summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* svga: Use consistent file names for dumping facilities.José Fonseca2009-11-2710-11/+11
|
* st/xorg: speculatively round textures up to nearest POTKeith Whitwell2009-11-271-0/+18
| | | | | I'm not sure if this is a great change, but helps with caching. Probably we want to turn this on/off on a driver-by-driver basis.
* st/xorg: free last fenceKeith Whitwell2009-11-271-0/+4
|
* svga: flush our command buffer after the 8th distinct render targetKeith Whitwell2009-11-273-1/+11
| | | | | | | | | | | | | | This helps improve the surface cache behaviour in the face of the large number of single-use render targets generated by EXA and the xorg state tracker. Without this we can reference hundreds of individual render targets from a command buffer, which leaves little scope for sharing or reuse of those targets. Flushing early means we can start reusing textures much sooner. This shouldn't have much effect on normal 3d rendering as it's pretty rare to have a command buffer with >8 different render targets in that world.
* svga: add DEBUG_CACHE optionKeith Whitwell2009-11-278-9/+53
|
* svga: Prevent potential null pointer dereference in vmw_surface.c.Vinson Lee2009-11-261-3/+5
|
* vmware/xorg: Remove gem includeJakob Bornecrantz2009-11-251-1/+0
|
* st/xorg: Use pipe clear hook for solid fills of whole pixmaps.Michel Dänzer2009-11-261-0/+6
| | | | Can give a little boost e.g. for anti-aliased text rendering.
* svga: hash the whole key, not just the first four bytesKeith Whitwell2009-11-261-3/+3
|
* st/xorg: Make sure DRI2 blits use GPU copy contents even for software fallback.Michel Dänzer2009-11-261-11/+19
| | | | | | | Fixes 3D apps not updating with a non-GL compositing manager and Option "2DAccel" "off". Also clean up a little pixmap vs. drawable mess.
* st/xorg: fix refcounting bugs introduced in earlier commitKeith Whitwell2009-11-262-0/+5
|
* st/xorg: formattingKeith Whitwell2009-11-261-9/+9
| | | | This directory needs indent run over it.
* st/xorg: remove debugging flushKeith Whitwell2009-11-261-1/+0
| | | | | Accidentally committed in 7fbdbad5c02e3d5bfbf0e641e2aec224e39fa974 ('st/xorg: consolidate some dest surface state setting')
* svga: Remove spurious argument to SVGA_DBG.José Fonseca2009-11-261-1/+1
|
* st/xorg: render throttling in block handlerKeith Whitwell2009-11-263-7/+22
| | | | | | | | | | | | | | Similar to the classic swapbuffer throttling in GL drivers, put an upper bound on the number of outstanding chunks of rendering the state tracker can generate -- where calling the block handler denotes a chunk. Currently that number is set at around 4 "chunks", but could be tweaked up or down. If a better measure for the amount of outstanding rendering is found, that would be fine too. As it stands, this improves interactivity by preventing the X server from queueing up arbitary amounts of rendering.
* st/xorg: split copy operation into prepare/copy/done phasesKeith Whitwell2009-11-265-128/+102
| | | | Any high-overhead one-off tasks are moved into the prepare hook.
* st/xorg: don't set up constant buffer for non-xv fragment shadersKeith Whitwell2009-11-262-30/+0
| | | | | | These currently don't reference any constants. Can add this back if newer shaders need them, but in the meantime don't create a new constant buffer every time we do a blit.
* st/xorg: split up shared Done callKeith Whitwell2009-11-263-53/+99
| | | | | | The two users of composite (Composite and Solid) now call a new xorg_composite_done() from their Done functions, while CopyDone is directly implemented on top of xorg_renderer.c.
