From a24a11e6b4e96bca817f854e0ffcce75d3eddd13 Mon Sep 17 00:00:00 2001 From: Chris Wilson Date: Thu, 14 Mar 2013 17:52:05 +0200 Subject: [PATCH] drm/i915: Resurrect ring kicking for semaphores, selectively Once we thought we got semaphores working, we disabled kicking the ring if hangcheck fired whilst waiting upon a ring as it was doing more harm than good: commit 4e0e90dcb8a7df1229c69e30abebb59b0b3c2a1f Author: Daniel Vetter Date: Wed Dec 14 13:56:58 2011 +0100 drm/i915: kicking rings stuck on semaphores considered harmful However, life is never that easy and semaphores are still causing problems whereby the value written by one ring (bcs) is not being propagated to the waiter (rcs). Thus the waiter never wakes up and we declare the GPU hung, which often has unfortunate consequences, even if we successfully reset the GPU. But the GPU is idle as it has completed the work, just didn't notify its clients. So we can detect the incomplete wait during hang check and probe the target ring to see if has indeed emitted the breadcrumb seqno following the work and then and only then kick the waiter. Based on a suggestion by Ben Widawsky. v2: cross-check wait with iphdr. fix signaller calculation. References: https://bugs.freedesktop.org/show_bug.cgi?id=54226 Signed-off-by: Chris Wilson Signed-off-by: Mika Kuoppala Cc: Daniel Vetter Cc: Ben Widawsky Acked-by: Ben Widawsky Signed-off-by: Daniel Vetter --- drivers/gpu/drm/i915/i915_irq.c | 40 +++++++++++++++++++++++++++++++++ 1 file changed, 40 insertions(+) diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c index 39621e50ca5e..63abf2fa1fa4 100644 --- a/drivers/gpu/drm/i915/i915_irq.c +++ b/drivers/gpu/drm/i915/i915_irq.c @@ -1787,6 +1787,37 @@ static bool i915_hangcheck_ring_idle(struct intel_ring_buffer *ring, bool *err) return false; } +static bool semaphore_passed(struct intel_ring_buffer *ring) +{ + struct drm_i915_private *dev_priv = ring->dev->dev_private; + u32 acthd = intel_ring_get_active_head(ring) & HEAD_ADDR; + struct intel_ring_buffer *signaller; + u32 cmd, ipehr, acthd_min; + + ipehr = I915_READ(RING_IPEHR(ring->mmio_base)); + if ((ipehr & ~(0x3 << 16)) != + (MI_SEMAPHORE_MBOX | MI_SEMAPHORE_COMPARE | MI_SEMAPHORE_REGISTER)) + return false; + + /* ACTHD is likely pointing to the dword after the actual command, + * so scan backwards until we find the MBOX. + */ + acthd_min = max((int)acthd - 3 * 4, 0); + do { + cmd = ioread32(ring->virtual_start + acthd); + if (cmd == ipehr) + break; + + acthd -= 4; + if (acthd < acthd_min) + return false; + } while (1); + + signaller = &dev_priv->ring[(ring->id + (((ipehr >> 17) & 1) + 1)) % 3]; + return i915_seqno_passed(signaller->get_seqno(signaller, false), + ioread32(ring->virtual_start+acthd+4)+1); +} + static bool kick_ring(struct intel_ring_buffer *ring) { struct drm_device *dev = ring->dev; @@ -1798,6 +1829,15 @@ static bool kick_ring(struct intel_ring_buffer *ring) I915_WRITE_CTL(ring, tmp); return true; } + + if (INTEL_INFO(dev)->gen >= 6 && + tmp & RING_WAIT_SEMAPHORE && + semaphore_passed(ring)) { + DRM_ERROR("Kicking stuck semaphore on %s\n", + ring->name); + I915_WRITE_CTL(ring, tmp); + return true; + } return false; } -- 2.30.2