drm-intel fixes since 2.6.35-rc4
Subsystems (2)
The following changes since commit 815c4163b6c8ebf8152f42b0a5fd015cfdcedc78:
Linux 2.6.35-rc4 (2010-07-04 20:22:50 -0700)
are available in the git repository at:
git://git.kernel.org/pub/scm/linux/kernel/git/anholt/drm-intel for-linus
There's been a massive bugfixing spree in the last couple of weeks .
jbarnes and chris have reviewed this series for sending to you, and I'm running it on my development system, so I feel pretty good about sending it despite being at rc6.
Adam Jackson (1):
drm/i915: Make G4X-style PLL search more permissive
Chris Wilson (4):
drm/i915: Explosion following OOM in do_execbuffer.
drm/i915: Clear any existing dither mode prior to enabling spatial dithering drm/i915: Use the correct scanout alignment for fbcon. drm/i915: Fix panel fitting regression since 734b4157
Jesse Barnes (7):
drm/i915: handle shared framebuffers when flipping drm/i915: add PANEL_UNLOCK_REGS definition drm/i915: make sure eDP panel is turned on drm/i915: disable FBC when more than one pipe is active drm/i915: don't free non-existent compressed llb on ILK+ drm/i915: fix deadlock in fb teardown drm/i915: add pipe A force quirks to i915 driver
Thomas Bächler (1):
gpu/drm/i915: Add a blacklist to omit modeset on LID open
Tim Gardner (1):
agp/intel: Use the correct mask to detect i830 aperture size.
drivers/char/agp/intel-gtt.c | 13 +++-- drivers/gpu/drm/i915/i915_debugfs.c | 3 + drivers/gpu/drm/i915/i915_dma.c | 2 +- drivers/gpu/drm/i915/i915_drv.h | 5 ++ drivers/gpu/drm/i915/i915_gem.c | 3 +- drivers/gpu/drm/i915/i915_reg.h | 1 + drivers/gpu/drm/i915/intel_display.c | 109 ++++++++++++++++++++++++++++++---- drivers/gpu/drm/i915/intel_dp.c | 53 ++++++++++++++++- drivers/gpu/drm/i915/intel_drv.h | 3 + drivers/gpu/drm/i915/intel_fb.c | 4 +- drivers/gpu/drm/i915/intel_lvds.c | 23 +++++++ 11 files changed, 197 insertions(+), 22 deletions(-)
Quoted 6 lines · Eric Anholt
The following changes since commit
815c4163b6c8ebf8152f42b0a5fd015cfdcedc78:Linux 2.6.35-rc4 (2010-07-04 20:22:50 -0700)
are available in the git repository at:
git://git.kernel.org/pub/scm/linux/kernel/git/anholt/drm-intel for-linus
Hmm. This seems to introduce a new warning:
drivers/gpu/drm/i915/intel_dp.c:744: warning:
‘ironlake_edp_panel_off’ defined but not used
when I pulled it.
And the warning appears to be due to a real bug: instead of calling ironlake_edp_panel_off(), the code calls ironlake_edp_backlight_off() twice. I assume that the _intention_ was to call both the backlight and panel "off" routines, rather than making doubly sure that just the backlight is off by turning it off twice.
Tssk, tssk. Sadly, I noticed this too late, so now it's pulled and pushed out.
Linus
Quoted 15 lines · On Mon, 26 Jul 2010 13:19:14 -0700 Linus Torvalds…
Quoted 6 lines · Eric Anholt
The following changes since commit
815c4163b6c8ebf8152f42b0a5fd015cfdcedc78:Linux 2.6.35-rc4 (2010-07-04 20:22:50 -0700)
are available in the git repository at:
git://git.kernel.org/pub/scm/linux/kernel/git/anholt/drm-intel for-linusHmm. This seems to introduce a new warning:
drivers/gpu/drm/i915/intel_dp.c:744: warning:
‘ironlake_edp_panel_off’ defined but not usedwhen I pulled it.
Quoted 7 lines
And the warning appears to be due to a real bug: instead of calling
ironlake_edp_panel_off(), the code callsironlake_edp_backlight_off()twice. I assume that the _intention_ was to call both the backlight and panel "off" routines, rather than making doubly sure that just the backlight is off by turning it off twice.Tssk, tssk. Sadly, I noticed this too late, so now it's pulled and pushed out.
Arg. Well at least the patch has been tested and is confirmed working. Fix for the warning on its way now (just going to test it to make extra sure it still works when the panel actually gets shut off!)
Signature · 1 line
Jesse Barnes, Intel Open Source Technology Center
Here you go. Tested on the problem machine, works fine.
Quoted 1 line
From
a9daaec5a40d572d11f689edac5cb1710c998b1eMon Sep 17 00:00:00 2001
From: Jesse Barnes <[email protected]>
Date: Mon, 26 Jul 2010 13:51:22 -0700
Subject: [PATCH] drm/i915: make sure we shut off the panel in eDP configs
Merge error from the last pull request. Making sure we shut the panel off is more correct and saves power.
--- drivers/gpu/drm/i915/intel_dp.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-)
drivers/gpu/drm/i915/intel_dp.c +1 -1
@@ -798,7 +798,7 @@ intel_dp_dpms(struct drm_encoder *encoder, int mode)intel_dp_link_down (intel_encoder, dp_priv->DP);if (IS_eDP (intel_encoder)) {ironlake_edp_backlight_off (dev); -ironlake_edp_backlight_off (dev); +ironlake_edp_panel_off (dev); } } }else {
Signature · 1 line
1.7.1
Quoted 15 lines · Linus Torvalds
Quoted 6 lines · Eric Anholt
The following changes since commit
815c4163b6c8ebf8152f42b0a5fd015cfdcedc78:Linux 2.6.35-rc4 (2010-07-04 20:22:50 -0700)
are available in the git repository at:
git://git.kernel.org/pub/scm/linux/kernel/git/anholt/drm-intel for-linusHmm. This seems to introduce a new warning:
drivers/gpu/drm/i915/intel_dp.c:744: warning:
‘ironlake_edp_panel_off’ defined but not usedwhen I pulled it.
Quoted 7 lines
And the warning appears to be due to a real bug: instead of calling
ironlake_edp_panel_off(), the code callsironlake_edp_backlight_off()twice. I assume that the _intention_ was to call both the backlight and panel "off" routines, rather than making doubly sure that just the backlight is off by turning it off twice.Tssk, tssk. Sadly, I noticed this too late, so now it's pulled and pushed out.
And I managed to botch that while hand-resolving the patch that was sent against the wrong tree instead of bouncing it. I'm going to be more strict on that in the future.