From: Chris Wilson <chris@chris-wilson.co.uk>
To: intel-gfx@lists.freedesktop.org
Subject: [PATCH 17/21] drm/i915: Repeat retiring of requests until the seqno is stable
Date: Sat, 16 Apr 2011 10:17:41 +0100 [thread overview]
Message-ID: <1302945465-32115-18-git-send-email-chris@chris-wilson.co.uk> (raw)
In-Reply-To: <1302945465-32115-1-git-send-email-chris@chris-wilson.co.uk>
The list walking over objects and requests to retire may take a
significant amount of time, enough time for the GPU to have finished
more batches. So repeat the list walking until the GPU seqno is stable.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
---
drivers/gpu/drm/i915/i915_gem.c | 66 ++++++++++++++++++++-------------------
1 files changed, 34 insertions(+), 32 deletions(-)
diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
index b9515ac..58e77d6 100644
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -1952,47 +1952,45 @@ i915_gem_retire_requests_ring(struct intel_ring_buffer *ring)
WARN_ON(i915_verify_lists(ring->dev));
- seqno = ring->get_seqno(ring);
+ do {
+ seqno = ring->get_seqno(ring);
- for (i = 0; i < ARRAY_SIZE(ring->sync_seqno); i++)
- if (seqno >= ring->sync_seqno[i])
- ring->sync_seqno[i] = 0;
+ while (!list_empty(&ring->request_list)) {
+ struct drm_i915_gem_request *request;
- while (!list_empty(&ring->request_list)) {
- struct drm_i915_gem_request *request;
+ request = list_first_entry(&ring->request_list,
+ struct drm_i915_gem_request,
+ list);
- request = list_first_entry(&ring->request_list,
- struct drm_i915_gem_request,
- list);
-
- if (!i915_seqno_passed(seqno, request->seqno))
- break;
+ if (!i915_seqno_passed(seqno, request->seqno))
+ break;
- trace_i915_gem_request_retire(ring, request->seqno);
+ trace_i915_gem_request_retire(ring, request->seqno);
- list_del(&request->list);
- i915_gem_request_remove_from_client(request);
- kfree(request);
- }
+ list_del(&request->list);
+ i915_gem_request_remove_from_client(request);
+ kfree(request);
+ }
- /* Move any buffers on the active list that are no longer referenced
- * by the ringbuffer to the flushing/inactive lists as appropriate.
- */
- while (!list_empty(&ring->active_list)) {
- struct drm_i915_gem_object *obj;
+ /* Move any buffers on the active list that are no longer referenced
+ * by the ringbuffer to the flushing/inactive lists as appropriate.
+ */
+ while (!list_empty(&ring->active_list)) {
+ struct drm_i915_gem_object *obj;
- obj= list_first_entry(&ring->active_list,
- struct drm_i915_gem_object,
- ring_list);
+ obj = list_first_entry(&ring->active_list,
+ struct drm_i915_gem_object,
+ ring_list);
- if (!i915_seqno_passed(seqno, obj->last_rendering_seqno))
- break;
+ if (!i915_seqno_passed(seqno, obj->last_rendering_seqno))
+ break;
- if (obj->base.write_domain != 0)
- i915_gem_object_move_to_flushing(obj);
- else
- i915_gem_object_move_to_inactive(obj);
- }
+ if (obj->base.write_domain != 0)
+ i915_gem_object_move_to_flushing(obj);
+ else
+ i915_gem_object_move_to_inactive(obj);
+ }
+ } while (seqno != ring->get_seqno(ring));
if (unlikely(ring->trace_irq_seqno &&
i915_seqno_passed(seqno, ring->trace_irq_seqno))) {
@@ -2000,6 +1998,10 @@ i915_gem_retire_requests_ring(struct intel_ring_buffer *ring)
