linux-stable.git/drivers/gpu/drm/amd, branch linux-6.16.y Linux kernel stable tree drm/amdgpu/gfx11: Add Cleaner Shader Support for GFX11.0.1/11.0.4 GPUs 2025-10-12T11:01:02+00:00 Srinivasan Shanmugam srinivasan.shanmugam@amd.com 2025-09-10T06:57:05+00:00 c6dc023a4428e96b55a1fa73e8196abce7d956ef [ Upstream commit c1b6b8c7706354b73196649c46b5e6d4d61c2f5c ] Enable the cleaner shader for additional GFX11.0.1/11.0.4 series GPUs to ensure data isolation among GPU tasks. The cleaner shader is tasked with clearing the Local Data Store (LDS), Vector General Purpose Registers (VGPRs), and Scalar General Purpose Registers (SGPRs), which helps avoid data leakage and guarantees the accuracy of computational results. This update extends cleaner shader support to GFX11.0.1/11.0.4 GPUs, previously available for GFX11.0.3. It enhances security by clearing GPU memory between processes and maintains a consistent GPU state across KGD and KFD workloads. Cc: Wasee Alam <wasee.alam@amd.com> Cc: Mario Sopena-Novales <mario.novales@amd.com> Cc: Christian König <christian.koenig@amd.com> Cc: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com> Acked-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> (cherry picked from commit 0a71ceb27f88a944c2de2808b67b2f46ac75076b) Signed-off-by: Sasha Levin <sashal@kernel.org> 
[ Upstream commit c1b6b8c7706354b73196649c46b5e6d4d61c2f5c ]

Enable the cleaner shader for additional GFX11.0.1/11.0.4 series GPUs to
ensure data isolation among GPU tasks. The cleaner shader is tasked with
clearing the Local Data Store (LDS), Vector General Purpose Registers
(VGPRs), and Scalar General Purpose Registers (SGPRs), which helps avoid
data leakage and guarantees the accuracy of computational results.

This update extends cleaner shader support to GFX11.0.1/11.0.4 GPUs,
previously available for GFX11.0.3. It enhances security by clearing GPU
memory between processes and maintains a consistent GPU state across KGD
and KFD workloads.

Cc: Wasee Alam <wasee.alam@amd.com>
Cc: Mario Sopena-Novales <mario.novales@amd.com>
Cc: Christian König <christian.koenig@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit 0a71ceb27f88a944c2de2808b67b2f46ac75076b)
Signed-off-by: Sasha Levin <sashal@kernel.org>
drm/amdgpu: Enable MES lr_compute_wa by default 2025-10-12T11:01:00+00:00 Mario Limonciello mario.limonciello@amd.com 2025-09-19T00:48:00+00:00 94129770ddbecd670187c45bd20fad6e5cca3e49 commit 1fb710793ce2619223adffaf981b1ff13cd48f17 upstream. The MES set resources packet has an optional bit 'lr_compute_wa' which can be used for preventing MES hangs on long compute jobs. Set this bit by default. Co-developed-by: Yifan Zhang <yifan1.zhang@amd.com> Signed-off-by: Yifan Zhang <yifan1.zhang@amd.com> Acked-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Mario Limonciello <mario.limonciello@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> 
commit 1fb710793ce2619223adffaf981b1ff13cd48f17 upstream.

The MES set resources packet has an optional bit 'lr_compute_wa'
which can be used for preventing MES hangs on long compute jobs.

Set this bit by default.

Co-developed-by: Yifan Zhang <yifan1.zhang@amd.com>
Signed-off-by: Yifan Zhang <yifan1.zhang@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
drm/amd/display: Only restore backlight after amdgpu_dm_init or dm_resume 2025-10-02T11:48:37+00:00 Matthew Schwartz matthew.schwartz@linux.dev 2025-09-11T17:48:51+00:00 3bf00f58a8075fe881890cb21b753105f231713d commit 44b0fed0a5947f54fd14255cd0766df952267bc5 upstream. On clients that utilize AMD_PRIVATE_COLOR properties for HDR support, brightness sliders can include a hardware controlled portion and a gamma-based portion. This is the case on the Steam Deck OLED when using gamescope with Steam as a client. When a user sets a brightness level while HDR is active, the gamma-based portion and/or hardware portion are adjusted to achieve the desired brightness. However, when a modeset takes place while the gamma-based portion is in-use, restoring the hardware brightness level overrides the user's overall brightness level and results in a mismatch between what the slider reports and the display's current brightness. To avoid overriding gamma-based brightness, only restore HW backlight level after boot or resume. This ensures that the backlight level is set correctly after the DC layer resets it while avoiding interference with subsequent modesets. Fixes: 7875afafba84 ("drm/amd/display: Fix brightness level not retained over reboot") Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/4551 Signed-off-by: Matthew Schwartz <matthew.schwartz@linux.dev> Reviewed-by: Mario Limonciello <mario.limonciello@amd.com> Signed-off-by: Mario Limonciello <mario.limonciello@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> (cherry picked from commit a490c8d77d500b5981e739be3d59c60cfe382536) Cc: stable@vger.kernel.org Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> 
commit 44b0fed0a5947f54fd14255cd0766df952267bc5 upstream.

