xenocara/driver/xf86-video-openchrome/ChangeLog
fcambus b3ae9f8426 Update openchrome to 0.6.176.
This fixes display on my VX900 systems.

OK matthieu@
2018-12-09 21:07:33 +00:00

2714 lines
88 KiB
Plaintext

2011-05-10 Bartosz Kosiorek <gang65@poczta.onet.pl>
Workaround EXA crash with new libcairo2 (#298)
* src/via_accel.c: (viaExaPrepareComposite):
2011-05-08 Bartosz Kosiorek <gang65@poczta.onet.pl>
Fix cursor garbage after suspend/resume for Xserver 1.10 (#405)
* src/via_video.c: (viaRestoreVideo):
2011-05-04 Bartosz Kosiorek <gang65@poczta.onet.pl>
Fix XAA displaying issues
* src/via_accel.c: (viaInitXAA):
2011-01-23 Bartosz Kosiorek <gang65@poczta.onet.pl>
Enable hardware cursor for VX900
* src/via_cursor.c: (viaHWCursorInit):
* src/via_mode.c: (ViaModeSet):
2010-12-16 Bartosz Kosiorek <gang65@poczta.onet.pl>
Merge vx900_branch - initial VX900 support
* src/via_accel.c: (viaFlushPCI), (viaDisableVQ),
(viaInitialize2DEngine), (viaAccelSync), (viaPitchHelper),
(viaInitXAA):
* src/via_bandwidth.c: (ViaSetPrimaryFIFO), (ViaSetSecondaryFIFO):
* src/via_bios.h:
* src/via_crtc.c: (ViaFirstCRTCSetMode), (ViaSecondCRTCSetMode):
* src/via_cursor.c: (viaHWCursorInit), (viaCursorStore),
(viaCursorRestore), (viaShowCursor), (viaHideCursor),
(viaSetCursorPosition), (viaLoadCursorImage), (viaSetCursorColors):
* src/via_driver.c: (VIASetupDefaultOptions), (VIAPreInit),
(VIALeaveVT), (VIASave), (VIARestore), (ViaMMIOEnable),
(ViaMMIODisable), (VIAMapFB), (VIAWriteMode), (VIACloseScreen):
* src/via_driver.h:
* src/via_id.c:
* src/via_id.h:
* src/via_mode.c: (ViaDFPDetect), (ViaOutputsDetect),
(ViaOutputsSelect), (ViaGetMemoryBandwidth), (ViaSetDotclock),
(ViaModeSet):
* src/via_mode.h:
* src/via_panel.c: (ViaPanelScaleDisable), (ViaPanelPreInit),
(ViaPanelGetSizeFromDDC):
* src/via_video.c: (DecideOverlaySupport):
* src/via_xvmc.c: (ViaInitXVMC):
2010-12-15 Bartosz Kosiorek <gang65@poczta.onet.pl>
Enable the new mode switch and panel support on K8M800 and VM800 chipsets
* src/via_driver.c: (VIASetupDefaultOptions), (VIAPreInit):
* src/via_mode.c: (ViaModeSet):
2010-11-10 Bartosz Kosiorek <gang65@poczta.onet.pl>
Replace the deprecated functions
xalloc/xrealloc/xfree/xcalloc with
malloc/realloc/free/calloc.
Refer to "/xserver/include/os.h"
* src/via_accel.c: (viaSetupCBuffer), (viaTearDownCBuffer),
(viaInitExa), (viaExitAccel), (viaFinishInitAccel):
* src/via_dga.c: (VIASetupDGAMode):
* src/via_dri.c: (VIAInitVisualConfigs), (VIADRIScreenInit),
(VIADRICloseScreen):
* src/via_driver.c: (VIAFreeRec), (VIAProbe), (VIAPreInit),
(VIACloseScreen):
* src/via_memcpy.c: (viaVidCopyInit):
* src/via_swov.c: (Upd_Video):
* src/via_vbe.c: (ViaVbeSetMode):
* src/via_video.c: (viaExitVideo), (viaStopVideo),
(viaDmaBlitImage):
* src/via_xvmc.c: (cleanupViaXvMC), (ViaCleanupXVMC),
(ViaXvMCCreateContext), (ViaXvMCCreateSurface),
(ViaXvMCCreateSubpicture), (ViaXvMCDestroyContext),
(ViaXvMCDestroySurface), (ViaXvMCDestroySubpicture),
(viaXvMCInitXv):
2010-10-24 Bartosz Kosiorek <gang65@poczta.onet.pl>
Siragon ML-6200 laptop support
* src/via_id.c:
2010-06-24 Jon Nettleton <jon.nettleton@gmail.com>
PM800 also uses the CME Engine. Setup the hqv_cme_regs
for it.
* src/via_swov.c: (VIAVidHWDiffInit):
2010-06-09 Bartosz Kosiorek <gang65@poczta.onet.pl>
Fix freeze on 64bit system for K8M800 chipset
* src/via_dri.c: (VIADRIAgpInit):
2010-05-04 Jon Nettleton <jon.nettleton@gmail.com>
Re-enable the Virtual Queue for the VX800/VX855 chipsets.
* src/via_accel.c: (viaDisableVQ), (viaInitialize2DEngine):
2010-05-04 Jon Nettleton <jon.nettleton@gmail.com>
Disable certain hardware clipping options for the VX855.
These cause the 2d engine to become unstable when in
16-bit mode.
* src/via_accel.c: (viaInitXAA):
2010-05-04 Jon Nettleton <jon.nettleton@gmail.com>
Add an undocumented option which allows certain I2C buses
to be probed at startup. This allows workarounds for custom
chipset makers that have used the VX855 I2C buses for other
purposes.
* src/via_bios.h:
* src/via_driver.c: (VIASetupDefaultOptions), (VIAPreInit):
* src/via_driver.h:
* src/via_i2c.c:
2010-05-04 Jon Nettleton <jon.nettleton@gmail.com>
Improve 2d performance on chipsets that don't have
AGP/PCIe support yet.
* src/via_accel.c: (viaSetupForScreenToScreenCopy),
(viaSetupForSolidFill), (viaSetupForMono8x8PatternFill),
(viaSetupForColor8x8PatternFill),
(viaSetupForCPUToScreenColorExpandFill),
(viaSubsequentScanlineCPUToScreenColorExpandFill),
(viaSetupForImageWrite), (viaSubsequentImageWriteRect),
(viaSetupForSolidLine), (viaSetupForDashedLine), (viaInitXAA):
2010-05-04 Jon Nettleton <jon.nettleton@gmail.com>
Put timeouts on our while statements. These codepaths
should be interrupted by a hardware state change, but
if something goes wrong they loop forevere. Let's try
and behave a little by putting a timeout on these loops.
* src/via_swov.c: (viaWaitHQVFlip), (viaWaitHQVFlipClear),
(viaWaitHQVDone):
* src/via_video.c: (Flip):
2010-05-04 Jon Nettleton <jon.nettleton@gmail.com>
Add XVideo support for the VX855 Chipset.
To support this chipset I have added HWDiff->HQVCmeRegs
that allows handling differing register values, and
HWDiff->dwNewScaleCtl which allows selection of a
new Video scaling engine needed for the VX800/VX855
chipsets.
* src/via.h:
* src/via_bandwidth.c: (ViaSetSecondaryFIFO):
* src/via_driver.h:
* src/via_swov.c: (SaveVideoRegister), (VIAVidHWDiffInit),
(viaOverlayHQVCalcZoomWidth), (viaOverlayHQVCalcZoomHeight),
(ViaSwovSurfaceCreate), (SetHQVFetch), (Upd_Video):
* src/via_swov.h:
2010-04-21 Bartosz Kosiorek <gang65@poczta.onet.pl>
Replace RegionsEqual with REGION_EQUAL and use
the xf86XVFillKeyHelperDrawable instead of xf86XVFillKeyHelper
* src/via_video.c: (viaReputImage), (viaPutImage):
2010-03-07 Bartosz Kosiorek <gang65@poczta.onet.pl>
Fix segfaults with EXA and XV (Ticket #359)
Tested on K8M890 and VN800
* src/via_video.c: (viaReputImage), (viaPutImage):
2010-02-10 Bartosz Kosiorek <gang65@poczta.onet.pl>
Fix bug with suspend and VT switch on VX800 and 64bit systems
* src/via_driver.h:
* src/via_video.c: (viaResetVideo), (viaSaveVideo),
(viaRestoreVideo), (viaExitVideo):
* src/via_video.h:
2010-01-24 Bartosz Kosiorek <gang65@poczta.onet.pl>
Fix starting address restore and save (initial 64-bit support)
* src/via_crtc.c: (ViaFirstCRTCSetMode),
(ViaFirstCRTCSetStartingAddress):
* src/via_dri.c: (VIADRIAgpInit):
* src/via_driver.c: (VIASave), (VIARestore):
* src/via_driver.h:
2009-12-04 Bartosz Kosiorek <gang65@poczta.onet.pl>
Enable new mode switch for VM800 chipsets
* src/via_driver.c: (VIASetupDefaultOptions):
2009-11-21 Bartosz Kosiorek <gang65@poczta.onet.pl>
Add option to enable unaccelerated RandR rotation ("SWRandR").
The accelerated option "HWRandR" is currently not implemented.
* src/openchrome.man:
* src/via_driver.c: (VIAPreInit):
2009-11-20 Bartosz Kosiorek <gang65@poczta.onet.pl>
Enabled new mode switch for PM800 chipset,
to resolve many bugs with resolution detecting and changing
(eg. switching to console)
* src/via_driver.c: (VIASetupDefaultOptions), (VIAPreInit):
2009-11-07 Bartosz Kosiorek <gang65@poczta.onet.pl>
Add more comments to ViaSetSecondaryFIFO, add panel scale support for
CLE266 and KM400, fix bug with malloc.
* src/via_bandwidth.c: (ViaSetSecondaryFIFO):
* src/via_panel.c: (ViaPanelScale), (ViaPanelGetNativeDisplayMode):
2009-09-26 Bartosz Kosiorek <gang65@poczta.onet.pl>
Save/restore ECK Clock Synthesizer
* src/via_driver.c: (VIASave), (VIARestore):
* src/via_driver.h:
2009-09-21 Bartosz Kosiorek <gang65@poczta.onet.pl>
Fixed bug with panel autodetection using
ViaPanelGetNativeModeFromScratchPad function
* src/via_panel.c:
2009-09-21 Bartosz Kosiorek <gang65@poczta.onet.pl>
Fixed problem with backlite (Ticket #308)
* src/via_lvds.c: (ViaLVDSSoftwarePowerFirstSequence),
(ViaLVDSSoftwarePowerSecondSequence),
(ViaLVDSHardwarePowerFirstSequence),
(ViaLVDSHardwarePowerSecondSequence), (ViaLVDSPower):
2009-09-15 Bartosz Kosiorek <gang65@poczta.onet.pl>
TV-out VT1625 chip support
* src/via_bios.h:
* src/via_crtc.c: (ViaFirstCRTCSetMode):
* src/via_display.c: (ViaDisplayEnableDVO), (ViaDisplayDisableDVO),
(ViaDisplaySetStreamOnDVO):
* src/via_driver.c: (VIASetupDefaultOptions), (VIAPreInit):
* src/via_mode.c: (ViaTVSetMode), (ViaOutputsSelect), (ViaModeSet):
* src/via_vt162x.c: (ViaSetTVClockSource), (VT1622ModeI2C),
(VT1622ModeCrtc):
* src/via_vt162x.h:
2009-09-15 Bartosz Kosiorek <gang65@poczta.onet.pl>
Cursor support speedup
* src/via_cursor.c: (viaHWCursorInit), (viaCursorStore),
(viaCursorRestore), (viaShowCursor), (viaHideCursor),
(viaSetCursorPosition), (viaLoadCursorImage), (viaLoadCursorARGB):
2009-09-14 Bartosz Kosiorek <gang65@poczta.onet.pl>
Removed two old variables:
pBIOSInfo->panelX
pBIOSInfo->panelY
and replaced it by:
pBIOSInfo->Panel->NativeMode->Width
pBIOSInfo->Panel->NativeMode->Height
* src/via_bios.h:
* src/via_driver.c: (VIASetupDefaultOptions):
* src/via_mode.c: (ViaPanelGetIndex), (VIASetLCDMode),
(ViaModePrimaryLegacy), (ViaModeSecondaryLegacy):
* src/via_swov.c: (SetVideoWindow):
* src/via_video.c: (DecideOverlaySupport):
2009-09-10 Bartosz Kosiorek <gang65@poczta.onet.pl>
Fix bug with DFP Power Off (ticket #317)
* src/via_mode.c: (ViaDFPPower):
2009-08-20 Bartosz Kosiorek <gang65@poczta.onet.pl>
Fix bug with wrong panel size value (http://www.openchrome.org/trac/ticket/301)
This bug appeared only when UseLegacyModeSwitch was TRUE
* src/via_driver.c: (VIASetupDefaultOptions):
* src/via_video.c: (DecideOverlaySupport):
2009-08-17 Bartosz Kosiorek <gang65@poczta.onet.pl>
Added support of the VIA OpenBook
* src/via_id.c:
2009-08-08 Bartosz Kosiorek <gang65@poczta.onet.pl>
Added support for screen rotate upside-down and remove "magic numbers"
* src/via_driver.c:
* src/via_driver.h:
* src/via_shadow.c:
2009-07-28 Jon Nettleton <jon.nettleton@gmail.com>
Forgot to remove an old Dot Clock entry from the table.
* src/via_mode.h:
2009-07-28 Jon Nettleton <jon.nettleton@gmail.com>
XO 1.5 Panel patch contributed by Xavier Bachelot.
Fixup some of the Dotclock code and add working plls
for the XO 1.5
* src/via_bios.h:
* src/via_mode.c: (ViaSetDotclock), (ViaSetPrimaryDotclock),
(ViaSetSecondaryDotclock):
* src/via_mode.h:
* src/via_panel.c:
2009-03-21 Xavier Bachelot <xavier@bachelot.org>
* src/via_bios.h:
* src/via_display.c: (ViaDisplaySetStreamOnDFP):
* src/via_driver.c: (VIAPreInit), (VIASave), (VIARestore),
(VIADPMS):
* src/via_driver.h:
* src/via_mode.c: (ViaDFPDetect), (ViaOutputsDetect),
(ViaOutputsSelect), (ViaDFPPower), (ViaModeSet):
CX700 integrated TMDS (Patch courtesy Tim Chen from VIA).
2009-02-19 Gabriel Mansi <gabriel-dot-mansi-at-gmail-dot-com>
* src/via_driver.c: (VIAScreenInit), (VIAWriteMode):
Initialize CRTC before a mode switch. Fix bug #260.
2009-01-26 Jon Nettleton <jon.nettleton@gmail.com>
* src/via_accel.c: (viaInitXAA), (viaExaCheckComposite):
a couple of small compatibility fixes for XAA and EXA.
2009-01-17 Jon Nettleton <jon.nettleton@gmail.com>
* libxvmc/viaLowLevel.c: (viaDMAInitTimeStamp):
* src/via_accel.c: (viaDisableVQ):
* src/via_video.c: (viaVideoFillPixmap):
viaLowLevel.c and via_accel.c are fixes by
Bartosz Kosiorek. The via_video.c is a syntax
fix by Robert Bridge. Thanks for the bug fixes.
2009-01-17 Jon Nettleton <jon.nettleton@gmail.com>
* configure.ac:
* src/via_accel.c: (viaInitAccel), (viaExitAccel),
(viaFinishInitAccel):
* src/via_driver.c: (VIASetup), (VIASetupDefaultOptions),
(VIAPreInit), (VIAWriteMode), (VIAInitialize3DEngine):
* src/via_driver.h:
* src/via_memory.c: (viaExaFBSave), (VIAFreeLinear),
(viaOffScreenLinear), (VIAInitLinear):
* src/via_priv.h:
Patch submitted by Robert Bridge, originally started by
Xavier Bachelot, to remove old EXA support. We are moving
forward with the driver and are no longer supporting the
old EXA initializations.
2009-01-14 Jon Nettleton <jon.nettleton@gmail.com>
* src/via_accel.c: (viaDisableVQ), (viaInitialize2DEngine),
(viaAccelSync):
According to VIA's code the P4M900 should have the same
acceleration initialization registers as the K8M890. This
patch fixes this and gives about a 10x increase in the
video benchmark numbers.
2009-01-14 Jon Nettleton <jon.nettletno@gmail.com>
* src/via_accel.c: (viaFlushPCI):
remove the check for VIA_3D_ENG_BUSY in the idle loop.
This causes xorg to chew a bit more cpu with XAA enabled,
and it causes the entire X server to dump if EXA is enabled.
I should investigate further, but for now I will just remove
it.
2009-01-14 Jon Nettleton <jon.nettleton@gmail.com>
* src/via_accel.c: (viaFlushPCI):
Oops forgot to change registers for the vx800 chipset
when waiting for Idle in the engines. I also added
back in waiting for the 3d engine to be idle. I am
not sure if this was removed for a reason.
2009-01-07 Jon Nettleton <jon.nettleton@gmail.com>
* src/via_bios.h:
* src/via_crtc.c: (via_xf86crtc_resize), (ViaPreInitCRTCConfig):
* src/via_driver.c: (VIAPreInit):
Temporary hack to fix a bug with newer Xorg and DRI
that crashes if CRTC is not setup. I am adding this
because I am working on this part of the code anyways
so it won't get lost in bit rot.
2009-01-06 Jon Nettleton <jon.nettleton@gmail.com>
* src/via_mode.c: (ViaModesMonitorFixup), (ViaModesAttachHelper):
Make sure that the monitor settings will handle our
internal Modes set for TV out and LCD Panels.
2009-01-05 Jon Nettleton <jon.nettleton@gmail.com>
* src/via_cursor.c: (viaHWCursorInit), (viaLoadCursorImage):
A couple more tweaks to fix mono hardware cursor on older
chipsets.
2009-01-05 Jon Nettleton <jon.nettleton@gmail.com>
* src/via_cursor.c: (viaHWCursorInit), (viaUseHWCursorARGB),
(viaUseHWCursor):
* src/via_driver.c: (VIAScreenInit):
Missed a couple of changes for the chipset specific
cursors. Moved chipset detection to HWInit. Removed
some extraneous debugging.
2009-01-04 Gabriel Mansi <gabriel-dot-mansi-at-gmail-dot-com>
* src/via.h:
* src/via_accel.c: (viaInitialize2DEngine):
* src/via_crtc.c: (ViaFirstCRTCSetMode), (ViaSecondCRTCSetMode):
* src/via_driver.c: (ViaMMIOEnable), (ViaMMIODisable),
(VIAWriteMode):
* src/via_swov.c: (SetFIFO_V3), (Upd_Video):
* src/via_video.c: (DecideOverlaySupport):
Initial XVideo support for VX800.
2009-01-04 Jon Nettleton <jon.nettleton@gmail.com>
* src/via_cursor.c: (viaHWCursorInit), (viaCursorSetFB),
(viaUseHWCursorARGB), (viaUseHWCursor), (viaLoadCursorImage),
(viaSetCursorColors), (viaLoadCursorARGB):
* src/via_driver.h:
Okay I tried to cheat. We need to have some chipset
specific intelligence. I have added this back into
the initial test to check for HWCursor support.
2009-01-04 Jon Nettleton <jon.nettleton@gmail.com>
* src/via_accel.c: (viaInitXAA), (viaInitAccel):
* src/via_crtc.c: (ViaSecondCRTCSetStartingAddress):
* src/via_cursor.c: (viaHWCursorInit), (viaCursorSetFB),
(viaCursorStore), (viaCursorRestore), (viaShowCursor),
(viaHideCursor), (viaSetCursorPosition), (viaUseHWCursorARGB),
(viaUseHWCursor), (viaLoadCursorImage), (viaSetCursorColors),
(viaLoadCursorARGB):
* src/via_dga.c: (VIADGASetMode):
* src/via_driver.c: (VIAFreeRec), (VIASetupDefaultOptions),
(VIAPreInit), (VIAEnterVT), (VIALeaveVT), (VIAScreenInit),
(VIACloseScreen):
* src/via_driver.h:
* src/via_regs.h:
This is a take at reworking HWCursor. It works off the
ttm_branch and the work done in the randr_branch. It only needs
marginal testing as HWCursor was already very broken.
2009-01-04 Jon Nettleton <jon.nettleton@gmail.com>
* src/via_driver.c: (VIAMapMMIO), (VIAMapFB):
Moved pVia->FrameBufferBase setting to within
VIAMapFB where it was before being called without
being set.
2009-01-04 Jon Nettleton <jon.nettleton@gmail.com>
* src/via_driver.c: (VIAPreInit):
Patch submitted by Robert Bridge to clean up compiler
warnings. Should close trac ticket #243. Thanks
RobbieAB.
2008-12-29 Jon Nettleton <jon.nettleton@gmail.com>
* src/via_accel.c: (viaInitXAA):
* src/via_driver.c: (VIAMapMMIO), (VIAMapFB):
Finally tracked down the XAA libpciaccess bug. Also
Realized I had not enabled write-combining on the main
framebuffer. This should give a small speed boost.
2008-12-26 Jon Nettleton <jon.nettleton@gmail.com>
* src/via_mode.h:
Added support for all Modes in ViaPanelNativeModes to
ViaPanelModes. Modes used are from Via's released driver
in ViaSupportModes.
2008-11-04 Jon Nettleton <jon.nettleton@gmail.com>
* src/via_crtc.c: (ViaFirstCRTCSetMode), (ViaSecondCRTCSetMode):
Found needed register writes that are needed so VX800 doesn't
crash when CRT is not enabled. I also removed NEEDS FIXING
comments from them as they are obviously needed.
2008-11-03 Jon Nettleton <jon.nettleton@gmail.com>
* src/via_accel.c: (viaSubsequentMono8x8PatternFillRect),
(viaSetupForSolidLine), (viaSubsequentSolidHorVertLine),
(viaSetupForDashedLine), (viaAccelMarkSync):
A few more acceleration tweaks for the VX800.
This should fix all the XAA rendering issues.
This should also allow 800x480 Panels to work without
custom modelines.
2008-11-02 Jon Nettleton <jon.nettleton@gmail.com>
* src/via_accel.c: (viaFlushPCI), (viaDisableVQ),
(viaInitialize2DEngine), (viaAccelSync), (viaPitchHelper),
(viaAccelClippingHelper), (viaAccelSolidHelper),
(viaAccelTransparentHelper), (viaAccelCopyHelper),
(viaSetupForScreenToScreenCopy), (viaSubsequentScreenToScreenCopy),
(viaSetupForSolidFill), (viaSubsequentSolidFillRect),
(viaSetupForMono8x8PatternFill),
(viaSubsequentMono8x8PatternFillRect),
(viaSetupForColor8x8PatternFill),
(viaSubsequentColor8x8PatternFillRect),
(viaSetupForCPUToScreenColorExpandFill),
(viaSubsequentScanlineCPUToScreenColorExpandFill),
(viaSetupForImageWrite), (viaSubsequentImageWriteRect),
(viaSetupForSolidLine), (viaSubsequentSolidTwoPointLine),
(viaSubsequentSolidHorVertLine), (viaSetupForDashedLine),
(viaInitXAA), (viaAccelMarkSync), (viaExaPrepareSolid),
(viaExaSolid), (viaExaPrepareCopy), (viaExaCopy):
* src/via_bandwidth.c: (ViaSetPrimaryFIFO), (ViaSetSecondaryFIFO):
* src/via_cursor.c: (viaCursorARGBInit), (viaCursorARGBShow),
(viaCursorARGBHide), (viaCursorARGBSetPosition),
(viaCursorRecInit):
* src/via_driver.c: (VIASetupDefaultOptions), (VIAPreInit),
(VIALeaveVT), (VIACloseScreen):
* src/via_driver.h:
* src/via_id.c:
* src/via_id.h:
* src/via_mode.c: (ViaOutputsSelect), (VIAGetPanelSize),
(ViaPanelGetIndex):
* src/via_mode.h:
* src/via_panel.c: (ViaPanelPreInit):
* src/via_regs.h:
* src/via_swov.c: (VIAVidHWDiffInit), (viaCalculateVideoColor),
(viaSetColorSpace), (ViaInitVideoStatusFlag), (ViaSetVidCtl),
(SetFIFO_V3_64or32or32), (SetFIFO_V3_64or32or16):
* src/via_video.c: (viaInitVideo):
Initial support for the VX800 chipset. This is based off the
Basic chipset support work of Xavier Bachelot and the inital
Acceleration work done by Harald Welte. I verified the changes
based off the most recent VIA codebaase and fixed a couple of
rendering issues.
Still needs more work on EXA ( font rendering glitches ) and
XVideo.
2008-05-26 Gabriel Mansi <gabriel.mansi@gmail.com>
* src/Makefile.am:
* src/via_accel.c: (viaInitAccel):
* src/via_bios.h:
* src/via_crtc.c: (ViaCRTCSetGraphicsRegisters),
(ViaCRTCSetAttributeRegisters), (ViaCRTCInit),
(ViaFirstCRTCSetMode), (ViaFirstCRTCSetStartingAddress),
(ViaSecondCRTCSetStartingAddress),
(ViaSecondCRTCHorizontalQWCount), (ViaSecondCRTCHorizontalOffset),
(ViaSecondCRTCSetMode), (ViaFirstCRTCModeValid),
(ViaSecondCRTCModeValid), (ViaShadowCRTCSetMode):
* src/via_cursor.c: (viaCursorARGBInit), (viaCursorSetFB),
(viaCursorHWInit), (viaCursorHWShow), (viaCursorHWHide),
(viaCursorShow), (viaCursorHide), (viaCursorLoadImage),
(viaCursorHWSetPosition), (viaCursorSetPosition),
(viaCursorSetColors), (viaCursorStore), (viaCursorRestore),
(viaCursorARGBShow), (viaCursorARGBHide),
(viaCursorARGBSetPosition), (viaCursorARGBUse),
(viaCursorARGBLoad), (viaCursorRecInit), (viaCursorRecDestroy),
(viaCursorHWUse):
* src/via_dga.c: (VIADGASetMode):
* src/via_display.c: (ViaSecondDisplayChannelEnable),
(ViaSecondDisplayChannelDisable), (ViaDisplayInit),
(ViaDisplayEnableSimultaneous), (ViaDisplayDisableSimultaneous),
(ViaDisplayEnableCRT), (ViaDisplayDisableCRT),
(ViaDisplaySetStreamOnCRT):
* src/via_driver.c: (VIAGetRec), (VIAFreeRec),
(VIASetupDefaultOptions), (VIAPreInit), (VIAEnterVT), (VIALeaveVT),
(ViaGammaDisable), (VIASave), (VIARestore), (ViaMMIOEnable),
(ViaMMIODisable), (VIAMapMMIO), (VIAUnmapMem), (VIALoadPalette),
(VIAScreenInit), (VIAWriteMode), (VIACloseScreen),
(VIAAdjustFrame), (VIADPMS):
* src/via_driver.h:
* src/via_lvds.c: (ViaLVDSPowerFirstSequence),
(ViaLVDSPowerSecondSequence), (ViaLVDSDFPPower),
(ViaLVDSPowerChannel), (ViaLVDSPower):
* src/via_mode.c: (ViaPrintMode), (ViaOutputsSelect),
(VIAGetPanelSize), (ViaPanelGetIndex), (ViaModesAttach),
(ViaValidMode), (VIASetLCDMode), (ViaModePrimaryLegacy),
(ViaModeSecondaryLegacy), (ViaLCDPower), (ViaModeFirstCRTC),
(ViaModeSecondCRTC), (ViaModeSet):
* src/via_panel.c: (ViaPanelLookUpModeIndex),
(ViaPanelGetNativeModeFromOption),
(ViaPanelGetNativeModeFromScratchPad), (ViaPanelScaleDisable),
(ViaPanelScale), (ViaPanelGetNativeDisplayMode), (ViaPanelPreInit),
(ViaPanelCenterMode), (ViaPanelGetSizeFromEDID),
(ViaPanelGetSizeFromDDCv1), (ViaPanelGetSizeFromDDCv2):
* src/via_regs.h:
* src/via_swov.c: (Upd_Video):
* src/via_timing.c: (froundf), (timingGenerateMode),
(viaTimingCvt), (viaTimingGtf):
* src/via_timing.h:
* src/via_vbe.c: (ViaVbeGetActiveDevices), (ViaVbeSetMode),
(ViaVbeDoDPMS):
* src/via_video.c: (DecideOverlaySupport), (viaSaveVideo),
(viaRestoreVideo):
Merge panel code from randr branch (rev. 427 to 553)
Add native mode setting for P4M890, P4M900, K8M890 and CX700
Add LVDS power functions for P4M900 and CX700
Add ARGB hardware cursor support
2008-03-11 Xavier Bachelot <xavier-at-bachelot-dot-org>
* configure.ac:
* src/via.h:
* src/via_dri.c: (VIADRIIrqInit), (VIADRIScreenInit):
* src/via_driver.c: (via_pci_device), (via_host_bridge),
(viaPciDeviceVga), (if), (switch):
* src/via_driver.h:
* src/via_id.c: (ViaCheckCardId):
Merge changes from pciaccess branch.
2008-02-19 Benno Schulenberg <bensberg-at-justemail-dot-net>
* src/via_driver.c: (VIAPreInit):
Dealing with the NoAccel option before the Rotate option, since
the latter overrides the first.
2008-01-28 Benno Schulenberg <bensberg-at-justemail-dot-net>
* src/Makefile.am:
Recreating svnversion.h only when a source file has changed.
2008-01-18 Benno Schulenberg <bensberg-at-justemail-dot-net>
* src/via_dri.c: (VIAInitVisualConfigs):
Condensing an unneeded code duplication.
2008-01-11 Benno Schulenberg <bensberg-at-justemail-dot-net>
* src/openchrome.man:
* src/via_driver.c: (VIASetupDefaultOptions), (VIAPreInit):
Removing the redundant HWCursor option.
2008-01-10 Benno Schulenberg <bensberg-at-justemail-dot-net>
* src/openchrome.man: Tweaking the text, clarifying the effect
of an option, giving the default value.
2008-01-08 Benno Schulenberg <bensberg-at-justemail-dot-net>
* src/via_accel.c: (viaBitExpandHelper), (viaPixelARGB8888):
Moving a little more computation into the helper function.
2007-10-29 Benno Schulenberg <bensberg-at-justemail-dot-net>
* src/via_driver.c: (VIAPreInit):
Redoing the second change of June 10: Printing VideoRAM again
only when it's actually been probed; plus further tiny code
cleanups: comments, line wraps, whitespace.
2007-10-29 Benno Schulenberg <bensberg-at-justemail-dot-net>
* src/via_driver.c: (VIAPreInit):
Redoing the first change of June 10: Putting the chipset and
revision print statements together at the beginning; moving
the reading of the VideoRAM option to after X has parsed the
config file; plus a few small layout tweaks.
2007-10-05 Gabriel Mansi <gabriel-dot-mansi-at-gmail-dot-com>
* src/via_swov.c: (Upd_Video):
Enable colorkey on secondary.
2007-10-03 Gabriel Mansi <gabriel-dot-mansi-at-gmail-dot-com>
* src/via_driver.c: (VIAPreInit):
Fix memory detection for P4M900 and CX700.
2007-09-12 Gabriel Mansi <gabriel-dot-mansi-at-gmail-dot-com>
* src/via_bios.h:
* src/via_driver.c: (VIAWriteMode), (VIAAdjustFrame),
(VIASwitchMode):
* src/via_mode.c: (ViaModeSecondaryVGAFetchCount),
(ViaModeSecondaryVGAOffset), (ViaModeSecondaryVGA):
* src/via_swov.c: (Upd_Video):
* src/via_vbe.c: (ViaVbeSetPanelMode), (ViaVbeSetMode):
Fix offset when using vbe modes on secondary.
2007-09-11 Xavier Bachelot <xavier-at-bachelot-dot-org>
* libxvmc/viaLowLevel.c: (initXvMCLowLevel):
* libxvmc/viaLowLevel.h:
* libxvmc/viaLowLevelPro.c: (initXvMCLowLevel):
* src/via.h:
* src/via_accel.c: (viaInitExa):
* src/via_bandwidth.c: (ViaSetPrimaryFIFO):
* src/via_driver.c: (VIASetupDefaultOptions), (VIAPreInit),
(VIASave), (VIARestore), (VIALoadPalette):
* src/via_driver.h:
* src/via_id.c:
* src/via_mode.c: (ViaOutputsDetect):
* src/via_priv.h:
* src/via_swov.c: (Upd_Video):
* src/via_vbe.c: (ViaVbeInitInt10), (ViaVbeGetRefreshRateIndex),
(ViaVbeGetActiveDevices), (ViaVbeSetActiveDevices),
(ViaVbeSetPanelExpansion), (ViaVbeSetRefresh), (ViaVbeSetMode),
(ViaVbeSaveRestore), (ViaVbeModePreInit):
* src/via_xvmc.c: (ViaInitXVMC), (ViaXvMCCreateContext):
Sync with experimental_branch.
2007-09-07 Gabriel Mansi <gabriel-dot-mansi-at-gmail-dot-com>
* unichrome/via_id.c:
Added Gigabyte GA-VM900M ID, reported by Alexandre Penasso Teixeira.
2007-09-07 Gabiel Mansi <gabriel-dot-mansi-at-gmail-dot-com>
* libxvmc/viaLowLevel.c: (initXvMCLowLevel):
* libxvmc/viaLowLevel.h:
* libxvmc/viaLowLevelPro.c: (initXvMCLowLevel):
* unichrome/via.h:
* unichrome/via_driver.c: (VIASetupDefaultOptions), (VIAPreInit),
(VIASave), (VIARestore), (VIALoadPalette):
* unichrome/via_driver.h:
* unichrome/via_mode.c: (ViaOutputsDetect):
* unichrome/via_swov.c: (Upd_Video):
* unichrome/via_vbe.c: (ViaVbeInitInt10),
(ViaVbeGetRefreshRateIndex), (ViaVbeGetActiveDevices),
(ViaVbeSetActiveDevices), (ViaVbeSetPanelExpansion),
(ViaVbeSetRefresh), (ViaVbeSetMode), (ViaVbeSaveRestore),
(ViaVbeModePreInit):
* unichrome/via_xvmc.c: (ViaInitXVMC), (ViaXvMCCreateContext):
Merged vn896_branch from revision 361 to 391.
2007-09-07 Gabriel Mansi <gabriel-dot-mansi-at-gmail-dot-com>
* unichrome/via_driver.c: (VIAPreInit):
* unichrome/via_mode.c: (ViaOutputsDetect):
* unichrome/via_vbe.c: (ViaVbeSetActiveDevices):
Disabled TV detection for P4M900.
2007-09-07 Gabriel Mansi <gabriel-dot-mansi-at-gmail-dot-com>
* unichrome/via_driver.h:
* unichrome/via_vbe.c: (ViaVbeInitInt10),
(ViaVbeGetRefreshRateIndex), (ViaVbeGetActiveDevices),
(ViaVbeSetActiveDevices), (ViaVbeSetPanelExpansion),
(ViaVbeSetRefresh), (ViaVbeSetMode), (ViaVbeModePreInit):
VBE code cleanup.
2007-09-02 Gabriel Mansi <gabriel-dot-mansi-at-gmail-dot-com>
* unichrome/via_driver.c: (VIASetupDefaultOptions), (VIASave),
(VIARestore):
* unichrome/via_vbe.c: (ViaVbeSetMode):
Disabled XvDMA for P4M900.
Modified VBE functions calling order.
2007-08-18 Gabriel Mansi <gabriel-dot-mansi-at-gmail-dot-com>
* unichrome/via_driver.c: (VIASetupDefaultOptions):
* unichrome/via_priv.h:
* unichrome/via_swov.c: (Upd_Video):
Xvideo code cleanup; added maximum resolution with
interpolation for all chipsets.
2007-08-03 Xavier Bachelot <xavier-at-bachelot-dot-org>
* src/via_bios.h:
* src/via_mode.h:
* src/via_video.c: (DecideOverlaySupport):
Add basic support for DDR667.
2007-07-31 Jon Nettleton <jon-dot-nettleton-at-gmail-dot-com>
* src/via_xvmc.c: (ViaInitXVMC):
Fixed the xvmc name we register. Was still looking for
the old via xvmc library names.
2007-07-04 Xavier Bachelot <xavier-at-bachelot-dot-org>
* configure.ac:
* src/via_driver.c: (VIAPreInit):
* src/via_driver.h:
Bump driver version to 0.2.900.
Do not log detected chipset 2 times.
Remove duplicate changelog entry.
2007-06-20 Gabriel Mansi <gabriel-dot-mansi-at-gmail-dot-com>
* unichrome/via_bandwidth.c: (ViaSetPrimaryFIFO)
FIFO settings for CX700.
2007-06-14 Jon Nettleton <jon-dot-nettleton-at-gmail-dot-com>
* configure.ac:
* libxvmc/viaLowLevel.h:
* src/via.h:
* src/via_accel.c: (viaFlushPCI), (viaInitPCIe), (viaEnableAgpVQ),
(viaEnablePCIeVQ), (viaDisableVQ), (viaInitialize2DEngine),
(viaAccelSync), (viaInitXAA), (viaInitExa), (viaInitAccel):
* src/via_bandwidth.c: (ViaSetPrimaryFIFO), (ViaSetSecondaryFIFO):
* src/via_dri.c: (VIADRIRingBufferInit):
* src/via_driver.c: (VIASetupDefaultOptions), (VIAPreInit),
(VIALeaveVT), (VIASave), (VIAScreenInit), (VIACloseScreen):
* src/via_driver.h:
* src/via_id.c: (ViaDoubleCheckCLE266Revision), (ViaCheckCardId):
* src/via_id.h:
* src/via_mode.c: (ViaGetMemoryBandwidth), (ViaModePrimary):
* src/via_mode.h:
* src/via_regs.h:
* src/via_swov.c: (viaWaitHQVFlip), (viaWaitHQVDone),
(VIAVidHWDiffInit), (viaCalculateVideoColor), (viaSetColorSpace),
(ViaInitVideoStatusFlag), (ViaSetVidCtl), (AddHQVSurface),
(SetFIFO_V3), (SetFIFO_V3_64or32or32), (SetFIFO_V3_64or32or16),
(SetColorKey), (SetChromaKey), (SetHQVFetch), (Upd_Video),
(VIAVidUpdateOverlay), (ViaOverlayHide):
* src/via_swov.h:
* src/via_vbe.c: (ViaVbeSetMode):
* src/via_video.c: (DecideOverlaySupport), (viaInitVideo), (Flip),
(viaDmaBlitImage), (viaPutImage):
* src/via_video.h:
* src/via_xvmc.c: (ViaInitXVMC):
Pulled in Changes from the experimental_branch including
- Partial support for VIA_K8M890, VIA_P4M900, VIA_CX700,
VIA_P4M890
- Added the VideoEngine option to help clean up code
- Ability to enable debugging with the configure script
2007-06-12 Gabriel Mansi <gabriel-dot-mansi-at-gmail-dot-com>
* unichrome/via.h:
* unichrome/via_dri.c: (VIADRIRingBufferInit):
* unichrome/via_driver.c:
* unichrome/via_id.c:
* unichrome/via_id.h:
* unichrome/via_swov.c: (viaCalculateVideoColor),
(viaSetColorSpace), (ViaInitVideoStatusFlag), (ViaSetVidCtl),
(SetFIFO_V3), (SetFIFO_V3_64or32or32), (SetFIFO_V3_64or32or16),
(Upd_Video):
* unichrome/via_vbe.c: (ViaVbeSetMode):
* unichrome/via_video.c: (DecideOverlaySupport):
Renamed VT3157 to VT3324.
Added check for null data block in vbe.
Remove version check for older version of Xorg.
2007-06-12 Gabriel Mansi <gabriel-dot-mansi-at-gmail-dot-com>
* unichrome/via.h:
Added missing includes.
2007-06-10 Benno Schulenberg <bensberg-at-justemail-dot-net>
* unichrome/via_driver.c: (VIAPreInit):
Print VideoRAM again only when it's actually been probed.
Further tiny code cleanups: comments, line wraps, whitespace.
2007-06-10 Benno Schulenberg <bensberg-at-justemail-dot-net>
* unichrome/via.man:
* unichrome/via_driver.c: (VIAPreInit):
Move reading of the VideoRAM option to after X has parsed the
config file. Also put the chipset and revision print statements
together and at the beginning. Plus several small layout tweaks.
2007-06-07 Jon Nettleton <jon-dot-nettleton-at-gmail-dot-com>
* Makefile.am:
* configure.ac:
* libxvmc/Makefile.am:
* man/Makefile.am:
* src/Makefile.am:
* src/via_driver.c:
* src/via_driver.h:
Renaming driver to openchrome_drv.so and XvMC library to
libchromeXvMC.so.
2007-06-06 Jon Nettleton <jon-dot-nettleton-at-gmail-dot-com>
* unichrome/via_driver.c:
Removed the chipset option definitions that aren't in
trunk yet.
2007-06-06 Jon Nettleton <jon-dot-nettleton-at-gmail-dot-com>
* unichrome/via_driver.c:
* unichrome/via_driver.h:
* unichrome/via.man:
Added per-chipset option defaults. This should allow the
driver to work out of the box with almost all hardware and
xorg.conf configurations.
2007-05-23 Jon Nettleton <jon-dot-nettleton-at-gmail-dot-com>
* unichrome/via.h:
Remove version check for older version of Xorg. This is
really not necessary anymore, and was causing some breakage
with xorg-server's new version numbering.
2007-05-18 Xavier Bachelot <xavier-at-bachelot-dot-org>
* libxvmc/Makefile.am:
* libxvmc/viaXvMC.c: (XvMCCreateContext):
Backport 2 fixes from Xorg tree.
http://gitweb.freedesktop.org/?p=xorg/driver/xf86-video-via.git;a=commit;h=362e03a38682bfcf366242c53444fa36b6052794
http://gitweb.freedesktop.org/?p=xorg/driver/xf86-video-via.git;a=commit;h=3cd7dac2b5a1c4bfb66bd1b67904d72dc08cbd0e
2007-05-15 Gabriel Mansi <gabriel-dot-mansi-at-gmail-dot-com>
* unichrome/via.h:
* unichrome/via_driver.c: (VIASetupDefaultOptions):
* unichrome/via_driver.h:
* unichrome/via_swov.c: (viaWaitHQVFlip), (SetColorKey),
(SetChromaKey), (Upd_Video):
* unichrome/via_video.c: (Flip), (viaDmaBlitImage), (viaPutImage):
Xv code cleanup.
2007-05-10 Jon Nettleton <jon-dot-nettleton-at-gmail-dot-com>
* unichrome/via_driver.c:
* unichrome/via_driver.h:
* unichrome/via.man:
Added per chipset option defaults. This should allow the
driver to install and work out of the box with almost all
hardware and xorg.conf configurations.
