summaryrefslogtreecommitdiff
path: root/src/gallium/winsys/drm/radeon/core/radeon_buffer.c
AgeCommit message (Collapse)Author
2010-02-14radeong: flush when changing tiling flags of BO referenced by CSMarek Olšák
2010-02-08drm/radeon: remove softpipe referencesKeith Whitwell
2010-01-28radeong: Slight improvement to buffer creation.Corbin Simpson
2010-01-25radeong: Clean up domain usages.Corbin Simpson
Split from Marek's immd-mode patch.
2010-01-25Revert "r300g,radeong: finish and enable the immediate mode"Corbin Simpson
This reverts commit 112239e9a66a155d36fe2ad0ab130e6f26eff298.
2010-01-24r300g,radeong: finish and enable the immediate modeMarek Olšák
Nearly 100% performance increase in glxgears.
2010-01-24r300g,radeong: set tiling flags in libdrmMarek Olšák
2010-01-18radeong: Don't allocate HW BOs for constantbufs.Corbin Simpson
We have broken 1000 FPS. Hell yes. Heavily inspired by Marek's patch, but using pipebuffer instead of a roll-your-own malloc.
2010-01-11radeong: Check for DONTBLOCK before flushing, remove dead code.Corbin Simpson
2009-12-18Revert "r300g: flush CS if a buffer being deleted is referenced by it"Corbin Simpson
As requested by just about everybody. I'm going back to actually reading patches before ACKing and pushing them. This reverts commit 417ce06306962a9355cbb35cefcdea1951b0ce85. Conflicts: src/gallium/winsys/drm/radeon/core/radeon_buffer.c
2009-12-18Merge branch 'pipe-format-simplify'Michal Krol
2009-12-18r300g: store own copy of flush_cb and flush data.Dave Airlie
don't go sneaking around inside the libdrm copy.
2009-12-17Move the remaining format pf_get_* functions to u_format.h.Michal Krol
Previously they depended on format blocks, but after removing those they started depending on format encoding.
2009-12-15r300g: flush CS if a buffer being deleted is referenced by itMarek Olšák
2009-12-04radeong: flush CS if a buffer being mapped is referenced by itMarek Olšák
Also, overlapping occlusion queries seems to work now.
2009-12-04r300g: fix warningsMarek Olšák
2009-11-30fixups for interface changes (mostly state trackers)Roland Scheidegger
2009-11-26gallium: fix more statetrackers/drivers for not using texture ↵Roland Scheidegger
width/height/depth arrays
2009-10-29r300g: Fix bytes_per_line calculation error while displaying surfaceCooper Yuan
2009-10-23r300g: last changes's typo, miss a include fileCooper Yuan
2009-10-23r300g: add flush_frontbuffer function to display video surfaceCooper Yuan
2009-09-06r300g: Fix some issues pointed out by compiler warnings.Michel Dänzer
2009-09-06r300g: Handle PIPE_BUFFER_USAGE_DONTBLOCK as intended.Michel Dänzer
Return NULL if the BO is busy, otherwise just map it.
2009-08-18radeon-gallium: Oh, look, we *do* already support DONTBLOCK.Corbin Simpson
Well, okay, the kernel doesn't, but that's no excuse for us! :3
2009-07-26radeon-gallium: If BO allocation fails, return NULL.Corbin Simpson
2009-07-13r300g, radeon: Whitespace fixes.Nicolai Hähnle
Signed-off-by: Corbin Simpson <MostAwesomeDude@gmail.com>
2009-07-05radeon-gallium: Use FLINK to do proper global buffers.Corbin Simpson
2009-07-05radeon-gallium: Wait for BOs before mapping them.Corbin Simpson
2009-05-20radeon-gallium: Add surface_buffer_create callback.Corbin Simpson
2009-05-01r300-gallium, radeon-winsys: Hide radeon_cs from r300 pipe.Corbin Simpson
2009-05-01r300-gallium, radeon-winsys: Space accounting.Corbin Simpson
It is no longer optional in current libdrm, so it was time to actually start counting our BOs.
2009-04-04radeon: A bit of indent and line break cleanup.Corbin Simpson
2009-03-10radeon-gallium: Unbreak non-RADEON_SOFTPIPE.Corbin Simpson
Hehe, sleep may be necessary now.
2009-03-10radeon-gallium: Unbreak RADEON_SOFTPIPE.Corbin Simpson
And there was much rejoicing.
2009-03-04radeon-gallium: DRI2 state tracker, part 3.Corbin Simpson
Properly setup the build process for adding in xorg and egl trackers, and finally put radeon_context and radeon_screen to bed.