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
514 lines
16 KiB
LLVM
514 lines
16 KiB
LLVM
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
|
|
; RUN: llc < %s -mtriple=aarch64-- | FileCheck %s --check-prefixes=CHECK,CHECK-SD
|
|
; RUN: llc < %s -mtriple=aarch64-- -global-isel -global-isel-abort=2 2>&1 | FileCheck %s --check-prefixes=CHECK,CHECK-GI
|
|
|
|
; CHECK-GI: warning: Instruction selection used fallback path for v16i4
|
|
; CHECK-GI-NEXT: warning: Instruction selection used fallback path for v16i1
|
|
|
|
define <16 x i8> @v16i8(<16 x i8> %x, <16 x i8> %y) nounwind {
|
|
; CHECK-LABEL: v16i8:
|
|
; CHECK: // %bb.0:
|
|
; CHECK-NEXT: uqadd v0.16b, v0.16b, v1.16b
|
|
; CHECK-NEXT: ret
|
|
%z = call <16 x i8> @llvm.uadd.sat.v16i8(<16 x i8> %x, <16 x i8> %y)
|
|
ret <16 x i8> %z
|
|
}
|
|
|
|
define <32 x i8> @v32i8(<32 x i8> %x, <32 x i8> %y) nounwind {
|
|
; CHECK-SD-LABEL: v32i8:
|
|
; CHECK-SD: // %bb.0:
|
|
; CHECK-SD-NEXT: uqadd v1.16b, v1.16b, v3.16b
|
|
; CHECK-SD-NEXT: uqadd v0.16b, v0.16b, v2.16b
|
|
; CHECK-SD-NEXT: ret
|
|
;
|
|
; CHECK-GI-LABEL: v32i8:
|
|
; CHECK-GI: // %bb.0:
|
|
; CHECK-GI-NEXT: uqadd v0.16b, v0.16b, v2.16b
|
|
; CHECK-GI-NEXT: uqadd v1.16b, v1.16b, v3.16b
|
|
; CHECK-GI-NEXT: ret
|
|
%z = call <32 x i8> @llvm.uadd.sat.v32i8(<32 x i8> %x, <32 x i8> %y)
|
|
ret <32 x i8> %z
|
|
}
|
|
|
|
define <64 x i8> @v64i8(<64 x i8> %x, <64 x i8> %y) nounwind {
|
|
; CHECK-SD-LABEL: v64i8:
|
|
; CHECK-SD: // %bb.0:
|
|
; CHECK-SD-NEXT: uqadd v2.16b, v2.16b, v6.16b
|
|
; CHECK-SD-NEXT: uqadd v0.16b, v0.16b, v4.16b
|
|
; CHECK-SD-NEXT: uqadd v1.16b, v1.16b, v5.16b
|
|
; CHECK-SD-NEXT: uqadd v3.16b, v3.16b, v7.16b
|
|
; CHECK-SD-NEXT: ret
|
|
;
|
|
; CHECK-GI-LABEL: v64i8:
|
|
; CHECK-GI: // %bb.0:
|
|
; CHECK-GI-NEXT: uqadd v0.16b, v0.16b, v4.16b
|
|
; CHECK-GI-NEXT: uqadd v1.16b, v1.16b, v5.16b
|
|
; CHECK-GI-NEXT: uqadd v2.16b, v2.16b, v6.16b
|
|
; CHECK-GI-NEXT: uqadd v3.16b, v3.16b, v7.16b
|
|
; CHECK-GI-NEXT: ret
|
|
%z = call <64 x i8> @llvm.uadd.sat.v64i8(<64 x i8> %x, <64 x i8> %y)
|
|
ret <64 x i8> %z
|
|
}
|
|
|
|
define <8 x i16> @v8i16(<8 x i16> %x, <8 x i16> %y) nounwind {
|
|
; CHECK-LABEL: v8i16:
