[libc-commits] [PATCH] D148717: [libc] Improve memcmp latency and codegen

[Tue Ly via Phabricator via libc-commits]

lntue added inline comments.


================
Comment at: libc/src/string/CMakeLists.txt:573
   add_memcpy(memcpy_x86_64_opt_sse4   COMPILE_OPTIONS -march=nehalem        REQUIRE SSE4_2)
-  add_memcpy(memcpy_x86_64_opt_avx2   COMPILE_OPTIONS -march=haswell        REQUIRE AVX2)
+  add_memcpy(memcpy_x86_64_opt_avx    COMPILE_OPTIONS -march=haswell        REQUIRE AVX)
   add_memcpy(memcpy_x86_64_opt_avx512 COMPILE_OPTIONS -march=skylake-avx512 REQUIRE AVX512F)
----------------
If you drop the requirement to AVX, should the compile option be `-march=sandybridge` instead?


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D148717/new/

https://reviews.llvm.org/D148717