maxjhandsome
/
llvm-project
forked from OSchip/llvm-project
1
0
Fork 0
Code Issues Pull Requests Packages Releases Wiki Activity

The fp16 <-> fp32 intrinsics are pure, mark them so they can be CSE'd, etc. 

llvm-svn: 98498
This commit is contained in:
Anton Korobeynikov 2010-03-14 18:42:04 +00:00
parent 77bbbf782a
commit c6273f28a9
1 changed files with 4 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
llvm/include/llvm/Intrinsics.td
View File

@ -452,10 +452,12 @@ def int_trap : Intrinsic<[llvm_void_ty]>,
// Intrisics to support half precision floating point format
def int_convert_to_fp16 : Intrinsic<[llvm_i16_ty],
[llvm_float_ty]>,
[llvm_float_ty],
[IntrNoMem]>,
GCCBuiltin<"__gnu_f2h_ieee">;
def int_convert_from_fp16 : Intrinsic<[llvm_float_ty],
[llvm_i16_ty]>,
[llvm_i16_ty],
[IntrNoMem]>,
GCCBuiltin<"__gnu_h2f_ieee">;
// These convert intrinsics are to support various conversions between