The only unique thing about my userland would be the path
by which it was built - that is, I've long forgotten what
I bootstrapped the box to -current from, and then at random
intervals, I have cvs updated and rebuilt kernel/userland.

Thinking I may have caught a bad libbfd, I pulled over the
formal 1.6 release and borrowed the libbfd.so.4.0 from that.
This alas did not appear to help.  I'd also swap out the libc,
but my July 29th copy is already past the copy with 1.6 release.

No big deal - just thought I'd post something in case anyone
else runs into this... or had and knew what boo boo I have made.

Cheers,

-- Ken

On Mon, Sep 23, 2002 at 03:53:22PM -0400, Lord Isildur wrote:
> is there anything unique about your userland? ive not seen this before 
> (or at least ive never _noticed_ it.. :) 
> 
> On Mon, 23 Sep 2002, Ken Wellsch wrote:
> 
> > While trying to refresh a Dec 5k/150 box from a current build
> > circa July 29th, I found I was unable to get past building
> > libc.  The summary is as follows, it appears that some of
> > the object files created are sparse (ls -l shows megabytes,
> > ls -s shows just a handful of blocks), and when "ar" tries
> > to build libc.a I run out of space.  I didn't see anything
> > like this on vax or sparc, both also build at the same time
> > back at the end of July.  So I don't know whether this is
> > just a pmax issue, or just my own issue based upon my unique
> > userland.  Cheers,  -- Ken Wellsch