All of lore.kernel.org
 help / color / mirror / Atom feed
From: Chris Wilson <chris@chris-wilson.co.uk>
To: intel-gfx@lists.freedesktop.org
Cc: Chris Wilson <chris@chris-wilson.co.uk>, stable@vger.kernel.org
Subject: [CI-ping 12/15] drm/i915: Force ringbuffers to not be at offset 0
Date: Tue, 12 Apr 2016 21:03:06 +0100	[thread overview]
Message-ID: <1460491389-8602-12-git-send-email-chris@chris-wilson.co.uk> (raw)
In-Reply-To: <1460491389-8602-1-git-send-email-chris@chris-wilson.co.uk>

For reasons unknown Sandybridge GT1 (at least) will eventually hang when
it encounters a ring wraparound at offset 0. The test case that
reproduces the bug reliably forces a large number of interrupted context
switches, thereby causing very frequent ring wraparounds, but there are
similar bug reports in the wild with the same symptoms, seqno writes
stop just before the wrap and the ringbuffer at address 0. It is also
timing crucial, but adding various delays hasn't helped pinpoint where
the window lies.

Whether the fault is restricted to the ringbuffer itself or the GTT
addressing is unclear, but moving the ringbuffer fixes all the hangs I
have been able to reproduce.

References: (e.g.) https://bugs.freedesktop.org/show_bug.cgi?id=93262
Testcase: igt/gem_exec_whisper/render-contexts-interruptible #snb-gt1
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: stable@vger.kernel.org
---
 drivers/gpu/drm/i915/intel_ringbuffer.c | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c
index 8391382431b2..904a8a276f6a 100644
--- a/drivers/gpu/drm/i915/intel_ringbuffer.c
+++ b/drivers/gpu/drm/i915/intel_ringbuffer.c
@@ -2096,10 +2096,12 @@ int intel_pin_and_map_ringbuffer_obj(struct drm_device *dev,
 	struct drm_i915_private *dev_priv = to_i915(dev);
 	struct i915_ggtt *ggtt = &dev_priv->ggtt;
 	struct drm_i915_gem_object *obj = ringbuf->obj;
+	/* Ring wraparound at offset 0 sometimes hangs. No idea why. */
+	unsigned flags = PIN_OFFSET_BIAS | 4096;
 	int ret;
 
 	if (HAS_LLC(dev_priv) && !obj->stolen) {
-		ret = i915_gem_obj_ggtt_pin(obj, PAGE_SIZE, 0);
+		ret = i915_gem_obj_ggtt_pin(obj, PAGE_SIZE, flags);
 		if (ret)
 			return ret;
 
@@ -2113,7 +2115,8 @@ int intel_pin_and_map_ringbuffer_obj(struct drm_device *dev,
 			goto err_unpin;
 		}
 	} else {
-		ret = i915_gem_obj_ggtt_pin(obj, PAGE_SIZE, PIN_MAPPABLE);
+		ret = i915_gem_obj_ggtt_pin(obj, PAGE_SIZE,
+					    flags | PIN_MAPPABLE);
 		if (ret)
 			return ret;
 
-- 
2.8.0.rc3


  parent reply	other threads:[~2016-04-12 20:03 UTC|newest]

Thread overview: 39+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-04-12 20:02 [CI-ping 01/15] drm/i915: Force clean compilation with -Werror Chris Wilson
2016-04-12 20:02 ` [CI-ping 02/15] drm/i915: Disentangle i915_drv.h includes Chris Wilson
2016-04-12 20:02 ` [CI-ping 03/15] drm/i915: Add GEM debugging Kconfig option Chris Wilson
2016-04-12 20:02 ` [CI-ping 04/15] drm/i915: Hide the atomic_read(reset_counter) behind a helper Chris Wilson
2016-04-12 20:02 ` [CI-ping 05/15] drm/i915: Simplify checking of GPU reset_counter in display pageflips Chris Wilson
2016-04-12 20:03 ` [CI-ping 06/15] drm/i915: Tighten reset_counter for reset status Chris Wilson
2016-04-12 20:03 ` [CI-ping 07/15] drm/i915: Store the reset counter when constructing a request Chris Wilson
2016-04-12 20:03 ` [CI-ping 08/15] drm/i915: Simplify reset_counter handling during atomic modesetting Chris Wilson
2016-04-12 20:03 ` [CI-ping 09/15] drm/i915: Prevent leaking of -EIO from i915_wait_request() Chris Wilson
2016-04-12 20:03 ` [CI-ping 10/15] drm/i915: Suppress error message when GPU resets are disabled Chris Wilson
2016-04-12 20:03 ` [CI-ping 11/15] drm/i915: Prevent machine death on Ivybridge context switching Chris Wilson
2016-04-13  9:33   ` Daniel Vetter
2016-04-13  9:33     ` Daniel Vetter
2016-04-18  9:50     ` [Intel-gfx] " Jani Nikula
2016-04-18  9:50       ` Jani Nikula
2016-04-20 13:26       ` [Intel-gfx] " Jani Nikula
2016-04-12 20:03 ` Chris Wilson [this message]
2016-04-13  9:34   ` [Intel-gfx] [CI-ping 12/15] drm/i915: Force ringbuffers to not be at offset 0 Daniel Vetter
2016-04-12 20:03 ` [CI-ping 13/15] drm/i915: Move the mb() following release-mmap into release-mmap Chris Wilson
2016-04-12 20:03 ` [CI-ping 14/15] drm/i915: Reorganise legacy context switch to cope with late failure Chris Wilson
2016-04-13  9:59   ` Daniel Vetter
2016-04-13 10:05     ` Chris Wilson
2016-04-13 14:16       ` [PATCH 1/2] drm/i915: Split out !RCS legacy context switching Chris Wilson
2016-04-13 14:16         ` [PATCH 2/2] drm/i915: Reorganise legacy context switch to cope with late failure Chris Wilson
2016-04-13 15:05           ` Daniel Vetter
2016-04-13 15:18             ` Chris Wilson
2016-04-13 14:56         ` [PATCH 1/2] drm/i915: Split out !RCS legacy context switching Daniel Vetter
2016-04-13 15:04           ` Chris Wilson
2016-04-12 20:03 ` [CI-ping 15/15] drm/i915: Late request cancellations are harmful Chris Wilson
2016-04-13  9:57   ` Daniel Vetter
2016-04-13  9:57     ` Daniel Vetter
2016-04-13 14:21     ` John Harrison
2016-04-19 12:35       ` Dave Gordon
2016-04-21 13:04         ` John Harrison
2016-04-22 22:57           ` John Harrison
2016-04-27 18:52             ` Dave Gordon
2016-04-18  9:46     ` [Intel-gfx] " Jani Nikula
2016-04-18  9:46       ` Jani Nikula
2016-04-14  8:45 ` ✗ Fi.CI.BAT: failure for series starting with [CI-ping,01/15] drm/i915: Force clean compilation with -Werror (rev3) Patchwork

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=1460491389-8602-12-git-send-email-chris@chris-wilson.co.uk \
    --to=chris@chris-wilson.co.uk \
    --cc=intel-gfx@lists.freedesktop.org \
    --cc=stable@vger.kernel.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.