[llvm] 1720787 - [Attributor][NFX] Pre-commit test case exposing a problem
Johannes Doerfert via llvm-commits
llvm-commits at lists.llvm.org
Wed Oct 27 19:04:51 PDT 2021
Author: Johannes Doerfert
Date: 2021-10-27T21:04:31-05:00
New Revision: 172078729fced990126cb9c0dc4fe19650069703
URL: https://github.com/llvm/llvm-project/commit/172078729fced990126cb9c0dc4fe19650069703
DIFF: https://github.com/llvm/llvm-project/commit/172078729fced990126cb9c0dc4fe19650069703.diff
LOG: [Attributor][NFX] Pre-commit test case exposing a problem
The test case is the IR of:
```
void func(float * restrict a, float *b, int N) {
N = 199;
#pragma omp parallel for
for (int i = 1; i < N; i++)
a[i] = b[i] + 1.0;
}
```
Added:
llvm/test/Transforms/Attributor/openmp_parallel.ll
Modified:
Removed:
################################################################################
diff --git a/llvm/test/Transforms/Attributor/openmp_parallel.ll b/llvm/test/Transforms/Attributor/openmp_parallel.ll
new file mode 100644
index 0000000000000..6d48cbea4e381
--- /dev/null
+++ b/llvm/test/Transforms/Attributor/openmp_parallel.ll
@@ -0,0 +1,304 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --function-signature --check-attributes --check-globals
+; RUN: opt -attributor -enable-new-pm=0 -attributor-manifest-internal -attributor-max-iterations-verify -attributor-annotate-decl-cs -attributor-max-iterations=12 -S < %s | FileCheck %s --check-prefixes=CHECK,NOT_CGSCC_NPM,NOT_CGSCC_OPM,NOT_TUNIT_NPM,IS__TUNIT____,IS________OPM,IS__TUNIT_OPM
+; RUN: opt -aa-pipeline=basic-aa -passes=attributor -attributor-manifest-internal -attributor-max-iterations-verify -attributor-annotate-decl-cs -attributor-max-iterations=12 -S < %s | FileCheck %s --check-prefixes=CHECK,NOT_CGSCC_OPM,NOT_CGSCC_NPM,NOT_TUNIT_OPM,IS__TUNIT____,IS________NPM,IS__TUNIT_NPM
+; RUN: opt -attributor-cgscc -enable-new-pm=0 -attributor-manifest-internal -attributor-annotate-decl-cs -S < %s | FileCheck %s --check-prefixes=CHECK,NOT_TUNIT_NPM,NOT_TUNIT_OPM,NOT_CGSCC_NPM,IS__CGSCC____,IS________OPM,IS__CGSCC_OPM
+; RUN: opt -aa-pipeline=basic-aa -passes=attributor-cgscc -attributor-manifest-internal -attributor-annotate-decl-cs -S < %s | FileCheck %s --check-prefixes=CHECK,NOT_TUNIT_NPM,NOT_TUNIT_OPM,NOT_CGSCC_OPM,IS__CGSCC____,IS________NPM,IS__CGSCC_NPM
+
+target datalayout = "e-m:e-p270:32:32-p271:32:32-p272:64:64-i64:64-f80:128-n8:16:32:64-S128"
+
+%struct.ident_t = type { i32, i32, i32, i32, i8* }
+
+ at 0 = private unnamed_addr constant [23 x i8] c";unknown;unknown;0;0;;\00", align 1
+ at 1 = private unnamed_addr constant %struct.ident_t { i32 0, i32 514, i32 0, i32 0, i8* getelementptr inbounds ([23 x i8], [23 x i8]* @0, i32 0, i32 0) }, align 8
+ at 2 = private unnamed_addr constant %struct.ident_t { i32 0, i32 2, i32 0, i32 0, i8* getelementptr inbounds ([23 x i8], [23 x i8]* @0, i32 0, i32 0) }, align 8
+
+; %a is write only, %b is read only, neither is captured or freed, or ...
+; FIXME: %a and %b are *not* readnone!
+
+;.
+; CHECK: @[[GLOB0:[0-9]+]] = private unnamed_addr constant [23 x i8] c"
+; CHECK: @[[GLOB1:[0-9]+]] = private unnamed_addr constant [[STRUCT_IDENT_T:%.*]] { i32 0, i32 514, i32 0, i32 0, i8* getelementptr inbounds ([23 x i8], [23 x i8]* @[[GLOB0]], i32 0, i32 0) }, align 8
+; CHECK: @[[GLOB2:[0-9]+]] = private unnamed_addr constant [[STRUCT_IDENT_T:%.*]] { i32 0, i32 2, i32 0, i32 0, i8* getelementptr inbounds ([23 x i8], [23 x i8]* @[[GLOB0]], i32 0, i32 0) }, align 8
+;.
+define dso_local void @func(float* nocapture %a, float* %b, i32 %N) local_unnamed_addr #0 {
+; IS__TUNIT_OPM: Function Attrs: nounwind uwtable
+; IS__TUNIT_OPM-LABEL: define {{[^@]+}}@func
+; IS__TUNIT_OPM-SAME: (float* nocapture nofree writeonly [[A:%.*]], float* nocapture nofree readonly [[B:%.*]], i32 [[N:%.*]]) local_unnamed_addr #[[ATTR0:[0-9]+]] {
+; IS__TUNIT_OPM-NEXT: entry:
+; IS__TUNIT_OPM-NEXT: [[A_ADDR:%.*]] = alloca float*, align 8
+; IS__TUNIT_OPM-NEXT: [[B_ADDR:%.*]] = alloca float*, align 8
+; IS__TUNIT_OPM-NEXT: [[N_ADDR:%.*]] = alloca i32, align 4
+; IS__TUNIT_OPM-NEXT: store float* [[A]], float** [[A_ADDR]], align 8
+; IS__TUNIT_OPM-NEXT: store float* [[B]], float** [[B_ADDR]], align 8
+; IS__TUNIT_OPM-NEXT: store i32 199, i32* [[N_ADDR]], align 4
+; IS__TUNIT_OPM-NEXT: call void (%struct.ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call(%struct.ident_t* noundef nonnull align 8 dereferenceable(24) @[[GLOB2]], i32 noundef 3, void (i32*, i32*, ...)* noundef bitcast (void (i32*, i32*, i32*, float**, float**)* @.omp_outlined. to void (i32*, i32*, ...)*), i32* nocapture nofree noundef nonnull readonly align 4 dereferenceable(4) [[N_ADDR]], float** nocapture nofree noundef nonnull readonly align 8 dereferenceable(8) [[A_ADDR]], float** nocapture nofree noundef nonnull readonly align 8 dereferenceable(8) [[B_ADDR]])
+; IS__TUNIT_OPM-NEXT: ret void
+;
+; IS__TUNIT_NPM: Function Attrs: nounwind uwtable
+; IS__TUNIT_NPM-LABEL: define {{[^@]+}}@func
+; IS__TUNIT_NPM-SAME: (float* nocapture nofree writeonly [[A:%.*]], float* nocapture nofree readonly [[B:%.*]], i32 [[N:%.*]]) local_unnamed_addr #[[ATTR0:[0-9]+]] {
+; IS__TUNIT_NPM-NEXT: entry:
+; IS__TUNIT_NPM-NEXT: [[A_ADDR:%.*]] = alloca float*, align 8
+; IS__TUNIT_NPM-NEXT: [[B_ADDR:%.*]] = alloca float*, align 8
+; IS__TUNIT_NPM-NEXT: [[N_ADDR:%.*]] = alloca i32, align 4
+; IS__TUNIT_NPM-NEXT: store float* [[A]], float** [[A_ADDR]], align 8
+; IS__TUNIT_NPM-NEXT: store float* [[B]], float** [[B_ADDR]], align 8
+; IS__TUNIT_NPM-NEXT: store i32 199, i32* [[N_ADDR]], align 4
+; IS__TUNIT_NPM-NEXT: call void (%struct.ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call(%struct.ident_t* noundef nonnull align 8 dereferenceable(24) @[[GLOB2]], i32 noundef 3, void (i32*, i32*, ...)* noundef bitcast (void (i32*, i32*, i32*, float**, float**)* @.omp_outlined. to void (i32*, i32*, ...)*), i32* noalias nocapture nofree noundef nonnull readonly align 4 dereferenceable(4) [[N_ADDR]], float** noalias nocapture nofree noundef nonnull readonly align 8 dereferenceable(8) [[A_ADDR]], float** noalias nocapture nofree noundef nonnull readonly align 8 dereferenceable(8) [[B_ADDR]])
+; IS__TUNIT_NPM-NEXT: ret void
+;
+; IS__CGSCC_OPM: Function Attrs: nounwind uwtable
+; IS__CGSCC_OPM-LABEL: define {{[^@]+}}@func
+; IS__CGSCC_OPM-SAME: (float* nocapture nofree readnone [[A:%.*]], float* nofree [[B:%.*]], i32 [[N:%.*]]) local_unnamed_addr #[[ATTR0:[0-9]+]] {
+; IS__CGSCC_OPM-NEXT: entry:
+; IS__CGSCC_OPM-NEXT: [[A_ADDR:%.*]] = alloca float*, align 8
+; IS__CGSCC_OPM-NEXT: [[B_ADDR:%.*]] = alloca float*, align 8
+; IS__CGSCC_OPM-NEXT: [[N_ADDR:%.*]] = alloca i32, align 4
