Commit 184ed645 authored by Simon Pilgrim's avatar Simon Pilgrim
Browse files

[X86] cvttpd2dq lowering has been supported for some time 

Tests in vec_fp_to_int.ll
parent a29dd911

llvm/lib/Target/X86/README-SSE.txt

+0 −9
Original line number Diff line number Diff line
@@ -839,12 +839,3 @@ similarly, v[0]-v[1] should match to hsubpd, and {v[0]-v[1], w[0]-w[1]} should 
turn into hsubpd also.



//===---------------------------------------------------------------------===//
 


define <2 x i32> @foo(<2 x double> %in) {

  %x = fptosi <2 x double> %in to <2 x i32>

  ret <2 x i32> %x

}



Should compile into cvttpd2dq instead of being scalarized into 2 cvttsd2si.



//===---------------------------------------------------------------------===//