[Bug 83658] New: Loop unrolling doesn't take into account whether it will decrease or increase performance

bugzilla-daemon at freedesktop.org bugzilla-daemon at freedesktop.org
Tue Sep 9 03:10:22 PDT 2014


https://bugs.freedesktop.org/show_bug.cgi?id=83658

          Priority: medium
            Bug ID: 83658
          Assignee: idr at freedesktop.org
            Blocks: 77547
           Summary: Loop unrolling doesn't take into account whether it
                    will decrease or increase performance
        QA Contact: intel-3d-bugs at lists.freedesktop.org
          Severity: normal
    Classification: Unclassified
                OS: Linux (All)
          Reporter: eero.t.tamminen at intel.com
          Hardware: x86-64 (AMD64)
            Status: NEW
           Version: git
         Component: Drivers/DRI/i965
           Product: Mesa

Currently loop unrolling is done in frontend.  This is done based on
heuristics, not on actual data from backend whether and how much unrolling
makes sense for given HW: will unrolling run out of registers for SIMD16, how
many instructions fit into cache...

Backend should provide this kind of information for unrolling and frontend
should utilize it (either before unrolling, to control how much to do it, or
after unrolling to revert it).

(Frontend probably should be fixed to push loop invariant code outside of loop
before devoting time to improving unrolling itself.)

-- 
You are receiving this mail because:
You are the QA Contact for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.freedesktop.org/archives/intel-3d-bugs/attachments/20140909/0fc31042/attachment.html>


More information about the intel-3d-bugs mailing list