Commits · ba04d3a620fe70bc7b29dd91e0bdfe5158046b47 · Roger Ferrer / llvm-epi-0.8

Browse files

Jul 26, 2016
- Make min follow the OCL 1.0 specs · 4c59714a
  Jan Vesely authored Jul 25, 2016
  4c59714a
Jul 23, 2016
- Implement cbrt builtin · d835b3f1
  Tom Stellard authored Jul 22, 2016
  d835b3f1
- Implement cosh builtin · 9cb070f9
  Tom Stellard authored Jul 22, 2016
  9cb070f9
- geometric/floatn.inc: Add vec8 and vec16 types · ff13926a
  Tom Stellard authored Jul 22, 2016
  ff13926a
Jul 22, 2016
- AMDGPU: Implement get_global_offset builtin · a82e080b
  Jan Vesely authored Jul 22, 2016
  a82e080b
- AMDGPU: Use clang intrinsics for workitem builtins · 74f02db9
  Jan Vesely authored Jul 22, 2016
  74f02db9
- ptx: Fix builtin names after clang r274770 · 7846c9b8
  Jan Vesely authored Jul 22, 2016
  7846c9b8
Jul 19, 2016
- amdgpu: Use right builtn for rsq · 633d749d
  Matt Arsenault authored Jul 19, 2016
  633d749d
Jul 18, 2016
- R600: Use new barrier intrinsic · 1ab0d9c1
  Matt Arsenault authored Jul 18, 2016
  1ab0d9c1
- Replace llvm.AMDGPU.ldexp with llvm.amdgcn.ldexp · b456c6dd
  Matt Arsenault authored Jul 18, 2016
  b456c6dd
Jun 17, 2016
- configure: Remove device specific defines · e97deffb
  Jan Vesely authored Jun 17, 2016
  e97deffb
- nvptx: Drop feature defines. · 5fd84d02
  Jan Vesely authored Jun 17, 2016
  5fd84d02
- 64 bit integers are legal in full profile without an extension · 3317f253
  Jan Vesely authored Jun 17, 2016
  3317f253
May 17, 2016
- math: Use single precision fmax in sp path · 973c1fa5
  Jan Vesely authored May 17, 2016
  973c1fa5
May 06, 2016
- math: Add erf ported from amd-builtins · c374cb76
  Jan Vesely authored May 06, 2016
  c374cb76
- math: Add fdim implementation · 55a8e0fd
  Aaron Watry authored May 06, 2016
  55a8e0fd
Apr 15, 2016
- prepare-builtins: Remove call to getGlobalContext() · 6cb18a09
  Tom Stellard authored Apr 15, 2016
  6cb18a09
Apr 07, 2016
- [AMDGPU] Implement get_local_size for amdgcn--amdhsa triple · f8a81f86
  Konstantin Zhuravlyov authored Apr 07, 2016
  f8a81f86
Mar 31, 2016
- Update copyright year to 2016. · 36a6eaeb
  Paul Robinson authored Mar 30, 2016
  36a6eaeb
Feb 24, 2016
- math: Fix ilogb(double) return type · 09f3c99a
  Aaron Watry authored Feb 24, 2016
  09f3c99a
Feb 23, 2016
- math: Add ilogb ported from amd-builtins · d6d04542
  Aaron Watry authored Feb 23, 2016
  d6d04542
Feb 17, 2016
- Add .gitignore for build directories · 34766ca3
  Matt Arsenault authored Feb 17, 2016
  34766ca3
- amdgcn: Use new workitem intrinsics · 45e6eaaa
  Matt Arsenault authored Feb 17, 2016
  45e6eaaa
Feb 13, 2016
- Update page to list supported targets · d4cd67ab
  Matt Arsenault authored Feb 13, 2016
  d4cd67ab
- Split sources for amdgcn and r600 · a48e15c6
  Matt Arsenault authored Feb 13, 2016
  a48e15c6
Feb 09, 2016
- configure: Remove llvm 3.6 defines · 6d870d2e
  Jan Vesely authored Feb 09, 2016
  6d870d2e
- configure: Remove cl_khr_fp64 for device that don't support doubles · 46b7dd99
  Jan Vesely authored Feb 09, 2016
  46b7dd99
- configure: Introduce per device defines · 4f38142f
  Jan Vesely authored Feb 09, 2016
  4f38142f
- math: Fix log2 vectorization on non-fp64 hw · 7fbb96b9
  Jan Vesely authored Feb 09, 2016
  7fbb96b9
Feb 08, 2016
- math: Add frexp ported from amd-builtins · 8872800e
  Aaron Watry authored Feb 08, 2016
  8872800e
Jan 27, 2016
- Implement modf math builtin · 37d19875
  Tom Stellard authored Jan 27, 2016
  37d19875
- Add _CLC_V_V_VP_VECTORIZE macro · a249f509
  Tom Stellard authored Jan 27, 2016
  a249f509
Dec 15, 2015
- AMDGPU: Add aliases for all VI targets · 0abd2814
  Tom Stellard authored Dec 15, 2015
  0abd2814
- AMDGPU: Add alias for tonga · 655680a2
  Tom Stellard authored Dec 15, 2015
  655680a2
Oct 06, 2015
- integer: remove explicit casts from _MIN definitions · 23faa5a1
  Aaron Watry authored Oct 06, 2015
  23faa5a1
Sep 29, 2015
- Implement tanh builtin · f51df5ba
  Niels Ole Salscheider authored Sep 29, 2015
  f51df5ba
Sep 21, 2015
- Add sampler defines. · e2bab44c
  Tom Stellard authored Sep 21, 2015
  e2bab44c
- Add image attribute defines. · 50dfd445
  Tom Stellard authored Sep 21, 2015
  50dfd445
- r600: Add image writing builtins. · a59fd49b
  Tom Stellard authored Sep 21, 2015
  a59fd49b
- r600: Add image reading builtins. · 9a7d4a94
  Tom Stellard authored Sep 21, 2015
  9a7d4a94