[llvm-bugs] [Bug 34229] New: Evaluate constant condition of if statement has no effect

via llvm-bugs llvm-bugs at lists.llvm.org
Fri Aug 18 00:57:38 PDT 2017


https://bugs.llvm.org/show_bug.cgi?id=34229

            Bug ID: 34229
           Summary: Evaluate constant condition of if statement has no
                    effect
           Product: clang
           Version: trunk
          Hardware: PC
                OS: All
            Status: NEW
          Severity: normal
          Priority: P
         Component: C++
          Assignee: unassignedclangbugs at nondot.org
          Reporter: bluechristlove at 163.com
                CC: dgregor at apple.com, llvm-bugs at lists.llvm.org

Simple Example:

int main()
{
  int x;
  if (x || 1) {}
}

Clang can not evaluate this condition and will emit IR like this:

define signext i32 @main() #0 {
entry:
  %retval = alloca i32, align 4
  %x = alloca i32, align 4
  store i32 0, i32* %retval, align 4
  %0 = load i32, i32* %x, align 4
  %tobool = icmp ne i32 %0, 0
  br i1 %tobool, label %if.then, label %lor.lhs.false

lor.lhs.false:                                    ; preds = %entry
  br i1 true, label %if.then, label %if.end

if.then:                                          ; preds = %lor.lhs.false,
%entry
  br label %if.end

if.end:                                           ; preds = %if.then,
%lor.lhs.false
  %1 = load i32, i32* %retval, align 4
  ret i32 %1
}

However, when we swap the position of LHS and RHS, Clang can recognize it:

define signext i32 @main() #0 {
entry:
  %x = alloca i32, align 4
  ret i32 0
}

The same condition contains x && 0 and so on. I investigate it initially, I
find we have code to do it like code "void
DataRecursiveIntBinOpEvaluator::process(EvalResult &Result)" But seems that
there is something wrong here in if stmt condition handling.

-- 
You are receiving this mail because:
You are on the CC list for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-bugs/attachments/20170818/a15b18c5/attachment.html>


More information about the llvm-bugs mailing list