On clients that utilize AMD_PRIVATE_COLOR properties for HDR support,
brightness sliders can include a hardware controlled portion and a
gamma-based portion. This is the case on the Steam Deck OLED when using
gamescope with Steam as a client.

When a user sets a brightness level while HDR is active, the gamma-based
portion and/or hardware portion are adjusted to achieve the desired
brightness. However, when a modeset takes place while the gamma-based
portion is in-use, restoring the hardware brightness level overrides the
user's overall brightness level and results in a mismatch between what
the slider reports and the display's current brightness.

To avoid overriding gamma-based brightness, only restore HW backlight
level after boot or resume. This ensures that the backlight level is
set correctly after the DC layer resets it while avoiding interference
with subsequent modesets.

Fixes: 7875afafba84 ("drm/amd/display: Fix brightness level not retained over reboot")
Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/4551
Signed-off-by: Matthew Schwartz <matthew.schwartz@linux.dev>
Reviewed-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit a490c8d77d500b5981e739be3d59c60cfe382536)
Cc: stable@vger.kernel.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
drm/amd/display: remove output_tf_change flag 2025-10-02T11:48:33+00:00 Melissa Wen mwen@igalia.com 2025-09-01T21:51:05+00:00 c6adf475f375ca1c0fd6c930a2002bf4873ef436 [ Upstream commit 41b1f9fcba62b06195e625bb88c1031102892439 ] Remove this flag as the driver stopped managing it individually since commit a4056c2a6344 ("drm/amd/display: use HW hdr mult for brightness boost"). After some back and forth it was reintroduced as a condition to `set_output_transfer_func()` in [1]. Without direct management, this flag only changes value when all surface update flags are set true on UPDATE_TYPE_FULL with no output TF status meaning. Fixes: bb622e0c0044 ("drm/amd/display: program output tf when required") [1] Signed-off-by: Melissa Wen <mwen@igalia.com> Reviewed-by: Alex Hung <alex.hung@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> (cherry picked from commit 752e6f283ec59ae007aa15a93d5a4b2eefa8cec9) Signed-off-by: Sasha Levin <sashal@kernel.org> 
[ Upstream commit 41b1f9fcba62b06195e625bb88c1031102892439 ]

Remove this flag as the driver stopped managing it individually since
commit a4056c2a6344 ("drm/amd/display: use HW hdr mult for brightness
boost"). After some back and forth it was reintroduced as a condition to
`set_output_transfer_func()` in [1]. Without direct management, this
flag only changes value when all surface update flags are set true on
UPDATE_TYPE_FULL with no output TF status meaning.

Fixes: bb622e0c0044 ("drm/amd/display: program output tf when required") [1]
Signed-off-by: Melissa Wen <mwen@igalia.com>
Reviewed-by: Alex Hung <alex.hung@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit 752e6f283ec59ae007aa15a93d5a4b2eefa8cec9)
Signed-off-by: Sasha Levin <sashal@kernel.org>
amd/amdkfd: correct mem limit calculation for small APUs 2025-10-02T11:48:29+00:00 Yifan Zhang yifan1.zhang@amd.com 2025-08-20T08:10:51+00:00 0bcc5ea4bb30d29ea741c2a4abc963f2752f797b [ Upstream commit 53503556273a5ead8b75534085e2dcb46e96f883 ] Current mem limit check leaks some GTT memory (reserved_for_pt reserved_for_ras + adev->vram_pin_size) for small APUs. Since carveout VRAM is tunable on APUs, there are three case regarding the carveout VRAM size relative to GTT: 1. 0 < carveout < gtt apu_prefer_gtt = true, is_app_apu = false 2. carveout > gtt / 2 apu_prefer_gtt = false, is_app_apu = false 3. 0 = carveout apu_prefer_gtt = true, is_app_apu = true It doesn't make sense to check below limitation in case 1 (default case, small carveout) because the values in the below expression are mixed with carveout and gtt. adev->kfd.vram_used[xcp_id] + vram_needed > vram_size - reserved_for_pt - reserved_for_ras - atomic64_read(&adev->vram_pin_size) gtt: kfd.vram_used, vram_needed, vram_size carveout: reserved_for_pt, reserved_for_ras, adev->vram_pin_size In case 1, vram allocation will go to gtt domain, skip vram check since ttm_mem_limit check already cover this allocation. Signed-off-by: Yifan Zhang <yifan1.zhang@amd.com> Reviewed-by: Mario Limonciello <mario.limonciello@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> (cherry picked from commit fa7c99f04f6dd299388e9282812b14e95558ac8e) Signed-off-by: Sasha Levin <sashal@kernel.org> 
[ Upstream commit 53503556273a5ead8b75534085e2dcb46e96f883 ]

Current mem limit check leaks some GTT memory (reserved_for_pt
reserved_for_ras + adev->vram_pin_size) for small APUs.

Since carveout VRAM is tunable on APUs, there are three case
regarding the carveout VRAM size relative to GTT:

1. 0 < carveout < gtt
   apu_prefer_gtt = true, is_app_apu = false

2. carveout > gtt / 2
   apu_prefer_gtt = false, is_app_apu = false

3. 0 = carveout
   apu_prefer_gtt = true, is_app_apu = true

It doesn't make sense to check below limitation in case 1
(default case, small carveout) because the values in the below
expression are mixed with carveout and gtt.

adev->kfd.vram_used[xcp_id] + vram_needed >
    vram_size - reserved_for_pt - reserved_for_ras -
    atomic64_read(&adev->vram_pin_size)

gtt: kfd.vram_used, vram_needed, vram_size
carveout: reserved_for_pt, reserved_for_ras, adev->vram_pin_size

In case 1, vram allocation will go to gtt domain, skip vram check
since ttm_mem_limit check already cover this allocation.

Signed-off-by: Yifan Zhang <yifan1.zhang@amd.com>
Reviewed-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit fa7c99f04f6dd299388e9282812b14e95558ac8e)
Signed-off-by: Sasha Levin <sashal@kernel.org>
drm/amdkfd: fix p2p links bug in topology 2025-10-02T11:48:28+00:00 Eric Huang jinhuieric.huang@amd.com 2025-08-25T13:50:49+00:00 a01d1325e0fbd64db82b695d36866204f99f56d0 [ Upstream commit ce42a3b581a9db10765eb835840b04dbe7972135 ] When creating p2p links, KFD needs to check XGMI link with two conditions, hive_id and is_sharing_enabled, but it is missing to check is_sharing_enabled, so add it to fix the error. Signed-off-by: Eric Huang <jinhuieric.huang@amd.com> Acked-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> (cherry picked from commit 36cc7d13178d901982da7a122c883861d98da624) Signed-off-by: Sasha Levin <sashal@kernel.org> 
[ Upstream commit ce42a3b581a9db10765eb835840b04dbe7972135 ]

When creating p2p links, KFD needs to check XGMI link
with two conditions, hive_id and is_sharing_enabled,
but it is missing to check is_sharing_enabled, so add
it to fix the error.

Signed-off-by: Eric Huang <jinhuieric.huang@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit 36cc7d13178d901982da7a122c883861d98da624)
Signed-off-by: Sasha Levin <sashal@kernel.org>
drm/amd: Only restore cached manual clock settings in restore if OD enabled 2025-09-25T09:16:49+00:00 Mario Limonciello mario.limonciello@amd.com 2025-09-16T01:59:02+00:00 898aaf78480b0c707b3a0bbe8e2138f08d296b77 commit f9b80514a7227c589291792cb6743b0ddf41c2bc upstream. If OD is not enabled then restoring cached clock settings doesn't make sense and actually leads to errors in resume. Check if enabled before restoring settings. Fixes: 4e9526924d09 ("drm/amd: Restore cached manual clock settings during resume") Reported-by: Jérôme Lécuyer <jerome.4a4c@gmail.com> Closes: https://lore.kernel.org/amd-gfx/0ffe2692-7bfa-4821-856e-dd0f18e2c32b@amd.com/T/#me6db8ddb192626360c462b7570ed7eba0c6c9733 Suggested-by: Jérôme Lécuyer <jerome.4a4c@gmail.com> Acked-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Mario Limonciello <mario.limonciello@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> (cherry picked from commit 1a4dd33cc6e1baaa81efdbe68227a19f51c50f20) Cc: stable@vger.kernel.org Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> 
commit f9b80514a7227c589291792cb6743b0ddf41c2bc upstream.

If OD is not enabled then restoring cached clock settings doesn't make
sense and actually leads to errors in resume.

Check if enabled before restoring settings.

