[LLVMbugs] [Bug 21122] New: Compiler overlooks vhadd operation given widen/add/shift/narrow operation

[bugzilla-daemon at llvm.org]

http://llvm.org/bugs/show_bug.cgi?id=21122

            Bug ID: 21122
           Summary: Compiler overlooks vhadd operation given
                    widen/add/shift/narrow operation
           Product: libraries
           Version: trunk
          Hardware: PC
                OS: Linux
            Status: NEW
          Severity: enhancement
          Priority: P
         Component: Backend: ARM
          Assignee: unassignedbugs at nondot.org
          Reporter: simon.hosie at arm.com
                CC: llvmbugs at cs.uiuc.edu
    Classification: Unclassified

Created attachment 13113
  --> http://llvm.org/bugs/attachment.cgi?id=13113&action=edit
addh.c

For the attached file and this command line:

clang --target=arm-linux-gnueabihf -mfpu=neon -O3 addh.c -S -o-

The compiler currently emits:

        vaddl.u8        q8, d1, d3
        vshrn.i16       d17, q8, #1
        vaddl.u8        q9, d0, d2
        vshrn.i16       d16, q9, #1
        vorr    q0, q8, q8
        bx      lr

Should be:

        vhadd.u8        q0, q0, q1
        bx      lr


I want to extend this bug to include things like vqadd (which might use min
instead of shrn) but I can't seem to wring the relevant LLVM code out of Clang.

-- 
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/20141001/8845a1cd/attachment.html>