http://llvm.org/bugs/show_bug.cgi?id=22265
Michael Kuperstein <[email protected]> changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |RESOLVED CC| |michael.m.kuperstein@intel. | |com Resolution|--- |DUPLICATE --- Comment #3 from Michael Kuperstein <[email protected]> --- This looks like a duplicate of PR15941 (albeit with a smaller reproducer). In David's reproducer, the volatile load gets folded into a CMOV (that came from the select), and the non-volatile load gets folded into the INC (that came from the add). Because of the volatility of the first load, we have a chain from the CMOV to the INC, but the CMOV also depends on the result of the INC (through EFLAGS), and we get a cycle. *** This bug has been marked as a duplicate of bug 15941 *** -- You are receiving this mail because: You are on the CC list for the bug.
_______________________________________________ LLVMbugs mailing list [email protected] http://lists.cs.uiuc.edu/mailman/listinfo/llvmbugs
