[LLVMbugs] [Bug 22640] New: r217953 degraded lsr optimization at -O3 for -march=nehalem in SciMark2's Sparse matmult benchmark

[bugzilla-daemon at llvm.org]

http://llvm.org/bugs/show_bug.cgi?id=22640

            Bug ID: 22640
           Summary: r217953 degraded lsr optimization at -O3 for
                    -march=nehalem in SciMark2's Sparse matmult benchmark
           Product: clang
           Version: 3.6
          Hardware: PC
                OS: All
            Status: NEW
          Severity: normal
          Priority: P
         Component: -New Bugs
          Assignee: unassignedclangbugs at nondot.org
          Reporter: howarth.mailing.lists at gmail.com
                CC: llvmbugs at cs.uiuc.edu
    Classification: Unclassified

The SciMark2's Sparse matmult benchmark performance is degraded by up to 24%
after r217953 at -O3 -march=nehalem. The performance regression is reversed by
passing "-mllvm -disable-lsr" to the compiler for the SparseCompRow.c source
file. A diff of the assembly generated for "-O3 -march=nehalem" without and
with  "-mllvm -disable-lsr" shows that additional spills are being introduced
by the lsr optimizations.

-- 
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/20150220/c5327df7/attachment.html>