[PATCH] drm/amdgpu: Partially revert commit 2dc80b006

Christian König ckoenig.leichtzumerken at gmail.com
Wed Jun 13 12:46:46 UTC 2018


Am 13.06.2018 um 13:40 schrieb Rex Zhu:
> Move the CG enablement out of delay worker thread.
>
> 1. CG/PG enablement are part of gpu hw ip initialize, we should
> wait for them complete. otherwise, there are some potential conflicts,
> for example, Suspend and CG enablement concurrently.
> 2. better run ib test after hw initialize completely. That is to say,
>     ib test should be after CG/PG enablement. otherwise, the test will
>     not cover the cg/pg/poweroff enable case.
>
> Signed-off-by: Rex Zhu <Rex.Zhu at amd.com>

Yeah, that thought came to my mind as well.

Essentially the IB test should simulate a submission from userspace to 
make sure that the stack is working as expected. I think it was just 
moved before CG/PG to avoid issues with that, which is actually not very 
clever.

Patch is Reviewed-by: Christian König <christian.koenig at amd.com>, but 
there could be some fallout we could need to deal with.

Thanks,
Christian.

> ---
>   drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 14 ++++++++------
>   1 file changed, 8 insertions(+), 6 deletions(-)
>
> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
> index 9647f54..90b78c7 100644
> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
> @@ -1709,10 +1709,6 @@ static int amdgpu_device_ip_late_set_cg_state(struct amdgpu_device *adev)
>   	if (amdgpu_emu_mode == 1)
>   		return 0;
>   
> -	r = amdgpu_ib_ring_tests(adev);
> -	if (r)
> -		DRM_ERROR("ib ring test failed (%d).\n", r);
> -
>   	for (i = 0; i < adev->num_ip_blocks; i++) {
>   		if (!adev->ip_blocks[i].status.valid)
>   			continue;
> @@ -1793,6 +1789,9 @@ static int amdgpu_device_ip_late_init(struct amdgpu_device *adev)
>   		}
>   	}
>   
> +	amdgpu_device_ip_late_set_cg_state(adev);
> +	amdgpu_device_ip_late_set_pg_state(adev);
> +
>   	queue_delayed_work(system_wq, &adev->late_init_work,
>   			   msecs_to_jiffies(AMDGPU_RESUME_MS));
>   
> @@ -1921,8 +1920,11 @@ static void amdgpu_device_ip_late_init_func_handler(struct work_struct *work)
>   {
>   	struct amdgpu_device *adev =
>   		container_of(work, struct amdgpu_device, late_init_work.work);
> -	amdgpu_device_ip_late_set_cg_state(adev);
> -	amdgpu_device_ip_late_set_pg_state(adev);
> +	int r;
> +
> +	r = amdgpu_ib_ring_tests(adev);
> +	if (r)
> +		DRM_ERROR("ib ring test failed (%d).\n", r);
>   }
>   
>   /**



More information about the amd-gfx mailing list