[PATCH] D16399: [X86][AVX512] add encoding and intrinsics for Multishift

Elena Demikhovsky via llvm-commits llvm-commits at lists.llvm.org
Sat Jan 30 23:29:16 PST 2016


delena added inline comments.

================
Comment at: E:/Work/LLVM/llvm/include/llvm/IR/IntrinsicsX86.td:2514
@@ -2513,1 +2513,3 @@
                    llvm_i8_ty, llvm_v8i64_ty,  llvm_i8_ty], [IntrNoMem]>;
+  def int_x86_avx512_mask_pmultishift_qb_128:
+        GCCBuiltin<"__builtin_ia32_vpmultishiftqb128_mask">,
----------------
The types you wrote here don't match GCC
return (__m512i) __builtin_ia32_vpmultishiftqb512_mask ((__v64qi) __X,
                                                          (__v64qi) __Y,
                                                          (__v64qi) __W,
                                                          (__mmask64) __M);


Repository:
  rL LLVM

http://reviews.llvm.org/D16399





More information about the llvm-commits mailing list