intel-gfx.lists.freedesktop.org archive mirror
 help / color / mirror / Atom feed
From: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
To: Chris Wilson <chris@chris-wilson.co.uk>,
	Intel-gfx@lists.freedesktop.org,
	Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Subject: Re: [PATCH v3 4/7] drm/i915: Cache LRC state page in the context
Date: Tue, 12 Jan 2016 12:54:25 +0000	[thread overview]
Message-ID: <5694F781.2050109@linux.intel.com> (raw)
In-Reply-To: <20160112121223.GC26866@nuc-i3427.alporthouse.com>


On 12/01/16 12:12, Chris Wilson wrote:
> On Tue, Jan 12, 2016 at 11:56:11AM +0000, Tvrtko Ursulin wrote:
>> From: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
>>
>> LRC lifetime is well defined so we can cache the page pointing
>> to the object backing store in the context in order to avoid
>> walking over the object SG page list from the interrupt context
>> without the big lock held.
>>
>> v2: Also cache the mapping. (Chris Wilson)
>> v3: Unmap on the error path.
>
> Then we only use the lrc_state_page in the unmapping, hardly worth the
> cost of saving it.

Ok.

Do you also know if this would now require any flushing or something if 
previously kunmap_atomic might have done something under the covers?

> The reg_state is better associated with the ring (since it basically
> contains the analog of the RING_HEAD and friends).

Hm, not sure. The page belongs to the object from that anonymous struct 
in intel_context so I think it is best to keep them together.

Regards,

Tvrtko


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

  reply	other threads:[~2016-01-12 12:54 UTC|newest]

Thread overview: 48+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-01-11 14:08 [PATCH v3 0/7] Misc cleanups and locking fixes Tvrtko Ursulin
2016-01-11 14:08 ` [PATCH 1/7] drm/i915/bdw+: Replace list_del+list_add_tail with list_move_tail Tvrtko Ursulin
2016-01-11 14:08 ` [PATCH 2/7] drm/i915: Do not call API requiring struct_mutex where it is not available Tvrtko Ursulin
2016-01-11 14:35   ` Chris Wilson
2016-01-12 11:41   ` [PATCH v2 " Tvrtko Ursulin
2016-01-12 15:47     ` Dave Gordon
2016-01-13 16:16       ` [PATCH v3] " Tvrtko Ursulin
2016-01-13 19:34         ` Chris Wilson
2016-01-11 14:08 ` [PATCH 3/7] drm/i915: Cache ringbuffer GTT address Tvrtko Ursulin
2016-01-11 14:30   ` Chris Wilson
2016-01-11 14:31   ` Chris Wilson
2016-01-11 15:06     ` Tvrtko Ursulin
2016-01-11 14:08 ` [PATCH 4/7] drm/i915: Cache LRC state page in the context Tvrtko Ursulin
2016-01-11 14:29   ` Chris Wilson
2016-01-11 15:07     ` Tvrtko Ursulin
2016-01-12 11:43     ` [PATCH v2 " Tvrtko Ursulin
2016-01-12 11:56     ` [PATCH v3 " Tvrtko Ursulin
2016-01-12 12:12       ` Chris Wilson
2016-01-12 12:54         ` Tvrtko Ursulin [this message]
2016-01-12 13:11           ` Chris Wilson
2016-01-12 16:45             ` Dave Gordon
2016-01-13  1:37               ` Chris Wilson
2016-01-13 13:49               ` Tvrtko Ursulin
2016-01-11 14:08 ` [PATCH 5/7] drm/i915: Don't need a timer to wake us up Tvrtko Ursulin
2016-01-11 14:33   ` Chris Wilson
2016-01-12 10:20     ` Tvrtko Ursulin
2016-01-11 14:08 ` [PATCH 6/7] drm/i915: Only grab timestamps when needed Tvrtko Ursulin
2016-01-11 14:36   ` Chris Wilson
2016-01-11 15:04     ` Tvrtko Ursulin
2016-01-12 15:52       ` Dave Gordon
2016-01-12 17:14         ` Daniel Vetter
2016-01-13 13:54           ` [PATCH v2] " Tvrtko Ursulin
2016-01-11 14:08 ` [PATCH 7/7] drm/i915: GEM operations need to be done under the big lock Tvrtko Ursulin
2016-01-11 14:38   ` Chris Wilson
2016-01-11 14:47     ` Tvrtko Ursulin
2016-01-11 15:00       ` Chris Wilson
2016-01-11 15:04         ` Chris Wilson
2016-01-11 15:16           ` Tvrtko Ursulin
2016-01-11 15:36             ` Ville Syrjälä
2016-01-11 16:56               ` Chris Wilson
2016-01-13 12:46                 ` Tvrtko Ursulin
2016-01-13 13:36                   ` Imre Deak
2016-01-13 14:11                     ` Tvrtko Ursulin
2016-01-13 14:32                       ` Chris Wilson
2016-01-13 14:41                         ` Imre Deak
2016-01-13 14:53                           ` Tvrtko Ursulin
2016-01-13 15:25                             ` Imre Deak
2016-01-13 15:55                               ` Daniel Vetter

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=5694F781.2050109@linux.intel.com \
    --to=tvrtko.ursulin@linux.intel.com \
    --cc=Intel-gfx@lists.freedesktop.org \
    --cc=chris@chris-wilson.co.uk \
    --cc=tvrtko.ursulin@intel.com \
    /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;
as well as URLs for NNTP newsgroup(s).