blob: becddaea31267a5a3af6ca1143bcf8ce51ffb2e8 [file] [log] [blame]
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
; RUN: llc -aarch64-sve-vector-bits-min=128 < %s | not grep ptrue
; RUN: llc -aarch64-sve-vector-bits-min=256 < %s | FileCheck %s -check-prefixes=CHECK,VBITS_GE_256
; RUN: llc -aarch64-sve-vector-bits-min=384 < %s | FileCheck %s -check-prefixes=CHECK,VBITS_GE_256
; RUN: llc -aarch64-sve-vector-bits-min=512 < %s | FileCheck %s -check-prefixes=CHECK,VBITS_GE_512
; RUN: llc -aarch64-sve-vector-bits-min=640 < %s | FileCheck %s -check-prefixes=CHECK,VBITS_GE_512
; RUN: llc -aarch64-sve-vector-bits-min=768 < %s | FileCheck %s -check-prefixes=CHECK,VBITS_GE_512
; RUN: llc -aarch64-sve-vector-bits-min=896 < %s | FileCheck %s -check-prefixes=CHECK,VBITS_GE_512
; RUN: llc -aarch64-sve-vector-bits-min=1024 < %s | FileCheck %s -check-prefixes=CHECK,VBITS_GE_1024
; RUN: llc -aarch64-sve-vector-bits-min=1152 < %s | FileCheck %s -check-prefixes=CHECK,VBITS_GE_1024
; RUN: llc -aarch64-sve-vector-bits-min=1280 < %s | FileCheck %s -check-prefixes=CHECK,VBITS_GE_1024
; RUN: llc -aarch64-sve-vector-bits-min=1408 < %s | FileCheck %s -check-prefixes=CHECK,VBITS_GE_1024
; RUN: llc -aarch64-sve-vector-bits-min=1536 < %s | FileCheck %s -check-prefixes=CHECK,VBITS_GE_1024
; RUN: llc -aarch64-sve-vector-bits-min=1664 < %s | FileCheck %s -check-prefixes=CHECK,VBITS_GE_1024
; RUN: llc -aarch64-sve-vector-bits-min=1792 < %s | FileCheck %s -check-prefixes=CHECK,VBITS_GE_1024
; RUN: llc -aarch64-sve-vector-bits-min=1920 < %s | FileCheck %s -check-prefixes=CHECK,VBITS_GE_1024
; RUN: llc -aarch64-sve-vector-bits-min=2048 < %s | FileCheck %s -check-prefixes=CHECK,VBITS_GE_2048
target triple = "aarch64-unknown-linux-gnu"
; Don't use SVE for 64-bit vectors.
define void @load_v2f32(ptr %a, ptr %b) #0 {
; CHECK-LABEL: load_v2f32:
; CHECK: // %bb.0:
; CHECK-NEXT: ldr d0, [x0]
; CHECK-NEXT: str d0, [x1]
; CHECK-NEXT: ret
%load = load <2 x float>, ptr %a
store <2 x float> %load, ptr %b
ret void
}
; Don't use SVE for 128-bit vectors.
define void @load_v4f32(ptr %a, ptr %b) #0 {
; CHECK-LABEL: load_v4f32:
; CHECK: // %bb.0:
; CHECK-NEXT: ldr q0, [x0]
; CHECK-NEXT: str q0, [x1]
; CHECK-NEXT: ret
%load = load <4 x float>, ptr %a
store <4 x float> %load, ptr %b
ret void
}
define void @load_v8f32(ptr %a, ptr %b) #0 {
; CHECK-LABEL: load_v8f32:
; CHECK: // %bb.0:
; CHECK-NEXT: ptrue p0.s, vl8
; CHECK-NEXT: ld1w { z0.s }, p0/z, [x0]
; CHECK-NEXT: st1w { z0.s }, p0, [x1]
; CHECK-NEXT: ret
%load = load <8 x float>, ptr %a
store <8 x float> %load, ptr %b
ret void
}
define void @load_v16f32(ptr %a, ptr %b) #0 {
; VBITS_GE_256-LABEL: load_v16f32:
; VBITS_GE_256: // %bb.0:
; VBITS_GE_256-NEXT: ptrue p0.s, vl8
