Skip to contentSkip to conversation

drm/i915: Wait for vblank in i9xx_disable_crtc() for gen 2 only

PATCH Superseded Ander Conselvan de Oliveira 5 participants 8 messages Feb 16, 2016 – Feb 18, 2016

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

PATCH 1/1 drm/i915: Wait for vblank in i9xx_disable_crtc() for gen 2 only

The wait for other gens was added in commit 564ed191f5d8 ("drm/i915:
gmch: fix stuck primary plane due to memory self-refresh mode") since that's necessary when disabling cxsr. However, cxsr disabling was later moved to intel_pre_disable_primary() in commit 87d4300a7dbc ("drm/i915:
Move intel_(pre_disable/post_enable)_primary to intel_display.c, and use it there.") and that function got its own vblank wait for cxsr in commit 262cd2e154c2 ("drm/i915: CHV DDR DVFS support and another watermark rewrite"). So remove the extra vblank wait from i9xx_crtc_distable().

---
 drivers/gpu/drm/i915/intel_display.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)
drivers/gpu/drm/i915/intel_display.c +2 -3
index 568eefc..3cb9383 100644
@@ -6317,10 +6317,9 @@ static void i9xx_crtc_disable(struct drm_crtc *crtc)
 	/*
 	 * On gen2 planes are double buffered but the pipe isn't, so we must
 	 * wait for planes to fully turn off before disabling the pipe.
-	 * We also need to wait on all gmch platforms because of the
-	 * self-refresh mode constraint explained above.
 	 */
-	intel_wait_for_vblank(dev, pipe);
+	if (IS_GEN2(dev))
+		intel_wait_for_vblank(dev, pipe);
 
 	for_each_encoder_on_crtc(dev, crtc, encoder)
 		encoder->disable(encoder);
Signature · 6 lines
2.4.3

_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
#1 lore raw

✗ Fi.CI.BAT: failure for drm/i915: Wait for vblank in i9xx_disable_crtc() for gen 2 only

== Summary ==

== Summary ==

Series 3496v1 drm/i915: Wait for vblank in i9xx_disable_crtc() for gen 2 only
http://patchwork.freedesktop.org/api/1.0/series/3496/revisions/1/mbox/

Test gem_ringfill:

        Subgroup basic-default-hang:
                incomplete -> PASS       (snb-dellxps)
Test kms_flip:
        Subgroup basic-flip-vs-modeset:
                pass       -> DMESG-WARN (ilk-hp8440p) UNSTABLE
Test kms_pipe_crc_basic:
        Subgroup suspend-read-crc-pipe-a:
                dmesg-warn -> PASS       (skl-i5k-2)
        Subgroup suspend-read-crc-pipe-b:
                pass       -> INCOMPLETE (hsw-gt2)
        Subgroup suspend-read-crc-pipe-c:
                pass       -> DMESG-WARN (skl-i5k-2)
                dmesg-warn -> PASS       (bsw-nuc-2)
Test pm_rpm:
        Subgroup basic-pci-d3-state:
                fail       -> PASS       (bdw-nuci7)
        Subgroup basic-rte:
                dmesg-warn -> PASS       (bsw-nuc-2)
bdw-nuci7        total:162  pass:152  dwarn:0   dfail:0   fail:0   skip:10 
bdw-ultra        total:165  pass:152  dwarn:0   dfail:0   fail:0   skip:13 
bsw-nuc-2        total:165  pass:136  dwarn:0   dfail:0   fail:0   skip:29 
byt-nuc          total:165  pass:140  dwarn:1   dfail:0   fail:0   skip:24 
hsw-brixbox      total:165  pass:151  dwarn:0   dfail:0   fail:0   skip:14 
hsw-gt2          total:127  pass:118  dwarn:0   dfail:0   fail:0   skip:8  
ilk-hp8440p      total:165  pass:115  dwarn:1   dfail:0   fail:1   skip:48 
skl-i5k-2        total:165  pass:149  dwarn:1   dfail:0   fail:0   skip:15 
snb-dellxps      total:165  pass:142  dwarn:0   dfail:0   fail:1   skip:22 

Results at /archive/results/CI_IGT_test/Patchwork_1418/

bd0b1a9aa8b7fdb2e06a5cbf1756ef93de2fa3fd drm-intel-nightly: 2016y-02m-16d-17h-53m-05s UTC integration manifest 3623c71eeb82092a0774b8b959a6a367ffca46c2 drm/i915: Wait for vblank in i9xx_disable_crtc() for gen 2 only