Fixes: 4e9526924d09 ("drm/amd: Restore cached manual clock settings during resume")
Reported-by: Jérôme Lécuyer <jerome.4a4c@gmail.com>
Closes: https://lore.kernel.org/amd-gfx/0ffe2692-7bfa-4821-856e-dd0f18e2c32b@amd.com/T/#me6db8ddb192626360c462b7570ed7eba0c6c9733
Suggested-by: Jérôme Lécuyer <jerome.4a4c@gmail.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit 1a4dd33cc6e1baaa81efdbe68227a19f51c50f20)
Cc: stable@vger.kernel.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
drm/amd/display: Allow RX6xxx & RX7700 to invoke amdgpu_irq_get/put 2025-09-25T09:16:49+00:00 Ivan Lipski ivan.lipski@amd.com 2025-09-02T20:20:09+00:00 20f87640ebe7636aec3dae64ea0503d30915746d commit 29a2f430475357f760679b249f33e7282688e292 upstream. [Why&How] As reported on https://gitlab.freedesktop.org/drm/amd/-/issues/3936, SMU hang can occur if the interrupts are not enabled appropriately, causing a vblank timeout. This patch reverts commit 5009628d8509 ("drm/amd/display: Remove unnecessary amdgpu_irq_get/put"), but only for RX6xxx & RX7700 GPUs, on which the issue was observed. This will re-enable interrupts regardless of whether the user space needed it or not. Fixes: 5009628d8509 ("drm/amd/display: Remove unnecessary amdgpu_irq_get/put") Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/3936 Suggested-by: Sun peng Li <sunpeng.li@amd.com> Reviewed-by: Sun peng Li <sunpeng.li@amd.com> Signed-off-by: Ivan Lipski <ivan.lipski@amd.com> Signed-off-by: Ray Wu <ray.wu@amd.com> Tested-by: Daniel Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> (cherry picked from commit 95d168b367aa28a59f94fc690ff76ebf69312c6d) Cc: stable@vger.kernel.org Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> 
commit 29a2f430475357f760679b249f33e7282688e292 upstream.

[Why&How]
As reported on https://gitlab.freedesktop.org/drm/amd/-/issues/3936,
SMU hang can occur if the interrupts are not enabled appropriately,
causing a vblank timeout.

This patch reverts commit 5009628d8509 ("drm/amd/display: Remove unnecessary
amdgpu_irq_get/put"), but only for RX6xxx & RX7700 GPUs, on which the
issue was observed.

This will re-enable interrupts regardless of whether the user space needed
it or not.

Fixes: 5009628d8509 ("drm/amd/display: Remove unnecessary amdgpu_irq_get/put")
Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/3936
Suggested-by: Sun peng Li <sunpeng.li@amd.com>
Reviewed-by: Sun peng Li <sunpeng.li@amd.com>
Signed-off-by: Ivan Lipski <ivan.lipski@amd.com>
Signed-off-by: Ray Wu <ray.wu@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit 95d168b367aa28a59f94fc690ff76ebf69312c6d)
Cc: stable@vger.kernel.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
drm/amdgpu: suspend KFD and KGD user queues for S0ix 2025-09-25T09:16:49+00:00 Alex Deucher alexander.deucher@amd.com 2025-09-17T16:42:11+00:00 1ff89f5627ef79151971bf2544c240ea5d9ba2e1 commit 9272bb34b066993f5f468b219b4a26ba3f2b25a1 upstream. We need to make sure the user queues are preempted so GFX can enter gfxoff. Reviewed-by: Mario Limonciello (AMD) <superm1@kernel.org> Tested-by: David Perry <david.perry@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> (cherry picked from commit f8b367e6fa1716cab7cc232b9e3dff29187fc99d) Cc: stable@vger.kernel.org Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> 
commit 9272bb34b066993f5f468b219b4a26ba3f2b25a1 upstream.

We need to make sure the user queues are preempted so
GFX can enter gfxoff.

Reviewed-by: Mario Limonciello (AMD) <superm1@kernel.org>
Tested-by: David Perry <david.perry@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit f8b367e6fa1716cab7cc232b9e3dff29187fc99d)
Cc: stable@vger.kernel.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
drm/amdkfd: add proper handling for S0ix 2025-09-25T09:16:49+00:00 Alex Deucher alexander.deucher@amd.com 2025-09-17T16:42:09+00:00 b477c5668ec61d93dc215a1ac970ec4d83d0ed55 commit 2ade36eaa9ac05e4913e9785df19c2cde8f912fb upstream. When in S0i3, the GFX state is retained, so all we need to do is stop the runlist so GFX can enter gfxoff. Reviewed-by: Mario Limonciello (AMD) <superm1@kernel.org> Tested-by: David Perry <david.perry@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> (cherry picked from commit 4bfa8609934dbf39bbe6e75b4f971469384b50b1) Cc: stable@vger.kernel.org Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> 
commit 2ade36eaa9ac05e4913e9785df19c2cde8f912fb upstream.

When in S0i3, the GFX state is retained, so all we need to do
is stop the runlist so GFX can enter gfxoff.

Reviewed-by: Mario Limonciello (AMD) <superm1@kernel.org>
Tested-by: David Perry <david.perry@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit 4bfa8609934dbf39bbe6e75b4f971469384b50b1)
Cc: stable@vger.kernel.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>