summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2010-04-27r300g: reserve CS space for index offsetMarek Olšák
This has caused CS overflow since the commit 671f1e1229877d459cfd76e2cf5ea89c8d881036. TA3D works here again.
2010-04-26llvmpipe: Remove unused variable.Vinson Lee
2010-04-27r300g: add support for more unaligned vertex formatsMarek Olšák
An aligned stride is still needed for it to work. I am slowly approaching the Sauerbraten milestone!
2010-04-27r300g: setup PSC first, then align vertex formatsMarek Olšák
This allows us to put correct swizzles for undefined components in PSC.
2010-04-27r300g: set undefined vertex swizzles to (0,0,0,1) respectivelyMarek Olšák
2010-04-26draw/llvm: implement fetch elts pathsZack Rusin
we were only running the llvm paths when the input elts were linear, now we can handle abritrary fetch elts arrays. we do this by generating two paths - linear and fetch_elts one and just selecting the right one at run time.
2010-04-26r600: add support for more rendering formatsAlex Deucher
2010-04-26r300g: disable point sprites (again)Marek Olšák
They cause lockups. The users doesn't seem to like them.
2010-04-26r300g: fix warnings when printing uint64_tMarek Olšák
Though not nice, this is the correct way.
2010-04-26r300g: fix warnings by using the const qualifierMarek Olšák
See also the libdrm commit af98ccf4dd5dcb1b904ec32b9bd1521e6bf7dda5.
2010-04-26r300g: use CS_OUT_TABLE betterMatt Turner
Reviewed-by: Corbin Simpson <MostAwesomeDude@gmail.com> Signed-off-by: Matt Turner <mattst88@gmail.com> Signed-off-by: Marek Olšák <maraeo@gmail.com>
2010-04-26r600: avoid setting invalid bit on r7xx for blitsAlex Deucher
2010-04-26gallivm: BGNFOR/ENDFOR fallthrough to BGNLOOP/ENDLOOPAlan Hourihane
2010-04-26draw: Always use the llvm middle end when available & enabled.José Fonseca
2010-04-26svga: Pass-through max_index to translate.José Fonseca
2010-04-26draw: Pass-through pipe_buffer::max_index to translate.José Fonseca
max_index must be observed to prevent crashes due to bad index data. I've been using this patch for some time without regressions. Some places, where we use internal vertex buffer, it is not entirely clear what max_index should be, so passing just ~0 to avoid regressions for now.
2010-04-26translate: Take and respect a max_index argument.José Fonseca
2010-04-26r300g: Rudimentary stats, printed on exit.Corbin Simpson
Whoo!
2010-04-26r300g: Add stats debug flag.Corbin Simpson
2010-04-26r600: enable VERT_RESULT_PSIZ - makes point size & attenuation workAndre Maasikas
doc additions: shader export ARRAY_BASE for EXPORT_POS: 60 is position, 61 is misc vec(VS_OUT_MISC_VEC - used here), 62, 63 are clip distance vectors(VS_OUT_CCDIST#) sorry for formating - there seem to be so many different styles in r600
2010-04-26r600: adjust point sprites after 911fa4a4a1Andre Maasikas
there's no more vp results for point coords so we cannot iterate over vp outputs. Use only Point.CoordReplace[i]
2010-04-26scons: Include headers from the local include directory first.Vinson Lee
Certain headers, such as GL/glew.h, are in both the Mesa include and the default installed include directories. On recent distros the needed symbols can be found in both places. On older distros the installed headers could be lacking symbols, so for a header that exists in both places, the local one should be found first.
2010-04-26os: Check for spurious wakeups in pipe_barrier_wait.Vinson Lee
The POSIX function pthread_cond_wait can have spurious wakeups when waiting on a condition variable. Add a 64-bit counter that is incremented whenever the barrier becomes full. A woken thread checks the counter. If the counter has not changed then it has been spuriously woken and goes back to sleep. If the counter has changed then it was properly signaled and exits the barrier. Tested on Mac OS X. This patch was based on ideas from Luca Barbieri.
2010-04-26r300g: Use table CS writing when applicable.Corbin Simpson
I just broke the 2000 fps barrier on glxgears. Yay?
2010-04-26r300g: Add CS table writing.Corbin Simpson
2010-04-26et/egl: Remove dead code.Vinson Lee
proc_name cannot be NULL after the switch statement. proc_name is assigned a value or the function has already returned.
2010-04-25nvfx: Add include path to SCons build.Vinson Lee
2010-04-26r300g: support index bias on r500 and DRM 2.3.0Marek Olšák
2010-04-26r300g: support rendering more than 65535 vertices in one batch on r500Marek Olšák
and disable batch splitting on these chipsets.
2010-04-26r300g: emit MSPOS regsMarek Olšák
2010-04-26r300g: force prefetch for non-indexed verticesMarek Olšák
2010-04-26r300g: use the dummy FS shader for shaders with zero instructionsMarek Olšák
2010-04-26r300g: allow unaligned vertex formats if the stride is dword-alignedMarek Olšák
2010-04-26r300g: abort on unsupported vertex formatsMarek Olšák
2010-04-26r300g: abort if a vertex buffer offset is not dword-alignedMarek Olšák
2010-04-25st/wgl: s/st_manager_create_api/st_gl_api_create/Vinson Lee
commit 0c572c6828b6a338b07a6860280b3a314a81662e replaced st_manager_create_api with st_gl_api_create. Fixes MSVC build.
2010-04-25nvfx: Add missing header.Vinson Lee
2010-04-26st/dri: Refactor dri_st_api into other filesJakob Bornecrantz
2010-04-26st/dri: Make st_framebuffer_iface the base for dri_drawableJakob Bornecrantz
2010-04-26st/dri: Make st_manager the base for dri_screenJakob Bornecrantz
2010-04-26st/dri: Add hooks for framebuffer functionsJakob Bornecrantz
2010-04-26st/dri: Make lookup_egl_image a hookJakob Bornecrantz
2010-04-26st_api: Remove st_moduleJakob Bornecrantz
The struct st_module isn't needed as it is the same thing as the st_api struct. That is they both represent the API. Instead just use a single function entry point to the the API.
2010-04-26llvmpipe: Respect pipe_sampler_view::swizzle_r/g/b/aJosé Fonseca
This allows u_sampler_view_default_dx9_template to do its magic on DX9.
2010-04-25draw: Respect pipe_vertex_buffer::max_index in llvm generated code.José Fonseca
Everybody should respect max_index, specially llvm generated code, which likes to eat vertices 4 at a time, so it may end up chew a bit a bit more than actually exists.
2010-04-25llvmpipe: No need to flush the caches for buffers.José Fonseca
2010-04-25softpipe: Make softpipe transfers in-order.José Fonseca
Transfer, being now a context operation, should happen in order with all other contexts operations. If there is rendering pending on the resource then the driver must flush and potentially wait itself internally. Instead of avoiding using transfers internally (as done in llvmpipe) I've opted to simply pass PIPE_TRANSFER_UNSYNCHRONIZED in all internal transfers, to avoid infinite recursion.
2010-04-25llvmpipe: Cleanup/improve llvmpipe_flush_resource usage.José Fonseca
Recognize PIPE_TRANSFER_UNSYNCHRONIZED and PIPE_TRANSFER_DONTBLOCK.
2010-04-25llvmpipe: Remove unused variable.Vinson Lee
2010-04-25llvmpipe: llvmpipe_flush_texture -> llvmpipe_flush_resourceJosé Fonseca