ATI Open Source Video Driver is the open-source alternative to the ATI/Radeon video driver.

ATI Open Source Video Driver is the open-source alternative to the ATI/Radeon video driver.

What's New in This Release:

Major changes:

· mach64, r128, radeon ported to libpciaccess
· massive restructuring of ati wrapper
· radeon support for r5xx, rs6xx, and r6xx chips using ATOMBIOS
· return of zaphod mode support
· radeon support for centered modes using scalers (selectable via
· output attributes)
· PAL tv-out fixed on supported chips
· initial support for render accel on r3xx/r4xx chips (rotation)
· fix TV option handling
· Xv RGB fixes
· XPRESS Xv fixes
· improve bios/driver interaction on radeon
· revert back to previous AGP mode behavior
· lots of bug fixes

Alex Deucher (134):
· Initial integration of Atom code and some of Dave's code.
· More of Dave's ATOM init code.
· Initial support for r4xx
· WIP: get ATOM crtc stuff working on r4xx
· for r4xx ATOM cards, just use ATOM for PLL
· Add atombios files
· First round of avivo support
· add additional connector types
· more avivo updates
· add avivo output stuff
· add support for initing external tmds via ATOM
· fix and move crtc source set up atombios_output.c
· pull in another of Dave's fixes
· make sure to assign gpio
· add pci ids for X1550PRO
· turn off vga control when using ext modes
· use atom to program plls on r4xx
· combine outputs based on id
· make sure i2c bus exists before using it
· rework crtc output source setup
· Add full parsing support for atom bios connector table
· fix from last commit
· atombios_dac_detect() takes care of primary vs tv dac itself
· fix INMC() and OUTMC() on !AVIVO chips
· WIP: sync up with the latest ATOM bios code in rhd
· WIP: more new ATOM integration work
· WIP: new atom code comples.
· Few fixes from the last commit.
· RADEON: step one in output rework
· RADEON: unify DDC line handling
· RADEON: unify connectortype handling
· fix typo in loop
· CRTs/DFPs may share a DVI port, but TV and CV don't
· Don't detect TV out for now
· Don't assign a gpio for TV
· fix logic in connector table check for TVs and switch counter to
· symbolic names
· RADEON: switch to using ATOM defines for bios device table
· RADEON: fixup for bios tables with wrong connector types
· RADEON: enable/disable the right TMDS controller
· RADEON: add LVDS atom support
· RADEON: store devices ids from bios for each driver output
· RADEON: make naming consistent and remove some cruft
· RADEON: major re-org and clean up of atom output control
· RADEON: provide clearer debugging info
· RADEON: attempt to fix crtc to output routing
· RADEON: fix crtc to output routing
· RADEON: reorder crtc/pll setup
· RADEON: add default connector table for avivo chips
· RADEON: move crtc output source selection into atombios_output.c
· RADEON: first pass at TV/Component video
· RADEON: re-org load detection for legacy chips
· RADEON: switch r4xx to atombios load detection
· RADEON: fix cursor offset on avivo chips
· RADEON: implement CLUT adjust support
· RADEON: small cleanup of pll code
· RADEON: disable atom pll set for r4xx cards
· RADEON: save/restore avivo crtc cursor control
· RADEON: remove unused cruft from last atom import
· RADEON: convert atombios connector table parsing to use ATOM structs
· RADEON: fix typo in previous commit
· RADEON: move GPIO lookup to a separate function