+; IS__CGSCC_OPM-NEXT: store float* [[A]], float** [[A_ADDR]], align 8
+; IS__CGSCC_OPM-NEXT: store float* [[B]], float** [[B_ADDR]], align 8
+; IS__CGSCC_OPM-NEXT: store i32 199, i32* [[N_ADDR]], align 4
+; IS__CGSCC_OPM-NEXT: call void (%struct.ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call(%struct.ident_t* noundef nonnull align 8 dereferenceable(24) @[[GLOB2]], i32 noundef 3, void (i32*, i32*, ...)* noundef bitcast (void (i32*, i32*, i32*, float**, float**)* @.omp_outlined. to void (i32*, i32*, ...)*), i32* nocapture nofree noundef nonnull readonly align 4 dereferenceable(4) [[N_ADDR]], float** nocapture nofree noundef nonnull readonly align 8 dereferenceable(8) [[A_ADDR]], float** nocapture nofree noundef nonnull readonly align 8 dereferenceable(8) [[B_ADDR]])
+; IS__CGSCC_OPM-NEXT: ret void
+;
+; IS__CGSCC_NPM: Function Attrs: nounwind uwtable
+; IS__CGSCC_NPM-LABEL: define {{[^@]+}}@func
+; IS__CGSCC_NPM-SAME: (float* nocapture nofree readnone [[A:%.*]], float* nofree [[B:%.*]], i32 [[N:%.*]]) local_unnamed_addr #[[ATTR0:[0-9]+]] {
+; IS__CGSCC_NPM-NEXT: entry:
+; IS__CGSCC_NPM-NEXT: [[A_ADDR:%.*]] = alloca float*, align 8
+; IS__CGSCC_NPM-NEXT: [[B_ADDR:%.*]] = alloca float*, align 8
+; IS__CGSCC_NPM-NEXT: [[N_ADDR:%.*]] = alloca i32, align 4
+; IS__CGSCC_NPM-NEXT: store float* [[A]], float** [[A_ADDR]], align 8
+; IS__CGSCC_NPM-NEXT: store float* [[B]], float** [[B_ADDR]], align 8
+; IS__CGSCC_NPM-NEXT: store i32 199, i32* [[N_ADDR]], align 4
+; IS__CGSCC_NPM-NEXT: call void (%struct.ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call(%struct.ident_t* noundef nonnull align 8 dereferenceable(24) @[[GLOB2]], i32 noundef 3, void (i32*, i32*, ...)* noundef bitcast (void (i32*, i32*, i32*, float**, float**)* @.omp_outlined. to void (i32*, i32*, ...)*), i32* noalias nocapture nofree noundef nonnull readonly align 4 dereferenceable(4) [[N_ADDR]], float** noalias nocapture nofree noundef nonnull readonly align 8 dereferenceable(8) [[A_ADDR]], float** noalias nocapture nofree noundef nonnull readonly align 8 dereferenceable(8) [[B_ADDR]])
+; IS__CGSCC_NPM-NEXT: ret void
+;
+entry:
+ %a.addr = alloca float*, align 8
+ %b.addr = alloca float*, align 8
+ %N.addr = alloca i32, align 4
+ store float* %a, float** %a.addr, align 8
+ store float* %b, float** %b.addr, align 8
+ store i32 199, i32* %N.addr, align 4
+ call void (%struct.ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call(%struct.ident_t* nonnull @2, i32 3, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, i32*, float**, float**)* @.omp_outlined. to void (i32*, i32*, ...)*), i32* nonnull %N.addr, float** nonnull %a.addr, float** nonnull %b.addr)
+ ret void
+}
+
+; FIXME: %N should not be loaded but 199 should be used.
+
+define internal void @.omp_outlined.(i32* noalias nocapture readonly %.global_tid., i32* noalias nocapture readnone %.bound_tid., i32* nocapture nonnull readonly align 4 dereferenceable(4) %N, float** nocapture nonnull readonly align 8 dereferenceable(8) %a, float** nocapture nonnull readonly align 8 dereferenceable(8) %b) #1 {
+; IS________OPM: Function Attrs: alwaysinline nofree norecurse nounwind uwtable
+; IS________OPM-LABEL: define {{[^@]+}}@.omp_outlined.
