Re: [PATCH 6/8] drm/amdgpu: meld together VM fragment and huge page handling

2018-11-12 Thread Christian König
Am 09.11.18 um 13:13 schrieb Koenig, Christian: Adding Aaron who initially reported the problem with this patch and suspend/resume. Am 08.11.18 um 20:35 schrieb Samuel Pitoiset: On 11/8/18 5:50 PM, Christian König wrote: Hi Samuel, yeah, that is a known issue and I was already looking into

Re: [PATCH 6/8] drm/amdgpu: meld together VM fragment and huge page handling

2018-11-09 Thread Koenig, Christian
Adding Aaron who initially reported the problem with this patch and suspend/resume. Am 08.11.18 um 20:35 schrieb Samuel Pitoiset: > On 11/8/18 5:50 PM, Christian König wrote: >> Hi Samuel, >> >> yeah, that is a known issue and I was already looking into that all >> day today. Problem is that I

Re: [PATCH 6/8] drm/amdgpu: meld together VM fragment and huge page handling

2018-11-08 Thread Samuel Pitoiset
On 11/8/18 5:50 PM, Christian König wrote: Hi Samuel, yeah, that is a known issue and I was already looking into that all day today. Problem is that I can't reproduce it reliable. Do you have any hint how to trigger that? E.g. memory exhaustion maybe? Hi Christian, Good to hear that you

Re: [PATCH 6/8] drm/amdgpu: meld together VM fragment and huge page handling

2018-11-08 Thread Christian König
Hi Samuel, yeah, that is a known issue and I was already looking into that all day today. Problem is that I can't reproduce it reliable. Do you have any hint how to trigger that? E.g. memory exhaustion maybe? Thanks, Christian. Am 08.11.18 um 17:17 schrieb Samuel Pitoiset: Hi, This change

Re: [PATCH 6/8] drm/amdgpu: meld together VM fragment and huge page handling

2018-11-08 Thread Samuel Pitoiset
Hi, This change introduces GPU hangs with F1 2017 and Rise of Tomb Raider on RADV/GFX9. I bisected the issue. Can you have a look? Thanks! On 9/12/18 10:54 AM, Christian König wrote: This optimizes the generating of PTEs by walking the hierarchy only once for a range and making changes as

[PATCH 6/8] drm/amdgpu: meld together VM fragment and huge page handling

2018-09-12 Thread Christian König
This optimizes the generating of PTEs by walking the hierarchy only once for a range and making changes as necessary. It allows for both huge (2MB) as well giant (1GB) pages to be used on Vega and Raven. Signed-off-by: Christian König Reviewed-by: Felix Kuehling Acked-by: Junwei Zhang ---