|
|
; CHECK: // %bb.0:
|
|
; CHECK-NEXT: uqadd v0.8h, v0.8h, v1.8h
|
|
; CHECK-NEXT: ret
|
|
%z = call <8 x i16> @llvm.uadd.sat.v8i16(<8 x i16> %x, <8 x i16> %y)
|
|
ret <8 x i16> %z
|
|
}
|
|
|
|
define <16 x i16> @v16i16(<16 x i16> %x, <16 x i16> %y) nounwind {
|
|
; CHECK-SD-LABEL: v16i16:
|
|
; CHECK-SD: // %bb.0:
|
|
; CHECK-SD-NEXT: uqadd v1.8h, v1.8h, v3.8h
|
|
; CHECK-SD-NEXT: uqadd v0.8h, v0.8h, v2.8h
|
|
; CHECK-SD-NEXT: ret
|
|
;
|
|
; CHECK-GI-LABEL: v16i16:
|
|
; CHECK-GI: // %bb.0:
|
|
; CHECK-GI-NEXT: uqadd v0.8h, v0.8h, v2.8h
|
|
; CHECK-GI-NEXT: uqadd v1.8h, v1.8h, v3.8h
|
|
; CHECK-GI-NEXT: ret
|
|
%z = call <16 x i16> @llvm.uadd.sat.v16i16(<16 x i16> %x, <16 x i16> %y)
|
|
ret <16 x i16> %z
|
|
}
|
|
|
|
define <32 x i16> @v32i16(<32 x i16> %x, <32 x i16> %y) nounwind {
|
|
; CHECK-SD-LABEL: v32i16:
|
|
; CHECK-SD: // %bb.0:
|
|
; CHECK-SD-NEXT: uqadd v2.8h, v2.8h, v6.8h
|
|
; CHECK-SD-NEXT: uqadd v0.8h, v0.8h, v4.8h
|
|
; CHECK-SD-NEXT: uqadd v1.8h, v1.8h, v5.8h
|
|
; CHECK-SD-NEXT: uqadd v3.8h, v3.8h, v7.8h
|
|
; CHECK-SD-NEXT: ret
|
|
;
|
|
; CHECK-GI-LABEL: v32i16:
|
|
; CHECK-GI: // %bb.0:
|
|
; CHECK-GI-NEXT: uqadd v0.8h, v0.8h, v4.8h
|
|
; CHECK-GI-NEXT: uqadd v1.8h, v1.8h, v5.8h
|
|
; CHECK-GI-NEXT: uqadd v2.8h, v2.8h, v6.8h
|
|
; CHECK-GI-NEXT: uqadd v3.8h, v3.8h, v7.8h
|
|
; CHECK-GI-NEXT: ret
|
|
%z = call <32 x i16> @llvm.uadd.sat.v32i16(<32 x i16> %x, <32 x i16> %y)
|
|
ret <32 x i16> %z
|
|
}
|
|
|
|
define void @v8i8(ptr %px, ptr %py, ptr %pz) nounwind {
|
|
; CHECK-LABEL: v8i8:
|
|
; CHECK: // %bb.0:
|
|
; CHECK-NEXT: ldr d0, [x0]
|
|
; CHECK-NEXT: ldr d1, [x1]
|
|
; CHECK-NEXT: uqadd v0.8b, v0.8b, v1.8b
|
|
; CHECK-NEXT: str d0, [x2]
|
|
; CHECK-NEXT: ret
|
|
%x = load <8 x i8>, ptr %px
|
|
%y = load <8 x i8>, ptr %py
|
|
%z = call <8 x i8> @llvm.uadd.sat.v8i8(<8 x i8> %x, <8 x i8> %y)
|
|
store <8 x i8> %z, ptr %pz
|
|
ret void
|
|
}
|
|
|
|
define void @v4i8(ptr %px, ptr %py, ptr %pz) nounwind {
|
|
; CHECK-SD-LABEL: v4i8:
|
|
; CHECK-SD: // %bb.0:
|
|
; CHECK-SD-NEXT: ldr s0, [x0]
|
|
; CHECK-SD-NEXT: ldr s1, [x1]
|
|
; CHECK-SD-NEXT: movi d2, #0xff00ff00ff00ff
|
|
; CHECK-SD-NEXT: uaddl v0.8h, v0.8b, v1.8b
