MERGE: strchr-speedup - speed up field name canonicalization



   Hi!

During oprofiling beast, I noticed that 6% of the total cpu usage while playing
a test song is spent in strchr(). The reason is that record fields (probably for
probes) are canonicalized to A-Za-z0-9, and that for every character in the
records' field name, a string of these 62 chars is searched using strchr(). I
rewrote the function which does canonicalization using one if statement, and
beast CPU usage drops by a few percent (and strchr() disappears from the
profile).

Please merge:

repo:   http://space.twc.de/public/git/stwbeast.git
branch: strchr-speedup

   Cu... Stefan
-- 
Stefan Westerfeld, Hamburg/Germany, http://space.twc.de/~stefan


[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]