-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 ATI 6.8 pre-release This is the long awaited radeon driver with randr 1.2 support. radeon 6.7 was pretty much a dead end, so it's been branched for those that still want to play with it. In addition to randr support there are quite a few other goodies: zero copy tfp, improved connector table parsing, TV-out support, and much more. Changes since 6.6.3: Adam Jackson (1): Undo unintended configure.ac change. airlied (9): split mode handling code into separate modes source file forgot SetPitch function is called radeon: initial move framebuffer away from zero prep. fixup EXA after previous commit make DDC function more generic add generic mode handling code from Intel driver add more functions from i830 driver make radeon modes use generic code move radeon over to not using X's mode management like the intel driver. Alain P?teut (1): enable i2c for PCI_CHIP_RV250_If. allows AIW fuctionality to work. Alex Deucher (165): radeon: restructure files along lines of Alex's superpatch radeon: add missing tv-out registers to register list radeon: add pre init controller function radeon: add save/restore of some new registers radeon: move to using Alex's controller structure radeon: make DVI_I default connector radeon: add reverse display option to switch displays on non-mobility cards radeon: re-organise FP and CRTC register setting routines tvdac handling was reversed between r200 and other chips fix LVDS plus internal TMDS remove bogus registers FP timing regs required for both internal and external TMDS Remove Bioshotkeys option. Leave the bios as configured at boot. Remove paneloff option. This is already covered by monitorlayout and Rework the output init code. Divide into separate functions Disable all ouputs on initial mode set. The ones we want will get clean up magic number Don't mess with the VIP bus on mobility chips. Seems to cause fan clean up previous VIP commit. update man page. fixes bug 6457 I think we can say r300 support is not experimental add option to force the DRI on for RN50 chips Add "DRI" option fix NULL dereference RADEON - update randr cursor handling, LVDS setup fix RADEON: Fix cursor handling for randr remove some old mergedfb cruft minor cleanup start to clean up MapControllers More re-org to allow more than 2 outputs enable all outputs on each crtc on mode restore more randr re-work randr checkpoint RADEON: randr driver re-org checkpoint RADEON: fix VT switch RADEON: more re-org RADEON: re-arrange output mode setting RADEON: switch output dpms to use RADEONEnableDisplay() RADEON: Move LVDS, TMDS, DAC properties to the output rec RADEON: Fix RMX after the last commit RADEON: remove hardcoded output limit and PortInfo stuff RADEON: don't need to pass pScrn to EnableDisplay() RADEON: fix up DDCConnected() RADEON: move crtc offset handling into init/save/restore functions RADEON: fix VT switch RADEON: several fixes RADEON: Remove dead code RADEON: re-org randr code RADEON: remove more dead code RADEON: remove old mergedfb and dualhead options that are no longer used RADEON: remove mergedfb, etc. cruft from man page RADEON: add new ConnectorTable option and re-add PanelSize option RADEON: additional cleanups RADEON: remove remnants of "old" multi-head support RADEON: Further cleanup RADEON: Lots of small fixes RADEON: Move DAC regs to their own Save() function RADEON: remove some debugging code RADEON: More cleanup RADEON: turn off all outputs in screeninit(). We'll turn on the ones we want later RADEON: more cleanup RADEON: cleanup RADEON: Change default LVDS i2c line for powerpc RADEON: if connector is VGA set TMDS to none Fix regular/"xinerama"/zaphod dualhead mode update to 6.6.192 for rc release RADEON: several updates RADEON: fix crtc1 (un)blanking code after last commit RADEON: Only add valid connectors RADEON: add output properties (not functional yet) RADEON: clean up some logic RADEON: clean up non-DDC probe, add detect for DVI RADEON: fixup last commit and add support for ext TMDS detect RADEON: more clean of last commits RADEON: implement auto/analog/digital output property for DVI-I (untested) RADEON: implement backlight control for LVDS RADEON: fix handling of DRI lock RADEON: Make sure LVDS and FP2 routing info gets written RADEON: make sure we have the lock when SYNCing RADEON: Switch SetOutputType() to use names rather than numbers RADEON: clean up and add comment regarding clones RADEON: first