changelog

x.org上有关ati系列显卡最新驱动

        the problem in the comment only concerned pre-production cards that Hui
        tested apparently. Also set back the default value to what it was
        initially, fixes some more cards that don't have a proper table (Mac
        Mini for example)

commit 36c04b88416905b9ae305583e8e34fefdbfafc50
Author: Michel Daenzer <[e-mail]>
Date:   Mon May 1 17:49:41 2006 +0000

    Bugzilla #6755 <https://bugs.freedesktop.org/show_bug.cgi?id=6755> Patch
        #5536 <https://bugs.freedesktop.org/attachment.cgi?id=5536>:
    Change HostDataBlit interface to take dst_offset_pitch and coordinates
        instead of just a destination pointer, as the latter is not sufficient
        with tiling. Also, use HW clipping to avoid overwriting destination
        data outside of the specified width.
    Adapt to new HostDataBlit interface. This fixes corruption with
        UploadToScreen to the front buffer (from exaPutImage).

commit b90d3710e67caad0686efab491c2b76ae877920c
Author: Michel Daenzer <[e-mail]>
Date:   Sat Apr 29 21:30:23 2006 +0000

    Bugzilla #1992 <https://bugs.freedesktop.org/show_bug.cgi?id=1992> Patch
        #4988 <https://bugs.freedesktop.org/attachment.cgi?id=4988>: Fix Option
        "DisplayPriority" "HIGH" for R300 family.

commit 30f3ce5c5ba205a23748def11de019ee727897f3
Author: Michel Daenzer <[e-mail]>
Date:   Sat Apr 29 20:34:57 2006 +0000

    Properly de-initialize EXA driver data, so EXA can be initialized again in
        the next generation.

commit d705429c230c9cc1645ae540a51dba62cae9f23e
Author: Michel Daenzer <[e-mail]>
Date:   Sat Apr 29 20:18:48 2006 +0000

    Fix tests for (un)forcing HW cursor. The broken tests caused problems when
        switching to console with 3D clients running.

commit 6142cb079d4a5844e4863c800baf4ae19c636af5
Author: Michel Daenzer <[e-mail]>
Date:   Sat Apr 29 20:12:38 2006 +0000

    Prefer double-buffered visuals. As a consequence, the root window's visual
        is double-buffered by default, which is desirable in a number of
        situations, in particular when running a GLX compositing manager with
        AIGLX.

commit eb841974d041b19461b7fb726687df8938b2e39b
Author: Eric Anholt <[e-mail]>
Date:   Fri Apr 28 02:55:08 2006 +0000

    Bug #6761: Fix font rendering in non-DRI mode by replacing RB2D_DSTCACHE_*
        with RB3D_DSTCACHE_*. The RB2D versions are read-only mirrors of the
        RB3D. This is masked when DRI is enabled because the DRM uses the right
        registers in its flushing.

commit 85e2ee6de6fd347aad66785ec7119f92013cd994
Author: Dave Airlie <[e-mail]>
Date:   Wed Apr 26 08:38:47 2006 +0000

    Make the ATI driver use the PCI BAR size to determine the MMIO register
        aperture size. Otherwise it tries to overmap things if they are up near
        the end of memory.

commit 13b19edf5accb7427436d3f0b1c9a9efb4fe98fa
Author: Eric Anholt <[e-mail]>
Date:   Wed Apr 26 07:26:15 2006 +0000

    Add support for some component-alpha operations (InReverse, OutReverse)
        necessary for testing upcoming general component alpha acceleration
        assistance in EXA. Tested with rendercheck on an r100.

commit 44a45bcb82d49349f4fdfae3e4b8f24f98d99b46
Author: Eric Anholt <[e-mail]>
Date:   Tue Apr 25 23:31:56 2006 +0000

    Use a CPPFLAGS with XORG_CFLAGS so that the exa test can actually find its
        headers and not fail. Also correct some nearby spelling.

commit cc1825b503ce7e785cbb4611fd177477a1081b6e
Author: Benjamin Herrenschmidt <[e-mail]>
Date:   Mon Apr 24 07:44:52 2006 +0000

