drm/amdgpu: Disable runtime PM for externally attached dGPUs
Subsystems (2)
PATCH 1/1 drm/amdgpu: Disable runtime PM for externally attached dGPUs
pci_is_thunderbolt_attached() requires an upstream PCI bridge with is_thunderbolt set from an Intel Thunderbolt VSEC. This does not cover the affected ASM4242 USB4 PCI hierarchy:
00:02.2
\- 0f:00.0 [1b21:2421]
+- 10:01.0 [1b21:2423] -> 45:00.0 -> 46:00.0
| -> 47:00.0 -> 48:00.0 -> 49:00.0 [1002:7590]
\- 10:03.0 -> 76:00.0 [1b21:2425] USB4 Host RouterThe host router is outside the GPU upstream bridge chain, leaving no ancestor with is_thunderbolt set. PCI core propagates DEVICE_REMOVABLE below the external-facing PCIe tunnel. Disable Runtime PM when either pci_is_thunderbolt_attached() or dev_is_removable() is true.
--- drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 7 +++++++ 1 file changed, 7 insertions(+)
drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +7 -0
@@ -609,6 +609,13 @@ void amdgpu_device_detect_runtime_pm_mode(struct amdgpu_device *adev)int bamaco_support; adev->pm.rpm_mode =AMDGPU_RUNPM_NONE ; +if (pci_is_thunderbolt_attached (adev->pdev) || +dev_is_removable (&adev->pdev->dev)) { +dev_info (adev->dev, +"Runtime PM disabled for externally attached device\n" ); +return ; + } + bamaco_support =amdgpu_device_supports_baco (adev);switch (amdgpu_runtime_pm) {
Signature · 1 line
2.54.0
AMD General
Thanks,
Candice
-----Original Message-----
From: Wang, Yang(Kevin) <[email protected]>
Sent: Wednesday, August 12, 2026 4:09 PM
To: [email protected]
Cc: Deucher, Alexander <[email protected]>; Zhang, Hawking <[email protected]>; Feng, Kenneth <[email protected]>; Li, Candice <[email protected]>
Subject: [PATCH] drm/amdgpu: Disable runtime PM for externally attached dGPUs
pci_is_thunderbolt_attached() requires an upstream PCI bridge with is_thunderbolt set from an Intel Thunderbolt VSEC. This does not cover the affected ASM4242 USB4 PCI hierarchy:
00:02.2
\- 0f:00.0 [1b21:2421]
+- 10:01.0 [1b21:2423] -> 45:00.0 -> 46:00.0
| -> 47:00.0 -> 48:00.0 -> 49:00.0 [1002:7590]
\- 10:03.0 -> 76:00.0 [1b21:2425] USB4 Host RouterThe host router is outside the GPU upstream bridge chain, leaving no ancestor with is_thunderbolt set. PCI core propagates DEVICE_REMOVABLE below the external-facing PCIe tunnel. Disable Runtime PM when either pci_is_thunderbolt_attached() or dev_is_removable() is true.
--- drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 7 +++++++ 1 file changed, 7 insertions(+)
drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +7 -0
@@ -609,6 +609,13 @@ void amdgpu_device_detect_runtime_pm_mode(struct amdgpu_device *adev)int bamaco_support; adev->pm.rpm_mode =AMDGPU_RUNPM_NONE ; +if (pci_is_thunderbolt_attached (adev->pdev) || +dev_is_removable (&adev->pdev->dev)) { +dev_info (adev->dev, +"Runtime PM disabled for externally attached device\n" ); +return ; + } + bamaco_support =amdgpu_device_supports_baco (adev);switch (amdgpu_runtime_pm) {
--
2.54.0
AMD General
-----Original Message-----
From: Wang, Yang(Kevin) <[email protected]>
Sent: Wednesday, August 12, 2026 4:09 PM
To: [email protected]
Cc: Deucher, Alexander <[email protected]>; Zhang, Hawking <[email protected]>; Feng, Kenneth <[email protected]>; Li, Candice <[email protected]>
Subject: [PATCH] drm/amdgpu: Disable runtime PM for externally attached dGPUs
pci_is_thunderbolt_attached() requires an upstream PCI bridge with is_thunderbolt set from an Intel Thunderbolt VSEC. This does not cover the affected ASM4242 USB4 PCI hierarchy:
00:02.2
\- 0f:00.0 [1b21:2421]
+- 10:01.0 [1b21:2423] -> 45:00.0 -> 46:00.0
| -> 47:00.0 -> 48:00.0 -> 49:00.0 [1002:7590]
\- 10:03.0 -> 76:00.0 [1b21:2425] USB4 Host RouterThe host router is outside the GPU upstream bridge chain, leaving no ancestor with is_thunderbolt set. PCI core propagates DEVICE_REMOVABLE below the external-facing PCIe tunnel. Disable Runtime PM when either pci_is_thunderbolt_attached() or dev_is_removable() is true.
--- drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 7 +++++++ 1 file changed, 7 insertions(+)
drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +7 -0
@@ -609,6 +609,13 @@ void amdgpu_device_detect_runtime_pm_mode(struct amdgpu_device *adev)int bamaco_support; adev->pm.rpm_mode =AMDGPU_RUNPM_NONE ; +if (pci_is_thunderbolt_attached (adev->pdev) || +dev_is_removable (&adev->pdev->dev)) { +dev_info (adev->dev, +"Runtime PM disabled for externally attached device\n" ); +return ; + } + bamaco_support =amdgpu_device_supports_baco (adev);switch (amdgpu_runtime_pm) {
--
2.54.0
Quoted 11 lines · Yang Wang
pci_is_thunderbolt_attached()requires an upstream PCI bridge withis_thunderboltset from an Intel Thunderbolt VSEC. This does not cover the affected ASM4242 USB4 PCI hierarchy:00:02.2
\- 0f:00.0 [1b21:2421] +- 10:01.0 [1b21:2423] -> 45:00.0 -> 46:00.0 | -> 47:00.0 -> 48:00.0 -> 49:00.0 [1002:7590] \- 10:03.0 -> 76:00.0 [1b21:2425] USB4 Host Router
Quoted 5 lines
The host router is outside the GPU upstream bridge chain, leaving no ancestor with
is_thunderboltset. PCI core propagatesDEVICE_REMOVABLEbelow the external-facing PCIe tunnel. Disable Runtime PM when eitherpci_is_thunderbolt_attached()ordev_is_removable()is true.
I could see BOCO being an issue, but shouldn't BACO work? That should be self contained to the driver.
Alex
Quoted 26 lines
Signed-off-by Yang Wang <[email protected]>--- drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 7 +++++++ 1 file changed, 7 insertions(+)drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +7
-1@@ -609,6 +609,13 @@ void amdgpu_device_detect_runtime_pm_mode(struct amdgpu_device *adev)int bamaco_support; adev->pm.rpm_mode =AMDGPU_RUNPM_NONE ; +if (pci_is_thunderbolt_attached (adev->pdev) || +dev_is_removable (&adev->pdev->dev)) { +dev_info (adev->dev, +"Runtime PM disabled for externally attached device\n" ); +return ; + } + bamaco_support =amdgpu_device_supports_baco (adev);switch (amdgpu_runtime_pm) { --2.54.0
AMD General
Quoted 24 lines
-----Original Message-----
From: Alex Deucher <[email protected]>
Sent: Wednesday, August 12, 2026 10:30 PM
To: Wang, Yang(Kevin) <[email protected]>
Cc: [email protected]; Deucher, Alexander <[email protected]>; Zhang, Hawking <[email protected]>; Feng, Kenneth <[email protected]>; Li, Candice <[email protected]>
Subject: Re: [PATCH] drm/amdgpu: Disable runtime PM for externally attached dGPUsQuoted 11 lines · Yang Wang
pci_is_thunderbolt_attached()requires an upstream PCI bridge withis_thunderboltset from an Intel Thunderbolt VSEC. This does not cover the affected ASM4242 USB4 PCI hierarchy:00:02.2
\- 0f:00.0 [1b21:2421] +- 10:01.0 [1b21:2423] -> 45:00.0 -> 46:00.0 | -> 47:00.0 -> 48:00.0 -> 49:00.0 [1002:7590] \- 10:03.0 -> 76:00.0 [1b21:2425] USB4 Host Router
Quoted 8 lines
Quoted 5 lines
The host router is outside the GPU upstream bridge chain, leaving no ancestor with
is_thunderboltset. PCI core propagatesDEVICE_REMOVABLEbelow the external-facing PCIe tunnel. Disable Runtime PM when eitherpci_is_thunderbolt_attached()ordev_is_removable()is true.I could see BOCO being an issue, but shouldn't BACO work? That should be self contained to the driver.
We reproduced the failure on the affected ASM4242 USB4 eGPU topology. Before applying the workaround, the driver selected BACO:
amdgpu 0000:49:00.0: Using BACO for runtime pm
amdgpu 0000:49:00.0: Failed to exit BACO state!
The failure therefore occurs in the BACO runtime‑resume path, not in a validated BOCO or _PR3 path.
