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
1096 lines
39 KiB
LLVM
1096 lines
39 KiB
LLVM
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
|
|
; RUN: llc -mtriple=riscv32 -mattr=+d,+zfh,+zvfh,+v,+m -target-abi=ilp32d \
|
|
; RUN: -verify-machineinstrs < %s | FileCheck %s
|
|
; RUN: llc -mtriple=riscv64 -mattr=+d,+zfh,+zvfh,+v,+m -target-abi=lp64d \
|
|
; RUN: -verify-machineinstrs < %s | FileCheck %s
|
|
|
|
define half @vreduce_fadd_nxv1f16(<vscale x 1 x half> %v, half %s) {
|
|
; CHECK-LABEL: vreduce_fadd_nxv1f16:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: vsetvli a0, zero, e16, mf4, ta, ma
|
|
; CHECK-NEXT: vfmv.s.f v9, fa0
|
|
; CHECK-NEXT: vfredusum.vs v8, v8, v9
|
|
; CHECK-NEXT: vfmv.f.s fa0, v8
|
|
; CHECK-NEXT: ret
|
|
%red = call reassoc half @llvm.vector.reduce.fadd.nxv1f16(half %s, <vscale x 1 x half> %v)
|
|
ret half %red
|
|
}
|
|
|
|
define half @vreduce_ord_fadd_nxv1f16(<vscale x 1 x half> %v, half %s) {
|
|
; CHECK-LABEL: vreduce_ord_fadd_nxv1f16:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: vsetvli a0, zero, e16, mf4, ta, ma
|
|
; CHECK-NEXT: vfmv.s.f v9, fa0
|
|
; CHECK-NEXT: vfredosum.vs v8, v8, v9
|
|
; CHECK-NEXT: vfmv.f.s fa0, v8
|
|
; CHECK-NEXT: ret
|
|
%red = call half @llvm.vector.reduce.fadd.nxv1f16(half %s, <vscale x 1 x half> %v)
|
|
ret half %red
|
|
}
|
|
|
|
define half @vreduce_fadd_nxv2f16(<vscale x 2 x half> %v, half %s) {
|
|
; CHECK-LABEL: vreduce_fadd_nxv2f16:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: vsetvli a0, zero, e16, mf2, ta, ma
|
|
; CHECK-NEXT: vfmv.s.f v9, fa0
|
|
; CHECK-NEXT: vfredusum.vs v8, v8, v9
|
|
; CHECK-NEXT: vfmv.f.s fa0, v8
|
|
; CHECK-NEXT: ret
|
|
%red = call reassoc half @llvm.vector.reduce.fadd.nxv2f16(half %s, <vscale x 2 x half> %v)
|
|
ret half %red
|
|
}
|
|
|
|
define half @vreduce_ord_fadd_nxv2f16(<vscale x 2 x half> %v, half %s) {
|
|
; CHECK-LABEL: vreduce_ord_fadd_nxv2f16:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: vsetvli a0, zero, e16, mf2, ta, ma
|
|
; CHECK-NEXT: vfmv.s.f v9, fa0
|
|
; CHECK-NEXT: vfredosum.vs v8, v8, v9
|
|
; CHECK-NEXT: vfmv.f.s fa0, v8
|
|
; CHECK-NEXT: ret
|
|
%red = call half @llvm.vector.reduce.fadd.nxv2f16(half %s, <vscale x 2 x half> %v)
|
|
ret half %red
|
|
}
|
|
|
|
define half @vreduce_fadd_nxv4f16(<vscale x 4 x half> %v, half %s) {
|
|
; CHECK-LABEL: vreduce_fadd_nxv4f16:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: vsetvli a0, zero, e16, m1, ta, ma
|
|
; CHECK-NEXT: vfmv.s.f v9, fa0
|
|
; CHECK-NEXT: vfredusum.vs v8, v8, v9
|
|
; CHECK-NEXT: vfmv.f.s fa0, v8
|
|
; CHECK-NEXT: ret
|
|
%red = call reassoc half @llvm.vector.reduce.fadd.nxv4f16(half %s, <vscale x 4 x half> %v)
|
|
ret half %red
|
|
}
|
|
|
|
define half @vreduce_ord_fadd_nxv4f16(<vscale x 4 x half> %v, half %s) {
|
|
; CHECK-LABEL: vreduce_ord_fadd_nxv4f16:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: vsetvli a0, zero, e16, m1, ta, ma
|
|
; CHECK-NEXT: vfmv.s.f v9, fa0
|
|
; CHECK-NEXT: vfredosum.vs v8, v8, v9
|
|
; CHECK-NEXT: vfmv.f.s fa0, v8
|
|
; CHECK-NEXT: ret
|
|
%red = call half @llvm.vector.reduce.fadd.nxv4f16(half %s, <vscale x 4 x half> %v)
|
|
ret half %red
|
|
}
|
|
|
|
define float @vreduce_fadd_nxv1f32(<vscale x 1 x float> %v, float %s) {
|
|
; CHECK-LABEL: vreduce_fadd_nxv1f32:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: vsetvli a0, zero, e32, mf2, ta, ma
|
|
; CHECK-NEXT: vfmv.s.f v9, fa0
|
|
; CHECK-NEXT: vfredusum.vs v8, v8, v9
|
|
; CHECK-NEXT: vfmv.f.s fa0, v8
|
|
; CHECK-NEXT: ret
|
|
%red = call reassoc float @llvm.vector.reduce.fadd.nxv1f32(float %s, <vscale x 1 x float> %v)
|
|
ret float %red
|
|
}
|
|
|
|
define float @vreduce_ord_fadd_nxv1f32(<vscale x 1 x float> %v, float %s) {
|
|
; CHECK-LABEL: vreduce_ord_fadd_nxv1f32:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: vsetvli a0, zero, e32, mf2, ta, ma
|
|
; CHECK-NEXT: vfmv.s.f v9, fa0
|
|
; CHECK-NEXT: vfredosum.vs v8, v8, v9
|
|
; CHECK-NEXT: vfmv.f.s fa0, v8
|
|
; CHECK-NEXT: ret
|
|
%red = call float @llvm.vector.reduce.fadd.nxv1f32(float %s, <vscale x 1 x float> %v)
|
|
ret float %red
|
|
}
|
|
|
|
define float @vreduce_fwadd_nxv1f32(<vscale x 1 x half> %v, float %s) {
|
|
; CHECK-LABEL: vreduce_fwadd_nxv1f32:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: vsetvli a0, zero, e32, m1, ta, ma
|
|
; CHECK-NEXT: vfmv.s.f v9, fa0
|
|
; CHECK-NEXT: vsetvli a0, zero, e16, mf4, ta, ma
|
|
; CHECK-NEXT: vfwredusum.vs v8, v8, v9
|
|
; CHECK-NEXT: vsetvli zero, zero, e32, mf2, ta, ma
|
|
; CHECK-NEXT: vfmv.f.s fa0, v8
|
|
; CHECK-NEXT: ret
|
|
%e = fpext <vscale x 1 x half> %v to <vscale x 1 x float>
|
|
%red = call reassoc float @llvm.vector.reduce.fadd.nxv1f32(float %s, <vscale x 1 x float> %e)
|
|
ret float %red
|
|
}
|
|
|
|
define float @vreduce_ord_fwadd_nxv1f32(<vscale x 1 x half> %v, float %s) {
|
|
; CHECK-LABEL: vreduce_ord_fwadd_nxv1f32:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: vsetvli a0, zero, e32, m1, ta, ma
|
|
; CHECK-NEXT: vfmv.s.f v9, fa0
|
|
; CHECK-NEXT: vsetvli a0, zero, e16, mf4, ta, ma
|
|
; CHECK-NEXT: vfwredosum.vs v8, v8, v9
|
|
; CHECK-NEXT: vsetvli zero, zero, e32, mf2, ta, ma
|
|
; CHECK-NEXT: vfmv.f.s fa0, v8
|
|
; CHECK-NEXT: ret
|
|
%e = fpext <vscale x 1 x half> %v to <vscale x 1 x float>
|
|
%red = call float @llvm.vector.reduce.fadd.nxv1f32(float %s, <vscale x 1 x float> %e)
