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
341 lines
19 KiB
LLVM
341 lines
19 KiB
LLVM
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
|
|
; RUN: llc -mtriple=aarch64-linux-gnu -mattr=+sme2 -mattr=+sme-i16i64 -mattr=+sme-f64f64 -force-streaming -verify-machineinstrs < %s | FileCheck %s
|
|
|
|
;
|
|
; SUB Multi-Single x2
|
|
;
|
|
|
|
define void @multi_vector_sub_write_single_za_vg1x2_i32(i32 %slice, <vscale x 4 x i32> %zn0, <vscale x 4 x i32> %zn1, <vscale x 4 x i32> %zm) {
|
|
; CHECK-LABEL: multi_vector_sub_write_single_za_vg1x2_i32:
|
|
; CHECK: // %bb.0:
|
|
; CHECK-NEXT: mov w8, w0
|
|
; CHECK-NEXT: sub za.s[w8, 0, vgx2], { z0.s, z1.s }, z2.s
|
|
; CHECK-NEXT: sub za.s[w8, 7, vgx2], { z0.s, z1.s }, z2.s
|
|
; CHECK-NEXT: ret
|
|
call void @llvm.aarch64.sme.sub.write.single.za.vg1x2.nxv4i32(i32 %slice,
|
|
<vscale x 4 x i32> %zn0, <vscale x 4 x i32> %zn1,
|
|
<vscale x 4 x i32> %zm)
|
|
%slice.7 = add i32 %slice, 7
|
|
call void @llvm.aarch64.sme.sub.write.single.za.vg1x2.nxv4i32(i32 %slice.7,
|
|
<vscale x 4 x i32> %zn0, <vscale x 4 x i32> %zn1,
|
|
<vscale x 4 x i32> %zm)
|
|
ret void
|
|
}
|
|
|
|
define void @multi_vector_sub_write_single_za_vg1x2_i64(i32 %slice, <vscale x 2 x i64> %zn0, <vscale x 2 x i64> %zn1, <vscale x 2 x i64> %zm) {
|
|
; CHECK-LABEL: multi_vector_sub_write_single_za_vg1x2_i64:
|
|
; CHECK: // %bb.0:
|
|
; CHECK-NEXT: mov w8, w0
|
|
; CHECK-NEXT: sub za.d[w8, 0, vgx2], { z0.d, z1.d }, z2.d
|
|
; CHECK-NEXT: sub za.d[w8, 7, vgx2], { z0.d, z1.d }, z2.d
|
|
; CHECK-NEXT: ret
|
|
call void @llvm.aarch64.sme.sub.write.single.za.vg1x2.nxv2i64(i32 %slice,
|
|
<vscale x 2 x i64> %zn0, <vscale x 2 x i64> %zn1,
|
|
<vscale x 2 x i64> %zm)
|
|
%slice.7 = add i32 %slice, 7
|
|
call void @llvm.aarch64.sme.sub.write.single.za.vg1x2.nxv2i64(i32 %slice.7,
|
|
<vscale x 2 x i64> %zn0, <vscale x 2 x i64> %zn1,
|
|
<vscale x 2 x i64> %zm)
|
|
ret void
|
|
}
|
|
|
|
;
|
|
; SUB Multi-Single x4
|
|
;
|
|
|
|
define void @multi_vector_sub_write_single_za_vg1x4_i32(i32 %slice, <vscale x 4 x i32> %zn0, <vscale x 4 x i32> %zn1,
|
|
; CHECK-LABEL: multi_vector_sub_write_single_za_vg1x4_i32:
|
|
; CHECK: // %bb.0:
|
|
; CHECK-NEXT: mov w8, w0
|
|
; CHECK-NEXT: sub za.s[w8, 0, vgx4], { z0.s - z3.s }, z4.s
|
|
; CHECK-NEXT: sub za.s[w8, 7, vgx4], { z0.s - z3.s }, z4.s
