diff options
| author | Daya S Khudia <dskhudia@fb.com> | 2019-01-15 01:40:03 +0300 |
|---|---|---|
| committer | Facebook Github Bot <facebook-github-bot@users.noreply.github.com> | 2019-01-15 01:42:31 +0300 |
| commit | 5e085f08d4039eb3b69de857df531c1bc2adca19 (patch) | |
| tree | 6b8c338a077eeb9054807317d0d76475e781464a | |
| parent | ae9f02719c40fb601ef26dc506b823bed02bfca6 (diff) | |
mac build fix (#58)
Summary:
Pull Request resolved: https://github.com/pytorch/FBGEMM/pull/58
Same as title
Reviewed By: jianyuh
Differential Revision: D13660391
fbshipit-source-id: ac3ff941587e8f17e1f8a69e16acc648df55d72b
| -rw-r--r-- | src/GroupwiseConvAcc32Avx2.cc | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/src/GroupwiseConvAcc32Avx2.cc b/src/GroupwiseConvAcc32Avx2.cc index 8298f4c..00a9571 100644 --- a/src/GroupwiseConvAcc32Avx2.cc +++ b/src/GroupwiseConvAcc32Avx2.cc @@ -148,7 +148,7 @@ void GenConvKernel<int32_t>::genConstForPermutations<inst_set_t::avx2>( // Permute to put 1st group to lower 128-bit and 2nd group in upper // 128-bit. // load 7, 5, 3, 1, 6, 4, 2, 0 in a 64-bit reg - a->mov(permute_const_reg, 0x0705030106040200); + a->mov(permute_const_reg, static_cast<asmjit::Imm>(0x0705030106040200)); a->movq(const_reg_xmm, permute_const_reg); // Zero extend 8 packed 8-bit integers in the low 8 bytes of const_reg_xmm to // 8 packed 32-bit integers in stPermRegAvx2_ |