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

#2 lore raw

Re: ✗ Fi.CI.BAT: failure for drm/i915: Wait for vblank in i9xx_disable_crtc() for gen 2 only

Quoted 11 lines · Patchwork

== Summary ==

Series 3496v1 drm/i915: Wait for vblank in i9xx_disable_crtc() for gen 2 only
http://patchwork.freedesktop.org/api/1.0/series/3496/revisions/1/mbox/

Test gem_ringfill:

        Subgroup basic-default-hang:
                incomplete -> PASS       (snb-dellxps)
Test kms_flip:
        Subgroup basic-flip-vs-modeset:
                pass       -> DMESG-WARN (ilk-hp8440p) UNSTABLE

https://bugs.freedesktop.org/show_bug.cgi?id=93787

Quoted 5 lines

Test kms_pipe_crc_basic:

        Subgroup suspend-read-crc-pipe-a:
                dmesg-warn -> PASS       (skl-i5k-2)
        Subgroup suspend-read-crc-pipe-b:
                pass       -> INCOMPLETE (hsw-gt2)

I'm pretty sure this is a fluke. The code the patch changes doesn't even run on HSW.

What is the normal procedure for this kind of issues?

Quoted 2 lines
        Subgroup suspend-read-crc-pipe-c:
                pass       -> DMESG-WARN (skl-i5k-2)

https://bugs.freedesktop.org/show_bug.cgi?id=93294

Quoted 17 lines
                dmesg-warn -> PASS       (bsw-nuc-2)
Test pm_rpm:
        Subgroup basic-pci-d3-state:
                fail       -> PASS       (bdw-nuci7)
        Subgroup basic-rte:
                dmesg-warn -> PASS       (bsw-nuc-2)
bdw-nuci7        total:162  pass:152  dwarn:0   dfail:0   fail:0   skip:10 
bdw-ultra        total:165  pass:152  dwarn:0   dfail:0   fail:0   skip:13 
bsw-nuc-2        total:165  pass:136  dwarn:0   dfail:0   fail:0   skip:29 
byt-nuc          total:165  pass:140  dwarn:1   dfail:0   fail:0   skip:24 
hsw-brixbox      total:165  pass:151  dwarn:0   dfail:0   fail:0   skip:14 
hsw-gt2          total:127  pass:118  dwarn:0   dfail:0   fail:0   skip:8  
ilk-hp8440p      total:165  pass:115  dwarn:1   dfail:0   fail:1   skip:48 
skl-i5k-2        total:165  pass:149  dwarn:1   dfail:0   fail:0   skip:15 
snb-dellxps      total:165  pass:142  dwarn:0   dfail:0   fail:1   skip:22 
Quoted 7 lines

Results at /archive/results/CI_IGT_test/Patchwork_1418/

bd0b1a9aa8b7fdb2e06a5cbf1756ef93de2fa3fd drm-intel-nightly: 2016y-02m-16d-17h -53m-05s UTC integration manifest 3623c71eeb82092a0774b8b959a6a367ffca46c2 drm/i915: Wait for vblank in i9xx_disable_crtc() for gen 2 only

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

maintainer
#3 lore raw
In reply to Ander Conselvan de Oliveira
Quoted 11 lines · Ander Conselvan de Oliveira

The wait for other gens was added in commit 564ed191f5d8 ("drm/i915:
gmch: fix stuck primary plane due to memory self-refresh mode") since that's necessary when disabling cxsr. However, cxsr disabling was later moved to intel_pre_disable_primary() in commit 87d4300a7dbc ("drm/i915:
Move intel_(pre_disable/post_enable)_primary to intel_display.c, and use it there.") and that function got its own vblank wait for cxsr in commit 262cd2e154c2 ("drm/i915: CHV DDR DVFS support and another watermark rewrite"). So remove the extra vblank wait from i9xx_crtc_distable().

Cc fixes or stable? Fixes which commit exactly of all those?

BR,
Jani.

