[LLVMbugs] [Bug 21042] New: fpowi handling on 64-bit architectures

[bugzilla-daemon at llvm.org]

http://llvm.org/bugs/show_bug.cgi?id=21042

            Bug ID: 21042
           Summary: fpowi handling on 64-bit architectures
           Product: new-bugs
           Version: 3.4
          Hardware: PC
                OS: All
            Status: NEW
          Severity: normal
          Priority: P
         Component: new bugs
          Assignee: unassignedbugs at nondot.org
          Reporter: rcgorton at cognitive-electronics.com
                CC: llvmbugs at cs.uiuc.edu
    Classification: Unclassified

Created attachment 13075
  --> http://llvm.org/bugs/attachment.cgi?id=13075&action=edit
a partially compiled to .ll example

When compiling the attached .ll file, compilation for 64-bit targets assert
with "do not know how to promote this operator's operand".  sparcv9, powerpc,
aarch64.

The culprit is the fpowi() opcode.  I tried to add cases in
PromoteIntegerResult and PromoteIntegerOperand, but I get other asserts. 
Either I'm missing something in my implementation (quite possible) or there is
a deeper problem with assumptions about the datatypes of opcodes.

The original source is slamchf77.f compiled with gcc4.6.0 + dragonegg to a .ll
file.

Richard Gorton
rcgorton at cog-e.com

-- 
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/20140923/295a40fb/attachment.html>