+; IS________OPM-SAME: (i32* noalias nocapture nofree readonly [[DOTGLOBAL_TID_:%.*]], i32* noalias nocapture nofree readnone [[DOTBOUND_TID_:%.*]], i32* nocapture nofree noundef nonnull readonly align 4 dereferenceable(4) [[N:%.*]], float** nocapture nofree noundef nonnull readonly align 8 dereferenceable(8) [[A:%.*]], float** nocapture nofree noundef nonnull readonly align 8 dereferenceable(8) [[B:%.*]]) #[[ATTR1:[0-9]+]] {
+; IS________OPM-NEXT: entry:
+; IS________OPM-NEXT: [[DOTOMP_LB:%.*]] = alloca i32, align 4
+; IS________OPM-NEXT: [[DOTOMP_UB:%.*]] = alloca i32, align 4
+; IS________OPM-NEXT: [[DOTOMP_STRIDE:%.*]] = alloca i32, align 4
+; IS________OPM-NEXT: [[DOTOMP_IS_LAST:%.*]] = alloca i32, align 4
+; IS________OPM-NEXT: [[TMP0:%.*]] = load i32, i32* [[N]], align 4
+; IS________OPM-NEXT: [[SUB2:%.*]] = add nsw i32 [[TMP0]], -2
+; IS________OPM-NEXT: [[CMP:%.*]] = icmp sgt i32 [[TMP0]], 1
+; IS________OPM-NEXT: br i1 [[CMP]], label [[OMP_PRECOND_THEN:%.*]], label [[OMP_PRECOND_END:%.*]]
+; IS________OPM: omp.precond.then:
+; IS________OPM-NEXT: [[TMP1:%.*]] = bitcast i32* [[DOTOMP_LB]] to i8*
+; IS________OPM-NEXT: call void @llvm.lifetime.start.p0i8(i64 noundef 4, i8* nocapture nofree noundef nonnull align 4 dereferenceable(4) [[TMP1]])
+; IS________OPM-NEXT: store i32 0, i32* [[DOTOMP_LB]], align 4
+; IS________OPM-NEXT: [[TMP2:%.*]] = bitcast i32* [[DOTOMP_UB]] to i8*
+; IS________OPM-NEXT: call void @llvm.lifetime.start.p0i8(i64 noundef 4, i8* nocapture nofree noundef nonnull align 4 dereferenceable(4) [[TMP2]])
+; IS________OPM-NEXT: store i32 [[SUB2]], i32* [[DOTOMP_UB]], align 4
+; IS________OPM-NEXT: [[TMP3:%.*]] = bitcast i32* [[DOTOMP_STRIDE]] to i8*
+; IS________OPM-NEXT: call void @llvm.lifetime.start.p0i8(i64 noundef 4, i8* nocapture nofree noundef nonnull align 4 dereferenceable(4) [[TMP3]])
+; IS________OPM-NEXT: store i32 1, i32* [[DOTOMP_STRIDE]], align 4
+; IS________OPM-NEXT: [[TMP4:%.*]] = bitcast i32* [[DOTOMP_IS_LAST]] to i8*
+; IS________OPM-NEXT: call void @llvm.lifetime.start.p0i8(i64 noundef 4, i8* nocapture nofree noundef nonnull align 4 dereferenceable(4) [[TMP4]])
+; IS________OPM-NEXT: store i32 0, i32* [[DOTOMP_IS_LAST]], align 4
+; IS________OPM-NEXT: [[TMP5:%.*]] = load i32, i32* [[DOTGLOBAL_TID_]], align 4
+; IS________OPM-NEXT: call void @__kmpc_for_static_init_4(%struct.ident_t* noundef nonnull align 8 dereferenceable(24) @[[GLOB1]], i32 [[TMP5]], i32 noundef 34, i32* noundef nonnull align 4 dereferenceable(4) [[DOTOMP_IS_LAST]], i32* noundef nonnull align 4 dereferenceable(4) [[DOTOMP_LB]], i32* noundef nonnull align 4 dereferenceable(4) [[DOTOMP_UB]], i32* noundef nonnull align 4 dereferenceable(4) [[DOTOMP_STRIDE]], i32 noundef 1, i32 noundef 1)
+; IS________OPM-NEXT: [[TMP6:%.*]] = load i32, i32* [[DOTOMP_UB]], align 4