|
|
ret float %red
|
|
}
|
|
|
|
define float @vreduce_fadd_nxv2f32(<vscale x 2 x float> %v, float %s) {
|
|
; CHECK-LABEL: vreduce_fadd_nxv2f32:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: vsetvli a0, zero, e32, m1, ta, ma
|
|
; CHECK-NEXT: vfmv.s.f v9, fa0
|
|
; CHECK-NEXT: vfredusum.vs v8, v8, v9
|
|
; CHECK-NEXT: vfmv.f.s fa0, v8
|
|
; CHECK-NEXT: ret
|
|
%red = call reassoc float @llvm.vector.reduce.fadd.nxv2f32(float %s, <vscale x 2 x float> %v)
|
|
ret float %red
|
|
}
|
|
|
|
define float @vreduce_ord_fadd_nxv2f32(<vscale x 2 x float> %v, float %s) {
|
|
; CHECK-LABEL: vreduce_ord_fadd_nxv2f32:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: vsetvli a0, zero, e32, m1, ta, ma
|
|
; CHECK-NEXT: vfmv.s.f v9, fa0
|
|
; CHECK-NEXT: vfredosum.vs v8, v8, v9
|
|
; CHECK-NEXT: vfmv.f.s fa0, v8
|
|
; CHECK-NEXT: ret
|
|
%red = call float @llvm.vector.reduce.fadd.nxv2f32(float %s, <vscale x 2 x float> %v)
|
|
ret float %red
|
|
}
|
|
|
|
define float @vreduce_fwadd_nxv2f32(<vscale x 2 x half> %v, float %s) {
|
|
; CHECK-LABEL: vreduce_fwadd_nxv2f32:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: vsetvli a0, zero, e32, m1, ta, ma
|
|
; CHECK-NEXT: vfmv.s.f v9, fa0
|
|
; CHECK-NEXT: vsetvli zero, zero, e16, mf2, ta, ma
|
|
; CHECK-NEXT: vfwredusum.vs v8, v8, v9
|
|
; CHECK-NEXT: vsetvli zero, zero, e32, m1, ta, ma
|
|
; CHECK-NEXT: vfmv.f.s fa0, v8
|
|
; CHECK-NEXT: ret
|
|
%e = fpext <vscale x 2 x half> %v to <vscale x 2 x float>
|
|
%red = call reassoc float @llvm.vector.reduce.fadd.nxv2f32(float %s, <vscale x 2 x float> %e)
|
|
ret float %red
|
|
}
|
|
|
|
define float @vreduce_ord_fwadd_nxv2f32(<vscale x 2 x half> %v, float %s) {
|
|
; CHECK-LABEL: vreduce_ord_fwadd_nxv2f32:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: vsetvli a0, zero, e32, m1, ta, ma
|
|
; CHECK-NEXT: vfmv.s.f v9, fa0
|
|
; CHECK-NEXT: vsetvli zero, zero, e16, mf2, ta, ma
|
|
; CHECK-NEXT: vfwredosum.vs v8, v8, v9
|
|
; CHECK-NEXT: vsetvli zero, zero, e32, m1, ta, ma
|
|
; CHECK-NEXT: vfmv.f.s fa0, v8
|
|
; CHECK-NEXT: ret
|
|
%e = fpext <vscale x 2 x half> %v to <vscale x 2 x float>
|
|
%red = call float @llvm.vector.reduce.fadd.nxv2f32(float %s, <vscale x 2 x float> %e)
|
|
ret float %red
|
|
}
|
|
|
|
define float @vreduce_fadd_nxv4f32(<vscale x 4 x float> %v, float %s) {
|
|
; CHECK-LABEL: vreduce_fadd_nxv4f32:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: vsetvli a0, zero, e32, m2, ta, ma
|
|
; CHECK-NEXT: vfmv.s.f v10, fa0
|
|
; CHECK-NEXT: vfredusum.vs v8, v8, v10
|
|
; CHECK-NEXT: vfmv.f.s fa0, v8
|
|
; CHECK-NEXT: ret
|
|
%red = call reassoc float @llvm.vector.reduce.fadd.nxv4f32(float %s, <vscale x 4 x float> %v)
|
|
ret float %red
|
|
}
|
|
|
|
define float @vreduce_ord_fadd_nxv4f32(<vscale x 4 x float> %v, float %s) {
|
|
; CHECK-LABEL: vreduce_ord_fadd_nxv4f32:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: vsetvli a0, zero, e32, m2, ta, ma
|
|
; CHECK-NEXT: vfmv.s.f v10, fa0
|
|
; CHECK-NEXT: vfredosum.vs v8, v8, v10
|
|
; CHECK-NEXT: vfmv.f.s fa0, v8
|
|
; CHECK-NEXT: ret
|
|
%red = call float @llvm.vector.reduce.fadd.nxv4f32(float %s, <vscale x 4 x float> %v)
|
|
ret float %red
|
|
}
|
|
|
|
define float @vreduce_fwadd_nxv4f32(<vscale x 4 x half> %v, float %s) {
|
|
; CHECK-LABEL: vreduce_fwadd_nxv4f32:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: vsetvli a0, zero, e32, m1, ta, ma
|
|
; CHECK-NEXT: vfmv.s.f v9, fa0
|
|
; CHECK-NEXT: vsetvli a0, zero, e16, m1, ta, ma
|
|
; CHECK-NEXT: vfwredusum.vs v8, v8, v9
|
|
; CHECK-NEXT: vsetvli zero, zero, e32, m2, ta, ma
|
|
; CHECK-NEXT: vfmv.f.s fa0, v8
|
|
; CHECK-NEXT: ret
|
|
%e = fpext <vscale x 4 x half> %v to <vscale x 4 x float>
|
|
%red = call reassoc float @llvm.vector.reduce.fadd.nxv4f32(float %s, <vscale x 4 x float> %e)
|
|
ret float %red
|
|
}
|
|
|
|
define float @vreduce_ord_fwadd_nxv4f32(<vscale x 4 x half> %v, float %s) {
|
|
; CHECK-LABEL: vreduce_ord_fwadd_nxv4f32:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: vsetvli a0, zero, e32, m1, ta, ma
|
|
; CHECK-NEXT: vfmv.s.f v9, fa0
|
|
; CHECK-NEXT: vsetvli a0, zero, e16, m1, ta, ma
|
|
; CHECK-NEXT: vfwredosum.vs v8, v8, v9
|
|
; CHECK-NEXT: vsetvli zero, zero, e32, m2, ta, ma
|
|
; CHECK-NEXT: vfmv.f.s fa0, v8
|
|
; CHECK-NEXT: ret
|
|
%e = fpext <vscale x 4 x half> %v to <vscale x 4 x float>
|
|
%red = call float @llvm.vector.reduce.fadd.nxv4f32(float %s, <vscale x 4 x float> %e)
|
|
ret float %red
|
|
}
|
|
|
|
define double @vreduce_fadd_nxv1f64(<vscale x 1 x double> %v, double %s) {
|
|
; CHECK-LABEL: vreduce_fadd_nxv1f64:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: vsetvli a0, zero, e64, m1, ta, ma
|
|
; CHECK-NEXT: vfmv.s.f v9, fa0
|
|
; CHECK-NEXT: vfredusum.vs v8, v8, v9
|
|
; CHECK-NEXT: vfmv.f.s fa0, v8
|
|
; CHECK-NEXT: ret
|
|
%red = call reassoc double @llvm.vector.reduce.fadd.nxv1f64(double %s, <vscale x 1 x double> %v)
|
|
ret double %red
|
|
}
|
|
|
|
define double @vreduce_ord_fadd_nxv1f64(<vscale x 1 x double> %v, double %s) {
|
|
; CHECK-LABEL: vreduce_ord_fadd_nxv1f64:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: vsetvli a0, zero, e64, m1, ta, ma
|
|
; CHECK-NEXT: vfmv.s.f v9, fa0
|
|
; CHECK-NEXT: vfredosum.vs v8, v8, v9
|
|
; CHECK-NEXT: vfmv.f.s fa0, v8
|
|
; CHECK-NEXT: ret
|
|
%red = call double @llvm.vector.reduce.fadd.nxv1f64(double %s, <vscale x 1 x double> %v)
