active because we care about "is the GPU awake?" not about "is the GPU awake and something else prevents suspend?". Tested-by: Mario Limonciello <mario.limonciello@amd.com> Reviewed-by: Kenneth Feng <kenneth.feng@amd.com> Signed-off-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 
Don't wake up the GPU for reading pm values. Instead, take a runtime
powermanagement ref when trying to read it if and only if the GPU is
already awake.

This avoids spurious wake ups (eg: from applets).

We use pm_runtime_get_if_in_active because we care about "is the GPU awake?"
not about "is the GPU awake and something else prevents suspend?".

Tested-by: Mario Limonciello <mario.limonciello@amd.com>
Reviewed-by: Kenneth Feng <kenneth.feng@amd.com>
Signed-off-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
drm/amd/pm: don't update runpm last_usage on debugfs getter 2024-10-07T18:03:45+00:00 Pierre-Eric Pelloux-Prayer pierre-eric.pelloux-prayer@amd.com 2024-09-16T14:35:15+00:00 ef532206a94357de1fbd259364265e7e0758d8af Reading pm values from the GPU shouldn't prevent it to be suspended by resetting the last active timestamp (eg: if an background app monitors GPU sensors every second, it would prevent the autosuspend sequence to trigger). Tested-by: Mario Limonciello <mario.limonciello@amd.com> Reviewed-by: Kenneth Feng <kenneth.feng@amd.com> Signed-off-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 
Reading pm values from the GPU shouldn't prevent it to be suspended
by resetting the last active timestamp (eg: if an background app
monitors GPU sensors every second, it would prevent the autosuspend
sequence to trigger).

Tested-by: Mario Limonciello <mario.limonciello@amd.com>
Reviewed-by: Kenneth Feng <kenneth.feng@amd.com>
Signed-off-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
drm/amd/pm: use pm_runtime_resume_and_get 2024-10-07T18:02:25+00:00 Pierre-Eric Pelloux-Prayer pierre-eric.pelloux-prayer@amd.com 2024-09-06T11:51:39+00:00 ded57e495158175c61f14662741ab4bd613c5bd0 Using pm_runtime_resume_and_get over pm_runtime_get_sync is recommended. Tested-by: Mario Limonciello <mario.limonciello@amd.com> Reviewed-by: Kenneth Feng <kenneth.feng@amd.com> Signed-off-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 
Using pm_runtime_resume_and_get over pm_runtime_get_sync is recommended.

Tested-by: Mario Limonciello <mario.limonciello@amd.com>
Reviewed-by: Kenneth Feng <kenneth.feng@amd.com>
Signed-off-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
drm/amd/pm: fix rpm refcount handling on error 2024-10-07T18:02:05+00:00 Pierre-Eric Pelloux-Prayer pierre-eric.pelloux-prayer@amd.com 2024-07-26T09:42:40+00:00 b0fca3275526db94e64106fa86473bb3a8ea795f pm_runtime_put_autosuspend must be called from all exit paths. Tested-by: Mario Limonciello <mario.limonciello@amd.com> Reviewed-by: Kenneth Feng <kenneth.feng@amd.com> Signed-off-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 
pm_runtime_put_autosuspend must be called from all exit paths.

Tested-by: Mario Limonciello <mario.limonciello@amd.com>
Reviewed-by: Kenneth Feng <kenneth.feng@amd.com>
Signed-off-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Documentation/amdgpu: Add PM policy documentation 2024-05-20T20:20:26+00:00 Lijo Lazar lijo.lazar@amd.com 2024-05-13T06:34:50+00:00 f88e570d4b1fb97dc4d1b27235757cedec7707d3 Add documentation about the newly added pm_policy node in sysfs. Signed-off-by: Lijo Lazar <lijo.lazar@amd.com> Acked-by: Alex Deucher <alexander.deucher@amd.com> Reviewed-by: Asad Kamal <asad.kamal@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 
Add documentation about the newly added pm_policy node in sysfs.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Asad Kamal <asad.kamal@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
drm/amd/pm: Remove legacy interface for xgmi plpd 2024-05-17T21:40:39+00:00 Lijo Lazar lijo.lazar@amd.com 2024-03-04T13:25:27+00:00 546e6309d14cfdc584b92eb6f2274618e629b3f5 Replace the legacy interface with amdgpu_dpm_set_pm_policy to set XGMI PLPD mode. Also, xgmi_plpd_policy sysfs node is not used by any client. Remove that as well. Signed-off-by: Lijo Lazar <lijo.lazar@amd.com> Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com> Reviewed-by: Asad Kamal <asad.kamal@amd.com> Acked-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 
Replace the legacy interface with amdgpu_dpm_set_pm_policy to set XGMI
PLPD mode. Also, xgmi_plpd_policy sysfs node is not used by any client.
Remove that as well.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Asad Kamal <asad.kamal@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>