2007-05-07 Gabriel Mansi <gabriel-dot-mansi-at-gmail-dot-com>
* libxvmc/viaLowLevel.h:
* unichrome/via_accel.c: (viaFlushPCI), (viaDisableVQ),
(viaAccelSync):
* unichrome/via_bandwidth.c: (ViaSetPrimaryFIFO),
(ViaSetSecondaryFIFO):
* unichrome/via_dri.c: (VIADRIRingBufferInit):
* unichrome/via_driver.c: (VIAPreInit):
* unichrome/via_id.c:
* unichrome/via_swov.c: (viaWaitHQVFlip), (VIAVidHWDiffInit),
(viaSetColorSpace), (ViaInitVideoStatusFlag), (ViaSetVidCtl),
(SetFIFO_V3_64or32or32), (SetFIFO_V3_64or32or16), (SetColorKey),
(SetChromaKey), (Upd_Video):
* unichrome/via_video.c: (DecideOverlaySupport), (Flip),
(viaDmaBlitImage), (viaPutImage):
Fix Xv for P4M890.
2007-04-30 Gabriel Mansi <gabriel-dot-mansi-at-gmail-dot-com>
* unichrome/via_swov.c: (Upd_Video):
* unichrome/via_xvmc.c: (ViaInitXVMC):
Disable interpolation when the source width is greater than 800,
otherwise the picture wraps around the screen (bug #109).
Fix a typo in ViaInitXVMC (bug #111).
2007-04-14 Benno Schulenberg <bensberg-at-justemail-dot-net>
* unichrome/via_id.c: (ViaDoubleCheckCLE266Revision),
(ViaCheckCardId):
VN -> VM, message tweaks, and trailing white space fixes.
2007-03-28 Gabriel Mansi <gabriel-dot-mansi-at-gmail-dot-com>
* unichrome/via_bandwidth.c: (ViaSetPrimaryFIFO):
* unichrome/via_swov.c: (viaWaitHQVFlip), (VIAVidHWDiffInit),
(viaSetColorSpace), (ViaInitVideoStatusFlag), (ViaSetVidCtl),
(SetFIFO_V3_64or32or32), (SetFIFO_V3_64or32or16), (SetColorKey),
(SetChromaKey), (Upd_Video):
* unichrome/via_video.c: (Flip), (viaDmaBlitImage), (viaPutImage):
Fix Xv for CX700.
2007-03-20 Xavier Bachelot <xavier-at-bachelot-dot-org>
* unichrome/via_bandwidth.c: (ViaSetPrimaryFIFO),
(ViaSetSecondaryFIFO):
* unichrome/via_dri.c: (VIADRIRingBufferInit):
* unichrome/via_driver.c:
* unichrome/via_id.c:
* unichrome/via_id.h:
* unichrome/via_mode.c: (ViaGetMemoryBandwidth), (ViaModePrimary):
* unichrome/via_mode.h:
* unichrome/via_swov.c: (VIAVidHWDiffInit),
(viaCalculateVideoColor), (viaSetColorSpace),
(ViaInitVideoStatusFlag), (SetFIFO_V3), (SetFIFO_V3_64or32or32):
* unichrome/via_video.c: (viaInitVideo):
* unichrome/via_xvmc.c: (ViaInitXVMC):
Initial support for P4M890.
2007-03-09 Gabriel Mansi <gabriel-dot-mansi-at-gmail-dot-com>
* unichrome/via_accel.c: (viaFlushPCI), (viaAccelSync),
(viaInitXAA):
* unichrome/via_bandwidth.c: (ViaSetPrimaryFIFO),
(ViaSetSecondaryFIFO):
* unichrome/via_driver.c: (VIAPreInit), (VIALeaveVT),
(VIACloseScreen):
* unichrome/via_id.c:
* unichrome/via_id.h:
* unichrome/via_mode.c: (ViaGetMemoryBandwidth), (ViaModePrimary):
* unichrome/via_mode.h:
* unichrome/via_swov.c: (viaWaitHQVFlip), (VIAVidHWDiffInit),
(viaCalculateVideoColor), (viaSetColorSpace),
(ViaInitVideoStatusFlag), (ViaSetVidCtl), (SetFIFO_V3_64or32or32),
(SetFIFO_V3_64or32or16), (SetColorKey), (SetChromaKey),
(Upd_Video):
* unichrome/via_video.c: (DecideOverlaySupport), (viaInitVideo),
(Flip), (viaDmaBlitImage), (viaPutImage):
* unichrome/via_xvmc.c: (ViaInitXVMC):
Initial support for P4M900.
2007-03-08 Xavier Bachelot <xavier-at-bachelot-dot-org>
* unichrome/via_bandwidth.c: (ViaSetPrimaryFIFO),
(ViaSetSecondaryFIFO):
* unichrome/via_dri.c: (VIADRIRingBufferInit):
* unichrome/via_driver.c:
* unichrome/via_id.c:
* unichrome/via_id.h:
* unichrome/via_mode.c: (ViaGetMemoryBandwidth), (ViaModePrimary):
* unichrome/via_mode.h:
* unichrome/via_swov.c: (VIAVidHWDiffInit),
(viaCalculateVideoColor), (viaSetColorSpace),
(ViaInitVideoStatusFlag), (SetFIFO_V3), (SetFIFO_V3_64or32or32):
* unichrome/via_video.c: (DecideOverlaySupport), (viaInitVideo):
* unichrome/via_xvmc.c: (ViaInitXVMC):
Initial support for CX700.
2007-03-04 Benno Schulenberg <bensberg-at-justemail-dot-net>
* unichrome/via.man:
Textual tweaks and options ordered alphabetically.
2007-02-16 Benno Schulenberg <bensberg-at-justemail-dot-net>
* unichrome/via_accel.c: (viaSetClippingRectangle),
(viaAccelSolidHelper), (viaAccelCopyHelper),
(viaSubsequentMono8x8PatternFillRect),
(viaSubsequentColor8x8PatternFillRect),
(viaSubsequentSolidTwoPointLine), (viaSubsequentSolidHorVertLine):
Limit x value to sixteen bits. Fixes a rendering glitch reported
by Marg Huijgen <mark-dot-sf-dot-net-at-huijgen-dot-tk>; solution
found by Thomas.
2007-02-10 Thomas Hellström <thomas-at-tungstengraphics-dot-com>
* unichrome/via_accel.c: (viaFlushDRIEnabled),
(viaSubsequentSolidTwoPointLine):
Fix line-clipping bug reported by Manuel Bilderbeek
<manuel-dot-bilderbeek-at-oce-dot-com>.
2007-02-06 Benno Schulenberg <bensberg-at-justemail-dot-net>
* unichrome/via_driver.c:
Try CR39 when CR34 says zero during memory detection on
the CLE266. Idea from Luc Verhaegen, patch by Mark Huijgen
<mark-dot-sf-dot-net-at-huijgen-dot-tk>.
2007-01-20 Benno Schulenberg <bensberg-at-justemail-dot-net>
* ChangeLog
Switch encoding of ChangeLog file to Unicode.
Whitespace and format adjustments.
2007-01-20 Benno Schulenberg <bensberg-at-justemail-dot-net>
* unichrome/via_driver.c: (VIALoadRgbLut), (VIALoadPalette),
(VIAScreenInit):
Add ability to change gamma setting. Original patch by
Gonzalo A. de la Vega <gadelavega-at-gmail-dot-com>,
adapted from Luc Verhaegen's implementation.
2006-12-28 Thomas Hellström <thomas-at-tungstengraphics-dot-com>
* unichrome/via_accel.c: (viaExaTexUploadToScreen):
Work around an obscure hardware limitation when texturing from
AGP.
2006-12-28 Thomas Hellström <thomas-at-tungstengraphics-dot-com>
* unichrome/via_dri.c: (VIASetAgpMode):
Fix AGP mode setting.
2006-12-28 Thomas Hellström <thomas-at-tungstengraphics-dot-com>
Use non-power-of-two textures for EXA whenever possible.
2006-12-18 Jon Nettleton <jon-dot-nettleton-at-gmail-dot-com>
* unichrome/via_xvmc.c:
Thomas needs a good drm test which means we need XvMC. I
have re-enabled it for testing.
2006-12-15 Jon Nettleton <jon-dot-nettleton-at-gmail-dot-com>
* unichrome/via.h:
* unichrome/via_swov.c:
This is my cleanup commit of the excellent patch that
Gabriel Mansi <gabriel-dot-mansi-at-gmail-dot-com> provided,
which finishes the XVideo work on the VT3336 chipset. Nice
work Gabriel.
2006-12-13 Jon Nettleton <jon-dot-nettleton-at-gmail-dot-com>
* unichrome/via_video.c:
Don't use HQV Offset for HQV_CONTROL.
2006-12-11 Jon Nettleton <jon-dot-nettleton-at-gmail-dot-com>
* unichrome/via_video.c:
* unichrome/via_swov.c:
Looks like the VT3336 has the same offset as the VT3259
for the hardware overlay.
2006-12-10 Jon Nettleton <jon-dot-nettleton-at-gmail-dot-com>
* unichrome/via_memcpy.c:
* unichrome/via_swov.c:
Removed the previous changes; I am not ready to rewrite
the assembly code for x86_64.
The ColorSpace registers were being set up incorrectly
for VT3336.
2006-12-09 Jon Nettleton <jon-dot-nettleton-at-gmail-dot-com>
* unichrome/via_memcpy.c:
The ifdef's in memcpy.c only support i386 arch; should
also support i86_64.
2006-12-09 Jon Nettleton <jon-dot-nettleton-at-gmail-dot-com>
* unichrome/via_xvmc.c:
K8M890 does not support XvMC like the other chipsets;
disable it for now.
2006-12-09 Jon Nettleton <jon-dot-nettleton-at-gmail-dot-com>
* unichrome/via_video.h:
* unichrome/via_swov.h:
* unichrome/via_swov.c:
Made K8M890 XV FIFO settings match K8M800.
Enabled XV_DEBUG.
2006-12-07 Jon Nettleton <jon-dot-nettleton-at-gmail-dot-com>
* unichrome/via_mode.h:
Updated the BandwidthTable for K8M890 chipset.
2006-12-07 Thomas Hellström <thomas-at-tungstengraphics-dot-com>
* unichrome/via_driver.c: (VIAEnterVT):
Initialize the AGP ring buffer before command submission.
2006-12-07 Thomas Hellström <thomas-at-tungstengraphics-dot-com>
* unichrome/via_driver.c: (VIAEnterVT):
Blank and clear screen when entering VT.
2006-12-07 Thomas Hellström <thomas-at-tungstengraphics-dot-com>
* unichrome/via_driver.c: (VIAEnterVT):
Blank and clear screen when entering VT.
2006-12-07 Thomas Hellström <thomas-at-tungstengraphics-dot-com>
* unichrome/via_accel.c: (viaAccelDMADownload),
(viaExaDownloadFromScreen), (viaInitExa):
Indent via_accel.c.
2006-12-07 Thomas Hellström <thomas-at-tungstengraphics-dot-com>
* unichrome/via_accel.c: (viaAccelDMADownload):
Fix download from screen which was reusing kernel argument
data that was altered by the kernel.
2006-12-07 Thomas Hellström <thomas-at-tungstengraphics-dot-com>
* unichrome/via_accel.c: (viaAccelDMADownload),
(viaExaDownloadFromScreen):
Reimplement download from screen to something more easily
debuggable.
2006-12-06 Jon Nettleton <jon-dot-nettleton-at-gmail-dot-com>
* unichrome/via_driver.c:
Fixed some of the boolean option changes that are
inverses such as DisableIRQ.
2006-12-06 Thomas Hellström <thomas-at-tungstengraphics-dot-com>
* unichrome/via_memory.c: (viaExaFBSave), (viaOffScreenLinear):
Cause a "FatalError" if EXA is used with an un-patched Xserver,
and an illegal save locked memory action is requested.
2006-12-06 Thomas Hellström <thomas-at-tungstengraphics-dot-com>
* unichrome/via_driver.c: (VIAInitialize3DEngine):
Proper register names in VIAInitialize3DEngine.
2006-12-06 Thomas Hellström <thomas-at-tungstengraphics-dot-com>
* unichrome/via_driver.c: (VIALeaveVT), (VIAWriteMode),
(VIACloseScreen):
* unichrome/via_vbe.c: (ViaVbeSetMode):
Now that we initialize the 3D engine every time we write
mode, make sure this is done before setting up VQ and AGP
so that those settings persist. The GPU will be unstable
otherwise.
Also fix some code comments.
2006-12-06 Thomas Hellström <thomas-at-tungstengraphics-dot-com>
* unichrome/via_driver.c: (VIALeaveVT):
Force 3D context upload after a VT switch.
2006-12-06 Thomas Hellström <thomas-at-tungstengraphics-dot-com>
* unichrome/via_dri.c: (viaDRIFBMemcpy), (viaDRIOffscreenSave):
Fix up previous commit for > 16MB sizes.
2006-12-06 Thomas Hellström <thomas-at-tungstengraphics-dot-com>
* unichrome/via_dri.c: (viaDRIOffscreenSave),
(viaDRIOffscreenRestore):
* unichrome/via_driver.c: (VIAEnterVT), (VIALeaveVT):
* unichrome/via_driver.h:
Back up DRI offscreen memory before leaving VT and restore it
when entering VT. It may be overwritten in between.
Use PCI DMA blit for this if available.
2006-12-05 Thomas Hellström <thomas-at-tungstengraphics-dot-com>
* unichrome/via_driver.c: (VIAWriteMode), (VIAInitialize3DEngine):
* unichrome/via_vbe.c: (ViaVbeSetMode):
With EXA + 3D we need to reinitialize the 3D engine after
a mode switch (possibly the soft reset + 2D engine
initialization). Otherwise the GPU will hang.
2006-12-05 Thomas Hellström <thomas-at-tungstengraphics-dot-com>
* unichrome/via_driver.c: (VIAPreInit):
Fix the dma2d option setting.
2006-12-05 Thomas Hellström <thomas-at-tungstengraphics-dot-com>
* unichrome/via_bandwidth.c: (ViaSetPrimaryFIFO):
Use a more aggressive fifo setting for CLE266 Cx.
Boosts performance.
Please report problems with this.
2006-12-05 Jon Nettleton <jon-dot-nettleton-at-gmail-dot-com>
* unichrome/via_video.c:
* unichrome/via_swov.c:
Added missing hardware overlay support for VT3336.
