summaryrefslogtreecommitdiffstats
path: root/src/gallium
Commit message (Collapse)AuthorAgeFilesLines
* gallivm: Add fallthrough comment at end of case statement.Vinson Lee2010-04-021-0/+1
| | | | | The code is correct. Tell Coverity that the fallthrough case is intentional.
* gallium/util: print \n after DXTn printfLuca Barbieri2010-04-021-3/+3
|
* softpipe: remove S3TC init, since it's done on-demand nowLuca Barbieri2010-04-021-2/+0
|
* gallium/util: remove extraneous semicolonBrian Paul2010-04-011-1/+1
|
* gallium: remove old comment about return valuesBrian Paul2010-04-011-1/+1
|
* gallium/util: revert util_format_init additionLuca Barbieri2010-04-024-13/+1
| | | | | | | | | | Putting calls to util_format_init all over the codebase is infeasible. Instead, half float tables are pregenerated, and the s3tc library is loaded on demand. I believe this is a solution that combines performance, cleanliness, flexibility and portability.
* gallium/util: add util_format_is_supported to check for pack/unpackLuca Barbieri2010-04-024-25/+39
| | | | | This improves the code by making it more readable, and removes special knowledge of S3TC and other formats from softpipe.
* gallium/util: load s3tc on demandLuca Barbieri2010-04-022-107/+151
| | | | | | | | | | | | | | | | This changes the S3TC function pointers to be initialized to stubs that load the S3TC library and then delegate to the real functions. If the S3TC library fails to load, the function pointers are replaced with a "nop" function. The code is also changed to attempt to load the library only one time.c Note that unlike checking for a flag, this method has no performance cost at all. The use of the "nop" functions also allows to avoid most checks, that are only preserved when the function does non-trivial work.
* gallium/util: pregenerate half float tablesLuca Barbieri2010-04-026-170/+191
| | | | | | | | This solution avoids the issue of how to run the initializers and also allows those pages (and the parts of them in processor caches) to be shared between multiple processes. The drawback is slightly higher library size.
* st/egl: Add a missing break.Chia-I Wu2010-04-021-0/+1
| | | | The swtich in egl_g3d_st_framebuffer_validate misses a break.
* gallium/util: s/inline/INLINE/Vinson Lee2010-04-011-1/+1
| | | | Fixes MSVC build.
* nv50: call util_format_initLuca Barbieri2010-04-021-0/+3
| | | | Needed to fetch static vertex attributes.
* gallium/util: add util_format_init that inits s3tc and util_halfLuca Barbieri2010-04-025-6/+23
| | | | | | | | Switch from auto-init to explicit init for util_half per Brian Paul's indication. NOTE: this is probably broken because not enough things call util_format_init. Will be fixed shortly
* gallium/util: add one-time initialization helperLuca Barbieri2010-04-021-0/+15
|
* gallium/util: use #pragma section instead of #pragma data_segLuca Barbieri2010-04-021-1/+1
| | | | | | | | | | | They apparently both declare the section, but #pragma data_seg also puts all subsequent definitions in the section, which is undesirable. This should be the correct solution, and is actually used by the reference I cited (but I forgot to do it in my code). Untested, let me know if it doesn't work.
* gallium/util: reindent u_half.c and u_half.h with Mesa coding styleLuca Barbieri2010-04-022-107/+112
| | | | Sorry, forgout about that.
* Revert "util: Init half-float tables on demand."Luca Barbieri2010-04-023-14/+57
| | | | This reverts commit 950300eb255f0e3507bf2757d16c3b5bc8ff3471.
* util: Init half-float tables on demand.Corbin Simpson2010-04-013-57/+14
| | | | Gets rid of unnecessary delays on startup and compiler-specific hax.
