[PATCH] D18049: AMDGPU i16 implementation

[Wei Ding via llvm-commits]

wdng updated this revision to Diff 50775.
wdng added a comment.

1. global-extload-i16.ll contains extload tests for i16->i32, i16->i64. Test for load/store i8->i16 is added into file global-extload-i8.ll
2. Dead codes have been removed.
3. Fixed indentation issues.
4. Fixed scalar zext, used S_MOV_B32 to materialize the 0
5. Added tests for shl, sra, sub
6. Removed min/max, which will be committed into another separate patch.


Repository:
  rL LLVM

http://reviews.llvm.org/D18049

Files:
  lib/Target/AMDGPU/VIInstructions.td
  test/CodeGen/AMDGPU/global-extload-i8.ll
  test/CodeGen/AMDGPU/max.i16.ll
  test/CodeGen/AMDGPU/min_test.ll
  test/CodeGen/AMDGPU/shl.i16.ll
  test/CodeGen/AMDGPU/shl.ll
  test/CodeGen/AMDGPU/sra.i16.ll
  test/CodeGen/AMDGPU/sra.ll
  test/CodeGen/AMDGPU/sub.i16.ll
  test/CodeGen/AMDGPU/sub.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D18049.50775.patch
Type: text/x-patch
Size: 45823 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20160315/2826d0f1/attachment.bin>