· RADEON: handle HMDI properly (untested) and fix some merge leftovers
· RADEON: fix rotation on avivo chips
· RADEON: rework PLL calculation
· RADEON: fix typo in previous cursor fix
· RADEON: clean up units in PLL calculation
· RADEON: fix sclock printout
· RADEON: fix typo from merge
· RADEON: post div tweaks for legacy radeon
· RADEON: more PLL tweaks
· RADEON: fix another merge error
· RADEON: re-org legacy crtc/output code into separate files
· RADEON: remove no longer used radeon_display.c
· RADEON: fix rn50 reversion from last merge
· RADEON: more re-org. move XAA Mem init to radeon_accel.c
· RADEON: more re-org
· RADEON: clean up prototypes
· RADEON: fix pll input setup on mac cards
· RADEON: Add "IgnoreLidStatus" option
· RADEON: restore crtc regs before VGA regs.
· RADEON: default "IgnoreLidStatus" to TRUE
· RADEON: fix PAL tv-out
· RADEON: fix typo that broke tv load detection
· RADEON: Make sure all old IGP chips have HasSingleDac set
· RADEON: add support for ATOM component video table
· RADEON: preliminary support for ATOM RMX
· RADEON: various avivo tv-out related clean-ups
· RADEON: fix crash when setting rotation in the config file
· RADEON: fix compile
· RADEON: improve ntsc image centering
· RADEON: further fixup for pScrn->pScreen issue
· RADEON: fix tvdac load detection at server start up
· R128: Like powerpc, don't use VGA by default on sparc
· RADEON: Fix TVStandard option
· RADEON: add default outputs if no connected devices
· RADEON: Make default output actually work...
· RADEON: restore FP2 regs before external encoders
· RADEON: clean up output handling
· ATOMBIOS: disable the scaler by default on avivo cards
· R300: First pass at render accel
· R300: only enable render accel on non-IGP r3xx/r4xx chips for now
· R300: only init3d on r3xx
· RADEON: add a message about render accel on newer cards
· AVIVO: Add support for RMX
· RADEON: use radeon_output->Flags for tracking RMX rather than mode->Flags
· RADEON: make sure EXA Composite is actually disabled on XPRESS chips.
· ATOM: Use LVDS edid from bios if available
· RADEON: grab pll_in_min/pll_in_max from bios tables if available
· RADEON: get dac2 adj values from the bios tables
· RADEON: re-work i2c for DDC
· RADEON: adjust PAL mode hstart
· Revert "RADEON: adjust PAL mode hstart"
· RADEON: Add new RV380 pci id
· Seems the default is more reliable... we can't win.
· RADEON: update man page to reflect AGP 1x default revert
· RADEON: Implement "center" mode for RMX on legacy radeons
· RADEON: remove unused "aspect" scaler option
· RADEON: remove redundant RADEONDisableDisplays()
· RADEON: don't restore LVDS_PLL_CNTL for now
· RADEON: rework i2c handling
· RADEON: sync up with latest upstream versions
· R6xx: fix ddc after my i2c rework
· RADEON: Implement proper (hopefully) BIOS/DRIVER interaction for COM bios
· RADEON: Implement proper (hopefully) BIOS/DRIVER interaction for ATOM bios
· RADEON: always restore crtc1 before crtc0 when using both crtcs
· R6xx: make sure we set up the HDP base properly
· R6xx: add missing objects
· RADEON: make sure we always set up the FP crtc shadow registers for crtc0
· R6xx: bios scratch regs moved.
· R6xx: fix up use of bios scratch regs to reflect the new offsets
· R6xx: when both crtcs are in use make sure they are both enabled
· RADEON: disable LVDS if panel size is 0x0
· ATOM: fix RS6xx connector table parsing and fix some warnings
· Bump for 6.8 release

