http://llvm.org/bugs/show_bug.cgi?id=18136
Tim Northover <[email protected]> changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |RESOLVED Resolution|--- |FIXED --- Comment #3 from Tim Northover <[email protected]> --- Thanks for the test-case. I believe it's the fact that the VLDMIA_UPD is at the beginning of the block that messes up the tracking rather than isCSRestore. I actually think the entire idea of tracking FirstPop like that was misguided: the folding *has* to happen precisely where we'd insert an sp-update to be valid, which is the entire reason MBBI exists. We should just be using that. This one should be fixed in r196493. I really hope it's the last. -- 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