+; IS________OPM-NEXT: [[CMP4:%.*]] = icmp sgt i32 [[TMP6]], [[SUB2]]
+; IS________OPM-NEXT: [[COND:%.*]] = select i1 [[CMP4]], i32 [[SUB2]], i32 [[TMP6]]
+; IS________OPM-NEXT: store i32 [[COND]], i32* [[DOTOMP_UB]], align 4
+; IS________OPM-NEXT: [[TMP7:%.*]] = load i32, i32* [[DOTOMP_LB]], align 4
+; IS________OPM-NEXT: [[CMP513:%.*]] = icmp sgt i32 [[TMP7]], [[COND]]
+; IS________OPM-NEXT: br i1 [[CMP513]], label [[OMP_LOOP_EXIT:%.*]], label [[OMP_INNER_FOR_BODY_LR_PH:%.*]]
+; IS________OPM: omp.inner.for.body.lr.ph:
+; IS________OPM-NEXT: [[TMP8:%.*]] = load float*, float** [[B]], align 8
+; IS________OPM-NEXT: [[TMP9:%.*]] = load float*, float** [[A]], align 8
+; IS________OPM-NEXT: [[TMP10:%.*]] = sext i32 [[TMP7]] to i64
+; IS________OPM-NEXT: [[TMP11:%.*]] = sext i32 [[COND]] to i64
+; IS________OPM-NEXT: br label [[OMP_INNER_FOR_BODY:%.*]]
+; IS________OPM: omp.inner.for.body:
+; IS________OPM-NEXT: [[INDVARS_IV:%.*]] = phi i64 [ [[INDVARS_IV_NEXT:%.*]], [[OMP_INNER_FOR_BODY]] ], [ [[TMP10]], [[OMP_INNER_FOR_BODY_LR_PH]] ]
+; IS________OPM-NEXT: [[INDVARS_IV_NEXT]] = add nsw i64 [[INDVARS_IV]], 1
+; IS________OPM-NEXT: [[ARRAYIDX:%.*]] = getelementptr inbounds float, float* [[TMP8]], i64 [[INDVARS_IV_NEXT]]
+; IS________OPM-NEXT: [[TMP12:%.*]] = load float, float* [[ARRAYIDX]], align 4
+; IS________OPM-NEXT: [[CONV7:%.*]] = fadd float [[TMP12]], 1.000000e+00
+; IS________OPM-NEXT: [[ARRAYIDX9:%.*]] = getelementptr inbounds float, float* [[TMP9]], i64 [[INDVARS_IV_NEXT]]
+; IS________OPM-NEXT: store float [[CONV7]], float* [[ARRAYIDX9]], align 4
+; IS________OPM-NEXT: [[CMP5:%.*]] = icmp slt i64 [[INDVARS_IV]], [[TMP11]]
+; IS________OPM-NEXT: br i1 [[CMP5]], label [[OMP_INNER_FOR_BODY]], label [[OMP_LOOP_EXIT]]
+; IS________OPM: omp.loop.exit:
+; IS________OPM-NEXT: call void @__kmpc_for_static_fini(%struct.ident_t* noundef nonnull align 8 dereferenceable(24) @[[GLOB1]], i32 [[TMP5]])
+; IS________OPM-NEXT: call void @llvm.lifetime.end.p0i8(i64 noundef 4, i8* nocapture nofree noundef nonnull align 4 dereferenceable(4) [[TMP4]])
+; IS________OPM-NEXT: call void @llvm.lifetime.end.p0i8(i64 noundef 4, i8* nocapture nofree noundef nonnull align 4 dereferenceable(4) [[TMP3]])
+; IS________OPM-NEXT: call void @llvm.lifetime.end.p0i8(i64 noundef 4, i8* nocapture nofree noundef nonnull align 4 dereferenceable(4) [[TMP2]])
+; IS________OPM-NEXT: call void @llvm.lifetime.end.p0i8(i64 noundef 4, i8* nocapture nofree noundef nonnull align 4 dereferenceable(4) [[TMP1]])
+; IS________OPM-NEXT: br label [[OMP_PRECOND_END]]
+; IS________OPM: omp.precond.end:
+; IS________OPM-NEXT: ret void
+;
+; IS________NPM: Function Attrs: alwaysinline nofree norecurse nounwind uwtable
+; IS________NPM-LABEL: define {{[^@]+}}@.omp_outlined.
