summaryrefslogtreecommitdiff
path: root/src
AgeCommit message (Collapse)Author
2009-11-24util: Describe a few more formats.José Fonseca
2009-11-24llvmpipe: Use assert instead of abort. Only verify functions on debug builds.José Fonseca
2009-11-24llvmpipe: Fix memory leak.José Fonseca
2009-11-24llvmpipe: Use the generic conversion routine for depths.José Fonseca
This allows for z32f depth format to work correctly.
2009-11-24llvmpipe: Be more conservative with the supported formats.José Fonseca
We'll likely support much more formats, but doing this allows to run more testsuites without immediately hit assertion failures.
2009-11-24wgl: Fix copy'n'paste typo in comment.José Fonseca
2009-11-24llvmpipe: Fix typo in comparison operator.José Fonseca
2009-11-24tgsi: Document Declaration Semantic token and FACE semantic name.Michal Krol
2009-11-24slang: Fix allocation size.Michal Krol
We don't need 16K+ to store a single pointer.
2009-11-24mesa: Fix pointer arithmetic.Michal Krol
2009-11-24tgsi: Implement predicated instructions in exec.Michal Krol
2009-11-24tgsi: Add ureg_DECL_loop().Michal Krol
2009-11-24tgsi: Account for gallium shader token representation changes.Michal Krol
2009-11-24gallium: Refactor the instruction predicate TGSI token.Michal Krol
Rename it to tgsi_instruction_predicate -- it's no longer an extended token. Its presence is indicated by a new flag in tgsi_instruction that indicates whether an instruction is predicated. Also, change predicate index representation to match the other tokens that specify register indices.
2009-11-24r300: fix swtcl bo leak problem.Dave Airlie
We can get a lot of swtcl bo allocations - need to probably abstract this a bit further. Signed-off-by: Dave Airlie <airlied@redhat.com>
2009-11-24radeon/r200/r300/r600: make bo mapping be explicitDave Airlie
This moves the bo mapping outside the DMA layer and makes it explicit, this should in theory make it simpler to split the clean up the dma/cmdbuf linkage that I created before that is broken. Tested on: r600, rv380 (tcl/no-tcl), rv200 (tcl/no-tcl) Signed-off-by: Dave Airlie <airlied@redhat.com>
2009-11-23mesa: use gcc __builtin_popcount()Brian Paul
2009-11-23r300g: use util_bitcount()Brian Paul
2009-11-23egl: use util_bitcount()Brian Paul
2009-11-23gallium/util: added util_bitcount()Brian Paul
2009-11-23Merge commit 'origin/mesa_7_7_branch'Maciej Cencora
2009-11-23radeon: fix errors in miptree related functionMaciej Cencora
- typo - memory leak - off by one (spotted by airlied)
2009-11-23svga: Scrub Makefiles a bitJakob Bornecrantz
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.
2009-11-23slang: Check return value from emit_instruction().Michal Krol
2009-11-23tgsi: Clamp the source argument in micro_exp2() to avoid Inf.Michal Krol
2009-11-23tgsi: Remove code that actually had no effect.Michal Krol
2009-11-23softpipe: Initialise TGSI machine's Face.Michal Krol
2009-11-23tgsi: Fix POSITION and FACE fragment shader inputs.Michal Krol
2009-11-23tgsi: Bring BGNFOR/ENDFOR implementation up to spec.Michal Krol
2009-11-23tgsi: Fake TXD implementation in exec.Michal Krol
2009-11-23tgsi: Add execution debugging facilities to exec.Michal Krol
2009-11-23slang: Be more robust with memory in concat_shaders().Michal Krol
2009-11-23mesa: Initialize variable in _mesa_get_texel_fetch_func.Vinson Lee
2009-11-23glx: Prevent potential null pointer deference in driCreateContext.Vinson Lee
2009-11-23glu/sgi: Fix memory leak in bitmapBuild2DMipmaps.Vinson Lee
2009-11-23glu/sgi: Fix memory leak in gluScaleImage3D.Vinson Lee
2009-11-22r600 : add support for shader instruction trunc and discard.Richard Li
2009-11-23r600: hopefully fix segfault.Dave Airlie
2009-11-23r600: fix inline issuesDave Airlie
2009-11-22r600 : add stack depth calculation, enable CF pop.Richard Li
2009-11-22r600 : use cf for all pop now, left optimization for future.Richard Li
2009-11-22r300: fix VP source conflict resolution on 64-bit machinesMaciej Cencora
On 32bit machines we were lucky because the sizeof(reg) == sizeof(rc_src_register). On 64bit machines pointers are 8 bytes long, so we were overwriting other data.
2009-11-22r300: fix SIN/COS/SCS instructions for R300 fpMaciej Cencora
2009-11-22glu/sgi: Fix memory leak in gluScaleImage.Vinson Lee
2009-11-22glu: Fix memory leak in __gl_meshMakeEdge.Vinson Lee
2009-11-22svga: Prevent potential null pointer deference in vmw_surface.c.Vinson Lee
2009-11-22svga: Fix memory leak in vmw_screen_ioctl.cVinson Lee
2009-11-22st/xorg: Prevent potential null pointer deference in xorg_xv.c.Vinson Lee
2009-11-22st/xorg: Reorder cases in switch statement.Vinson Lee
Silences missing break statement warning.
2009-11-21r300g: Build fix.Corbin Simpson
Oops.