2006-12-04 Jon Nettleton <jon-dot-nettleton-at-gmail-dot-com>
* unichrome/via_driver.c:
Added patch to fix the K8M890 hang on VT switch; submitted
by Gabriel Mansi <gabriel-dot-mansi-at-gmail-dot-com>.
2006-12-01 Thomas Hellström <thomas-at-tungstengraphics-dot-com>
* unichrome/via.man:
* unichrome/via_bios.h:
* unichrome/via_dri.c: (VIADRIScreenInit):
* unichrome/via_driver.c: (VIAPreInit):
* unichrome/via_driver.h:
Remove unused options.
First pass cleaning up option handling.
Make most boolean options work as expected;
(that is, accepting both a "true" and a "false" argument).
2006-11-30 Jon Nettleton <jon-dot-nettleton-at-gmail-dot-com>
* unichrome/via_mode.c:
* unichrome/via_accel.c:
Added patch to fix some of the K8M890 logic; submitted by
Gabriel Mansi <gabriel-dot-mansi-at-gmail-dot-com>.
2006-11-28 Jon Nettleton <jon-dot-nettleton-at-gmail-dot-com>
* unichrome/via_bandwidth.c:
* unichrome/via_accel.c:
Patched via_bandwidth.c for dumb mistake I made. Thanks, Gabriel
Mansi, for the second set of eyes. Move HWCursor memory allocation
even further in the driver initialization.
2006-11-25 Jon Nettleton <jon-dot-nettleton-at-gmail-dot-com>
* unichrome/via_id.c:
* unichrome/via_accel.c:
Added additional IDs for the K8M890 cards.
Changed memory allocation order for the HW cursor and 2D engine.
2006-10-29 Thomas Hellström <thomas-at-tungstengraphics-dot-com>
* libxvmc/viaLowLevel.c:
* libxvmc/viaLowLevelPro.c:
* unichrome/via_regs.h:
Via managed to mix up the 2D / 3D engine busy status bits.
Funny we never noticed this before.
2006-10-24 Thomas Hellström <thomas-at-tungstengraphics-dot-com>
* unichrome/via_xvmc.c: (ViaXvMCCreateContext):
Remove VT3314 (CN700, VM800 etc.) mpeg2 AGP DMA support.
2006-10-11 Xavier Bachelot <xb_ml-at-kelkoo-dot-net>
reviewed by: Ivor
* unichrome/via_bios.h:
* unichrome/via_id.c:
* unichrome/via_mode.h:
* unichrome/via_video.c: (DecideOverlaySupport):
DDR533 memory support and small cleanup.
Fujitsu/Siemens Amilo L7320 pci id (reported by DeNayGo).
2006-10-03 Thomas Hellström <thomas-at-tungstengraphics-dot-com>
* unichrome/via_video.c: (viaPutImage):
Removed some stray debug messages.
2006-10-02 Benno Schulenberg <bensberg-at-justemail-dot-net>
* unichrome/via_driver.c: (VIASave), (VIARestore):
Moved a line to after declarations for picky gcc-2.96,
tweaked comments and debugging lines.
2006-09-27 Thomas Hellström <thomas-at-tungstengraphics-dot-com>
* unichrome/via_memory.c:
Don't include drm headers in non-dri build.
2006-09-25 Benno Schulenberg <bensberg-at-justemail-dot-net>
* unichrome/Makefile.am:
Mention the overall revision number in the log, not just
the driver code revision.
2006-09-24 Thomas Hellström <thomas-at-tungstengraphics-dot-com>
* libxvmc/viaLowLevel.c: (initXvMCLowLevel):
* libxvmc/viaLowLevelPro.c: (initXvMCLowLevel):
* libxvmc/viaXvMC.c: (XvMCSetAttribute):
Fix a mutex unlocking issue (Luc Verhaegen).
Have the lowlevel drivers check for the correct chipid.
2006-09-24 Michal Ludvig <michal-at-logix-dot-cz>
* man/Makefile.am:
Fix man-page symlink for out-of-tree build (trac ticket #50).
2006-09-22 Jon Nettleton <jon-dot-nettleton-at-gmail-dot-com>
* unichrome/via_drmclient.h:
Include stdint.h instead of re-typedefining CARD32.
2006-09-22 Jon Nettleton <jon-dot-nettleton-at-gmail-dot-com>
* src/via_id.h
* src/via_video.c
* src/via_mode.c
* src/via_mode.h
* src/via_driver.c
* src/via_swov.c
* src/via_bandwidth.c
* src/via_accel.c
* src/via_vbe.c
* src/via_xvmc.c
* src/via_id.c
Initial code for support of VT3336 cards.
2006-09-17 Benno Schulenberg <bensberg-at-justemail-dot-net>
* unichrome/via_driver.c: (VIASave):
Put in a few extra debugging lines.
2006-09-05 Benno Schulenberg <bensberg-at-justemail-dot-net>
* unichrome/via_swov.c:
Comment and whitespace tweaks.
2006-09-05 Benno Schulenberg <bensberg-at-justemail-dot-net>
* unichrome/via_id.c:
Add an ID (trac ticket #67), note identical numbers, remove
a duplicate, remove obsolete unknown, remove trailing tab.
2006-08-27 Benno Schulenberg <bensberg-at-justemail-dot-net>
* unichrome/via_id.c:
Textual tweaks.
2006-08-27 Benno Schulenberg <bensberg-at-justemail-dot-net>
* unichrome/via_dri.c: (VIADRIAgpInit):
Rearrange declarations to fix compilation problem with gcc-2.96,
reported by Nikolai V. Ivanyushin <nvi-at-sven-dot-ru>.
2006-08-22 Benno Schulenberg <bensberg-at-justemail-dot-net>
* unichrome/via_accel.c:
Whitespace adjustment; getting back close to Xorg.
Comment tweaks and #endif clarifications.
2006-08-19 Benno Schulenberg <bensberg-at-justemail-dot-net>
* unichrome/via_swov.c: (viaWaitVideoCommandFire),
(SetVideoWindow):
Comment tweaks.
2006-08-18 Benno Schulenberg <bensberg-at-justemail-dot-net>
* unichrome/via_driver.h:
Move definition of AGP_SIZE out of #ifdef VIA_HAVE_EXA block,
to make the driver compile again for Xorg-6.8.2.
2006-08-10 Thomas Hellström <thomas-at-tungstengraphics-dot-com>
* unichrome/via_mode.c: (ViaModeDotClockTranslate):
* unichrome/via_video.c: (viaPaintColorkey):
Re-enable Ivor's Dotclock computation.
Enable accelerated redirected colorkey painting for Xv.
2006-07-28 Thomas Hellström <thomas-at-tungstengraphics-dot-com>
* unichrome/via_driver.c: (VIAPreInit):
...And get default AGP size correct as well.
2006-07-28 Thomas Hellström <thomas-at-tungstengraphics-dot-com>
* unichrome/via_accel.c: (viaIsAGP):
* unichrome/via_dri.c: (VIADRIAgpInit):
Fix false AGP detection and make sure we have the correct AGP
size.
2006-07-27 Thomas Hellström <thomas-at-tungstengraphics-dot-com>
* unichrome/via.man:
* unichrome/via_accel.c: (viaExaUploadToScratch),
(viaExaCheckComposite), (viaIsAGP), (viaInitExa), (viaInitAccel):
* unichrome/via_ch7xxx.c: (ViaCH7xxxInit):
* unichrome/via_dri.c: (VIADRIAgpInit):
* unichrome/via_driver.c: (VIAPreInit), (VIAEnterVT), (VIALeaveVT),
(VIAScreenInit), (VIACloseScreen):
* unichrome/via_driver.h:
Changes donated by "PConRails, LLC" and "Tungsten Graphics, Inc.":
EXA and memory management optimizations, Manpage update.
New options:
ExaScratchSize
MaxDRIMem
AGPMem
VBESaveRestore
2006-07-27 Thomas Hellström <thomas-at-tungstengraphics-dot-com>
* unichrome/via.h:
* unichrome/via_accel.c:
* unichrome/via_dri.h:
* unichrome/via_driver.h:
* unichrome/via_mode.c: (ViaModeDotClockTranslate):
* unichrome/via_swov.c: (ViaOverlayHide):
* unichrome/via_video.c: (DecideOverlaySupport),
(viaVideoFillPixmap), (viaPaintColorkey), (viaReputImage),
(viaPutImage):
Changes donated by "PConRails, LLC" and "Tungsten Graphics, Inc.":
Video optimizations:
Paint colorkey correctly under composite manager.
Make sure the driver compiles with Xorg git.
Don't reset the primary display FIFO after XV. It's never touched
by the Xv code, and slows things down in VBE mode.
2006-07-11 Thomas Hellström <thomas-at-tungstengraphics-dot-com>
* libxvmc/viaXvMC.c:
* unichrome/via_dri.c:
Change DRM major compatibility from 3.x.x to 2.x.x.
2006-07-10 Thomas Hellström <thomas-at-tungstengraphics-dot-com>
* libxvmc/driDrawable.c:
* libxvmc/viaLowLevel.c:
* libxvmc/viaLowLevelPro.c: (finish_header_agp),
(computeHQVScaleAndFilter):
* libxvmc/viaXvMC.c: (yOffs), (vOffs), (uOffs), (defaultQMatrices),
(releaseDecoder), (grabDecoder), (setupAttribDesc),
(releaseAttribDesc), (releaseContextResources),
(XvMCCreateContext), (XvMCDestroyContext), (XvMCCreateSurface),
(XvMCDestroySurface), (XvMCPutSlice2), (XvMCPutSlice),
(updateXVOverlay), (XvMCPutSurface), (debugControl),
(XvMCBeginSurface), (XvMCSyncSurface), (XvMCLoadQMatrix),
(XvMCRenderSurface), (XvMCCreateBlocks), (XvMCDestroyBlocks),
(XvMCCreateMacroBlocks), (XvMCDestroyMacroBlocks),
(XvMCCreateSubpicture), (XvMCSetSubpicturePalette), (findOverlap),
(XvMCClearSubpicture), (XvMCCompositeSubpicture),
(XvMCBlendSubpicture), (XvMCBlendSubpicture2),
(XvMCSyncSubpicture), (XvMCFlushSubpicture),
(XvMCDestroySubpicture), (XvMCGetSubpictureStatus),
(XvMCFlushSurface), (XvMCGetSurfaceStatus), (XvMCQueryAttributes),
(XvMCSetAttribute), (XvMCGetAttribute), (XvMCHideSurface):
* libxvmc/xf86dri.c:
Remove the "config.h" includes from libxvmc, as they break it.
I wonder how the openChrome libXvMC has been working?
Indent viaXvMC.c.
2006-07-10 Thomas Hellström <thomas-at-tungstengraphics-dot-com>
* libxvmc/viaXvMC.c: (XvMCCreateContext):
* unichrome/via_dri.c: (VIADRIRingBufferInit), (VIADRIScreenInit):
* unichrome/via_xvmc.c: (ViaInitXVMC):
Update DRM version checking.
2006-07-10 Thomas Hellström <thomas-at-tungstengraphics-dot-com>
* unichrome/via_swov.c: (viaWaitVideoCommandFire):
Increase video command fire timeout.
2006-07-10 Thomas Hellström <thomas-at-tungstengraphics-dot-com>
* unichrome/via_memory.c: (viaOffScreenLinear):
Fix dri close screen segfault caused by previous commit.
2006-06-15 Thomas Hellström <thomas-at-tungstengraphics-dot-com>
* unichrome/via_dri.c: (VIADRIFBInit):
* unichrome/via_driver.h:
* unichrome/via_memory.c: (viaOffScreenLinear), (VIAAllocLinear):
Fix severe dri memory allocation bug. (Reported by Trevor Kramer)
2006-05-11 Ivor Hewitt <ivor-at-ivor-dot-org>
* configure.ac:
* unichrome/via.h:
* unichrome/via_accel.c:
* unichrome/via_cursor.c:
* unichrome/via_dri.c:
* unichrome/via_driver.h:
* unichrome/via_memcpy.c:
* unichrome/via_memory.c:
* unichrome/via_mode.c:
* unichrome/via_shadow.c:
* unichrome/via_swov.c:
* unichrome/via_vbe.c:
* unichrome/via_vgahw.c:
* unichrome/via_video.c:
* unichrome/via_xvmc.c:
- Xorg xv ABI compatibility patch. (Marcin Kurek)
2006-04-14 Ivor Hewitt <ivor-at-ivor-dot-org>
* libxvmc/Imakefile:
- Unbreak makefile.
2006-03-19 Thomas Hellström <thomas-at-tungstengraphics-dot-com>
* libxvmc/Makefile.am:
* unichrome/Makefile.am:
- Backport Makefile changes form xorg.
2006-03-17 Thomas Hellström <thomas-at-tungstengraphics-dot-com>
* unichrome/via_driver.c:
- Update the built-against macro to recognize Xorg if present.
(Reported by Eric Anholt)
2006-03-17 Thomas Hellström <thomas-at-tungstengraphics-dot-com>
* configure.ac:
- Update compatibility check for XvMC.
2006-03-17 Thomas Hellström <thomas-at-tungstengraphics-dot-com>
* unichrome/via_accel.c: (viaInitExa):
* unichrome/via_driver.c: (VIAPreInit):
- Support EXA 2.0.
2006-03-13 Thomas Hellström <thomas-at-tungstengraphics-dot-com>
* unichrome/via_accel.c: (viaSetupForSolidLine),
(viaSubsequentSolidTwoPointLine), (viaSetupForDashedLine):
* unichrome/via_driver.h:
- Possible fix for XAA bug rendering solid lines as dashed.
(Reported by Lewin Edwards)
2006-03-08 Thomas Hellström <thomas-at-tungstengraphics-dot-com>
* libxvmc/Imakefile:
* unichrome/via_swov.c:
* unichrome/via_swov.h:
* unichrome/via_video.c: (viaInitVideo), (viaExitVideo),
* unichrome/via_video.h:
* unichrome/via_xvmc.c: (initViaXvMC), (cleanupViaXvMC), (stride),
* unichrome/via_xvmc.h:
* unichrome/via_xvpriv.h:
- Run indent on above files after syncing them with Xorg.
2006-02-23 Xavier Bachelot <xb_ml AT kelkoo DOT net>
* configure.ac:
- Fix man pages suffix.
- Bump driver version number to be equal to xorg.
