public inbox for intel-gfx@lists.freedesktop.org
 help / color / mirror / Atom feed
From: "Ville Syrjälä" <ville.syrjala@linux.intel.com>
To: Chris Wilson <chris@chris-wilson.co.uk>, intel-gfx@lists.freedesktop.org
Subject: Re: [PATCH v2 3/6] drm/i915: Use WC copies on !llc platforms for the command parser
Date: Tue, 1 Dec 2015 19:32:43 +0200	[thread overview]
Message-ID: <20151201173243.GJ4437@intel.com> (raw)
In-Reply-To: <20151120152203.GF16141@nuc-i3427.alporthouse.com>

On Fri, Nov 20, 2015 at 03:22:03PM +0000, Chris Wilson wrote:
> On Fri, Nov 20, 2015 at 05:05:05PM +0200, Ville Syrjälä wrote:
> > On Fri, Nov 20, 2015 at 10:55:58AM +0000, Chris Wilson wrote:
> > > Since we blow the TLB caches by using kmap/kunmap, we may as well go the
> > > whole hog and see if declaring our destination page as WC is faster than
> > > keeping it as WB and using clflush. It should be!
> > 
> > Is this description for another patch? I can't see any WC stuff in
> > there.
> 
> No, just badly written. (Admittedly at one point I did experiment with
> remapping the pages as WC and discovered that did stop_macheine()...)
> 
> drm/i915: Perform inline clflushes from the command parser
> 
> On incoherent architectures, we can avoid having to clflush the
> destination shadow batch as a separate pass by inlining the call to
> clflush whilst we already have the kmap_atomic() around for the page.

With the new description:
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>

-- 
Ville Syrjälä
Intel OTC
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

  reply	other threads:[~2015-12-01 17:32 UTC|newest]

Thread overview: 26+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-11-20 10:55 cmdparser overhead reduction Chris Wilson
2015-11-20 10:55 ` [PATCH v2 1/6] drm/i915: Eliminate vmap overhead for cmd parser Chris Wilson
2015-11-20 14:41   ` Ville Syrjälä
2015-11-20 14:52     ` Chris Wilson
2015-11-20 15:31     ` [PATCH v3] " Chris Wilson
2015-11-25 19:51       ` Ville Syrjälä
2015-11-25 20:13         ` Chris Wilson
2015-11-25 21:15           ` Ville Syrjälä
2015-11-20 10:55 ` [PATCH v2 2/6] drm/i915: Cache last cmd descriptor when parsing Chris Wilson
2015-11-20 15:08   ` Ville Syrjälä
2015-11-20 15:44     ` Chris Wilson
2015-12-01 17:30   ` Ville Syrjälä
2015-11-20 10:55 ` [PATCH v2 3/6] drm/i915: Use WC copies on !llc platforms for the command parser Chris Wilson
2015-11-20 15:05   ` Ville Syrjälä
2015-11-20 15:22     ` Chris Wilson
2015-12-01 17:32       ` Ville Syrjälä [this message]
2015-11-20 10:55 ` [PATCH v2 4/6] drm/i915: Reduce arithmetic operations during cmd parser lookup Chris Wilson
2015-11-20 15:02   ` Ville Syrjälä
2015-11-20 10:56 ` [PATCH v2 5/6] drm/i915: Reduce pointer indirection " Chris Wilson
2015-11-20 15:27   ` Ville Syrjälä
2015-11-20 15:34     ` Chris Wilson
2015-11-20 15:47       ` Ville Syrjälä
2015-11-23  8:09         ` Jani Nikula
2015-12-01 17:39     ` Ville Syrjälä
2015-11-20 10:56 ` [PATCH v2 6/6] drm/i915: Improve hash function for the command parser Chris Wilson
2015-11-20 15:13   ` Ville Syrjälä

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=20151201173243.GJ4437@intel.com \
    --to=ville.syrjala@linux.intel.com \
    --cc=chris@chris-wilson.co.uk \
    --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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox