From: Chris Wilson <chris@chris-wilson.co.uk>
To: Daniel Vetter <daniel@ffwll.ch>
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>,
intel-gfx <intel-gfx@lists.freedesktop.org>
Subject: Re: [PATCH 3/5] drm/i915: fall through pwrite_gtt_slow to the shmem slow path
Date: Sun, 30 Oct 2011 08:35:34 +0000 [thread overview]
Message-ID: <d08817$219t3v@azsmga001.ch.intel.com> (raw)
In-Reply-To: <20110920113556.GA2865@phenom.ffwll.local>
On Tue, 20 Sep 2011 13:35:56 +0200, Daniel Vetter <daniel@ffwll.ch> wrote:
> On Sat, Sep 17, 2011 at 09:57:05PM +0100, Chris Wilson wrote:
> > And whilst you are here, can you incorporate this patch?
> > else if (obj->gtt_space &&
> > + obj->map_and_fenceable &&
> > obj->base.write_domain != I915_GEM_DOMAIN_CPU) {
> > ret = i915_gem_object_pin(obj, 0, true);
>
> Not sure this is good. On !llc machines, gtt_pwrite is much faster, so we
> want to move the objects into the mappable part of the gtt to benefit from
> that. Without this, they'll just stay wherever they are.
This turns out to be a big win for machines where it avoids the pipeline
stall due to the eviction of an active page and allowing us to utilize
the full GTT for vertex data.
10% on pnv, 60% on snb for x11perf -aa10text. (Though on SNB this is
eclipsed by using LLC and a test for obj->cache_level). And it is likely
to be an improvement with geometry bound game benchmarks like openarena.
-Chris
--
Chris Wilson, Intel Open Source Technology Centre
next prev parent reply other threads:[~2011-10-30 8:35 UTC|newest]
Thread overview: 18+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-09-17 18:55 [PATCH 0/5] Fixup pwrite/pread with gtt mmapped user addresses Daniel Vetter
2011-09-17 18:55 ` [PATCH 1/5] io-mapping: ensure io_mapping_map_atomic _is_ atomic Daniel Vetter
2011-09-17 20:43 ` Chris Wilson
2011-09-17 18:55 ` [PATCH 2/5] drm/i915: drop KM_USER0 argument to k(un)map_atomic Daniel Vetter
2011-09-17 20:44 ` Chris Wilson
2011-09-23 16:25 ` Daniel Vetter
2011-09-17 18:55 ` [PATCH 3/5] drm/i915: fall through pwrite_gtt_slow to the shmem slow path Daniel Vetter
2011-09-17 20:57 ` Chris Wilson
2011-09-20 11:35 ` Daniel Vetter
2011-10-30 8:35 ` Chris Wilson [this message]
2011-09-17 18:55 ` [PATCH 4/5] drm/i915: rewrite shmem_pwrite_slow to use copy_from_user Daniel Vetter
2011-09-17 20:52 ` Daniel Vetter
2011-09-17 21:00 ` Chris Wilson
2011-09-18 8:44 ` Daniel Vetter
2011-09-17 18:55 ` [PATCH 5/5] drm/i915: rewrite shmem_pread_slow to use copy_to_user Daniel Vetter
2011-09-17 21:04 ` Chris Wilson
2011-09-17 23:07 ` [PATCH 0/5] Fixup pwrite/pread with gtt mmapped user addresses Ben Widawsky
2011-10-20 21:14 ` Keith Packard
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to='d08817$219t3v@azsmga001.ch.intel.com' \
--to=chris@chris-wilson.co.uk \
--cc=daniel.vetter@ffwll.ch \
--cc=daniel@ffwll.ch \
--cc=intel-gfx@lists.freedesktop.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.