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
114 lines
7.7 KiB
LLVM
114 lines
7.7 KiB
LLVM
; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
|
|
; RUN: opt -S -passes='require<profile-summary>,function(codegenprepare)' < %s | FileCheck %s
|
|
|
|
target triple = "aarch64-unknown-linux-gnu"
|
|
|
|
%struct.a = type { i32, i32 }
|
|
@c = external dso_local global %struct.a, align 4
|
|
@glob_array = internal unnamed_addr constant [16 x i32] [i32 1, i32 1, i32 2, i32 3, i32 5, i32 8, i32 13, i32 21, i32 34, i32 55, i32 89, i32 144, i32 233, i32 377, i32 610, i32 987], align 16
|
|
|
|
define <vscale x 4 x i32> @splat_base(ptr %base, <vscale x 4 x i64> %index, <vscale x 4 x i1> %mask) #0 {
|
|
; CHECK-LABEL: @splat_base(
|
|
; CHECK-NEXT: [[TMP1:%.*]] = getelementptr i32, ptr [[BASE:%.*]], <vscale x 4 x i64> [[INDEX:%.*]]
|
|
; CHECK-NEXT: [[RES:%.*]] = call <vscale x 4 x i32> @llvm.masked.gather.nxv4i32.nxv4p0(<vscale x 4 x ptr> align 4 [[TMP1]], <vscale x 4 x i1> [[MASK:%.*]], <vscale x 4 x i32> undef)
|
|
; CHECK-NEXT: ret <vscale x 4 x i32> [[RES]]
|
|
;
|
|
%broadcast.splatinsert = insertelement <vscale x 4 x ptr> undef, ptr %base, i32 0
|
|
%broadcast.splat = shufflevector <vscale x 4 x ptr> %broadcast.splatinsert, <vscale x 4 x ptr> undef, <vscale x 4 x i32> zeroinitializer
|
|
%gep = getelementptr i32, <vscale x 4 x ptr> %broadcast.splat, <vscale x 4 x i64> %index
|
|
%res = call <vscale x 4 x i32> @llvm.masked.gather.nxv4i32.nxv4p0(<vscale x 4 x ptr> %gep, i32 4, <vscale x 4 x i1> %mask, <vscale x 4 x i32> undef)
|
|
ret <vscale x 4 x i32> %res
|
|
}
|
|
|
|
define <vscale x 4 x i32> @splat_struct(ptr %base, <vscale x 4 x i1> %mask) #0 {
|
|
; CHECK-LABEL: @splat_struct(
|
|
; CHECK-NEXT: [[TMP1:%.*]] = getelementptr [[STRUCT_A:%.*]], ptr [[BASE:%.*]], i64 0, i32 1
|
|
; CHECK-NEXT: [[TMP2:%.*]] = getelementptr i32, ptr [[TMP1]], <vscale x 4 x i64> zeroinitializer
|
|
; CHECK-NEXT: [[RES:%.*]] = call <vscale x 4 x i32> @llvm.masked.gather.nxv4i32.nxv4p0(<vscale x 4 x ptr> align 4 [[TMP2]], <vscale x 4 x i1> [[MASK:%.*]], <vscale x 4 x i32> undef)
|
|
; CHECK-NEXT: ret <vscale x 4 x i32> [[RES]]
|
|
;
|
|
%gep = getelementptr %struct.a, ptr %base, <vscale x 4 x i64> zeroinitializer, i32 1
|
|
%res = call <vscale x 4 x i32> @llvm.masked.gather.nxv4i32.nxv4p0(<vscale x 4 x ptr> %gep, i32 4, <vscale x 4 x i1> %mask, <vscale x 4 x i32> undef)
|
|
ret <vscale x 4 x i32> %res
|
|
}
|
|
|
|
define <vscale x 4 x i32> @scalar_index(ptr %base, i64 %index, <vscale x 4 x i1> %mask) #0 {
|
|
; CHECK-LABEL: @scalar_index(
|
|
; CHECK-NEXT: [[TMP1:%.*]] = getelementptr i32, ptr [[BASE:%.*]], i64 [[INDEX:%.*]]
|
|
; CHECK-NEXT: [[TMP2:%.*]] = getelementptr i32, ptr [[TMP1]], <vscale x 4 x i64> zeroinitializer
|
|
