summaryrefslogtreecommitdiff
path: root/src/gallium/auxiliary
AgeCommit message (Expand)AuthorFilesLines
2013-07-20util/u_math: Define NAN/INFINITY macros for MSVC.José Fonseca1-0/+4
2013-07-19gallivm: add a log function that handles edge casesZack Rusin2-0/+21
2013-07-19gallivm: export unordered/ordered cmp to a common functionZack Rusin1-283/+158
2013-07-19gallivm: handle -inf, inf and nan's in sin/cos instructionsZack Rusin2-0/+49
2013-07-19gallivm: add a version of log2 which handles edge casesZack Rusin3-6/+65
2013-07-19gallivm: fix edge cases in exp2Zack Rusin1-3/+7
2013-07-19gallivm: handle nan's in min/maxZack Rusin6-52/+482
2013-07-17gallivm: (trivial) simplify lp_build_cos/lp_build_sin a tiny bitRoland Scheidegger1-7/+6
2013-07-17util/u_format_s3tc: handle srgb formats correctly.Roland Scheidegger2-185/+254
2013-07-17gallium/util: use explicily sized types for {un, }pack_rgba_{s, u}intEmil Velikov2-8/+8
2013-07-17llvmpipe: use MCJIT on ARM and AArch64Kyle McMartin1-1/+1
2013-07-16llvmpipe: support sRGB framebuffersRoland Scheidegger2-4/+54
2013-07-14gallivm: (trivial) use constant instead of exp2f() functionRoland Scheidegger1-2/+3
2013-07-13gallivm: handle srgb-to-linear and linear-to-srgb conversionsRoland Scheidegger6-7/+332
2013-07-13gallivm: better support for fast rsqrtRoland Scheidegger2-16/+63
2013-07-11gallium: fixup definitions of the rsq and sqrtZack Rusin2-13/+8
2013-07-12util/u_format: Comment out half float denormal test case.José Fonseca1-0/+5
2013-07-12gallivm: Eliminate redundant lp_build_select calls.José Fonseca1-12/+2
2013-07-12tgsi: rename the TGSI fragment kill opcodesBrian Paul14-46/+44
2013-07-12tgsi: fix-up KILP commentsBrian Paul2-5/+3
2013-07-12tgsi: exec TGSI_OPCODE_SQRT as a scalar instruction, not vectorBrian Paul1-1/+1
2013-07-12tgsi: use X component of the second operand in exec_scalar_binary()Brian Paul1-1/+1
2013-07-12os: add os_get_process_name() functionBrian Paul3-0/+133
2013-07-12hud: silence some MSVC warningsBrian Paul1-8/+8
2013-07-12util: add casts to silence MSVC warnings in u_blit.cBrian Paul1-14/+14
2013-07-12tgsi: s/unsigned/int/ to silence MSVC warningBrian Paul1-1/+1
2013-07-10util/u_math: Use xmmintrin.h whenever possible.José Fonseca1-9/+17
2013-07-09util: treat denorm'ed floats like zeroZack Rusin4-0/+72
2013-07-05gallivm: (trivial) fix using one lod instead of per-quad lod for texel fetchRoland Scheidegger1-1/+2
2013-07-05gallivm: Remove bogus assert.José Fonseca1-4/+1
2013-07-05gallivm: (trivial) fix bogus assertion for per-element lod with 1d resourcesRoland Scheidegger2-2/+1
2013-07-04gallivm: do per-pixel lod calculations for explicit lodRoland Scheidegger9-125/+193
2013-07-03draw: fix overflows in the indexed rendering pathsZack Rusin4-43/+159
2013-07-03draw/llvm: index overflows if it's greater than elt maxZack Rusin1-1/+1
2013-07-03postprocess: move second temporary assertion into isolated configurationMatthew McClure1-2/+2
2013-07-02gallivm: Simplify intrinsic name construction.José Fonseca1-23/+10
2013-07-02gallivm: Fix MSVC build.José Fonseca1-8/+7
2013-07-02gallivm: Fix indirect immediate registers.José Fonseca1-2/+2
2013-06-28draw/translate: fix instancingZack Rusin13-24/+93
2013-06-28draw: fix incorrect clipper invocation statisticsZack Rusin1-6/+0
2013-06-28draw/gallivm: export overflow arithmetic to its own fileZack Rusin4-44/+234
2013-06-28draw: check for integer overflows in instance computationZack Rusin2-0/+7
2013-06-28draw: check for an integer overflow when computing strideZack Rusin1-10/+43
2013-06-28draw: account for elem size when computing overflowZack Rusin1-7/+23
2013-06-30st/mesa: handle SNORM formats in generic CopyPixels pathMarek Olšák2-0/+23
2013-06-27postprocess: handle partial intialization failures.Matthew McClure7-95/+281
2013-06-26hud: add float casts to silence MSVC warningsBrian Paul1-49/+49
2013-06-26hud: include stdio.h since we use fprintf(), fscanf(), etcBrian Paul1-0/+2
2013-06-26hud: add cast to silence MSVC warningBrian Paul1-1/+1
2013-06-26os: add cast in os_time_sleep() to silence MSVC warningBrian Paul1-1/+1