wrong code with "opt -gvn -licm -instcombine -loop-rotate -loop-unroll -simplifycfg -licm -loop-reroll" on x86_64-linux-gnu

[zhendongsu]

Bugzilla Link	34639
Version	trunk
OS	All

Extended Description

Tested with trunk revision 313398.

$ clangpolly -v 
clang version 6.0.0 (http://llvm.org/git/clang.git f5779ed24ea68e59509ac648aac13c197170ab4d)
Target: x86_64-unknown-linux-gnu
Thread model: posix
InstalledDir: /home/su/bin
Found candidate GCC installation: /usr/lib/gcc/i686-linux-gnu/4.9
Found candidate GCC installation: /usr/lib/gcc/i686-linux-gnu/4.9.4
Found candidate GCC installation: /usr/lib/gcc/i686-linux-gnu/5
Found candidate GCC installation: /usr/lib/gcc/i686-linux-gnu/5.3.0
Found candidate GCC installation: /usr/lib/gcc/x86_64-linux-gnu/4.4
Found candidate GCC installation: /usr/lib/gcc/x86_64-linux-gnu/4.4.7
Found candidate GCC installation: /usr/lib/gcc/x86_64-linux-gnu/4.6
Found candidate GCC installation: /usr/lib/gcc/x86_64-linux-gnu/4.6.4
Found candidate GCC installation: /usr/lib/gcc/x86_64-linux-gnu/4.7
Found candidate GCC installation: /usr/lib/gcc/x86_64-linux-gnu/4.7.3
Found candidate GCC installation: /usr/lib/gcc/x86_64-linux-gnu/4.8
Found candidate GCC installation: /usr/lib/gcc/x86_64-linux-gnu/4.8.5
Found candidate GCC installation: /usr/lib/gcc/x86_64-linux-gnu/4.9
Found candidate GCC installation: /usr/lib/gcc/x86_64-linux-gnu/4.9.4
Found candidate GCC installation: /usr/lib/gcc/x86_64-linux-gnu/5
Found candidate GCC installation: /usr/lib/gcc/x86_64-linux-gnu/5.3.0
Found candidate GCC installation: /usr/lib/gcc/x86_64-linux-gnu/6
Found candidate GCC installation: /usr/lib/gcc/x86_64-linux-gnu/6.2.0
Selected GCC installation: /usr/lib/gcc/x86_64-linux-gnu/4.9
Candidate multilib: .;@m64
Candidate multilib: 32;@m32
Candidate multilib: x32;@mx32
Selected multilib: .;@m64
$ 
$ clangpolly -O3 -mllvm -disable-llvm-optzns -c -emit-llvm -o small.bc small.c
$ optpolly -gvn -licm -instcombine -loop-rotate -loop-unroll -simplifycfg -licm -loop-reroll -o small-opt.bc small.bc
$ clangpolly small-opt.bc
$ ./a.out
3
$ 
$ clangpolly -O0 small.c; ./a.out
1
$ 

int printf (const char *, ...);

int a, b, c[1][3];

int main ()
{
  for (; a < 1; a++)
    for (b = 0; b < 3; b++)
      c[a][b] = 1;
  printf ("%d\n", a);
  return 0; 
}

[fhahn]

Still reproduces on current main: https://clang.godbolt.org/z/cPao5j31z

Culprit is likely -loop-reroll, which has known issues.

IR reproducer:

; ModuleID = '/app/example.c'
source_filename = "/app/example.c"
target datalayout = "e-m:e-i64:64-f80:128-n8:16:32:64-S128"
target triple = "x86_64-unknown-linux-gnu"

@a = common dso_local global i32 0, align 4
@b = common dso_local global i32 0, align 4
@c = common dso_local global [1 x [3 x i32]] zeroinitializer, align 4
@.str = private unnamed_addr constant [4 x i8] c"%d\0A\00", align 1

