RPCS3/llvm
1
0
Fork 0
mirror of https://github.com/RPCS3/llvm.git synced 2024-12-28 07:05:03 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity
b1d081230e
llvm/test/CodeGen/ARM/2011-11-29-128bitArithmetics.ll
Evan Cheng d10eab0a95 Use vld1 / vst2 for unaligned v2f64 load / store. e.g. Use vld1.16 for 2-byte 
aligned address. Based on patch by David Peixotto.

Also use vld1.64 / vst1.64 with 128-bit alignment to take advantage of alignment
hints. rdar://12090772, rdar://12238782


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@164089 91177308-0d34-0410-b5e6-96231b3b80d8
2012-09-18 01:42:45 +00:00

334 lines
7.7 KiB
LLVM
Raw Blame History

; RUN: llc < %s -march=arm -mcpu=cortex-a9 | FileCheck %s
@A = global <4 x float> <float 0., float 1., float 2., float 3.>
define void @test_sqrt(<4 x float>* %X) nounwind {
; CHECK: test_sqrt:
; CHECK: movw r1, :lower16:{{.*}}
; CHECK: movt r1, :upper16:{{.*}}
; CHECK: vld1.64 {{.*}}, [r1, :128]
; CHECK: vsqrt.f32 {{s[0-9]+}}, {{s[0-9]+}}
; CHECK: vsqrt.f32 {{s[0-9]+}}, {{s[0-9]+}}
; CHECK: vsqrt.f32 {{s[0-9]+}}, {{s[0-9]+}}
; CHECK: vsqrt.f32 {{s[0-9]+}}, {{s[0-9]+}}
; CHECK: vst1.64 {{.*}}
L.entry:
%0 = load <4 x float>* @A, align 16
%1 = call <4 x float> @llvm.sqrt.v4f32(<4 x float> %0)
store <4 x float> %1, <4 x float>* %X, align 16
ret void
}
declare <4 x float> @llvm.sqrt.v4f32(<4 x float>) nounwind readonly
define void @test_cos(<4 x float>* %X) nounwind {
; CHECK: test_cos:
; CHECK: movw [[reg0:r[0-9]+]], :lower16:{{.*}}
; CHECK: movt [[reg0]], :upper16:{{.*}}
; CHECK: vld1.64
; CHECK: {{v?mov(.32)?}} r0,
; CHECK: bl {{.*}}cosf
; CHECK: {{v?mov(.32)?}} r0,
; CHECK: bl {{.*}}cosf
; CHECK: {{v?mov(.32)?}} r0,
; CHECK: bl {{.*}}cosf
; CHECK: {{v?mov(.32)?}} r0,
; CHECK: bl {{.*}}cosf
; CHECK: vst1.64
L.entry:
%0 = load <4 x float>* @A, align 16
%1 = call <4 x float> @llvm.cos.v4f32(<4 x float> %0)
store <4 x float> %1, <4 x float>* %X, align 16
ret void
}
declare <4 x float> @llvm.cos.v4f32(<4 x float>) nounwind readonly
define void @test_exp(<4 x float>* %X) nounwind {
; CHECK: test_exp:
; CHECK: movw [[reg0:r[0-9]+]], :lower16:{{.*}}
; CHECK: movt [[reg0]], :upper16:{{.*}}
; CHECK: vld1.64
; CHECK: {{v?mov(.32)?}} r0,
; CHECK: bl {{.*}}expf
; CHECK: {{v?mov(.32)?}} r0,
; CHECK: bl {{.*}}expf
; CHECK: {{v?mov(.32)?}} r0,
; CHECK: bl {{.*}}expf
; CHECK: {{v?mov(.32)?}} r0,
; CHECK: bl {{.*}}expf
; CHECK: vst1.64
L.entry:
%0 = load <4 x float>* @A, align 16
%1 = call <4 x float> @llvm.exp.v4f32(<4 x float> %0)
store <4 x float> %1, <4 x float>* %X, align 16