    Fix enabling of external VGA output on some cards with dual entity setup
        instead of MergedFB. This was broken due to bogus manipulations of a
        register between the 2 entities. This fix is a big "violent" in that it
        always enables the analog output regardless of what is plugged. It
        should work though (at the expense of some additional power
        consumption). A proper fix should be easy to do once Alex gets his
        output mapping rework in.

commit a66145bc9d487f7e19237cf292f12750cb959f61
Author: Dave Airlie <[e-mail]>
Date:   Fri Apr 21 00:38:44 2006 +0000

    Bugzilla 6672: fix interrupts on radeon since memmap changes

commit f1f057df9aaba42986d9e20afe6f609a90318333
Author: Dave Airlie <[e-mail]>
Date:   Fri Apr 21 00:34:43 2006 +0000

    Add support for backing up the PCIE GART table on VT switch. This makes
        suspend/resume work a lot better on PCIE cards.

commit 31396df3b0101571bb85a4a66b67f5a8b0c992a9
Author: Adam Jackson <[e-mail]>
Date:   Sat Apr 8 01:19:17 2006 +0000

    Bump to 6.6.0 and release. This module's versioning is completely screwed,
        so until we split out mach64/r128/radeon please just increment the
        package number whenever you make significant changes.

commit d706f494ed22afd369991e8c75a2d4f5aed15305
Author: Aaron Plattner <[e-mail]>
Date:   Fri Apr 7 23:06:02 2006 +0000

    Add a DrawablePtr argument to the XV functions to pave the way for
        redirected video.

commit f94685aec26b65f60c00a4b1bcd8f26bd80eb6e4
Author: Eric Anholt <[e-mail]>
Date:   Tue Apr 4 07:42:36 2006 +0000

    Bug #5632: Fix up ati.4 to explain that it is a wrapper driver for the
        other three. While here, touch up radeon.4, and add an authors list
        gleaned from checking copyright headers.

commit 58c9b1d84f514755945e7a0d25fd5530af34ce95
Author: Eric Anholt <[e-mail]>
Date:   Sun Apr 2 21:45:12 2006 +0000

    Add missing .cvsignore lines.

commit f1ce6170ef21e1f9769337c3138cd8d65c33e40a
Author: Alex Deucher <[e-mail]>
Date:   Sat Apr 1 23:02:40 2006 +0000

    - Fix dpi when switching from clone to dualhead with MergedFB.
    - Add ConstantDPI option to force a particlar dpi across mode changes Both
        based on Thomas Winischhofer's sis code.

commit 607f18cfbfc2a7b5509f8b567bf510fce31b361e
Author: Benjamin Herrenschmidt <[e-mail]>
Date:   Mon Mar 27 06:12:57 2006 +0000

    Fix a regression of my latest fix that broke color tiling on some setups.
        RADEONRestoreMemMapRegisters() must not clear CRTC_OFFSET_CNTL and
        friends unconditionally but only when there is a map change or it will
        clear them after the DRI updates the AGP location, thus turning tiling
        back off on the CRTC. Also remove a duplicate call to
        RADEONInitCommonRegisters().

commit bf1c4471137f10b7de9d79519ca2240058997e18
Author: Roland Scheidegger <[e-mail]>
Date:   Thu Mar 23 18:04:17 2006 +0000

    commit the actual changes instead of just the Changelog...

commit aa5fff9bae4818bbf99e8d9caa8f2ce9905294ca
Author: Roland Scheidegger <[e-mail]>
Date:   Thu Mar 23 17:59:25 2006 +0000

    Use programmable tap coefficients for (hopefully) better video image
        quality when downscaling. Remove unnecessary code programming the tap
        coefficients twice. Disable some code which is supposed to improve
        filtering quality but causes very visible image quality problems at
        least with rv250 when upscaling. Fix potentially broken FOURCC_RGB16T
        format (untested if it was broken, untested if it works now...). Change
        some magic numbers to (new) macro defines.