|
|
ret double %red
|
|
}
|
|
|
|
define double @vreduce_fwadd_nxv1f64(<vscale x 1 x float> %v, double %s) {
|
|
; CHECK-LABEL: vreduce_fwadd_nxv1f64:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: vsetvli a0, zero, e64, m1, ta, ma
|
|
; CHECK-NEXT: vfmv.s.f v9, fa0
|
|
; CHECK-NEXT: vsetvli zero, zero, e32, mf2, ta, ma
|
|
; CHECK-NEXT: vfwredusum.vs v8, v8, v9
|
|
; CHECK-NEXT: vsetvli zero, zero, e64, m1, ta, ma
|
|
; CHECK-NEXT: vfmv.f.s fa0, v8
|
|
; CHECK-NEXT: ret
|
|
%e = fpext <vscale x 1 x float> %v to <vscale x 1 x double>
|
|
%red = call reassoc double @llvm.vector.reduce.fadd.nxv1f64(double %s, <vscale x 1 x double> %e)
|
|
ret double %red
|
|
}
|
|
|
|
define double @vreduce_ord_fwadd_nxv1f64(<vscale x 1 x float> %v, double %s) {
|
|
; CHECK-LABEL: vreduce_ord_fwadd_nxv1f64:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: vsetvli a0, zero, e64, m1, ta, ma
|
|
; CHECK-NEXT: vfmv.s.f v9, fa0
|
|
; CHECK-NEXT: vsetvli zero, zero, e32, mf2, ta, ma
|
|
; CHECK-NEXT: vfwredosum.vs v8, v8, v9
|
|
; CHECK-NEXT: vsetvli zero, zero, e64, m1, ta, ma
|
|
; CHECK-NEXT: vfmv.f.s fa0, v8
|
|
; CHECK-NEXT: ret
|
|
%e = fpext <vscale x 1 x float> %v to <vscale x 1 x double>
|
|
%red = call double @llvm.vector.reduce.fadd.nxv1f64(double %s, <vscale x 1 x double> %e)
|
|
ret double %red
|
|
}
|
|
|
|
define double @vreduce_fadd_nxv2f64(<vscale x 2 x double> %v, double %s) {
|
|
; CHECK-LABEL: vreduce_fadd_nxv2f64:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: vsetvli a0, zero, e64, m2, ta, ma
|
|
; CHECK-NEXT: vfmv.s.f v10, fa0
|
|
; CHECK-NEXT: vfredusum.vs v8, v8, v10
|
|
; CHECK-NEXT: vfmv.f.s fa0, v8
|
|
; CHECK-NEXT: ret
|
|
%red = call reassoc double @llvm.vector.reduce.fadd.nxv2f64(double %s, <vscale x 2 x double> %v)
|
|
ret double %red
|
|
}
|
|
|
|
define double @vreduce_ord_fadd_nxv2f64(<vscale x 2 x double> %v, double %s) {
|
|
; CHECK-LABEL: vreduce_ord_fadd_nxv2f64:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: vsetvli a0, zero, e64, m2, ta, ma
|
|
; CHECK-NEXT: vfmv.s.f v10, fa0
|
|
; CHECK-NEXT: vfredosum.vs v8, v8, v10
|
|
; CHECK-NEXT: vfmv.f.s fa0, v8
|
|
; CHECK-NEXT: ret
|
|
%red = call double @llvm.vector.reduce.fadd.nxv2f64(double %s, <vscale x 2 x double> %v)
|
|
ret double %red
|
|
}
|
|
|
|
define double @vreduce_fwadd_nxv2f64(<vscale x 2 x float> %v, double %s) {
|
|
; CHECK-LABEL: vreduce_fwadd_nxv2f64:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: vsetvli a0, zero, e64, m1, ta, ma
|
|
; CHECK-NEXT: vfmv.s.f v9, fa0
|
|
; CHECK-NEXT: vsetvli a0, zero, e32, m1, ta, ma
|
|
; CHECK-NEXT: vfwredusum.vs v8, v8, v9
|
|
; CHECK-NEXT: vsetvli zero, zero, e64, m2, ta, ma
|
|
; CHECK-NEXT: vfmv.f.s fa0, v8
|
|
; CHECK-NEXT: ret
|
|
%e = fpext <vscale x 2 x float> %v to <vscale x 2 x double>
|
|
%red = call reassoc double @llvm.vector.reduce.fadd.nxv2f64(double %s, <vscale x 2 x double> %e)
|
|
ret double %red
|
|
}
|
|
|
|
define double @vreduce_ord_fwadd_nxv2f64(<vscale x 2 x float> %v, double %s) {
|
|
; CHECK-LABEL: vreduce_ord_fwadd_nxv2f64:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: vsetvli a0, zero, e64, m1, ta, ma
|
|
; CHECK-NEXT: vfmv.s.f v9, fa0
|
|
; CHECK-NEXT: vsetvli a0, zero, e32, m1, ta, ma
|
|
; CHECK-NEXT: vfwredosum.vs v8, v8, v9
|
|
; CHECK-NEXT: vsetvli zero, zero, e64, m2, ta, ma
|
|
; CHECK-NEXT: vfmv.f.s fa0, v8
|
|
; CHECK-NEXT: ret
|
|
%e = fpext <vscale x 2 x float> %v to <vscale x 2 x double>
|
|
%red = call double @llvm.vector.reduce.fadd.nxv2f64(double %s, <vscale x 2 x double> %e)
|
|
ret double %red
|
|
}
|
|
|
|
define double @vreduce_fadd_nxv4f64(<vscale x 4 x double> %v, double %s) {
|
|
; CHECK-LABEL: vreduce_fadd_nxv4f64:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: vsetvli a0, zero, e64, m4, ta, ma
|
|
; CHECK-NEXT: vfmv.s.f v12, fa0
|
|
; CHECK-NEXT: vfredusum.vs v8, v8, v12
|
|
; CHECK-NEXT: vfmv.f.s fa0, v8
|
|
; CHECK-NEXT: ret
|
|
%red = call reassoc double @llvm.vector.reduce.fadd.nxv4f64(double %s, <vscale x 4 x double> %v)
|
|
ret double %red
|
|
}
|
|
|
|
define double @vreduce_ord_fadd_nxv4f64(<vscale x 4 x double> %v, double %s) {
|
|
; CHECK-LABEL: vreduce_ord_fadd_nxv4f64:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: vsetvli a0, zero, e64, m4, ta, ma
|
|
; CHECK-NEXT: vfmv.s.f v12, fa0
|
|
; CHECK-NEXT: vfredosum.vs v8, v8, v12
|
|
; CHECK-NEXT: vfmv.f.s fa0, v8
|
|
; CHECK-NEXT: ret
|
|
%red = call double @llvm.vector.reduce.fadd.nxv4f64(double %s, <vscale x 4 x double> %v)
|
|
ret double %red
|
|
}
|
|
|
|
define double @vreduce_fwadd_nxv4f64(<vscale x 4 x float> %v, double %s) {
|
|
; CHECK-LABEL: vreduce_fwadd_nxv4f64:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: vsetvli a0, zero, e64, m1, ta, ma
|
|
; CHECK-NEXT: vfmv.s.f v10, fa0
|
|
; CHECK-NEXT: vsetvli a0, zero, e32, m2, ta, ma
|
|
; CHECK-NEXT: vfwredusum.vs v8, v8, v10
|
|
; CHECK-NEXT: vsetvli zero, zero, e64, m4, ta, ma
|
|
; CHECK-NEXT: vfmv.f.s fa0, v8
|
|
; CHECK-NEXT: ret
|
|
%e = fpext <vscale x 4 x float> %v to <vscale x 4 x double>
|
|
%red = call reassoc double @llvm.vector.reduce.fadd.nxv4f64(double %s, <vscale x 4 x double> %e)
|
|
ret double %red
|
|
}
|
|
|
|
define double @vreduce_ord_fwadd_nxv4f64(<vscale x 4 x float> %v, double %s) {
|
|
; CHECK-LABEL: vreduce_ord_fwadd_nxv4f64:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: vsetvli a0, zero, e64, m1, ta, ma
|
|
; CHECK-NEXT: vfmv.s.f v10, fa0
|
|
; CHECK-NEXT: vsetvli a0, zero, e32, m2, ta, ma
|
|
; CHECK-NEXT: vfwredosum.vs v8, v8, v10
|
|
; CHECK-NEXT: vsetvli zero, zero, e64, m4, ta, ma
|
|
; CHECK-NEXT: vfmv.f.s fa0, v8
|
|
; CHECK-NEXT: ret
|
|
%e = fpext <vscale x 4 x float> %v to <vscale x 4 x double>
|
|
%red = call double @llvm.vector.reduce.fadd.nxv4f64(double %s, <vscale x 4 x double> %e)
|
|
ret double %red
|
|
}
|
|
|
|
define half @vreduce_fmin_nxv1f16(<vscale x 1 x half> %v) {
|
|
; CHECK-LABEL: vreduce_fmin_nxv1f16:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: vsetvli a0, zero, e16, mf4, ta, ma
|
|
; CHECK-NEXT: vfredmin.vs v8, v8, v8
|
|
; CHECK-NEXT: vfmv.f.s fa0, v8
|
|
; CHECK-NEXT: ret
|
|
%red = call half @llvm.vector.reduce.fmin.nxv1f16(<vscale x 1 x half> %v)
|
|
ret half %red
|
|
}
|
|
|
|
define half @vreduce_fmin_nxv1f16_nonans(<vscale x 1 x half> %v) #0 {
|
|
; CHECK-LABEL: vreduce_fmin_nxv1f16_nonans:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: vsetvli a0, zero, e16, mf4, ta, ma
|
|
; CHECK-NEXT: vfredmin.vs v8, v8, v8
|
|
; CHECK-NEXT: vfmv.f.s fa0, v8
|
|
; CHECK-NEXT: ret
|
|
%red = call nnan half @llvm.vector.reduce.fmin.nxv1f16(<vscale x 1 x half> %v)
|
|
ret half %red
|
|
}
|
|
|
|
define half @vreduce_fmin_nxv1f16_nonans_noinfs(<vscale x 1 x half> %v) #1 {
|
|
; CHECK-LABEL: vreduce_fmin_nxv1f16_nonans_noinfs:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: vsetvli a0, zero, e16, mf4, ta, ma
|
|
; CHECK-NEXT: vfredmin.vs v8, v8, v8
|
|
; CHECK-NEXT: vfmv.f.s fa0, v8
|
|
; CHECK-NEXT: ret
|
|
%red = call nnan ninf half @llvm.vector.reduce.fmin.nxv1f16(<vscale x 1 x half> %v)
|
|
ret half %red
|
|
}
|
|
|
|
define half @vreduce_fmin_nxv2f16(<vscale x 2 x half> %v) {
|
|
; CHECK-LABEL: vreduce_fmin_nxv2f16:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: vsetvli a0, zero, e16, mf2, ta, ma
|
|
; CHECK-NEXT: vfredmin.vs v8, v8, v8
|
|
; CHECK-NEXT: vfmv.f.s fa0, v8
|
|
; CHECK-NEXT: ret
|
|
%red = call half @llvm.vector.reduce.fmin.nxv2f16(<vscale x 2 x half> %v)
|
|
ret half %red
|
|
}
|
|
|
|
define half @vreduce_fmin_nxv4f16(<vscale x 4 x half> %v) {
|
|
; CHECK-LABEL: vreduce_fmin_nxv4f16:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: vsetvli a0, zero, e16, m1, ta, ma
|
|
; CHECK-NEXT: vfredmin.vs v8, v8, v8
|
|
; CHECK-NEXT: vfmv.f.s fa0, v8
|
|
; CHECK-NEXT: ret
|
|
%red = call half @llvm.vector.reduce.fmin.nxv4f16(<vscale x 4 x half> %v)
|
|
ret half %red
|
|
}
|
|
|
|
define half @vreduce_fmin_nxv64f16(<vscale x 64 x half> %v) {
|
|
; CHECK-LABEL: vreduce_fmin_nxv64f16:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: vsetvli a0, zero, e16, m8, ta, ma
|
|
; CHECK-NEXT: vfmin.vv v8, v8, v16
|
|
; CHECK-NEXT: vfredmin.vs v8, v8, v8
|
|
; CHECK-NEXT: vfmv.f.s fa0, v8
|
|
; CHECK-NEXT: ret
|
|
%red = call half @llvm.vector.reduce.fmin.nxv64f16(<vscale x 64 x half> %v)
|
|
ret half %red
|
|
}
|
|
|
|
define float @vreduce_fmin_nxv1f32(<vscale x 1 x float> %v) {
|
|
; CHECK-LABEL: vreduce_fmin_nxv1f32:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: vsetvli a0, zero, e32, mf2, ta, ma
|
|
; CHECK-NEXT: vfredmin.vs v8, v8, v8
|
|
; CHECK-NEXT: vfmv.f.s fa0, v8
|
|
; CHECK-NEXT: ret
|
|
%red = call float @llvm.vector.reduce.fmin.nxv1f32(<vscale x 1 x float> %v)
|
|
ret float %red
|
|
}
|
|
|
|
define float @vreduce_fmin_nxv1f32_nonans(<vscale x 1 x float> %v) {
|
|
; CHECK-LABEL: vreduce_fmin_nxv1f32_nonans:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: vsetvli a0, zero, e32, mf2, ta, ma
|
|
; CHECK-NEXT: vfredmin.vs v8, v8, v8
|
|
; CHECK-NEXT: vfmv.f.s fa0, v8
|
|
; CHECK-NEXT: ret
|
|
%red = call nnan float @llvm.vector.reduce.fmin.nxv1f32(<vscale x 1 x float> %v)
|
|
ret float %red
|
|
}
|
|
|
|
define float @vreduce_fmin_nxv1f32_nonans_noinfs(<vscale x 1 x float> %v) {
|
|
; CHECK-LABEL: vreduce_fmin_nxv1f32_nonans_noinfs:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: vsetvli a0, zero, e32, mf2, ta, ma
|
|
; CHECK-NEXT: vfredmin.vs v8, v8, v8
|
|
; CHECK-NEXT: vfmv.f.s fa0, v8
|
|
; CHECK-NEXT: ret
|
|
%red = call nnan ninf float @llvm.vector.reduce.fmin.nxv1f32(<vscale x 1 x float> %v)
|
|
ret float %red
|
|
}
|
|
|
|
define float @vreduce_fmin_nxv2f32(<vscale x 2 x float> %v) {
|
|
; CHECK-LABEL: vreduce_fmin_nxv2f32:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: vsetvli a0, zero, e32, m1, ta, ma
|
|
; CHECK-NEXT: vfredmin.vs v8, v8, v8
|
|
; CHECK-NEXT: vfmv.f.s fa0, v8
|
|
; CHECK-NEXT: ret
|
|
%red = call float @llvm.vector.reduce.fmin.nxv2f32(<vscale x 2 x float> %v)
|
|
ret float %red
|
|
}
|
|
|
|
define float @vreduce_fmin_nxv4f32(<vscale x 4 x float> %v) {
|
|
; CHECK-LABEL: vreduce_fmin_nxv4f32:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: vsetvli a0, zero, e32, m2, ta, ma
|
|
; CHECK-NEXT: vfredmin.vs v8, v8, v8
|
|
; CHECK-NEXT: vfmv.f.s fa0, v8
|
|
; CHECK-NEXT: ret
|
|
%red = call float @llvm.vector.reduce.fmin.nxv4f32(<vscale x 4 x float> %v)
|
|
ret float %red
|
|
}
|
|
|
|
define float @vreduce_fmin_nxv32f32(<vscale x 32 x float> %v) {
|
|
; CHECK-LABEL: vreduce_fmin_nxv32f32:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: vsetvli a0, zero, e32, m8, ta, ma
|
|
; CHECK-NEXT: vfmin.vv v8, v8, v16
|
|
; CHECK-NEXT: vfredmin.vs v8, v8, v8
|
|
