[llvm-bugs] [Bug 24522] New: [AArch64 MachineCombine] Enhance/Add support for general reassociation to reduce the critical path

[via llvm-bugs]

https://llvm.org/bugs/show_bug.cgi?id=24522

            Bug ID: 24522
           Summary: [AArch64 MachineCombine] Enhance/Add support for
                    general reassociation to reduce the critical path
           Product: libraries
           Version: trunk
          Hardware: PC
                OS: Windows NT
            Status: NEW
          Severity: normal
          Priority: P
         Component: Backend: AArch64
          Assignee: unassignedbugs at nondot.org
          Reporter: mcrosier at codeaurora.org
                CC: gberry at codeaurora.org, james.molloy at arm.com,
                    kristof.beyls at arm.com, llvm-bugs at lists.llvm.org,
                    spatel+llvm at rotateright.com
    Classification: Unclassified

The x86 target supports general reassociation optimizations in the machine
combiner.  We should support/extend the same type of optimization for AArch64.

Here are a few of the related commits (not necessarily all inclusive):

r239486: [x86] Add a reassociation optimization to increase ILP via the
MachineCombiner pass

r240361: [x86] generalize reassociation optimization in machine combiner to 2
instructions

r241515: [x86] extend machine combiner reassociation optimization to SSE scalar
adds

r241752: [x86] enable machine combiner reassociations for scalar
single-precision multiplies

r241871: [x86] enable machine combiner reassociations for scalar
double-precision adds


IIRC, this might improve the performance of 188.ammp.

-- 
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/20150820/ac0fa386/attachment.html>