You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@httpd.apache.org by Dean Gaudet <dg...@arctic.org> on 1997/07/01 00:54:55 UTC
ok it's broken!
I committed to HEAD:
- Doug's DCE/SFIO patch
- http_main revamp/speedup patch
- http_config run_method speedup patch
- LFLAGS -> LDFLAGS (don't forget to tweak your Configuration)
- Makefile dependency cleanup
- make_dirstr/directory_walk speedup
A few months back I profiled the server heavily and developed the
run_method speedup and the make_dirstr/directory_walk speedup. After I
did that, gprof profiles were relatively flat with palloc() being the big
hog. So I stopped approaching it from that angle and now I'm looking at
the system performance angle (i.e. strace is my friend ;).
It works for me under light testing, and everything else was tested
heavier when I originally wrote it.
That should be about it for me on HEAD today.
Dean
Re: ok it's broken!
Posted by Dean Gaudet <dg...@arctic.org>.
On Mon, 30 Jun 1997, Dean Gaudet wrote:
> palloc() being the big hog
P.S. This isn't an indication that palloc() itself is slow, just that it's
called far too much. But I couldn't see any really obvious good wins.
Dean