Skip to contentSkip to conversation

SWDEV-206718 drm/amdgpu: Fix tdr3 could hang with slow compute issue

PATCH In review Emily Deng 2 participants 3 messages Oct 9, 2019 – Oct 14, 2019 2 messages not in the archive

Subsystems (2)
lore
Share & tools
lore

Emily Deng

author
#0 lore raw

PATCH 1/1 SWDEV-206718 drm/amdgpu: Fix tdr3 could hang with slow compute issue

When index is 1, need to set compute ring timeout for sriov and passthrough.

---
 drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 5 ++++-
 drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c    | 6 ++++--
 2 files changed, 8 insertions(+), 3 deletions(-)
drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +4 -1
index 53ce227..2f5a015 100644
@@ -2664,8 +2664,11 @@ static int amdgpu_device_get_job_timeout_settings(struct amdgpu_device *adev)
 		 * There is only one value specified and
 		 * it should apply to all non-compute jobs.
 		 */
-		if (index == 1)
+		if (index == 1) {
 			adev->sdma_timeout = adev->video_timeout = adev->gfx_timeout;
+			if (amdgpu_sriov_vf(adev) || amdgpu_passthrough(adev))
+				adev->compute_timeout = adev->gfx_timeout;
+		}
 	}
 
 	return ret;
drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c +4 -2
index a88ea74..311abc8 100644
@@ -250,9 +250,11 @@ module_param_named(msi, amdgpu_msi, int, 0444);
  * By default(with no lockup_timeout settings), the timeout for all non-compute(GFX, SDMA and Video)
  * jobs is 10000. And there is no timeout enforced on compute jobs.
  */
-MODULE_PARM_DESC(lockup_timeout, "GPU lockup timeout in ms (default: 10000 for non-compute jobs and infinity timeout for compute jobs."
+MODULE_PARM_DESC(lockup_timeout, "GPU lockup timeout in ms (default: for bare metal 10000 for non-compute jobs and infinity timeout for compute jobs; "
+		"for passthrough or sriov, 10000 for all jobs."
 		" 0: keep default value. negative: infinity timeout), "
-		"format is [Non-Compute] or [GFX,Compute,SDMA,Video]");
+		"format: for bare metal [Non-Compute] or [GFX,Compute,SDMA,Video]; "
+		"for passthrough or sriov [all jobs] or [GFX,Compute,SDMA,Video].");
 module_param_string(lockup_timeout, amdgpu_lockup_timeout, sizeof(amdgpu_lockup_timeout), 0444);
 
 /**
Signature · 6 lines
2.7.4

_______________________________________________
amd-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/amd-gfx

A message that is not in the archive (sent off-list or not yet received): 1570618306-11560-1-git-send-email-Emily.Deng-5C7GfCeVMHo@public.gmane.org

#2 lore raw

Ping....

Best wishes
Emily Deng

Quoted 62 lines

-----Original Message-----
From: Emily Deng <[email protected]>
Sent: Wednesday, October 9, 2019 6:52 PM
To: [email protected]
Cc: Deng, Emily <[email protected]>
Subject: [PATCH] SWDEV-206718 drm/amdgpu: Fix tdr3 could hang with slow compute issue

When index is 1, need to set compute ring timeout for sriov and passthrough.

---
drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 5 ++++-
drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c | 6 ++++--

2 files changed, 8 insertions(+), 3 deletions(-)
drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +4 -1
index 53ce227..2f5a015 100644
@@ -2664,8 +2664,11 @@ static int
 amdgpu_device_get_job_timeout_settings(struct amdgpu_device *adev)
 		 * There is only one value specified and
 		 * it should apply to all non-compute jobs.
 		 */
-		if (index == 1)
+		if (index == 1) {
 			adev->sdma_timeout = adev->video_timeout = adev-
 >gfx_timeout;
+			if (amdgpu_sriov_vf(adev) ||
 amdgpu_passthrough(adev))
+				adev->compute_timeout = adev->gfx_timeout;
+		}
	}

	return ret;
drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c +0 -1
index a88ea74..311abc8 100644
@@ -250,9 +250,11 @@ module_param_named(msi, amdgpu_msi, int, 0444);
 * By default(with no lockup_timeout settings), the timeout for all non-
 compute(GFX, SDMA and Video)
 * jobs is 10000. And there is no timeout enforced on compute jobs.
 */
-MODULE_PARM_DESC(lockup_timeout, "GPU lockup timeout in ms (default:

10000 for non-compute jobs and infinity timeout for compute jobs." +MODULE_PARM_DESC(lockup_timeout, "GPU lockup timeout in ms (default:
for bare metal 10000 for non-compute jobs and infinity timeout for compute jobs; "

drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c +1 -0
+		"for passthrough or sriov, 10000 for all jobs."