|
|
; CHECK-NEXT: ret
|
|
<vscale x 4 x i32> %zn2, <vscale x 4 x i32> %zn3,
|
|
<vscale x 4 x i32> %zm) {
|
|
call void @llvm.aarch64.sme.sub.write.single.za.vg1x4.nxv4i32(i32 %slice,
|
|
<vscale x 4 x i32> %zn0, <vscale x 4 x i32> %zn1,
|
|
<vscale x 4 x i32> %zn2, <vscale x 4 x i32> %zn3,
|
|
<vscale x 4 x i32> %zm)
|
|
%slice.7 = add i32 %slice, 7
|
|
call void @llvm.aarch64.sme.sub.write.single.za.vg1x4.nxv4i32(i32 %slice.7,
|
|
<vscale x 4 x i32> %zn0, <vscale x 4 x i32> %zn1,
|
|
<vscale x 4 x i32> %zn2, <vscale x 4 x i32> %zn3,
|
|
<vscale x 4 x i32> %zm)
|
|
ret void
|
|
}
|
|
|
|
define void @multi_vector_sub_write_single_za_vg1x4_i64(i32 %slice,
|
|
; CHECK-LABEL: multi_vector_sub_write_single_za_vg1x4_i64:
|
|
; CHECK: // %bb.0:
|
|
; CHECK-NEXT: mov w8, w0
|
|
; CHECK-NEXT: sub za.d[w8, 0, vgx4], { z0.d - z3.d }, z4.d
|
|
; CHECK-NEXT: sub za.d[w8, 7, vgx4], { z0.d - z3.d }, z4.d
|
|
; CHECK-NEXT: ret
|
|
<vscale x 2 x i64> %zn0, <vscale x 2 x i64> %zn1,
|
|
<vscale x 2 x i64> %zn2, <vscale x 2 x i64> %zn3,
|
|
<vscale x 2 x i64> %zm) {
|
|
call void @llvm.aarch64.sme.sub.write.single.za.vg1x4.nxv2i64(i32 %slice,
|
|
<vscale x 2 x i64> %zn0, <vscale x 2 x i64> %zn1,
|
|
<vscale x 2 x i64> %zn2, <vscale x 2 x i64> %zn3,
|
|
<vscale x 2 x i64> %zm)
|
|
%slice.7 = add i32 %slice, 7
|
|
call void @llvm.aarch64.sme.sub.write.single.za.vg1x4.nxv2i64(i32 %slice.7,
|
|
<vscale x 2 x i64> %zn0, <vscale x 2 x i64> %zn1,
|
|
<vscale x 2 x i64> %zn2, <vscale x 2 x i64> %zn3,
|
|
<vscale x 2 x i64> %zm)
|
|
ret void
|
|
}
|
|
|
|
;
|
|
; SUB Multi-Multi x2
|
|
;
|
|
|
|
define void @multi_vector_sub_write_za_vg1x2_i32(i32 %slice, <vscale x 4 x i32> %zn0, <vscale x 4 x i32> %zn1,
|
|
; CHECK-LABEL: multi_vector_sub_write_za_vg1x2_i32:
|
|
; CHECK: // %bb.0:
|
|
; CHECK-NEXT: mov w8, w0
|
|
; CHECK-NEXT: sub za.s[w8, 0, vgx2], { z0.s, z1.s }, { z2.s, z3.s }
|
|
; CHECK-NEXT: sub za.s[w8, 7, vgx2], { z0.s, z1.s }, { z2.s, z3.s }
|
|
; CHECK-NEXT: ret
|
|
<vscale x 4 x i32> %zm1, <vscale x 4 x i32> %zm2) {
|
|
call void @llvm.aarch64.sme.sub.write.za.vg1x2.nxv4i32(i32 %slice,
|
|
<vscale x 4 x i32> %zn0, <vscale x 4 x i32> %zn1,
|
|
<vscale x 4 x i32> %zm1, <vscale x 4 x i32> %zm2)
|
|
%slice.7 = add i32 %slice, 7
|
|
call void @llvm.aarch64.sme.sub.write.za.vg1x2.nxv4i32(i32 %slice.7,
|
|
<vscale x 4 x i32> %zn0, <vscale x 4 x i32> %zn1,
|
|
<vscale x 4 x i32> %zm1, <vscale x 4 x i32> %zm2)
|
|
ret void
|
|
}
|
|
|
|
|
|
define void @multi_vector_sub_write_za_vg1x2_i64(i32 %slice, <vscale x 2 x i64> %zn0, <vscale x 2 x i64> %zn1,
|
|
; CHECK-LABEL: multi_vector_sub_write_za_vg1x2_i64:
|
|
; CHECK: // %bb.0:
|
|
; CHECK-NEXT: mov w8, w0
|
|
; CHECK-NEXT: sub za.d[w8, 0, vgx2], { z0.d, z1.d }, { z2.d, z3.d }
|
|
; CHECK-NEXT: sub za.d[w8, 7, vgx2], { z0.d, z1.d }, { z2.d, z3.d }
|
|
; CHECK-NEXT: ret
|
|
<vscale x 2 x i64> %zm1, <vscale x 2 x i64> %zm2) {
|
|
call void @llvm.aarch64.sme.sub.write.za.vg1x2.nxv2i64(i32 %slice,
|
|
<vscale x 2 x i64> %zn0, <vscale x 2 x i64> %zn1,
|
|
<vscale x 2 x i64> %zm1, <vscale x 2 x i64> %zm2)
|
|
%slice.7 = add i32 %slice, 7
|
|
call void @llvm.aarch64.sme.sub.write.za.vg1x2.nxv2i64(i32 %slice.7,
|
|
<vscale x 2 x i64> %zn0, <vscale x 2 x i64> %zn1,
|
|
<vscale x 2 x i64> %zm1, <vscale x 2 x i64> %zm2)
|
|
ret void
|
|
}
|
|
|
|
|
|
;
|
|
; SUB Multi-Multi x4
|
|
;
|
|
|
|
define void @multi_vector_sub_write_za_vg1x4_i32(i32 %slice, <vscale x 4 x i32> %zn0, <vscale x 4 x i32> %zn1,
|
|
; CHECK-LABEL: multi_vector_sub_write_za_vg1x4_i32:
|
|
; CHECK: // %bb.0:
|
|
; CHECK-NEXT: mov w8, w0
|
|
; CHECK-NEXT: sub za.s[w8, 0, vgx4], { z0.s - z3.s }, { z4.s - z7.s }
|
|
; CHECK-NEXT: sub za.s[w8, 7, vgx4], { z0.s - z3.s }, { z4.s - z7.s }
|
|
; CHECK-NEXT: ret
|
|
<vscale x 4 x i32> %zn2, <vscale x 4 x i32> %zn3,
|
|
<vscale x 4 x i32> %zm0, <vscale x 4 x i32> %zm1,
|
|
<vscale x 4 x i32> %zm2, <vscale x 4 x i32> %zm3) {
|
|
call void @llvm.aarch64.sme.sub.write.za.vg1x4.nxv4i32(i32 %slice,
|
|
<vscale x 4 x i32> %zn0, <vscale x 4 x i32> %zn1,
|
|
<vscale x 4 x i32> %zn2, <vscale x 4 x i32> %zn3,
|
|
<vscale x 4 x i32> %zm0, <vscale x 4 x i32> %zm1,
|
|
<vscale x 4 x i32> %zm2, <vscale x 4 x i32> %zm3)
|
|
%slice.7 = add i32 %slice, 7
|
|
call void @llvm.aarch64.sme.sub.write.za.vg1x4.nxv4i32(i32 %slice.7,
|
|
<vscale x 4 x i32> %zn0, <vscale x 4 x i32> %zn1,
|
|
<vscale x 4 x i32> %zn2, <vscale x 4 x i32> %zn3,
|
|
<vscale x 4 x i32> %zm0, <vscale x 4 x i32> %zm1,
|
|
<vscale x 4 x i32> %zm2, <vscale x 4 x i32> %zm3)
|
|
ret void
|
|
}
|
|
|
|