We do not have an actual test case to verify the BOCO case, but it should be disabled together for safety. Note that the final target state for BOCO is D3cold, while D3hot is an intermediate state.
It could also be problematic if the device cannot fully transition into D3cold, (though this is only my guess and not yet confirmed)
So, should we keep pursuing this issue, or limit our workaround scope to BAMXCO only?
Best Regards,
Kevin
Best Regards,
Kevin
Quoted 31 lines
Alex
drivers/gpu/drm/amd/amdgpu/amdgpu_device.c void
Quoted 11 lines
Signed-off-by Yang Wang <[email protected]>--- drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 7 +++++++ 1 file changed, 7 insertions(+)drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +0
-0@@ -609,6 +609,13 @@ void
amdgpu_device_detect_runtime_pm_mode(struct amdgpu_device *adev)Quoted 16 lines
drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +7
-1int bamaco_support; adev->pm.rpm_mode =AMDGPU_RUNPM_NONE ; +if (pci_is_thunderbolt_attached (adev->pdev) || +dev_is_removable (&adev->pdev->dev)) { +dev_info (adev->dev, +"Runtime PM disabled for externally attached device\n" ); +return ; + } + bamaco_support =amdgpu_device_supports_baco (adev);switch (amdgpu_runtime_pm) { --2.54.0
Quoted 42 lines · Yang(Kevin)
AMD General
Quoted 24 lines
-----Original Message-----
From: Alex Deucher <[email protected]>
Sent: Wednesday, August 12, 2026 10:30 PM
To: Wang, Yang(Kevin) <[email protected]>
Cc: [email protected]; Deucher, Alexander <[email protected]>; Zhang, Hawking <[email protected]>; Feng, Kenneth <[email protected]>; Li, Candice <[email protected]>
Subject: Re: [PATCH] drm/amdgpu: Disable runtime PM for externally attached dGPUsQuoted 11 lines · Yang Wang
pci_is_thunderbolt_attached()requires an upstream PCI bridge withis_thunderboltset from an Intel Thunderbolt VSEC. This does not cover the affected ASM4242 USB4 PCI hierarchy:00:02.2
\- 0f:00.0 [1b21:2421] +- 10:01.0 [1b21:2423] -> 45:00.0 -> 46:00.0 | -> 47:00.0 -> 48:00.0 -> 49:00.0 [1002:7590] \- 10:03.0 -> 76:00.0 [1b21:2425] USB4 Host RouterQuoted 8 lines
Quoted 5 lines
The host router is outside the GPU upstream bridge chain, leaving no ancestor with
is_thunderboltset. PCI core propagatesDEVICE_REMOVABLEbelow the external-facing PCIe tunnel. Disable Runtime PM when eitherpci_is_thunderbolt_attached()ordev_is_removable()is true.I could see BOCO being an issue, but shouldn't BACO work? That should be self contained to the driver.
We reproduced the failure on the affected ASM4242 USB4 eGPU topology. Before applying the workaround, the driver selected BACO:
amdgpu 0000:49:00.0: Using BACO for runtime pm
amdgpu 0000:49:00.0: Failed to exit BACO state!
Quoted 8 lines
The failure therefore occurs in the BACO runtime‑resume path, not in a validated BOCO or
_PR3path.We do not have an actual test case to verify the BOCO case, but it should be disabled together for safety. Note that the final target state for BOCO is D3cold, while D3hot is an intermediate state.
It could also be problematic if the device cannot fully transition into D3cold, (though this is only my guess and not yet confirmed)So, should we keep pursuing this issue, or limit our workaround scope to BAMXCO only?
This change is fine for now as it fixes the issue, but it would be nice to understand if there is a fundamental compatibility issue with BACO and USB4 in general.
Alex
Quoted 37 lines
Best Regards,
KevinBest Regards,
KevinQuoted 31 lines
Alex
drivers/gpu/drm/amd/amdgpu/amdgpu_device.c void
Quoted 11 lines
Signed-off-by Yang Wang <[email protected]>--- drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 7 +++++++ 1 file changed, 7 insertions(+)drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +0
-0@@ -609,6 +609,13 @@ void
amdgpu_device_detect_runtime_pm_mode(struct amdgpu_device *adev)Quoted 16 lines
drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +7
-1int bamaco_support; adev->pm.rpm_mode =AMDGPU_RUNPM_NONE ; +if (pci_is_thunderbolt_attached (adev->pdev) || +dev_is_removable (&adev->pdev->dev)) { +dev_info (adev->dev, +"Runtime PM disabled for externally attached device\n" ); +return ; + } + bamaco_support =amdgpu_device_supports_baco (adev);switch (amdgpu_runtime_pm) { --2.54.0