replica
/
hanchenye-llvm-project
2
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

[Fix] OpenMP parallel loop detection for the isl backend 

There was a bug in the IslAst which caused that no more outermost
  parallel loops were detected/checked after a parallel outermost loop
  of depth 1.

  + Test case attached

llvm-svn: 217452
This commit is contained in:
Johannes Doerfert 2014-09-09 17:03:54 +00:00
parent 0910605af6
commit 8e95dc657e
2 changed files with 56 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

3
polly/lib/CodeGen/IslAst.cpp
View File

@ -249,7 +249,8 @@ astBuildAfterFor(__isl_take isl_ast_node *Node, __isl_keep isl_ast_build *Build,
else else
Payload->IsInnermostParallel = Payload->IsInnermostParallel =
astScheduleDimIsParallel(Build, BuildInfo->Deps, Payload); astScheduleDimIsParallel(Build, BuildInfo->Deps, Payload);
} else if (Payload->IsOutermostParallel) }
if (Payload->IsOutermostParallel)
BuildInfo->InParallelFor = false; BuildInfo->InParallelFor = false;
isl_id_free(Id); isl_id_free(Id);

54
polly/test/Isl/Ast/OpenMP/multiple_loops_outer_parallel.ll Normal file
View File

@ -0,0 +1,54 @@
; RUN: opt %loadPolly -polly-ast -polly-ast-detect-parallel -analyze < %s | FileCheck %s
;
; void jd(int *A) {
; CHECK: #pragma omp parallel for
; for (int i = 0; i < 1024; i++)
; A[i] = 1;
; CHECK: #pragma omp parallel for
; for (int i = 0; i < 1024; i++)
; A[i] = A[i] * 2;
; }
;
target datalayout = "e-m:e-i64:64-f80:128-n8:16:32:64-S128"
define void @jd(i32* %A) {
entry:
br label %for.cond
for.cond: ; preds = %for.inc, %entry
%indvars.iv1 = phi i64 [ %indvars.iv.next2, %for.inc ], [ 0, %entry ]
%exitcond3 = icmp ne i64 %indvars.iv1, 1024
br i1 %exitcond3, label %for.body, label %for.end
for.body: ; preds = %for.cond
%arrayidx = getelementptr inbounds i32* %A, i64 %indvars.iv1
store i32 1, i32* %arrayidx, align 4
br label %for.inc
for.inc: ; preds = %for.body
%indvars.iv.next2 = add nuw nsw i64 %indvars.iv1, 1
br label %for.cond
for.end: ; preds = %for.cond
br label %for.cond2
for.cond2: ; preds = %for.inc9, %for.end
%indvars.iv = phi i64 [ %indvars.iv.next, %for.inc9 ], [ 0, %for.end ]
%exitcond = icmp ne i64 %indvars.iv, 1024
br i1 %exitcond, label %for.body4, label %for.end11
for.body4: ; preds = %for.cond2
%arrayidx6 = getelementptr inbounds i32* %A, i64 %indvars.iv
%tmp = load i32* %arrayidx6, align 4
%mul = shl nsw i32 %tmp, 1
%arrayidx8 = getelementptr inbounds i32* %A, i64 %indvars.iv
store i32 %mul, i32* %arrayidx8, align 4
br label %for.inc9
for.inc9: ; preds = %for.body4
%indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
br label %for.cond2
for.end11: ; preds = %for.cond2
ret void
}