+; IS________NPM-SAME: (i32* noalias nocapture nofree readonly [[DOTGLOBAL_TID_:%.*]], i32* noalias nocapture nofree readnone [[DOTBOUND_TID_:%.*]], i32* noalias nocapture nofree noundef nonnull readonly align 4 dereferenceable(4) [[N:%.*]], float** noalias nocapture nofree noundef nonnull readonly align 8 dereferenceable(8) [[A:%.*]], float** noalias nocapture nofree noundef nonnull readonly align 8 dereferenceable(8) [[B:%.*]]) #[[ATTR1:[0-9]+]] {
+; IS________NPM-NEXT: entry:
+; IS________NPM-NEXT: [[DOTOMP_LB:%.*]] = alloca i32, align 4
+; IS________NPM-NEXT: [[DOTOMP_UB:%.*]] = alloca i32, align 4
+; IS________NPM-NEXT: [[DOTOMP_STRIDE:%.*]] = alloca i32, align 4
+; IS________NPM-NEXT: [[DOTOMP_IS_LAST:%.*]] = alloca i32, align 4
+; IS________NPM-NEXT: [[TMP0:%.*]] = load i32, i32* [[N]], align 4
+; IS________NPM-NEXT: [[SUB2:%.*]] = add nsw i32 [[TMP0]], -2
+; IS________NPM-NEXT: [[CMP:%.*]] = icmp sgt i32 [[TMP0]], 1
+; IS________NPM-NEXT: br i1 [[CMP]], label [[OMP_PRECOND_THEN:%.*]], label [[OMP_PRECOND_END:%.*]]
+; IS________NPM: omp.precond.then:
+; IS________NPM-NEXT: [[TMP1:%.*]] = bitcast i32* [[DOTOMP_LB]] to i8*
+; IS________NPM-NEXT: call void @llvm.lifetime.start.p0i8(i64 noundef 4, i8* nocapture nofree noundef nonnull align 4 dereferenceable(4) [[TMP1]])
+; IS________NPM-NEXT: store i32 0, i32* [[DOTOMP_LB]], align 4
+; IS________NPM-NEXT: [[TMP2:%.*]] = bitcast i32* [[DOTOMP_UB]] to i8*
+; IS________NPM-NEXT: call void @llvm.lifetime.start.p0i8(i64 noundef 4, i8* nocapture nofree noundef nonnull align 4 dereferenceable(4) [[TMP2]])
+; IS________NPM-NEXT: store i32 [[SUB2]], i32* [[DOTOMP_UB]], align 4
+; IS________NPM-NEXT: [[TMP3:%.*]] = bitcast i32* [[DOTOMP_STRIDE]] to i8*
+; IS________NPM-NEXT: call void @llvm.lifetime.start.p0i8(i64 noundef 4, i8* nocapture nofree noundef nonnull align 4 dereferenceable(4) [[TMP3]])
+; IS________NPM-NEXT: store i32 1, i32* [[DOTOMP_STRIDE]], align 4
+; IS________NPM-NEXT: [[TMP4:%.*]] = bitcast i32* [[DOTOMP_IS_LAST]] to i8*
+; IS________NPM-NEXT: call void @llvm.lifetime.start.p0i8(i64 noundef 4, i8* nocapture nofree noundef nonnull align 4 dereferenceable(4) [[TMP4]])
+; IS________NPM-NEXT: store i32 0, i32* [[DOTOMP_IS_LAST]], align 4
+; IS________NPM-NEXT: [[TMP5:%.*]] = load i32, i32* [[DOTGLOBAL_TID_]], align 4
+; IS________NPM-NEXT: call void @__kmpc_for_static_init_4(%struct.ident_t* noundef nonnull align 8 dereferenceable(24) @[[GLOB1]], i32 [[TMP5]], i32 noundef 34, i32* noundef nonnull align 4 dereferenceable(4) [[DOTOMP_IS_LAST]], i32* noundef nonnull align 4 dereferenceable(4) [[DOTOMP_LB]], i32* noundef nonnull align 4 dereferenceable(4) [[DOTOMP_UB]], i32* noundef nonnull align 4 dereferenceable(4) [[DOTOMP_STRIDE]], i32 noundef 1, i32 noundef 1)
+; IS________NPM-NEXT: [[TMP6:%.*]] = load i32, i32* [[DOTOMP_UB]], align 4
+; IS________NPM-NEXT: [[CMP4:%.*]] = icmp sgt i32 [[TMP6]], [[SUB2]]
+; IS________NPM-NEXT: [[COND:%.*]] = select i1 [[CMP4]], i32 [[SUB2]], i32 [[TMP6]]
+; IS________NPM-NEXT: store i32 [[COND]], i32* [[DOTOMP_UB]], align 4
+; IS________NPM-NEXT: [[TMP7:%.*]] = load i32, i32* [[DOTOMP_LB]], align 4
+; IS________NPM-NEXT: [[CMP513:%.*]] = icmp sgt i32 [[TMP7]], [[COND]]
