GCC Testsuite failures was Re: GCC-3.4.2 PCH failures (segmentation faults)

Ken Moffat ken at kenmoffat.uklinux.net
Fri Nov 12 09:04:39 PST 2004


On Fri, 12 Nov 2004, Ken Moffat wrote:

>
>  Well, my problems are probably very different from Ilja's - thanks to
> his comments on the testsuite logs I can now see that most of the
> failures follow error messages of the form
>
> sh: /usr/local/bin/stty: No such file or directory
>
> (note the /usr/local).  So far I've confirmed that my script set the
> prefix to /usr, will dig around a bit, it's probably something stupid
> I've done.
>

And yes, somebody please pass me a brown paper bag.  Looks like a PATH
problem - the idiot sitting at the keyboard here thought it might be a
good idea to use a fairly recent compiler, which happened to be in
/usr/local, to start off chapter 5 because the host is debian and has
sometimes had quite old stuff in it.

A good compiler and associated binutils in /usr/local and on the PATH -
seems fine.  A version of coreutils from who-knows-when sitting in
/usr/local and on the PATH - not so fine.  Specifically, in chapter 5,
expect hard-codes -DSTTY_BIN=\"/usr/local/bin/stty\" into its
compilation.  In chapter 5 this is of course fine, but in chapter 6...

Posting so that maybe others will find this answer in the future.

Ken
-- 
 das eine Mal als Tragödie, das andere Mal als Farce




More information about the lfs-support mailing list