Re: [PATCH 1/1] drm/amdkfd: Consistently apply noretry setting

2019-07-12 Thread Liu, Shaoyun
Reviewed-by : Shaoyun.liu < shaoyun@amd.com> I manually applied the change and  verified on vega10 on visualization  which we do have  retry vm fault issue . With  this patch , the no-retry interrupt delivered as expected and I don't see  the interrupt storm. Regards shaoyun.liu On 2019-

Re: [PATCH 1/1] drm/amdkfd: Consistently apply noretry setting

2019-07-04 Thread Kuehling, Felix
On 2019-07-04 1:16 p.m., Yang, Philip wrote: > On 2019-07-04 12:02 p.m., Kuehling, Felix wrote: >> On 2019-07-03 6:19 p.m., Yang, Philip wrote: >>> amdgpu_noretry default value is 0, this will generate VM fault storm >>> because the vm fault is not recovered. It may slow down the machine and >>> ne

Re: [PATCH 1/1] drm/amdkfd: Consistently apply noretry setting

2019-07-04 Thread Yang, Philip
On 2019-07-04 12:02 p.m., Kuehling, Felix wrote: > On 2019-07-03 6:19 p.m., Yang, Philip wrote: >> amdgpu_noretry default value is 0, this will generate VM fault storm >> because the vm fault is not recovered. It may slow down the machine and >> need reboot after application VM fault. Maybe change

Re: [PATCH 1/1] drm/amdkfd: Consistently apply noretry setting

2019-07-04 Thread Kuehling, Felix
On 2019-07-03 6:19 p.m., Yang, Philip wrote: > amdgpu_noretry default value is 0, this will generate VM fault storm > because the vm fault is not recovered. It may slow down the machine and > need reboot after application VM fault. Maybe change default value to 1? This is the same as the current b

Re: [PATCH 1/1] drm/amdkfd: Consistently apply noretry setting

2019-07-04 Thread Koenig, Christian
Patches to silent retry faults are already on the mailing list. But I need to get back from vacation to take a closer look at this once more. Christian. Am 04.07.2019 00:19 schrieb "Yang, Philip" : amdgpu_noretry default value is 0, this will generate VM fault storm because the vm fault is not r

Re: [PATCH 1/1] drm/amdkfd: Consistently apply noretry setting

2019-07-03 Thread Yang, Philip
amdgpu_noretry default value is 0, this will generate VM fault storm because the vm fault is not recovered. It may slow down the machine and need reboot after application VM fault. Maybe change default value to 1? Other than that, this is reviewed by Philip Yang On 2019-07-02 3:05 p.m., Kuehli

Re: [PATCH 1/1] drm/amdkfd: Consistently apply noretry setting

2019-07-02 Thread Kuehling, Felix
Ping. Christian, Philip, any opinion about this patch? On 2019-06-21 8:20 p.m., Kuehling, Felix wrote: > Apply the same setting to SH_MEM_CONFIG and VM_CONTEXT1_CNTL. This > makes the noretry param no longer KFD-specific. On GFX10 I'm not > changing SH_MEM_CONFIG in this commit because GFX10 has