[ANNOUNCE] xf86-video-intel 2.2.0
jbarnes at virtuousgeek.org
Thu Nov 15 12:48:48 PST 2007
Well, this is it, Intel driver release 2.2.0, available from all the usual
locations. Thanks are owed to the many bug reporters, testers and developers
who contributed to this release.
Since the 2.1.0 release, over 160 bugs have been resolved. That still leaves
96 open bugs (yay for breaking the 100 barrier), but at this rate the next
release should be bug free! :) Moreover, I think many of the open bugs are
duplicates of bugs fixed in this release, or just need to be tested with this
release before being closed out, so things are looking good on the bug front.
Again, this driver should be run with the latest released versions of the
- X server 1.4.x
- Mesa 7.0.x
- DRM modules from your respective kernel
Running with more recent versions should be possible, but you'll likely hit
bugs in the upstream trees due to active development going on there.
If you find problems with this release, don't hesitate to file them at
bugs.freedesktop.org. There have been many changes since the last release, so
we undoubtedly added a few new bugs in the process.
The (large) git changelog for this release is shown below.
Alan Coopersmith (2):
Clear compiler error: "void functions cannot return values"
Fix builds without DRI
Brice Goglin (2):
Fix typo in intel.man
Define INTEL_VERSION_MAJOR/MINOR/PATCH using PACKAGE_VERSION_*
Carl Worth (1):
Allow 965 composite acceleration to A8 destinations.
Dave Airlie (9):
intel: don't try and use TTM memory manager with old libdrm interface
intel: actually 2.3.1 should be good enough
intel: oops I commited pixman local workaround - undo it
strip out remainder of drmmm code in driver
i965: fix memcpy of the sf_kernel when a mask is needed
i965: increase composite vertex buffer size and alignment to be safe
intel: don't setup texOffsetStart unless using EXA
i915: add support for render to a8
intel: make block handler hook happen no matter what video is in use
Dmitry Babrovich (1):
Bug #12457: Let the ch7xxx driver probe the 7301
Eric Anholt (53):
Fix some physical address handling for >4GB addresses.
When TTM is available, use it instead of manual AGP allocations when possible.
Use new drmBOSetPin interface instead of NO_EVICT/NO_MOVE buffers.
Use the dontMapFramebuffer option available with DRIINFO 5.4.
Delay the first screen pixmap update to CreateScreenResources.
Pin some buffer objects at creation time, which can't be moved yet.
Mark DRI buffers as shareable, and pass their buffer handles through the SAREA.
Quirk away the nonexistent TV connector on the Panasonic CF-Y4.
Add the file mode for bios_dumper output so it doesn't have 000 permissions.
Fix accumulated whitespace nits in i830_exa.c
Fix EXA rendering with tiled front buffer on pre-965.
Bug #11593: Remove dead struct vch_bdb_20 which was angering the sun compiler.
Attempt to fix several front buffer tiling failure cases.
Fix stack-smashing in the last commit.
Don't force tiling on if it is disabled in configuration but fbc is possible.
Add #if 0-ed fence debugging code. It's noisy, and of little use to most.
Add tiling information to BO layout description.
Use i830_memory.c instead of the AA's allocator for XV buffers.
Replace AA allocator usage with i830_memory.c for RandR rotation.
Tune acceleration architecture allocator sizes down.
Remove logic for supporting i915tex_dri.so vs. i915_dri.so.
Disable TV/VGA output on the CH7017/7018 devices.
Enable CH7017/7018 DVO driver by default.
Suppress I2C failure error messages during DVO device detection.
Remove more DVO probing noise.
Restore building of pciaccess-based tools even if the server doesn't use it.
Fix probing of the sample CH7017 device I found by allowing GPIO overrides.
Bring the CH7017 driver closer to spec.
Fix formatting of error message.
Don't leak buffer object allocations in i830_reset_allocations().
Revert setting of exa_965_state as LIFETIME_FIXED.
Don't set overlay registers LIFETIME_FIXED. It always uses the current offset.
Replace setting of LIFETIME_FIXED on cursors with just updating the offsets.
Fix potential use-after-free in XV overlay code on video stop.
In i830_allocate_memory_bo, bind if we control the VT, not on lifetime-fixed.
Refuse to allocate LIFETIME_FIXED objects in buffer objects.
Move drmMMLock to after we have unbound our (pinned) buffers.
Rework DRI buffer mappings and sarea setup to allow for moving buffers.
Delay SAREA and mapping setup until EnterVT when using the memory manager.
Allow front/back/depth to move over the lifetime of the server.
Move tiling fence register setup to bind time instead of allocate time.
Update memory manager sizing for the current set of LIFETIME_FIXED bufffers.
Don't double-free the memory manager allocation.
Use mprotect on unbound AGP memory to attempt to catch use while unbound.
Fix failure in tiling setup on non-power-of-two allocations on pre-965.
In update_front_offset(), set pScrn->fbOffset even at EnterVT during init.
Bump pciaccess version requirement for new API usage.
Add some nickle scripts for looking at PLL issues.
Warn in the log if we choose a PLL clock that's way out of line.
In the clock graph, draw the VCO as erasures in the lines representing clocks.
Fix typo in my hand-application of rglowery's patch.
Bug #12059: Add ch7019 to the list of supported devices for ch7017.
Hong Liu (1):
Fix pixmap offset
Jesse Barnes (55):
Enable framebuffer compression (use Option "FrameBufferCompression"
Framebuffer compression changes:
Fixup line length buffer padding, add kludge for front buffer tile
Revert discard alpha change, requires other fixes to work.