commit 17c52ff15ebe2cf220d5988ee13b599c41799886
Author: Benjamin Herrenschmidt <[e-mail]>
Date:   Thu Mar 23 01:37:15 2006 +0000

    Call RADEONRestoreCommonRegisters() before RADEONRestoreCrtc2Registers()
        instead of after. I can't find any good reason why we did it backward
        until now and it definitely causes problems as the workaround for VT
        switch in RADEONRestoreCommonRegisters() will break output to the
        secondary head in various situations. Also does an unrelated minor
        update to a comment.

commit 673e4d7501861066fdd0874919a4524b820b944b
Author: Kristian H酶gsberg <[e-mail]>
Date:   Thu Mar 23 00:06:56 2006 +0000

    Add atipciids.h.

commit 5dcb4152dca6ac1fe9342f44eb48c0198b15772e
Author: Kristian H酶gsberg <[e-mail]>
Date:   Wed Mar 22 22:30:14 2006 +0000

    Convert use of xf86fopen() and other xf86 wrapped libc symbols to use libc
        symbols directly. The xf86* versions aren't supposed to be used
        directly.
    Drop libc wrapper; don't include xf86_ansic.h and add includes now missing.

commit 21acdf371e3cf913c5affbd3e86641cfb115311c
Author: Benjamin Herrenschmidt <[e-mail]>
Date:   Sat Mar 18 00:08:24 2006 +0000

    Clear the offsets when updating the memory map instead of when enabling the
        CRTCs and try to make that code more reliable. Doesn't disable the
        CRTCs beforehand.

commit 323ecb92e40d71c5ef994b41b6d8dedba6dd6203
Author: Roland Scheidegger <[e-mail]>
Date:   Fri Mar 17 03:00:53 2006 +0000

    Fix various small cosmetic issues. Change a driver message, get the order
        right for requesting drm versions, replace the use of some numbers with
        the respective macro defines in radeon_video.c, and add some more macro
        defines. None of that really matters.

commit dc552edf191e14249421f8b27146874f16b80c1a
Author: Benjamin Herrenschmidt <[e-mail]>
Date:   Thu Mar 16 21:53:58 2006 +0000

    Add missing return statement (the driver would not start randomly), add
        RV410 to the list of new generation cards for the memory map setup and
        fix a bogus message.

commit 901409f47cc990721c1648b1c08e23450e6a4452
Author: Benjamin Herrenschmidt <[e-mail]>
Date:   Thu Mar 16 04:32:22 2006 +0000

    Fix incorrect CRTC2_OFFSET values when using old-style dual head (not
        MergedFB). Fixes Xinerama for me.

commit 4d6656062129da0489eb4bc898871379ba891d8f
Author: Benjamin Herrenschmidt <[e-mail]>
Date:   Thu Mar 16 03:09:34 2006 +0000

    Add various workarounds that seem to fix some remaning lockup scenarios I'm
        experiencing with the driver when setting the memory map. Some of the
        magic delays are a bit dodgy but they seem to work, I suppose I can't
        do better now without help from ATI. Also removed some really too noisy
        debug messages.

commit 36799e3ea7ddde5fce528e8bc17ea942933a9111
Author: Benjamin Herrenschmidt <[e-mail]>
Date:   Wed Mar 15 04:03:37 2006 +0000

    Fix various issues with DRI & server recycle by putting DRICloseSreen in
        the normal CloseScreen callback chain at a spot that matches the call
        to DRIFinishScreenInit in order to unwrap at the right time.

commit 50923b6a9325592edc8b02b17646f96b9b8c4a01
Author: Luc Verhaegen <[e-mail]>
Date:   Wed Mar 15 00:46:33 2006 +0000

    Add some autoconf to keep this driver from building EXA when a slightly
        older server with older exa is installed.
    The way in which this problem was treated by the actually related
        developers explains a lot about this driver.

commit 74f4ed4275866475b3c182fae962df2fc0946b85
Author: Benjamin Herrenschmidt <[e-mail]>
Date:   Mon Mar 13 02:57:17 2006 +0000

