Messages in this thread | | | From | Eric Anholt <> | Subject | [git pull] drm-intel fixes for 2.6.33rc3 | Date | Wed, 06 Jan 2010 10:17:15 -0800 |
| |
The following changes since commit 8bea8672edfca7ec5f661cafb218f1205863b343: Stephen Rothwell (1): mfd: compile fix for twl4030 renaming
are available in the git repository at:
git://git.kernel.org/pub/scm/linux/kernel/git/anholt/drm-intel drm-intel-next
The only scary part in here should be the new execbuf2 ioctl. A little background on that one: We've been passing around this patch and hammering it out for much longer than its author date suggests. I think the first iteration was around july or so. It'll let us fix a big performance hit on pre-965 chipsets, particularly when windows or textures get to 1024 pixels wide (but reducing in severity when they get to around 1280 pixels wide). We've now had multiple userland developers spend quite a while trying to track down that strange performance valley (at a suddenly popular again resolution). I've been testing the compatibility side of it, while Chris Wilson's been using the interface a lot in his performance work.
Highlights for other people: Removing render reclock may fix hangs some people have seen since we introduced the power stuff. More Ironlake love. And a hibernate fix.
Andrew Lutomirski (1): drm/i915: Fix RC6 suspend/resume
Andrew Morton (1): drm/i915: fix unused var
Chris Wilson (2): drm/i915: Hold struct mutex whilst pinning power context bo. drm/i915: Permit pinning whilst the device is 'suspended'
Daniel Vetter (1): drm/i915: fix order of fence release wrt flushing
Eric Anholt (1): drm/i915: In the debugfs interface, unmap our address instead of the page's.
Jesse Barnes (3): drm/i915: only enable hotplug for detected outputs drm/i915: execbuf2 support drm/i915: remove render reclock support
Kristian Høgsberg (3): drm/i915: Move PCI IDs into i915 driver drm/i915: Implement IS_* macros using static tables drm/i915: Track whether cursor needs physical address in intel_device_info
Matthew Garrett (1): drm/i915: Don't check for lid presence when detecting LVDS
Tobias Klauser (1): drm/i915: Storage class should be before const qualifier
Zhao Yakui (6): drm/i915: Add MALATA PC-81005 to ACPI LID quirk list drm/i915: Update LVDS connector status when receiving ACPI LID event drm/i915: Enable/disable the dithering for LVDS based on VBT setting drm/i915: Make the BPC in FDI rx/transcoder be consistent with that in pipeconf on Ironlake drm/i915: Select the correct BPC for LVDS on Ironlake drm/i915: Add DP dpll limit on ironlake and use existing DPLL search function
Zhenyu Wang (3): drm/i915: implement new pm ops for i915 drm/i915: Reload hangcheck timer too for Ironlake drm/i915: remove full registers dump debug
drivers/gpu/drm/i915/i915_debugfs.c | 31 +--- drivers/gpu/drm/i915/i915_dma.c | 26 +- drivers/gpu/drm/i915/i915_drv.c | 174 +++++++++++++- drivers/gpu/drm/i915/i915_drv.h | 123 +++++------ drivers/gpu/drm/i915/i915_gem.c | 249 +++++++++++++++----- drivers/gpu/drm/i915/i915_gem_tiling.c | 46 ++-- drivers/gpu/drm/i915/i915_irq.c | 32 ++- drivers/gpu/drm/i915/i915_reg.h | 11 +- drivers/gpu/drm/i915/i915_suspend.c | 12 - drivers/gpu/drm/i915/intel_crt.c | 2 + drivers/gpu/drm/i915/intel_display.c | 414 ++++++++++++++------------------ drivers/gpu/drm/i915/intel_dp.c | 6 + drivers/gpu/drm/i915/intel_hdmi.c | 5 + drivers/gpu/drm/i915/intel_lvds.c | 83 ++----- drivers/gpu/drm/i915/intel_sdvo.c | 3 + drivers/gpu/drm/i915/intel_tv.c | 2 + include/drm/i915_drm.h | 54 ++++ 17 files changed, 736 insertions(+), 537 deletions(-) [unhandled content-type:application/pgp-signature] | |