Should code that is directly using the builtins themselves (like __builtin_ia32_pblendw256) be optimized too? If so wouldn't it be better to, for example, leave _mm256_blend_epi16 as is, remove __builtin_ia32_pblendw256 from BuiltinsX86.def and make it a #define to shufflevector?
On 2 May 2014 20:55, Filipe Cabecinhas <[email protected]> wrote: > Hi eli.friedman, craig.topper, > > Most of the clang header patch by Simon Pilgrim @ SCEE. > Also fixed (or added) clang tests for these intrinsics. > > LLVM tests to make sure we get the blend instruction out of these > shufflevectors are at http://reviews.llvm.org/D3600 > > http://reviews.llvm.org/D3601 > > Files: > lib/Headers/avx2intrin.h > lib/Headers/avxintrin.h > lib/Headers/smmintrin.h > test/CodeGen/avx-builtins.c > test/CodeGen/avx2-builtins.c > test/CodeGen/sse-builtins.c > > _______________________________________________ > cfe-commits mailing list > [email protected] > http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits > _______________________________________________ cfe-commits mailing list [email protected] http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits
