mirror of
https://github.com/RPCS3/llvm-mirror.git
synced 2024-11-23 19:23:23 +01:00
6db76aaf10
This is still a work in progress but most of the NEON instruction set is supported. llvm-svn: 73919
27 lines
1.1 KiB
LLVM
27 lines
1.1 KiB
LLVM
; RUN: llvm-as < %s | llc -march=arm -mattr=+neon > %t
|
|
; RUN: grep {vshrn\\.i16} %t | count 1
|
|
; RUN: grep {vshrn\\.i32} %t | count 1
|
|
; RUN: grep {vshrn\\.i64} %t | count 1
|
|
|
|
define <8 x i8> @vshrns8(<8 x i16>* %A) nounwind {
|
|
%tmp1 = load <8 x i16>* %A
|
|
%tmp2 = call <8 x i8> @llvm.arm.neon.vshiftn.v8i8(<8 x i16> %tmp1, <8 x i16> < i16 -8, i16 -8, i16 -8, i16 -8, i16 -8, i16 -8, i16 -8, i16 -8 >)
|
|
ret <8 x i8> %tmp2
|
|
}
|
|
|
|
define <4 x i16> @vshrns16(<4 x i32>* %A) nounwind {
|
|
%tmp1 = load <4 x i32>* %A
|
|
%tmp2 = call <4 x i16> @llvm.arm.neon.vshiftn.v4i16(<4 x i32> %tmp1, <4 x i32> < i32 -16, i32 -16, i32 -16, i32 -16 >)
|
|
ret <4 x i16> %tmp2
|
|
}
|
|
|
|
define <2 x i32> @vshrns32(<2 x i64>* %A) nounwind {
|
|
%tmp1 = load <2 x i64>* %A
|
|
%tmp2 = call <2 x i32> @llvm.arm.neon.vshiftn.v2i32(<2 x i64> %tmp1, <2 x i64> < i64 -32, i64 -32 >)
|
|
ret <2 x i32> %tmp2
|
|
}
|
|
|
|
declare <8 x i8> @llvm.arm.neon.vshiftn.v8i8(<8 x i16>, <8 x i16>) nounwind readnone
|
|
declare <4 x i16> @llvm.arm.neon.vshiftn.v4i16(<4 x i32>, <4 x i32>) nounwind readnone
|
|
declare <2 x i32> @llvm.arm.neon.vshiftn.v2i32(<2 x i64>, <2 x i64>) nounwind readnone
|