[PATCH] D42086: [X86] Teach LowerBUILD_VECTOR to recognize pair-wise splats of 32-bit elements and use a 64-bit broadcast

Simon Pilgrim via Phabricator via llvm-commits llvm-commits at lists.llvm.org
Tue Jan 16 09:45:52 PST 2018


RKSimon added inline comments.


================
Comment at: test/CodeGen/X86/avx512-intrinsics-fast-isel.ll:490
+; X32-NEXT:    vmovd {{.*#+}} xmm1 = mem[0],zero,zero,zero
+; X32-NEXT:    vpinsrd $1, {{[0-9]+}}(%esp), %xmm1, %xmm1
 ; X32-NEXT:    kmovw %eax, %k1
----------------
We'd gain from INSERT_VECTOR_ELT support being added to EltsFromConsecutiveLoads - merging multiple consecutive scalar loads into a single scalar load+insert into a zero/undef vector.


https://reviews.llvm.org/D42086





More information about the llvm-commits mailing list