; CHECK-NEXT: vfmv.f.s fa0, v8
|
|
; CHECK-NEXT: ret
|
|
%red = call float @llvm.vector.reduce.fmin.nxv32f32(<vscale x 32 x float> %v)
|
|
ret float %red
|
|
}
|
|
|
|
define double @vreduce_fmin_nxv1f64(<vscale x 1 x double> %v) {
|
|
; CHECK-LABEL: vreduce_fmin_nxv1f64:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: vsetvli a0, zero, e64, m1, ta, ma
|
|
; CHECK-NEXT: vfredmin.vs v8, v8, v8
|
|
; CHECK-NEXT: vfmv.f.s fa0, v8
|
|
; CHECK-NEXT: ret
|
|
%red = call double @llvm.vector.reduce.fmin.nxv1f64(<vscale x 1 x double> %v)
|
|
ret double %red
|
|
}
|
|
|
|
define double @vreduce_fmin_nxv1f64_nonans(<vscale x 1 x double> %v) {
|
|
; CHECK-LABEL: vreduce_fmin_nxv1f64_nonans:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: vsetvli a0, zero, e64, m1, ta, ma
|
|
; CHECK-NEXT: vfredmin.vs v8, v8, v8
|
|
; CHECK-NEXT: vfmv.f.s fa0, v8
|
|
; CHECK-NEXT: ret
|
|
%red = call nnan double @llvm.vector.reduce.fmin.nxv1f64(<vscale x 1 x double> %v)
|
|
ret double %red
|
|
}
|
|
|
|
define double @vreduce_fmin_nxv1f64_nonans_noinfs(<vscale x 1 x double> %v) {
|
|
; CHECK-LABEL: vreduce_fmin_nxv1f64_nonans_noinfs:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: vsetvli a0, zero, e64, m1, ta, ma
|
|
; CHECK-NEXT: vfredmin.vs v8, v8, v8
|
|
; CHECK-NEXT: vfmv.f.s fa0, v8
|
|
; CHECK-NEXT: ret
|
|
%red = call nnan ninf double @llvm.vector.reduce.fmin.nxv1f64(<vscale x 1 x double> %v)
|
|
ret double %red
|
|
}
|
|
|
|
define double @vreduce_fmin_nxv2f64(<vscale x 2 x double> %v) {
|
|
; CHECK-LABEL: vreduce_fmin_nxv2f64:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: vsetvli a0, zero, e64, m2, ta, ma
|
|
; CHECK-NEXT: vfredmin.vs v8, v8, v8
|
|
; CHECK-NEXT: vfmv.f.s fa0, v8
|
|
; CHECK-NEXT: ret
|
|
%red = call double @llvm.vector.reduce.fmin.nxv2f64(<vscale x 2 x double> %v)
|
|
ret double %red
|
|
}
|
|
|
|
define double @vreduce_fmin_nxv4f64(<vscale x 4 x double> %v) {
|
|
; CHECK-LABEL: vreduce_fmin_nxv4f64:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: vsetvli a0, zero, e64, m4, ta, ma
|
|
; CHECK-NEXT: vfredmin.vs v8, v8, v8
|
|
; CHECK-NEXT: vfmv.f.s fa0, v8
|
|
; CHECK-NEXT: ret
|
|
%red = call double @llvm.vector.reduce.fmin.nxv4f64(<vscale x 4 x double> %v)
|
|
ret double %red
|
|
}
|
|
|
|
define double @vreduce_fmin_nxv16f64(<vscale x 16 x double> %v) {
|
|
; CHECK-LABEL: vreduce_fmin_nxv16f64:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: vsetvli a0, zero, e64, m8, ta, ma
|
|
; CHECK-NEXT: vfmin.vv v8, v8, v16
|
|
; CHECK-NEXT: vfredmin.vs v8, v8, v8
|
|
; CHECK-NEXT: vfmv.f.s fa0, v8
|
|
; CHECK-NEXT: ret
|
|
%red = call double @llvm.vector.reduce.fmin.nxv16f64(<vscale x 16 x double> %v)
|
|
ret double %red
|
|
}
|
|
|
|
define half @vreduce_fmax_nxv1f16(<vscale x 1 x half> %v) {
|
|
; CHECK-LABEL: vreduce_fmax_nxv1f16:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: vsetvli a0, zero, e16, mf4, ta, ma
|
|
; CHECK-NEXT: vfredmax.vs v8, v8, v8
|
|
; CHECK-NEXT: vfmv.f.s fa0, v8
|
|
; CHECK-NEXT: ret
|
|
%red = call half @llvm.vector.reduce.fmax.nxv1f16(<vscale x 1 x half> %v)
|
|
ret half %red
|
|
}
|
|
|
|
define half @vreduce_fmax_nxv1f16_nonans(<vscale x 1 x half> %v) #0 {
|
|
; CHECK-LABEL: vreduce_fmax_nxv1f16_nonans:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: vsetvli a0, zero, e16, mf4, ta, ma
|
|
; CHECK-NEXT: vfredmax.vs v8, v8, v8
|
|
; CHECK-NEXT: vfmv.f.s fa0, v8
|
|
; CHECK-NEXT: ret
|
|
%red = call nnan half @llvm.vector.reduce.fmax.nxv1f16(<vscale x 1 x half> %v)
|
|
ret half %red
|
|
}
|
|
|
|
define half @vreduce_fmax_nxv1f16_nonans_noinfs(<vscale x 1 x half> %v) #1 {
|
|
; CHECK-LABEL: vreduce_fmax_nxv1f16_nonans_noinfs:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: vsetvli a0, zero, e16, mf4, ta, ma
|
|
; CHECK-NEXT: vfredmax.vs v8, v8, v8
|
|
; CHECK-NEXT: vfmv.f.s fa0, v8
|
|
; CHECK-NEXT: ret
|
|
%red = call nnan ninf half @llvm.vector.reduce.fmax.nxv1f16(<vscale x 1 x half> %v)
|
|
ret half %red
|
|
}
|
|
|
|
define half @vreduce_fmax_nxv2f16(<vscale x 2 x half> %v) {
|
|
; CHECK-LABEL: vreduce_fmax_nxv2f16:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: vsetvli a0, zero, e16, mf2, ta, ma
|
|
; CHECK-NEXT: vfredmax.vs v8, v8, v8
|
|
; CHECK-NEXT: vfmv.f.s fa0, v8
|
|
; CHECK-NEXT: ret
|
|
%red = call half @llvm.vector.reduce.fmax.nxv2f16(<vscale x 2 x half> %v)
|
|
ret half %red
|
|
}
|
|
|
|
define half @vreduce_fmax_nxv4f16(<vscale x 4 x half> %v) {
|
|
; CHECK-LABEL: vreduce_fmax_nxv4f16:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: vsetvli a0, zero, e16, m1, ta, ma
|
|
; CHECK-NEXT: vfredmax.vs v8, v8, v8
|
|
; CHECK-NEXT: vfmv.f.s fa0, v8
|
|
; CHECK-NEXT: ret
|
|
%red = call half @llvm.vector.reduce.fmax.nxv4f16(<vscale x 4 x half> %v)
|
|
ret half %red
|
|
}
|
|
|
|
define half @vreduce_fmax_nxv64f16(<vscale x 64 x half> %v) {
|
|
; CHECK-LABEL: vreduce_fmax_nxv64f16:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: vsetvli a0, zero, e16, m8, ta, ma
|
|
; CHECK-NEXT: vfmax.vv v8, v8, v16
|
|
; CHECK-NEXT: vfredmax.vs v8, v8, v8
|
|
; CHECK-NEXT: vfmv.f.s fa0, v8
|
|
; CHECK-NEXT: ret
|
|
%red = call half @llvm.vector.reduce.fmax.nxv64f16(<vscale x 64 x half> %v)
|
|
ret half %red
|
|
}
|
|
|
|
define float @vreduce_fmax_nxv1f32(<vscale x 1 x float> %v) {
|
|
; CHECK-LABEL: vreduce_fmax_nxv1f32:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: vsetvli a0, zero, e32, mf2, ta, ma