" 0: keep default value. negative: infinity timeout), "

drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c +1 -1
-		"format is [Non-Compute] or [GFX,Compute,SDMA,Video]");
+		"format: for bare metal [Non-Compute] or

[GFX,Compute,SDMA,Video]; "

drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c +1 -0
+		"for passthrough or sriov [all jobs] or

[GFX,Compute,SDMA,Video]."); module_param_string(lockup_timeout, amdgpu_lockup_timeout, sizeof(amdgpu_lockup_timeout), 0444);

Quoted 3 lines

/**
--
2.7.4

_______________________________________________
amd-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/amd-gfx

A message that is not in the archive (sent off-list or not yet received): MN2PR12MB2975AFBB743A6C57493F96878F960-rweVpJHSKToFlvJWC7EAqwdYzm3356FpvxpqHgZTriW3zl9H0oFU5g@public.gmane.org

#4 lore raw

________________________________
From: amd-gfx <amd-gfx-bounces-PD4FTy7X32lNgt0PjOBp9y5qC8QIuHrW@public.gmane.org> on behalf of Deng, Emily <[email protected]>
Sent: Saturday, October 12, 2019 1:36 AM
To: Deng, Emily <[email protected]>; [email protected] <[email protected]>
Subject: RE: [PATCH] SWDEV-206718 drm/amdgpu: Fix tdr3 could hang with slow compute issue

Ping....

Best wishes
Emily Deng

Quoted 62 lines

-----Original Message-----
From: Emily Deng <[email protected]>
Sent: Wednesday, October 9, 2019 6:52 PM
To: [email protected]
Cc: Deng, Emily <[email protected]>
Subject: [PATCH] SWDEV-206718 drm/amdgpu: Fix tdr3 could hang with slow compute issue

When index is 1, need to set compute ring timeout for sriov and passthrough.

---
drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 5 ++++-
drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c | 6 ++++--

2 files changed, 8 insertions(+), 3 deletions(-)
drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +4 -1
index 53ce227..2f5a015 100644
@@ -2664,8 +2664,11 @@ static int
 amdgpu_device_get_job_timeout_settings(struct amdgpu_device *adev)
                * There is only one value specified and
                * it should apply to all non-compute jobs.
                */
-              if (index == 1)
+              if (index == 1) {
                       adev->sdma_timeout = adev->video_timeout = adev-
 >gfx_timeout;
+                      if (amdgpu_sriov_vf(adev) ||
 amdgpu_passthrough(adev))
+                              adev->compute_timeout = adev->gfx_timeout;
+              }
       }

       return ret;
drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c +0 -1
index a88ea74..311abc8 100644
@@ -250,9 +250,11 @@ module_param_named(msi, amdgpu_msi, int, 0444);
 * By default(with no lockup_timeout settings), the timeout for all non-
 compute(GFX, SDMA and Video)
 * jobs is 10000. And there is no timeout enforced on compute jobs.
 */
-MODULE_PARM_DESC(lockup_timeout, "GPU lockup timeout in ms (default:

10000 for non-compute jobs and infinity timeout for compute jobs." +MODULE_PARM_DESC(lockup_timeout, "GPU lockup timeout in ms (default:
for bare metal 10000 for non-compute jobs and infinity timeout for compute jobs; "

drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c +2 -1
+              "for passthrough or sriov, 10000 for all jobs."
               " 0: keep default value. negative: infinity timeout), "
-              "format is [Non-Compute] or [GFX,Compute,SDMA,Video]");
+              "format: for bare metal [Non-Compute] or

[GFX,Compute,SDMA,Video]; "
+ "for passthrough or sriov [all jobs] or
[GFX,Compute,SDMA,Video]."); module_param_string(lockup_timeout, amdgpu_lockup_timeout, sizeof(amdgpu_lockup_timeout), 0444);

Quoted 3 lines

/**
--
2.7.4

_______________________________________________
amd-gfx mailing list [email protected]
https://lists.freedesktop.org/mailman/listinfo/amd-gfx

_______________________________________________
amd-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/amd-gfx