From d41a3c2be178783c85e05025265ab58fbb4d4ce1 Mon Sep 17 00:00:00 2001
From: Chris Wilson <chris@chris-wilson.co.uk>
Date: Mon, 7 Aug 2017 13:19:19 +0100
Subject: [PATCH] drm/i915: Clear lost context-switch interrupts across reset
Git-commit: d41a3c2be178783c85e05025265ab58fbb4d4ce1
Patch-mainline: v4.13-rc7
References: FATE#322643 bsc#1055900
No-fix: 64f09f00caf0a7cb40a8c0b85789bacba0f51d9e
During a global reset, we disable the irq. As we disable the irq, the
hardware may be raising a GT interrupt that we then ignore, leaving it
pending in the GTIIR. After the reset, we then re-enable the irq,
triggering the pending interrupt. However, that interrupt was for the
stale state from before the reset, and the contents of the CSB buffer
are now invalid.
V2: Add a comment to make it clear that the double clear is purely my
paranoia.
Reported-by: "Dong, Chuanxiao" <chuanxiao.dong@intel.com>
Fixes: 821ed7df6e2a ("drm/i915: Update reset path to fix incomplete requests")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: "Dong, Chuanxiao" <chuanxiao.dong@intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Michal Winiarski <michal.winiarski@intel.com>
Cc: Michel Thierry <michel.thierry@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20170807121919.30165-1-chris@chris-wilson.co.uk
Link: https://patchwork.freedesktop.org/patch/msgid/20170818090509.5363-1-chris@chris-wilson.co.uk
Reviewed-by: Michel Thierry <michel.thierry@intel.com>
(cherry picked from commit 64f09f00caf0a7cb40a8c0b85789bacba0f51d9e)
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Acked-by: Takashi Iwai <tiwai@suse.de>
---
drivers/gpu/drm/i915/intel_lrc.c | 23 ++++++++++++++++++++++-
1 file changed, 22 insertions(+), 1 deletion(-)
--- a/drivers/gpu/drm/i915/intel_lrc.c
+++ b/drivers/gpu/drm/i915/intel_lrc.c
@@ -1221,6 +1221,14 @@ static int intel_init_workaround_bb(stru
return ret;
}
+static u8 gtiir[] = {
+ [RCS] = 0,
+ [BCS] = 0,
+ [VCS] = 1,
+ [VCS2] = 1,
+ [VECS] = 3,
+};
+
static int gen8_init_common_ring(struct intel_engine_cs *engine)
{
struct drm_i915_private *dev_priv = engine->i915;
@@ -1245,9 +1253,22 @@ static int gen8_init_common_ring(struct
DRM_DEBUG_DRIVER("Execlists enabled for %s\n", engine->name);
- /* After a GPU reset, we may have requests to replay */
+ GEM_BUG_ON(engine->id >= ARRAY_SIZE(gtiir));
+
+ /*
+ * Clear any pending interrupt state.
+ *
+ * We do it twice out of paranoia that some of the IIR are double
+ * buffered, and if we only reset it once there may still be
+ * an interrupt pending.
+ */
+ I915_WRITE(GEN8_GT_IIR(gtiir[engine->id]),
+ GT_CONTEXT_SWITCH_INTERRUPT << engine->irq_shift);
+ I915_WRITE(GEN8_GT_IIR(gtiir[engine->id]),
+ GT_CONTEXT_SWITCH_INTERRUPT << engine->irq_shift);
clear_bit(ENGINE_IRQ_EXECLIST, &engine->irq_posted);
+ /* After a GPU reset, we may have requests to replay */
submit = false;
for (n = 0; n < ARRAY_SIZE(engine->execlist_port); n++) {
if (!port_isset(&port[n]))