+; IS________NPM-NEXT: br i1 [[CMP513]], label [[OMP_LOOP_EXIT:%.*]], label [[OMP_INNER_FOR_BODY_LR_PH:%.*]]
+; IS________NPM: omp.inner.for.body.lr.ph:
+; IS________NPM-NEXT: [[TMP8:%.*]] = load float*, float** [[B]], align 8
+; IS________NPM-NEXT: [[TMP9:%.*]] = load float*, float** [[A]], align 8
+; IS________NPM-NEXT: [[TMP10:%.*]] = sext i32 [[TMP7]] to i64
+; IS________NPM-NEXT: [[TMP11:%.*]] = sext i32 [[COND]] to i64
+; IS________NPM-NEXT: br label [[OMP_INNER_FOR_BODY:%.*]]
+; IS________NPM: omp.inner.for.body:
+; IS________NPM-NEXT: [[INDVARS_IV:%.*]] = phi i64 [ [[INDVARS_IV_NEXT:%.*]], [[OMP_INNER_FOR_BODY]] ], [ [[TMP10]], [[OMP_INNER_FOR_BODY_LR_PH]] ]
+; IS________NPM-NEXT: [[INDVARS_IV_NEXT]] = add nsw i64 [[INDVARS_IV]], 1
+; IS________NPM-NEXT: [[ARRAYIDX:%.*]] = getelementptr inbounds float, float* [[TMP8]], i64 [[INDVARS_IV_NEXT]]
+; IS________NPM-NEXT: [[TMP12:%.*]] = load float, float* [[ARRAYIDX]], align 4
+; IS________NPM-NEXT: [[CONV7:%.*]] = fadd float [[TMP12]], 1.000000e+00
+; IS________NPM-NEXT: [[ARRAYIDX9:%.*]] = getelementptr inbounds float, float* [[TMP9]], i64 [[INDVARS_IV_NEXT]]
+; IS________NPM-NEXT: store float [[CONV7]], float* [[ARRAYIDX9]], align 4
+; IS________NPM-NEXT: [[CMP5:%.*]] = icmp slt i64 [[INDVARS_IV]], [[TMP11]]
+; IS________NPM-NEXT: br i1 [[CMP5]], label [[OMP_INNER_FOR_BODY]], label [[OMP_LOOP_EXIT]]
+; IS________NPM: omp.loop.exit:
+; IS________NPM-NEXT: call void @__kmpc_for_static_fini(%struct.ident_t* noundef nonnull align 8 dereferenceable(24) @[[GLOB1]], i32 [[TMP5]])
+; IS________NPM-NEXT: call void @llvm.lifetime.end.p0i8(i64 noundef 4, i8* nocapture nofree noundef nonnull align 4 dereferenceable(4) [[TMP4]])
+; IS________NPM-NEXT: call void @llvm.lifetime.end.p0i8(i64 noundef 4, i8* nocapture nofree noundef nonnull align 4 dereferenceable(4) [[TMP3]])
+; IS________NPM-NEXT: call void @llvm.lifetime.end.p0i8(i64 noundef 4, i8* nocapture nofree noundef nonnull align 4 dereferenceable(4) [[TMP2]])
+; IS________NPM-NEXT: call void @llvm.lifetime.end.p0i8(i64 noundef 4, i8* nocapture nofree noundef nonnull align 4 dereferenceable(4) [[TMP1]])
+; IS________NPM-NEXT: br label [[OMP_PRECOND_END]]
+; IS________NPM: omp.precond.end:
+; IS________NPM-NEXT: ret void
+;
+entry:
+ %.omp.lb = alloca i32, align 4
+ %.omp.ub = alloca i32, align 4
+ %.omp.stride = alloca i32, align 4
+ %.omp.is_last = alloca i32, align 4
+ %0 = load i32, i32* %N, align 4
+ %sub2 = add nsw i32 %0, -2
+ %cmp = icmp sgt i32 %0, 1
+ br i1 %cmp, label %omp.precond.then, label %omp.precond.end
+
+omp.precond.then: ; preds = %entry
+ %1 = bitcast i32* %.omp.lb to i8*
+ call void @llvm.lifetime.start.p0i8(i64 4, i8* nonnull %1) #3
+ store i32 0, i32* %.omp.lb, align 4
+ %2 = bitcast i32* %.omp.ub to i8*
+ call void @llvm.lifetime.start.p0i8(i64 4, i8* nonnull %2) #3
+ store i32 %sub2, i32* %.omp.ub, align 4
+ %3 = bitcast i32* %.omp.stride to i8*
+ call void @llvm.lifetime.start.p0i8(i64 4, i8* nonnull %3) #3
+ store i32 1, i32* %.omp.stride, align 4
+ %4 = bitcast i32* %.omp.is_last to i8*
+ call void @llvm.lifetime.start.p0i8(i64 4, i8* nonnull %4) #3
+ store i32 0, i32* %.omp.is_last, align 4
+ %5 = load i32, i32* %.global_tid., align 4