; Function Attrs: nounwind uwtable
define dso_local i32 @main() #0 {
entry:
  %retval = alloca i32, align 4
  store i32 0, i32* %retval, align 4
  br label %for.cond

for.cond:                                         ; preds = %for.inc6, %entry
  %0 = load i32, i32* @a, align 4, !tbaa !2
  %cmp = icmp slt i32 %0, 1
  br i1 %cmp, label %for.body, label %for.end8

for.body:                                         ; preds = %for.cond
  store i32 0, i32* @b, align 4, !tbaa !2
  br label %for.cond1

for.cond1:                                        ; preds = %for.inc, %for.body
  %1 = load i32, i32* @b, align 4, !tbaa !2
  %cmp2 = icmp slt i32 %1, 3
  br i1 %cmp2, label %for.body3, label %for.end

for.body3:                                        ; preds = %for.cond1
  %2 = load i32, i32* @a, align 4, !tbaa !2
  %idxprom = sext i32 %2 to i64
  %arrayidx = getelementptr inbounds [1 x [3 x i32]], [1 x [3 x i32]]* @c, i64 0, i64 %idxprom
  %3 = load i32, i32* @b, align 4, !tbaa !2
  %idxprom4 = sext i32 %3 to i64
  %arrayidx5 = getelementptr inbounds [3 x i32], [3 x i32]* %arrayidx, i64 0, i64 %idxprom4
  store i32 1, i32* %arrayidx5, align 4, !tbaa !2
  br label %for.inc

for.inc:                                          ; preds = %for.body3
  %4 = load i32, i32* @b, align 4, !tbaa !2
  %inc = add nsw i32 %4, 1
  store i32 %inc, i32* @b, align 4, !tbaa !2
  br label %for.cond1

for.end:                                          ; preds = %for.cond1
  br label %for.inc6

for.inc6:                                         ; preds = %for.end
  %5 = load i32, i32* @a, align 4, !tbaa !2
  %inc7 = add nsw i32 %5, 1
  store i32 %inc7, i32* @a, align 4, !tbaa !2
  br label %for.cond

for.end8:                                         ; preds = %for.cond
  %6 = load i32, i32* @a, align 4, !tbaa !2
  %call = call i32 (i8*, ...) @printf(i8* getelementptr inbounds ([4 x i8], [4 x i8]* @.str, i32 0, i32 0), i32 %6)
  ret i32 0
}

declare dso_local i32 @printf(i8*, ...) #1

attributes #0 = { nounwind uwtable "correctly-rounded-divide-sqrt-fp-math"="false" "disable-tail-calls"="false" "less-precise-fpmad"="false" "no-frame-pointer-elim"="false" "no-infs-fp-math"="false" "no-jump-tables"="false" "no-nans-fp-math"="false" "no-signed-zeros-fp-math"="false" "no-trapping-math"="false" "stack-protector-buffer-size"="8" "target-cpu"="x86-64" "target-features"="+fxsr,+mmx,+sse,+sse2,+x87" "unsafe-fp-math"="false" "use-soft-float"="false" }
attributes #1 = { "correctly-rounded-divide-sqrt-fp-math"="false" "disable-tail-calls"="false" "less-precise-fpmad"="false" "no-frame-pointer-elim"="false" "no-infs-fp-math"="false" "no-nans-fp-math"="false" "no-signed-zeros-fp-math"="false" "no-trapping-math"="false" "stack-protector-buffer-size"="8" "target-cpu"="x86-64" "target-features"="+fxsr,+mmx,+sse,+sse2,+x87" "unsafe-fp-math"="false" "use-soft-float"="false" }

!llvm.module.flags = !{!0}
!llvm.ident = !{!1}

!0 = !{i32 1, !"wchar_size", i32 4}
!1 = !{!"clang version 7.0.0 (tags/RELEASE_700/final 342594)"}
!2 = !{!3, !3, i64 0}
!3 = !{!"int", !4, i64 0}
!4 = !{!"omnipotent char", !5, i64 0}
!5 = !{!"Simple C/C++ TBAA"}

[nikic]

The LoopReroll pass has been removed by #80972.