; CHECK-NEXT: [[RES:%.*]] = call <vscale x 4 x i32> @llvm.masked.gather.nxv4i32.nxv4p0(<vscale x 4 x ptr> align 4 [[TMP2]], <vscale x 4 x i1> [[MASK:%.*]], <vscale x 4 x i32> undef)
|
|
; CHECK-NEXT: ret <vscale x 4 x i32> [[RES]]
|
|
;
|
|
%broadcast.splatinsert = insertelement <vscale x 4 x ptr> undef, ptr %base, i32 0
|
|
%broadcast.splat = shufflevector <vscale x 4 x ptr> %broadcast.splatinsert, <vscale x 4 x ptr> undef, <vscale x 4 x i32> zeroinitializer
|
|
%gep = getelementptr i32, <vscale x 4 x ptr> %broadcast.splat, i64 %index
|
|
%res = call <vscale x 4 x i32> @llvm.masked.gather.nxv4i32.nxv4p0(<vscale x 4 x ptr> %gep, i32 4, <vscale x 4 x i1> %mask, <vscale x 4 x i32> undef)
|
|
ret <vscale x 4 x i32> %res
|
|
}
|
|
|
|
define <vscale x 4 x i32> @splat_index(ptr %base, i64 %index, <vscale x 4 x i1> %mask) #0 {
|
|
; CHECK-LABEL: @splat_index(
|
|
; CHECK-NEXT: [[TMP1:%.*]] = getelementptr i32, ptr [[BASE:%.*]], i64 [[INDEX:%.*]]
|
|
; CHECK-NEXT: [[TMP2:%.*]] = getelementptr i32, ptr [[TMP1]], <vscale x 4 x i64> zeroinitializer
|
|
; CHECK-NEXT: [[RES:%.*]] = call <vscale x 4 x i32> @llvm.masked.gather.nxv4i32.nxv4p0(<vscale x 4 x ptr> align 4 [[TMP2]], <vscale x 4 x i1> [[MASK:%.*]], <vscale x 4 x i32> undef)
|
|
; CHECK-NEXT: ret <vscale x 4 x i32> [[RES]]
|
|
;
|
|
%broadcast.splatinsert = insertelement <vscale x 4 x i64> undef, i64 %index, i32 0
|
|
%broadcast.splat = shufflevector <vscale x 4 x i64> %broadcast.splatinsert, <vscale x 4 x i64> undef, <vscale x 4 x i32> zeroinitializer
|
|
%gep = getelementptr i32, ptr %base, <vscale x 4 x i64> %broadcast.splat
|
|
%res = call <vscale x 4 x i32> @llvm.masked.gather.nxv4i32.nxv4p0(<vscale x 4 x ptr> %gep, i32 4, <vscale x 4 x i1> %mask, <vscale x 4 x i32> undef)
|
|
ret <vscale x 4 x i32> %res
|
|
}
|
|
|
|
define <vscale x 4 x i32> @test_global_array(<vscale x 4 x i64> %indxs, <vscale x 4 x i1> %mask) #0 {
|
|
; CHECK-LABEL: @test_global_array(
|
|
; CHECK-NEXT: [[TMP1:%.*]] = getelementptr i32, ptr @glob_array, <vscale x 4 x i64> [[INDXS:%.*]]
|
|
; CHECK-NEXT: [[G:%.*]] = call <vscale x 4 x i32> @llvm.masked.gather.nxv4i32.nxv4p0(<vscale x 4 x ptr> align 4 [[TMP1]], <vscale x 4 x i1> [[MASK:%.*]], <vscale x 4 x i32> undef)
|
|
; CHECK-NEXT: ret <vscale x 4 x i32> [[G]]
|
|
;
|
|
%p = getelementptr inbounds [16 x i32], ptr @glob_array, i64 0, <vscale x 4 x i64> %indxs
|
|
%g = call <vscale x 4 x i32> @llvm.masked.gather.nxv4i32.nxv4p0(<vscale x 4 x ptr> %p, i32 4, <vscale x 4 x i1> %mask, <vscale x 4 x i32> undef)
|
|
ret <vscale x 4 x i32> %g
|
|
}
|
|
|
|
define <vscale x 4 x i32> @global_struct_splat(<vscale x 4 x i1> %mask) #0 {
|
|
; CHECK-LABEL: @global_struct_splat(
|
|
; CHECK-NEXT: [[TMP1:%.*]] = call <vscale x 4 x i32> @llvm.masked.gather.nxv4i32.nxv4p0(<vscale x 4 x ptr> align 4 shufflevector (<vscale x 4 x ptr> insertelement (<vscale x 4 x ptr> poison, ptr getelementptr ([[STRUCT_A:%.*]], ptr @c, i64 0, i32 1), i64 0), <vscale x 4 x ptr> poison, <vscale x 4 x i32> zeroinitializer), <vscale x 4 x i1> [[MASK:%.*]], <vscale x 4 x i32> undef)
|
|
; CHECK-NEXT: ret <vscale x 4 x i32> [[TMP1]]
|
|
;
|
|
%1 = insertelement <vscale x 4 x ptr> undef, ptr @c, i32 0
|
|
%2 = shufflevector <vscale x 4 x ptr> %1, <vscale x 4 x ptr> undef, <vscale x 4 x i32> zeroinitializer
|
|
%3 = getelementptr %struct.a, <vscale x 4 x ptr> %2, <vscale x 4 x i64> zeroinitializer, i32 1
|
|
%4 = call <vscale x 4 x i32> @llvm.masked.gather.nxv4i32.nxv4p0(<vscale x 4 x ptr> %3, i32 4, <vscale x 4 x i1> %mask, <vscale x 4 x i32> undef)
|
|
ret <vscale x 4 x i32> %4
|
|
}
|
|
|
|
define <vscale x 4 x i32> @splat_ptr_gather(ptr %ptr, <vscale x 4 x i1> %mask, <vscale x 4 x i32> %passthru) #0 {
|
|
; CHECK-LABEL: @splat_ptr_gather(
|
|
; CHECK-NEXT: [[TMP1:%.*]] = getelementptr i32, ptr [[PTR:%.*]], <vscale x 4 x i64> zeroinitializer
|
|
; CHECK-NEXT: [[TMP2:%.*]] = call <vscale x 4 x i32> @llvm.masked.gather.nxv4i32.nxv4p0(<vscale x 4 x ptr> align 4 [[TMP1]], <vscale x 4 x i1> [[MASK:%.*]], <vscale x 4 x i32> [[PASSTHRU:%.*]])
|
|
; CHECK-NEXT: ret <vscale x 4 x i32> [[TMP2]]
|
|
;
|
|
%1 = insertelement <vscale x 4 x ptr> undef, ptr %ptr, i32 0
|
|
%2 = shufflevector <vscale x 4 x ptr> %1, <vscale x 4 x ptr> undef, <vscale x 4 x i32> zeroinitializer
|
|
%3 = call <vscale x 4 x i32> @llvm.masked.gather.nxv4i32.nxv4p0(<vscale x 4 x ptr> %2, i32 4, <vscale x 4 x i1> %mask, <vscale x 4 x i32> %passthru)
|
|
ret <vscale x 4 x i32> %3
|
|
}
|
|
|
|
define void @splat_ptr_scatter(ptr %ptr, <vscale x 4 x i1> %mask, <vscale x 4 x i32> %val) #0 {
|
|
; CHECK-LABEL: @splat_ptr_scatter(
|
|
; CHECK-NEXT: [[TMP1:%.*]] = getelementptr i32, ptr [[PTR:%.*]], <vscale x 4 x i64> zeroinitializer
|
|
; CHECK-NEXT: call void @llvm.masked.scatter.nxv4i32.nxv4p0(<vscale x 4 x i32> [[VAL:%.*]], <vscale x 4 x ptr> align 4 [[TMP1]], <vscale x 4 x i1> [[MASK:%.*]])
|
|
; CHECK-NEXT: ret void
|
|
;
|
|
%1 = insertelement <vscale x 4 x ptr> undef, ptr %ptr, i32 0
|
|
%2 = shufflevector <vscale x 4 x ptr> %1, <vscale x 4 x ptr> undef, <vscale x 4 x i32> zeroinitializer
|
|
call void @llvm.masked.scatter.nxv4i32.nxv4p0(<vscale x 4 x i32> %val, <vscale x 4 x ptr> %2, i32 4, <vscale x 4 x i1> %mask)
|
|
ret void
|
|
}
|
|
|
|
declare <vscale x 4 x i32> @llvm.masked.gather.nxv4i32.nxv4p0(<vscale x 4 x ptr>, i32, <vscale x 4 x i1>, <vscale x 4 x i32>)
|
|
declare void @llvm.masked.scatter.nxv4i32.nxv4p0(<vscale x 4 x i32>, <vscale x 4 x ptr>, i32, <vscale x 4 x i1>)
|
|
|
|
attributes #0 = { "target-features"="+sve" }
|