; VBITS_GE_256-NEXT: mov x8, #8 // =0x8
; VBITS_GE_256-NEXT: ld1w { z0.s }, p0/z, [x0, x8, lsl #2]
; VBITS_GE_256-NEXT: ld1w { z1.s }, p0/z, [x0]
; VBITS_GE_256-NEXT: st1w { z0.s }, p0, [x1, x8, lsl #2]
; VBITS_GE_256-NEXT: st1w { z1.s }, p0, [x1]
; VBITS_GE_256-NEXT: ret
;
; VBITS_GE_512-LABEL: load_v16f32:
; VBITS_GE_512: // %bb.0:
; VBITS_GE_512-NEXT: ptrue p0.s, vl16
; VBITS_GE_512-NEXT: ld1w { z0.s }, p0/z, [x0]
; VBITS_GE_512-NEXT: st1w { z0.s }, p0, [x1]
; VBITS_GE_512-NEXT: ret
;
; VBITS_GE_1024-LABEL: load_v16f32:
; VBITS_GE_1024: // %bb.0:
; VBITS_GE_1024-NEXT: ptrue p0.s, vl16
; VBITS_GE_1024-NEXT: ld1w { z0.s }, p0/z, [x0]
; VBITS_GE_1024-NEXT: st1w { z0.s }, p0, [x1]
; VBITS_GE_1024-NEXT: ret
;
; VBITS_GE_2048-LABEL: load_v16f32:
; VBITS_GE_2048: // %bb.0:
; VBITS_GE_2048-NEXT: ptrue p0.s, vl16
; VBITS_GE_2048-NEXT: ld1w { z0.s }, p0/z, [x0]
; VBITS_GE_2048-NEXT: st1w { z0.s }, p0, [x1]
; VBITS_GE_2048-NEXT: ret
%load = load <16 x float>, ptr %a
store <16 x float> %load, ptr %b
ret void
}
define void @load_v32f32(ptr %a, ptr %b) #0 {
; VBITS_GE_256-LABEL: load_v32f32:
; VBITS_GE_256: // %bb.0:
; VBITS_GE_256-NEXT: ptrue p0.s, vl8
; VBITS_GE_256-NEXT: mov x8, #16 // =0x10
; VBITS_GE_256-NEXT: mov x9, #24 // =0x18
; VBITS_GE_256-NEXT: mov x10, #8 // =0x8
; VBITS_GE_256-NEXT: ld1w { z0.s }, p0/z, [x0, x8, lsl #2]
; VBITS_GE_256-NEXT: ld1w { z1.s }, p0/z, [x0, x9, lsl #2]
; VBITS_GE_256-NEXT: ld1w { z2.s }, p0/z, [x0, x10, lsl #2]
; VBITS_GE_256-NEXT: ld1w { z3.s }, p0/z, [x0]
; VBITS_GE_256-NEXT: st1w { z0.s }, p0, [x1, x8, lsl #2]
; VBITS_GE_256-NEXT: st1w { z1.s }, p0, [x1, x9, lsl #2]
; VBITS_GE_256-NEXT: st1w { z2.s }, p0, [x1, x10, lsl #2]
; VBITS_GE_256-NEXT: st1w { z3.s }, p0, [x1]
; VBITS_GE_256-NEXT: ret
;
; VBITS_GE_512-LABEL: load_v32f32:
; VBITS_GE_512: // %bb.0:
; VBITS_GE_512-NEXT: ptrue p0.s, vl16
; VBITS_GE_512-NEXT: mov x8, #16 // =0x10
; VBITS_GE_512-NEXT: ld1w { z0.s }, p0/z, [x0, x8, lsl #2]
; VBITS_GE_512-NEXT: ld1w { z1.s }, p0/z, [x0]
; VBITS_GE_512-NEXT: st1w { z0.s }, p0, [x1, x8, lsl #2]
; VBITS_GE_512-NEXT: st1w { z1.s }, p0, [x1]
; VBITS_GE_512-NEXT: ret
;
; VBITS_GE_1024-LABEL: load_v32f32:
; VBITS_GE_1024: // %bb.0:
; VBITS_GE_1024-NEXT: ptrue p0.s, vl32
; VBITS_GE_1024-NEXT: ld1w { z0.s }, p0/z, [x0]
; VBITS_GE_1024-NEXT: st1w { z0.s }, p0, [x1]
; VBITS_GE_1024-NEXT: ret
;
; VBITS_GE_2048-LABEL: load_v32f32:
; VBITS_GE_2048: // %bb.0:
; VBITS_GE_2048-NEXT: ptrue p0.s, vl32
; VBITS_GE_2048-NEXT: ld1w { z0.s }, p0/z, [x0]
; VBITS_GE_2048-NEXT: st1w { z0.s }, p0, [x1]
; VBITS_GE_2048-NEXT: ret
%load = load <32 x float>, ptr %a
store <32 x float> %load, ptr %b
ret void
}
define void @load_v64f32(ptr %a, ptr %b) #0 {
; VBITS_GE_256-LABEL: load_v64f32:
; VBITS_GE_256: // %bb.0:
; VBITS_GE_256-NEXT: ptrue p0.s, vl8
; VBITS_GE_256-NEXT: mov x8, #8 // =0x8
; VBITS_GE_256-NEXT: mov x9, #24 // =0x18
; VBITS_GE_256-NEXT: mov x10, #16 // =0x10
; VBITS_GE_256-NEXT: mov x11, #48 // =0x30
; VBITS_GE_256-NEXT: mov x12, #40 // =0x28
; VBITS_GE_256-NEXT: mov x13, #56 // =0x38
; VBITS_GE_256-NEXT: mov x14, #32 // =0x20
; VBITS_GE_256-NEXT: ld1w { z0.s }, p0/z, [x0, x11, lsl #2]
; VBITS_GE_256-NEXT: ld1w { z1.s }, p0/z, [x0, x8, lsl #2]
; VBITS_GE_256-NEXT: ld1w { z2.s }, p0/z, [x0, x13, lsl #2]
; VBITS_GE_256-NEXT: ld1w { z3.s }, p0/z, [x0, x9, lsl #2]
; VBITS_GE_256-NEXT: ld1w { z4.s }, p0/z, [x0, x10, lsl #2]
; VBITS_GE_256-NEXT: ld1w { z5.s }, p0/z, [x0, x14, lsl #2]
; VBITS_GE_256-NEXT: ld1w { z6.s }, p0/z, [x0, x12, lsl #2]
; VBITS_GE_256-NEXT: ld1w { z7.s }, p0/z, [x0]
; VBITS_GE_256-NEXT: st1w { z0.s }, p0, [x1, x11, lsl #2]
; VBITS_GE_256-NEXT: st1w { z2.s }, p0, [x1, x13, lsl #2]
; VBITS_GE_256-NEXT: st1w { z5.s }, p0, [x1, x14, lsl #2]
; VBITS_GE_256-NEXT: st1w { z6.s }, p0, [x1, x12, lsl #2]
; VBITS_GE_256-NEXT: st1w { z4.s }, p0, [x1, x10, lsl #2]
; VBITS_GE_256-NEXT: st1w { z3.s }, p0, [x1, x9, lsl #2]
; VBITS_GE_256-NEXT: st1w { z1.s }, p0, [x1, x8, lsl #2]
; VBITS_GE_256-NEXT: st1w { z7.s }, p0, [x1]
; VBITS_GE_256-NEXT: ret
;
; VBITS_GE_512-LABEL: load_v64f32:
; VBITS_GE_512: // %bb.0:
; VBITS_GE_512-NEXT: ptrue p0.s, vl16
; VBITS_GE_512-NEXT: mov x8, #32 // =0x20
; VBITS_GE_512-NEXT: mov x9, #48 // =0x30
; VBITS_GE_512-NEXT: mov x10, #16 // =0x10
; VBITS_GE_512-NEXT: ld1w { z0.s }, p0/z, [x0, x8, lsl #2]
; VBITS_GE_512-NEXT: ld1w { z1.s }, p0/z, [x0, x9, lsl #2]
; VBITS_GE_512-NEXT: ld1w { z2.s }, p0/z, [x0, x10, lsl #2]
; VBITS_GE_512-NEXT: ld1w { z3.s }, p0/z, [x0]
; VBITS_GE_512-NEXT: st1w { z0.s }, p0, [x1, x8, lsl #2]
; VBITS_GE_512-NEXT: st1w { z1.s }, p0, [x1, x9, lsl #2]
; VBITS_GE_512-NEXT: st1w { z2.s }, p0, [x1, x10, lsl #2]
; VBITS_GE_512-NEXT: st1w { z3.s }, p0, [x1]
; VBITS_GE_512-NEXT: ret
;
; VBITS_GE_1024-LABEL: load_v64f32:
; VBITS_GE_1024: // %bb.0:
; VBITS_GE_1024-NEXT: ptrue p0.s, vl32
; VBITS_GE_1024-NEXT: mov x8, #32 // =0x20
; VBITS_GE_1024-NEXT: ld1w { z0.s }, p0/z, [x0, x8, lsl #2]
; VBITS_GE_1024-NEXT: ld1w { z1.s }, p0/z, [x0]
; VBITS_GE_1024-NEXT: st1w { z0.s }, p0, [x1, x8, lsl #2]
; VBITS_GE_1024-NEXT: st1w { z1.s }, p0, [x1]
; VBITS_GE_1024-NEXT: ret
;
; VBITS_GE_2048-LABEL: load_v64f32:
; VBITS_GE_2048: // %bb.0:
; VBITS_GE_2048-NEXT: ptrue p0.s, vl64
; VBITS_GE_2048-NEXT: ld1w { z0.s }, p0/z, [x0]
; VBITS_GE_2048-NEXT: st1w { z0.s }, p0, [x1]
; VBITS_GE_2048-NEXT: ret
%load = load <64 x float>, ptr %a
store <64 x float> %load, ptr %b
ret void
}
attributes #0 = { "target-features"="+sve" }