This is the mail archive of the
mailing list for the binutils project.
Re: -Bdirect linking patch
- From: michael meeks <michael dot meeks at novell dot com>
- To: Ulrich Drepper <drepper at redhat dot com>
- Cc: binutils <binutils at sources dot redhat dot com>
- Date: Fri, 28 Oct 2005 16:19:55 +0100
- Subject: Re: -Bdirect linking patch
- References: <email@example.com> <firstname.lastname@example.org>
- Reply-to: michael dot meeks at novell dot com
On Fri, 2005-10-28 at 07:30 -0700, Ulrich Drepper wrote:
> michael meeks wrote:
> > So - I have an evily hackish prototype for glibc & binutils that gives
> > me a really nice speedup with relocation processing. It's inspired by
> > the Solaris -Bdirect feature of the same name, and is implemented in the
> > same way - by adding a section containing a 16bit word per .dynsym entry
> > with an index into the libraries DT_NEEDED table, built at library link
> > time.
> Forget it, we have prelinking, which is much more efficient.
Prelinking doesn't work for dlopened libraries - which accounts for the
vast majority of OO.o code, and an increasing amount of eg. KDE & other
desktop code; and AFAICS it's rather difficult to improve that with
Furthermore -Bdirect linking requires no regular per-end-user
'prelinking' phase to cope with changed libraries. It is also (as can
perhaps be deduced from the patch) inordinately simple, so shouldn't
complicate & clutter the code too much.
The hope is that it will remove the anecdotal embarrassments that OO.o
starts quicker under Wine on the same Linux box than natively - due in
large part (I guess) to the linking burning nearly 3 seconds on my
2.8GHz machine. [ and of course faster yet on Win32 ].
Or am I missing something ?
email@example.com <><, Pseudo Engineer, itinerant idiot