mirror of
https://github.com/c64scene-ar/llvm-6502.git
synced 2024-12-28 04:33:05 +00:00
7b4b522ec8
We couldn't cope if the first mask element was UNDEF before, which isn't ideal. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@206588 91177308-0d34-0410-b5e6-96231b3b80d8
49 lines
3.3 KiB
LLVM
49 lines
3.3 KiB
LLVM
; RUN: llc < %s -verify-machineinstrs -mtriple=aarch64-none-linux-gnu -mattr=+neon | FileCheck %s
|
|
; arm64 has a separate copy due to intrinsics
|
|
|
|
define <4 x i32> @copyTuple.QPair(i8* %a, i8* %b) {
|
|
; CHECK-LABEL: copyTuple.QPair:
|
|
; CHECK: orr v{{[0-9]+}}.16b, v{{[0-9]+}}.16b, v{{[0-9]+}}.16b
|
|
; CHECK: orr v{{[0-9]+}}.16b, v{{[0-9]+}}.16b, v{{[0-9]+}}.16b
|
|
; CHECK: ld2 {{{v[0-9]+}}.s, {{v[0-9]+}}.s}[{{[0-9]+}}], [x{{[0-9]+|sp}}]
|
|
entry:
|
|
%vld = tail call { <4 x i32>, <4 x i32> } @llvm.arm.neon.vld2lane.v4i32(i8* %a, <4 x i32> <i32 -1, i32 -1, i32 -1, i32 -1>, <4 x i32> <i32 2, i32 2, i32 2, i32 2>, i32 0, i32 4)
|
|
%extract = extractvalue { <4 x i32>, <4 x i32> } %vld, 0
|
|
%vld1 = tail call { <4 x i32>, <4 x i32> } @llvm.arm.neon.vld2lane.v4i32(i8* %b, <4 x i32> %extract, <4 x i32> <i32 -1, i32 -1, i32 -1, i32 -1>, i32 1, i32 4)
|
|
%vld1.fca.0.extract = extractvalue { <4 x i32>, <4 x i32> } %vld1, 0
|
|
ret <4 x i32> %vld1.fca.0.extract
|
|
}
|
|
|
|
define <4 x i32> @copyTuple.QTriple(i8* %a, i8* %b, <4 x i32> %c) {
|
|
; CHECK-LABEL: copyTuple.QTriple:
|
|
; CHECK: orr v{{[0-9]+}}.16b, v{{[0-9]+}}.16b, v{{[0-9]+}}.16b
|
|
; CHECK: orr v{{[0-9]+}}.16b, v{{[0-9]+}}.16b, v{{[0-9]+}}.16b
|
|
; CHECK: orr v{{[0-9]+}}.16b, v{{[0-9]+}}.16b, v{{[0-9]+}}.16b
|
|
; CHECK: ld3 {{{v[0-9]+}}.s, {{v[0-9]+}}.s, {{v[0-9]+}}.s}[{{[0-9]+}}], [x{{[0-9]+|sp}}]
|
|
entry:
|
|
%vld = tail call { <4 x i32>, <4 x i32>, <4 x i32> } @llvm.arm.neon.vld3lane.v4i32(i8* %a, <4 x i32> <i32 -1, i32 -1, i32 -1, i32 -1>, <4 x i32> %c, <4 x i32> %c, i32 0, i32 4)
|
|
%extract = extractvalue { <4 x i32>, <4 x i32>, <4 x i32> } %vld, 0
|
|
%vld1 = tail call { <4 x i32>, <4 x i32>, <4 x i32> } @llvm.arm.neon.vld3lane.v4i32(i8* %b, <4 x i32> %extract, <4 x i32> <i32 -1, i32 -1, i32 -1, i32 -1>, <4 x i32> %c, i32 1, i32 4)
|
|
%vld1.fca.0.extract = extractvalue { <4 x i32>, <4 x i32>, <4 x i32> } %vld1, 0
|
|
ret <4 x i32> %vld1.fca.0.extract
|
|
}
|
|
|
|
define <4 x i32> @copyTuple.QQuad(i8* %a, i8* %b, <4 x i32> %c) {
|
|
; CHECK-LABEL: copyTuple.QQuad:
|
|
; CHECK: orr v{{[0-9]+}}.16b, v{{[0-9]+}}.16b, v{{[0-9]+}}.16b
|
|
; CHECK: orr v{{[0-9]+}}.16b, v{{[0-9]+}}.16b, v{{[0-9]+}}.16b
|
|
; CHECK: orr v{{[0-9]+}}.16b, v{{[0-9]+}}.16b, v{{[0-9]+}}.16b
|
|
; CHECK: orr v{{[0-9]+}}.16b, v{{[0-9]+}}.16b, v{{[0-9]+}}.16b
|
|
; CHECK: ld4 {{{v[0-9]+}}.s, {{v[0-9]+}}.s, {{v[0-9]+}}.s, {{v[0-9]+}}.s}[{{[0-9]+}}], [x{{[0-9]+|sp}}]
|
|
entry:
|
|
%vld = tail call { <4 x i32>, <4 x i32>, <4 x i32>, <4 x i32> } @llvm.arm.neon.vld4lane.v4i32(i8* %a, <4 x i32> <i32 -1, i32 -1, i32 -1, i32 -1>, <4 x i32> %c, <4 x i32> %c, <4 x i32> %c, i32 0, i32 4)
|
|
%extract = extractvalue { <4 x i32>, <4 x i32>, <4 x i32>, <4 x i32> } %vld, 0
|
|
%vld1 = tail call { <4 x i32>, <4 x i32>, <4 x i32>, <4 x i32> } @llvm.arm.neon.vld4lane.v4i32(i8* %b, <4 x i32> %extract, <4 x i32> <i32 -1, i32 -1, i32 -1, i32 -1>, <4 x i32> %c, <4 x i32> %c, i32 1, i32 4)
|
|
%vld1.fca.0.extract = extractvalue { <4 x i32>, <4 x i32>, <4 x i32>, <4 x i32> } %vld1, 0
|
|
ret <4 x i32> %vld1.fca.0.extract
|
|
}
|
|
|
|
declare { <4 x i32>, <4 x i32> } @llvm.arm.neon.vld2lane.v4i32(i8*, <4 x i32>, <4 x i32>, i32, i32)
|
|
declare { <4 x i32>, <4 x i32>, <4 x i32> } @llvm.arm.neon.vld3lane.v4i32(i8*, <4 x i32>, <4 x i32>, <4 x i32>, i32, i32)
|
|
declare { <4 x i32>, <4 x i32>, <4 x i32>, <4 x i32> } @llvm.arm.neon.vld4lane.v4i32(i8*, <4 x i32>, <4 x i32>, <4 x i32>, <4 x i32>, i32, i32)
|