Quoted 21 lines
---
 drivers/gpu/drm/i915/intel_display.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)
drivers/gpu/drm/i915/intel_display.c +2 -3
index 568eefc..3cb9383 100644
@@ -6317,10 +6317,9 @@ static void i9xx_crtc_disable(struct drm_crtc *crtc)
 	/*
 	 * On gen2 planes are double buffered but the pipe isn't, so we must
 	 * wait for planes to fully turn off before disabling the pipe.
-	 * We also need to wait on all gmch platforms because of the
-	 * self-refresh mode constraint explained above.
 	 */
-	intel_wait_for_vblank(dev, pipe);
+	if (IS_GEN2(dev))
+		intel_wait_for_vblank(dev, pipe);
 
 	for_each_encoder_on_crtc(dev, crtc, encoder)
 		encoder->disable(encoder);
Signature · 5 lines
Jani Nikula, Intel Open Source Technology Center
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
#4 lore raw
Quoted 11 lines · Jani Nikula
Quoted 9 lines · Ander Conselvan de Oliveira

The wait for other gens was added in commit 564ed191f5d8 ("drm/i915:
gmch: fix stuck primary plane due to memory self-refresh mode") since that's necessary when disabling cxsr. However, cxsr disabling was later moved to intel_pre_disable_primary() in commit 87d4300a7dbc ("drm/i915:
Move intel_(pre_disable/post_enable)_primary to intel_display.c, and use it there.") and that function got its own vblank wait for cxsr in commit 262cd2e154c2 ("drm/i915: CHV DDR DVFS support and another watermark rewrite"). So remove the extra vblank wait from i9xx_crtc_distable().

Quoted 5 lines
Quoted 3 lines

Cc: Kalyan Kondapally <[email protected]>
Signed-off-by: Ander Conselvan de Oliveira < [email protected]>

Cc fixes or stable? Fixes which commit exactly of all those?
Fixes 262cd2e154c2 ("drm/i915: CHV DDR DVFS support and another watermark

rewrite")

Kalyan was seeing some screen corruption with BSW when suspending. He did some experimenting with moving vblank waits around and it seemed to help. When I dug through the history of the code I noticed that we now have a double wait for vblank for cxsr, but I wasn't able to confirm this is actually the fix for that screen corruption. So this may be a candidate for fixes or even stable, but I'm not entirely sure.

I attached an image of what the display corruption looks like, in case someone has any clues.

Thanks,
Ander

Quoted 26 lines

BR,
Jani.

Quoted 21 lines
---
 drivers/gpu/drm/i915/intel_display.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)
drivers/gpu/drm/i915/intel_display.c +0 -0
index 568eefc..3cb9383 100644
@@ -6317,10 +6317,9 @@ static void i9xx_crtc_disable(struct drm_crtc *crtc)
 	/*
 	 * On gen2 planes are double buffered but the pipe isn't, so we

must

drivers/gpu/drm/i915/intel_display.c +2 -3
 	 * wait for planes to fully turn off before disabling the pipe.
-	 * We also need to wait on all gmch platforms because of the
-	 * self-refresh mode constraint explained above.
 	 */
-	intel_wait_for_vblank(dev, pipe);
+	if (IS_GEN2(dev))
+		intel_wait_for_vblank(dev, pipe);
Quoted 3 lines

drivers/gpu/drm/i915/intel_display.c static void i9xx_crtc_disable(struct drm_crtc *crtc) · line 6325

Quoted 2 lines
drivers/gpu/drm/i915/intel_display.c +0 -0
 	for_each_encoder_on_crtc(dev, crtc, encoder)
 		encoder->disable(encoder);

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

#5 lore raw
In reply to Jani Nikula
Quoted 10 lines · Jani Nikula
Quoted 9 lines · Ander Conselvan de Oliveira

The wait for other gens was added in commit 564ed191f5d8 ("drm/i915:
gmch: fix stuck primary plane due to memory self-refresh mode") since that's necessary when disabling cxsr. However, cxsr disabling was later moved to intel_pre_disable_primary() in commit 87d4300a7dbc ("drm/i915:
Move intel_(pre_disable/post_enable)_primary to intel_display.c, and use it there.") and that function got its own vblank wait for cxsr in commit 262cd2e154c2 ("drm/i915: CHV DDR DVFS support and another watermark rewrite"). So remove the extra vblank wait from i9xx_crtc_distable().

Quoted 4 lines
Quoted 2 lines
Cc fixes or stable? Fixes which commit exactly of all those?

Why do we want to put a fix to remove a harmless roughly 16ms delay to stable? Or do I miss something here?
-Daniel

Quoted 33 lines

BR,
Jani.

