[ANNOUNCE] xf86-video-ati 6.12.191
Dave Airlie
airlied at gmail.com
Tue Mar 2 17:43:59 PST 2010
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Subject: [ANNOUNCE] xf86-video-ati 6.12.191
To: xorg-announce at lists.freedesktop.org
CC: xorg at lists.freedesktop.org
This is a pre-release for the upcoming 6.13 release of the X.org ati/radeon
driver.
Major highlights:
KMS/DRI2 support for all families r100->r700 for KMS driver in Linux 2.6.33.
Initial evergreen KMS support for planned 2.6.34 driver.
Userspace modesetting support for r600/r700 families
experimental userspace modesetting support for evergreen families
r600/r700 acceleration support.
Userspace modesetting basic power management controls
textured Xv controls (brightness, contrast, etc.)
Displayport support.
Improved Zaphod support - per-screen output assignment.
CustomEDID option - force the driver to use EDID from a file
and lots of quirks and fixes and optimisations.
Thanks to all contributiors,
Dave.
Adam Jackson (5):
atom: Fix phantom VGA connector when HDMI-B present.
Cosmetic cleanup to digital output detection
Add a (harmlessly) missing break statement.
Fix misleading filenames in DRI setup failure messages
radeon: fix dri2 struct zeroing
Alan Coopersmith (3):
Add link to patch submission instructions to README
Update Sun license notices to current X.Org standard form
Fix configure with --disable-dri
Alex Deucher (235):
Bump version post release
R6xx/R7xx: set proper 3D client driver name for r600
radeon: chip family typo
R6xx/R7xx: cleanup macro
radeon: remove old crtc size heuristics
radeon: fix tmds_pll randr property
Add unified ClockGating Option
radeon: Add functions to set sclk/mclk on r1xx-r4xx
radeon: Add functions to set sclk/mclk using atombios
Add ForceLowPowerMode Option
Add support for setting the number PCIE lanes
Set default low power PCIE lanes to 2
Add DynamicPM Option
Unify ForceLowPowerMode and DynamicPM Options
man page: fix missing break
PM: don't touch clock gating regs unless ClockGating option is set
Change ForceLowPowerMode to 4 pcie lanes
Revert "DCE3/3.2: cleanup dpms after previous changes"
R3xx/R4xx EXA: set explicit tex cache regions
RS690/RS740: fix connector enumeration in some cases
RS400/RS480: attempt to fix ClockGating on RS4xx chips
PM: fix reversed logic in pm block handler
Tex vid: split by family
R1xx tex vid: append verts for clip boxes
R2xx tex vid: append verts for clip boxes
R1xx: add support for native planar textured Xv
Tex vid: general cleanup
R200: clean code, always use shader based csc
R6xx/R7xx: implement Xv attributes
tex vid: fix attribute setup typo for XV_COLORSPACE
R3xx/R4xx: Implement shader-based csc for packed formats
R5xx: add shader-based csc
R3xx/R4xx: fix up planar shader
Tex vid: remove remnants of XV_HWPLANAR
R3xx/R4xx: set tex caching for Y texture when doing planar rendering
R3xx/R5xx: only apply Xv attributes if bicubic is disabled
Update Xv info in man page
Remove old messages
Endian fixes for object table
Tex vid: cleanup and unify common code
radeon PM: add some quirks
R4xx: avoid a PM related hang
Tex vid: rework part 2
RS690: add quirk for acer board
Tex vid: fix top clipping
radeon pll: add support for fractional feedback divs
AVIVO: fix for cursor corruption issue
radeon PM: update quirks
r3xx/r5xx: fix typo in last commit
radeon: minor cleanup and warning fix
r2xx/r3xx/r4xx: further i2c fixups
AVIVO: make sure cursor width isn't negative
AVIVO: disable fractional feedback divs
RV410: SE variants only have 1 quad pipe
AVIVO: fix cursor corruption/flickering issues
R6xx/r7xx: fix CURSOR_SWAPPING_* macros
AVIVO: better fix for cursor flickering/corruption issues
