This is the mail archive of the
java@gcc.gnu.org
mailing list for the Java project.
Re: dynamic library cost (was RE: libtool, java woes)
- To: "Boehm, Hans" <hans_boehm at hp dot com>
- Subject: Re: dynamic library cost (was RE: libtool, java woes)
- From: Jakub Jelinek <jakub at redhat dot com>
- Date: Tue, 10 Apr 2001 23:52:53 +0200
- Cc: "'Alexandre Oliva'" <aoliva at redhat dot com>, tromey at redhat dot com, Jeff Sturm <jsturm at one-point dot com>, java at gcc dot gnu dot org, gcc at gcc dot gnu dot org
- References: <140D21516EC2D3119EE7009027876644049B5D4C@hplex1.hpl.hp.com>
- Reply-To: Jakub Jelinek <jakub at redhat dot com>
On Tue, Apr 10, 2001 at 09:15:40AM -0700, Boehm, Hans wrote:
> > From: Alexandre Oliva [mailto:aoliva@redhat.com]
> >
> > We either need libsupc++ to be a shared library, that both libstdc++
> > and libgcj depend on, or have libgcj depend on libstdc++.
> >
> One thing to keep in mind with all of this:
>
> I'm seeing significant overheads as a result of dynamic library calls. On a
> PII/300 machine, a (single-threaded) loop containing only free(malloc(8))
> runs more than 20% faster when it's linked statically. This is similar on
> an Itanium machine. I'm not sure how significant that is for typical C
> programs, nor how frequently something like libsupc++ is called. I believe
> it is currently very significant for libgcj and calls to the garbage
> collector library from libgcj.
Not to mention the dynamic linking overhead at startup (or dlopen) time
which every additional shared library adds.
Jakub