http://llvm.org/bugs/show_bug.cgi?id=8113
Andrew Trick <[email protected]> changed: What |Removed |Added ---------------------------------------------------------------------------- Status|REOPENED |RESOLVED Resolution| |FIXED --- Comment #4 from Andrew Trick <[email protected]> 2011-11-11 16:39:12 CST --- I celebrated this bug's anniversary by trying to reproduce it again. It turns out there were a bunch of issues at play. The main problem I think is that LSR generates negative offsets. ARM ISEL doesn't fold those into loads. Thumb mode does, but uses different encoding for zero offset and negative offset. In the end, I gave up on this test case and just added my own: ldrd-memop.ll The fix I came up with in r144409 is slightly different than the proposed patch. I ran the full llvm test suite for both cortex-a8 and a9, and both thumb and -mno-thumb. -- Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email ------- 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
