]> git.baikalelectronics.ru Git - kernel.git/commitdiff
drm/amdgpu: When the VCN(1.0) block is suspended, powergating is explicitly enabled
authorchen gong <curry.gong@amd.com>
Thu, 9 Dec 2021 11:47:10 +0000 (19:47 +0800)
committerAlex Deucher <alexander.deucher@amd.com>
Fri, 17 Dec 2021 17:47:29 +0000 (12:47 -0500)
Play a video on the raven (or PCO, raven2) platform, and then do the S3
test. When resume, the following error will be reported:

amdgpu 0000:02:00.0: [drm:amdgpu_ring_test_helper [amdgpu]] *ERROR* ring
vcn_dec test failed (-110)
[drm:amdgpu_device_ip_resume_phase2 [amdgpu]] *ERROR* resume of IP block
<vcn_v1_0> failed -110
amdgpu 0000:02:00.0: amdgpu: amdgpu_device_ip_resume failed (-110).
PM: dpm_run_callback(): pci_pm_resume+0x0/0x90 returns -110

[why]
When playing the video: The power state flag of the vcn block is set to
POWER_STATE_ON.

When doing suspend: There is no change to the power state flag of the
vcn block, it is still POWER_STATE_ON.

When doing resume: Need to open the power gate of the vcn block and set
the power state flag of the VCN block to POWER_STATE_ON.
But at this time, the power state flag of the vcn block is already
POWER_STATE_ON. The power status flag check in the "8f2cdef drm/amd/pm:
avoid duplicate powergate/ungate setting" patch will return the
amdgpu_dpm_set_powergating_by_smu function directly.
As a result, the gate of the power was not opened, causing the
subsequent ring test to fail.

[how]
In the suspend function of the vcn block, explicitly change the power
state flag of the vcn block to POWER_STATE_OFF.

BugLink: https://gitlab.freedesktop.org/drm/amd/-/issues/1828
Signed-off-by: chen gong <curry.gong@amd.com>
Reviewed-by: Evan Quan <evan.quan@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org
drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c

index d54d720b3cf68307a8533e1704dbc73e816ecdaa..3799226defc0910ec2a2b4a036f146d67a680896 100644 (file)
@@ -246,6 +246,13 @@ static int vcn_v1_0_suspend(void *handle)
 {
        int r;
        struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+       bool idle_work_unexecuted;
+
+       idle_work_unexecuted = cancel_delayed_work_sync(&adev->vcn.idle_work);
+       if (idle_work_unexecuted) {
+               if (adev->pm.dpm_enabled)
+                       amdgpu_dpm_enable_uvd(adev, false);
+       }
 
        r = vcn_v1_0_hw_fini(adev);
        if (r)