Remove tiling kludge. May need more fixes for 965.
Re-add tiling kludge, but only for 965.
FBC and tiling changes
Fix tiling and fb compression defaults for 965 (not yet fully supported).
Fix debug output in fbc enable/disable routines. Add logic to make sure fbc
Update man page with current behavior.
Fix naming of FBC plane enable bits (mistakenly called them pipes earlier).
Fix manpage to reflect default behavior.
Legacy backlight changes:
Tiled rendering & fbc fixes:
Limit Solid & Copy offsets to 4k when rendering to tiled targets
More tiled rendering fixes: - check for tiling, not just offset in PrepareSolid - combine pI830->tiling
and frontbuffer checks into new exaPixmapTiled function for readability
Remove 4k offset checks from Copy & Solid hooks.
Fixup pitch in Prepare* functions, since actual hooks may
Tiling fixes for 965
Enable tiling by default on 965.
Cleanup tiling and FBC driver output.
Disambiguate plane and pipe mapping, use plane A on pipe B on pre-965 LVDS
Only swap planes and pipes if DRM supports it
Fix crash in ScreenInit
Fix plane/pipe mapping compat code
Remove unused plane->pipe mappings from SAREA private
Go back to using old drm_i915_flip_t field name
Fix palette save/restore
intel_reg_dumper - dump VGA AR registers too
Framebuffer compression fix: front buffer may not be at fence 0
Check DPLL status before writing PIPEnCONF regs
Setup 3D state at EnterVT time
Fix overlay destination clamping
Clear current video crtc on DPMS off
Default to EXA
Bump driver version to 2.1.99 in preparation for 2.2 release
Update man page to reflect EXA by default
Remove unused 'palette_enable' variable
Add 'ret' variable to i830_allocator_init
Fix typo in PLL enable check
Improve backlight control
Remove harmless error message
Don't stop ring before restoring hw state
Add FBC registers to register dump output
Correct FBC debug message
Fix kernel get_brightness function
Add more FBC regs
Disable FBC by default on 965GM
Add 1920x1080 mode to TV out
Adjust default TV out paramaters
Bump version to 2.2.0
Keith Packard (24):
Eliminate bogus (and harmful) blanking adjustment for load detect.
Ensure pipe/output active before doing load detection.
Remove hard-coded CRT blanking frobbing for load detection.
Clean up tv mode name allocation and copy.
Set DSPATILEOFF/DSPBTILEOFF to handle 965 tiled frame buffers.
Save/restore tile-mode offset registers DSPATILEOFF and DSPBTILEOFF
Make sure XV_PIPE is used whenever possible.
Thinkpad X61s has no TV out
Lenovo 201a is x60s, not x61s
Sort quirk table, add Dell Latitude X1
Intel driver configuration (only) changes for X server libpciaccess usage.
Change IS_Ixxx tests to work with or without libpciaccess.
Add libpciaccess declarations to I810Rec and I830Rec.
Change DRI interface to fill in PCI data from new libpciaccess structure.
Mechanical API conversions for libpciaccess.
i810_driver.c changes for libpciaccess.
i830_driver.c changes for libpciaccess.
Add register defines for hw binning
Limit TV formats to those supported by current connection
Don't set supported TV formats until after RandR initialized.
Switch to pci_device_map_range/pci_device_unmap_range APIs.
Quirk for Samsung Q35 which has no TV output.
Include xf86mm.h if using XF86DRI_MM
Review PLL spreadsheet and update register ranges.
Kyle McMartin (2):
Dell XPS 1330 has no TV out
toshiba satellite u300 has no tv out
Michel Dänzer (2):
Fix build against pre-pci-rework xserver.
Adapt to libdrm buffer object API changes.
Regis Prevot (1):
Ignore detection of TV output on Panasonic CF-Y7
AOpen 965GM mini pc has no LVDS
Fix a crash in TV mode handling by initializing the prev field of modes.
Thomas Hellstrom (1):
Adapt to DRM Lockfree and setStatus changes.
Wang Zhenyu (3):
Fix i915 rendering for tiled buffer
Add quirk support
Add another Lenovo TV output quirk
Zhenyu Wang (15):
Fix device id info for 945GME, 965GME
Fix a typo in i915 render
Update Lenovo TV quirk info
Fix seg fault introduced in tiling patch when TV detect
Fix i915 a8 color buffer blending
Fix G33 GTT stolen mem range
Pin cursor, overlay(no physical) and exa state buffers
EXA: fix tiled dest rendering on i8XX chips
Revert "Fix G33 GTT stolen mem range"
Fix allocation reset for really not free BO allocator
Don't setup fb compression if fb is not tiling
Remove one redundant line.
Don't enable fbc with XAA or tiling is off.
Move fb compression reg definition into i810_reg.h
remove unnecessary i830_reg.h includes
git tag: xf86-video-intel-2.2.0
MD5: e90d44adc819e8c5c76a1abde525a963 xf86-video-intel-2.2.0.tar.bz2
SHA1: 4737388e290fb20fe86399876ed36d4854e4e69f xf86-video-intel-2.2.0.tar.bz2
MD5: 7cab9695a391f07880bf03dba291f6ee xf86-video-intel-2.2.0.tar.gz
SHA1: 8fcb63ff419e6ceadc2ee51acca0903987ec45bc xf86-video-intel-2.2.0.tar.gz
More information about the xorg-announce