|
|
; CHECK-NEXT: vfredmax.vs v8, v8, v8
|
|
; CHECK-NEXT: vfmv.f.s fa0, v8
|
|
; CHECK-NEXT: ret
|
|
%red = call float @llvm.vector.reduce.fmax.nxv1f32(<vscale x 1 x float> %v)
|
|
ret float %red
|
|
}
|
|
|
|
define float @vreduce_fmax_nxv1f32_nonans(<vscale x 1 x float> %v) {
|
|
; CHECK-LABEL: vreduce_fmax_nxv1f32_nonans:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: vsetvli a0, zero, e32, mf2, ta, ma
|
|
; CHECK-NEXT: vfredmax.vs v8, v8, v8
|
|
; CHECK-NEXT: vfmv.f.s fa0, v8
|
|
; CHECK-NEXT: ret
|
|
%red = call nnan float @llvm.vector.reduce.fmax.nxv1f32(<vscale x 1 x float> %v)
|
|
ret float %red
|
|
}
|
|
|
|
define float @vreduce_fmax_nxv1f32_nonans_noinfs(<vscale x 1 x float> %v) {
|
|
; CHECK-LABEL: vreduce_fmax_nxv1f32_nonans_noinfs:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: vsetvli a0, zero, e32, mf2, ta, ma
|
|
; CHECK-NEXT: vfredmax.vs v8, v8, v8
|
|
; CHECK-NEXT: vfmv.f.s fa0, v8
|
|
; CHECK-NEXT: ret
|
|
%red = call nnan ninf float @llvm.vector.reduce.fmax.nxv1f32(<vscale x 1 x float> %v)
|
|
ret float %red
|
|
}
|
|
|
|
define float @vreduce_fmax_nxv2f32(<vscale x 2 x float> %v) {
|
|
; CHECK-LABEL: vreduce_fmax_nxv2f32:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: vsetvli a0, zero, e32, m1, ta, ma
|
|
; CHECK-NEXT: vfredmax.vs v8, v8, v8
|
|
; CHECK-NEXT: vfmv.f.s fa0, v8
|
|
; CHECK-NEXT: ret
|
|
%red = call float @llvm.vector.reduce.fmax.nxv2f32(<vscale x 2 x float> %v)
|
|
ret float %red
|
|
}
|
|
|
|
define float @vreduce_fmax_nxv4f32(<vscale x 4 x float> %v) {
|
|
; CHECK-LABEL: vreduce_fmax_nxv4f32:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: vsetvli a0, zero, e32, m2, ta, ma
|
|
; CHECK-NEXT: vfredmax.vs v8, v8, v8
|
|
; CHECK-NEXT: vfmv.f.s fa0, v8
|
|
; CHECK-NEXT: ret
|
|
%red = call float @llvm.vector.reduce.fmax.nxv4f32(<vscale x 4 x float> %v)
|
|
ret float %red
|
|
}
|
|
|
|
define float @vreduce_fmax_nxv32f32(<vscale x 32 x float> %v) {
|
|
; CHECK-LABEL: vreduce_fmax_nxv32f32:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: vsetvli a0, zero, e32, m8, ta, ma
|
|
; CHECK-NEXT: vfmax.vv v8, v8, v16
|
|
; CHECK-NEXT: vfredmax.vs v8, v8, v8
|
|
; CHECK-NEXT: vfmv.f.s fa0, v8
|
|
; CHECK-NEXT: ret
|
|
%red = call float @llvm.vector.reduce.fmax.nxv32f32(<vscale x 32 x float> %v)
|
|
ret float %red
|
|
}
|
|
|
|
define double @vreduce_fmax_nxv1f64(<vscale x 1 x double> %v) {
|
|
; CHECK-LABEL: vreduce_fmax_nxv1f64:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: vsetvli a0, zero, e64, m1, ta, ma
|
|
; CHECK-NEXT: vfredmax.vs v8, v8, v8
|
|
; CHECK-NEXT: vfmv.f.s fa0, v8
|
|
; CHECK-NEXT: ret
|
|
%red = call double @llvm.vector.reduce.fmax.nxv1f64(<vscale x 1 x double> %v)
|
|
ret double %red
|
|
}
|
|
|
|
define double @vreduce_fmax_nxv1f64_nonans(<vscale x 1 x double> %v) {
|
|
; CHECK-LABEL: vreduce_fmax_nxv1f64_nonans:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: vsetvli a0, zero, e64, m1, ta, ma
|
|
; CHECK-NEXT: vfredmax.vs v8, v8, v8
|
|
; CHECK-NEXT: vfmv.f.s fa0, v8
|
|
; CHECK-NEXT: ret
|
|
%red = call nnan double @llvm.vector.reduce.fmax.nxv1f64(<vscale x 1 x double> %v)
|
|
ret double %red
|
|
}
|
|
|
|
define double @vreduce_fmax_nxv1f64_nonans_noinfs(<vscale x 1 x double> %v) {
|
|
; CHECK-LABEL: vreduce_fmax_nxv1f64_nonans_noinfs:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: vsetvli a0, zero, e64, m1, ta, ma
|
|
; CHECK-NEXT: vfredmax.vs v8, v8, v8
|
|
; CHECK-NEXT: vfmv.f.s fa0, v8
|
|
; CHECK-NEXT: ret
|
|
%red = call nnan ninf double @llvm.vector.reduce.fmax.nxv1f64(<vscale x 1 x double> %v)
|
|
ret double %red
|
|
}
|
|
|
|
define double @vreduce_fmax_nxv2f64(<vscale x 2 x double> %v) {
|
|
; CHECK-LABEL: vreduce_fmax_nxv2f64:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: vsetvli a0, zero, e64, m2, ta, ma
|
|
; CHECK-NEXT: vfredmax.vs v8, v8, v8
|
|
; CHECK-NEXT: vfmv.f.s fa0, v8
|
|
; CHECK-NEXT: ret
|
|
%red = call double @llvm.vector.reduce.fmax.nxv2f64(<vscale x 2 x double> %v)
|
|
ret double %red
|
|
}
|
|
|
|
define double @vreduce_fmax_nxv4f64(<vscale x 4 x double> %v) {
|
|
; CHECK-LABEL: vreduce_fmax_nxv4f64:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: vsetvli a0, zero, e64, m4, ta, ma
|
|
; CHECK-NEXT: vfredmax.vs v8, v8, v8
|
|
; CHECK-NEXT: vfmv.f.s fa0, v8
|
|
; CHECK-NEXT: ret
|
|
%red = call double @llvm.vector.reduce.fmax.nxv4f64(<vscale x 4 x double> %v)
|
|
ret double %red
|
|
}
|
|
|
|
define double @vreduce_fmax_nxv16f64(<vscale x 16 x double> %v) {
|
|
; CHECK-LABEL: vreduce_fmax_nxv16f64:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: vsetvli a0, zero, e64, m8, ta, ma
|
|
; CHECK-NEXT: vfmax.vv v8, v8, v16
|
|
; CHECK-NEXT: vfredmax.vs v8, v8, v8
|
|
; CHECK-NEXT: vfmv.f.s fa0, v8
|
|
; CHECK-NEXT: ret
|
|
%red = call double @llvm.vector.reduce.fmax.nxv16f64(<vscale x 16 x double> %v)
|
|
ret double %red
|
|
}
|
|
|
|
define float @vreduce_nsz_fadd_nxv1f32(<vscale x 1 x float> %v, float %s) {
|
|
; CHECK-LABEL: vreduce_nsz_fadd_nxv1f32:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: vsetvli a0, zero, e32, mf2, ta, ma
|
|
; CHECK-NEXT: vfmv.s.f v9, fa0
|
|
; CHECK-NEXT: vfredusum.vs v8, v8, v9
|
|
; CHECK-NEXT: vfmv.f.s fa0, v8
|
|
; CHECK-NEXT: ret
|
|
%red = call reassoc nsz float @llvm.vector.reduce.fadd.nxv1f32(float %s, <vscale x 1 x float> %v)
|
|
ret float %red
|
|
}
|
|
|
|
; Test Widen VECREDUCE_SEQ_FADD
|
|
|
|
define half @vreduce_ord_fadd_nxv3f16(<vscale x 3 x half> %v, half %s) {
|
|
; CHECK-LABEL: vreduce_ord_fadd_nxv3f16:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: csrr a0, vlenb
|
|
; CHECK-NEXT: srli a1, a0, 3
|
|
; CHECK-NEXT: srli a0, a0, 2
|
|
; CHECK-NEXT: add a0, a0, a1
|
|
; CHECK-NEXT: vsetivli zero, 1, e16, m1, ta, ma
|
|
; CHECK-NEXT: vfmv.s.f v9, fa0
|
|
; CHECK-NEXT: vsetvli zero, a0, e16, m1, ta, ma
|
|
; CHECK-NEXT: vfredosum.vs v9, v8, v9
|
|
; CHECK-NEXT: vfmv.f.s fa0, v9
|
|
; CHECK-NEXT: ret
|
|
%red = call half @llvm.vector.reduce.fadd.nxv3f16(half %s, <vscale x 3 x half> %v)
|
|
ret half %red
|
|
}
|
|
|
|
define half @vreduce_ord_fadd_nxv6f16(<vscale x 6 x half> %v, half %s) {
|
|
; CHECK-LABEL: vreduce_ord_fadd_nxv6f16:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: csrr a0, vlenb
|
|
; CHECK-NEXT: srli a1, a0, 2
|
|
; CHECK-NEXT: sub a0, a0, a1
|
|
; CHECK-NEXT: vsetivli zero, 1, e16, m1, ta, ma
|
|
; CHECK-NEXT: vfmv.s.f v10, fa0
|
|
; CHECK-NEXT: vsetvli zero, a0, e16, m2, ta, ma
|
|
; CHECK-NEXT: vfredosum.vs v10, v8, v10
|
|
; CHECK-NEXT: vfmv.f.s fa0, v10
|
|
; CHECK-NEXT: ret
|
|
%red = call half @llvm.vector.reduce.fadd.nxv6f16(half %s, <vscale x 6 x half> %v)
|
|
ret half %red
|
|
}
|
|
|
|
define half @vreduce_ord_fadd_nxv10f16(<vscale x 10 x half> %v, half %s) {
|
|
; CHECK-LABEL: vreduce_ord_fadd_nxv10f16:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: csrr a0, vlenb
|
|
; CHECK-NEXT: srli a1, a0, 2
|
|
; CHECK-NEXT: add a0, a0, a1
|
|
; CHECK-NEXT: vsetivli zero, 1, e16, m1, ta, ma
|
|
; CHECK-NEXT: vfmv.s.f v12, fa0
|
|
; CHECK-NEXT: vsetvli zero, a0, e16, m4, ta, ma
|
|
; CHECK-NEXT: vfredosum.vs v12, v8, v12
|
|
; CHECK-NEXT: vfmv.f.s fa0, v12
|
|
; CHECK-NEXT: ret
|
|
%red = call half @llvm.vector.reduce.fadd.nxv10f16(half %s, <vscale x 10 x half> %v)
|
|
ret half %red
|
|
}
|
|
|
|
define half @vreduce_ord_fadd_nxv12f16(<vscale x 12 x half> %v, half %s) {
|
|
; CHECK-LABEL: vreduce_ord_fadd_nxv12f16:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: csrr a0, vlenb
|
|
; CHECK-NEXT: srli a1, a0, 1
|
|
; CHECK-NEXT: add a0, a0, a1
|
|
; CHECK-NEXT: vsetivli zero, 1, e16, m1, ta, ma
|
|
; CHECK-NEXT: vfmv.s.f v12, fa0
|
|
; CHECK-NEXT: vsetvli zero, a0, e16, m4, ta, ma
|
|
; CHECK-NEXT: vfredosum.vs v12, v8, v12
|
|
; CHECK-NEXT: vfmv.f.s fa0, v12
|
|
; CHECK-NEXT: ret
|
|
%red = call half @llvm.vector.reduce.fadd.nxv12f16(half %s, <vscale x 12 x half> %v)
|
|
ret half %red
|
|
}
|
|
|
|
; Test Widen vector reduce type (fadd/fmin/fmax)
|
|
define half @vreduce_fadd_nxv3f16(<vscale x 3 x half> %v, half %s) {
|
|
; CHECK-LABEL: vreduce_fadd_nxv3f16:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: csrr a0, vlenb
|
|
; CHECK-NEXT: vsetivli zero, 1, e16, m1, ta, ma
|
|
; CHECK-NEXT: vfmv.s.f v9, fa0
|
|
; CHECK-NEXT: srli a1, a0, 3
|
|
; CHECK-NEXT: srli a0, a0, 2
|
|
; CHECK-NEXT: add a0, a0, a1
|
|
; CHECK-NEXT: lui a1, 1048568
|
|
; CHECK-NEXT: vmv.s.x v10, a1
|
|
; CHECK-NEXT: vsetvli zero, a0, e16, m1, ta, ma
|
|
; CHECK-NEXT: vfredusum.vs v10, v8, v9
|
|
; CHECK-NEXT: vfmv.f.s fa0, v10
|
|
; CHECK-NEXT: ret
|
|
%red = call reassoc half @llvm.vector.reduce.fadd.nxv3f16(half %s, <vscale x 3 x half> %v)
|
|
ret half %red
|
|
}
|
|
|
|
define half @vreduce_fadd_nxv6f16(<vscale x 6 x half> %v, half %s) {
|
|
; CHECK-LABEL: vreduce_fadd_nxv6f16:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: csrr a0, vlenb
|
|
; CHECK-NEXT: vsetivli zero, 1, e16, m1, ta, ma
|
|
; CHECK-NEXT: vfmv.s.f v10, fa0
|
|
; CHECK-NEXT: srli a1, a0, 2
|
|
; CHECK-NEXT: sub a0, a0, a1
|
|
; CHECK-NEXT: lui a1, 1048568
|
|
; CHECK-NEXT: vmv.s.x v11, a1
|
|
; CHECK-NEXT: vsetvli zero, a0, e16, m2, ta, ma
|
|
; CHECK-NEXT: vfredusum.vs v11, v8, v10
|
|
; CHECK-NEXT: vfmv.f.s fa0, v11
|
|
; CHECK-NEXT: ret
|
|
%red = call reassoc half @llvm.vector.reduce.fadd.nxv6f16(half %s, <vscale x 6 x half> %v)
|
|
ret half %red
|
|
}
|
|
|
|
define half @vreduce_fmin_nxv10f16(<vscale x 10 x half> %v) {
|
|
; CHECK-LABEL: vreduce_fmin_nxv10f16:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: csrr a0, vlenb
|
|
; CHECK-NEXT: srli a1, a0, 2
|
|
; CHECK-NEXT: add a0, a0, a1
|
|
; CHECK-NEXT: lui a1, 8
|
|
; CHECK-NEXT: addi a1, a1, -512
|
|
; CHECK-NEXT: vsetivli zero, 1, e16, m1, ta, ma
|
|
; CHECK-NEXT: vmv.s.x v12, a1
|
|
; CHECK-NEXT: vsetvli zero, a0, e16, m4, ta, ma
|
|
; CHECK-NEXT: vfredmin.vs v12, v8, v12
|
|
; CHECK-NEXT: vfmv.f.s fa0, v12
|
|
; CHECK-NEXT: ret
|
|
%red = call half @llvm.vector.reduce.fmin.nxv10f16(<vscale x 10 x half> %v)
|
|
ret half %red
|
|
}
|
|
|
|
define half @vreduce_fmax_nxv12f16(<vscale x 12 x half> %v) {
|
|
; CHECK-LABEL: vreduce_fmax_nxv12f16:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: csrr a0, vlenb
|
|
; CHECK-NEXT: srli a1, a0, 1
|
|
; CHECK-NEXT: add a0, a0, a1
|
|
; CHECK-NEXT: li a1, -512
|
|
; CHECK-NEXT: vsetivli zero, 1, e16, m1, ta, ma
|
|
; CHECK-NEXT: vmv.s.x v12, a1
|
|
; CHECK-NEXT: vsetvli zero, a0, e16, m4, ta, ma
|
|
; CHECK-NEXT: vfredmax.vs v12, v8, v12
|
|
; CHECK-NEXT: vfmv.f.s fa0, v12
|
|
; CHECK-NEXT: ret
|
|
%red = call half @llvm.vector.reduce.fmax.nxv12f16(<vscale x 12 x half> %v)
|
|
ret half %red
|
|
}
|
|
|
|
define float @vreduce_fmul_nxv1f32(<vscale x 1 x float> %v, float %s) {
