[PATCH 2/2] drm/amgpu: Check return value of amdgpu_device_baco_enter/exit
Ma, Jun
majun at amd.com
Thu Feb 29 02:53:58 UTC 2024
On 2/28/2024 7:58 PM, Lazar, Lijo wrote:
>
>
> On 2/28/2024 5:14 PM, Ma Jun wrote:
>> Check return value of amdgpu_device_baco_enter/exit and print
>> warning message because these errors may cause runtime resume failure
>>
>> Signed-off-by: Ma Jun <Jun.Ma2 at amd.com>
>> ---
>> drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c | 8 ++++++--
>> 1 file changed, 6 insertions(+), 2 deletions(-)
>>
>> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
>> index 8b83c26dc018..11e089cfb465 100644
>> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
>> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
>> @@ -2705,7 +2705,9 @@ static int amdgpu_pmops_runtime_suspend(struct device *dev)
>> } else if (adev->pm.rpm_mode == AMDGPU_RUNPM_BOCO) {
>> /* nothing to do */
>> } else if (adev->pm.rpm_mode == AMDGPU_RUNPM_BACO) {
>> - amdgpu_device_baco_enter(drm_dev);
>> + ret = amdgpu_device_baco_enter(drm_dev);
>> + if (ret)
>> + dev_warn(&pdev->dev, "warning: device fails to enter baco. ret=%d\n", ret);
>> }
>
> If the intention is to print for any baco entry/exit failure, it's
> better to put the print inside enter/exit functions itself. Also, since
> these are triggered through PMFW, any PMFW fail will be printed. So this
> may not be that useful.
>
Thanks, I will move it into enter/exit functions.
This check is mainly used for other error cases not for PMFW failure.
Regards,
Ma Jun
> Thanks,
> Lijo
>
>>
>> dev_dbg(&pdev->dev, "asic/device is runtime suspended\n");
>> @@ -2745,7 +2747,9 @@ static int amdgpu_pmops_runtime_resume(struct device *dev)
>> */
>> pci_set_master(pdev);
>> } else if (adev->pm.rpm_mode == AMDGPU_RUNPM_BACO) {
>> - amdgpu_device_baco_exit(drm_dev);
>> + ret = amdgpu_device_baco_exit(drm_dev);
>> + if (ret)
>> + dev_warn(&pdev->dev, "warning: device fails to exit from baco. ret=%d\n", ret);
>> }
>> ret = amdgpu_device_resume(drm_dev, false);
>> if (ret) {
More information about the amd-gfx
mailing list