Re: contrib/gcc_update bug?

[Bruce Korb <bkorb at sco dot COM>]

Richard Earnshaw wrote:
> 
> This can happen if gcc_update is itself updated and the change includes an
> alteration in the list of things to touch.  Maybe gcc_update could be made
> intelligent enough to detect this; I don't know.  In the mean time, if
> gcc_update is changed by an update, it is best to run it again.

Very plausible.  Not true.  Remember, I did a complete pull and *then*
ran the update.  The likely cause is that gcc_update is *missing*
a file name that needs touching.  This be a bug.  :)

> > My bootstrap failed as in below.  This was immediately after
> > a complete refresh this morning, followed by running:
> >
> >   cd egcs ; sh contrib/gcc_update --touch
> >
> > I don't think we should be trying to update anything in the
> > source tree, yes?
> >
> >
> > gcc  -DIN_GCC    -g -O2   -DHAVE_CONFIG_H  -I. -I../../egcs/gcc \
> >   -I../../egcs/gcc/config -I../../egcs/gcc/../include -I./f \
> >            `echo ../../egcs/gcc/f/intdoc.c | sed 's,^\./,,'` -o f/intdoc; \
> >           f/intdoc > ../../egcs/gcc/f/intdoc.texi; \
> >           rm f/intdoc f/ansify f/intdoc.h0; \
> >         else true; fi
> > sh: ../../egcs/gcc/f/intdoc.texi: cannot create
> > *** Error code 1 (bu21)
> > *** Error code 1 (bu21)
> >
> > $ chmod +w ../egcs/gcc/f/intdoc.texi