comparing kernel socket performance
Greg Lindahl
lindahl@cs.virginia.edu
Sun, 6 Sep 1998 18:41:29 -0400
> You are mistaken. Only a kernel and (application which uses > 1024
> fds) recompile is neccessary to take advantage of thousands of
> descriptors.
Well, that's nice to hear; still not as good as other Unixes, but if
you bothered to tell us where this patch could be found, then maybe
someone could use it. I looked at the Squid pages and the link for
"linux binaries and patches" isn't pingable.
> Squid is unique in that it has cached fd's for files and it has
> sockets. IRC servers are mainly socket bound. This can be gotten
> around by simply writing the code correctly; meaning, use shared
> memory segments or unix domains sockets to share common data, then
> run multiple processes in accept-on-common-fd and select loops. (or
> SIGIO/SIGPOLL driven I/O).
That isn't "writing the code correctly", that's a blecherous
workaround when it's much cleaner to fix libc and/or the kernel to
allow a reasonable number of fd's per process.
> If I was inclined to care, I'd write an IRC server the *right* way
> rather than the pseudo-college-student-Eric-Allman-wannabe-hacker
> way.
It would be much better for you to fix or write code than throw around
childish insults at the authors of free programs.
-- g