* [PATCH] drm/amdgpu: change the default timeout for kernel compute queues
@ 2021-05-04 15:03 Alex Deucher
2021-05-05 6:34 ` Christian König
0 siblings, 1 reply; 2+ messages in thread
From: Alex Deucher @ 2021-05-04 15:03 UTC (permalink / raw)
To: amd-gfx; +Cc: Alex Deucher
Change to 60s. This matches what we already do in virtualization.
Infinite timeout can lead to deadlocks in the kernel.
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 8 +++-----
drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c | 4 ++--
2 files changed, 5 insertions(+), 7 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
index 357f9405f1aa..1b60f8205f15 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -3167,8 +3167,8 @@ static int amdgpu_device_get_job_timeout_settings(struct amdgpu_device *adev)
int ret = 0;
/*
- * By default timeout for non compute jobs is 10000.
- * And there is no timeout enforced on compute jobs.
+ * By default timeout for non compute jobs is 10000
+ * and 60000 for compute jobs.
* In SR-IOV or passthrough mode, timeout for compute
* jobs are 60000 by default.
*/
@@ -3177,10 +3177,8 @@ static int amdgpu_device_get_job_timeout_settings(struct amdgpu_device *adev)
if (amdgpu_sriov_vf(adev))
adev->compute_timeout = amdgpu_sriov_is_pp_one_vf(adev) ?
msecs_to_jiffies(60000) : msecs_to_jiffies(10000);
- else if (amdgpu_passthrough(adev))
- adev->compute_timeout = msecs_to_jiffies(60000);
else
- adev->compute_timeout = MAX_SCHEDULE_TIMEOUT;
+ adev->compute_timeout = msecs_to_jiffies(60000);
if (strnlen(input, AMDGPU_MAX_TIMEOUT_PARAM_LENGTH)) {
while ((timeout_setting = strsep(&input, ",")) &&
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
index ad8a2c391bbd..d3d756585db4 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
@@ -287,9 +287,9 @@ module_param_named(msi, amdgpu_msi, int, 0444);
* for SDMA and Video.
*
* 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.
+ * jobs is 10000. The timeout for compute is 60000.
*/
-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; "
+MODULE_PARM_DESC(lockup_timeout, "GPU lockup timeout in ms (default: for bare metal 10000 for non-compute jobs and 60000 for compute jobs; "
"for passthrough or sriov, 10000 for all jobs."
" 0: keep default value. negative: infinity timeout), "
"format: for bare metal [Non-Compute] or [GFX,Compute,SDMA,Video]; "
--
2.30.2
_______________________________________________
amd-gfx mailing list
amd-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/amd-gfx
^ permalink raw reply related [flat|nested] 2+ messages in thread
* Re: [PATCH] drm/amdgpu: change the default timeout for kernel compute queues
2021-05-04 15:03 [PATCH] drm/amdgpu: change the default timeout for kernel compute queues Alex Deucher
@ 2021-05-05 6:34 ` Christian König
0 siblings, 0 replies; 2+ messages in thread
From: Christian König @ 2021-05-05 6:34 UTC (permalink / raw)
To: Alex Deucher, amd-gfx
Am 04.05.21 um 17:03 schrieb Alex Deucher:
> Change to 60s. This matches what we already do in virtualization.
> Infinite timeout can lead to deadlocks in the kernel.
>
> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
> ---
> drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 8 +++-----
> drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c | 4 ++--
> 2 files changed, 5 insertions(+), 7 deletions(-)
>
> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
> index 357f9405f1aa..1b60f8205f15 100644
> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
> @@ -3167,8 +3167,8 @@ static int amdgpu_device_get_job_timeout_settings(struct amdgpu_device *adev)
> int ret = 0;
>
> /*
> - * By default timeout for non compute jobs is 10000.
> - * And there is no timeout enforced on compute jobs.
> + * By default timeout for non compute jobs is 10000
> + * and 60000 for compute jobs.
> * In SR-IOV or passthrough mode, timeout for compute
> * jobs are 60000 by default.
> */
> @@ -3177,10 +3177,8 @@ static int amdgpu_device_get_job_timeout_settings(struct amdgpu_device *adev)
> if (amdgpu_sriov_vf(adev))
> adev->compute_timeout = amdgpu_sriov_is_pp_one_vf(adev) ?
> msecs_to_jiffies(60000) : msecs_to_jiffies(10000);
> - else if (amdgpu_passthrough(adev))
> - adev->compute_timeout = msecs_to_jiffies(60000);
> else
> - adev->compute_timeout = MAX_SCHEDULE_TIMEOUT;
> + adev->compute_timeout = msecs_to_jiffies(60000);
>
> if (strnlen(input, AMDGPU_MAX_TIMEOUT_PARAM_LENGTH)) {
> while ((timeout_setting = strsep(&input, ",")) &&
> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
> index ad8a2c391bbd..d3d756585db4 100644
> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
> @@ -287,9 +287,9 @@ module_param_named(msi, amdgpu_msi, int, 0444);
> * for SDMA and Video.
> *
> * 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.
> + * jobs is 10000. The timeout for compute is 60000.
> */
> -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; "
> +MODULE_PARM_DESC(lockup_timeout, "GPU lockup timeout in ms (default: for bare metal 10000 for non-compute jobs and 60000 for compute jobs; "
> "for passthrough or sriov, 10000 for all jobs."
> " 0: keep default value. negative: infinity timeout), "
> "format: for bare metal [Non-Compute] or [GFX,Compute,SDMA,Video]; "
_______________________________________________
amd-gfx mailing list
amd-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/amd-gfx
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2021-05-05 6:35 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-05-04 15:03 [PATCH] drm/amdgpu: change the default timeout for kernel compute queues Alex Deucher
2021-05-05 6:34 ` Christian König
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).