shylie/llvm-project
1
0
Fork 0
Code Issues Pull Requests Actions 6 Packages Projects Releases Wiki Activity
llvm-project/llvm/test/Transforms/LoopVectorize/operand-bundles.ll
Florian Hahn d316fb0797
[VPlan] Replicate VPScalarIVStepsRecipe by VF outside replicate regions. (#170053) 
Extend replicateByVF to also handle VPScalarIVStepsRecipe. To do so, the
patch adds a new lane operand to VPScalarIVStepsRecipe, which is only
added when replicating. This enables removing a number of lane 0
computations. The lane operand will also be used to explicitly replicate
replicate regions in a follow-up.

Depends on https://github.com/llvm/llvm-project/pull/169796
Depends on https://github.com/llvm/llvm-project/pull/170906

PR: https://github.com/llvm/llvm-project/pull/170053
2026-03-05 12:42:20 +00:00

227 lines
11 KiB
LLVM
Raw Permalink Blame History

; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --check-globals none --version 6
; RUN: opt -p loop-vectorize -force-vector-width=4 -S %s | FileCheck %s
define void @call_loop_invariant_operand_bundle(ptr %dst, {float, float} %sv) {
; CHECK-LABEL: define void @call_loop_invariant_operand_bundle(
; CHECK-SAME: ptr [[DST:%.*]], { float, float } [[SV:%.*]]) {
; CHECK-NEXT: [[ENTRY:.*:]]
; CHECK-NEXT: br label %[[VECTOR_PH:.*]]
; CHECK: [[VECTOR_PH]]:
; CHECK-NEXT: [[TMP0:%.*]] = extractvalue { float, float } [[SV]], 0
; CHECK-NEXT: [[BROADCAST_SPLATINSERT:%.*]] = insertelement <4 x float> poison, float [[TMP0]], i64 0
; CHECK-NEXT: [[BROADCAST_SPLAT:%.*]] = shufflevector <4 x float> [[BROADCAST_SPLATINSERT]], <4 x float> poison, <4 x i32> zeroinitializer
; CHECK-NEXT: [[TMP1:%.*]] = extractvalue { float, float } [[SV]], 1
; CHECK-NEXT: [[BROADCAST_SPLATINSERT1:%.*]] = insertelement <4 x float> poison, float [[TMP1]], i64 0
; CHECK-NEXT: [[BROADCAST_SPLAT2:%.*]] = shufflevector <4 x float> [[BROADCAST_SPLATINSERT1]], <4 x float> poison, <4 x i32> zeroinitializer
; CHECK-NEXT: br label %[[VECTOR_BODY:.*]]
; CHECK: [[VECTOR_BODY]]:
; CHECK-NEXT: [[INDEX:%.*]] = phi i32 [ 0, %[[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], %[[VECTOR_BODY]] ]
; CHECK-NEXT: [[TMP2:%.*]] = getelementptr float, ptr [[DST]], i32 [[INDEX]]
; CHECK-NEXT: [[TMP3:%.*]] = call <4 x float> @llvm.pow.v4f32(<4 x float> [[BROADCAST_SPLAT]], <4 x float> [[BROADCAST_SPLAT2]]) [ "deopt"(float 1.000000e+01) ]
; CHECK-NEXT: store <4 x float> [[TMP3]], ptr [[TMP2]], align 4
; CHECK-NEXT: [[INDEX_NEXT]] = add nuw i32 [[INDEX]], 4
; CHECK-NEXT: [[TMP4:%.*]] = icmp eq i32 [[INDEX_NEXT]], 1000
; CHECK-NEXT: br i1 [[TMP4]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
; CHECK: [[MIDDLE_BLOCK]]:
; CHECK-NEXT: br label %[[EXIT:.*]]
; CHECK: [[EXIT]]:
; CHECK-NEXT: ret void
;
entry:
br label %loop
loop:
%iv = phi i32 [ 0, %entry ], [ %iv.next, %loop ]
%a = extractvalue { float, float } %sv, 0
%b = extractvalue { float, float } %sv, 1
