derek.holloway
7502bf64a4
CodeQL / Analyze (actions) (push) Canceled after 0s
CodeQL / Analyze (python) (push) Canceled after 0s
pre-commit / pre-commit (push) Canceled after 0s
Multi-Arch CI / setup (push) Canceled after 0s
Publish build_manylinux_x86_64 images / publish_build_manylinux_x86_64 (push) Canceled after 0s
Setup (Multi-Arch) / setup (push) Canceled after 0s
Multi-Arch CI / Linux::${{ fromJSON(needs.setup.outputs.linux_build_config || '{}').build_variant_label || 'skip' }} (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Publish no_rocm_image_ubi10 image / publish_no_rocm_image_ubi10 (push) Canceled after 0s
Multi-Arch CI / Windows::${{ fromJSON(needs.setup.outputs.windows_build_config || '{}').build_variant_label || 'skip' }} (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Multi-Arch CI / Manifest Diff (push) Canceled after 0s
TheRock Manifest Diff Report / Generate Manifest Diff Report (push) Canceled after 0s
Multi-Arch CI / CI Summary (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04 (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04_ocl_rt images / publish_no_rocm_image_ubuntu24_04_ocl_rt (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_rocgdb (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04_rocgdb images / publish_no_rocm_image_ubuntu24_04_rocgdb (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_ocl_rt (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Unit Tests / Unit Tests :: ubuntu-24.04 (push) Canceled after 0s
Unit Tests / Unit Tests :: windows-2022 (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_media (push) Canceled after 0s
Unit Tests / Unit Tests Summary (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_openmpi (push) Canceled after 0s
203 lines
7.1 KiB
LLVM
203 lines
7.1 KiB
LLVM
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
|
|
; RUN: llc -mtriple=aarch64-linux-gnu -mattr=+sve < %s | FileCheck %s
|
|
|
|
; EXTRACT VECTOR ELT
|
|
|
|
define i32 @promote_extract_2i32_idx(<vscale x 2 x i32> %a, i32 %idx) {
|
|
; CHECK-LABEL: promote_extract_2i32_idx:
|
|
; CHECK: // %bb.0:
|
|
; CHECK-NEXT: mov w8, w0
|
|
; CHECK-NEXT: whilels p0.d, xzr, x8
|
|
; CHECK-NEXT: lastb x0, p0, z0.d
|
|
; CHECK-NEXT: // kill: def $w0 killed $w0 killed $x0
|
|
; CHECK-NEXT: ret
|
|
%ext = extractelement <vscale x 2 x i32> %a, i32 %idx
|
|