define void @multi_vector_sub_write_za_vg1x4_i64(i32 %slice, <vscale x 2 x i64> %zn0, <vscale x 2 x i64> %zn1,
|
|
; CHECK-LABEL: multi_vector_sub_write_za_vg1x4_i64:
|
|
; CHECK: // %bb.0:
|
|
; CHECK-NEXT: mov w8, w0
|
|
; CHECK-NEXT: sub za.d[w8, 0, vgx4], { z0.d - z3.d }, { z4.d - z7.d }
|
|
; CHECK-NEXT: sub za.d[w8, 7, vgx4], { z0.d - z3.d }, { z4.d - z7.d }
|
|
; CHECK-NEXT: ret
|
|
<vscale x 2 x i64> %zn2, <vscale x 2 x i64> %zn3,
|
|
<vscale x 2 x i64> %zm0, <vscale x 2 x i64> %zm1,
|
|
<vscale x 2 x i64> %zm2, <vscale x 2 x i64> %zm3) {
|
|
call void @llvm.aarch64.sme.sub.write.za.vg1x4.nxv2i64(i32 %slice,
|
|
<vscale x 2 x i64> %zn0, <vscale x 2 x i64> %zn1,
|
|
<vscale x 2 x i64> %zn2, <vscale x 2 x i64> %zn3,
|
|
<vscale x 2 x i64> %zm0, <vscale x 2 x i64> %zm1,
|
|
<vscale x 2 x i64> %zm2, <vscale x 2 x i64> %zm3)
|
|
%slice.7 = add i32 %slice, 7
|
|
call void @llvm.aarch64.sme.sub.write.za.vg1x4.nxv2i64(i32 %slice.7,
|
|
<vscale x 2 x i64> %zn0, <vscale x 2 x i64> %zn1,
|
|
<vscale x 2 x i64> %zn2, <vscale x 2 x i64> %zn3,
|
|
<vscale x 2 x i64> %zm0, <vscale x 2 x i64> %zm1,
|
|
<vscale x 2 x i64> %zm2, <vscale x 2 x i64> %zm3)
|
|
ret void
|
|
}
|
|
|
|
|
|
;
|
|
; SUB and accumulate into ZA
|
|
;
|
|
|
|
; x2
|
|
define void @multi_vector_sub_za_vg1x2_i32(i32 %slice, <vscale x 4 x i32> %zn0, <vscale x 4 x i32> %zn1) {
|
|
; CHECK-LABEL: multi_vector_sub_za_vg1x2_i32:
|
|
; CHECK: // %bb.0:
|
|
; CHECK-NEXT: mov w8, w0
|
|
; CHECK-NEXT: sub za.s[w8, 0, vgx2], { z0.s, z1.s }
|
|
; CHECK-NEXT: sub za.s[w8, 7, vgx2], { z0.s, z1.s }
|
|
; CHECK-NEXT: ret
|
|
call void @llvm.aarch64.sme.sub.za32.vg1x2.nxv4i32(i32 %slice,<vscale x 4 x i32> %zn0, <vscale x 4 x i32> %zn1)
|
|
%slice.7 = add i32 %slice, 7
|
|
call void @llvm.aarch64.sme.sub.za32.vg1x2.nxv4i32(i32 %slice.7, <vscale x 4 x i32> %zn0, <vscale x 4 x i32> %zn1)
|
|
ret void
|
|
}
|
|
|
|
define void @multi_vector_sub_za_vg1x2_i64(i32 %slice, <vscale x 2 x i64> %zn0, <vscale x 2 x i64> %zn1) {
|
|
; CHECK-LABEL: multi_vector_sub_za_vg1x2_i64:
|
|
; CHECK: // %bb.0:
|
|
; CHECK-NEXT: mov w8, w0
|
|
; CHECK-NEXT: sub za.d[w8, 0, vgx2], { z0.d, z1.d }
|
|
; CHECK-NEXT: sub za.d[w8, 7, vgx2], { z0.d, z1.d }
|
|
; CHECK-NEXT: ret
|
|
call void @llvm.aarch64.sme.sub.za64.vg1x2.nxv2i64(i32 %slice, <vscale x 2 x i64> %zn0, <vscale x 2 x i64> %zn1)
