All of lore.kernel.org
 help / color / mirror / Atom feed
* [PATCH] RFT drm/i915/execlists: Flush memory before signaling ELSQ
@ 2018-10-31  9:02 Chris Wilson
  2018-10-31 12:14 ` ✓ Fi.CI.BAT: success for " Patchwork
                   ` (28 more replies)
  0 siblings, 29 replies; 32+ messages in thread
From: Chris Wilson @ 2018-10-31  9:02 UTC (permalink / raw)
  To: intel-gfx

We observe that the ordering of writes for a CS event is not as strong
from the GPU as we would like, and that on occasions we see the
ringbuffer tail updated before the event is written into the ringbuffer,
leading us to reuse the stale data.

Through around a big hammer to try and batter ELSQ into submission with
the presumption that perhaps the UC mmio write is not flushing our
writes into the context images.

References: https://bugs.freedesktop.org/show_bug.cgi?id=108315
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
---
 drivers/gpu/drm/i915/intel_lrc.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/intel_lrc.c b/drivers/gpu/drm/i915/intel_lrc.c
index 22b57b8926fc..ba61849fbb9b 100644
--- a/drivers/gpu/drm/i915/intel_lrc.c
+++ b/drivers/gpu/drm/i915/intel_lrc.c
@@ -454,8 +454,10 @@ static void execlists_submit_ports(struct intel_engine_cs *engine)
 	}
 
 	/* we need to manually load the submit queue */
-	if (execlists->ctrl_reg)
+	if (execlists->ctrl_reg) {
+		wmb(); /* XXX Big hammer or paper? XXX */
 		writel(EL_CTRL_LOAD, execlists->ctrl_reg);
+	}
 
 	execlists_clear_active(execlists, EXECLISTS_ACTIVE_HWACK);
 }
-- 
2.19.1

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

^ permalink raw reply related	[flat|nested] 32+ messages in thread

end of thread, other threads:[~2018-11-02 11:30 UTC | newest]

Thread overview: 32+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2018-10-31  9:02 [PATCH] RFT drm/i915/execlists: Flush memory before signaling ELSQ Chris Wilson
2018-10-31 12:14 ` ✓ Fi.CI.BAT: success for " Patchwork
2018-10-31 14:05 ` Patchwork
2018-10-31 14:35 ` Patchwork
2018-10-31 14:49 ` [PATCH] " Mika Kuoppala
2018-10-31 15:06   ` Chris Wilson
2018-10-31 15:29 ` ✓ Fi.CI.BAT: success for " Patchwork
2018-10-31 15:51 ` ✓ Fi.CI.IGT: " Patchwork
2018-10-31 16:03 ` ✓ Fi.CI.BAT: " Patchwork
2018-10-31 16:37 ` Patchwork
2018-10-31 16:43   ` Chris Wilson
2018-10-31 17:39 ` Patchwork
2018-10-31 17:55 ` ✓ Fi.CI.IGT: " Patchwork
2018-10-31 18:20 ` ✓ Fi.CI.BAT: " Patchwork
2018-10-31 19:17 ` ✓ Fi.CI.IGT: " Patchwork
2018-10-31 20:40 ` Patchwork
2018-10-31 21:04 ` ✓ Fi.CI.BAT: " Patchwork
2018-10-31 22:02 ` ✓ Fi.CI.IGT: " Patchwork
2018-10-31 22:11 ` ✓ Fi.CI.BAT: " Patchwork
2018-10-31 23:21 ` ✓ Fi.CI.IGT: " Patchwork
2018-11-01  1:24 ` Patchwork
2018-11-01  2:47 ` Patchwork
2018-11-01  6:12 ` Patchwork
2018-11-01  8:16 ` Patchwork
2018-11-01  9:06 ` ✓ Fi.CI.BAT: " Patchwork
2018-11-01 11:39 ` ✓ Fi.CI.IGT: " Patchwork
2018-11-01 14:37 ` ✓ Fi.CI.BAT: " Patchwork
2018-11-01 16:24 ` ✓ Fi.CI.IGT: " Patchwork
2018-11-01 17:44 ` ✓ Fi.CI.BAT: " Patchwork
2018-11-01 19:54 ` ✓ Fi.CI.IGT: " Patchwork
2018-11-02  9:49 ` ✓ Fi.CI.BAT: " Patchwork
2018-11-02 11:30 ` ✓ Fi.CI.IGT: " Patchwork

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.