fstat64 filesize problem

Kelledin kelledin+LFS at skarpsey.dyndns.org
Thu Feb 13 16:32:15 PST 2003


On Thursday 13 February 2003 05:02 pm, Greg Schafer wrote:
> On Thu, Feb 13, 2003 at 04:53:05PM -0600, Kelledin wrote:
> > On Thursday 13 February 2003 04:39 pm, Greg Schafer wrote:
> > > On Thu, Feb 13, 2003 at 04:26:13PM -0600, Kelledin wrote:
> > > > blowfish-crypt support, and I configure it with
> > > > --enable-kernel=2.4.0.
> > > >
> > > > BTW, do NOT use --enable-kernel=current, as it
> > > > introduced some obscure brokenness last time I tried it.
> > >
> > > This looks like FUD, What was the problem?
> >
> > When I did that with my somewhat modified 2.4.18 kernel on
> > x86, it produced an /sbin/ldconfig that would never add
> > anything to /etc/ld.so.cache.  Unfortunately, I only
> > discovered that when I got around to installing XFree86 and
> > added /usr/X11R6/lib to ld.so.conf--redoing the entire
> > system at that point was annoying, to say the least.
> >
> > Removing all my extra glibc patches and going back to
> > LFS-standard glibc (with --enable-kernel=current) didn't fix
> > it. Removing --enable-kernel completely, or paring it back
> > to --enable-kernel=2.4.0, fixed the problem.  I haven't hit
> > any other glibc brokenness since then.
> >
> > Perhaps it's some brokenness in glibc, or perhaps it's some
> > brokenness specifically in my modified 2.4.18.  Even if it
> > is kernel brokenness, though, it kind of highlights that
> > --enable-kernel=current is not always a "safe" choice.
>
> Ahh, did you configure for i386 instead of i686? Somethng like
> i386-pc-linux-gnu? If so, that is a known bug with glibc
> compiled with gcc-3.2.
>
> http://sources.redhat.com/ml/libc-alpha/2003-01/msg00308.html
>
> Fixed in glibc cvs, should be in glibc-2.3.2. This came up on
> lfs-support list only a few weeks ago.

Yup, that could very well be it (I compiled for 
i486-incept-linux-gnu, which is pretty close).  It might also 
explain why the problem mysteriously didn't come up on my Alpha 
box...

-- 
Kelledin
"If a server crashes in a server farm and no one pings it, does 
it still cost four figures to fix?"
-- 
Unsubscribe: send email to listar at linuxfromscratch.org
and put 'unsubscribe lfs-dev' in the subject header of the message



More information about the lfs-dev mailing list