2006-02-21 Ivor Hewitt <ivor-at-ivor-dot-org>
* unichrome/via_bios.h:
* unichrome/via_driver.c: (VIAPreInit):
* unichrome/via_mode.h:
* unichrome/via_priv.h:
* unichrome/via_swov.c: (viaOverlayGetV1V3Format),
(viaOverlayGetSrcStartAddress), (viaOverlayGetFetch),
(CreateSurface), (ViaSwovSurfaceCreate), (ViaSwovSurfaceDestroy),
(VIAVidUpdateOverlay), (ViaOverlayHide):
* unichrome/via_video.c: (viaInitVideo), (Flip), (viaDmaBlitImage),
(viaPutImage), (viaQueryImageAttributes):
- Lets have RV32 too.
- Add 12x8 panel.
2006-02-10 Ivor Hewitt <ivor-at-ivor-dot-org>
* unichrome/via.h:
* unichrome/via_bandwidth.c: (ViaSetPrimaryFIFO),
(ViaSetSecondaryFIFO):
* unichrome/via_dri.c: (VIADRIRingBufferInit):
* unichrome/via_mode.c: (ViaModePrimary):
* unichrome/via_swov.c: (VIAVidHWDiffInit),
(viaOverlayGetSrcStartAddress), (viaOverlayGetFetch),
(viaCalculateVideoColor), (viaSetColorSpace),
(ViaInitVideoStatusFlag), (ViaSetVidCtl), (SetFIFO_V3),
(SetFIFO_V3_64or32or32), (SetFIFO_V3_64or32or16), (SetColorKey),
(SetChromaKey), (SetHQVFetch), (Upd_Video):
* unichrome/via_video.c: (DecideOverlaySupport), (viaInitVideo),
(viaDmaBlitImage):
* unichrome/via_xvmc.c: (ViaInitXVMC), (ViaXvMCCreateContext):
- Get some code exercised for VM800, let's see where it crumbles.
2006-01-29 Thomas Hellström <thomas-at-tungstengraphics-dot-com>
* unichrome/via_vbe.c: (ViaVbeSetMode):
- Fix VBE refresh setting.
2006-01-29 Thomas Hellström <thomas-at-tungstengraphics-dot-com>
* unichrome/via_accel.c: (viaInitExa):
* unichrome/via_driver.c: (VIAPreInit):
- Merge EXA branch revisions 152:155
2006-01-29 Thomas Hellström <thomas-at-tungstengraphics-dot-com>
* unichrome/via_swov.c:
* unichrome/via_video.c: (viaXvError), (viaReputImage),
(viaSetupAdaptors), (viaDmaBlitImage), (viaPutImage):
* unichrome/via_xvpriv.h:
- Removed time.h and sys/time.h includes from via_swov.c
(Reported by George E Sollish).
- More verbose Xv error reporting.
2006-01-27 Thomas Hellström <thomas-at-tungstengraphics-dot-com>
* unichrome/via.man:
* unichrome/via_3d.c:
* unichrome/via_accel.c: (viaAccelPlaneMaskHelper), (viaInitExa),
(viaInitAccel), (viaExitAccel), (viaFinishInitAccel):
* unichrome/via_driver.c: (VIAPreInit):
* unichrome/via_driver.h:
- Merge EXA branch revisions 138:152
2006-01-27 Thomas Hellström <thomas-at-tungstengraphics-dot-com>
* unichrome/via_mode.c: (ViaModePrimary):
- Import Luc's fix for sometimes-blanking CRTs on CLE266.
(Luc Verhaegen)
2006-01-27 Thomas Hellström <thomas-at-tungstengraphics-dot-com>
* unichrome/via_swov.c: (viaWaitVideoCommandFire):
- The previous commit causes problems with older X versions.
Avoid using gettimeofday(). Reported by Paul Bender.
2006-01-26 Thomas Hellström <thomas-at-tungstengraphics-dot-com>
* unichrome/via_swov.c: (viaTimeDiff), (viaWaitVideoCommandFire):
- Make the Xv viaWaitVideoCommandFire loop time out. Better than server hang?
2006-01-26 Thomas Hellström <thomas-at-tungstengraphics-dot-com>
* unichrome/via.man:
* unichrome/via_driver.c: (VIAPreInit):
* unichrome/via_driver.h:
* unichrome/via_video.c: (viaInitVideo):
- Added a "NoXVDMA" option to disable PCI DMA for Xv image transfers.
May turn out useful to vlc users.
- Updated man page.
2006-01-26 Thomas Hellström <thomas-at-tungstengraphics-dot-com>
* unichrome/via_driver.c:
- Added vgaHWGetIndex to vgaHW symbol list.
(Reported by Ole Sandum)
2006-01-26 Thomas Hellström <thomas-at-tungstengraphics-dot-com>
* unichrome/via_swov.c: (ViaOverlayHide):
- Fix HQV offset bug.
2006-01-26 Thomas Hellström <thomas-at-tungstengraphics-dot-com>
* libxvmc/driDrawable.c: (drawStamp), (getDRIDrawableInfoLocked),
(driDestroyHashContents):
* libxvmc/driDrawable.h:
* libxvmc/viaLowLevel.c: (hwlLock), (hwlUnlock), (timeDiff),
(viaDMATimeStampLowLevel), (viaDMAWaitTimeStamp),
(viaDMAInitTimeStamp), (viaDMACleanupTimeStamp),
(viaMpegGetStatus), (viaMpegIsBusy), (syncDMA), (syncVideo),
(syncAccel), (syncMpeg), (pciFlush), (agpFlush),
(flushXvMCLowLevel), (flushPCIXvMCLowLevel), (pciCommand),
(viaMpegSetSurfaceStride), (viaVideoSetSWFLipLocked),
(viaVideoSWFlipLocked), (viaMpegSetFB), (viaMpegBeginPicture),
(viaMpegReset), (viaMpegWriteSlice), (viaVideoSubPictureOffLocked),
(viaVideoSubPictureLocked), (viaBlit), (syncXvMCLowLevel),
(initXvMCLowLevel), (closeXvMCLowLevel):
* libxvmc/viaLowLevel.h:
* libxvmc/viaLowLevelPro.c: (initHQVShadow),
(setHQVHWDeinterlacing), (setHQVDeblocking), (setHQVStartAddress),
(setHQVColorSpaceConversion), (setHQVFetchLine), (setHQVScale),
(setHQVSingleDestination), (setHQVDeinterlacing),
(setHQVTripleBuffer), (finish_header_agp), (hwlLock), (hwlUnlock),
(timeDiff), (viaDMATimeStampLowLevel), (viaDMAWaitTimeStamp),
(viaDMAInitTimeStamp), (viaDMACleanupTimeStamp),
(viaMpegGetStatus), (viaMpegIsBusy), (syncDMA), (syncVideo),
(syncAccel), (syncMpeg), (pciFlush), (agpFlush),
(uploadHQVDeinterlace), (uploadHQVShadow), (flushXvMCLowLevel),
(flushPCIXvMCLowLevel), (viaMpegSetSurfaceStride),
(viaVideoSetSWFLipLocked), (viaVideoSWFlipLocked), (viaMpegSetFB),
(viaMpegBeginPicture), (viaMpegReset), (viaMpegWriteSlice),
(viaVideoSubPictureOffLocked), (viaVideoSubPictureLocked),
(viaBlit), (syncXvMCLowLevel), (updateLowLevelBuf),
(cleanupLowLevelBuf), (releaseXvMCLowLevel), (initXvMCLowLevel),
(setLowLevelLocking), (closeXvMCLowLevel), (computeDownScaling),
(computeHQVScaleAndFilter), (setupBackBuffer):
* libxvmc/viaXvMCPriv.h:
* libxvmc/vldXvMC.h:
* libxvmc/xf86dri.c: (uniDRIDestroyContext),
(uniDRICreateDrawable), (uniDRIDestroyDrawable),
(uniDRIGetDrawableInfo):
* libxvmc/xf86dri.h:
* libxvmc/xf86dristr.h:
- Run indent on libXvMC. Sync XvMC with Xorg 7.0.
2006-01-26 Thomas Hellström <thomas-at-tungstengraphics-dot-com>
* unichrome/via_dri.c: (VIADRIKernelInit), (VIADRIMapInit):
* unichrome/via_driver.h:
- Add a framebuffer map for DRI. Don't just assume that the
framebuffer map handle is the physical address.
2006-01-12 Thomas Hellström <unichrome-at-shipmail-dot-org>
* unichrome/via_accel.c: (viaExaDownloadFromScreen),
(viaExaTexUploadToScreen), (viaExaUploadToScreen),
(viaExaCheckComposite):
* unichrome/via_driver.h:
- Exa branch revision 132:138 merge.
2006-01-12 Thomas Hellström <unichrome-at-shipmail-dot-org>
* unichrome/via_swov.c: (CreateSurface):
* unichrome/via_video.c: (viaInitVideo), (Flip), (viaDmaBlitImage):
- Fix planar Xv Flipping and surface allocation size.
(Reported by Tim, MagicITX)
- Fix Xv dmablit strides and bump drm version for dmablit to 2.9.0, since
there is a direction bug in earlier versions.
2006-01-09 Thomas Hellström <unichrome-at-shipmail-dot-org>
* libxvmc/viaXvMC.c: (releaseContextResources):
* unichrome/via_swov.c: (CreateSurface), (ViaSwovSurfaceDestroy):
- Mark already destroyed XvMC contexts as bad, in case clients
try to destroy them twice.
- Don't try to destroy YV12 buffers when exiting XvMC video.
2006-01-08 Thomas Hellström <unichrome-at-shipmail-dot-org>
* unichrome/via_accel.c: (viaPixelARGB8888):
- Merge exa_branch 130:132. Rendering errors.
- Changelog typo.
2006-01-08 Thomas Hellström <unichrome-at-shipmail-dot-org>
* unichrome/via_3d.c: (viaSet3DFlags), (viaSet3DCompositeOperator):
* unichrome/via_accel.c: (viaIsAGP):
- Merge exa_branch 127:130. Rendering errors.
2006-01-08 Thomas Hellström <unichrome-at-shipmail-dot-org>
* unichrome/Makefile.am:
- Modular build fix. (Boris Dusek)
2006-01-07 Thomas Hellström <unichrome-at-shipmail-dot-org>
* unichrome/Imakefile:
* unichrome/via_accel.c: (viaDumpDMA), (viaFlushPCI),
(viaFlushDRIEnabled), (viaSetupCBuffer), (viaAccelClippingHelper),
(viaAccelSolidHelper), (viaAccelPlaneMaskHelper),
(viaAccelTransparentHelper), (viaAccelCopyHelper),
(viaSetupForScreenToScreenCopy), (viaSubsequentScreenToScreenCopy),
(viaSetupForSolidFill), (viaSubsequentSolidFillRect),
(viaSetupForMono8x8PatternFill),
(viaSubsequentMono8x8PatternFillRect),
(viaSetupForColor8x8PatternFill),
(viaSubsequentColor8x8PatternFillRect),
(viaSetupForCPUToScreenColorExpandFill),
(viaSubsequentScanlineCPUToScreenColorExpandFill),
(viaSetupForImageWrite), (viaSubsequentImageWriteRect),
(viaSetupForSolidLine), (viaSubsequentSolidTwoPointLine),
(viaSubsequentSolidHorVertLine), (viaSetupForDashedLine),
(viaAccelMarkSync), (viaExaPrepareSolid), (viaExaSolid),
(viaExaPrepareCopy), (viaExaCopy), (viaExaCompositePictDesc),
(viaExaPrintComposite), (viaBitExpandHelper), (viaPixelARGB8888),
(viaExpandablePixel), (viaCheckUpload), (viaOrder),
(viaExaDownloadFromScreen), (viaExaTexUploadToScreen),
(viaExaUploadToScreen), (viaExaUploadToScratch),
(viaExaCheckComposite), (viaIsAGP), (viaIsOffscreen),
(viaExaPrepareComposite), (viaExaComposite), (viaInitExa),
(viaInitAccel), (viaExitAccel), (viaFinishInitAccel),
(viaAccelBlitRect), (viaAccelFillRect), (viaAccelSyncMarker):
* unichrome/via_dga.c:
* unichrome/via_dmabuffer.h:
* unichrome/via_dri.c: (VIADRIFBInit):
* unichrome/via_driver.c: (VIALeaveVT), (VIAScreenInit),
(VIAWriteMode), (VIACloseScreen):
* unichrome/via_driver.h:
- Merge in exa_branch from revision 104 to 127. Exa Composite acceleration.
See the branch Changelog for a detailed description of changes.
2006-01-06 Thomas Hellström
* unichrome/via_accel.c: (viaExaDownloadFromScreen),
(viaExaUploadToScreen):
* unichrome/via_video.c: (viaDmaBlitImage):
- Update Xv blit to new blit-combining feature of DRM.
(Idea from Luc Verhaegen/Unichrome).
- Prepare for ugly via_drm.h dmablit IOCTL arg change and via_drm.h
versioning.
2005-12-29 Thomas Hellström <unichrome-at-shipmail-dot-org>
* unichrome/via_video.c: (viaDmaBlitImage):
- Fix Xv YUY2/RVXX dmaBlit stride.
2005-12-26 Thomas Hellström <unichrome-at-shipmail-dot-org>
* libxvmc/Makefile.am:
- Modular built two identical XvMC libs. The pro lib was not a pro lib.
Fix this. (Tim Dodge)
2005-12-15 Thomas Hellström <unichrome-at-shipmail-dot-org>
* unichrome/via_accel.c: (viaInitAccel):
- Make sure the accel marker system is properly initialized.
2005-12-08 Thomas Hellström <unichrome-at-shipmail-dot-org>
* unichrome/via_driver.c: (VIAScreenInit):
* unichrome/via_memory.c: (VIAAllocLinear):
- Protect the accelerated initial sceen clearing with a DRI lock.
- Submit the correct context for drm memory allocation.
2005-12-08 Thomas Hellström <unichrome-at-shipmail-dot-org>
* unichrome/via_ch7xxx.c: (CH7xxxModeI2C):
* unichrome/via_dri.c: (VIADRIAgpInit):
* unichrome/via_driver.c: (VIAScreenInit):
* unichrome/via_id.c:
- Moved the memset from previous commit to after dri is initialized, so
that AGP command submission will work.
- Silenced some warnings.
2005-12-08 Thomas Hellström <unichrome-at-shipmail-dot-org>
* unichrome/via_driver.c: (VIAScreenInit):
* unichrome/via_mode.c: (ViaModePrimary):
* unichrome/via_vbe.c: (ViaVbeSetMode):
- Moved memset to blank screen from ModeSetting to ScreenInit, in
an attempt to avoid long standing hangs on K8M800. Also use
the 2D engine for this if acceleration is enabled.
