[Bug rtl-optimization/59466] New: Slow code generation by LRA for memory addresses on PPC

[vmakarov at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=59466

            Bug ID: 59466
           Summary: Slow code generation by LRA for memory addresses on
                    PPC
           Product: gcc
           Version: 4.9.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: rtl-optimization
          Assignee: unassigned at gcc dot gnu.org
          Reporter: vmakarov at gcc dot gnu.org

Investigating SPEC2000 code (e.g. bzip2.c) generated by LRA on PPC using -O3
-mtune=power7 -mlra, sometimes I see code like

addis 9,2,.LC77@toc@ha
addi 9,9,.LC77@toc@l
ld 9,0(9)

Although there is a faster and smaller code

addis 9,2,.LC77@toc@ha
ld 9,.LC77@toc@l(9)