summaryrefslogtreecommitdiff
path: root/generic
AgeCommit message (Expand)AuthorFilesLines
2016-06-1764 bit integers are legal in full profile without an extensionJan Vesely1-5/+12
2016-05-17math: Use single precision fmax in sp pathJan Vesely1-1/+1
2016-05-06math: Add erf ported from amd-builtinsJan Vesely4-0/+413
2016-05-06math: Add fdim implementationAaron Watry6-0/+86
2016-02-24math: Fix ilogb(double) return typeAaron Watry1-1/+1
2016-02-23math: Add ilogb ported from amd-builtinsAaron Watry6-0/+68
2016-02-09math: Fix log2 vectorization on non-fp64 hwJan Vesely1-0/+2
2016-02-08math: Add frexp ported from amd-builtinsAaron Watry7-0/+151
2016-01-27Implement modf math builtinTom Stellard6-0/+120
2016-01-27Add _CLC_V_V_VP_VECTORIZE macroTom Stellard1-0/+22
2015-10-06integer: remove explicit casts from _MIN definitionsAaron Watry1-3/+3
2015-09-29Implement tanh builtinNiels Ole Salscheider5-0/+195
2015-09-21Add sampler defines.Tom Stellard1-0/+18
2015-09-21Add image attribute defines.Tom Stellard2-0/+32
2015-09-21r600: Add image writing builtins.Tom Stellard1-0/+7
2015-09-21r600: Add image reading builtins.Tom Stellard1-0/+13
2015-09-21Add image attribute getter builtinsTom Stellard4-0/+30
2015-09-15integer: Update integer limits to comply with specAaron Watry1-6/+6
2015-08-13Remove files accidentally not removed in r244310Jeroen Ketema2-9/+0
2015-07-24Fix double implementation of logTom Stellard4-3/+73
2015-07-24Implement accurate log2 functionTom Stellard8-6/+517
2015-07-24Use llvm intrinsics for native_log and native_log2Tom Stellard9-2/+216
2015-07-10Fix implementation of sqrt v2Tom Stellard8-6/+168
2015-05-13Use a more accurate implementation for expTom Stellard2-13/+85
2015-05-13Implement exp2 using OpenCL C rather than using an intrinsicTom Stellard8-6/+303
2015-05-12Implement sin for double typesTom Stellard1-7/+16
2015-05-12Implement cos for double typesTom Stellard5-7/+289
2015-05-12Implement atan2pi builtinTom Stellard5-0/+270
2015-05-12Implement atan2 for doublesTom Stellard3-2/+412
2015-05-09math: limit half_sqrt to single precisionJan Vesely2-6/+4
2015-05-09geometric: Limit fast_{distance,length} functions to single precisionJan Vesely4-25/+6
2015-05-09Fix ldexp fp64 build errorJan Vesely1-1/+1
2015-05-09Implement fast_normalize builtin v4Tom Stellard7-0/+125
2015-05-08Implement half_rsqrt builtin v3Tom Stellard6-0/+88
2015-05-06Move ldexp soft implementation to a separate fileJan Vesely4-100/+138
2015-05-06Implement sinpi builtinJan Vesely5-0/+137
2015-05-06math: Add ldexp implementationTom Stellard4-0/+172
2015-05-06Implement ldexp for R600/SITom Stellard4-0/+74
2015-05-06Fix implementation of normalize builtinTom Stellard2-6/+152
2015-04-29Allow compilation depending to the LLVM versionTom Stellard2-0/+9
2015-04-24Fix compilation warnings without cl_khr_fp64Jan Vesely3-6/+32
2015-04-23Implement fract builtinTom Stellard6-0/+130
2015-04-20configure: Add --enable-runtime-subnormal optionTom Stellard6-0/+75
2015-04-07Implement atanh builtinTom Stellard5-0/+162
2015-04-07Implement acosh builtinTom Stellard5-0/+176
2015-04-02Implement atanpi builtinTom Stellard5-0/+231
2015-04-02Implement asinpi builtinTom Stellard5-0/+219
2015-04-02Implement asinh builtinTom Stellard7-0/+466
2015-04-02Implement acospi builtinTom Stellard5-0/+221
2015-03-31Implement fmax using __builtin_fmaxTom Stellard3-8/+28
generated by cgit v1.2.3 (git 2.39.1) at 2024-05-15 15:32:20 +0000