* llvmpipe: Support sampling from PIPE_FORMAT_R32_FLOAT.José Fonseca2010-04-012-4/+45
|
* util: Declare .CRT$XCU data segment.Michal Krol2010-04-011-0/+1
|
* scons: Remove `util/u_gctors.cpp'.Michal Krol2010-04-011-1/+0
|
* gallium/util: rewrite global constructor system for half floats (GCC/MSVC only!)Luca Barbieri2010-04-015-26/+55
| | | | | | | | | | | | | | | | NOTE: this commit will cause Gallium to fail to build on any compiler except GCC, the Microsoft C compiler and compatible compilers that claim to be one of those. This commit removes the u_gctors.cpp mechanism, in favor of using compiler-specific syntax to add global constructors from C files. This solves the problem of u_gctors.o not being pulled from static libraries and avoids using C++. However, it needs compiler-specific support for every compiler. The Microsoft C compiler support has not been tested.
* llvmpipe: Support sampling from signed and mixed siged formats.José Fonseca2010-04-011-9/+64
|
* st/python: Allow to sample only in the [0,1] range.José Fonseca2010-04-013-23/+46
|
* st/python: Always use softpipe as reference driver.José Fonseca2010-04-014-11/+17
|
* gallium/util: add copyright header to u_half.cLuca Barbieri2010-04-011-2/+37
|
* llvmpipe: More tweaks to the supported texture formats.José Fonseca2010-04-011-10/+6
|
* util: Declare util_half_init_tables only once.José Fonseca2010-04-013-4/+15
|
* util: Use u_math.h's union fi instead of redefining it.José Fonseca2010-04-011-4/+5
|
* util: Add support for other DXTn RGBA formats.José Fonseca2010-04-012-22/+298
|
* util: Get DXT1_RGB format working correctly.José Fonseca2010-04-011-1/+54
|
* util: Add test case for PIPE_FORMAT_DXT1_RGB.José Fonseca2010-04-011-0/+32
|
* llvmpipe: Fix build...José Fonseca2010-04-011-0/+23
|
* util: Allow to have block format test casesJosé Fonseca2010-04-012-449/+458
|
* util: Generate correct format conversions for half floats.Michal Krol2010-04-011-39/+59
|
* util: Use u_half to perform half <--> float conversions.Michal Krol2010-04-011-47/+6
|
* gallium: Integrate util_half with scons.Michal Krol2010-04-012-0/+3
|
* gallium/util: add fast half float conversion functionsLuca Barbieri2010-04-014-0/+199
| | | | | | | | | | | | | | | | | 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.
* util: Add half float test cases.José Fonseca2010-04-011-0/+35
|
* llvmpipe: Fix (un)swizzling, broken due to use of VOID channels.José Fonseca2010-04-011-2/+10
|
* r300g: Remove unnecessary header.Vinson Lee2010-03-311-1/+0
|
* util/format: fix big endian compilation by not trying to byteswap single bytesLuca Barbieri2010-04-011-7/+9
| | | | | Conceivably, we could also have a dummy util_bswap8, but it seems better to not emit it in the first place.
* gallivm: convert floats to doublesZack Rusin2010-03-311-2/+9
| | | | | printf can't handle floats, convert them to doubles so that we can actually print floats.
* r300g: add RGBA16F colorbuffer supportMarek Olšák2010-04-011-5/+8
| | | | Disabled by default due to unresolved IP issues.
* r300g: add RGTC texture supportMarek Olšák2010-04-013-1/+56
| | | | The CS checker already knows about this.
* r300g: format handling cleanupMarek Olšák2010-04-012-24/+16
|
* softpipe: Use S3TC when avaiilable,José Fonseca2010-03-311-5/+8
|
* util: Hook into libtxc_dxtn.so (WIP).José Fonseca2010-03-316-5/+738
|
* llvmpipe: Don't call unused generate_clamp().José Fonseca2010-03-311-2/+0
|
* libgl-xlib: Fix missing X11 symbols.José Fonseca2010-03-311-0/+2
|
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-14 00:23:10 +0000