drm/amdgpu: skip the noirq suspend reset for a switcheroo-parked GPU
Subsystems (3)
PATCH 1/1 drm/amdgpu: skip the noirq suspend reset for a switcheroo-parked GPU
amdgpu_pmops_suspend_noirq() resets the ASIC unconditionally. When the GPU has been parked by vga_switcheroo it has neither power nor a PCIe link, so the reset cannot reach it: pci_set_power_state() reports the device as inaccessible and amdgpu_asic_reset() returns -EINVAL. A failure there aborts the entire noirq suspend phase, and with it the system suspend, so the machine cannot sleep at all while the GPU is switched off.
amdgpu_device_prepare(), amdgpu_device_suspend() and amdgpu_device_resume() all bail out early on DRM_SWITCH_POWER_OFF. This callback was added later, for an unrelated reason, and did not inherit the check. nouveau guards every one of its PM callbacks the same way.
Bail out the same way here. On a single-GPU system switch_power_state is never DRM_SWITCH_POWER_OFF, so this is a no-op there.
Found on a MacBookPro11,5, where the Radeon is powered down through apple-gmux so that the internal panel can be driven by the iGPU instead. Every suspend failed in amdgpu_pmops_suspend_noirq() while the card was off; with this check a full S3 cycle completes.
--- drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c | 8 ++++++++ 1 file changed, 8 insertions(+)
drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c +8 -0
@@ -2664,6 +2664,14 @@ static int amdgpu_pmops_suspend_noirq(struct device *dev)struct amdgpu_device *adev =drm_to_adev (drm_dev);int r; +/* +* A GPU parked by vga_switcheroo has no power and no PCIe link, so the +* ASIC reset below would fail and abort the whole noirq suspend phase. +* Bail out like amdgpu_device_prepare/suspend/resume() already do. +*/ +if (drm_dev->switch_power_state ==DRM_SWITCH_POWER_OFF ) +return 0 ; +if (amdgpu_acpi_should_gpu_reset (adev)) {amdgpu_device_lock_reset_domain (adev->reset_domain); r =amdgpu_asic_reset (adev);
base-commit: 50d05c7c76c96b90462f24debacca971d2e86713
Signature · 1 line
2.55.0
Public
Applied. Thanks!
Quoted 67 lines
-----Original Message-----
From: Theo Andersen Carton <[email protected]>
Sent: Thursday, September 17, 2026 3:15 PM
To: Deucher, Alexander <[email protected]>; Koenig, Christian <[email protected]>
Cc: [email protected]; [email protected]; [email protected]; amd- [email protected]; [email protected]; linux- [email protected]; Theo Andersen Carton <[email protected]>; [email protected]
Subject: [PATCH] drm/amdgpu: skip the noirq suspend reset for a switcheroo- parked GPU
amdgpu_pmops_suspend_noirq()resets the ASIC unconditionally. When the GPU has been parked byvga_switcherooit has neither power nor a PCIe link, so the reset cannot reach it:pci_set_power_state()reports the device as inaccessible andamdgpu_asic_reset()returns-EINVAL. A failure there aborts the entire noirq suspend phase, and with it the system suspend, so the machine cannot sleep at all while the GPU is switched off.
amdgpu_device_prepare(),amdgpu_device_suspend()andamdgpu_device_resume()all bail out early onDRM_SWITCH_POWER_OFF. This callback was added later, for an unrelated reason, and did not inherit the check. nouveau guards every one of its PM callbacks the same way.Bail out the same way here. On a single-GPU system
switch_power_stateis neverDRM_SWITCH_POWER_OFF, so this is a no-op there.Found on a MacBookPro11,5, where the Radeon is powered down through apple-gmux so that the internal panel can be driven by the iGPU instead. Every suspend failed in
amdgpu_pmops_suspend_noirq()while the card was off; with this check a full S3 cycle completes.Fixes 9e051720f9d3 ("drm/amdgpu: Ensure HDA function is suspendedbefore ASIC reset")
Signed-off-by Theo Andersen Carton <[email protected]>--- drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c | 8 ++++++++ 1 file changed, 8 insertions(+)drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c +2
-0@@ -2664,6 +2664,14 @@ static int amdgpu_pmops_suspend_noirq(struct device *dev)struct amdgpu_device *adev =drm_to_adev (drm_dev);int r; +/* +* A GPU parked by vga_switcheroo has no power and no PCIe link, so the
+ * ASIC reset below would fail and abort the whole noirq suspend
phase.drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c +5
-0+* Bail out like amdgpu_device_prepare/suspend/resume() already do. +*/ +if (drm_dev->switch_power_state ==DRM_SWITCH_POWER_OFF ) +return 0 ; +if (amdgpu_acpi_should_gpu_reset (adev)) {amdgpu_device_lock_reset_domain (adev->reset_domain); r =amdgpu_asic_reset (adev);base-commit:
50d05c7c76c96b90462f24debacca971d2e86713--
2.55.0