All of lore.kernel.org
 help / color / mirror / Atom feed
From: Chris Wilson <chris@chris-wilson.co.uk>
To: intel-gfx@lists.freedesktop.org
Subject: [PATCH 1/7] drm/i915: Flush submission tasklet after bumping priority
Date: Mon,  7 May 2018 10:25:21 +0100	[thread overview]
Message-ID: <20180507092527.7359-1-chris@chris-wilson.co.uk> (raw)

When called from process context tasklet_schedule() defers itself to
ksoftirqd. From experience this may cause unacceptable latencies of over
200ms in executing the submission tasklet, our goal is to reprioritise
the HW execution queue and trigger preemption immediately, so convert
the rcu protection over to include a kick of the tasklet.

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
---
 drivers/gpu/drm/i915/i915_gem.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
index 5ece6ae4bdff..03cd30001b5d 100644
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -578,10 +578,10 @@ static void __fence_set_priority(struct dma_fence *fence,
 	rq = to_request(fence);
 	engine = rq->engine;
 
-	rcu_read_lock();
+	local_bh_disable(); /* RCU serialisation for set-wedged protection */
 	if (engine->schedule)
 		engine->schedule(rq, attr);
-	rcu_read_unlock();
+	local_bh_enable(); /* kick the tasklets if queues were reprioritised */
 }
 
 static void fence_set_priority(struct dma_fence *fence,
-- 
2.17.0

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

             reply	other threads:[~2018-05-07  9:26 UTC|newest]

Thread overview: 19+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-05-07  9:25 Chris Wilson [this message]
2018-05-07  9:25 ` [PATCH 2/7] drm/i915: Combine set-wedged protection and tasklet kicking Chris Wilson
2018-05-07 12:09   ` Mika Kuoppala
2018-05-07 12:16     ` Chris Wilson
2018-05-07  9:25 ` [PATCH 3/7] drm/i915/execlists: Make submission tasklet hardirq safe Chris Wilson
2018-05-07  9:25 ` [PATCH 4/7] drm/i915/guc: " Chris Wilson
2018-05-07  9:25 ` [PATCH 5/7] drm/i915/execlists: Direct submit onto idle engines Chris Wilson
2018-05-07  9:25 ` [PATCH 6/7] drm/i915/execlists: Direct submission from irq handler Chris Wilson
2018-05-07  9:25 ` [PATCH 7/7] drm/i915: Speed up idle detection by kicking the tasklets Chris Wilson
2018-05-07  9:35   ` [PATCH v3] " Chris Wilson
2018-05-07 14:08     ` kbuild test robot
2018-05-07  9:33 ` ✗ Fi.CI.CHECKPATCH: warning for series starting with [1/7] drm/i915: Flush submission tasklet after bumping priority Patchwork
2018-05-07  9:35 ` ✗ Fi.CI.SPARSE: " Patchwork
2018-05-07  9:52 ` ✗ Fi.CI.BAT: failure " Patchwork
2018-05-07  9:57 ` ✗ Fi.CI.CHECKPATCH: warning for series starting with [1/7] drm/i915: Flush submission tasklet after bumping priority (rev2) Patchwork
2018-05-07  9:59 ` ✗ Fi.CI.SPARSE: " Patchwork
2018-05-07 10:14 ` ✗ Fi.CI.BAT: failure " Patchwork
2018-05-07 11:15 ` [PATCH 1/7] drm/i915: Flush submission tasklet after bumping priority Mika Kuoppala
2018-05-07 11:19   ` Chris Wilson

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=20180507092527.7359-1-chris@chris-wilson.co.uk \
    --to=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 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.