%addr = getelementptr float, ptr %dst, i32 %iv
%p = call float @llvm.pow.f32(float %a, float %b) [ "deopt"(float 10.0) ]
store float %p, ptr %addr
%iv.next = add nsw i32 %iv, 1
%cond = icmp ne i32 %iv.next, 1000
br i1 %cond, label %loop, label %exit
exit:
ret void
}
define void @call_unknown_operand_bundle(ptr %dst, {float, float} %sv) {
; CHECK-LABEL: define void @call_unknown_operand_bundle(
; CHECK-SAME: ptr [[DST:%.*]], { float, float } [[SV:%.*]]) {
; CHECK-NEXT: [[ENTRY:.*:]]
; CHECK-NEXT: br label %[[VECTOR_PH:.*]]
; CHECK: [[VECTOR_PH]]:
; CHECK-NEXT: [[TMP0:%.*]] = extractvalue { float, float } [[SV]], 0
; CHECK-NEXT: [[BROADCAST_SPLATINSERT:%.*]] = insertelement <4 x float> poison, float [[TMP0]], i64 0
; CHECK-NEXT: [[BROADCAST_SPLAT:%.*]] = shufflevector <4 x float> [[BROADCAST_SPLATINSERT]], <4 x float> poison, <4 x i32> zeroinitializer
; CHECK-NEXT: [[TMP1:%.*]] = extractvalue { float, float } [[SV]], 1
; CHECK-NEXT: [[BROADCAST_SPLATINSERT1:%.*]] = insertelement <4 x float> poison, float [[TMP1]], i64 0
; CHECK-NEXT: [[BROADCAST_SPLAT2:%.*]] = shufflevector <4 x float> [[BROADCAST_SPLATINSERT1]], <4 x float> poison, <4 x i32> zeroinitializer
; CHECK-NEXT: br label %[[VECTOR_BODY:.*]]
; CHECK: [[VECTOR_BODY]]:
; CHECK-NEXT: [[INDEX:%.*]] = phi i32 [ 0, %[[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], %[[VECTOR_BODY]] ]
; CHECK-NEXT: [[TMP2:%.*]] = getelementptr float, ptr [[DST]], i32 [[INDEX]]
; CHECK-NEXT: [[TMP3:%.*]] = call <4 x float> @llvm.pow.v4f32(<4 x float> [[BROADCAST_SPLAT]], <4 x float> [[BROADCAST_SPLAT2]]) [ "unknown"(ptr null) ]
; CHECK-NEXT: store <4 x float> [[TMP3]], ptr [[TMP2]], align 4
; CHECK-NEXT: [[INDEX_NEXT]] = add nuw i32 [[INDEX]], 4
; CHECK-NEXT: [[TMP4:%.*]] = icmp eq i32 [[INDEX_NEXT]], 1000
; CHECK-NEXT: br i1 [[TMP4]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP3:![0-9]+]]
; CHECK: [[MIDDLE_BLOCK]]:
; CHECK-NEXT: br label %[[EXIT:.*]]
; CHECK: [[EXIT]]:
; CHECK-NEXT: ret void
;
entry:
br label %loop
loop:
%iv = phi i32 [ 0, %entry ], [ %iv.next, %loop ]
%a = extractvalue { float, float } %sv, 0
%b = extractvalue { float, float } %sv, 1
%addr = getelementptr float, ptr %dst, i32 %iv
%p = call float @llvm.pow.f32(float %a, float %b) [ "unknown"(ptr null) ]
store float %p, ptr %addr
%iv.next = add nsw i32 %iv, 1
%cond = icmp ne i32 %iv.next, 1000
br i1 %cond, label %loop, label %exit
exit:
ret void
}
define void @call_cold_operand_bundle(ptr %dst, {float, float} %sv) {
; CHECK-LABEL: define void @call_cold_operand_bundle(
; CHECK-SAME: ptr [[DST:%.*]], { float, float } [[SV:%.*]]) {
; CHECK-NEXT: [[ENTRY:.*:]]
; CHECK-NEXT: br label %[[VECTOR_PH:.*]]
; CHECK: [[VECTOR_PH]]:
; CHECK-NEXT: [[TMP0:%.*]] = extractvalue { float, float } [[SV]], 0
; CHECK-NEXT: [[BROADCAST_SPLATINSERT:%.*]] = insertelement <4 x float> poison, float [[TMP0]], i64 0