pass at rotation (not working yet) RADEON: more rotation work (still not there) RADEON: remove some dead code from the last commit RADEON: number the outputs if there are more than one of the same RADEON: remove some cruft RADEON: add support for RMX on DVI, make sure RMX is only enabled on crtc1 RADEON: no need to go through all the crtcs, we've got what we need RADEON: fix indenting RADEON: fixup RADEONDisplayVideo() to better handle the crtc being used RADEON: implement support for posting cards based on x86 bios tables (untested) RADEON: fix some issues with bios table init code RADEON: make sure RADEON_LVDS_ON is cleared when turning LVDS off RADEON: make sure to clear RADEON_LVDS_ON when turning LVDS off RADEON: additional LVDS off fix (missed in previous commit) RADEON: additional LVDS off fix (missed in previous commit) RADEON: yet another LVDS... RADEON: only touch LVDS on mobility chips RADEON: only touch LVDS on mobility chips RADEON: Split FP reg restore to only touch regs relevant to the output RADEON: bump crtc max to 1200 so 1600x1200 will work out of the box RADEON: tiling and FB size fixes RADEON: fix possible segfault on mobility chips with MM tables RADEON: fix possible segfault on mobility chips with MM tables RADEON: attempt to fix cursor on big endian (only tested on x86) RADEON: make sure we set the stride properly RADEON: make sure we unblank in RADEONRestore() RADEON: Switch cursor back to memcpy() RADEON: set info->CurrentLayout.displayWidth in RADEONSetPitch() RADEON: fix corruption after 3D apps run RADEON: move crtc base setups to new functions RADEON: factor out surface_cntl init into one function RADEON: simplify console restore RADEON: save crtc on/off state for console restore RADEON: Fix tiling on r1xx and r2xx RADEON: set default values for crtc_offset_cntls R128: don't clip modes to panel on laptops when only using the CRT port RADEON: fix palette when depth != 24 RADEON: fix randr on r4xx RADEON: hack around console restore hang on r4xx RADEON: write out saved vclk and pixclk values RADEON: Compute PLL VCO gain RADEON: Initial pass at integrated tv out support RADEON: add info about tv out code and authorship RADEON: tv-out fixes. works now. tested on rv350. RADEON: fix name of tv output RADEON: more fixes... RADEON: convert hard coded tv out values to calculations RADEON: remove unused elements RADEON: limit tv modes to the only one we can program at the moment RADEON: move tv dac enable to enabledisplay() RADEON: add missing break RADEON: fix tv-out enable/disable RADEON: avoid a divide by 0 and only save tv out regs if the chip has them RADEON: Change indexing of TV constants table in preparation for standard re-work RADEON: Major rework of BIOS table parsing RADEON: attempt to do the right thing for standards other than PAL or NTSC RADEON: refactor output init to handle multiple DVI or VGA RADEON: Fix DVI-I support in ATOM bios connector table parsing RADEON: fixes RADEON: typo RADEON: switch RMX attribute to string type RADEON: make backlight attribute lower case RADEON: switch DVI-I monitor type attribute to string RADEON: add tv out properties RADEON: Always set MonType in radeon_detect() RADEON: minor fix to legacy bios connector table RADEON: Make sure the default TV standard is supported RADEON: fix ConnectorTable option after connector table re-work RADEON: fix typo in comment RADEON: fix connector setup when there's no bios tables RADEON: order the VGA and DVI ports correctly RADEON: Add load detection RADEON: Make sure RMX is always programmed for crtc1 RADEON: Turn off tv encoder in disableoutputs() RADEON: Implement improved tv load detection for r300 RADEON: minor cleanup RADEON: make sure crtc routing is correct in r300_detect_tv() RADEON: Remove RADEONRestoreMode() RADEON: fix Xv clipping and overlay sourcing RADEON: turn off TVCLK when blanking tv encoder RADEON: Add DefaultConnectorTable option RADEON: fix tv-out on R3xx RADEON: more tv out fixes and clean up RADEON: Fix color problem on pre-R3xx chips tv-out Andrew Randrianasulu (2): RADEON: Fix VT switch hangs RADEON: Fix VT switch hangs Brice Goglin (2): MACH64: avoid crash in Xv code Minor fixes in the manpages Daniel Stone (2): Makefile.am: fix changelog hook Makefile.am: re-add ChangeLog to CLEANFILES Dave Airlie (125): radeon: restore secondary head behaviour radeon: re-add DAC power on/off code around detect radeon: re-add ATI patch for rn50 modes radeon: add tv dac adj lookup radeon: move InitFPRegisters up to align with superpatch radeon: initialise controller structures radeon: minor cleanup of masking for crtc registers radeon: copy of overupdated bios hotkeys code from Alex radeon: fixup InitFPRegisters function call to not pass info radeon: cleanup DacPowerSet radeon: cleanup DPMS set on/off code to use new controller structure radeon: move code around in InitCrtcRegisters radeon; put back old TV DAC code for now radeon: add enable display function radeon: refactor blank/unblank radeon: move comment radeon: add simple map controllers call radeon: add some info to the map controllers radeon: flesh out mapcontrollers radeon: some minor alignments and cleanups for info-> radeon: make GetConnectedMonitors a static void radeon: change function name to Displays rather than Monitors radeon: align some output with superpatch radeon: align tabs and make display detection like superpatch radeon: bring over final set of changes for radeon_display.c from susperpatch radeon: setup the radeon init function to use secondary radeon: initial CRTC register setting alignment radeon: finish radeon_driver.c import for register changes radeon: add some missing bits of superpatch for crtc registers radeon: put back in reverseddc code remove all CVS annotations from the ATI driver files radeon : add back tv-dac register write radeon: cleanup some bits from superpatch radeon: don't move connectors around if LVDS connector is connector 1 radeon: check DDC info for both connectors not just the pScrn one. update version number to 6.6.99 in preparation for a 6.7 branch at some stage always turn on DAC for scanning the CRT connect. radeon: set preferred modes radeon: separate static controller setup from non-static add initial controller binding, remove uses of mergedfb test add radeon Init2 function, to abstract some mergedfb mode setting radeon: abstract init display bandwidth from mergedfb move HasCRTC2 to pRADEONEnt from info make CRTCs malloc'ed rather than part of Ent radeon: dynamically allocate connectors move crtc to connector mapping to a connector to crtc mapping. re-program bios 4 scratch so we know if the LCD panel is there post startup move monitor detection into single function Add radeon randr 1.2 initial attempt at support update radeon driver to Intel driver xf86Crtc interfaces hook up randr crtc setting bring over latest code from Intel git repo. hook up mode detect to new randr code remove some code that moved into server fix up output namings use getddcmodes properly fix remnant of old code fixup even more remnants of old code add randr pre init add displayWidth set add setdpi call radeon: don't touch Controller[1] on single CRTC cards. radeon: fixup fastwrites to be always off bring radeon randr code inline with intel randr code more minor fixes to get radeon up to speed make restore mode non-static add more files for back compat fixup some function calls hook up detection of output fix typo hook up crtc modesetting fix cursor handling fix LUT set binding for new heads hook up DPMS through xf86 minor randr1.2 fixups always create a second CRTC I think this should solve any problems radeon: destroy and fixup LVDS crtc disable dpms on/off as it was writing regs before we wanted to radeon: move blank/unblank to use randr use more randr-ish names for variables radeon: get randr-1.2 mode switching mostly working on my laptop comment out unblanking on VT switch make dri work again move some cursor code around for show/hide remove mergedfb remove alot more mergedfb support add back overlay CRTC chooser this may not work yet remove unused merged code radeon: fixup some of the video code after randr remove more mergedfb relics removed even more mergedfb remains remove isused and isactive remove unused restore palette function fix LVDS by moving bios reading around radeon: move some i2c registers out into common register file make radeon randr build against master server cleanup radeon code against master server fix typo radeon: add option to set pci aperture size from config file update number to 6.6.191 for rc release radeon: fix up crtc debug dereference problem radeon: add support for enabling direct rendering on RS480 radeon: fix build since patches for IBM don't actually build radeon: fix build on older server radeon: add support for DDC on some laptop chipsets radeon: add support for DDC on some laptop chipsets radeon: another attempt at fixing the mergedfb refresh rate rs480: only has single dac rs480: make second crtc work with magic number in magic register. rs480: more unknown regs radeon: add bios quirk for nx6125 monid remove these syncs, at least on rs480 it doesn't break randr-1.2: make native mode preferred randr-1.2: add support for vblank on both heads randr-1.2 increase dri minor version for mesa to know we can do vbl on both radeon: disable vbl interrupts when no 3d is running on a new enough drm radeon: disable irqs at server start until 3D app starts radeon: disable irqs at server start until 3D app starts radeon: cleanup some pieces of the dpms/blank register programming ati: add -Wall for gcc radeon: remove unused variables update configure.ac for 6.6.193 release updated release numbering for randr 1.2 radeon: cleanup some warnings David Airlie (14): fix up issue with bindings and mergedfb fixup init disp bandwidth radeon: fixup problem with cursor not going off radeon: always reset cursor regs update to latest intel codebase modulo using their mode set add locks for crtcs and some missing lines from intel update fixup callers to crtcsetmode fixes my cursors move i2c buses into outputs bring over setproperty from intel code fixup mode probing a bit neater rename some variables to output from connector get rid of mergetype and displaytype from radeon.h remove all the locking glxgears runs fine over screen changes without it remove assignments to pOutput before it is created David Woodhouse (1): Fix radeon vs fbdev pitch calculation. Eamon Walsh (2): Convert callers of LookupWindow() to dixLookupWindow(). Revert "Convert callers of LookupWindow() to dixLookupWindow()." George Sapountzis (48): Bug 6242: [mach64] Use private DMA buffers. [mach64] Fix build for !defined(USE_EXA). [mach64] EXA: reserve DRI buffers only when DRI is active. Handle .99 in version patch level. Officially require libdrm 2.2.0 [mach64] Use asynchronous DMA mode by default. [mach64] Prefer double-buffered visuals for GLX. Use portable XORG_RELEASE_VERSION macro. Drop AVOID_NON_PCI. Drop probing for VGA, VGAWonder, 8514, Mach32. Keep PCI mach64 only (atipreinit.c, atimode.c), drop: Keep PCI mach64 only, drop: Clean pATI->Adapter and pATI->VGAAdapter. We always use the accelerator CRTC within the xserver. Always use the linear aperture. Banked memory is no longer needed. Drop support for non-programmable clocks. Minor refactoring of module (un)loading for TV_OUT. Drop checking of sparse I/O bases conflicts Sane probe, based on work by Luc Verhaegen <libv at skynet.be>. Add option "probe_sparse" to force probing sparse I/O bases. Drop pATI->Chipset, no longer used. Cleanup checks for pATI->PCIInfo. Minor refactoring of pATI->Block0Base computation. Drop non-PCI support from atividmem.c Consolidate atiaccel.c w/ atimach64accel.c, part 1. Consolidate atiaccel.c w/ atimach64accel.c, part 2. Consolidate aticursor.c w/ atimach64cursor.c Consolidate atixv.c w/ atimach64xv.c, part 1. Consolidate atixv.c w/ atimach64xv.c, part 2. [mach64] Enable RENDER acceleration on the Pro variants. [mach64] PreInit: Use goto bail. [mach64] PreInit: pitchInc is local. [mach64] PreInit: maxPitch, minor cosmetic. [mach64] Consolidate adjustments of mode timings, part 1. [mach64] Consolidate adjustments of mode timings, part 2. [mach64] Consolidate adjustments of mode timings, part 3. [mach64] Xv: use single surface and encoding structs. [mach64] Bug 5586: overlay scaler limited to 720 pixels on ATI Rage Pro. Move atiprobe.c from ati to atimisc. Move atioption.c from ati to atimisc. Move {atimach64,r128,radeon}_probe.c from ati to subdrivers. Drop now unneeded _X_EXPORT's. Fold FillIn() back to Probe(). Drop probing by driver name from "Device" section. [mach64] Set pATI->ChipRevision correctly, instead of overriding. [mach64] Use Mach64Chipsets[] instead of ATIChipNames[]. Move atichip.c from ati to atimisc. Henry Zhao (5): fix bug 6966 (Henry Zhao, Shem Multinymous) improved results for mergedfb xrandr refresh rates - bug 6966 9337: EDID modes do not participate in validation for CRT monitor 10205: Radeon driver's own mode validation code does not work properly Update CRT2pScrn->monitor->Last to reflect the iLisa Wu (1): Fix crash with no valid mode in xorg.conf's modelist and empty Virtual Jesse Barnes (6): Update for new CRTC resize hooks. Add prepare/commit hooks to output and crtc func table - fix randr 1.2 on pre-RV350 chips by enabling DAC_MACRO_CNTL writes