radeon: fix several dpms issues
AVIVO: first pass at display watermark setup
R3xx-R5xx: don't set TX_OFFSET_RS in RS_INST_COUNT
RS690: gpio/connector fixes
r2xx-r4xx: fix typo in last i2c fix and clarify hw i2c pin sel
Remove some old debugging info
r3xx-r5xx: switch to 1/12 subpixel precision
AVIVO: move cusor offscreen when disabling
radeon: switch to EXA by default
AVIVO: fix cursoe corruption bands for real
AVIVO: set default cursor mode in cursor init
R3xx/R4xx tex vid: increase the guardband limit for rendering with a tri
R3xx/R4xx tex vid: no need to check if IS_R300_3D
R7xx: add support for reading disabled bios images
RV770: add missing pci id
R3xx-R5xx: do EXA transforms in the vertex shader
R6xx/R7xx: do EXA transforms in the vertex shader
R6xx/R7xx Xv: normalize texture coordinates in the vertex shader
radeon: try harder when checking if a card is posted
R100: switch to normalized texture coordinates
r4xx: Add some missing pci ids
Pre-atom: Improve engine clock setup code
R3xx/R4xx: increase EXA render target limits
radeon: remove dead code
Add RV740 (HD4770) support
atom: connector quirk
R6xx/r7xx: don't clamp tex coords in composite VS
r3xx/r4xx/r5xx: refactor exa prepare composite
r3xx/r4xx/r5xx: clip 3D rendering to destination buffer
Consolidate indentical RADEONDoneSolid/RADEONDoneCopy functions
r1xx/r2xx: clip 3D rendering to destination buffer
Add PLL flag to prefer frequencies <= the target freq
Re-enable frac feedback dividers on AVIVO hardware
Only enable frac fb divs on rs600/rs690/rs740 for now
Fix count in 3d state setup after 43374c7420e378918bec062f4cbd581f16adb6f0
Add some missing M96 pci ids
radeon: call hdmi-b connectors dvi in randr
EXA: fix the build on servers without EXA_HANDLES_PIXMAPS
Fix the build for the !XF86DRM_MODE case
R6xx/R7xx: fix hangs on x2 cards with PM options
RS740: fixup display latency setting as per rs690
RV280: Add an AGP quirk
RV280: another AGP quirk
Add configure option to disable kms support
R3/4/5xx: fix bicubic Xv filtering with KMS changes
R3/4/5xx: only upload the bicubic texture once
Add an r420 AGP quirk
R6xx/r7xx: reduce the amount of default state setup
R6xx/R7xx: clip rendering to destination surface dimensions
RS780: fill in MC access functions
radeon: Set PCI/PCIE bus type properly per asic
radeon: fix typo in object header to connector type conversion
radeon: reload bicubic Xv texture on VT switch
radeon: use XAA in some cases
radeon: add some new r7xx pci ids
RS880: enable accel
radeon: re-add asus connector quirks
radeon: fix compile with !kms
Add KMS EXA UploadToScreen hook.
RV280: add agp quirk
AVIVO: add support for DisplayPriority HIGH
rs600: add support for DisplayPriority HIGH
RV740: disable small DFS transfers
r6xx/r7xx: move more common state to default state setup
R6xx/r7xx: add begin/end batch macros
r6xx/r7xx: first pass at kms accel support
r6xx/r7xx EXA: WIP
r6xx/r7xx: more cs exa wip
r6xx/r7xx: fix flipped domains
R6xx/r7xx: unmap vb bo when done
r6xx/r7xx: fix reloc for vtx buffer
r6xx/r7xx: fix some define problems in Xv code
r6xx/r7xx: more WIP
r6xx/r7xx: set EXA_HANDLES_PIXMAPS
r1xx: fix textured video
r6xx/r7xx: various CS fixes from Dave
radeon: add some notes about DVO setup
RV280: add agp quirk
radeon: replace dga code with DiDGA
r600: add CS UTS and DFS hooks
atom: fix typo in asus quirks
kms: init dga using xf86DiDGAInit
kms: disable dga
radeon: remove old pre-randr rotation cruft
r600: fix bo accounting for Xv
radeon: fix vline handling for kms
kms/r600: add support for vline relocs
radeon: fix segfault in MMIO path in RADEONInit3DEngine
R6/7xx EXA: Add support for BGRA picture formats.
