[PATCH] [AArch64] Inline memcpy() as a sequence of ldp-stp with 64-bit registers

Sergey Dmitrouk sdmitrouk at accesssoftek.com
Thu Nov 6 08:57:29 PST 2014


Changed code to process pairs of loads and stores rather than all of them inside a basic block.
Removed changes that are not directly related to new pass.

http://reviews.llvm.org/D6054

Files:
  lib/Target/AArch64/AArch64.h
  lib/Target/AArch64/AArch64LoadStoreInterleave.cpp
  lib/Target/AArch64/AArch64TargetMachine.cpp
  lib/Target/AArch64/CMakeLists.txt
  test/CodeGen/AArch64/arm64-variadic-aapcs.ll
  test/CodeGen/AArch64/arm64-virtual_base.ll
  test/CodeGen/AArch64/func-calls.ll
  test/CodeGen/AArch64/memcpy-f128.ll
  test/CodeGen/AArch64/optimal-load-store-pairs.ll
-------------- next part --------------
A non-text attachment was scrubbed...
Name: D6054.15881.patch
Type: text/x-patch
Size: 16512 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20141106/a2e40645/attachment.bin>


More information about the llvm-commits mailing list