RADEON: RADEON: Add cscope files to .gitignore Lisa Wu (2): RADEON: make sure RADEON_CRTC2_CRT2_ON bit state is properly accounted for RADEON: fix console restore on r3xx and r4xx Luc Verhaegen (2): Clean up PortInfo to CRTC mapping AGPFastWrite risk reduction. Matthias Hopf (5): Fixed typo in mode list generation. Fix crash if MergedFB and secondary head not found Fix inconsistent use of Mode lists. Set sync polarity restriction flags even for non-"digital separate" monitors. Disable RMX for IBM Lewis server. Matthieu Herrb (3): Fix build without XF86DRI Fix build without XF86DRI Fix build whithout XF86DRI Michel Daenzer (1): Bug #6756: Attempt to fix repeat picture acceleration. Michel D?nzer (40): Remove some dead code. radeon: Always call RADEONGetAccessibleVRAM(). radeon: Fix function name typo which could probably lead to an unresolved symbol. radeon: Add support for page flipping with EXA. radeon: Don't try to synchronize pages when the CP isn't running. radeon: Track screen damage all the time for page flipping. radeon: Use xf86XVFillKeyHelperDrawable() when available. radeon: Pitch doesn't matter when height == 1. Fix compile warnings. radeon: Bug #8989: Fix page flipping artifacts with XAA. Fix build warning. Increase default overlay scaler buffer width for RV350. Make log file output clear about origin of used overlay scaler buffer width. Fix radeon compile warnings. radeon: Make cache flushing for R300 series less invasive. radeon: Move Xv option handling code into RADEONPreInitXv(). radeon: Default to leaving AGP transfer mode and fast writes unchanged. radeon: Advertise 2D engine limits to EXA when we're not using the 3D engine. radeon: Fix build warnings. atimisc: Fix strict aliasing violation flagged by -Wall. radeon: Make sure 3D clients will re-upload textures to video RAM after LeaveVT. radeon: Disable CP line acceleration on RV280s. radeon: Minor BlockHandler cleanups. radeon: Unify code to release the CP. radeon: Only sync to hardware when really necessary with EXA. Fix advertised minimum minor version of the DRI module. radeon: Fix typo. radeon: Guard some MergedFB specific code with info->MergedFB tests. radeon: Don't shrink virtual size based on secondary modes. radeon: Always sort modes when adding to list. radeon: Link nearest modes by default for clone mode. radeon: Suppress debugging output by default. radeon: HW cursor cleanup. radeon: Provide new DRI texOffsetStart hook when available with EXA. radeon: Lots of warning fixes. radeon: Fix some more ErrorFs when setting mode. radeon: Don't loop indefinitely if no mode matches detected panel size. radeon: Fix panel size detection from registers with stretched mode programmed. radeon: Fix panel size detection from registers with stretched mode programmed. radeon: Fully zero-initialize info->CRT2pScrn->monitor. Paul TBBle Hampson (2): Fix for infinite loop in RADEONGetLVDSInfo RADEONProbePLLParameters sets pll->reference_div, and Roland Scheidegger (9): radeon: Check ROM more thoroughly before treating it as an x86 BIOS. radeon: bug #1462, predownscale to make HD video work. radeon: use overlay scaler native planar yuv capability. radeon: fix up packed yuv broken by planar yuv patch radeon: avoid unnecessary OUTPLL/INPLL calls when displaying video fix uninitialized mode regs (bug 9495) fix uninitialized mode regs (bug 9495) fix alignment issues with planar yuv and a bug with packed uyvy add missing part of last commit... Ronald Wahl (1): Hook up DFP power to DPMS events. Tilman Sauerbeck (1): Fixed 'make dist'. Timo Jyrinki (1): ati: fix detection of some newer radeons using ati wrapper Tormod Volden (2): RADEON: limit PanelPwrDly to 2000 ms on ATOM bios as per legacy bios RADEON: limit PanelPwrDly to 2000 ms on ATOM bios as per legacy bios git tag: xf86-video-ati-6.7.191 http://xorg.freedesktop.org/archive/individual/driver/xf86-video-ati-6.7.191.tar.bz2 MD5: f350217d7b56890ba714f0479866210f xf86-video-ati-6.7.191.tar.bz2 SHA1: 077b8bf872cdfc31e781b5c0f81bb8f17cc648dd xf86-video-ati-6.7.191.tar.bz2 http://xorg.freedesktop.org/archive/individual/driver/xf86-video-ati-6.7.191.tar.gz MD5: a4fb5391d53cdf4388a745d16cebfaba xf86-video-ati-6.7.191.tar.gz SHA1: 6fd4152b8f1ec506614bd92c978984914198ec21 xf86-video-ati-6.7.191.tar.gz -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.6 (GNU/Linux) iD8DBQFGzQRbm07k+YR03kARArXeAKCR2IJTdGTmZ3OdhKYczcSw80fVxwCeNhth OvnU0BazDBmWk34pRc2KGyc=90UE -----END PGP SIGNATURE-----