Arkadiusz Miskiewicz (2):
· RADEON: various cleanups
· configure.ac fixes

Brice Goglin (1):
· RADEON: print the name of the output when printing the EDID

Dave Airlie (76):
· r5xx: fix typo for crtc offset
· r5xx: fix typo for crtc offset
· Add copyright headers
· radeon: rename a large section of avivo regs to documented names
· remove avivo_reg.h
· atombios: fixup warnings in atombios files
· avivo i2c: consolidate the avivo i2c code
· make i2c unlock/lock registers for gpios
· add missing hpd register
· avivo: fixup some i2c stuff
· add x1900xt support
· fix type for r520 agp code
· restore avivo memory map registers at correct places
· i2c: fix bus enable stuff
· r520: nail i2c enable/disable issue
· add firegl card on ajaxs machine
· add z3ro's pciids
· makes 2-headed cursor work
· r5xx: cleanups after last merge
· r5xx: add 71c5 for macbook pro
· atombios: add all r5xx and r6xx pci ids
· atombios: add warnings for r500 and r600
· r600: block r600 startup due to lack of memory controller info
· r600: get r600 to work non-accelerated.
· r600: use standard memory controller setup paths
· rs690 is !> r600
· r600: add memory controller regs from AMD
· LVDS on r500/r600 needs some work disable mobile chips for now
· add missing files for make dist
· re-enable mobility chips
· avivo: fix typo in register saving
· r500: make vt switch work again for me
· atombios: add initial object header parsing for r600 cards
· atombios: use values from object header
· r600: add hi agp address for mc
· fix silly spaces
· radeon: fix up memory mapping issues for vt switch
· r5xx: cleanup pll code..
· r500: set default minimum pll freq
· legacy: fix fb/agp read/writes
· radeon: add in pll spread spectrum workaround
· powerpc: build fixes from last merge
· radeon: move savedreg/modereg into entity instead of info
· radeon: bring back zaphod all is forgiven.
· don't add fboffset to info->FB it already is mapped at the offset
· disable tiling for zaphod heads
· add support for building against legacy servers similiar to
· Intel codepaths
· add more to configure.ac for using out-of-tree mode src
· fixup some warnings
· more zaphod fixes - some other work maybe needed
· radeon: cleanup pitch calculation and make r600 work again
· r600: add shadow support to r600 driver to at least make 2d useable
· fixup shadow setup on !r600
· fix big endian build since zaphod fixups
· more endian related fixage
· set ddc line correctly post-merge
· fixup clip test include handling
· fix mode bandwidth configure check
· atombios: add support for reading tv standard from atombios
· atombios: enable support for using tv timings
· atombios: enable TV-out detection code - tv out works for me with this
· r600: fix tv-out output naming
· PLL/r600: tweak pll to pick first one found instead of keeping going
· r600: fixup crash on unknown output type
· r600: change devices list depending on connector
· r500: tvout avoid doing dpms here it makes my tv mode go all crappy
· atombios: i2c lines are all done with sw so we can accept any of them.
· i2c: a line mux of 0 is valid
· fixup register 6594 save/restore
· r300: remove r300 specific vertex emission
· r300: move fragprog setup code to prepare composite for now
· add rv670 pciids
· make distcheck pass
· mach64: fix non pciaccess build
· r128/radeon: fix build without pciaccess
· r128/radeon: hopeful fix for non pci access code

David Miller (2):
· [RADEON]: Like powerpc, don't use VGA by default on sparc.
· [RADEON]: Add missing break in SCLK calculation.

George Sapountzis (31):
· [mach64] Drop symbol lists.
· Drop symbol lists from r128, radeon, theatre.
· r128: pci-rework conversion
· radeon: remove stray _X_EXPORT
· radeon: remove stray include
· ati wrapper: use pci ids
· atimisc: rename to mach64
· mach64: add version (need not match with ati)
· mach64: load for both "ati" and "mach64" as driver names
· ati wrapper: add DriverRec's and use them
· ati wrapper: xf86PciInfo.h is enough
· use stand-alone drivers for each chip family.
· mach64: workaround for corruption at upper-left
· mach64: add hint for sparc and minor cosmetic.
· mach64: BIOSBase is no longer used
· mach64: factor out BIOS clock parsing
· mach64: cosmetic
· mach64: consolidate refclk #1
· mach64: consolidate refclk #2
· mach64: complement hint for sparc
· mach64: factor out BIOS multimedia parsing
· mach64: minor cosmetic, DAC type
· mach64: minor cosmetic, LCD panel id
· mach64: factor out BIOS panel info parsing
· mach64: minor cosmetic, I2C type
· mach64: clean probe a little
· ati: drop duplicate pci-id defines
· r128: do not compile in PciChipsets twice
· ati wrapper: translate Device lines
· ati: convert to pci probe
· r128/radeon: probably correct fix for non pci access code

George Wu (3):
· Add LVTMA PWRSEQ registers to fix VT switching for LVDS
· RADEON/R600: small code cleanup
· r600: might as well fix VT for R600

Kristian Høgsberg (1):
· RADEON: fix crtc routing on r4xx cards when using atom to init DVO chip

Kusanagi Kouichi (1):
· radeon: Partial fix for XVideo RGB image distortions.

Maciej Cencora (2):
· atombios: initial rs690 patches
· RS690: Implement MC idle check

Michel Dänzer (2):
· radeon: Miscellaneous warning fixes.
· radeon: Adapt manpage to reality wrt AGP options.

Roland Scheidegger (1):
· bring back to life planar-to-packed conversion for rs4xx