ret void
}
declare <4 x float> @llvm.exp.v4f32(<4 x float>) nounwind readonly
define void @test_exp2(<4 x float>* %X) nounwind {
; CHECK: test_exp2:
; CHECK: movw [[reg0:r[0-9]+]], :lower16:{{.*}}
; CHECK: movt [[reg0]], :upper16:{{.*}}
; CHECK: vld1.64
; CHECK: {{v?mov(.32)?}} r0,
; CHECK: bl {{.*}}exp2f
; CHECK: {{v?mov(.32)?}} r0,
; CHECK: bl {{.*}}exp2f
; CHECK: {{v?mov(.32)?}} r0,
; CHECK: bl {{.*}}exp2f
; CHECK: {{v?mov(.32)?}} r0,
; CHECK: bl {{.*}}exp2f
; CHECK: vst1.64
L.entry:
%0 = load <4 x float>* @A, align 16
%1 = call <4 x float> @llvm.exp2.v4f32(<4 x float> %0)
store <4 x float> %1, <4 x float>* %X, align 16
ret void
}
declare <4 x float> @llvm.exp2.v4f32(<4 x float>) nounwind readonly
define void @test_log10(<4 x float>* %X) nounwind {
; CHECK: test_log10:
; CHECK: movw [[reg0:r[0-9]+]], :lower16:{{.*}}
; CHECK: movt [[reg0]], :upper16:{{.*}}
; CHECK: vld1.64
; CHECK: {{v?mov(.32)?}} r0,
; CHECK: bl {{.*}}log10f
; CHECK: {{v?mov(.32)?}} r0,
; CHECK: bl {{.*}}log10f
; CHECK: {{v?mov(.32)?}} r0,
; CHECK: bl {{.*}}log10f
; CHECK: {{v?mov(.32)?}} r0,
; CHECK: bl {{.*}}log10f
; CHECK: vst1.64
L.entry:
%0 = load <4 x float>* @A, align 16
%1 = call <4 x float> @llvm.log10.v4f32(<4 x float> %0)
store <4 x float> %1, <4 x float>* %X, align 16
ret void
}
declare <4 x float> @llvm.log10.v4f32(<4 x float>) nounwind readonly
define void @test_log(<4 x float>* %X) nounwind {
; CHECK: test_log:
; CHECK: movw [[reg0:r[0-9]+]], :lower16:{{.*}}
; CHECK: movt [[reg0]], :upper16:{{.*}}
; CHECK: vld1.64
; CHECK: {{v?mov(.32)?}} r0,
; CHECK: bl {{.*}}logf
; CHECK: {{v?mov(.32)?}} r0,
; CHECK: bl {{.*}}logf
; CHECK: {{v?mov(.32)?}} r0,
; CHECK: bl {{.*}}logf
; CHECK: {{v?mov(.32)?}} r0,
; CHECK: bl {{.*}}logf
; CHECK: vst1.64
L.entry:
%0 = load <4 x float>* @A, align 16
%1 = call <4 x float> @llvm.log.v4f32(<4 x float> %0)
store <4 x float> %1, <4 x float>* %X, align 16
ret void
}
declare <4 x float> @llvm.log.v4f32(<4 x float>) nounwind readonly
define void @test_log2(<4 x float>* %X) nounwind {
; CHECK: test_log2:
; CHECK: movw [[reg0:r[0-9]+]], :lower16:{{.*}}
; CHECK: movt [[reg0]], :upper16:{{.*}}
; CHECK: vld1.64
; CHECK: {{v?mov(.32)?}} r0,
; CHECK: bl {{.*}}log2f
; CHECK: {{v?mov(.32)?}} r0,
; CHECK: bl {{.*}}log2f
; CHECK: {{v?mov(.32)?}} r0,
; CHECK: bl {{.*}}log2f
; CHECK: {{v?mov(.32)?}} r0,
; CHECK: bl {{.*}}log2f
; CHECK: vst1.64
L.entry:
%0 = load <4 x float>* @A, align 16
%1 = call <4 x float> @llvm.log2.v4f32(<4 x float> %0)
store <4 x float> %1, <4 x float>* %X, align 16
ret void
}