R1/2xx EXA: Minimise number of draw primitives used for Composite operations.
radeon: clamp the internal FB map to the aperture size
ATOM: add support for AdjustDisplayPll table
ATOM: reorder crtc dpms based on bios recommendations
ATOM: rework crtc modeset
r600 EXA: properly check planemask
r600 EXA: fix some compile warnings
r600 EXA: fix up mask reg mixup
ATOM: minor cleanups to pll setup
radeon: add XV_CRTC attribute for textured video
radeon: fixup CustomEDID option
ATOM: fix up get clock info
ATOM: use indirect addressing for regs >= 0x10000
R7xx: program additional CUR/GRPH regs for 40 bit addresses
radeon: clamp the internal FB map to the aperture size v2
atom: loosen pll min output limits
radeon: fix DVO on AVIVO chips
ATOM: fix up DVO for DCE 3.x
radeon/atom: fix tv-out on r5xx
ATOM/DCE3+: fix up usPixelClock calculation for Transmitter tables
DCE3+: call transmitter init on mode set
EXA: fallback if no pMaskPicture->pDrawable
Revert "Fix warning about unused defined labels in radeon_exa_funcs.c."
radeon: properly fix unused label warnings
IGP: some IGP chips report as AGP
r600/r700: typo, fix mask of DB_ALPHA_TO_MASK
radeon: fix support for external tmds on some legacy chips
radeon: man page updates
r600: fix num format in vtx fetch
r600: remove un-needed format conversions
radeon: fix potential memory leak in ddc code
radeon: deal with connectors sourced to the same encoder
radeon: pull previous displayport work into master
more dp cleanup
DP wip
wip
DP: hack that makes DP work.
r300: render target limit is 2560
dp fix timing
DCE3.2: fix uniphy2 dvi issues
dce3: CV1/TV1OutputControl tables didn't go away until dce3.2
radeon: clean up DP code
atom: pull misc mode info for lvds panel mode from bios tables
radeon: only read RADEON_PPLL_REF_DIV for ref div on pre-avivo
AVIVO: add new PLL code
ATOM: split set base/format into separate functions
ATOM: add new power table defs
radeon: add cvt timing if we only have panel w/h
radeon: fix LVDS power sequence on Mac cards
radeon: updated ObjectID.h
radeon: add initial support for eDP (embedded DisplayPort)
radeon: fallback to i2c for edid if aux fails
radeon/kms: add new strings for eDP and TV
rv100: reject modes >135 Mhz with DVI
rv100: fix typo in fix for bug 25992
ATOM: Upstream parser updates
KMS: fix no accel option
radeon: use mmio bar size rather than hardcoded number for register ops
radeon: minor pll updates
avivo: add some hotplug detect regs
radeon: clean up dig encoder handling
r500: fix CS section size missmatch in Xv code
atom: update to the latest upstream atombios.h changes
evergreen: add chip enums
evergreen: add register and utility defines
evergreen: add base asic support
evergreen: add hw cursor support
evergreen: add lut support
evergreen: add crtc set base/format support
evergreen: add atom support for digital outputs
evergreen: add atombios crtc/pll functions
evergreen: add pci ids
evergreen: fix units on frac_fb_div
evergreen: DP requires coherent mode
evergreen: pll fixes
evergreen: blank/unblank DP in dpms calls
evergreen: use external clock source for DP PHY
r600: reduce number of cache flushes
evergreen: minor clean up in transmitter setup
r500: fix relocs for Xv planar video
radeon: consolidate crtc selector for vline wait
r6xx/r7xx: implement EXA vline support
radeon: add ZaphodHeads option
RS600: add connector quirk
radeon: update new pll algo
AVIVO: set frac fb divider pll flag
radeon: fixes for zaphodheads option
update ZaphodHeads option
rv740: disable dfs workaround for drm 1.32+
Bryce Harrington (3):
radeon: AGPMode quirk needed for HP Omnibook 6200
radeon: AGPMode quirk needed for IBM Thinkpad T40 with Mobility M7 LW
radeon: AGPMode quirk needed for SiS
Cooper Yuan (1):
add more triple-head server ID
Corbin Simpson (1):
dri: be more useful if disabling DRI due to KMS/UMS mismatch.