Quoted 21 lines
---
 drivers/gpu/drm/i915/intel_display.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)
drivers/gpu/drm/i915/intel_display.c +2 -3
index 568eefc..3cb9383 100644
@@ -6317,10 +6317,9 @@ static void i9xx_crtc_disable(struct drm_crtc *crtc)
 	/*
 	 * On gen2 planes are double buffered but the pipe isn't, so we must
 	 * wait for planes to fully turn off before disabling the pipe.
-	 * We also need to wait on all gmch platforms because of the
-	 * self-refresh mode constraint explained above.
 	 */
-	intel_wait_for_vblank(dev, pipe);
+	if (IS_GEN2(dev))
+		intel_wait_for_vblank(dev, pipe);
 
 	for_each_encoder_on_crtc(dev, crtc, encoder)
 		encoder->disable(encoder);
Signature · 5 lines
Jani Nikula, Intel Open Source Technology Center
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Signature · 7 lines
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
maintainer
#6 lore raw
Quoted 11 lines · Daniel Vetter
Quoted 10 lines · Jani Nikula
Quoted 9 lines · Ander Conselvan de Oliveira

The wait for other gens was added in commit 564ed191f5d8 ("drm/i915:
gmch: fix stuck primary plane due to memory self-refresh mode") since that's necessary when disabling cxsr. However, cxsr disabling was later moved to intel_pre_disable_primary() in commit 87d4300a7dbc ("drm/i915:
Move intel_(pre_disable/post_enable)_primary to intel_display.c, and use it there.") and that function got its own vblank wait for cxsr in commit 262cd2e154c2 ("drm/i915: CHV DDR DVFS support and another watermark rewrite"). So remove the extra vblank wait from i9xx_crtc_distable().

Quoted 7 lines
Quoted 4 lines
Quoted 2 lines
Cc fixes or stable? Fixes which commit exactly of all those?

Why do we want to put a fix to remove a harmless roughly 16ms delay to stable? Or do I miss something here?

I missed something here. Nevermind, sorry for the noise. :)

BR,
Jani.

Quoted 35 lines

-Daniel

Quoted 33 lines

BR,
Jani.

Quoted 21 lines
---
 drivers/gpu/drm/i915/intel_display.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)
drivers/gpu/drm/i915/intel_display.c +2 -3
index 568eefc..3cb9383 100644
@@ -6317,10 +6317,9 @@ static void i9xx_crtc_disable(struct drm_crtc *crtc)
 	/*
 	 * On gen2 planes are double buffered but the pipe isn't, so we must
 	 * wait for planes to fully turn off before disabling the pipe.
-	 * We also need to wait on all gmch platforms because of the
-	 * self-refresh mode constraint explained above.
 	 */
-	intel_wait_for_vblank(dev, pipe);
+	if (IS_GEN2(dev))
+		intel_wait_for_vblank(dev, pipe);
 
 	for_each_encoder_on_crtc(dev, crtc, encoder)
 		encoder->disable(encoder);
Signature · 5 lines
Jani Nikula, Intel Open Source Technology Center
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Signature · 5 lines
Jani Nikula, Intel Open Source Technology Center
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Reviewed-by on v1 1/1
#7 lore raw
In reply to Ander Conselvan de Oliveira
Quoted 11 lines · Ander Conselvan de Oliveira

The wait for other gens was added in commit 564ed191f5d8 ("drm/i915:
gmch: fix stuck primary plane due to memory self-refresh mode") since that's necessary when disabling cxsr. However, cxsr disabling was later moved to intel_pre_disable_primary() in commit 87d4300a7dbc ("drm/i915:
Move intel_(pre_disable/post_enable)_primary to intel_display.c, and use it there.") and that function got its own vblank wait for cxsr in commit 262cd2e154c2 ("drm/i915: CHV DDR DVFS support and another watermark rewrite"). So remove the extra vblank wait from i9xx_crtc_distable().

Quoted 28 lines
---
 drivers/gpu/drm/i915/intel_display.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)
drivers/gpu/drm/i915/intel_display.c +2 -3
index 568eefc..3cb9383 100644
@@ -6317,10 +6317,9 @@ static void i9xx_crtc_disable(struct drm_crtc *crtc)
 	/*
 	 * On gen2 planes are double buffered but the pipe isn't, so we must
 	 * wait for planes to fully turn off before disabling the pipe.
-	 * We also need to wait on all gmch platforms because of the
-	 * self-refresh mode constraint explained above.
 	 */
-	intel_wait_for_vblank(dev, pipe);
+	if (IS_GEN2(dev))
+		intel_wait_for_vblank(dev, pipe);
 
 	for_each_encoder_on_crtc(dev, crtc, encoder)
 		encoder->disable(encoder);
Signature · 6 lines
2.4.3

_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Signature · 7 lines
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx