derek/TheRock
1
0 Fork 0
Code Issues Pull Requests Actions 1 Packages Projects Releases Wiki Activity
Files
1a182e2dac58879f215c0fad0f5f78c76dcafe56
TheRock/compiler/amd-llvm/llvm/test/CodeGen/AArch64/sve2-fcopysign.ll
T
derek.holloway 7502bf64a4
CodeQL / Analyze (actions) (push) Canceled after 0s
Details
CodeQL / Analyze (python) (push) Canceled after 0s
Details
pre-commit / pre-commit (push) Canceled after 0s
Details
Multi-Arch CI / setup (push) Canceled after 0s
Details
Publish build_manylinux_x86_64 images / publish_build_manylinux_x86_64 (push) Canceled after 0s
Details
Setup (Multi-Arch) / setup (push) Canceled after 0s
Details
Multi-Arch CI / Linux::${{ fromJSON(needs.setup.outputs.linux_build_config || '{}').build_variant_label || 'skip' }} (push) Canceled after 0s
Details
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Details
Publish no_rocm_image_ubi10 image / publish_no_rocm_image_ubi10 (push) Canceled after 0s
Details
Multi-Arch CI / Windows::${{ fromJSON(needs.setup.outputs.windows_build_config || '{}').build_variant_label || 'skip' }} (push) Canceled after 0s
Details
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Details
Multi-Arch CI / Manifest Diff (push) Canceled after 0s
Details
TheRock Manifest Diff Report / Generate Manifest Diff Report (push) Canceled after 0s
Details
Multi-Arch CI / CI Summary (push) Canceled after 0s
Details
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04 (push) Canceled after 0s
Details
Publish no_rocm_image_ubuntu24_04_ocl_rt images / publish_no_rocm_image_ubuntu24_04_ocl_rt (push) Canceled after 0s
Details
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Details
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Details
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_rocgdb (push) Canceled after 0s
Details
Publish no_rocm_image_ubuntu24_04_rocgdb images / publish_no_rocm_image_ubuntu24_04_rocgdb (push) Canceled after 0s
Details
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_ocl_rt (push) Canceled after 0s
Details
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Details
Unit Tests / Unit Tests :: ubuntu-24.04 (push) Canceled after 0s
Details
Unit Tests / Unit Tests :: windows-2022 (push) Canceled after 0s
Details
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_media (push) Canceled after 0s
Details
Unit Tests / Unit Tests Summary (push) Canceled after 0s
Details
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_openmpi (push) Canceled after 0s
Details
Flatten subprojects into a single base monorepo
2026-09-02 16:34:47 -07:00

442 lines
18 KiB
LLVM
Raw Blame History

; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
; RUN: llc -mattr=+sve2,+bf16 < %s -o - | FileCheck %s
target triple = "aarch64-unknown-linux-gnu"
;
; llvm.copysign.nxv2f16
;
define <vscale x 2 x half> @copysign_nxv2f16_nxv2f16(<vscale x 2 x half> %a, <vscale x 2 x half> %b) {
; CHECK-LABEL: copysign_nxv2f16_nxv2f16:
; CHECK: // %bb.0:
; CHECK-NEXT: mov z2.h, #32767 // =0x7fff
; CHECK-NEXT: bsl z0.d, z0.d, z1.d, z2.d
; CHECK-NEXT: ret
%r = call <vscale x 2 x half> @llvm.copysign.nxv2f16(<vscale x 2 x half> %a, <vscale x 2 x half> %b)
ret <vscale x 2 x half> %r
}
define <vscale x 2 x half> @copysign_nxv2f16_nxv2f32(<vscale x 2 x half> %a, <vscale x 2 x float> %b) {
; CHECK-LABEL: copysign_nxv2f16_nxv2f32:
; CHECK: // %bb.0:
; CHECK-NEXT: ptrue p0.d
; CHECK-NEXT: mov z2.h, #32767 // =0x7fff
; CHECK-NEXT: fcvt z1.h, p0/m, z1.s
; CHECK-NEXT: bsl z0.d, z0.d, z1.d, z2.d
; CHECK-NEXT: ret
%b.trunc = fptrunc <vscale x 2 x float> %b to <vscale x 2 x half>
%r = call <vscale x 2 x half> @llvm.copysign.nxv2f16(<vscale x 2 x half> %a, <vscale x 2 x half> %b.trunc)
ret <vscale x 2 x half> %r
}
define <vscale x 2 x half> @copysign_nxv2f16_nxv2f64(<vscale x 2 x half> %a, <vscale x 2 x double> %b) {
; CHECK-LABEL: copysign_nxv2f16_nxv2f64:
; CHECK: // %bb.0:
; CHECK-NEXT: ptrue p0.d
; CHECK-NEXT: mov z2.h, #32767 // =0x7fff
; CHECK-NEXT: fcvt z1.h, p0/m, z1.d
; CHECK-NEXT: bsl z0.d, z0.d, z1.d, z2.d
; CHECK-NEXT: ret
%b.trunc = fptrunc <vscale x 2 x double> %b to <vscale x 2 x half>
%r = call <vscale x 2 x half> @llvm.copysign.nxv2f16(<vscale x 2 x half> %a, <vscale x 2 x half> %b.trunc)
ret <vscale x 2 x half> %r
}
;
; llvm.copysign.nxv4f16
;
define <vscale x 4 x half> @copysign_nxv4f16_nxv4f16(<vscale x 4 x half> %a, <vscale x 4 x half> %b) {
; CHECK-LABEL: copysign_nxv4f16_nxv4f16:
; CHECK: // %bb.0:
; CHECK-NEXT: mov z2.h, #32767 // =0x7fff
; CHECK-NEXT: bsl z0.d, z0.d, z1.d, z2.d
; CHECK-NEXT: ret
%r = call <vscale x 4 x half> @llvm.copysign.nxv4f16(<vscale x 4 x half> %a, <vscale x 4 x half> %b)
ret <vscale x 4 x half> %r
}
define <vscale x 4 x half> @copysign_nxv4f16_nxv4f32(<vscale x 4 x half> %a, <vscale x 4 x float> %b) {
; CHECK-LABEL: copysign_nxv4f16_nxv4f32:
; CHECK: // %bb.0:
; CHECK-NEXT: ptrue p0.s
; CHECK-NEXT: mov z2.h, #32767 // =0x7fff
; CHECK-NEXT: fcvt z1.h, p0/m, z1.s
; CHECK-NEXT: bsl z0.d, z0.d, z1.d, z2.d
; CHECK-NEXT: ret
%b.trunc = fptrunc <vscale x 4 x float> %b to <vscale x 4 x half>
%r = call <vscale x 4 x half> @llvm.copysign.nxv4f16(<vscale x 4 x half> %a, <vscale x 4 x half> %b.trunc)
ret <vscale x 4 x half> %r
}
define <vscale x 4 x half> @copysign_nxv4f16_nxv4f64(<vscale x 4 x half> %a, <vscale x 4 x double> %b) {
; CHECK-LABEL: copysign_nxv4f16_nxv4f64:
; CHECK: // %bb.0:
; CHECK-NEXT: ptrue p0.d
; CHECK-NEXT: mov z3.h, #32767 // =0x7fff
; CHECK-NEXT: fcvt z2.h, p0/m, z2.d
; CHECK-NEXT: fcvt z1.h, p0/m, z1.d
; CHECK-NEXT: uzp1 z1.s, z1.s, z2.s
; CHECK-NEXT: bsl z0.d, z0.d, z1.d, z3.d
; CHECK-NEXT: ret
%b.trunc = fptrunc <vscale x 4 x double> %b to <vscale x 4 x half>
%r = call <vscale x 4 x half> @llvm.copysign.nxv4f16(<vscale x 4 x half> %a, <vscale x 4 x half> %b.trunc)
ret <vscale x 4 x half> %r
}
;
; llvm.copysign.nxv8f16
;
define <vscale x 8 x half> @copysign_nxv8f16_nxv8f16(<vscale x 8 x half> %a, <vscale x 8 x half> %b) {
; CHECK-LABEL: copysign_nxv8f16_nxv8f16:
; CHECK: // %bb.0:
; CHECK-NEXT: mov z2.h, #32767 // =0x7fff
; CHECK-NEXT: bsl z0.d, z0.d, z1.d, z2.d
; CHECK-NEXT: ret
%r = call <vscale x 8 x half> @llvm.copysign.nxv8f16(<vscale x 8 x half> %a, <vscale x 8 x half> %b)
ret <vscale x 8 x half> %r
}
define <vscale x 8 x half> @copysign_nxv8f16_nxv8f32(<vscale x 8 x half> %a, <vscale x 8 x float> %b) {
; CHECK-LABEL: copysign_nxv8f16_nxv8f32:
; CHECK: // %bb.0:
; CHECK-NEXT: ptrue p0.s
; CHECK-NEXT: mov z3.h, #32767 // =0x7fff
; CHECK-NEXT: fcvt z2.h, p0/m, z2.s
; CHECK-NEXT: fcvt z1.h, p0/m, z1.s
; CHECK-NEXT: uzp1 z1.h, z1.h, z2.h
; CHECK-NEXT: bsl z0.d, z0.d, z1.d, z3.d
; CHECK-NEXT: ret
%b.trunc = fptrunc <vscale x 8 x float> %b to <vscale x 8 x half>
%r = call <vscale x 8 x half> @llvm.copysign.nxv8f16(<vscale x 8 x half> %a, <vscale x 8 x half> %b.trunc)
ret <vscale x 8 x half> %r
}
define <vscale x 8 x half> @copysign_nxv8f16_nxv8f64(<vscale x 8 x half> %a, <vscale x 8 x double> %b) {
; CHECK-LABEL: copysign_nxv8f16_nxv8f64:
; CHECK: // %bb.0:
; CHECK-NEXT: ptrue p0.d
; CHECK-NEXT: fcvt z4.h, p0/m, z4.d
; CHECK-NEXT: fcvt z3.h, p0/m, z3.d
; CHECK-NEXT: fcvt z2.h, p0/m, z2.d
; CHECK-NEXT: fcvt z1.h, p0/m, z1.d
; CHECK-NEXT: uzp1 z3.s, z3.s, z4.s
; CHECK-NEXT: uzp1 z1.s, z1.s, z2.s
; CHECK-NEXT: mov z2.h, #32767 // =0x7fff
; CHECK-NEXT: uzp1 z1.h, z1.h, z3.h
; CHECK-NEXT: bsl z0.d, z0.d, z1.d, z2.d
; CHECK-NEXT: ret
%b.trunc = fptrunc <vscale x 8 x double> %b to <vscale x 8 x half>
%r = call <vscale x 8 x half> @llvm.copysign.nxv8f16(<vscale x 8 x half> %a, <vscale x 8 x half> %b.trunc)
ret <vscale x 8 x half> %r
}
;
; llvm.copysign.nxv2f32
;
define <vscale x 2 x float> @copysign_nxv2f32_nxv2f16(<vscale x 2 x float> %a, <vscale x 2 x half> %b) {
; CHECK-LABEL: copysign_nxv2f32_nxv2f16:
; CHECK: // %bb.0:
; CHECK-NEXT: ptrue p0.d
; CHECK-NEXT: mov z2.s, #0x7fffffff
; CHECK-NEXT: fcvt z1.s, p0/m, z1.h
; CHECK-NEXT: bsl z0.d, z0.d, z1.d, z2.d
; CHECK-NEXT: ret
%b.ext = fpext <vscale x 2 x half> %b to <vscale x 2 x float>
%r = call <vscale x 2 x float> @llvm.copysign.nxv2f32(<vscale x 2 x float> %a, <vscale x 2 x float> %b.ext)
ret <vscale x 2 x float> %r
}
define <vscale x 2 x float> @copysign_nxv2f32_nxv2f32(<vscale x 2 x float> %a, <vscale x 2 x float> %b) {
; CHECK-LABEL: copysign_nxv2f32_nxv2f32:
; CHECK: // %bb.0:
; CHECK-NEXT: mov z2.s, #0x7fffffff
; CHECK-NEXT: bsl z0.d, z0.d, z1.d, z2.d
; CHECK-NEXT: ret
%r = call <vscale x 2 x float> @llvm.copysign.nxv2f32(<vscale x 2 x float> %a, <vscale x 2 x float> %b)
ret <vscale x 2 x float> %r
}
define <vscale x 2 x float> @copysign_nxv2f32_nxv2f64(<vscale x 2 x float> %a, <vscale x 2 x double> %b) {
; CHECK-LABEL: copysign_nxv2f32_nxv2f64:
; CHECK: // %bb.0:
; CHECK-NEXT: ptrue p0.d
; CHECK-NEXT: mov z2.s, #0x7fffffff
; CHECK-NEXT: fcvt z1.s, p0/m, z1.d
; CHECK-NEXT: bsl z0.d, z0.d, z1.d, z2.d
; CHECK-NEXT: ret
%b.trunc = fptrunc <vscale x 2 x double> %b to <vscale x 2 x float>
%r = call <vscale x 2 x float> @llvm.copysign.nxv2f32(<vscale x 2 x float> %a, <vscale x 2 x float> %b.trunc)
ret <vscale x 2 x float> %r
}
define <vscale x 2 x float> @copysign_nxv2f32_nxv2bf16(<vscale x 2 x float> %a, <vscale x 2 x bfloat> %b) {
; CHECK-LABEL: copysign_nxv2f32_nxv2bf16:
; CHECK: // %bb.0:
; CHECK-NEXT: mov z2.s, #0x7fffffff
; CHECK-NEXT: lsl z1.s, z1.s, #16
; CHECK-NEXT: bsl z0.d, z0.d, z1.d, z2.d
; CHECK-NEXT: ret
%tmp0 = fpext <vscale x 2 x bfloat> %b to <vscale x 2 x float>
%r = call <vscale x 2 x float> @llvm.copysign.nxv2f32(<vscale x 2 x float> %a, <vscale x 2 x float> %tmp0)
ret <vscale x 2 x float> %r
}
;
; llvm.copysign.nxv4f32
;
define <vscale x 4 x float> @copysign_nxv4f32_nxv4f16(<vscale x 4 x float> %a, <vscale x 4 x half> %b) {
; CHECK-LABEL: copysign_nxv4f32_nxv4f16:
; CHECK: // %bb.0:
; CHECK-NEXT: ptrue p0.s
; CHECK-NEXT: mov z2.s, #0x7fffffff
; CHECK-NEXT: fcvt z1.s, p0/m, z1.h
; CHECK-NEXT: bsl z0.d, z0.d, z1.d, z2.d
; CHECK-NEXT: ret
%b.ext = fpext <vscale x 4 x half> %b to <vscale x 4 x float>
%r = call <vscale x 4 x float> @llvm.copysign.nxv4f32(<vscale x 4 x float> %a, <vscale x 4 x float> %b.ext)
ret <vscale x 4 x float> %r
}
define <vscale x 4 x float> @copysign_nxv4f32_nxv4f32(<vscale x 4 x float> %a, <vscale x 4 x float> %b) {
; CHECK-LABEL: copysign_nxv4f32_nxv4f32:
; CHECK: // %bb.0:
; CHECK-NEXT: mov z2.s, #0x7fffffff
; CHECK-NEXT: bsl z0.d, z0.d, z1.d, z2.d
; CHECK-NEXT: ret
%r = call <vscale x 4 x float> @llvm.copysign.nxv4f32(<vscale x 4 x float> %a, <vscale x 4 x float> %b)
ret <vscale x 4 x float> %r
}
define <vscale x 4 x float> @copysign_nxv4f32_nxv4f64(<vscale x 4 x float> %a, <vscale x 4 x double> %b) {
; CHECK-LABEL: copysign_nxv4f32_nxv4f64:
; CHECK: // %bb.0:
; CHECK-NEXT: ptrue p0.d
; CHECK-NEXT: mov z3.s, #0x7fffffff
; CHECK-NEXT: fcvt z2.s, p0/m, z2.d
; CHECK-NEXT: fcvt z1.s, p0/m, z1.d
; CHECK-NEXT: uzp1 z1.s, z1.s, z2.s
; CHECK-NEXT: bsl z0.d, z0.d, z1.d, z3.d
; CHECK-NEXT: ret
%b.trunc = fptrunc <vscale x 4 x double> %b to <vscale x 4 x float>
%r = call <vscale x 4 x float> @llvm.copysign.nxv4f32(<vscale x 4 x float> %a, <vscale x 4 x float> %b.trunc)
ret <vscale x 4 x float> %r
}
define <vscale x 4 x float> @copysign_nxv4f32_nxv4bf16(<vscale x 4 x float> %a, <vscale x 4 x bfloat> %b) {
; CHECK-LABEL: copysign_nxv4f32_nxv4bf16:
; CHECK: // %bb.0:
; CHECK-NEXT: mov z2.s, #0x7fffffff
; CHECK-NEXT: lsl z1.s, z1.s, #16
; CHECK-NEXT: bsl z0.d, z0.d, z1.d, z2.d
; CHECK-NEXT: ret
%tmp0 = fpext <vscale x 4 x bfloat> %b to <vscale x 4 x float>
%r = call <vscale x 4 x float> @llvm.copysign.nxv4f32(<vscale x 4 x float> %a, <vscale x 4 x float> %tmp0)
ret <vscale x 4 x float> %r
}
;
; llvm.copysign.nxv2f64
;
define <vscale x 2 x double> @copysign_nxv2f64_nxv2f16(<vscale x 2 x double> %a, <vscale x 2 x half> %b) {
; CHECK-LABEL: copysign_nxv2f64_nxv2f16:
; CHECK: // %bb.0:
; CHECK-NEXT: ptrue p0.d
; CHECK-NEXT: mov z2.d, #0x7fffffffffffffff
; CHECK-NEXT: fcvt z1.d, p0/m, z1.h
; CHECK-NEXT: bsl z0.d, z0.d, z1.d, z2.d
; CHECK-NEXT: ret
%b.ext = fpext <vscale x 2 x half> %b to <vscale x 2 x double>
%r = call <vscale x 2 x double> @llvm.copysign.nxv2f64(<vscale x 2 x double> %a, <vscale x 2 x double> %b.ext)
ret <vscale x 2 x double> %r
}
define <vscale x 2 x double> @copysign_nxv2f64_nxv2f32(<vscale x 2 x double> %a, <vscale x 2 x float> %b) {
; CHECK-LABEL: copysign_nxv2f64_nxv2f32:
; CHECK: // %bb.0:
; CHECK-NEXT: ptrue p0.d
; CHECK-NEXT: mov z2.d, #0x7fffffffffffffff
; CHECK-NEXT: fcvt z1.d, p0/m, z1.s
; CHECK-NEXT: bsl z0.d, z0.d, z1.d, z2.d
; CHECK-NEXT: ret
%b.ext = fpext <vscale x 2 x float> %b to <vscale x 2 x double>
%r = call <vscale x 2 x double> @llvm.copysign.nxv2f64(<vscale x 2 x double> %a, <vscale x 2 x double> %b.ext)
ret <vscale x 2 x double> %r
}
define <vscale x 2 x double> @copysign_nxv2f64_nxv2f64(<vscale x 2 x double> %a, <vscale x 2 x double> %b) {
; CHECK-LABEL: copysign_nxv2f64_nxv2f64:
; CHECK: // %bb.0:
; CHECK-NEXT: mov z2.d, #0x7fffffffffffffff
; CHECK-NEXT: bsl z0.d, z0.d, z1.d, z2.d
; CHECK-NEXT: ret
%r = call <vscale x 2 x double> @llvm.copysign.nxv2f64(<vscale x 2 x double> %a, <vscale x 2 x double> %b)
ret <vscale x 2 x double> %r
}
define <vscale x 2 x double> @copysign_nxv2f64_nxv2bf16(<vscale x 2 x double> %a, <vscale x 2 x bfloat> %b) {
; CHECK-LABEL: copysign_nxv2f64_nxv2bf16:
; CHECK: // %bb.0:
; CHECK-NEXT: lsl z1.s, z1.s, #16
; CHECK-NEXT: ptrue p0.d
; CHECK-NEXT: mov z2.d, #0x7fffffffffffffff
; CHECK-NEXT: fcvt z1.d, p0/m, z1.s
; CHECK-NEXT: bsl z0.d, z0.d, z1.d, z2.d
; CHECK-NEXT: ret
%b.ext = fpext <vscale x 2 x bfloat> %b to <vscale x 2 x double>
%r = call <vscale x 2 x double> @llvm.copysign.nxv2f64(<vscale x 2 x double> %a, <vscale x 2 x double> %b.ext)
ret <vscale x 2 x double> %r
}
;
; llvm.copysign.nxv2bf16
;
define <vscale x 2 x bfloat> @copysign_nxv2bf16_nxv2bf16(<vscale x 2 x bfloat> %a, <vscale x 2 x bfloat> %b) {
; CHECK-LABEL: copysign_nxv2bf16_nxv2bf16:
; CHECK: // %bb.0:
; CHECK-NEXT: mov z2.h, #32767 // =0x7fff
; CHECK-NEXT: bsl z0.d, z0.d, z1.d, z2.d
; CHECK-NEXT: ret
%r = call <vscale x 2 x bfloat> @llvm.copysign.nxv2bf16(<vscale x 2 x bfloat> %a, <vscale x 2 x bfloat> %b)
ret <vscale x 2 x bfloat> %r
}
define <vscale x 2 x bfloat> @copysign_nxv2bf16_nxv2f32(<vscale x 2 x bfloat> %a, <vscale x 2 x float> %b) {
; CHECK-LABEL: copysign_nxv2bf16_nxv2f32:
; CHECK: // %bb.0:
; CHECK-NEXT: ptrue p0.d
; CHECK-NEXT: mov z2.h, #32767 // =0x7fff
; CHECK-NEXT: bfcvt z1.h, p0/m, z1.s
; CHECK-NEXT: bsl z0.d, z0.d, z1.d, z2.d
; CHECK-NEXT: ret
%tmp0 = fptrunc <vscale x 2 x float> %b to <vscale x 2 x bfloat>
%r = call <vscale x 2 x bfloat> @llvm.copysign.nxv2bf16(<vscale x 2 x bfloat> %a, <vscale x 2 x bfloat> %tmp0)
ret <vscale x 2 x bfloat> %r
}
define <vscale x 2 x bfloat> @copysign_nxv2bf16_nxv2f64(<vscale x 2 x bfloat> %a, <vscale x 2 x double> %b) {
; CHECK-LABEL: copysign_nxv2bf16_nxv2f64:
; CHECK: // %bb.0:
; CHECK-NEXT: ptrue p0.d
; CHECK-NEXT: mov z2.h, #32767 // =0x7fff
; CHECK-NEXT: fcvtx z1.s, p0/m, z1.d
; CHECK-NEXT: bfcvt z1.h, p0/m, z1.s
; CHECK-NEXT: bsl z0.d, z0.d, z1.d, z2.d
; CHECK-NEXT: ret
%tmp0 = fptrunc <vscale x 2 x double> %b to <vscale x 2 x bfloat>
%r = call <vscale x 2 x bfloat> @llvm.copysign.nxv2bf16(<vscale x 2 x bfloat> %a, <vscale x 2 x bfloat> %tmp0)
ret <vscale x 2 x bfloat> %r
}
;
; llvm.copysign.nxv2bf16
;
define <vscale x 4 x bfloat> @copysign_nxv4bf16_nxv4bf16(<vscale x 4 x bfloat> %a, <vscale x 4 x bfloat> %b) {
; CHECK-LABEL: copysign_nxv4bf16_nxv4bf16:
; CHECK: // %bb.0:
; CHECK-NEXT: mov z2.h, #32767 // =0x7fff
; CHECK-NEXT: bsl z0.d, z0.d, z1.d, z2.d
; CHECK-NEXT: ret
%r = call <vscale x 4 x bfloat> @llvm.copysign.nxv4bf16(<vscale x 4 x bfloat> %a, <vscale x 4 x bfloat> %b)
ret <vscale x 4 x bfloat> %r
}
define <vscale x 4 x bfloat> @copysign_nxv4bf16_nxv4f32(<vscale x 4 x bfloat> %a, <vscale x 4 x float> %b) {
; CHECK-LABEL: copysign_nxv4bf16_nxv4f32:
; CHECK: // %bb.0:
; CHECK-NEXT: ptrue p0.s
; CHECK-NEXT: mov z2.h, #32767 // =0x7fff
; CHECK-NEXT: bfcvt z1.h, p0/m, z1.s
; CHECK-NEXT: bsl z0.d, z0.d, z1.d, z2.d
; CHECK-NEXT: ret
%b.trunc = fptrunc <vscale x 4 x float> %b to <vscale x 4 x bfloat>
%r = call <vscale x 4 x bfloat> @llvm.copysign.nxv4bf16(<vscale x 4 x bfloat> %a, <vscale x 4 x bfloat> %b.trunc)
ret <vscale x 4 x bfloat> %r
}
define <vscale x 4 x bfloat> @copysign_nxv4bf16_nxv4f64(<vscale x 4 x bfloat> %a, <vscale x 4 x double> %b) {
; CHECK-LABEL: copysign_nxv4bf16_nxv4f64:
; CHECK: // %bb.0:
; CHECK-NEXT: ptrue p0.d
; CHECK-NEXT: mov z3.h, #32767 // =0x7fff
; CHECK-NEXT: fcvtx z2.s, p0/m, z2.d
; CHECK-NEXT: fcvtx z1.s, p0/m, z1.d
; CHECK-NEXT: bfcvt z2.h, p0/m, z2.s
; CHECK-NEXT: bfcvt z1.h, p0/m, z1.s
; CHECK-NEXT: uzp1 z1.s, z1.s, z2.s
; CHECK-NEXT: bsl z0.d, z0.d, z1.d, z3.d
; CHECK-NEXT: ret
%b.trunc = fptrunc <vscale x 4 x double> %b to <vscale x 4 x bfloat>
%r = call <vscale x 4 x bfloat> @llvm.copysign.nxv4bf16(<vscale x 4 x bfloat> %a, <vscale x 4 x bfloat> %b.trunc)
ret <vscale x 4 x bfloat> %r
}
;
; llvm.copysign.nxv8bf16
;
define <vscale x 8 x bfloat> @copysign_nxv8bf16_nxv8bf16(<vscale x 8 x bfloat> %a, <vscale x 8 x bfloat> %b) {
; CHECK-LABEL: copysign_nxv8bf16_nxv8bf16:
; CHECK: // %bb.0:
; CHECK-NEXT: mov z2.h, #32767 // =0x7fff
; CHECK-NEXT: bsl z0.d, z0.d, z1.d, z2.d
; CHECK-NEXT: ret
%r = call <vscale x 8 x bfloat> @llvm.copysign.nxv8bf16(<vscale x 8 x bfloat> %a, <vscale x 8 x bfloat> %b)
ret <vscale x 8 x bfloat> %r
}
define <vscale x 8 x bfloat> @copysign_nxv8bf16_nxv8f32(<vscale x 8 x bfloat> %a, <vscale x 8 x float> %b) {
; CHECK-LABEL: copysign_nxv8bf16_nxv8f32:
; CHECK: // %bb.0:
; CHECK-NEXT: ptrue p0.s
; CHECK-NEXT: mov z3.h, #32767 // =0x7fff
; CHECK-NEXT: bfcvt z2.h, p0/m, z2.s
; CHECK-NEXT: bfcvt z1.h, p0/m, z1.s
; CHECK-NEXT: uzp1 z1.h, z1.h, z2.h
; CHECK-NEXT: bsl z0.d, z0.d, z1.d, z3.d
; CHECK-NEXT: ret
%b.trunc = fptrunc <vscale x 8 x float> %b to <vscale x 8 x bfloat>
%r = call <vscale x 8 x bfloat> @llvm.copysign.nxv8bf16(<vscale x 8 x bfloat> %a, <vscale x 8 x bfloat> %b.trunc)
ret <vscale x 8 x bfloat> %r
}
define <vscale x 8 x bfloat> @copysign_nxv8bf16_nxv8f64(<vscale x 8 x bfloat> %a, <vscale x 8 x double> %b) {
; CHECK-LABEL: copysign_nxv8bf16_nxv8f64:
; CHECK: // %bb.0:
; CHECK-NEXT: ptrue p0.d
; CHECK-NEXT: fcvtx z4.s, p0/m, z4.d
; CHECK-NEXT: fcvtx z3.s, p0/m, z3.d
; CHECK-NEXT: fcvtx z2.s, p0/m, z2.d
; CHECK-NEXT: fcvtx z1.s, p0/m, z1.d
; CHECK-NEXT: bfcvt z4.h, p0/m, z4.s
; CHECK-NEXT: bfcvt z3.h, p0/m, z3.s
; CHECK-NEXT: bfcvt z2.h, p0/m, z2.s
; CHECK-NEXT: bfcvt z1.h, p0/m, z1.s
; CHECK-NEXT: uzp1 z3.s, z3.s, z4.s
; CHECK-NEXT: uzp1 z1.s, z1.s, z2.s
; CHECK-NEXT: mov z2.h, #32767 // =0x7fff
; CHECK-NEXT: uzp1 z1.h, z1.h, z3.h
; CHECK-NEXT: bsl z0.d, z0.d, z1.d, z2.d
; CHECK-NEXT: ret
%b.trunc = fptrunc <vscale x 8 x double> %b to <vscale x 8 x bfloat>
%r = call <vscale x 8 x bfloat> @llvm.copysign.nxv8bf16(<vscale x 8 x bfloat> %a, <vscale x 8 x bfloat> %b.trunc)
ret <vscale x 8 x bfloat> %r
}
declare <vscale x 2 x half> @llvm.copysign.nxv2f16(<vscale x 2 x half> %a, <vscale x 2 x half> %b)
declare <vscale x 4 x half> @llvm.copysign.nxv4f16(<vscale x 4 x half> %a, <vscale x 4 x half> %b)
declare <vscale x 8 x half> @llvm.copysign.nxv8f16(<vscale x 8 x half> %a, <vscale x 8 x half> %b)
declare <vscale x 2 x float> @llvm.copysign.nxv2f32(<vscale x 2 x float> %a, <vscale x 2 x float> %b)
declare <vscale x 4 x float> @llvm.copysign.nxv4f32(<vscale x 4 x float> %a, <vscale x 4 x float> %b)
declare <vscale x 2 x double> @llvm.copysign.nxv2f64(<vscale x 2 x double> %a, <vscale x 2 x double> %b)
declare <vscale x 2 x bfloat> @llvm.copysign.nxv2bf16(<vscale x 2 x bfloat> %a, <vscale x 2 x bfloat> %b)
declare <vscale x 4 x bfloat> @llvm.copysign.nxv4bf16(<vscale x 4 x bfloat> %a, <vscale x 4 x bfloat> %b)
declare <vscale x 8 x bfloat> @llvm.copysign.nxv8bf16(<vscale x 8 x bfloat> %a, <vscale x 8 x bfloat> %b)
Reference in new issue View Git Blame Copy Permalink