summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2010-04-01util: Allow to have block format test casesJosé Fonseca
2010-04-01util: Generate correct format conversions for half floats.Michal Krol
2010-04-01util: Use u_half to perform half <--> float conversions.Michal Krol
2010-04-01gallium: Integrate util_half with scons.Michal Krol
2010-04-01gallium/util: add fast half float conversion functionsLuca Barbieri
This adds a fast half float conversion facility to Gallium. Mesa already contains such a facility, but using a much worse algorithm. This one is an implementation of www.fox-toolkit.org/ftp/fasthalffloatconversion.pdf and uses a branch-less algorithm with some lookup tables small enough to fit in the L1 cache. Ideally, Mesa should start using these functions too, but I'm not sure how to arrange that with the current build system. A new "u_gctors.cpp" is added that defines a global C++ constructor allowing to initialize to conversion lookup tables at library init.
2010-04-01gallium/unit: Install test in build/xxx/binJosé Fonseca
2010-04-01util: Add half float test cases.José Fonseca
2010-04-01llvmpipe: Fix (un)swizzling, broken due to use of VOID channels.José Fonseca
2010-03-31mesa: Remove unnecessary header.Vinson Lee
2010-03-31r300g: Remove unnecessary header.Vinson Lee
2010-04-01util/format: fix big endian compilation by not trying to byteswap single bytesLuca Barbieri
Conceivably, we could also have a dummy util_bswap8, but it seems better to not emit it in the first place.
2010-03-31gallivm: convert floats to doublesZack Rusin
printf can't handle floats, convert them to doubles so that we can actually print floats.
2010-04-01r300g: add RGBA16F colorbuffer supportMarek Olšák
Disabled by default due to unresolved IP issues.
2010-04-01r300g: add RGTC texture supportMarek Olšák
The CS checker already knows about this.
2010-04-01r300g: format handling cleanupMarek Olšák
2010-03-31softpipe: Use S3TC when avaiilable,José Fonseca
2010-03-31util: Hook into libtxc_dxtn.so (WIP).José Fonseca
2010-03-31llvmpipe: Don't call unused generate_clamp().José Fonseca
2010-03-31libgl-xlib: Fix missing X11 symbols.José Fonseca
2010-03-31util: Make the accessors bidimensional again.José Fonseca
Otherwise there's no way to unpack blocks with height >1
2010-03-31util: First stab at half-float conversion.Michal Krol
2010-03-31util: Describe PIPE_FORMAT_NONE as PIPE_FORMAT_R8_USCALED.José Fonseca
Avoids the need to special case PIPE_FORMAT_NONE so often. Conflicts: src/gallium/auxiliary/util/u_format_table.py
2010-03-31util: Silence pointer to integer size mismatch errors.José Fonseca
2010-03-31util: Add callback to fetch a single pixel.José Fonseca
2010-03-31python/tests: Read the number of tests to execute from command line.José Fonseca
2010-03-31mesa: Use a consistent name of the external s3tc library for all windows ↵José Fonseca
compilers.
2010-03-31scons: Force to consider the util/u_format_pack.py for util/u_format_table.c.José Fonseca
2010-03-31scons: Don't always include/links X11 headers/libs.José Fonseca
Move it into a separate tool. TODO: Needs to be "tooled" in each SConscript that uses it.
2010-03-31scons: Don't always link expat.José Fonseca
Only needed for DRI. Should be linked in DRI specific subdirectories.
2010-03-31llvmpipe: Disable threads by default on embedded.José Fonseca
2010-03-31os: Temporarily use posix thread for embedded too.Kurt Daverman
2010-03-31scons: Allow any toolchain to be specified in the command line.José Fonseca
2010-03-31util: Remove the hand written u_tile.c functions for the ones that are code ↵José Fonseca
generated correctly. Not that the code generated are any better, but to quickly detect any regression in the code generated ones from now on.
2010-03-31util: Use u_format_pack.py's code instead of u_format_access.py.José Fonseca
2010-03-31llvmpipe: Don't rely on u_format_access.pyJosé Fonseca
2010-03-31Merge branch 'gallium-new-formats'Roland Scheidegger
Conflicts: src/gallium/auxiliary/util/u_format.csv src/gallium/auxiliary/util/u_format_access.py src/gallium/auxiliary/util/u_format_pack.py
2010-03-31util: Put the format pack/unpack functions in the description table.José Fonseca
2010-03-31util: Make pack/unpack operate on spans.José Fonseca
2010-03-31st/mesa: use BITFIELD64_BIT to access shader OutputsWritten in more placesBen Skeggs
2010-03-31util: Make util_format_xxx_pack_xxx take pointer as arguments.José Fonseca
2010-03-31util: Fix PIPE_FORMAT_B5G5R5X1_UNORM test case.José Fonseca
2010-03-31util: Use VOID type for X channels in format descriptions.José Fonseca
2010-03-31python/tests: Run trhough tests randomly.José Fonseca
2010-03-31gallivm: Fix USCALED translation. Minor Cleanups.José Fonseca
2010-03-31llvmpipe: More accurate format capability exporting.José Fonseca
2010-03-31llvmpipe: Drop the aos format conversion.José Fonseca
It's unused and incomplete. Still in git history if necessary in future.
2010-03-31util: Generalize lp_format_is_rgba8 into util_format_is_rgba8_variant.José Fonseca
2010-03-31python/tests: Rename test name.José Fonseca
We only test blitting from a surface -- we don't test any sampler state.
2010-03-31python/tests: Test all existing gallium formats.José Fonseca
2010-03-31python/tests: Speed up the tests by reusing the same pipe context.José Fonseca