Re: [PATCH] drm/amdgpu: Add gpu_recovery parameter

2017-12-18 Thread Deucher, Alexander
...@gmail.com Subject: RE: [PATCH] drm/amdgpu: Add gpu_recovery parameter You can add my RB But to be honest, the current bare-metal GPU recover approach still look not good enough especially that soft_rest checking parts: 1) not all engine/IP on all version are implemented for this, and it's

RE: [PATCH] drm/amdgpu: Add gpu_recovery parameter

2017-12-18 Thread Liu, Monk
top.org Cc: mar...@gmail.com Subject: Re: [PATCH] drm/amdgpu: Add gpu_recovery parameter On 12/14/2017 03:52 AM, Christian König wrote: > Am 14.12.2017 um 08:19 schrieb Liu, Monk: >>> Problem with this is that amdgpu_check_soft_reset will not be >>> called, this function which

Re: [PATCH] drm/amdgpu: Add gpu_recovery parameter

2017-12-15 Thread Andrey Grodzovsky
s hang , it may just busy BR Monk -Original Message- From: Grodzovsky, Andrey Sent: 2017年12月13日 20:53 To: Koenig, Christian ; amd-gfx@lists.freedesktop.org Cc: Liu, Monk ; mar...@gmail.com Subject: Re: [PATCH] drm/amdgpu: Add gpu_recovery parameter On 12/13/2017 07:20 AM,

Re: [PATCH] drm/amdgpu: Add gpu_recovery parameter

2017-12-14 Thread Andrey Grodzovsky
On 12/14/2017 02:16 AM, Liu, Monk wrote: Andrey You patch looks breaks the logic for SRIOV, please check function "xgpu_ai_mailbox_flr_work" This function manually triggers GPU_RECOVER by the will of hypervisor. Your check of : + if (!amdgpu_gpu_recovery) { + DRM_INFO("GP

Re: [PATCH] drm/amdgpu: Add gpu_recovery parameter

2017-12-14 Thread Christian König
0:53 To: Koenig, Christian ; amd-gfx@lists.freedesktop.org Cc: Liu, Monk ; mar...@gmail.com Subject: Re: [PATCH] drm/amdgpu: Add gpu_recovery parameter On 12/13/2017 07:20 AM, Christian König wrote: Am 12.12.2017 um 20:16 schrieb Andrey Grodzovsky: Add new parameter to control GPU recovery pr

RE: [PATCH] drm/amdgpu: Add gpu_recovery parameter

2017-12-13 Thread Liu, Monk
if some IP is busy or not, but busy may not prove the engine is hang , it may just busy BR Monk -Original Message- From: Grodzovsky, Andrey Sent: 2017年12月13日 20:53 To: Koenig, Christian ; amd-gfx@lists.freedesktop.org Cc: Liu, Monk ; mar...@gmail.com Subject: Re: [PATCH] drm/amdgp

RE: [PATCH] drm/amdgpu: Add gpu_recovery parameter

2017-12-13 Thread Liu, Monk
Andrey You patch looks breaks the logic for SRIOV, please check function "xgpu_ai_mailbox_flr_work" This function manually triggers GPU_RECOVER by the will of hypervisor. Your check of : + if (!amdgpu_gpu_recovery) { + DRM_INFO("GPU recovery disabled.\n"); + ret

Re: [PATCH] drm/amdgpu: Add gpu_recovery parameter

2017-12-13 Thread Christian König
Am 13.12.2017 um 13:53 schrieb Andrey Grodzovsky: On 12/13/2017 07:20 AM, Christian König wrote: Am 12.12.2017 um 20:16 schrieb Andrey Grodzovsky: Add new parameter to control GPU recovery procedure. Retire old way of disabling GPU recovery by setting lockup_timeout == 0 and set default for

Re: [PATCH] drm/amdgpu: Add gpu_recovery parameter

2017-12-13 Thread Andrey Grodzovsky
On 12/13/2017 07:20 AM, Christian König wrote: Am 12.12.2017 um 20:16 schrieb Andrey Grodzovsky: Add new parameter to control GPU recovery procedure. Retire old way of disabling GPU recovery by setting lockup_timeout == 0 and set default for lockup_timeout to 10s. Signed-off-by: Andrey Grod

Re: [PATCH] drm/amdgpu: Add gpu_recovery parameter

2017-12-13 Thread Christian König
Am 12.12.2017 um 20:16 schrieb Andrey Grodzovsky: Add new parameter to control GPU recovery procedure. Retire old way of disabling GPU recovery by setting lockup_timeout == 0 and set default for lockup_timeout to 10s. Signed-off-by: Andrey Grodzovsky --- drivers/gpu/drm/amd/amdgpu/amdgpu.h

Re: [PATCH] drm/amdgpu: Add gpu_recovery parameter

2017-12-12 Thread Alex Deucher
On Tue, Dec 12, 2017 at 2:16 PM, Andrey Grodzovsky wrote: > Add new parameter to control GPU recovery procedure. > Retire old way of disabling GPU recovery by setting lockup_timeout == 0 and > set default for lockup_timeout to 10s. > > Signed-off-by: Andrey Grodzovsky > --- > drivers/gpu/drm/amd