|
|
; CHECK-LABEL: vreduce_fmul_nxv1f32:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: li a0, 0
|
|
; CHECK-NEXT: csrr a1, vlenb
|
|
; CHECK-NEXT: srli a1, a1, 3
|
|
; CHECK-NEXT: vsetivli zero, 1, e32, mf2, ta, ma
|
|
; CHECK-NEXT: .LBB75_1: # %rdx.loop
|
|
; CHECK-NEXT: # =>This Inner Loop Header: Depth=1
|
|
; CHECK-NEXT: vslidedown.vx v9, v8, a0
|
|
; CHECK-NEXT: vfmv.f.s fa5, v9
|
|
; CHECK-NEXT: addi a0, a0, 1
|
|
; CHECK-NEXT: fmul.s fa0, fa0, fa5
|
|
; CHECK-NEXT: bne a0, a1, .LBB75_1
|
|
; CHECK-NEXT: # %bb.2: # %rdx.exit
|
|
; CHECK-NEXT: ret
|
|
%red = call reassoc float @llvm.vector.reduce.fmul.nxv1f32(float %s, <vscale x 1 x float> %v)
|
|
ret float %red
|
|
}
|
|
|
|
define float @vreduce_ord_fmul_nxv1f32(<vscale x 1 x float> %v, float %s) {
|
|
; CHECK-LABEL: vreduce_ord_fmul_nxv1f32:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: li a0, 0
|
|
; CHECK-NEXT: csrr a1, vlenb
|
|
; CHECK-NEXT: srli a1, a1, 3
|
|
; CHECK-NEXT: vsetivli zero, 1, e32, mf2, ta, ma
|
|
; CHECK-NEXT: .LBB76_1: # %rdx.loop
|
|
; CHECK-NEXT: # =>This Inner Loop Header: Depth=1
|
|
; CHECK-NEXT: vslidedown.vx v9, v8, a0
|
|
; CHECK-NEXT: vfmv.f.s fa5, v9
|
|
; CHECK-NEXT: addi a0, a0, 1
|
|
; CHECK-NEXT: fmul.s fa0, fa0, fa5
|
|
; CHECK-NEXT: bne a0, a1, .LBB76_1
|
|
; CHECK-NEXT: # %bb.2: # %rdx.exit
|
|
; CHECK-NEXT: ret
|
|
%red = call float @llvm.vector.reduce.fmul.nxv1f32(float %s, <vscale x 1 x float> %v)
|
|
ret float %red
|
|
}
|
|
|
|
define float @vreduce_fmul_nxv2f32(<vscale x 2 x float> %v, float %s) {
|
|
; CHECK-LABEL: vreduce_fmul_nxv2f32:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: li a0, 0
|
|
; CHECK-NEXT: csrr a1, vlenb
|
|
; CHECK-NEXT: srli a1, a1, 2
|
|
; CHECK-NEXT: vsetivli zero, 1, e32, m1, ta, ma
|
|
; CHECK-NEXT: .LBB77_1: # %rdx.loop
|
|
; CHECK-NEXT: # =>This Inner Loop Header: Depth=1
|
|
; CHECK-NEXT: vslidedown.vx v9, v8, a0
|
|
; CHECK-NEXT: vfmv.f.s fa5, v9
|
|
; CHECK-NEXT: addi a0, a0, 1
|
|
; CHECK-NEXT: fmul.s fa0, fa0, fa5
|
|
; CHECK-NEXT: bne a0, a1, .LBB77_1
|
|
; CHECK-NEXT: # %bb.2: # %rdx.exit
|
|
; CHECK-NEXT: ret
|
|
%red = call reassoc float @llvm.vector.reduce.fmul.nxv2f32(float %s, <vscale x 2 x float> %v)
|
|
ret float %red
|
|
}
|
|
|
|
define float @vreduce_ord_fmul_nxv2f32(<vscale x 2 x float> %v, float %s) {
|
|
; CHECK-LABEL: vreduce_ord_fmul_nxv2f32:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: li a0, 0
|
|
; CHECK-NEXT: csrr a1, vlenb
|
|
; CHECK-NEXT: srli a1, a1, 2
|
|
; CHECK-NEXT: vsetivli zero, 1, e32, m1, ta, ma
|
|
; CHECK-NEXT: .LBB78_1: # %rdx.loop
|
|
; CHECK-NEXT: # =>This Inner Loop Header: Depth=1
|
|
; CHECK-NEXT: vslidedown.vx v9, v8, a0
|
|
; CHECK-NEXT: vfmv.f.s fa5, v9
|
|
; CHECK-NEXT: addi a0, a0, 1
|
|
; CHECK-NEXT: fmul.s fa0, fa0, fa5
|
|
; CHECK-NEXT: bne a0, a1, .LBB78_1
|
|
; CHECK-NEXT: # %bb.2: # %rdx.exit
|
|
; CHECK-NEXT: ret
|
|
%red = call float @llvm.vector.reduce.fmul.nxv2f32(float %s, <vscale x 2 x float> %v)
|
|
ret float %red
|
|
}
|
|
|
|
define float @vreduce_fmul_nxv4f32(<vscale x 4 x float> %v, float %s) {
|
|
; CHECK-LABEL: vreduce_fmul_nxv4f32:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: li a0, 0
|
|
; CHECK-NEXT: csrr a1, vlenb
|
|
; CHECK-NEXT: srli a1, a1, 1
|
|
; CHECK-NEXT: vsetivli zero, 1, e32, m2, ta, ma
|
|
; CHECK-NEXT: .LBB79_1: # %rdx.loop
|
|
; CHECK-NEXT: # =>This Inner Loop Header: Depth=1
|
|
; CHECK-NEXT: vslidedown.vx v10, v8, a0
|
|
; CHECK-NEXT: vfmv.f.s fa5, v10
|
|
; CHECK-NEXT: addi a0, a0, 1
|
|
; CHECK-NEXT: fmul.s fa0, fa0, fa5
|
|
; CHECK-NEXT: bne a0, a1, .LBB79_1
|
|
; CHECK-NEXT: # %bb.2: # %rdx.exit
|
|
; CHECK-NEXT: ret
|
|
%red = call reassoc float @llvm.vector.reduce.fmul.nxv4f32(float %s, <vscale x 4 x float> %v)
|
|
ret float %red
|
|
}
|
|
|
|
define double @vreduce_fmul_nxv1f64(<vscale x 1 x double> %v, double %s) {
|
|
; CHECK-LABEL: vreduce_fmul_nxv1f64:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: li a0, 0
|
|
; CHECK-NEXT: csrr a1, vlenb
|
|
; CHECK-NEXT: srli a1, a1, 3
|
|
; CHECK-NEXT: vsetivli zero, 1, e64, m1, ta, ma
|
|
; CHECK-NEXT: .LBB80_1: # %rdx.loop
|
|
; CHECK-NEXT: # =>This Inner Loop Header: Depth=1
|
|
; CHECK-NEXT: vslidedown.vx v9, v8, a0
|
|
; CHECK-NEXT: vfmv.f.s fa5, v9
|
|
; CHECK-NEXT: addi a0, a0, 1
|
|
; CHECK-NEXT: fmul.d fa0, fa0, fa5
|
|
; CHECK-NEXT: bne a0, a1, .LBB80_1
|
|
; CHECK-NEXT: # %bb.2: # %rdx.exit
|
|
; CHECK-NEXT: ret
|
|
%red = call reassoc double @llvm.vector.reduce.fmul.nxv1f64(double %s, <vscale x 1 x double> %v)
|
|
ret double %red
|
|
}
|
|
|
|
define double @vreduce_ord_fmul_nxv1f64(<vscale x 1 x double> %v, double %s) {
|
|
; CHECK-LABEL: vreduce_ord_fmul_nxv1f64:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: li a0, 0
|
|
; CHECK-NEXT: csrr a1, vlenb
|
|
; CHECK-NEXT: srli a1, a1, 3
|
|
; CHECK-NEXT: vsetivli zero, 1, e64, m1, ta, ma
|
|
; CHECK-NEXT: .LBB81_1: # %rdx.loop
|
|
; CHECK-NEXT: # =>This Inner Loop Header: Depth=1
|
|
; CHECK-NEXT: vslidedown.vx v9, v8, a0
|
|
; CHECK-NEXT: vfmv.f.s fa5, v9
|
|
; CHECK-NEXT: addi a0, a0, 1
|
|
; CHECK-NEXT: fmul.d fa0, fa0, fa5
|
|
; CHECK-NEXT: bne a0, a1, .LBB81_1
|
|
; CHECK-NEXT: # %bb.2: # %rdx.exit
|
|
; CHECK-NEXT: ret
|
|
%red = call double @llvm.vector.reduce.fmul.nxv1f64(double %s, <vscale x 1 x double> %v)
|
|
ret double %red
|
|
}
|