    Fix typo in the function setting the color space transformation factors for
        old radeon

commit 65625c6c2053792cd14b54c904bf46f360ef1565
Author: Benjamin Herrenschmidt <[e-mail]>
Date:   Mon Mar 13 00:09:44 2006 +0000

    Earliest fix introduced a bug due to another chicken & egg problem in card
        detection. Now split PreInitConfig into PreInitChipType and PreInitVRAM
        so that PreInitDRI can be done just in between. Restores proper AGPx4
        functionality (when enabled)

commit 3a290115af41b0ecb94eb201d7e7abb42ebbfec6
Author: Benjamin Herrenschmidt <[e-mail]>
Date:   Sun Mar 12 06:53:27 2006 +0000

    Stop using xf86PciInfo.h, instead use a local copy of the PCI IDs we need
        in atipciids.h so we can update the ATI driver independently of the
        server when new chips are added

commit d6b8f0a5eeb332563ec3ec7082f03be1797d26d8
Author: Alan Coopersmith <[e-mail]>
Date:   Sat Mar 11 04:19:47 2006 +0000

    Add #ifdef XF86DRI guards around references to info->allowPageFlip and
        info->depthMoves, since they are only defined if building with DRI.

commit 69fe90bc501df9bd9e74b894bfca79a4025725bd
Author: Roland Scheidegger <[e-mail]>
Date:   Fri Mar 10 13:00:49 2006 +0000

    Use "FBTexPercent" option for XAA too (only for increasing memory reserved
        for textures).

commit 8033d7423267fc1b3725df6a077dd910b586daaf
Author: Benjamin Herrenschmidt <[e-mail]>
Date:   Thu Mar 9 23:26:27 2006 +0000

    More memory map fixes !!! This time, we add a way to get the DRM kernel
        module version early, in order to properly limit the framebuffer size
        when a module that doesn't support the new map is loaded. That will
        improve backward compatibility. I also made the minimum required kernel
        module version 1.3 which hopefully shouldn't be a problem for anybody,
        and cleaned up the ordering in which we test for DRM version to enable
        features

commit c637939d3c4ce5cdddbc5200d5ffbca4c3927050
Author: Michel Daenzer <[e-mail]>
Date:   Thu Mar 9 15:41:16 2006 +0000

    RN50: Skip modes that exceed memory bandwidth.
    Bugzilla #5766 <https://bugs.freedesktop.org/show_bug.cgi?id=5766> Patch
        #4636 <https://bugs.freedesktop.org/attachment.cgi?id=4636>
    - Acknowledge that RN50 only has one CRTC, and use this to distinguish it
        from RV100.
    - Fix detection of RN50 memory type and bus width.
    - Model RN50 memory bandwidth limits by capping the pixel clock range based
        on memory clock, bpp and memory bus width. (ATI Technologies Inc.)

commit 890b95a3ddad7634f3aea635cb236b221e902acc
Author: Eric Anholt <[e-mail]>
Date:   Thu Mar 9 06:06:25 2006 +0000

    Update drivers that support EXA for the new EXA ABI. This consists of
        moving all the accel and card members into the driver ptr, filling in
        the exa_major/ exa_minor fields, and always using LoadSubModule so we
        can check the module version up front. Only tested on ATI.

commit e19ccfa18fee046c54d7a9401e4e0098b2300598
Author: Michel Daenzer <[e-mail]>
Date:   Fri Mar 3 16:41:41 2006 +0000

    Add Option "FBTexPercent" to override the amount of video RAM reserved for
        OpenGL textures with EXA.

commit ca00035eccb6e3495b27f8e64f2d016d8d2a9ce2
Author: Benjamin Herrenschmidt <[e-mail]>
Date:   Thu Mar 2 02:05:17 2006 +0000

    Fix the fix ... I didn't interpret PciInfo->size properly and forgot that
        the function works in Kb not bytes... Ooops.

commit b2145aea36bb035bff048366c607b967d70fff49
Author: Benjamin Herrenschmidt <[e-mail]>
Date:   Wed Mar 1 21:35:14 2006 +0000