; CHECK-NEXT: [[BROADCAST_SPLAT:%.*]] = shufflevector <4 x float> [[BROADCAST_SPLATINSERT]], <4 x float> poison, <4 x i32> zeroinitializer
; CHECK-NEXT: [[TMP1:%.*]] = extractvalue { float, float } [[SV]], 1
; CHECK-NEXT: [[BROADCAST_SPLATINSERT1:%.*]] = insertelement <4 x float> poison, float [[TMP1]], i64 0
; CHECK-NEXT: [[BROADCAST_SPLAT2:%.*]] = shufflevector <4 x float> [[BROADCAST_SPLATINSERT1]], <4 x float> poison, <4 x i32> zeroinitializer
; CHECK-NEXT: br label %[[VECTOR_BODY:.*]]
; CHECK: [[VECTOR_BODY]]:
; CHECK-NEXT: [[INDEX:%.*]] = phi i32 [ 0, %[[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], %[[VECTOR_BODY]] ]
; CHECK-NEXT: [[TMP2:%.*]] = getelementptr float, ptr [[DST]], i32 [[INDEX]]
; CHECK-NEXT: [[TMP3:%.*]] = call <4 x float> @llvm.pow.v4f32(<4 x float> [[BROADCAST_SPLAT]], <4 x float> [[BROADCAST_SPLAT2]]) [ "cold"() ]
; CHECK-NEXT: store <4 x float> [[TMP3]], ptr [[TMP2]], align 4
; CHECK-NEXT: [[INDEX_NEXT]] = add nuw i32 [[INDEX]], 4
; CHECK-NEXT: [[TMP4:%.*]] = icmp eq i32 [[INDEX_NEXT]], 1000
; CHECK-NEXT: br i1 [[TMP4]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP4:![0-9]+]]
; CHECK: [[MIDDLE_BLOCK]]:
; CHECK-NEXT: br label %[[EXIT:.*]]
; CHECK: [[EXIT]]:
; CHECK-NEXT: ret void
;
entry:
br label %loop
loop:
%iv = phi i32 [ 0, %entry ], [ %iv.next, %loop ]
%a = extractvalue { float, float } %sv, 0
%b = extractvalue { float, float } %sv, 1
%addr = getelementptr float, ptr %dst, i32 %iv
%p = call float @llvm.pow.f32(float %a, float %b) [ "cold"() ]
store float %p, ptr %addr
%iv.next = add nsw i32 %iv, 1
%cond = icmp ne i32 %iv.next, 1000
br i1 %cond, label %loop, label %exit
exit:
ret void
}
define void @assume_loop_variant_operand_bundle(ptr noalias %a, ptr noalias %b) {
; CHECK-LABEL: define void @assume_loop_variant_operand_bundle(
; CHECK-SAME: ptr noalias [[A:%.*]], ptr noalias [[B:%.*]]) {
; CHECK-NEXT: [[ENTRY:.*:]]
; CHECK-NEXT: br label %[[VECTOR_PH:.*]]
; CHECK: [[VECTOR_PH]]:
; CHECK-NEXT: br label %[[VECTOR_BODY:.*]]
; CHECK: [[VECTOR_BODY]]:
; CHECK-NEXT: [[INDEX:%.*]] = phi i64 [ 0, %[[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], %[[VECTOR_BODY]] ]
; CHECK-NEXT: [[TMP1:%.*]] = add i64 [[INDEX]], 1
; CHECK-NEXT: [[TMP2:%.*]] = add i64 [[INDEX]], 2
; CHECK-NEXT: [[TMP3:%.*]] = add i64 [[INDEX]], 3
; CHECK-NEXT: [[TMP8:%.*]] = getelementptr inbounds float, ptr [[B]], i64 [[INDEX]]
; CHECK-NEXT: [[WIDE_LOAD:%.*]] = load <4 x float>, ptr [[TMP8]], align 4
; CHECK-NEXT: tail call void @llvm.assume(i1 true) [ "align"(ptr [[A]], i64 [[INDEX]]) ]
; CHECK-NEXT: tail call void @llvm.assume(i1 true) [ "align"(ptr [[A]], i64 [[TMP1]]) ]
; CHECK-NEXT: tail call void @llvm.assume(i1 true) [ "align"(ptr [[A]], i64 [[TMP2]]) ]
; CHECK-NEXT: tail call void @llvm.assume(i1 true) [ "align"(ptr [[A]], i64 [[TMP3]]) ]
; CHECK-NEXT: [[TMP5:%.*]] = fadd <4 x float> [[WIDE_LOAD]], splat (float 1.000000e+00)
