Re: drm/amdgpu: invalidate L2 before SDMA IBs (on gfx10)

2020-05-01 Thread Marek Olšák
> > _ > > Monk Liu|GPU Virtualization Team |AMD > > [image: sig-cloud-gpu] > > > > *From:* Liu, Monk > *Sent:* Wednesday, April 29, 2020 5:06 PM > *To:* 'Marek Olšák' ; amd-gfx > mailing list > ; Koenig, Christian >

RE: drm/amdgpu: invalidate L2 before SDMA IBs (on gfx10)

2020-04-29 Thread Koenig, Christian
lto:amd-gfx@lists.freedesktop.org>; Koenig, Christian <mailto:christian.koe...@amd.com> Subject: RE: drm/amdgpu: invalidate L2 before SDMA IBs (on gfx10) Hi @Koenig, Christian<mailto:christian.koe...@amd.com> & Marek I still have some concerns regarding Marek’s patch, correct

RE: drm/amdgpu: invalidate L2 before SDMA IBs (on gfx10)

2020-04-29 Thread Liu, Monk
Koenig, Christian Sent: Wednesday, April 29, 2020 5:38 PM To: Liu, Monk ; Marek Olšák ; amd-gfx mailing list Subject: Re: drm/amdgpu: invalidate L2 before SDMA IBs (on gfx10) Well from my understanding I think that a G2LC invalidation is still necessary before an IB executes. The problem is t

Re: drm/amdgpu: invalidate L2 before SDMA IBs (on gfx10)

2020-04-29 Thread Christian König
a fence signaled _ Monk Liu|GPU Virtualization Team |AMD sig-cloud-gpu *From:* Liu, Monk *Sent:* Wednesday, April 29, 2020 5:06 PM *To:* 'Marek Olšák' ; amd-gfx mailing list ; Koenig, Christian *Subject:* RE: drm/amdgpu: invalidate L2 before SDMA IBs

RE: drm/amdgpu: invalidate L2 before SDMA IBs (on gfx10)

2020-04-29 Thread Liu, Monk
and invalidate upon a fence signaled _ Monk Liu|GPU Virtualization Team |AMD [sig-cloud-gpu] From: Liu, Monk Sent: Wednesday, April 29, 2020 5:06 PM To: 'Marek Olšák' ; amd-gfx mailing list ; Koenig, Christian Subject: RE: drm/amdgpu: invalidate L2 before SDMA IBs

RE: drm/amdgpu: invalidate L2 before SDMA IBs (on gfx10)

2020-04-29 Thread Liu, Monk
Hi @Koenig, Christian & Marek I still have some concerns regarding Marek’s patch, correct me if I’m wrong See that Marek put a SDMA_OP_GCR_REQ before emitting IB, to make sure SDMA won’t get stale cache data during the IB execution. But that “SDMA_OP_GCR_REQ”

Re: drm/amdgpu: invalidate L2 before SDMA IBs (on gfx10)

2020-04-27 Thread Marek Olšák
for the new > packet in the emi IB function? > > Alex > -- > *From:* amd-gfx on behalf of > Christian König > *Sent:* Sunday, April 26, 2020 4:52 AM > *To:* Marek Olšák > *Cc:* amd-gfx mailing list > *Subject:* Re: drm/amdgpu: invalida

Re: drm/amdgpu: invalidate L2 before SDMA IBs (on gfx10)

2020-04-27 Thread Deucher, Alexander
function? Alex From: amd-gfx on behalf of Christian König Sent: Sunday, April 26, 2020 4:52 AM To: Marek Olšák Cc: amd-gfx mailing list Subject: Re: drm/amdgpu: invalidate L2 before SDMA IBs (on gfx10) Thought so, we should try to get this get it committed

Re: drm/amdgpu: invalidate L2 before SDMA IBs (on gfx10)

2020-04-26 Thread Christian König
Thought so, we should try to get this get it committed ASAP. And maybe add a CC: stable tag as well. Patch is Reviewed-by: Christian König . Thanks, Christian. Am 26.04.20 um 02:28 schrieb Marek Olšák: Not without a mandatory firmware update. The gcr packet support for IBs was added into the

Re: drm/amdgpu: invalidate L2 before SDMA IBs (on gfx10)

2020-04-25 Thread Marek Olšák
Not without a mandatory firmware update. The gcr packet support for IBs was added into the sdma firmware just two weeks ago. Marek On Sat., Apr. 25, 2020, 11:04 Christian König, < ckoenig.leichtzumer...@gmail.com> wrote: > Could we do this in userspace as well? > > Am 25.04.20 um 10:52 schrieb

Re: drm/amdgpu: invalidate L2 before SDMA IBs (on gfx10)

2020-04-25 Thread Pierre-Eric Pelloux-Prayer
Hi Marek, With this patch applied I could replay a trace (that usually caused a sdma timeout on the first run) several times in a row without any issues. So you can tag the commit as: Tested-by: Pierre-Eric Pelloux-Prayer Thanks, Pierre-Eric On 25/04/2020 10:52, Marek Olšák wrote: > This