[ANNOUNCE] xf86-video-radeonhd 1.3.0 Release

Matthias Hopf mhopf at suse.de
Fri Oct 9 09:34:05 PDT 2009


Announcing the 1.3.0 Release of the xf86-video-radeonhd driver.


RadeonHD is the X.org X11 driver for AMD GPG (ATI) r5xx/r6xx/r7xx chipsets.
The development is driven by a community of open source developers, supported
by Novell and AMD at the time of writing.
AMD provides free documentation for the chipsets.

Note the wiki on http://wiki.x.org/wiki/radeonhd


Version 1.3.0 improvements:

  - Added support for RV740, M92, M93, M97.
  - Added support for HDMI audio on RS690 and R700.
  - Added support for power management.
  - Implemented almost correct analysis of PowerPlay AtomBIOS tables.
  - 2D acceleration (EXA) is enabled by default now, except on RV740.
  - Backlight handling finally fixed - compatible with xbacklight 1.1.1.
  - Overhaul of memory controller setup. Fixes many "MC not idle" issues.
  - Overhaul of cursor handling. Fixes most cursor bugs.
  - Selectable color space for XVideo on R6xx/R7xx.
  - Tons of bug fixes (DDC, EXA, LUT, RandR, AtomBIOS).
  - More quirk table entries.
  - Shave support (cleaner compilation output).
  - All warnings fixed.
  - Some start of Developer's documentation in README.coding.


git tag: 1.3.0

http://xorg.freedesktop.org/archive/individual/driver/xf86-video-radeonhd-1.3.0.tar.bz2
MD5:  7b6641aa9d836f1621b9b220ad6771b8  xf86-video-radeonhd-1.3.0.tar.bz2
SHA1: 4cdcdbcdc6ec7cd4caa19afdbfc34a8bec461f56  xf86-video-radeonhd-1.3.0.tar.bz2

http://xorg.freedesktop.org/archive/individual/driver/xf86-video-radeonhd-1.3.0.tar.gz
MD5:  d635f5f0f6ab821662502ce87dc7077a  xf86-video-radeonhd-1.3.0.tar.gz
SHA1: fd85910b540e64639155f94d136bd07c314f0bf5  xf86-video-radeonhd-1.3.0.tar.gz



git ShortLog:
-------------

Alex Deucher (6):
      Fix build on systems without pci_device_enable()
      RS690: fix EXA corruption
      Fix build on systems without pci_device_enable()
      Fix scissor offsets for r5xx in last patch
      r7xx: add some new pci ids
      rs880: enable accel

Christian Koenig (2):
      Initial RS690 HDMI Audio support.
      R700 HDMI audio support.

Christian König (1):
      Silence audio stream option.

Dave Airlie (3):
      radeonhd: update for resources/RAC API removal
      radeonhd: further RAC/resources changes
      radeonhd: change to using ABI version check

David Morrison (1):
      LUT: Fix syntax error in 59085c4a

Egbert Eich (12):
      Remove superfluous function arguments.
      Spelling fixes.
      Move test for backlight support enabled to proper place.
      Reenabling AtomBIOS controlled backlight.
      Add better fallback heuristics for acceleration methods.
      Clean up most warnings.
      Add ACPI controlled Backlight support for Linux.
      PM: Removed unnneeded define.
      ID: Added quirk entry for HIS Excalibur Radeon XT1650 Pro IceQ 256M
      Properites: Get HW for Backlight setting on every query.
      Improve test for disabled differential clock driver.
      Improve DAC load detection on RS780.

Florian Forster (1):
      man/radeonhd.man: Fix two typos.

Hans Ulrich Niedermann (2):
      Fix RHDRegWrite macro invocation breakage
      Hide README, radeon.man non-updates if --enable-shave

Jakub Zawadzki (1):
      rhd_conntest: mmap() returns -1 on error, not NULL

Javeed Shaikh (1):
      Xv: Fix YUV scaling and use Rec709 coefficients

Luc Verhaegen (4):
      Cursor: fix logic error in cursor visibility checking.
      DRI: Fix VBlankInterrupt setting when not using RandR1.2.
      Revert "More idle/flush swaps."
      Revert "Fix softlocks on rs690. Idle commands have to be flushed to be of any use."

Marc Dietrich (2):
      silence some compiler warnings
      Fix softlocks on rs690. Idle commands have to be flushed to be of any use.

Marvin (1):
      Port lockup fix from radeon

Matthias Hopf (74):
      Extend ugly hack for CARD64 to rhd_mc.h.
      Consolidated chip identifiers.
      Extend ugly hack for CARD64 to rhd_mc.h. This time actually working.
      FB mapping cleanup didn't reflect changes to allocation. Fixes #21233.
      Fix #13405: Cursor corruption
      Nuke some trailing whitespace.
      man: cleanup
      rhd_dump: move pci_device_enable to correct case (pciaccess)
      utils: Only allow -e if pci_device_enable() is available.
      modes: Fixup broken panel modes if possible
      Fix switch construction in LVDSPropertyControl
      Add quirk entry for 0x7146:0x174B:0x0940. Bug #21947.
      Add some more PCI Ids, improve RV740 handling, fix descriptions.
      Add AtomBIOS based static powermanagement enablement
      Initialize fresh connector object correctly in AtomBIOS case.
      Add quirk table entry for Asus F3JR.
      Adapt for nuked xf86LoaderReqSymLists() in Xserver 1.6.99.1.
      randr: Remove useless error messages when fetching unknown/invalid properties.
      Move setting of Private->Modes to correct place.
      Refactor AtomBIOS option handling.
      randr: Add AtomBIOS property for on-the-fly setting. Experimental.
      randr: Export EDID property correctly.
      randr: XA_STRING property values are not necessarily 0-terminated.
      randr: Reformat AtomBIOS parameter correctly when setting.
      Add quirks table entry for ASUS EAH3450.
      Add Developer's documentation. Or better: Some start of.
      Add shave (nuking excessive libtool output).
      Rewrite to use structured data for power settings.
      Rework powermanagement: Introduce the notion of levels, change interface.
      Consolidate prefixes for ATOM_*.
      Add code fetch chip clock and voltage limits and validate them.
      Typo.
      Improve chip limit validations.
      Improve output for powermanagement.
      Improve chip limit validations. Again.
      Improve chip limit validations. Again. 2.
      Consolidate structures in rhd_pm.c and rhd_atombios.c.
      Remove warnings in rhd_pm.c on 32bit.
      Import known good chip settings from AtomBIOS.
      Add check for -Wno-unknown-pragmas in configure.ac.
      Fix C++ comments warnings in reverse engineered AtomBIOS.
      Validate current clocks - some AtomBIOSes provide broken values.
      Work around ParseTable returning CD_SUCCESS if the command table doesn't exist
      Fix autoconf issues.
      Typo.
      RS690 MC idle: Wait for system idle instead of sequencer idle.
      Rename registers in PCIE space.
      MC: Major overhaul for RS600, RS690, RS760, RS780, RS880.
      Fix logic bug in sideport memory detection.
      INSTALL: xorgversion.m4 is obsolete for a long time already.
      Non-existent AtomBIOS command tables should issue a warning only.
      Consider Backlight property to be a known good property now.
      Add atombios_rev.h to files list.
      Also combine outputs if only the second of them has .dual set.
      Fix segfault due to missing Card NULL check.
      atombios.h prohibits -std=c99 due to unnamed unions
      More idle/flush swaps.
      Fix typo in non-DRI case.
      Add ugly hack for uint32_t to rhd_pm as well.
      Add quirk table entry for ATI Radeon X1550.
      Add quirk table entry for Wyse R-Class.
      RV740: disable DFS and Composite
      Disable DRI by default if shadowfb is explicitly selected.
      Fix missing parenthesis.
      RV740: Choose shadowfb by default.
      randr: Select virtual large enough for typical dual-monitor situations
      pm: Do NOT set idle engine clocks to default/2 unless ForceLowPowerMode is used.
      pm: Use minimum known working frequency instead of default/2.
      pm: Ignore validation when setting negative engine/mem frequencies.
      pm: Improve wording of lowPowerModeEngineClock logging.
      i2c: Hardware may need a while to indicate availability to the host
      README.coding: Add rhd_acpi.[ch] and atombios_rev.h
      pm: Comment out currently unused variable.
      Bump to 1.3.0. Update README.

Michael Cree (3):
      Added code to use libpciaccess
      Added code to use libpciaccess and access PCI ROM
      Updated Makefile in utils/conntest to enable link with libpciaccess.

Peter Hutterer (1):
      radeonhd: Update to xextproto 7.1 support.

Rafał Miłecki (12):
      clean power management code
      PM: Save and restore engine clock on VT switching
      PM: fix broken engine clock setting logic
      PM: add RHDGetDefaultMemoryClock
      PM: get rid of EnableForced field
      PM: add memory downclocking. Commented out for now.
      RandR: fix return in DigPropertyControl
      RandR: add returning HDMI property value
      AtomBIOS: add pixel clock's ucFracFbDiv to debugging messages
      PLL: RV620 PLL shutdowning fix (sometimes we need unused PLL running)
      PM: Fix printing units and variable typo in memory code
      Fix build with --disable-atombios

Roland Scheidegger (1):
      R6xx/R7xx: fix pixel centers

Thomas Jaeger (1):
      Fall back to software for unsupported repeat modes

Tristan Hoffmann (1):
      man: Add lowpower options.

Yang Zhao (22):
      AtomBIOS: add hooks to call memory and engine clock getters and setters
      PM: Add functions for get/setting engine and memory clocks
      PM: Add ForceLowPowerMode and LowPowerModeEngineClock options
      Cursor: Do not change cursor mode when disabling; fixes bug #13405
      Cursor: refactor RandR cursor code
      Cursor: Fix remaning corruption cases
      Cursor: refactor non-RandR cursor code
      Cursor: clean-up: eliminate Width and Height from rhdCursor struct
      Cursor: optimize lockCursor()
      MC: Remove duplicated definition of SRBM_STATUS
      Raise warnings when 2D acceleration initialization fails
      Fix incorrectly reporting XAA/EXA init as failing for r5xx
      Xv: Make color space selectable on r6xx/r7xx
      Make EXA the default AccelMethod on r5xx
      Fallback to ShadowFB on r6xx/r7xx if DRI pre-init fails
      Use DRM and EXA by default on r6xx/r7xx
      Do not log about "Unknown card detected" when AtomBIOS is available
      Put back logging of PCI IDs not in database during probing.
      LUT: Change internal functions to reflect hardware
      LUT: Use DC_LUT_30_COLOR for saving/restoring LUT
      LUT: Make sure LUT_RW_INDEX is reset to 0 before loading LUT
      LUT: Fix RHDLUTCopyForRR() not actually modifying LUT


Matthias

-- 
Matthias Hopf <mhopf at suse.de>      __        __   __
Maxfeldstr. 5 / 90409 Nuernberg   (_   | |  (_   |__          mat at mshopf.de
Phone +49-911-74053-715           __)  |_|  __)  |__  R & D   www.mshopf.de
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 197 bytes
Desc: not available
URL: <http://lists.x.org/archives/xorg/attachments/20091009/9d725a58/attachment.pgp>


More information about the xorg mailing list