RPCSX/llvm
1
0
Fork 0
mirror of https://github.com/RPCSX/llvm.git synced 2025-01-25 13:45:34 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity

Add isel patterns for v2f32 / v4f32 neon.vbsl intrinsics. rdar://12471808

Browse Source 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@165673 91177308-0d34-0410-b5e6-96231b3b80d8
This commit is contained in:
Evan Cheng 2012-10-10 23:06:34 +00:00
parent 6f9e66e1be
commit 6b61491de3
2 changed files with 26 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

8
lib/Target/ARM/ARMInstrNEON.td
View File

@ -4500,6 +4500,10 @@ def : Pat<(v2i32 (int_arm_neon_vbsl (v2i32 DPR:$src1),
(v2i32 DPR:$Vn), (v2i32 DPR:$Vm))),
(VBSLd DPR:$src1, DPR:$Vn, DPR:$Vm)>,
Requires<[HasNEON]>;
def : Pat<(v2f32 (int_arm_neon_vbsl (v2f32 DPR:$src1),
(v2f32 DPR:$Vn), (v2f32 DPR:$Vm))),
(VBSLd DPR:$src1, DPR:$Vn, DPR:$Vm)>,
Requires<[HasNEON]>;
def : Pat<(v2i32 (or (and DPR:$Vn, DPR:$Vd),
(and DPR:$Vm, (vnotd DPR:$Vd)))),
@ -4525,6 +4529,10 @@ def : Pat<(v4i32 (int_arm_neon_vbsl (v4i32 QPR:$src1),
(v4i32 QPR:$Vn), (v4i32 QPR:$Vm))),
(VBSLq QPR:$src1, QPR:$Vn, QPR:$Vm)>,
Requires<[HasNEON]>;
def : Pat<(v4f32 (int_arm_neon_vbsl (v4f32 QPR:$src1),
(v4f32 QPR:$Vn), (v4f32 QPR:$Vm))),
(VBSLq QPR:$src1, QPR:$Vn, QPR:$Vm)>,
Requires<[HasNEON]>;
def : Pat<(v4i32 (or (and QPR:$Vn, QPR:$Vd),
(and QPR:$Vm, (vnotq QPR:$Vd)))),

18
test/CodeGen/ARM/vbsl.ll
View File

@ -1,5 +1,7 @@
; RUN: llc < %s -march=arm -mattr=+neon | FileCheck %s
; rdar://12471808
define <8 x i8> @v_bsli8(<8 x i8>* %A, <8 x i8>* %B, <8 x i8>* %C) nounwind {
;CHECK: v_bsli8:
;CHECK: vbsl
@ -125,6 +127,13 @@ define <2 x i32> @f3(<2 x i32> %a, <2 x i32> %b, <2 x i32> %c) nounwind readnone
ret <2 x i32> %vbsl3.i
}
define <2 x float> @f4(<2 x float> %a, <2 x float> %b, <2 x float> %c) nounwind readnone optsize ssp {
; CHECK: f4:
; CHECK: vbsl
%vbsl4.i = tail call <2 x float> @llvm.arm.neon.vbsl.v2f32(<2 x float> %a, <2 x float> %b, <2 x float> %c) nounwind
ret <2 x float> %vbsl4.i
}
define <16 x i8> @g1(<16 x i8> %a, <16 x i8> %b, <16 x i8> %c) nounwind readnone optsize ssp {
; CHECK: g1:
; CHECK: vbsl
@ -146,9 +155,18 @@ define <4 x i32> @g3(<4 x i32> %a, <4 x i32> %b, <4 x i32> %c) nounwind readnone
ret <4 x i32> %vbsl3.i
}
define <4 x float> @g4(<4 x float> %a, <4 x float> %b, <4 x float> %c) nounwind readnone optsize ssp {
; CHECK: g4:
; CHECK: vbsl
%vbsl4.i = tail call <4 x float> @llvm.arm.neon.vbsl.v4f32(<4 x float> %a, <4 x float> %b, <4 x float> %c) nounwind
ret <4 x float> %vbsl4.i
}
declare <4 x i32> @llvm.arm.neon.vbsl.v4i32(<4 x i32>, <4 x i32>, <4 x i32>) nounwind readnone
declare <8 x i16> @llvm.arm.neon.vbsl.v8i16(<8 x i16>, <8 x i16>, <8 x i16>) nounwind readnone
declare <16 x i8> @llvm.arm.neon.vbsl.v16i8(<16 x i8>, <16 x i8>, <16 x i8>) nounwind readnone
declare <2 x i32> @llvm.arm.neon.vbsl.v2i32(<2 x i32>, <2 x i32>, <2 x i32>) nounwind readnone
declare <4 x i16> @llvm.arm.neon.vbsl.v4i16(<4 x i16>, <4 x i16>, <4 x i16>) nounwind readnone
declare <8 x i8> @llvm.arm.neon.vbsl.v8i8(<8 x i8>, <8 x i8>, <8 x i8>) nounwind readnone
declare <2 x float> @llvm.arm.neon.vbsl.v2f32(<2 x float>, <2 x float>, <2 x float>) nounwind readnone
declare <4 x float> @llvm.arm.neon.vbsl.v4f32(<4 x float>, <4 x float>, <4 x float>) nounwind readnone