; CHECK-NEXT: [[TMP10:%.*]] = getelementptr inbounds float, ptr [[A]], i64 [[INDEX]]
; CHECK-NEXT: store <4 x float> [[TMP5]], ptr [[TMP10]], align 4
; CHECK-NEXT: [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 4
; CHECK-NEXT: [[TMP11:%.*]] = icmp eq i64 [[INDEX_NEXT]], 1600
; CHECK-NEXT: br i1 [[TMP11]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP5:![0-9]+]]
; CHECK: [[MIDDLE_BLOCK]]:
; CHECK-NEXT: br label %[[EXIT:.*]]
; CHECK: [[EXIT]]:
; CHECK-NEXT: ret void
;
entry:
br label %loop
loop:
%iv = phi i64 [ 0, %entry ], [ %iv.next, %loop ]
%arrayidx = getelementptr inbounds float, ptr %b, i64 %iv
%0 = load float, ptr %arrayidx, align 4
%cmp1 = fcmp ogt float %0, 1.000000e+02
tail call void @llvm.assume(i1 true) [ "align"(ptr %a, i64 %iv) ]
%add = fadd float %0, 1.000000e+00
%arrayidx5 = getelementptr inbounds float, ptr %a, i64 %iv
store float %add, ptr %arrayidx5, align 4
%iv.next = add nuw nsw i64 %iv, 1
%exitcond = icmp eq i64 %iv, 1599
br i1 %exitcond, label %exit, label %loop
exit:
ret void
}
define void @assume_cold_operand_bundle(ptr noalias %a, ptr noalias %b) {
; CHECK-LABEL: define void @assume_cold_operand_bundle(
; CHECK-SAME: ptr noalias [[A:%.*]], ptr noalias [[B:%.*]]) {
; CHECK-NEXT: [[ENTRY:.*:]]
; CHECK-NEXT: br label %[[VECTOR_PH:.*]]
; CHECK: [[VECTOR_PH]]:
; CHECK-NEXT: tail call void @llvm.assume(i1 true) [ "cold"() ]
; CHECK-NEXT: br label %[[VECTOR_BODY:.*]]
; CHECK: [[VECTOR_BODY]]:
; CHECK-NEXT: [[INDEX:%.*]] = phi i64 [ 0, %[[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], %[[VECTOR_BODY]] ]
; CHECK-NEXT: [[TMP0:%.*]] = getelementptr inbounds float, ptr [[B]], i64 [[INDEX]]
; CHECK-NEXT: [[WIDE_LOAD:%.*]] = load <4 x float>, ptr [[TMP0]], align 4
; CHECK-NEXT: [[TMP1:%.*]] = fadd <4 x float> [[WIDE_LOAD]], splat (float 1.000000e+00)
; CHECK-NEXT: [[TMP2:%.*]] = getelementptr inbounds float, ptr [[A]], i64 [[INDEX]]
; CHECK-NEXT: store <4 x float> [[TMP1]], ptr [[TMP2]], align 4
; CHECK-NEXT: [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 4
; CHECK-NEXT: [[TMP3:%.*]] = icmp eq i64 [[INDEX_NEXT]], 1600
; CHECK-NEXT: br i1 [[TMP3]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP6:![0-9]+]]
; CHECK: [[MIDDLE_BLOCK]]:
; CHECK-NEXT: br label %[[EXIT:.*]]
; CHECK: [[EXIT]]:
; CHECK-NEXT: ret void
;
entry:
br label %loop
loop:
%iv = phi i64 [ 0, %entry ], [ %iv.next, %loop ]
%arrayidx = getelementptr inbounds float, ptr %b, i64 %iv
%0 = load float, ptr %arrayidx, align 4
%cmp1 = fcmp ogt float %0, 1.000000e+02
tail call void @llvm.assume(i1 true) [ "cold"() ]
%add = fadd float %0, 1.000000e+00
%arrayidx5 = getelementptr inbounds float, ptr %a, i64 %iv
store float %add, ptr %arrayidx5, align 4
%iv.next = add nuw nsw i64 %iv, 1
%exitcond = icmp eq i64 %iv, 1599
br i1 %exitcond, label %exit, label %loop
exit:
ret void
}
Reference in New Issue View Git Blame Copy Permalink