[PATCH 10/47] drm/i915: Suppress switch_mm emission between the same aliasing_ppgtt

Chris Wilson chris at chris-wilson.co.uk
Sat Feb 18 00:03:48 UTC 2017


When switching between contexts using the aliasing_ppgtt, the VM is
shared. We don't need to reload the PD registers unless they are dirty.

Martin Peres reported an issue that looks like corruption between
Haswell context switches, bisecting to commit f9326be5f1d3 ("drm/i915:
Rearrange switch_context to load the aliasing ppgtt on first use").
Switching between the same mm (the aliasing_ppgtt is used for all
contexts in this case) should be a nop, but appears to trigger some
side-effects in the context switch. However, as we know the switch
is redundant in this case, we can skip it and continue to ignore the
issue until somebody feels strong enough to investigate full-ppgtt on
gen7 again!

Except.. Martin was using full-ppgtt which is not supported as it
doesn't work correctly yet. So whilst the bisect did yield valuable
information about the failures, the fix should not have any user impact
under default settings.

Fixes: f9326be5f1d3 ("drm/i915: Rearrange switch_context to load the aliasing ppgtt on first use")
Reported-by: Martin Peres <martin.peres at linux.intel.com>
Signed-off-by: Chris Wilson <chris at chris-wilson.co.uk>
Cc: Martin Peres <martin.peres at linux.intel.com>
Reviewed-by: Joonas Lahtinen <joonas.lahtinen at linux.intel.com>
---
 drivers/gpu/drm/i915/i915_gem_context.c | 14 ++++++++------
 1 file changed, 8 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_gem_context.c b/drivers/gpu/drm/i915/i915_gem_context.c
index 39273a407592..ee4097bfb94a 100644
--- a/drivers/gpu/drm/i915/i915_gem_context.c
+++ b/drivers/gpu/drm/i915/i915_gem_context.c
@@ -707,10 +707,10 @@ static inline bool skip_rcs_switch(struct i915_hw_ppgtt *ppgtt,
 }
 
 static bool
-needs_pd_load_pre(struct i915_hw_ppgtt *ppgtt,
-		  struct intel_engine_cs *engine,
-		  struct i915_gem_context *to)
+needs_pd_load_pre(struct i915_hw_ppgtt *ppgtt, struct intel_engine_cs *engine)
 {
+	struct i915_hw_ppgtt *last_ppgtt;
+
 	if (!ppgtt)
 		return false;
 
@@ -719,7 +719,9 @@ needs_pd_load_pre(struct i915_hw_ppgtt *ppgtt,
 		return true;
 
 	/* Same context without new entries, skip */
-	if (engine->legacy_active_context == to &&
+	last_ppgtt =
+		engine->legacy_active_context->ppgtt ?: engine->i915->mm.aliasing_ppgtt;
+	if (last_ppgtt == ppgtt &&
 	    !(intel_engine_flag(engine) & ppgtt->pd_dirty_rings))
 		return false;
 
@@ -763,7 +765,7 @@ static int do_rcs_switch(struct drm_i915_gem_request *req)
 	if (skip_rcs_switch(ppgtt, engine, to))
 		return 0;
 
-	if (needs_pd_load_pre(ppgtt, engine, to)) {
+	if (needs_pd_load_pre(ppgtt, engine)) {
 		/* Older GENs and non render rings still want the load first,
 		 * "PP_DCLV followed by PP_DIR_BASE register through Load
 		 * Register Immediate commands in Ring Buffer before submitting
@@ -860,7 +862,7 @@ int i915_switch_context(struct drm_i915_gem_request *req)
 		struct i915_hw_ppgtt *ppgtt =
 			to->ppgtt ?: req->i915->mm.aliasing_ppgtt;
 
-		if (needs_pd_load_pre(ppgtt, engine, to)) {
+		if (needs_pd_load_pre(ppgtt, engine)) {
 			int ret;
 
 			trace_switch_mm(engine, to);
-- 
2.11.0



More information about the Intel-gfx-trybot mailing list