[llvm-bugs] [Bug 27577] New: Inefficient code gen for cmove

via llvm-bugs llvm-bugs at lists.llvm.org
Fri Apr 29 13:27:06 PDT 2016


https://llvm.org/bugs/show_bug.cgi?id=27577

            Bug ID: 27577
           Summary: Inefficient code gen for cmove
           Product: clang
           Version: trunk
          Hardware: PC
                OS: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: LLVM Codegen
          Assignee: unassignedclangbugs at nondot.org
          Reporter: hjl.tools at gmail.com
                CC: david.l.kreitzer at intel.com, llvm-bugs at lists.llvm.org,
                    zia.ansari at intel.com
    Classification: Unclassified

On x86-64, at -O2, for

---
extern int a, b;

int foo(int cond)
{
   return cond == -1 ? a : b;
}
---

clang 3.7 and 3.9 generate

---
    cmpl    $-1, %edi
    movl    $a, %eax
    movl    $b, %ecx
    cmoveq    %rax, %rcx
    movl    (%rcx), %eax
    retq
---

---
    cmpl    $-1, %edi
    movl    a(%rip), %eax
    cmovne    b(%rip), %eax
    ret
---

is better.

-- 
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/20160429/624fa57a/attachment-0001.html>


More information about the llvm-bugs mailing list