2005-12-07 Thomas Hellström <unichrome-at-shipmail-dot-org>
* Makefile.in:
- Removed Makefile.in since it is generated by the build process.
2005-12-06 Jon Nettleton <jon-dot-nettleton-at-gmail-dot-com>
* Makefile.am:
* Makefile.in:
- Added Makefile.am and Makefile.in to allow running make
in the root directory.
2005-12-06 Thomas Hellström <unichrome-at-shipmail-dot-org>
* acinclude.m4:
* autogen.sh:
* configure.ac:
* libxvmc/Makefile.am:
* libxvmc/driDrawable.c:
* libxvmc/viaLowLevel.c:
* libxvmc/viaLowLevelPro.c:
* libxvmc/viaXvMC.c:
* libxvmc/viaXvMCPriv.h:
* libxvmc/xf86dri.c:
* man/Makefile.am:
* unichrome/Makefile.am:
* unichrome/via_accel.c:
* unichrome/via_bandwidth.c:
* unichrome/via_ch7xxx.c:
* unichrome/via_cursor.c:
* unichrome/via_dga.c:
* unichrome/via_dri.c:
* unichrome/via_driver.c:
* unichrome/via_i2c.c:
* unichrome/via_id.c:
* unichrome/via_memcpy.c:
* unichrome/via_memory.c:
* unichrome/via_mode.c:
* unichrome/via_shadow.c:
* unichrome/via_vbe.c:
* unichrome/via_vgahw.c:
* unichrome/via_vt162x.c:
* unichrome/via_xvmc.c:
- Adjust for modular build.
2005-12-04 Thomas Hellström <unichrome-at-shipmail-dot-org>
* unichrome/via_vt162x.c: (VT1622ModeCrtc):
- Fix trac item #13.
This will probably also make TV-out work on some K8M/N800 laptops,
and will probaly fix TV-out after a resume.
2005-12-02 Thomas Hellström <unichrome-at-shipmail-dot-org>
* libxvmc/Imakefile:
* libxvmc/xf86dri.c:
- 64 bit fixes on libviaXvMC*
2005-12-02 Thomas Hellström <unichrome-at-shipmail-dot-org>
* unichrome/via_accel.c: (viaAccelPlaneMaskHelper),
(viaAccelTransparentHelper), (viaSetupForScreenToScreenCopy),
(viaSetupForSolidFill), (viaSetupForMono8x8PatternFill),
(viaSetupForColor8x8PatternFill),
(viaSetupForCPUToScreenColorExpandFill), (viaSetupForImageWrite),
(viaSetupForSolidLine), (viaSubsequentSolidTwoPointLine),
(viaSetupForDashedLine), (viaAccelMarkSync), (viaExaPrepareSolid),
(viaExaPrepareCopy), (viaInitAccel), (viaDGABlitRect),
(viaDGAFillRect):
* unichrome/via_driver.c:
* unichrome/via_driver.h:
- Merge exa_branch revisions 91:104:
Add support for EXA planemasks.
Tell loader about some missing EXA functions (Reported by Adam Jackson).
Fix broken XAA dashed lines.
2005-12-01 Thomas Hellström <unichrome-at-shipmail-dot-org>
* libxvmc/Imakefile:
* unichrome/Imakefile:
* unichrome/via_driver.c:
- Remove drmCommandWriteRead from drm loader symbols again.
Ivor had already added it :)
- Fix a typo in unichrome/Imakefile that broke build on x86_64.
- Add $(PICFLAGS) to libxvmc Makefile to make it build on x86_64.
2005-12-01 Thomas Hellström <unichrome-at-shipmail-dot-org>
* unichrome/via_driver.c:
* unichrome/via_memory.c: (VIAFreeLinear):
- Add drmCommandWriteRead to drm loader symbols.
- Make the drm freemem call drmCommandWrite.
2005-11-29 Thomas Hellström <unichrome-at-shipmail-dot-org>
* unichrome/via_accel.c: (viaInitAccel):
* unichrome/via_memory.c: (VIAAllocLinear):
- Merge exa_branch revisions 86:91
2005-11-29 Thomas Hellström <unichrome-at-shipmail-dot-org>
* unichrome/Imakefile:
- Yet another drm.h include fixup. For 6.8 and earlier?
2005-11-28 Ivor Hewitt <ivor-at-ivor-dot-org>
* unichrome/via_id.c:
- Just another id. (Tobias Blom)
2005-11-28 Thomas Hellström <unichrome-at-shipmail-dot-org>
* unichrome/Imakefile:
- Include the correct drm.h file. This is important for 64 bit builds.
2005-11-26 Thomas Hellström <unichrome-at-shipmail-dot-org>
* unichrome/via_accel.c: (viaFlushDRIEnabled), (viaSetupCBuffer):
- Merged EXA branch changes between revisions 84 and 86.
2005-11-23 Thomas Hellström <unichrome-at-shipmail-dot-org>
* unichrome/via_accel.c: (viaFlushPCI), (viaFlushDRIEnabled),
(viaSetupCBuffer), (viaTearDownCBuffer), (viaEnableVQ),
(viaDisableVQ), (viaAccelSetMode), (viaInitialize2DEngine),
(viaAccelSync), (viaSetClippingRectangle),
(viaAccelClippingHelper), (viaAccelSolidHelper),
(viaAccelTransparentHelper), (viaAccelCopyHelper),
(viaSetupForScreenToScreenCopy), (viaSubsequentScreenToScreenCopy),
(viaSetupForSolidFill), (viaSubsequentSolidFillRect),
(viaSetupForMono8x8PatternFill),
(viaSubsequentMono8x8PatternFillRect),
(viaSetupForColor8x8PatternFill),
(viaSubsequentColor8x8PatternFillRect),
(viaSetupForCPUToScreenColorExpandFill),
(viaSubsequentScanlineCPUToScreenColorExpandFill),
(viaSetupForImageWrite), (viaSubsequentImageWriteRect),
(viaSetupForSolidLine), (viaSubsequentSolidTwoPointLine),
(viaSubsequentSolidHorVertLine), (viaSetupForDashedLine),
(viaSubsequentDashedTwoPointLine), (viaInitXAA),
(viaAccelMarkSync), (viaAccelWaitMarker), (viaExaPrepareSolid),
(viaExaSolid), (viaExaPrepareCopy), (viaExaCopy),
(viaExaDownloadFromScreen), (viaExaUploadToScreen), (viaInitExa),
(viaInitAccel), (viaExitAccel), (viaDGABlitRect), (viaDGAFillRect):
- Merged EXA branch changes between revisions 74 and 84. Mainly
indentations and minor bugfixes.
2005-11-15 Thomas Hellström <unichrome-at-shipmail-dot-org>
* unichrome/via_swov.c: (ViaSwovSurfaceCreate),
(ViaSwovSurfaceDestroy):
* unichrome/via_video.c: (viaSetupAdaptors):
* unichrome/via_xvpriv.h:
- Fix Xv surface destruction and re-allocation.
(Reported by Cedric Berger).
2005-11-15 Thomas Hellström <unichrome-at-shipmail-dot-org>
* unichrome/Imakefile:
* unichrome/via_accel.c: (viaTearDownCBuffer),
(viaAccelSolidHelper), (viaAccelCopyHelper),
(viaExaDownloadFromScreen), (viaExaUploadToScreen):
* unichrome/via_driver.c: (VIAScreenInit):
- Merged changes in exa_branch revisions 67 to 74. For a detailed
ChangeLog, see that branch.
2005-11-09 Thomas Hellström <unichrome-at-shipmail-dot-org>
* unichrome/Imakefile:
* unichrome/via_accel.c: (viaFlushPCI), (viaFlushDRIEnabled),
(viaSetupCBuffer), (viaTearDownCBuffer), (viaInitAgp),
(viaEnableVQ), (viaDisableVQ), (viaAccelSetMode),
(viaInitialize2DEngine), (viaAccelSync), (viaSetClippingRectangle),
(viaDisableClipping), (viaAccelClippingHelper),
(viaAccelSolidHelper), (viaAccelTransparentHelper),
(viaAccelCopyHelper), (viaSetupForScreenToScreenCopy),
(viaSubsequentScreenToScreenCopy), (viaSetupForSolidFill),
(viaSubsequentSolidFillRect), (viaSetupForMono8x8PatternFill),
(viaSubsequentMono8x8PatternFillRect),
(viaSetupForColor8x8PatternFill),
(viaSubsequentColor8x8PatternFillRect),
(viaSetupForCPUToScreenColorExpandFill),
(viaSubsequentScanlineCPUToScreenColorExpandFill),
(viaSetupForImageWrite), (viaSubsequentImageWriteRect),
(viaSetupForSolidLine), (viaSubsequentSolidTwoPointLine),
(viaSubsequentSolidHorVertLine), (viaSetupForDashedLine),
(viaSubsequentDashedTwoPointLine), (viaInitXAA),
(viaAccelMarkSync), (viaAccelWaitMarker), (viaExaPrepareSolid),
(viaExaSolid), (viaExaDoneSolidCopy), (viaExaPrepareCopy),
(viaExaCopy), (viaExaDownloadFromScreen), (viaExaUploadToScreen),
(viaInitExa), (viaInitAccel), (viaExitAccel), (viaDGABlitRect),
(viaDGAFillRect), (viaDGAWaitMarker):
* unichrome/via_cursor.c: (VIALoadCursorImage):
* unichrome/via_dga.c:
* unichrome/via_dri.c: (VIADRIFBInit), (VIADRIScreenInit),
(VIADRICloseScreen), (VIADRIFinishScreenInit):
* unichrome/via_driver.c: (VIASetup), (VIAPreInit), (VIALeaveVT),
(VIAScreenInit), (VIAWriteMode), (VIACloseScreen), (VIASwitchMode):
* unichrome/via_driver.h:
* unichrome/via_memcpy.c: (viaVidCopyInit):
* unichrome/via_memory.c: (VIAFreeLinear), (offScreenLinear),
(VIAAllocLinear), (VIAInitLinear):
* unichrome/via_priv.h:
* unichrome/via_vbe.c: (ViaVbeSetMode):
- Merge exa_branch from revision 52 to 67. For a detailed Changelog,
check that branch.
2005-11-09 Ivor Hewitt <ivor-at-ivor-dot-org>
* unichrome/via_driver.c: (VIAPreInit):
* unichrome/via_mode.c: (ViaModeDotClockTranslate):
- Fix pro-mode and remove pro warning. (Paul Bender)
2005-11-08 Ivor Hewitt <ivor-at-ivor-dot-org>
* unichrome/via_driver.c: (VIASwitchMode):
- Don't kickVblank with no DRI enabled. (Joris van Rantwijk)
2005-11-02 Ivor Hewitt <ivor-at-ivor-dot-org>
* unichrome/via_mode.c: (ViaGetMemoryBandwidth):
* unichrome/via_mode.h:
- Give VM800 some bandwidth.
2005-11-01 Thomas Hellström <unichrome-at-shipmail-dot-org>
* libxvmc/viaLowLevel.c: (viaDMAInitTimeStamp):
* libxvmc/viaLowLevelPro.c: (viaDMAInitTimeStamp),
(updateLowLevelBuf):
* unichrome/via_drmclient.h:
- Fix compilation errors caused by the drm / dri update.
(reported by Tarun Kripalani)
2005-10-31 Thomas Hellström <unichrome-at-shipmail-dot-org>
* unichrome/via_accel.c: (VIAInitAccel):
* unichrome/via_cursor.c: (VIAHWCursorInit):
- Second attempt at fixing the hw cursor issue.
There were actually two errors involved.
2005-10-31 Thomas Hellström <unichrome-at-shipmail-dot-org>
* unichrome/via_xvpriv.h:
- Fix stray include left over by previous commit.
2005-10-31 Ivor Hewitt <ivor-at-ivor-dot-org>
* unichrome/via_id.c:
- Mitac 8889 device id. (Daniel Schindler)
2005-10-30 Thomas Hellström <unichrome-at-shipmail-dot-org>
* unichrome/via_accel.c: (VIAInitAccel):
- Revert the hardware cursor fix since it for some strange reason
makes hw cursors white. Need to figure out why.
2005-10-30 Thomas Hellström <unichrome-at-shipmail-dot-org>
* unichrome/via_video.c: (viaInitVideo), (viaSetupAdaptors),
(nv12cp), (viaDmaBlitImage), (viaPutImage), (nv12Blit):
* unichrome/via_xvpriv.h:
- Enable DMA Xv image transfers also for PM8X0 / CN400.
YV12 to NV12 conversion is still software but now takes place
in system memory instead of during transfer to framebuffer memory.
CPU usage is much improved on PM8X0. Less so on CN400.
Hardware YV12 to NV12 conversion using the blitter was implemented
and tested but was very slow, probably due to the fact that blitting
only took place one column at a time. Since sync on the 2D engine
currently requires a busy-wait, the approach was not only slow, but
also CPU-consuming.
2005-10-27 Thomas Hellström <unichrome-at-shipmail-dot-org>
* unichrome/via_video.c: (viaExitVideo), (viaSaveVideo),
(viaRestoreVideo), (VIAVidAdjustFrame):
- Fix server crash when compiled without -DXvExtension.
2005-10-27 Thomas Hellström <unichrome-at-shipmail-dot-org>
* unichrome/via_dri.c: (VIADRIRingBufferInit), (VIADRIScreenInit),
(VIADRIFinishScreenInit), (VIADRIMapInit):
* unichrome/via_dri.h:
* unichrome/via_driver.h:
* unichrome/via_drm.h:
* unichrome/via_memory.c: (VIAFreeLinear), (VIAAllocLinear):
* unichrome/via_video.c: (viaInitVideo):
* unichrome/via_xvmc.c: (ViaInitXVMC):
- Sync dri handling with Xorg. This should hopefully make it run
with Mesa 6.4 again. (Luc Verhaegen, Eric Anholt, me)
2005-10-27 Thomas Hellström <unichrome-at-shipmail-dot-org>
* unichrome/via_video.c: (viaPutImage):
- Make via_video.c compile without -DXF86DRI. Was broken with the
dmablit commit. (Reported by Luc Verhaegen)
2005-10-27 Thomas Hellström <unichrome-at-shipmail-dot-org>
* unichrome/via_accel.c: (VIAInitAccel):
- Fix HW cursor memory allocation.
2005-10-18 Ivor Hewitt <ivor-at-ivor-dot-org>
* unichrome/via_bandwidth.c: (ViaSetPrimaryFIFO),
(ViaSetSecondaryFIFO):
* unichrome/via_driver.c: (VIAPreInit):
* unichrome/via_id.c:
* unichrome/via_id.h:
* unichrome/via_swov.c: (VIAVidHWDiffInit):
- Remaining device id's lying about.
- Remove PM800 conditionals.
2005-10-18 Ivor Hewitt <ivor-at-ivor-dot-org>
* unichrome/via_driver.c: (VIAPreInit):
* unichrome/via_id.c:
* unichrome/via_id.h:
* unichrome/via_mode.h:
- Device id additions. AsRock P4VM800 reported by Mariano Benedettini
2005-09-25 Thomas Hellström <unichrome-at-shipmail-dot-org>
* unichrome/via.h:
* unichrome/via_dri.c: (VIADRIRingBufferInit),
(VIADRIFinishScreenInit):
* unichrome/via_dri.h:
* unichrome/via_driver.h:
* unichrome/via_drm.h:
* unichrome/via_priv.h:
* unichrome/via_swov.c: (viaWaitHQVFlip), (viaWaitHQVFlipClear),
(viaWaitHQVDone), (viaOverlayGetV1V3Format),
(viaOverlayGetSrcStartAddress), (viaOverlayGetFetch),
(viaSetColorSpace), (ViaSetVidCtl), (AddHQVSurface),
(ViaSwovSurfaceCreate), (ViaSwovSurfaceDestroy),
(SetFIFO_V3_64or32or32), (SetFIFO_V3_64or32or16), (SetColorKey),
(SetChromaKey), (SetHQVFetch), (Upd_Video), (VIAVidUpdateOverlay),
(ViaOverlayHide):
* unichrome/via_video.c: (DecideOverlaySupport), (viaResetVideo),
(viaSaveVideo), (viaExitVideo), (viaInitVideo), (viaReputImage),
(viaSetupAdaptors), (viaStopVideo), (Flip), (nv12cp),
(viaDmaBlitImage), (viaPutImage), (viaQueryImageAttributes):
* unichrome/via_xvmc.c: (ViaInitXVMC):
* unichrome/via_xvpriv.h:
Big Xv update.
- Bring Xv in sync with Xorg, which includes RV15 and RV16 support.
- Add support for PCI DMA Xv image transfer. This requires DRM >= 2.7.0,
and does not yet work on Unichrome Pro group A, since YV12->NV12
hardware conversion is not yet implemented.
2005-08-13 Thomas Hellström <unichrome-at-shipmail-dot-org>
* unichrome/via_swov.c: (ViaSwovSurfaceDestroy):
- Destroy video surfaces on video exit / stop.
2005-08-13 Thomas Hellström <unichrome-at-shipmail-dot-org>
* unichrome/via_driver.c: (VIAAdjustFrame)
- Fix Xv panning also for VBEModes.
2005-08-12 Thomas Hellström <unichrome-at-shipmail-dot-org>
* unichrome/via_driver.c: (VIAEnterVT):
- Fix segfault on EnterVT when DRI is not enabled.
(Joris van Rantwijk)
2005-08-12 Thomas Hellström <unichrome-at-shipmail-dot-org>
* unichrome/via_priv.h:
* unichrome/via_swov.c: (VIAVidUpdateOverlay):
* unichrome/via_video.c: (RegionsEqual), (viaReputImage),
(viaSetupAdaptors), (VIAVidAdjustFrame):
- Fix Xv panning. Speed up overlay updates on window moves.
2005-08-12 Thomas Hellström <unichrome-at-shipmail-dot-org>
* unichrome/via_video.c: (viaExitVideo):
- Fix stopvideo for all ports and adaptors.
2005-08-10 Ivor Hewitt <ivor-at-ivor-dot-org>
* unichrome/Imakefile:
* unichrome/via_bios.h:
* unichrome/via_ch7xxx.c: (CH7xxxPrintRegs), (ViaCH7xxxDetect),
(CH7xxxSave), (CH7xxxRestore), (CH7xxxDACSenseI2C),
(CH7xxxDACSense), (CH7011ModeIndex), (CH7019ModeIndex),
(CH7xxxModeValid), (CH7xxxModeI2C), (CH7xxxModeCrtc),
(CH7xxxTVPower), (CH7019LCDPower), (ViaCH7xxxInit):
* unichrome/via_ch7xxx.h:
* unichrome/via_driver.c: (VIAPreInit):
* unichrome/via_id.c:
* unichrome/via_mode.c: (ViaTVDetect), (ViaTVInit):
* unichrome/via_vt162x.c: (ViaVT162xDetect), (VT1625DACSenseI2C),
(VT1625DACSense), (VT1621ModeIndex), (VT1622ModeIndex),
(VT1625ModeValid), (VT1622ModeI2C), (VT1622ModeCrtc),
(VT1625Power), (ViaVT162xInit):
* unichrome/via_vt162x.h:
- Initial steps to incorporating CH7xxx and VT1625 support;
merging patches from David George, Philip Prindeville, Terry Lewis.
2005-07-30 Ivor Hewitt <ivor-at-ivor-dot-org>
* libxvmc/Imakefile:
* unichrome/via_mode.c: (ViaGetMemoryBandwidth),
(ViaModePrimaryVGA), (ViaModePrimary):
* unichrome/via_swov.c:
- Merge fixes from xorg for typos in mode setting and sync set
2005-07-18 Ivor Hewitt <ivor-at-ivor-dot-org>
* unichrome/via_id.c:
- Additional card-id. Reported by Philip Prindeville.
2005-07-04 Thomas Hellström <unichrome-at-shipmail-dot-org>
* unichrome/via_driver.c: (VIAPreInit), (VIAScreenInit):
* unichrome/via_id.c:
- Avoid calling DPMS to blank screen during init with VBEModes.
- Add another card ID.
2005-06-29 Ivor Hewitt <ivor-at-ivor-dot-org>
* unichrome/via_id.c:
- Additional card-ids. Patch from Xavier Bachelot
2005-06-26 Thomas Hellström <unichrome-at-shipmail-dot-org>
* libxvmc/viaLowLevelPro.c: (finish_header_agp), (syncDMA),
(syncVideo), (pciFlush), (agpFlush), (uploadHQVDeinterlace),
(uploadHQVShadow), (flushXvMCLowLevel), (flushPCIXvMCLowLevel),
(viaMpegSetSurfaceStride), (viaVideoSetSWFLipLocked),
(viaVideoSWFlipLocked), (viaMpegSetFB), (viaMpegBeginPicture),
(viaMpegReset), (viaMpegWriteSlice), (viaVideoSubPictureOffLocked),
(viaVideoSubPictureLocked), (viaBlit), (syncXvMCLowLevel),
(releaseXvMCLowLevel), (initXvMCLowLevel), (closeXvMCLowLevel):
- Clean up the VIDEO_DMA ifdef mess before xorg inclusion.
- Make VIDEO_DMA wait for DMA quiescent before flipping, pending proper
HQV locking.
- Disable VIDEO_DMA to improve latency pending proper HQV locking.
2005-06-17 Thomas Hellström <unichrome-at-shipmail-dot-org>
* unichrome/via_mode.c: (ViaSetPrimaryDotclock),
(ViaSetSecondaryDotclock), (ViaComputeProDotClock),
(ViaModeDotClockTranslate):
- Free dotclock on Unichrome Pro. Pls report problems with
unstable clocks.
- Minor 64-bit fixes.
2005-06-16 Thomas Hellström <unichrome-at-shipmail-dot-org>
* unichrome/via_xvmc.c: (ViaXvMCCreateContext),
(ViaXvMCCreateSurface), (ViaXvMCCreateSubpicture):
* unichrome/via_xvmc.h:
- XvMC 64 bit fixes. XvMC now runs on K8M800 64-bit Linux.
2005-06-12 Thomas Hellström <unichrome-at-shipmail-dot-org>
* unichrome/via_bios.h:
* unichrome/via_i2c.c:
* unichrome/via_mode.c: (ViaTVRestore):
* unichrome/via_vt162x.c: (VT162xPrintRegs), (ViaVT162xDetect),
(VT162xSave), (VT162xRestore), (ViaVT162xInit):
- Fix up the vt1623 encoder: Correct device ID, number of registers
and most of all reading from the i2cbus it is sitting on.
2005-06-11 Thomas Hellström <unichrome-at-shipmail-dot-org>
* unichrome/via_dri.c: (VIADRIScreenInit):
- Update via_dri.c to compile with new libdri.
2005-06-01 Thomas Hellström <unichrome-at-shipmail-dot-org>
* unichrome/via_mode.c: (ViaModePrimary):
* unichrome/via_vt162x.h:
- Fix TV-out on Unichrome Pro. To make it work CRT+TV needs to be
enabled in BIOS. Still it will hang the machine or possibly only
the display chip on X server exit.
- Import the 720x576Noscale mode for vt1622A/vt1623 from the Unichrome
driver (Terry Barnaby)
2005-05-25 Thomas Hellström <unichrome-at-shipmail-dot-org>
* libxvmc/viaLowLevelPro.c: (setHQVDeinterlacing):
* libxvmc/viaXvMC.c:
* unichrome/via_driver.c:
* unichrome/via_mode.c: (ViaModePrimary):
* unichrome/via_mode.h:
* unichrome/via_vt162x.c: (ViaVT162xDetect):
- Updated incorrect field order in libViaXvMCPro.
- Fixed warnings about unresolved int10 symbols in the Unichrome driver.
- Updated Unichrome Pro dotclocks - also for TV.
- Made the driver recognize the vt1623 TV encoder on the EPIA SP13000.
- Removed unnecessary define in viaXvMC.c.
2005-05-16 Thomas Hellström <unichrome-at-shipmail-dot-org>
* libxvmc/driDrawable.c: (getDRIDrawableInfoLocked):
* libxvmc/viaXvMC.c: (releaseContextResources),
(XvMCCreateContext):
* libxvmc/xf86dri.c: (uniDRIDestroyContext),
(uniDRICreateDrawable), (uniDRIDestroyDrawable),
(uniDRIGetDrawableInfo):
* libxvmc/xf86dri.h:
- Changed DRI client function names since they clash with
via_dri.so, causing strange problems in, for example, xine.
2005-05-16 Thomas Hellström <unichrome-at-shipmail-dot-org>
* unichrome/via_dri.c: (VIASetAgpMode), (VIADRIAgpInit):
- Added better AGP mode control to the DDX.
2005-05-11 Thomas Hellström <unichrome-at-shipmail-dot-org>
* prepare-ChangeLogSVN.pl:
- Added the above file for generation of ChangeLog Entries.
2005-05-11 Thomas Hellström <unichrome-at-shipmail-dot-org>
* libxvmc/driDrawable.c:
* libxvmc/viaLowLevel.c: (hwlLock), (hwlUnlock),
(setAGPSyncLowLevel), (viaDMATimeStampLowLevel),
(flushXvMCLowLevel), (flushPCIXvMCLowLevel),
(viaMpegSetSurfaceStride), (viaVideoSetSWFLipLocked),
(viaVideoSWFlipLocked), (viaMpegSetFB), (viaMpegBeginPicture),
(viaMpegReset), (viaMpegWriteSlice), (viaVideoSubPictureOffLocked),
(viaVideoSubPictureLocked), (viaBlit), (syncXvMCLowLevel),
(initXvMCLowLevel), (setLowLevelLocking), (closeXvMCLowLevel):
* libxvmc/viaLowLevel.h:
* libxvmc/viaLowLevelPro.c: (initHQVShadow),
(setHQVHWDeinterlacing), (setHQVDeblocking), (setHQVStartAddress),
(setHQVColorSpaceConversion), (setHQVFetchLine), (setHQVScale),
(setHQVSingleDestination), (setHQVDeinterlacing),
(setHQVTripleBuffer), (hwlLock), (hwlUnlock), (setAGPSyncLowLevel),
(viaDMATimeStampLowLevel), (viaDMAInitTimeStamp),
(uploadHQVDeinterlace), (uploadHQVShadow), (flushXvMCLowLevel),
(flushPCIXvMCLowLevel), (viaMpegSetSurfaceStride),
(viaVideoSetSWFLipLocked), (viaVideoSWFlipLocked), (viaMpegSetFB),
(viaMpegBeginPicture), (viaMpegReset), (viaMpegWriteSlice),
(viaVideoSubPictureOffLocked), (viaVideoSubPictureLocked),
(viaBlit), (syncXvMCLowLevel), (updateLowLevelBuf),
(cleanupLowLevelBuf), (initXvMCLowLevel), (setLowLevelLocking),
(closeXvMCLowLevel), (computeDownScaling),
(computeHQVScaleAndFilter), (setupBackBuffer):
* libxvmc/viaXvMC.c: (grabDecoder), (releaseContextResources),
(XvMCCreateContext), (XvMCCreateSurface), (XvMCPutSlice2),
(XvMCPutSlice), (XvMCPutSurface), (XvMCBeginSurface),
(XvMCSyncSurface), (XvMCLoadQMatrix), (XvMCCreateSubpicture),
(XvMCSetSubpicturePalette), (XvMCClearSubpicture),
(XvMCCompositeSubpicture), (XvMCBlendSubpicture2),
(XvMCSyncSubpicture), (XvMCDestroySubpicture), (XvMCFlushSurface),
(XvMCGetSurfaceStatus), (XvMCQueryAttributes), (XvMCSetAttribute),
(XvMCGetAttribute), (XvMCHideSurface):
* libxvmc/viaXvMCPriv.h:
* unichrome/via_swov.c: (ViaSwovSurfaceCreate):
* unichrome/via_video.c: (viaSetupAdaptors):
* unichrome/via_xvpriv.h:
- Added ChangeLog to the repository.
- Made the DDX update the overlay if a change in surface format occured
while it's dimensions are still the same.
- Made the XvMCLowLevel struct opaque (void *) and not defined in the
header file. This allows for different definitions for different hardware
- Added a HQV shadow which holds the HQV context for the Unichrome Pro.
The purpose is twofold:
1. If someone else touched the HQV since we last used it, we can upload the
whole context.
2. We do not need to read from the HQV, which is bad because we have to
halt the DMA engine to do that.
- Added and activated mpeg deblocking for Unichrome Pro.
- Added a number of HQV utility functions for the upcoming video engine free
output. These are not activated yet. (Unichrome Pro.)
- Added HW deinterlacing functions. Deactivated, since I cannot see any visible
difference. (Unichrome Pro.)