|
|
%slice.7 = add i32 %slice, 7
|
|
call void @llvm.aarch64.sme.sub.za64.vg1x2.nxv2i64(i32 %slice.7, <vscale x 2 x i64> %zn0, <vscale x 2 x i64> %zn1)
|
|
ret void
|
|
}
|
|
|
|
define void @multi_vector_sub_za_vg1x2_f32(i32 %slice, <vscale x 4 x float> %zn0, <vscale x 4 x float> %zn1) {
|
|
; CHECK-LABEL: multi_vector_sub_za_vg1x2_f32:
|
|
; CHECK: // %bb.0:
|
|
; CHECK-NEXT: mov w8, w0
|
|
; CHECK-NEXT: fsub za.s[w8, 0, vgx2], { z0.s, z1.s }
|
|
; CHECK-NEXT: fsub za.s[w8, 7, vgx2], { z0.s, z1.s }
|
|
; CHECK-NEXT: ret
|
|
call void @llvm.aarch64.sme.sub.za32.vg1x2.nxv4f32(i32 %slice,
|
|
<vscale x 4 x float> %zn0, <vscale x 4 x float> %zn1)
|
|
%slice.7 = add i32 %slice, 7
|
|
call void @llvm.aarch64.sme.sub.za32.vg1x2.nxv4f32(i32 %slice.7,
|
|
<vscale x 4 x float> %zn0, <vscale x 4 x float> %zn1)
|
|
ret void
|
|
}
|
|
|
|
define void @multi_vector_sub_za_vg1x2_f64(i32 %slice, <vscale x 2 x double> %zn0, <vscale x 2 x double> %zn1) {
|
|
; CHECK-LABEL: multi_vector_sub_za_vg1x2_f64:
|
|
; CHECK: // %bb.0:
|
|
; CHECK-NEXT: mov w8, w0
|
|
; CHECK-NEXT: fsub za.d[w8, 0, vgx2], { z0.d, z1.d }
|
|
; CHECK-NEXT: fsub za.d[w8, 7, vgx2], { z0.d, z1.d }
|
|
; CHECK-NEXT: ret
|
|
call void @llvm.aarch64.sme.sub.za64.vg1x2.nxv2f64(i32 %slice,
|
|
<vscale x 2 x double> %zn0, <vscale x 2 x double> %zn1)
|
|
%slice.7 = add i32 %slice, 7
|
|
call void @llvm.aarch64.sme.sub.za64.vg1x2.nxv2f64(i32 %slice.7,
|
|
<vscale x 2 x double> %zn0, <vscale x 2 x double> %zn1)
|
|
ret void
|
|
}
|
|
|
|
; x4
|
|
|
|
define void @multi_vector_sub_za_vg1x4_i32(i32 %slice,
|
|
; CHECK-LABEL: multi_vector_sub_za_vg1x4_i32:
|
|
; CHECK: // %bb.0:
|
|
; CHECK-NEXT: mov w8, w0
|
|
; CHECK-NEXT: sub za.s[w8, 0, vgx4], { z0.s - z3.s }
|
|
; CHECK-NEXT: sub za.s[w8, 7, vgx4], { z0.s - z3.s }
|
|
; CHECK-NEXT: ret
|
|
<vscale x 4 x i32> %zn0, <vscale x 4 x i32> %zn1,
|
|
<vscale x 4 x i32> %zn2, <vscale x 4 x i32> %zn3) {
|
|
call void @llvm.aarch64.sme.sub.za32.vg1x4.nxv4i32(i32 %slice,
|
|
<vscale x 4 x i32> %zn0, <vscale x 4 x i32> %zn1,
|
|
<vscale x 4 x i32> %zn2, <vscale x 4 x i32> %zn3)
|
|
%slice.7 = add i32 %slice, 7
|
|
call void @llvm.aarch64.sme.sub.za32.vg1x4.nxv4i32(i32 %slice.7,
|
|
<vscale x 4 x i32> %zn0, <vscale x 4 x i32> %zn1,
|
|
<vscale x 4 x i32> %zn2, <vscale x 4 x i32> %zn3)
