From mboxrd@z Thu Jan 1 00:00:00 1970 From: =?ISO-8859-1?Q?Christian_K=F6nig?= Subject: Re: [PATCH 0/5] radeon: Write-combined CPU mappings of BOs in GTT Date: Thu, 17 Jul 2014 12:09:20 +0200 Message-ID: <53C7A0D0.6080202@vodafone.de> References: <1405591275-14461-1-git-send-email-michel@daenzer.net> Mime-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1"; Format="flowed" Content-Transfer-Encoding: quoted-printable Return-path: Received: from pegasos-out.vodafone.de (pegasos-out.vodafone.de [80.84.1.38]) by gabe.freedesktop.org (Postfix) with ESMTP id 218DB6E279 for ; Thu, 17 Jul 2014 03:09:35 -0700 (PDT) In-Reply-To: <1405591275-14461-1-git-send-email-michel@daenzer.net> List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dri-devel-bounces@lists.freedesktop.org Sender: "dri-devel" To: =?ISO-8859-1?Q?Michel_D=E4nzer?= , dri-devel@lists.freedesktop.org, mesa-dev@lists.freedesktop.org List-Id: dri-devel@lists.freedesktop.org Am 17.07.2014 12:01, schrieb Michel D=E4nzer: > In order to try and improve X(Shm)PutImage performance with glamor, I > implemented support for write-combined CPU mappings of BOs in GTT. > > This did provide a nice speedup, but to my surprise, using VRAM instead > of write-combined GTT turned out to be even faster in general on my > Kaveri machine, both for the internal GPU and for discrete GPUs. > > However, I've kept the changes from GTT to VRAM separated, in case this > turns out to be a loss on other setups. > > Kernel patches: > > [PATCH 1/5] drm/radeon: Remove radeon_gart_restore() > [PATCH 2/5] drm/radeon: Pass GART page flags to > [PATCH 3/5] drm/radeon: Allow write-combined CPU mappings of BOs in > [PATCH 4/5] drm/radeon: Use write-combined CPU mappings of rings and Those four are Reviewed-by: Christian K=F6nig > [PATCH 5/5] drm/radeon: Use VRAM for indirect buffers on >=3D SI I'm still not very keen with this change since I still don't understand = the reason why it's faster than with GTT. Definitely needs more testing = on a wider range of systems. Maybe limit it to APUs for now? Regards, Christian. > > Mesa patches: > > [PATCH 1/5] winsys/radeon: Use separate caching buffer managers for > [PATCH 2/5] r600g/radeonsi: Use write-combined CPU mappings of some > [PATCH 3/5] r600g/radeonsi: Prefer VRAM for CPU -> GPU streaming > [PATCH 4/5] r600g,radeonsi: Use write-combined persistent GTT > [PATCH 5/5] r600g,radeonsi: Prefer VRAM for persistent mappings > _______________________________________________ > dri-devel mailing list > dri-devel@lists.freedesktop.org > http://lists.freedesktop.org/mailman/listinfo/dri-devel