Commits · 4e72300929d9450f13190cd264d40268af08427c · Roger Ferrer / llvm-epi-0.8

Browse files

Mar 06, 2018
- maxmag: Condition variable needs to be the same bitwidth as operands · 4e723009
  Jan Vesely authored Mar 06, 2018
  4e723009
- Move cl_khr_fp64 exntension enablement to gentype include lists · dbaf6d0f
  Jan Vesely authored Mar 06, 2018
  dbaf6d0f
Feb 06, 2018
- Add vstore_half_rte implementation · 1c570566
  Jan Vesely authored Feb 06, 2018
  1c570566
- Add vstore_half_rtp implementation · f2d876ae
  Jan Vesely authored Feb 06, 2018
  f2d876ae
- Add vstore_half_rtn implementation · 2655312c
  Jan Vesely authored Feb 06, 2018
  2655312c
- Add vstore_half_rtz implementation · d526a2b6
  Jan Vesely authored Feb 06, 2018
  d526a2b6
- vstore_half: Add support for custom rounding functions · 187ec005
  Jan Vesely authored Feb 06, 2018
  187ec005
- vstore_half: Make sure the helper function is always inline · 87036d27
  Jan Vesely authored Feb 06, 2018
  87036d27
Feb 01, 2018
- half_powr: Implement using powr · 3b8b4eb6
  Jan Vesely authored Feb 01, 2018
  3b8b4eb6
Jan 31, 2018
- math.h: Use logical operations instead of bit operations for readability · a75677c2
  Jan Vesely authored Jan 31, 2018
  a75677c2
Jan 29, 2018
- math.h: Set HAVE_HW_FMA32 based on compiler provided macro · 0ecb5e51
  Jan Vesely authored Jan 29, 2018
  0ecb5e51
Jan 19, 2018
- tanpi: Port from amd_builtins · 7013857f
  Jan Vesely authored Jan 19, 2018
  7013857f
- tan: Port from amd_builtins · 03937bde
  Jan Vesely authored Jan 19, 2018
  03937bde
Jan 18, 2018
- half_divide: Implement using x/y · 44e0522c
  Jan Vesely authored Jan 18, 2018
  44e0522c
- half_tan: Implement using tan · 2813b4f8
  Jan Vesely authored Jan 18, 2018
  2813b4f8
- half_sin: Implement using sin · bf38fae8
  Jan Vesely authored Jan 18, 2018
  bf38fae8
- half_recip: Implement using 1/x · 398108b9
  Jan Vesely authored Jan 18, 2018
  398108b9
- half_log2: Implement using log2 · a1aba44f
  Jan Vesely authored Jan 18, 2018
  a1aba44f
- half_log10: Implement using log10 · b3b72af4
  Jan Vesely authored Jan 18, 2018
  b3b72af4
- half_log: Implement using log · 68520238
  Jan Vesely authored Jan 18, 2018
  68520238
- half_exp10: Implement using exp10 · aa4c3899
  Jan Vesely authored Jan 18, 2018
  aa4c3899
- half_exp2: Implement using exp2 · 3c0e19b6
  Jan Vesely authored Jan 18, 2018
  3c0e19b6
- half_exp: Implement using exp · caa9000b
  Jan Vesely authored Jan 18, 2018
  caa9000b
- half_cos: Implement using cos · b5d55606
  Jan Vesely authored Jan 18, 2018
  b5d55606
- half_sqrt: Cleanup implementation · e53ae3b5
  Jan Vesely authored Jan 18, 2018
  e53ae3b5
- half_rsqrt: Cleanup implementation · a95db144
  Jan Vesely authored Jan 18, 2018
  a95db144
Jan 17, 2018
- rootn: Port from amd_builtins · fe8e00bc
  Jan Vesely authored Jan 17, 2018
  fe8e00bc
- powr: Port from amd_builtins · c45ec604
  Jan Vesely authored Jan 17, 2018
  c45ec604
- pown: Port from amd_builtins · 5efc8fe3
  Jan Vesely authored Jan 17, 2018
  5efc8fe3
- pow: Port from amd_builtins · cc5c65b2
  Jan Vesely authored Jan 17, 2018
  cc5c65b2
Nov 15, 2017
- math: Implement minmag · fe7c0457
  Jan Vesely authored Nov 15, 2017
  fe7c0457
- math: Implement maxmag · 7ba243cc
  Jan Vesely authored Nov 15, 2017
  7ba243cc
Nov 14, 2017
- native_powr: Switch implementation to native_exp2 and native_log2 · 383fbd05
  Jan Vesely authored Nov 14, 2017
  383fbd05
Nov 13, 2017
- native_divide: provide function implementation instead of macro · f38b40da
  Jan Vesely authored Nov 13, 2017
  f38b40da
- native_recip: provide function implementation instead of macro · 1b9825f9
  Jan Vesely authored Nov 13, 2017
  1b9825f9
- native_rsqrt: Switch implementation to 1 / native_sqrt · a6758c94
  Jan Vesely authored Nov 13, 2017
  a6758c94
- native_tan: Switch implementation to use native_sin/native_cos · 541a3f07
  Jan Vesely authored Nov 13, 2017
  541a3f07
- math: Use precomputed constant for log2(10.0) · 79b75662
  Jan Vesely authored Nov 13, 2017
  79b75662
Nov 10, 2017
- native_exp10: Switch implementation to llvm intrinsic · 6b4a6254
  Jan Vesely authored Nov 10, 2017
  6b4a6254
- native_sqrt: Switch implementation to llvm intrinsic · 4301e6d0
  Jan Vesely authored Nov 10, 2017
  4301e6d0