|
|
; CHECK-SD-NEXT: umin v0.4h, v0.4h, v2.4h
|
|
; CHECK-SD-NEXT: uzp1 v0.8b, v0.8b, v0.8b
|
|
; CHECK-SD-NEXT: str s0, [x2]
|
|
; CHECK-SD-NEXT: ret
|
|
;
|
|
; CHECK-GI-LABEL: v4i8:
|
|
; CHECK-GI: // %bb.0:
|
|
; CHECK-GI-NEXT: ldr s0, [x0]
|
|
; CHECK-GI-NEXT: ldr s1, [x1]
|
|
; CHECK-GI-NEXT: uqadd v0.8b, v0.8b, v1.8b
|
|
; CHECK-GI-NEXT: str s0, [x2]
|
|
; CHECK-GI-NEXT: ret
|
|
%x = load <4 x i8>, ptr %px
|
|
%y = load <4 x i8>, ptr %py
|
|
%z = call <4 x i8> @llvm.uadd.sat.v4i8(<4 x i8> %x, <4 x i8> %y)
|
|
store <4 x i8> %z, ptr %pz
|
|
ret void
|
|
}
|
|
|
|
define void @v2i8(ptr %px, ptr %py, ptr %pz) nounwind {
|
|
; CHECK-SD-LABEL: v2i8:
|
|
; CHECK-SD: // %bb.0:
|
|
; CHECK-SD-NEXT: ldr h0, [x0]
|
|
; CHECK-SD-NEXT: ldr h1, [x1]
|
|
; CHECK-SD-NEXT: movi d2, #0x0000ff000000ff
|
|
; CHECK-SD-NEXT: ushll v0.8h, v0.8b, #0
|
|
; CHECK-SD-NEXT: ushll v1.8h, v1.8b, #0
|
|
; CHECK-SD-NEXT: uaddl v0.4s, v0.4h, v1.4h
|
|
; CHECK-SD-NEXT: umin v0.2s, v0.2s, v2.2s
|
|
; CHECK-SD-NEXT: mov s1, v0.s[1]
|
|
; CHECK-SD-NEXT: str b0, [x2]
|
|
; CHECK-SD-NEXT: stur b1, [x2, #1]
|
|
; CHECK-SD-NEXT: ret
|
|
;
|
|
; CHECK-GI-LABEL: v2i8:
|
|
; CHECK-GI: // %bb.0:
|
|
; CHECK-GI-NEXT: ldr b0, [x0]
|
|
; CHECK-GI-NEXT: ldr b1, [x1]
|
|
; CHECK-GI-NEXT: add x8, x0, #1
|
|
; CHECK-GI-NEXT: add x9, x1, #1
|
|
; CHECK-GI-NEXT: ld1 { v0.b }[1], [x8]
|
|
; CHECK-GI-NEXT: ld1 { v1.b }[1], [x9]
|
|
; CHECK-GI-NEXT: add x8, x2, #1
|
|
; CHECK-GI-NEXT: uqadd v0.8b, v0.8b, v1.8b
|
|
; CHECK-GI-NEXT: st1 { v0.b }[0], [x2]
|
|
; CHECK-GI-NEXT: st1 { v0.b }[1], [x8]
|
|
; CHECK-GI-NEXT: ret
|
|
%x = load <2 x i8>, ptr %px
|
|
%y = load <2 x i8>, ptr %py
|
|
%z = call <2 x i8> @llvm.uadd.sat.v2i8(<2 x i8> %x, <2 x i8> %y)
|
|
store <2 x i8> %z, ptr %pz
|
|
ret void
|
|
}
|
|
|
|
define void @v4i16(ptr %px, ptr %py, ptr %pz) nounwind {
|
|
; CHECK-LABEL: v4i16:
|
|
; CHECK: // %bb.0:
|
|
; CHECK-NEXT: ldr d0, [x0]
|
|
; CHECK-NEXT: ldr d1, [x1]
|
|
; CHECK-NEXT: uqadd v0.4h, v0.4h, v1.4h
|
|
; CHECK-NEXT: str d0, [x2]
|
|
; CHECK-NEXT: ret
|
|
%x = load <4 x i16>, ptr %px
|
|
%y = load <4 x i16>, ptr %py
|
|
%z = call <4 x i16> @llvm.uadd.sat.v4i16(<4 x i16> %x, <4 x i16> %y)
|
|
store <4 x i16> %z, ptr %pz
|
|
ret void
|
|
}
|
|
|
|
define void @v2i16(ptr %px, ptr %py, ptr %pz) nounwind {
|
|
; CHECK-SD-LABEL: v2i16:
|
|
; CHECK-SD: // %bb.0:
|
|
; CHECK-SD-NEXT: ldr s0, [x0]
|
|
; CHECK-SD-NEXT: ldr s1, [x1]
|
|
; CHECK-SD-NEXT: movi d2, #0x00ffff0000ffff
|
|
; CHECK-SD-NEXT: uaddl v0.4s, v0.4h, v1.4h
|
|
; CHECK-SD-NEXT: umin v0.2s, v0.2s, v2.2s
|
|
; CHECK-SD-NEXT: mov s1, v0.s[1]
|
|
; CHECK-SD-NEXT: str h0, [x2]
|
|
; CHECK-SD-NEXT: str h1, [x2, #2]
|
|
; CHECK-SD-NEXT: ret
|
|
;
|
|
; CHECK-GI-LABEL: v2i16:
|
|
; CHECK-GI: // %bb.0:
|
|
; CHECK-GI-NEXT: ldr h0, [x0]
|
|
; CHECK-GI-NEXT: ldr h1, [x1]
|
|
; CHECK-GI-NEXT: add x8, x0, #2
|
|
; CHECK-GI-NEXT: add x9, x1, #2
|
|
; CHECK-GI-NEXT: ld1 { v0.h }[1], [x8]
|
|
; CHECK-GI-NEXT: ld1 { v1.h }[1], [x9]
|
|
; CHECK-GI-NEXT: add x8, x2, #2
|
|
; CHECK-GI-NEXT: uqadd v0.4h, v0.4h, v1.4h
|
|
; CHECK-GI-NEXT: str h0, [x2]
|
|
; CHECK-GI-NEXT: st1 { v0.h }[1], [x8]
|
|
; CHECK-GI-NEXT: ret
|
|
%x = load <2 x i16>, ptr %px
|
|
%y = load <2 x i16>, ptr %py
|
|
%z = call <2 x i16> @llvm.uadd.sat.v2i16(<2 x i16> %x, <2 x i16> %y)
|
|
store <2 x i16> %z, ptr %pz
|
|
ret void
|
|
}
|
|
|
|
define <12 x i8> @v12i8(<12 x i8> %x, <12 x i8> %y) nounwind {
|
|
; CHECK-LABEL: v12i8:
|
|
; CHECK: // %bb.0:
|
|
; CHECK-NEXT: uqadd v0.16b, v0.16b, v1.16b
|
|
; CHECK-NEXT: ret
|
|
%z = call <12 x i8> @llvm.uadd.sat.v12i8(<12 x i8> %x, <12 x i8> %y)
|
|
ret <12 x i8> %z
|
|
}
|
|
|
|
define void @v12i16(ptr %px, ptr %py, ptr %pz) nounwind {
|
|
; CHECK-SD-LABEL: v12i16:
|
|
; CHECK-SD: // %bb.0:
|
|
; CHECK-SD-NEXT: ldp q0, q3, [x1]
|
|
; CHECK-SD-NEXT: ldp q1, q2, [x0]
|
|
; CHECK-SD-NEXT: uqadd v0.8h, v1.8h, v0.8h
|
|
; CHECK-SD-NEXT: uqadd v1.8h, v2.8h, v3.8h
|
|
; CHECK-SD-NEXT: str q0, [x2]
|
|
; CHECK-SD-NEXT: str d1, [x2, #16]
|
|
; CHECK-SD-NEXT: ret
|
|
;
|
|
; CHECK-GI-LABEL: v12i16:
|
|
; CHECK-GI: // %bb.0:
|
|
; CHECK-GI-NEXT: ldr q0, [x0]
|
|
; CHECK-GI-NEXT: ldr q1, [x1]
|
|
; CHECK-GI-NEXT: ldr d2, [x0, #16]
|
|
; CHECK-GI-NEXT: ldr d3, [x1, #16]
|
|
; CHECK-GI-NEXT: uqadd v0.8h, v0.8h, v1.8h
|
|
; CHECK-GI-NEXT: uqadd v1.4h, v2.4h, v3.4h
|
|
; CHECK-GI-NEXT: str q0, [x2]
|
|
; CHECK-GI-NEXT: str d1, [x2, #16]
|
|
; CHECK-GI-NEXT: ret
|
|
%x = load <12 x i16>, ptr %px
|
|
%y = load <12 x i16>, ptr %py
|
|
%z = call <12 x i16> @llvm.uadd.sat.v12i16(<12 x i16> %x, <12 x i16> %y)
|
|
store <12 x i16> %z, ptr %pz
|
|
ret void
|
|
}
|
|
|
|
define void @v1i8(ptr %px, ptr %py, ptr %pz) nounwind {
|
|
; CHECK-SD-LABEL: v1i8:
|
|
; CHECK-SD: // %bb.0:
|
|
; CHECK-SD-NEXT: ldr b0, [x0]
|
|
; CHECK-SD-NEXT: ldr b1, [x1]
|
|
; CHECK-SD-NEXT: uqadd v0.8b, v0.8b, v1.8b
|
|
; CHECK-SD-NEXT: str b0, [x2]
|
|
; CHECK-SD-NEXT: ret
|
|
;
|
|
; CHECK-GI-LABEL: v1i8:
|
|
; CHECK-GI: // %bb.0:
|
|
; CHECK-GI-NEXT: ldrb w8, [x0]
|
|
; CHECK-GI-NEXT: ldrb w9, [x1]
|
|
; CHECK-GI-NEXT: add w8, w8, w9
|
|
; CHECK-GI-NEXT: cmp w8, w8, uxtb
|
|
; CHECK-GI-NEXT: csinv w8, w8, wzr, eq
|
|
; CHECK-GI-NEXT: strb w8, [x2]
|
|
; CHECK-GI-NEXT: ret
|
|
%x = load <1 x i8>, ptr %px
|
|
%y = load <1 x i8>, ptr %py
|
|
%z = call <1 x i8> @llvm.uadd.sat.v1i8(<1 x i8> %x, <1 x i8> %y)
|
|
store <1 x i8> %z, ptr %pz
|
|
ret void
|
|
}
|
|
|
|
define void @v1i16(ptr %px, ptr %py, ptr %pz) nounwind {
|
|
; CHECK-SD-LABEL: v1i16:
|
|
; CHECK-SD: // %bb.0:
|
|
; CHECK-SD-NEXT: ldr h0, [x0]
|
|
; CHECK-SD-NEXT: ldr h1, [x1]
|
|
; CHECK-SD-NEXT: uqadd v0.4h, v0.4h, v1.4h
|
|
; CHECK-SD-NEXT: str h0, [x2]
|
|
; CHECK-SD-NEXT: ret
|
|
;
|
|
; CHECK-GI-LABEL: v1i16:
|
|
; CHECK-GI: // %bb.0:
|
|
; CHECK-GI-NEXT: ldrh w8, [x0]
|
|
; CHECK-GI-NEXT: ldrh w9, [x1]
|
|
; CHECK-GI-NEXT: add w8, w8, w9
|
|
; CHECK-GI-NEXT: cmp w8, w8, uxth
|
|
; CHECK-GI-NEXT: csinv w8, w8, wzr, eq
|
|
; CHECK-GI-NEXT: strh w8, [x2]
|
|
; CHECK-GI-NEXT: ret
|
|
%x = load <1 x i16>, ptr %px
|
|
%y = load <1 x i16>, ptr %py
|
|
%z = call <1 x i16> @llvm.uadd.sat.v1i16(<1 x i16> %x, <1 x i16> %y)
|
|
store <1 x i16> %z, ptr %pz
|
|
ret void
|
|
}
|
|
|
|
define <16 x i4> @v16i4(<16 x i4> %x, <16 x i4> %y) nounwind {
|
|
; CHECK-LABEL: v16i4:
|
|
; CHECK: // %bb.0:
|
|
; CHECK-NEXT: movi v2.16b, #15
|
|
; CHECK-NEXT: and v1.16b, v1.16b, v2.16b
|
|
; CHECK-NEXT: and v0.16b, v0.16b, v2.16b
|
|
; CHECK-NEXT: add v0.16b, v0.16b, v1.16b
|
|
; CHECK-NEXT: umin v0.16b, v0.16b, v2.16b
|
|
; CHECK-NEXT: ret
|
|
%z = call <16 x i4> @llvm.uadd.sat.v16i4(<16 x i4> %x, <16 x i4> %y)
|
|
ret <16 x i4> %z
|
|
}
|
|
|
|
define <16 x i1> @v16i1(<16 x i1> %x, <16 x i1> %y) nounwind {
|
|
; CHECK-LABEL: v16i1:
|
|
; CHECK: // %bb.0:
|
|
; CHECK-NEXT: orr v0.16b, v0.16b, v1.16b
|
|
; CHECK-NEXT: ret
|
|
%z = call <16 x i1> @llvm.uadd.sat.v16i1(<16 x i1> %x, <16 x i1> %y)
|
|
ret <16 x i1> %z
|
|
}
|
|
|
|
define void @v1i32(ptr %px, ptr %py, ptr %pz) nounwind {
|
|
; CHECK-SD-LABEL: v1i32:
|
|
; CHECK-SD: // %bb.0:
|
|
; CHECK-SD-NEXT: ldr s0, [x0]
|
|
; CHECK-SD-NEXT: ldr s1, [x1]
|
|
; CHECK-SD-NEXT: uqadd v0.2s, v0.2s, v1.2s
|
|
; CHECK-SD-NEXT: str s0, [x2]
|
|
; CHECK-SD-NEXT: ret
|
|
;
|
|
; CHECK-GI-LABEL: v1i32:
|
|
; CHECK-GI: // %bb.0:
|
|
; CHECK-GI-NEXT: ldr w8, [x0]
|
|
; CHECK-GI-NEXT: ldr w9, [x1]
|
|
; CHECK-GI-NEXT: adds w8, w8, w9
|
|
; CHECK-GI-NEXT: cset w9, hs
|
|
; CHECK-GI-NEXT: tst w9, #0x1
|
|
; CHECK-GI-NEXT: csinv w8, w8, wzr, eq
|
|
; CHECK-GI-NEXT: str w8, [x2]
|
|
; CHECK-GI-NEXT: ret
|
|
%x = load <1 x i32>, ptr %px
|
|
%y = load <1 x i32>, ptr %py
|
|
%z = call <1 x i32> @llvm.uadd.sat.v1i32(<1 x i32> %x, <1 x i32> %y)
|
|
store <1 x i32> %z, ptr %pz
|
|
ret void
|
|
}
|
|
|
|
define <2 x i32> @v2i32(<2 x i32> %x, <2 x i32> %y) nounwind {
|
|
; CHECK-LABEL: v2i32:
|
|
; CHECK: // %bb.0:
|
|
; CHECK-NEXT: uqadd v0.2s, v0.2s, v1.2s
|
|
; CHECK-NEXT: ret
|
|
%z = call <2 x i32> @llvm.uadd.sat.v2i32(<2 x i32> %x, <2 x i32> %y)
|
|
ret <2 x i32> %z
|
|
}
|
|
|
|
define <4 x i32> @v4i32(<4 x i32> %x, <4 x i32> %y) nounwind {
|
|
; CHECK-LABEL: v4i32:
|
|
; CHECK: // %bb.0:
|
|
; CHECK-NEXT: uqadd v0.4s, v0.4s, v1.4s
|
|
; CHECK-NEXT: ret
|
|
%z = call <4 x i32> @llvm.uadd.sat.v4i32(<4 x i32> %x, <4 x i32> %y)
|
|
ret <4 x i32> %z
|
|
}
|
|
|
|
define <8 x i32> @v8i32(<8 x i32> %x, <8 x i32> %y) nounwind {
|
|
; CHECK-SD-LABEL: v8i32:
|
|
; CHECK-SD: // %bb.0:
|
|
; CHECK-SD-NEXT: uqadd v1.4s, v1.4s, v3.4s
|
|
; CHECK-SD-NEXT: uqadd v0.4s, v0.4s, v2.4s
|
|
; CHECK-SD-NEXT: ret
|
|
;
|
|
; CHECK-GI-LABEL: v8i32:
|
|
; CHECK-GI: // %bb.0:
|
|
; CHECK-GI-NEXT: uqadd v0.4s, v0.4s, v2.4s
|
|
; CHECK-GI-NEXT: uqadd v1.4s, v1.4s, v3.4s
|
|
; CHECK-GI-NEXT: ret
|
|
%z = call <8 x i32> @llvm.uadd.sat.v8i32(<8 x i32> %x, <8 x i32> %y)
|
|
ret <8 x i32> %z
|
|
}
|
|
|
|
define <16 x i32> @v16i32(<16 x i32> %x, <16 x i32> %y) nounwind {
|
|
; CHECK-SD-LABEL: v16i32:
|
|
; CHECK-SD: // %bb.0:
|
|
; CHECK-SD-NEXT: uqadd v2.4s, v2.4s, v6.4s
|
|
; CHECK-SD-NEXT: uqadd v0.4s, v0.4s, v4.4s
|
|
; CHECK-SD-NEXT: uqadd v1.4s, v1.4s, v5.4s
|
|
; CHECK-SD-NEXT: uqadd v3.4s, v3.4s, v7.4s
|
|
; CHECK-SD-NEXT: ret
|
|
;
|
|
; CHECK-GI-LABEL: v16i32:
|
|
; CHECK-GI: // %bb.0:
|
|
; CHECK-GI-NEXT: uqadd v0.4s, v0.4s, v4.4s
|
|
; CHECK-GI-NEXT: uqadd v1.4s, v1.4s, v5.4s
|
|
; CHECK-GI-NEXT: uqadd v2.4s, v2.4s, v6.4s
|
|
; CHECK-GI-NEXT: uqadd v3.4s, v3.4s, v7.4s
|
|
; CHECK-GI-NEXT: ret
|
|
%z = call <16 x i32> @llvm.uadd.sat.v16i32(<16 x i32> %x, <16 x i32> %y)
|
|
ret <16 x i32> %z
|
|
}
|
|
|
|
define void @v1i64(ptr %px, ptr %py, ptr %pz) nounwind {
|
|
; CHECK-SD-LABEL: v1i64:
|
|
; CHECK-SD: // %bb.0:
|
|
; CHECK-SD-NEXT: ldr d0, [x0]
|
|
; CHECK-SD-NEXT: ldr d1, [x1]
|
|
; CHECK-SD-NEXT: uqadd d0, d0, d1
|
|
; CHECK-SD-NEXT: str d0, [x2]
|
|
; CHECK-SD-NEXT: ret
|
|
;
|
|
; CHECK-GI-LABEL: v1i64:
|
|
; CHECK-GI: // %bb.0:
|
|
; CHECK-GI-NEXT: ldr x8, [x0]
|
|
; CHECK-GI-NEXT: ldr x9, [x1]
|
|
; CHECK-GI-NEXT: adds x8, x8, x9
|
|
; CHECK-GI-NEXT: cset w9, hs
|
|
; CHECK-GI-NEXT: tst w9, #0x1
|
|
; CHECK-GI-NEXT: csinv x8, x8, xzr, eq
|
|
; CHECK-GI-NEXT: str x8, [x2]
|
|
; CHECK-GI-NEXT: ret
|
|
%x = load <1 x i64>, ptr %px
|
|
%y = load <1 x i64>, ptr %py
|
|
%z = call <1 x i64> @llvm.uadd.sat.v1i64(<1 x i64> %x, <1 x i64> %y)
|
|
store <1 x i64> %z, ptr %pz
|
|
ret void
|
|
}
|
|
|
|
define <2 x i64> @v2i64(<2 x i64> %x, <2 x i64> %y) nounwind {
|
|
; CHECK-LABEL: v2i64:
|
|
; CHECK: // %bb.0:
|
|
; CHECK-NEXT: uqadd v0.2d, v0.2d, v1.2d
|
|
; CHECK-NEXT: ret
|
|
%z = call <2 x i64> @llvm.uadd.sat.v2i64(<2 x i64> %x, <2 x i64> %y)
|
|
ret <2 x i64> %z
|
|
}
|
|
|
|
define <4 x i64> @v4i64(<4 x i64> %x, <4 x i64> %y) nounwind {
|
|
; CHECK-SD-LABEL: v4i64:
|
|
; CHECK-SD: // %bb.0:
|
|
; CHECK-SD-NEXT: uqadd v1.2d, v1.2d, v3.2d
|
|
; CHECK-SD-NEXT: uqadd v0.2d, v0.2d, v2.2d
|
|
; CHECK-SD-NEXT: ret
|
|
;
|
|
; CHECK-GI-LABEL: v4i64:
|
|
; CHECK-GI: // %bb.0:
|
|
; CHECK-GI-NEXT: uqadd v0.2d, v0.2d, v2.2d
|
|
; CHECK-GI-NEXT: uqadd v1.2d, v1.2d, v3.2d
|
|
; CHECK-GI-NEXT: ret
|
|
%z = call <4 x i64> @llvm.uadd.sat.v4i64(<4 x i64> %x, <4 x i64> %y)
|
|
ret <4 x i64> %z
|
|
}
|
|
|
|
define <8 x i64> @v8i64(<8 x i64> %x, <8 x i64> %y) nounwind {
|
|
; CHECK-SD-LABEL: v8i64:
|
|
; CHECK-SD: // %bb.0:
|
|
; CHECK-SD-NEXT: uqadd v2.2d, v2.2d, v6.2d
|
|
; CHECK-SD-NEXT: uqadd v0.2d, v0.2d, v4.2d
|
|
; CHECK-SD-NEXT: uqadd v1.2d, v1.2d, v5.2d
|
|
; CHECK-SD-NEXT: uqadd v3.2d, v3.2d, v7.2d
|
|
; CHECK-SD-NEXT: ret
|
|
;
|
|
; CHECK-GI-LABEL: v8i64:
|
|
; CHECK-GI: // %bb.0:
|
|
; CHECK-GI-NEXT: uqadd v0.2d, v0.2d, v4.2d
|
|
; CHECK-GI-NEXT: uqadd v1.2d, v1.2d, v5.2d
|
|
; CHECK-GI-NEXT: uqadd v2.2d, v2.2d, v6.2d
|
|
; CHECK-GI-NEXT: uqadd v3.2d, v3.2d, v7.2d
|
|
; CHECK-GI-NEXT: ret
|
|
%z = call <8 x i64> @llvm.uadd.sat.v8i64(<8 x i64> %x, <8 x i64> %y)
|
|
ret <8 x i64> %z
|
|
}
|
|
|
|
define <2 x i128> @v2i128(<2 x i128> %x, <2 x i128> %y) nounwind {
|
|
; CHECK-SD-LABEL: v2i128:
|
|
; CHECK-SD: // %bb.0:
|
|
; CHECK-SD-NEXT: adds x8, x0, x4
|
|
; CHECK-SD-NEXT: adcs x9, x1, x5
|
|
; CHECK-SD-NEXT: csinv x0, x8, xzr, lo
|
|
; CHECK-SD-NEXT: csinv x1, x9, xzr, lo
|
|
; CHECK-SD-NEXT: adds x8, x2, x6
|
|
; CHECK-SD-NEXT: adcs x9, x3, x7
|
|
; CHECK-SD-NEXT: csinv x2, x8, xzr, lo
|
|
; CHECK-SD-NEXT: csinv x3, x9, xzr, lo
|
|
; CHECK-SD-NEXT: ret
|
|
;
|
|
; CHECK-GI-LABEL: v2i128:
|
|
; CHECK-GI: // %bb.0:
|
|
; CHECK-GI-NEXT: adds x8, x0, x4
|
|
; CHECK-GI-NEXT: adcs x9, x1, x5
|
|
; CHECK-GI-NEXT: cset w10, hs
|
|
; CHECK-GI-NEXT: tst w10, #0x1
|
|
; CHECK-GI-NEXT: csinv x0, x8, xzr, eq
|
|
; CHECK-GI-NEXT: csinv x1, x9, xzr, eq
|
|
; CHECK-GI-NEXT: adds x8, x2, x6
|
|
; CHECK-GI-NEXT: adcs x9, x3, x7
|
|
; CHECK-GI-NEXT: cset w10, hs
|
|
; CHECK-GI-NEXT: tst w10, #0x1
|
|
; CHECK-GI-NEXT: csinv x2, x8, xzr, eq
|
|
; CHECK-GI-NEXT: csinv x3, x9, xzr, eq
|
|
; CHECK-GI-NEXT: ret
|
|
%z = call <2 x i128> @llvm.uadd.sat.v2i128(<2 x i128> %x, <2 x i128> %y)
|
|
ret <2 x i128> %z
|
|
}
|