+ call void @__kmpc_for_static_init_4(%struct.ident_t* nonnull @1, i32 %5, i32 34, i32* nonnull %.omp.is_last, i32* nonnull %.omp.lb, i32* nonnull %.omp.ub, i32* nonnull %.omp.stride, i32 1, i32 1) #3
+ %6 = load i32, i32* %.omp.ub, align 4
+ %cmp4 = icmp sgt i32 %6, %sub2
+ %cond = select i1 %cmp4, i32 %sub2, i32 %6
+ store i32 %cond, i32* %.omp.ub, align 4
+ %7 = load i32, i32* %.omp.lb, align 4
+ %cmp513 = icmp sgt i32 %7, %cond
+ br i1 %cmp513, label %omp.loop.exit, label %omp.inner.for.body.lr.ph
+
+omp.inner.for.body.lr.ph: ; preds = %omp.precond.then
+ %8 = load float*, float** %b, align 8
+ %9 = load float*, float** %a, align 8
+ %10 = sext i32 %7 to i64
+ %11 = sext i32 %cond to i64
+ br label %omp.inner.for.body
+
+omp.inner.for.body: ; preds = %omp.inner.for.body, %omp.inner.for.body.lr.ph
+ %indvars.iv = phi i64 [ %indvars.iv.next, %omp.inner.for.body ], [ %10, %omp.inner.for.body.lr.ph ]
+ %indvars.iv.next = add nsw i64 %indvars.iv, 1
+ %arrayidx = getelementptr inbounds float, float* %8, i64 %indvars.iv.next
+ %12 = load float, float* %arrayidx, align 4
+ %conv7 = fadd float %12, 1.000000e+00
+ %arrayidx9 = getelementptr inbounds float, float* %9, i64 %indvars.iv.next
+ store float %conv7, float* %arrayidx9, align 4
+ %cmp5 = icmp slt i64 %indvars.iv, %11
+ br i1 %cmp5, label %omp.inner.for.body, label %omp.loop.exit
+
+omp.loop.exit: ; preds = %omp.inner.for.body, %omp.precond.then
+ call void @__kmpc_for_static_fini(%struct.ident_t* nonnull @1, i32 %5)
+ call void @llvm.lifetime.end.p0i8(i64 4, i8* nonnull %4) #3
+ call void @llvm.lifetime.end.p0i8(i64 4, i8* nonnull %3) #3
+ call void @llvm.lifetime.end.p0i8(i64 4, i8* nonnull %2) #3
+ call void @llvm.lifetime.end.p0i8(i64 4, i8* nonnull %1) #3
+ br label %omp.precond.end
+
+omp.precond.end: ; preds = %omp.loop.exit, %entry
+ ret void
+}
+
+; Function Attrs: argmemonly nounwind willreturn
+declare void @llvm.lifetime.start.p0i8(i64 immarg, i8* nocapture) #2
+
+; Function Attrs: argmemonly nounwind willreturn
+declare void @llvm.lifetime.end.p0i8(i64 immarg, i8* nocapture) #2
+
+declare dso_local void @__kmpc_for_static_init_4(%struct.ident_t*, i32, i32, i32*, i32*, i32*, i32*, i32, i32) local_unnamed_addr
+
+; Function Attrs: nounwind
+declare void @__kmpc_for_static_fini(%struct.ident_t*, i32) local_unnamed_addr #3
+
+; Function Attrs: nounwind
+declare !callback !1 void @__kmpc_fork_call(%struct.ident_t*, i32, void (i32*, i32*, ...)*, ...) local_unnamed_addr #3
+
+attributes #0 = { nounwind uwtable }
+attributes #1 = { alwaysinline nofree norecurse nounwind uwtable }
+attributes #2 = { nounwind }
+
+!llvm.module.flags = !{!0}
+
+!0 = !{i32 7, !"openmp", i32 50}
+!1 = !{!2}
+!2 = !{i64 2, i64 -1, i64 -1, i1 true}
+;.
+; CHECK: attributes #[[ATTR0:[0-9]+]] = { nounwind uwtable }
+; CHECK: attributes #[[ATTR1:[0-9]+]] = { alwaysinline nofree norecurse nounwind uwtable }
+; CHECK: attributes #[[ATTR2:[0-9]+]] = { argmemonly nofree nosync nounwind willreturn }
+;.
+; CHECK: [[META0:![0-9]+]] = !{i32 7, !"openmp", i32 50}
+; CHECK: [[META1:![0-9]+]] = !{!2}
+; CHECK: [[META2:![0-9]+]] = !{i64 2, i64 -1, i64 -1, i1 true}
+;.
More information about the llvm-commits
mailing list