ret i32 %ext
|
|
}
|
|
|
|
define i8 @split_extract_32i8_idx(<vscale x 32 x i8> %a, i32 %idx) {
|
|
; CHECK-LABEL: split_extract_32i8_idx:
|
|
; CHECK: // %bb.0:
|
|
; CHECK-NEXT: str x29, [sp, #-16]! // 8-byte Folded Spill
|
|
; CHECK-NEXT: addvl sp, sp, #-2
|
|
; CHECK-NEXT: .cfi_escape 0x0f, 0x08, 0x8f, 0x10, 0x92, 0x2e, 0x00, 0x40, 0x1e, 0x22 // sp + 16 + 16 * VG
|
|
; CHECK-NEXT: .cfi_offset w29, -16
|
|
; CHECK-NEXT: rdvl x8, #2
|
|
; CHECK-NEXT: mov w9, w0
|
|
; CHECK-NEXT: str z1, [sp, #1, mul vl]
|
|
; CHECK-NEXT: sub x8, x8, #1
|
|
; CHECK-NEXT: str z0, [sp]
|
|
; CHECK-NEXT: cmp x9, x8
|
|
; CHECK-NEXT: csel x8, x9, x8, lo
|
|
; CHECK-NEXT: mov x9, sp
|
|
; CHECK-NEXT: ldrb w0, [x9, x8]
|
|
; CHECK-NEXT: addvl sp, sp, #2
|
|
; CHECK-NEXT: ldr x29, [sp], #16 // 8-byte Folded Reload
|
|
; CHECK-NEXT: ret
|
|
%ext = extractelement <vscale x 32 x i8> %a, i32 %idx
|
|
ret i8 %ext
|
|
}
|
|
|
|
define i16 @split_extract_16i16_idx(<vscale x 16 x i16> %a, i32 %idx) {
|
|
; CHECK-LABEL: split_extract_16i16_idx:
|
|
; CHECK: // %bb.0:
|
|
; CHECK-NEXT: str x29, [sp, #-16]! // 8-byte Folded Spill
|
|
; CHECK-NEXT: addvl sp, sp, #-2
|
|
; CHECK-NEXT: .cfi_escape 0x0f, 0x08, 0x8f, 0x10, 0x92, 0x2e, 0x00, 0x40, 0x1e, 0x22 // sp + 16 + 16 * VG
|
|
; CHECK-NEXT: .cfi_offset w29, -16
|
|
; CHECK-NEXT: rdvl x8, #1
|
|
; CHECK-NEXT: mov w9, w0
|
|
; CHECK-NEXT: str z1, [sp, #1, mul vl]
|
|
; CHECK-NEXT: sub x8, x8, #1
|
|
; CHECK-NEXT: str z0, [sp]
|
|
; CHECK-NEXT: cmp x9, x8
|
|
; CHECK-NEXT: csel x8, x9, x8, lo
|
|
; CHECK-NEXT: mov x9, sp
|
|
; CHECK-NEXT: ldrh w0, [x9, x8, lsl #1]
|
|
; CHECK-NEXT: addvl sp, sp, #2
|
|
; CHECK-NEXT: ldr x29, [sp], #16 // 8-byte Folded Reload
|
|
; CHECK-NEXT: ret
|
|
%ext = extractelement <vscale x 16 x i16> %a, i32 %idx
|
|
ret i16 %ext
|
|
}
|
|
|
|
define i32 @split_extract_8i32_idx(<vscale x 8 x i32> %a, i32 %idx) {
|
|
; CHECK-LABEL: split_extract_8i32_idx:
|
|
; CHECK: // %bb.0:
|
|
; CHECK-NEXT: str x29, [sp, #-16]! // 8-byte Folded Spill
|
|
; CHECK-NEXT: addvl sp, sp, #-2
|
|
; CHECK-NEXT: .cfi_escape 0x0f, 0x08, 0x8f, 0x10, 0x92, 0x2e, 0x00, 0x40, 0x1e, 0x22 // sp + 16 + 16 * VG
|
|
; CHECK-NEXT: .cfi_offset w29, -16
|
|
; CHECK-NEXT: cnth x8
|
|
; CHECK-NEXT: mov w9, w0
|
|
; CHECK-NEXT: str z1, [sp, #1, mul vl]
|
|
; CHECK-NEXT: sub x8, x8, #1
|
|
; CHECK-NEXT: str z0, [sp]
|
|
; CHECK-NEXT: cmp x9, x8
|
|
; CHECK-NEXT: csel x8, x9, x8, lo
|
|
; CHECK-NEXT: mov x9, sp
|
|
; CHECK-NEXT: ldr w0, [x9, x8, lsl #2]
|
|
; CHECK-NEXT: addvl sp, sp, #2
|
|
; CHECK-NEXT: ldr x29, [sp], #16 // 8-byte Folded Reload
|
|
; CHECK-NEXT: ret
|
|
%ext = extractelement <vscale x 8 x i32> %a, i32 %idx
|
|
ret i32 %ext
|
|
}
|
|
|
|
define i64 @split_extract_8i64_idx(<vscale x 8 x i64> %a, i32 %idx) {
|
|
; CHECK-LABEL: split_extract_8i64_idx:
|
|
; CHECK: // %bb.0:
|
|
; CHECK-NEXT: str x29, [sp, #-16]! // 8-byte Folded Spill
|
|
; CHECK-NEXT: addvl sp, sp, #-4
|
|
; CHECK-NEXT: .cfi_escape 0x0f, 0x09, 0x8f, 0x10, 0x92, 0x2e, 0x00, 0x11, 0x20, 0x1e, 0x22 // sp + 16 + 32 * VG
|
|
; CHECK-NEXT: .cfi_offset w29, -16
|
|
; CHECK-NEXT: cnth x8
|
|
; CHECK-NEXT: mov w9, w0
|
|
; CHECK-NEXT: str z3, [sp, #3, mul vl]
|
|
; CHECK-NEXT: sub x8, x8, #1
|
|
; CHECK-NEXT: str z2, [sp, #2, mul vl]
|
|
; CHECK-NEXT: cmp x9, x8
|
|
; CHECK-NEXT: str z1, [sp, #1, mul vl]
|
|
; CHECK-NEXT: str z0, [sp]
|
|
; CHECK-NEXT: csel x8, x9, x8, lo
|
|
; CHECK-NEXT: mov x9, sp
|
|
; CHECK-NEXT: ldr x0, [x9, x8, lsl #3]
|
|
; CHECK-NEXT: addvl sp, sp, #4
|
|
; CHECK-NEXT: ldr x29, [sp], #16 // 8-byte Folded Reload
|
|
; CHECK-NEXT: ret
|
|
%ext = extractelement <vscale x 8 x i64> %a, i32 %idx
|
|
ret i64 %ext
|
|
}
|
|
|
|
; EXTRACT VECTOR ELT, CONSTANT IDX
|
|
|
|
define i16 @promote_extract_4i16(<vscale x 4 x i16> %a) {
|
|
; CHECK-LABEL: promote_extract_4i16:
|
|
; CHECK: // %bb.0:
|
|
; CHECK-NEXT: mov w0, v0.s[1]
|
|
; CHECK-NEXT: ret
|
|
%ext = extractelement <vscale x 4 x i16> %a, i32 1
|
|
ret i16 %ext
|
|
}
|
|
|
|
define i8 @split_extract_32i8(<vscale x 32 x i8> %a) {
|
|
; CHECK-LABEL: split_extract_32i8:
|
|
; CHECK: // %bb.0:
|
|
; CHECK-NEXT: umov w0, v0.b[3]
|
|
; CHECK-NEXT: ret
|
|
%ext = extractelement <vscale x 32 x i8> %a, i32 3
|
|
ret i8 %ext
|
|
}
|
|
|
|
define i16 @split_extract_16i16(<vscale x 16 x i16> %a) {
|
|
; CHECK-LABEL: split_extract_16i16:
|
|
; CHECK: // %bb.0:
|
|
; CHECK-NEXT: str x29, [sp, #-16]! // 8-byte Folded Spill
|
|
; CHECK-NEXT: addvl sp, sp, #-2
|
|
; CHECK-NEXT: .cfi_escape 0x0f, 0x08, 0x8f, 0x10, 0x92, 0x2e, 0x00, 0x40, 0x1e, 0x22 // sp + 16 + 16 * VG
|
|
; CHECK-NEXT: .cfi_offset w29, -16
|
|
; CHECK-NEXT: rdvl x8, #1
|
|
; CHECK-NEXT: mov w9, #128 // =0x80
|
|
; CHECK-NEXT: str z1, [sp, #1, mul vl]
|
|
; CHECK-NEXT: sub x8, x8, #1
|
|
; CHECK-NEXT: str z0, [sp]
|
|
; CHECK-NEXT: cmp x8, #128
|
|
; CHECK-NEXT: csel x8, x8, x9, lo
|
|
; CHECK-NEXT: mov x9, sp
|
|
; CHECK-NEXT: ldrh w0, [x9, x8, lsl #1]
|
|
; CHECK-NEXT: addvl sp, sp, #2
|
|
; CHECK-NEXT: ldr x29, [sp], #16 // 8-byte Folded Reload
|
|
; CHECK-NEXT: ret
|
|
%ext = extractelement <vscale x 16 x i16> %a, i32 128
|
|
ret i16 %ext
|
|
}
|
|
|
|
define i32 @split_extract_16i32(<vscale x 16 x i32> %a) {
|
|
; CHECK-LABEL: split_extract_16i32:
|
|
; CHECK: // %bb.0:
|
|
; CHECK-NEXT: str x29, [sp, #-16]! // 8-byte Folded Spill
|
|
; CHECK-NEXT: addvl sp, sp, #-4
|
|
; CHECK-NEXT: .cfi_escape 0x0f, 0x09, 0x8f, 0x10, 0x92, 0x2e, 0x00, 0x11, 0x20, 0x1e, 0x22 // sp + 16 + 32 * VG
|
|
; CHECK-NEXT: .cfi_offset w29, -16
|
|
; CHECK-NEXT: rdvl x8, #1
|
|
; CHECK-NEXT: mov w9, #34464 // =0x86a0
|
|
; CHECK-NEXT: str z3, [sp, #3, mul vl]
|
|
; CHECK-NEXT: movk w9, #1, lsl #16
|
|
; CHECK-NEXT: sub x8, x8, #1
|
|
; CHECK-NEXT: str z2, [sp, #2, mul vl]
|
|
; CHECK-NEXT: cmp x8, x9
|
|
; CHECK-NEXT: str z1, [sp, #1, mul vl]
|
|
; CHECK-NEXT: str z0, [sp]
|
|
; CHECK-NEXT: csel x8, x8, x9, lo
|
|
; CHECK-NEXT: mov x9, sp
|
|
; CHECK-NEXT: ldr w0, [x9, x8, lsl #2]
|
|
; CHECK-NEXT: addvl sp, sp, #4
|
|
; CHECK-NEXT: ldr x29, [sp], #16 // 8-byte Folded Reload
|
|
; CHECK-NEXT: ret
|
|
%ext = extractelement <vscale x 16 x i32> %a, i32 100000
|
|
ret i32 %ext
|
|
}
|
|
|
|
define i64 @split_extract_4i64(<vscale x 4 x i64> %a) {
|
|
; CHECK-LABEL: split_extract_4i64:
|
|
; CHECK: // %bb.0:
|
|
; CHECK-NEXT: str x29, [sp, #-16]! // 8-byte Folded Spill
|
|
; CHECK-NEXT: addvl sp, sp, #-2
|
|
; CHECK-NEXT: .cfi_escape 0x0f, 0x08, 0x8f, 0x10, 0x92, 0x2e, 0x00, 0x40, 0x1e, 0x22 // sp + 16 + 16 * VG
|
|
; CHECK-NEXT: .cfi_offset w29, -16
|
|
; CHECK-NEXT: cntw x8
|
|
; CHECK-NEXT: mov w9, #10 // =0xa
|
|
; CHECK-NEXT: str z1, [sp, #1, mul vl]
|
|
; CHECK-NEXT: sub x8, x8, #1
|
|
; CHECK-NEXT: str z0, [sp]
|
|
; CHECK-NEXT: cmp x8, #10
|
|
; CHECK-NEXT: csel x8, x8, x9, lo
|
|
; CHECK-NEXT: mov x9, sp
|
|
; CHECK-NEXT: ldr x0, [x9, x8, lsl #3]
|
|
; CHECK-NEXT: addvl sp, sp, #2
|
|
; CHECK-NEXT: ldr x29, [sp], #16 // 8-byte Folded Reload
|
|
; CHECK-NEXT: ret
|
|
%ext = extractelement <vscale x 4 x i64> %a, i32 10
|
|
ret i64 %ext
|
|
}
|