* st/xorg: set up rasterizer state in initKeith Whitwell2009-11-263-17/+8
|
* st/xorg: remove redundant clipping codeKeith Whitwell2009-11-261-120/+0
|
* st/xorg: don't bother with cso save and restore in copy funcKeith Whitwell2009-11-261-19/+0
|
* st/xorg: consolidate some dest surface state settingKeith Whitwell2009-11-265-105/+50
|
* glu/sgi: Fix memory leak in gluBuild2DMipmapLevelsCore.Vinson Lee2009-11-261-0/+1
|
* vmware/xorg: Link against libkmsJakob Bornecrantz2009-11-241-0/+1
| | | | If the system doesn't have libkms installed it wont try to link against it.
* vmware/xorg: Clean Makefile a bitJakob Bornecrantz2009-11-241-3/+5
|
* vmware/xorg: Don't link against libdrm_intelJakob Bornecrantz2009-11-241-1/+1
|
* vmware/xorg: Stage driver in lib/galliumJakob Bornecrantz2009-11-241-1/+8
|
* st/xorg: Pass mode types from the kernel to XJakob Bornecrantz2009-11-241-1/+5
|
* st/xorg: Add libkms integrationJakob Bornecrantz2009-11-244-122/+353
|
* st/xorg: Touch up xorg_crtc.cJakob Bornecrantz2009-11-241-0/+6
|
* st/xorg: Rename driver functionsJakob Bornecrantz2009-11-241-76/+84
|
* st/xorg: Rename output functionsJakob Bornecrantz2009-11-241-18/+18
|
* st/xorg: Rename dri2 functionsJakob Bornecrantz2009-11-241-17/+17
|
* st/xorg: Standardise all function names defined in xorg_tracker.hJakob Bornecrantz2009-11-246-20/+19
|
* svga: try harder to make the cachable flag workKeith Whitwell2009-11-252-6/+19
| | | | It doesn't though.
* st/xorg: accelerate src luminanceZack Rusin2009-11-253-21/+38
|
* util: also print out memory statisticsZack Rusin2009-11-251-0/+7
|
* svga: revert packing of surface keyKeith Whitwell2009-11-251-3/+3
| | | | Over-ambitious packing of values broke my cursor.
* radeon: fix context destroy needing lock for flushing.Dave Airlie2009-11-251-5/+4
| | | | | | | | | Thanks to Intel code which I've just stolen pretty much as usual. This fixes fdo bug 22851 which is a dri1 regression since rewrite. Tested by: fpiobaf (Fabio) on #radeon Signed-off-by: Dave Airlie <[email protected]>
* glu/sgi: Fix memory leak in gluBuild1DMipmapLevelsCore.Vinson Lee2009-11-251-0/+1
|
* glu/sgi: Fix memory leak in gluBuild2DMipmapLevelsCore.Vinson Lee2009-11-251-0/+1
|
* glu/sgi: Fix memory leak in gluBuild3DMipmapLevelsCore.Vinson Lee2009-11-251-0/+1
|
* svga: cache textures as well as buffersKeith Whitwell2009-11-245-117/+167
|
* st/xorg: use surface_copy for blits if availableKeith Whitwell2009-11-242-3/+36
| | | | | Even if its not available, we really want to be coalescing blit operations better.
* svga: Filter out pendantic and ansi flagsJakob Bornecrantz2009-11-241-0/+3
| | | | | Rather have the driver compile without the flags then having to disable them.
* glu/sgi: Fix memory leak in gluBuild3DMipmapLevelsCore.Vinson Lee2009-11-241-0/+1
|
* dri: Fix potential null pointer deference in dri_put_drawable.Vinson Lee2009-11-241-4/+4
|
* radeon: fix errors in miptree related functionMaciej Cencora2009-11-231-3/+8
| | | | | | - typo - memory leak - off by one (spotted by airlied)
* Merge branch 'mesa_7_7_branch' of ↵Maciej Cencora2009-11-2312-32/+29
|\ | | | | | | http://anongit.freedesktop.org/git/mesa/mesa into mesa_7_7_branch
| * svga: Scrub Makefiles a bitJakob Bornecrantz2009-11-232-20/+2
| | | | | | | | | | | | Remove x86 specific hacks. Not that they will ever be used on none x86 arches, but they are built by default. And the way the flags where added was a hack.
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-18 17:50:06 +0000