|
|
ret void
|
|
}
|
|
|
|
define void @multi_vector_sub_za_vg1x4_i64(i32 %slice,
|
|
; CHECK-LABEL: multi_vector_sub_za_vg1x4_i64:
|
|
; CHECK: // %bb.0:
|
|
; CHECK-NEXT: mov w8, w0
|
|
; CHECK-NEXT: sub za.d[w8, 0, vgx4], { z0.d - z3.d }
|
|
; CHECK-NEXT: sub za.d[w8, 7, vgx4], { z0.d - z3.d }
|
|
; CHECK-NEXT: ret
|
|
<vscale x 2 x i64> %zn0, <vscale x 2 x i64> %zn1,
|
|
<vscale x 2 x i64> %zn2, <vscale x 2 x i64> %zn3) {
|
|
call void @llvm.aarch64.sme.sub.za64.vg1x4.nxv2i64(i32 %slice,
|
|
<vscale x 2 x i64> %zn0, <vscale x 2 x i64> %zn1,
|
|
<vscale x 2 x i64> %zn2, <vscale x 2 x i64> %zn3)
|
|
%slice.7 = add i32 %slice, 7
|
|
call void @llvm.aarch64.sme.sub.za64.vg1x4.nxv2i64(i32 %slice.7,
|
|
<vscale x 2 x i64> %zn0, <vscale x 2 x i64> %zn1,
|
|
<vscale x 2 x i64> %zn2, <vscale x 2 x i64> %zn3)
|
|
ret void
|
|
}
|
|
|
|
define void @multi_vector_sub_za_vg1x4_f32(i32 %slice,
|
|
; CHECK-LABEL: multi_vector_sub_za_vg1x4_f32:
|
|
; CHECK: // %bb.0:
|
|
; CHECK-NEXT: mov w8, w0
|
|
; CHECK-NEXT: fsub za.s[w8, 0, vgx4], { z0.s - z3.s }
|
|
; CHECK-NEXT: fsub za.s[w8, 7, vgx4], { z0.s - z3.s }
|
|
; CHECK-NEXT: ret
|
|
<vscale x 4 x float> %zn0, <vscale x 4 x float> %zn1,
|
|
<vscale x 4 x float> %zn2, <vscale x 4 x float> %zn3) {
|
|
call void @llvm.aarch64.sme.sub.za32.vg1x4.nxv4f32(i32 %slice,
|
|
<vscale x 4 x float> %zn0, <vscale x 4 x float> %zn1,
|
|
<vscale x 4 x float> %zn2, <vscale x 4 x float> %zn3)
|
|
%slice.7 = add i32 %slice, 7
|
|
call void @llvm.aarch64.sme.sub.za32.vg1x4.nxv4f32(i32 %slice.7,
|
|
<vscale x 4 x float> %zn0, <vscale x 4 x float> %zn1,
|
|
<vscale x 4 x float> %zn2, <vscale x 4 x float> %zn3)
|
|
ret void
|
|
}
|
|
|
|
define void @multi_vector_sub_za_vg1x4_f64(i32 %slice,
|
|
; CHECK-LABEL: multi_vector_sub_za_vg1x4_f64:
|
|
; CHECK: // %bb.0:
|
|
; CHECK-NEXT: mov w8, w0
|
|
; CHECK-NEXT: fsub za.d[w8, 0, vgx4], { z0.d - z3.d }
|
|
; CHECK-NEXT: fsub za.d[w8, 7, vgx4], { z0.d - z3.d }
|
|
; CHECK-NEXT: ret
|
|
<vscale x 2 x double> %zn0, <vscale x 2 x double> %zn1,
|
|
<vscale x 2 x double> %zn2, <vscale x 2 x double> %zn3) {
|
|
call void @llvm.aarch64.sme.sub.za64.vg1x4.nxv2f64(i32 %slice,
|
|
<vscale x 2 x double> %zn0, <vscale x 2 x double> %zn1,
|
|
<vscale x 2 x double> %zn2, <vscale x 2 x double> %zn3)
|
|
%slice.7 = add i32 %slice, 7
