Skip to contentSkip to conversation

drm/i915: avoid unclaimed registers when capturing the error state

PATCH In review Paulo Zanoni 3 participants 4 messages Dec 20, 2013 – Jun 2, 2014

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

PATCH 1/1 drm/i915: avoid unclaimed registers when capturing the error state

From: Paulo Zanoni <[email protected]>

We're iterating over the CPU transcoders, so check for the correct power domain.

This fixes many "unclaimed register" error messages.

This can be reproduced by the IGT test mentioned below, but we still get a FAIL when we run it.

Testcase: igt/kms_lip/flip-vs-panning-vs-hang

---
 drivers/gpu/drm/i915/intel_display.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

I didn't bisect to check, but it really really looks like a regression from "drm/i915: add intel_display_power_enabled_sw() for use in atomic ctx".

drivers/gpu/drm/i915/intel_display.c +2 -1
index 4d1357a..4d4b4bc 100644
@@ -11459,7 +11459,8 @@ intel_display_capture_error_state(struct drm_device *dev)
 		enum transcoder cpu_transcoder = transcoders[i];
 
 		error->transcoder[i].power_domain_on =
-			intel_display_power_enabled_sw(dev, POWER_DOMAIN_PIPE(i));
+			intel_display_power_enabled_sw(dev,
+				POWER_DOMAIN_TRANSCODER(cpu_transcoder));
 		if (!error->transcoder[i].power_domain_on)
 			continue;
 
Signature · 1 line
1.8.3.1
Reviewed-by on v1 1/1
#1 lore raw
Quoted 11 lines · Paulo Zanoni

From: Paulo Zanoni <[email protected]>

We're iterating over the CPU transcoders, so check for the correct power domain.

This fixes many "unclaimed register" error messages.

This can be reproduced by the IGT test mentioned below, but we still get a FAIL when we run it.

Testcase: igt/kms_lip/flip-vs-panning-vs-hang

Quoted 8 lines
---
 drivers/gpu/drm/i915/intel_display.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

I didn't bisect to check, but it really really looks like a regression from "drm/i915: add intel_display_power_enabled_sw() for use in atomic ctx".

Yep, that was an overlook on my part:/ The fix looks ok:

Quoted 8 lines
drivers/gpu/drm/i915/intel_display.c +0 -0
index 4d1357a..4d4b4bc 100644

drivers/gpu/drm/i915/intel_display.c intel_display_capture_error_state(struct drm_device *dev) · line 11466

Quoted 9 lines
drivers/gpu/drm/i915/intel_display.c +2 -1
@@ -11459,7 +11459,8 @@ intel_display_capture_error_state(struct drm_device *dev)
 		enum transcoder cpu_transcoder = transcoders[i];
 
 		error->transcoder[i].power_domain_on =
-			intel_display_power_enabled_sw(dev, POWER_DOMAIN_PIPE(i));
+			intel_display_power_enabled_sw(dev,
+				POWER_DOMAIN_TRANSCODER(cpu_transcoder));
 		if (!error->transcoder[i].power_domain_on)
 			continue;
 

_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

#2 lore raw
In reply to Paulo Zanoni
Quoted 5 lines · Paulo Zanoni

From: Paulo Zanoni <[email protected]>

We're iterating over the CPU transcoders, so check for the correct power domain.

Quoted 7 lines

This fixes many "unclaimed register" error messages.

This can be reproduced by the IGT test mentioned below, but we still get a FAIL when we run it.

Testcase: igt/kms_lip/flip-vs-panning-vs-hang

This of course breaks error capture on every platform that doesn't use rpm, and even then is spectacularly useless on those that do. -Chris

Signature · 1 line
Chris Wilson, Intel Open Source Technology Centre
#3 lore raw
Quoted 8 lines · Chris Wilson
Quoted 7 lines · Paulo Zanoni

From: Paulo Zanoni <[email protected]>

We're iterating over the CPU transcoders, so check for the correct power domain.

This fixes many "unclaimed register" error messages.

Quoted 8 lines
Quoted 5 lines

This can be reproduced by the IGT test mentioned below, but we still get a FAIL when we run it.

Testcase: igt/kms_lip/flip-vs-panning-vs-hang

This of course breaks error capture on every platform that doesn't use rpm, and even then is spectacularly useless on those that do.

It is broken atm, because of the bug in intel_display_power_enabled_sw(). With that fixed these checks should work everywhere, since POWER_DOMAIN_INIT is always held on platforms w/o RPM.

--Imre

_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx