[Bug regex/605] New: regex goes uselessly through slow paths

["bonzini at gnu dot org" <sourceware-bugzilla@sources.redhat.com>]

An improvement of 10-30% in the speed of the matcher (LC_ALL=C or other
single-byte character sets) can be achieved by avoiding that OP_PERIOD goes
through the slow multibyte paths.  This is done by the attached patch.

While this patch only affects SBCS (i.e. scripts, which are starting to use
LC_ALL=C to avoid bugs due to the collation rules of some languages), in
principle UTF-8 optimization could be fixed to be optimizable as well.  The
slowness of prune_impossible_nodes, alas, makes this impractical yet.

-- 
           Summary: regex goes uselessly through slow paths
           Product: glibc
           Version: unspecified
            Status: NEW
          Severity: normal
          Priority: P2
         Component: regex
        AssignedTo: bonzini at gnu dot org
        ReportedBy: bonzini at gnu dot org
                CC: glibc-bugs-regex at sources dot redhat dot com,glibc-
                    bugs at sources dot redhat dot com
OtherBugsDependingO 500
             nThis:


http://sources.redhat.com/bugzilla/show_bug.cgi?id=605

------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.