|
|
call void @llvm.aarch64.sme.sub.za64.vg1x4.nxv2f64(i32 %slice.7,
|
|
<vscale x 2 x double> %zn0, <vscale x 2 x double> %zn1,
|
|
<vscale x 2 x double> %zn2, <vscale x 2 x double> %zn3)
|
|
ret void
|
|
}
|
|
|
|
declare void@llvm.aarch64.sme.sub.write.single.za.vg1x2.nxv4i32(i32, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>)
|
|
declare void@llvm.aarch64.sme.sub.write.single.za.vg1x2.nxv2i64(i32, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>)
|
|
declare void@llvm.aarch64.sme.sub.write.single.za.vg1x4.nxv4i32(i32, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>,
|
|
<vscale x 4 x i32>, <vscale x 4 x i32>)
|
|
declare void@llvm.aarch64.sme.sub.write.single.za.vg1x4.nxv2i64(i32, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>,
|
|
<vscale x 2 x i64>, <vscale x 2 x i64>)
|
|
declare void@llvm.aarch64.sme.sub.write.za.vg1x2.nxv4i32(i32, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>)
|
|
declare void@llvm.aarch64.sme.sub.write.za.vg1x2.nxv2i64(i32, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>)
|
|
declare void@llvm.aarch64.sme.sub.write.za.vg1x4.nxv4i32(i32, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>)
|
|
declare void@llvm.aarch64.sme.sub.write.za.vg1x4.nxv2i64(i32, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>)
|
|
declare void@llvm.aarch64.sme.sub.za32.vg1x2.nxv4i32(i32, <vscale x 4 x i32>,<vscale x 4 x i32>)
|
|
declare void@llvm.aarch64.sme.sub.za64.vg1x2.nxv2i64(i32, <vscale x 2 x i64>,<vscale x 2 x i64>)
|
|
declare void@llvm.aarch64.sme.sub.za32.vg1x4.nxv4i32(i32, <vscale x 4 x i32>,<vscale x 4 x i32>,<vscale x 4 x i32>,<vscale x 4 x i32>)
|
|
declare void@llvm.aarch64.sme.sub.za64.vg1x4.nxv2i64(i32, <vscale x 2 x i64>,<vscale x 2 x i64>,<vscale x 2 x i64>, <vscale x 2 x i64>)
|
|
declare void@llvm.aarch64.sme.sub.za32.vg1x2.nxv4f32(i32, <vscale x 4 x float>, <vscale x 4 x float>)
|
|
declare void@llvm.aarch64.sme.sub.za64.vg1x2.nxv2f64(i32, <vscale x 2 x double>, <vscale x 2 x double>)
|
|
declare void@llvm.aarch64.sme.sub.za32.vg1x4.nxv4f32(i32, <vscale x 4 x float>, <vscale x 4 x float>,<vscale x 4 x float>, <vscale x 4 x float>)
|
|
declare void@llvm.aarch64.sme.sub.za64.vg1x4.nxv2f64(i32, <vscale x 2 x double>, <vscale x 2 x double>,<vscale x 2 x double>, <vscale x 2 x double>)
|