Commits · f517f1a5160577f03d03820020a81d85447bc2fe · Roger Ferrer / llvm-epi · GitLab

Jan 14, 2018
- [X86] Implement old kunpck intrinsics using vector ops on vXi1 instead of integer shift/and/or · f517f1a5
  Craig Topper authored Jan 14, 2018
  f517f1a5
Jan 08, 2018
- [X86] Replace cvt*2mask intrinsics with native IR using 'icmp slt X, zeroinitializer. · de91dff5
  Craig Topper authored Jan 08, 2018
  de91dff5
Jan 06, 2018
- Add support for a limited subset of TS 18661-3 math builtins. · dfecbe9a
  Benjamin Kramer authored Jan 06, 2018
  dfecbe9a
Jan 04, 2018
- [CGBuiltin] Handle unsigned mul overflow properly (PR35750) · bbafd507
  Vedant Kumar authored Jan 03, 2018
  bbafd507
Dec 27, 2017
- [x86][icelake][bitalg] · 2268576f
  Coby Tayree authored Dec 27, 2017
  2268576f
Dec 22, 2017
- [X86] Allow _mm_prefetch (both the header implementation and the builtin) to... · 170de4b4
  Craig Topper authored Dec 21, 2017
  170de4b4
Dec 21, 2017
- [AArch64] Enable fp16 data type for the Builtin for AArch64 only. · abb890b7
  Abderrazek Zaafrani authored Dec 21, 2017
  abb890b7
- [AARch64] Add ARMv8.2-A FP16 vector intrinsics · f58a132e
  Abderrazek Zaafrani authored Dec 21, 2017
  f58a132e
- [ubsan] Diagnose noreturn functions which return · 09b5bfdd
  Vedant Kumar authored Dec 21, 2017
  09b5bfdd
Dec 19, 2017
- Silence a bunch of implicit fallthrough warnings · f3b3ccda
  Adrian Prantl authored Dec 19, 2017
  f3b3ccda
Dec 16, 2017
- [X86] Implement kand/kandn/kor/kxor/kxnor/knot intrinsics using native IR. · 5028ace6
  Craig Topper authored Dec 16, 2017
  5028ace6
- [X86] Add builtins and tests for 128 and 256 bit vpopcntdq. · b846d1ff
  Craig Topper authored Dec 16, 2017
  b846d1ff
- [CodeGen] Specialize mixed-sign mul-with-overflow (fix PR34920) · fa5a0e59
  Vedant Kumar authored Dec 16, 2017
  fa5a0e59
Dec 14, 2017
- [CodeGen][X86] Implement _InterlockedCompareExchange128 intrinsic · 627f45fe
  Reid Kleckner authored Dec 14, 2017
  627f45fe
Dec 13, 2017
- [Hexagon] Intrinsic support for V62 and V65 · 5a655838
  Krzysztof Parzyszek authored Dec 13, 2017
  5a655838
Dec 02, 2017
- [CodeGen] fix mapping from fmod calls to frem instruction · 08fba37e
  Sanjay Patel authored Dec 02, 2017
  08fba37e
- [CodeGen] remove stale comment; NFC · 0c0f77d0
  Sanjay Patel authored Dec 02, 2017
  0c0f77d0
- [CodeGen] convert math libcalls/builtins to equivalent LLVM intrinsics · 3e287b4d
  Sanjay Patel authored Dec 01, 2017
  3e287b4d
Nov 30, 2017
- [XRay][clang] Introduce -fxray-always-emit-customevents · 1a5b10d5
  Dean Michael Berris authored Nov 30, 2017
  1a5b10d5
Nov 22, 2017
- [X86] Update CPUSupports code to reuse LLVM .def file [NFC] · 0a340ab3
  Erich Keane authored Nov 22, 2017
  0a340ab3
Nov 15, 2017
- Simplify CpuIs code to use include from LLVM · 8202521c
  Erich Keane authored Nov 15, 2017
  8202521c
Nov 13, 2017
- [CodeGen] fix const-ness of cbrt and fma · 33f83995
  Sanjay Patel authored Nov 13, 2017
  33f83995
Nov 09, 2017
- Fix a bug with the use of __builtin_bzero in a conditional expression. · 26d55e03
  John McCall authored Nov 09, 2017
  26d55e03
Nov 07, 2017
- [NVPTX] Implement __nvvm_atom_add_gen_d builtin. · da9e0bd3
  Justin Lebar authored Nov 07, 2017
  da9e0bd3
Nov 06, 2017
- [X86] Replace the mask cmpeq/cmple/cmplt/cmpgt/cmpge/cmpneq intrinsics with... · 57f96ac6
  Craig Topper authored Nov 06, 2017
  57f96ac6
Oct 31, 2017
- [CodeGen] map sqrt libcalls to llvm.sqrt when errno is not set · 7cb25a88
  Sanjay Patel authored Oct 31, 2017
  7cb25a88
Oct 14, 2017
- [OpenCL] Emit enqueued block as kernel · c2a87a05
  Yaxun Liu authored Oct 14, 2017
  c2a87a05
Oct 12, 2017
- [CUDA] Added __hmma_m16n16k16_* builtins to support mma instructions on sm_70 · 91cc00bd
  Artem Belevich authored Oct 12, 2017
  91cc00bd
Oct 11, 2017
- [X86] Add support for 'amdfam17h' to __builtin_cpu_is to match gcc. · 8c8e83a1
  Craig Topper authored Oct 11, 2017
  8c8e83a1
Oct 09, 2017
- AMDGPU: Add read_exec_lo/hi builtins · f12e3b84
  Matt Arsenault authored Oct 09, 2017
  f12e3b84
Oct 06, 2017
- Split X86::BI__builtin_cpu_init handling into own function[NFC] · 1fe643a6
  Erich Keane authored Oct 06, 2017
  1fe643a6
- Fix check strings in test case and use llvm::to_string instead of · a4638128
  Akira Hatanaka authored Oct 06, 2017
  a4638128
- [CodeGen] Emit a helper function for __builtin_os_log_format to reduce · 6b103bc1
  Akira Hatanaka authored Oct 06, 2017
  6b103bc1
Sep 26, 2017
- [NVPTX] added match.{any,all}.sync instructions, intrinsics & builtins. · bab95c70
  Artem Belevich authored Sep 26, 2017
  bab95c70
Sep 25, 2017
- Revert "[NVPTX] added match.{any,all}.sync instructions, intrinsics & builtins.", rL314135. · d31d5e6a
  Justin Lebar authored Sep 25, 2017
  d31d5e6a
- [NVPTX] added match.{any,all}.sync instructions, intrinsics & builtins. · 9941ee95
  Artem Belevich authored Sep 25, 2017
  9941ee95
Sep 16, 2017
- [WebAssembly] Restore __builtin_wasm_rethrow builtin · b29a17ba
  Heejin Ahn authored Sep 16, 2017
  b29a17ba
- [X86] Use native shuffle vector for the perm2f128 intrinsics · 8cd7b0cd
  Craig Topper authored Sep 15, 2017
  8cd7b0cd
- Remove __builtin_wasm_rethrow builtin · fa9e1fba
  Heejin Ahn authored Sep 15, 2017
  fa9e1fba
Sep 13, 2017
- [X86] [PATCH] [intrinsics] Lowering X86 ABS intrinsics to IR. (clang) · 3fba3c3b
  Uriel Korach authored Sep 13, 2017
  3fba3c3b