declare <4 x float> @llvm.log2.v4f32(<4 x float>) nounwind readonly
define void @test_pow(<4 x float>* %X) nounwind {
; CHECK: test_pow:
; CHECK: movw [[reg0:r[0-9]+]], :lower16:{{.*}}
; CHECK: movt [[reg0]], :upper16:{{.*}}
; CHECK: vld1.64
; CHECK: {{v?mov(.32)?}} r0,
; CHECK: bl {{.*}}powf
; CHECK: {{v?mov(.32)?}} r0,
; CHECK: bl {{.*}}powf
; CHECK: {{v?mov(.32)?}} r0,
; CHECK: bl {{.*}}powf
; CHECK: {{v?mov(.32)?}} r0,
; CHECK: bl {{.*}}powf
; CHECK: vst1.64
L.entry:
%0 = load <4 x float>* @A, align 16
%1 = call <4 x float> @llvm.pow.v4f32(<4 x float> %0, <4 x float> <float 2., float 2., float 2., float 2.>)
store <4 x float> %1, <4 x float>* %X, align 16
ret void
}
declare <4 x float> @llvm.pow.v4f32(<4 x float>, <4 x float>) nounwind readonly
define void @test_powi(<4 x float>* %X) nounwind {
; CHECK: test_powi:
; CHECK: movw [[reg0:r[0-9]+]], :lower16:{{.*}}
; CHECK: movt [[reg0]], :upper16:{{.*}}
; CHECK: vld1.64 {{.*}}, :128
; CHECK: vmul.f32 {{.*}}
; CHECK: vst1.64
L.entry:
%0 = load <4 x float>* @A, align 16
%1 = call <4 x float> @llvm.powi.v4f32(<4 x float> %0, i32 2)
store <4 x float> %1, <4 x float>* %X, align 16
ret void
}
declare <4 x float> @llvm.powi.v4f32(<4 x float>, i32) nounwind readonly
define void @test_sin(<4 x float>* %X) nounwind {
; CHECK: test_sin:
; CHECK: movw [[reg0:r[0-9]+]], :lower16:{{.*}}
; CHECK: movt [[reg0]], :upper16:{{.*}}
; CHECK: vld1.64
; CHECK: {{v?mov(.32)?}} r0,
; CHECK: bl {{.*}}sinf
; CHECK: {{v?mov(.32)?}} r0,
; CHECK: bl {{.*}}sinf
; CHECK: {{v?mov(.32)?}} r0,
; CHECK: bl {{.*}}sinf
; CHECK: {{v?mov(.32)?}} r0,
; CHECK: bl {{.*}}sinf
; CHECK: vst1.64
L.entry:
%0 = load <4 x float>* @A, align 16
%1 = call <4 x float> @llvm.sin.v4f32(<4 x float> %0)
store <4 x float> %1, <4 x float>* %X, align 16
ret void
}
declare <4 x float> @llvm.sin.v4f32(<4 x float>) nounwind readonly
define void @test_floor(<4 x float>* %X) nounwind {
; CHECK: test_floor:
; CHECK: movw [[reg0:r[0-9]+]], :lower16:{{.*}}
; CHECK: movt [[reg0]], :upper16:{{.*}}
; CHECK: vld1.64
; CHECK: {{v?mov(.32)?}} r0,
; CHECK: bl {{.*}}floorf
; CHECK: {{v?mov(.32)?}} r0,
; CHECK: bl {{.*}}floorf
; CHECK: {{v?mov(.32)?}} r0,
; CHECK: bl {{.*}}floorf
; CHECK: {{v?mov(.32)?}} r0,
; CHECK: bl {{.*}}floorf
; CHECK: vst1.64
L.entry:
%0 = load <4 x float>* @A, align 16
%1 = call <4 x float> @llvm.floor.v4f32(<4 x float> %0)
store <4 x float> %1, <4 x float>* %X, align 16
ret void
}
declare <4 x float> @llvm.floor.v4f32(<4 x float>) nounwind readonly
Reference in New Issue View Git Blame Copy Permalink