ring->trace_irq_seqno = 0;
}
+ for (i = 0; i < ARRAY_SIZE(ring->sync_seqno); i++)
+ if (seqno >= ring->sync_seqno[i])
+ ring->sync_seqno[i] = 0;
+
WARN_ON(i915_verify_lists(ring->dev));
}
--
1.7.4.1
next prev parent reply other threads:[~2011-04-16 9:18 UTC|newest]
Thread overview: 33+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-04-16 9:17 i915 next, post-llc Chris Wilson
2011-04-16 9:17 ` [PATCH 01/21] drm/i915: Cache GT fifo count for SandyBridge Chris Wilson
2011-04-16 9:17 ` [PATCH 02/21] drm/i915: Refactor pwrite/pread to use single copy of get_user_pages Chris Wilson
2011-04-16 9:17 ` [PATCH 03/21] drm/i915: s/addr & ~PAGE_MASK/offset_in_page(addr)/ Chris Wilson
2011-04-16 9:17 ` [PATCH 04/21] drm/i915: Maintain fenced gpu access until we flush the fence Chris Wilson
2011-04-16 9:17 ` [PATCH 05/21] drm/i915: Invalidate fenced read domains upon flush Chris Wilson
2011-04-16 9:17 ` [PATCH 06/21] drm/i915: Pass the fence register number to be written Chris Wilson
2011-04-16 9:17 ` [PATCH 07/21] drm/i915: Track fence setup separately from fenced object lifetime Chris Wilson
2011-04-16 13:20 ` Daniel Vetter
2011-04-16 9:17 ` [PATCH 08/21] drm/i915: Only print out the actual number of fences for i915_error_state Chris Wilson
2011-04-16 9:17 ` [PATCH 09/21] drm/i915: Implement direct support for 24 bit LVDS pixel format Chris Wilson
2011-04-16 9:17 ` [PATCH 10/21] drm/i915: Implement manual override of LVDS single/dual channel mode Chris Wilson
2011-04-16 9:17 ` [PATCH 11/21] drm/i915/tv: Use a direct pointer for tv_mode Chris Wilson
2011-04-16 9:17 ` [PATCH 12/21] drm/i915: Replace ironlake_compute_wm0 with g4x_compute_wm0 Chris Wilson
2011-04-16 9:17 ` [PATCH 13/21] drm/i915/crt: Explicitly return false if connected to a digital monitor Chris Wilson
2011-04-16 9:17 ` [PATCH 14/21] drm/i915/i2c: Convert from using GMBUS1 + reg_offset idiom to reg + 0 Chris Wilson
2011-04-16 9:17 ` [PATCH 15/21] drm/i915/gmbus: Reset the controller on initialisation Chris Wilson
2011-04-16 9:17 ` [PATCH 16/21] drm/i915: Retire requests before disabling pagefaults Chris Wilson
2011-04-16 13:44 ` Daniel Vetter
2011-04-16 9:17 ` Chris Wilson [this message]
2011-04-16 13:45 ` [PATCH 17/21] drm/i915: Repeat retiring of requests until the seqno is stable Daniel Vetter
2011-04-16 9:17 ` [PATCH 18/21] drm/i915: Split out i915_gem_object_move_to_ring() from execbuffer Chris Wilson
2011-04-16 13:54 ` Daniel Vetter
2011-04-16 14:18 ` Chris Wilson
2011-04-16 14:24 ` Daniel Vetter
2011-04-16 9:17 ` [PATCH 19/21] drm/i915: Enable the use of GPU semaphores whilst page-flipping Chris Wilson
2011-04-16 13:58 ` Daniel Vetter
2011-04-16 14:20 ` Chris Wilson
2011-04-16 9:17 ` [PATCH 20/21] drm/i915: Use a slab for object allocation Chris Wilson
2011-04-16 14:07 ` Daniel Vetter
2011-04-16 9:17 ` [PATCH 21/21] drm/i915: Introduce vmap (mapping of user pages into video memory) ioctl Chris Wilson
2011-04-18 14:58 ` Daniel Vetter
2011-04-19 6:20 ` 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=1302945465-32115-18-git-send-email-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 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).