Csillag Kristof (2):
radeon: add support for Custom EDID
Fix some issues with CustomEDID support
Dave Airlie (101):
radeon: legacy bandwidth calcs for later r3xx/r4xx cards.
radeon: fix DDC on rv410 VGA ports for non-atom use.
radeon: tweak ddc fix for all r300s.
r300/r500: set R clamp mode.
radeon: attempt to fix R clamping for repeat again
Revert "radeon: attempt to fix R clamping for repeat again"
Revert "r3xx/r5xx: fix typo in last commit"
Revert "r300/r500: set R clamp mode."
radeon: initial preparation for kms patch.
radeon: add KMS support (still disabled)
kms: no need to reference fifo slots
radeon/kms: fixup 3d state emission for KMS
radeon: enable KMS build
radeon: fix CS imbalance
kms: drop DGA
radeon/kms: add textured video BO size checks
radeon/dri2: call proper gem bo function
radeon/dri2: fix dri2 - needs newest libdrm_radeon
radeon: fix compile without kms
radeon: fix it properly.
radeon: fix up configure check better
radeon: don't set up overlay under KMS until we figure out how.
Revert "KMS CS fixes."
radeon: port to new space checking in libdrm interface
radeon: taken from kms-support + CS fixes commit.
radeon: move state changes into flush function
radeon: oops fix compile without libdrm
git: fix git-log to git log
radeon: add make dist headers
radeon: add shave support.
radeon: emit colorpitch relocs.
ati: update for resources/RAC API removal
ati: change to using ABI version check
radeon: fix r600 shadow fb mode with cursor allocation in wrong place
r200: fixup scissors for DDX.
radeon/tv: fix typo in crtc register modifications
r100/r200: dont emit wrong clamp modes.
kms: make tv out match the connector name
atom/tvout: set the correct scaler up for tv out
atombios: store dac detect result into save register.
atom: recall crtc source script after tv dpms on
radeon: move detected monitor type debugging.
radeon/kms: add initial colortiling support (disabled by default).
exa/cs: add DFS from GTT optimisation
radeon: don't spec any initial placement for pixmaps.
r600: fix bad unit in texture setup
r600: use exa fallback code for solid/copy codepaths.
radeon: afaics these aren't radeon regs but r128 regs.
r100/r200: seriously WRAP is illegal for rectangular textures
r600: more alignment fixups + vb map/unmap
kms: add property support.
r6xx: fix EXA crash with gtkperf -a
avivo: disable VGA rendering core when starting X.
r600: space check wasn't correct.
radeon: fix zaphod
r600/xv: fix dst bo write domain
r600: fix build
radeon: fix build without kms
kms: setup colormap and gamma correctly.
kms: don't use scratch pixmaps when copying fbcon.
kms: only touch gamma stuff on 1.7
radeon: fix rotation since mixed pixmaps.
drmmode: reload cursors on modeset
kms: don't use scratch pixmaps at all
Revert "kms: setup colormap and gamma correctly."
Revert "radeon: clamp the internal FB map to the aperture size"
kms: add cloning support.
kms: check for 0 mask for cloned outputs.
radeon: r100/r200 have a 2047 scissor limit
kms: recalculate the flush limits after screen resize.
radeon: avoid using hw pixmaps when we have little VRAM.
radeon: surround mixed pixmaps usage with ifdef check for old servers
radeon: fix warning in printf
r600: enable mixed pixmaps
radeon/kms: drop special r600 indirect flush
kms: allow prepare copy to fail without dying
r600: refactor code to help future acceleration speedups.
radeon: fix check for no work in operation
kms: flush cs on close screen.
r600: fixup problems with EXA operation reset for multiple ops
r600: enable multiple operations in one CS
r600/xv: drop inited 3d false in xv code
radeon/exa: change option to determine exa pixmap usage.
Revert "r600: enable multiple operations in one CS"
displayport: cleanup call dp from dpms on instead.
fixes displayport for me here
radeon: fix typo in unused as of yet CV code
atombios: CV mode retrival was broken.
dce3: no CV1 or TV1 Output Control tables anymore.
r600/accel: cleanup and merge a lot of the accel functions
r600: fix multi-operation in single batch support.
radeon: use new libdrm api
displayport: fix DDC on DVI->DP convertors.
radeon/kms: only enable EXA pixmaps under kms if we support render accel
radeon: add support for picking the digitial encoder
radeon: further digital encoder cleanups.
rs400/rs480: mc idle bit is bit 2 like original radeon
zaphod: hopefully fix kms + zaphod
pci: add HP vendor id
pciids: hopefully fix HP
radeon: bump configure.ac
Edward O'Callaghan (1):
Fix a warning
Fredrik Höglund (1):
Check if the composite op is supported in R200CheckComposite.
Gaetan Nadon (9):
xf86-video-ati: use XORG_CHANGELOG macro to create ChangeLog. #22611
.gitignore: use common defaults with custom section # 24239
Several driver modules do not have a ChangeLog target in Makefile.am #23814
INSTALL, NEWS, README or AUTHORS files are missing/incorrect #24206
INSTALL, NEWS, README or AUTHORS files are missing/incorrect #24206
Makefile.am: add ChangeLog and INSTALL on MAINTAINERCLEANFILES
src/Makefile.am: missing 3 header files in tarball
configure.ac: remove unused sdkdir=$(pkg-config...) statement
config: remove dead LINUXDOC macro usage
Ingmar Vanhassel (1):
Replace shave with automake silent-rules
James Le Cuirot (2):
zaphod: Fix entity instance assignment.
clean up kms zaphod handling
Jerome Glisse (9):
radeon: fix KMS shadowfb for r6xx and newer hw.
radeon/kms: add simple DownloadFromScreen implementation
radeon/r600: abstract vertex buffer for r600 & r700 hardware
radeon: Fix X recycling with KMS
radeon/kms: fallback to shadowfb if kernel report acceleration is off
kms: assume accel is working is info ioctl return -EINVAL
kms: Fix resizing when acceleration is disabled
r6xx/r7xx: emit relocation for FRAG & TILE buffer
r6xx/kms: when reseting BO force default state emission
Kenneth Graunke (1):
radeon: Remove gatos message from driver startup
Kristian Høgsberg (1):
dri2: Use drmGetDeviceNameFromFd() instead of open coded loop.
Kusanagi Kouichi (3):
R600: fix some warnings
radeon: Lift hardcoded limit from RADEONQueryImageAttributes
Remove unnecessary checks.
Luca Tettamanti (1):
Disable color tiling on r600/r700
Markus Gapp (1):
Add quirk for asus hd3450 board
Matt Turner (2):
Use/define RADEON_GPU_PAGE_SIZE instead of sprinkling 4096 everywhere.
Use RADEON_ALIGN instead of open coding it.
Matthias Hopf (1):
fix 200M freezes on VT switch if CRTC is disabled
Matthieu Herrb (1):
radeon: fix segfault if Option "NoAccel" is set
Matthijs Kooijman (1):
radeon: fix crtc2 dpms
Michael Olbrich (1):
use AC_CHECK_HEADER instead of AC_CHECK_FILE
Michel Dänzer (47):
Clear the framebuffer before initializing modes.
Pre-R600 EXA: Fix coordinate limits off-by-one error.
EXA: Set new EXA_SUPPORTS_OFFSCREEN_OVERLAPS flag when we have a 2D engine.
Bump version to mark Git snapshot.
Build fixes for -Werror that aren't ready for prime time yet.
Revert "Build fixes for -Werror that aren't ready for prime time yet."
Warning fixes.
Warning fixes.
Test BO pointer before passing it to radeon_bo_unref().
KMS CS fixes.
Fix DRI2 with current xserver Git.
KMS CS fixes, take two.
Don't always flush CS when switching from 2D to 3D.
Don't set EXA UploadTo/DownloadFromScreen hooks when they'd always return FALSE.
Pass on CS relocation offset values.
Fix RADEON_TRACE_FALL build.
Further non-KMS fixes / cleanups.
Enable vsync for DRI2 region copies.
Set the STALL bit in the CRTC_GUI_TRIG_VLINE registers.
Set CRTC active flag to TRUE after a successful KMS mode set.
Also drop DRM master in KMS CloseScreen.
Guard reference to CRTC active field.
KMS: Explicitly specify VRAM and GTT domains when allocating BOs for pixmaps.
Minor fixes for KMS EXA DownloadFromScreen hook.
Add support for EXA_MIXED_PIXMAPS in xserver master.
Properly let radeon_cs_flush_indirect() re-emit 2D state if necessary.
Throttle DRI2 buffer swaps / frontbuffer flushes.
EXA: RENDER repeat cleanups.
DRI2: Set tiling for depth/stencil buffers with all DRI2 interface versions.
EXA: Allocate pixmap BOs in VRAM.
KMS: DownloadFromScreen improvements.
KMS: Don't use a blit for UploadToScreen if the pixmap BO will be idle anyway.
KMS: Remove explicit radeon_bo_wait calls before radeon_bo_map.
EXA: RENDER repeat fix and more cleanup.
EXA: Fix Composite restart dst/src switcheroo.
EXA: Check for solid/gradient pictures the same way for all generations.
Fix some compiler warnings.
KMS: Don't hide HW cursor when it's updated.
KMS: Double-buffer textured video source image upload.
EXA: Don't always flush when switching between 2D/3D engines with KMS.
R3/5xx EXA: Minimise number of draw primitives used for Composite operations.
R3/5xx EXA: Add support for BGRA picture formats.
EXA: Bail earlier from PrepareCopy hook on trivially unsupportable conditions.
Fix KMS on big endian machines.
Don't call radeon_cs_flush_indirect() without KMS.
Allocate BOs for 'video offscreen surfaces' and 'TV-in' in VRAM.
EXA: Fix order of coordinates passed to radeon_pick_best_crtc().
Pauli Nieminen (5):
radeon: Fix DRI2BufferPtr to be DRI2Buffer2Ptr for xserver 1.6.
radeon: Add missing libdrm_radeon cflags to compiler paramaters.
Fix configuration to disable KMS if not finding libdrm_radeon.
Use memcpy instead of memove for RADEONSwapCopy.
Allocate Xv buffers to GTT.
Peter Hutterer (1):
Update to xextproto 7.1 support.
Roel Kluin (1):
radeon: Fix duplicated bit settings
Roland Scheidegger (3):
r200/r300: implement brightness/contrast/hue/saturation/gamma controls for textured video
R6xx/R7xx: fix pixel centers
r100/r200: fix section size mismatch for textured video
Tomáš Chvátal (1):
Fix warning about unused defined labels in radeon_exa_funcs.c.
git tag: xf86-video-ati-6.12.191
http://xorg.freedesktop.org/archive/individual/driver/xf86-video-ati-6.12.191.tar.bz2
MD5: 998076ecb256b1bfc0177da2e239d8de xf86-video-ati-6.12.191.tar.bz2
SHA1: 361b88644939f28b70c38c7ce7550f68270ebd90 xf86-video-ati-6.12.191.tar.bz2
http://xorg.freedesktop.org/archive/individual/driver/xf86-video-ati-6.12.191.tar.gz
MD5: b9d315a0ecf9fe5d0f4737891874f9d8 xf86-video-ati-6.12.191.tar.gz
SHA1: 2fc53ee54444d82bdf7a0aa2fd54d3923c75ca87 xf86-video-ati-6.12.191.tar.gz
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.10 (GNU/Linux)
iEYEARECAAYFAkuNvosACgkQ6acWQe8Wxxq/yQCgnU2hPr+ELD29YfAoI11H1foW
TxoAn1vDaKhwHGOHQPN3/WFrlBXNBe/g
=x+rT
-----END PGP SIGNATURE-----
More information about the xorg-announce
mailing list