[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
rtems-980618 / posix's limits.h
- Date: Thu, 25 Jun 1998 13:44:22 +0200
- From: corsepiu at faw.uni-ulm.de (Ralf Corsepius)
- Subject: rtems-980618 / posix's limits.h
Joel Sherrill wrote:
> This is a common problem. It usually indicates that you are using the
> default gcc limits.h -- not the one installed after the C library is
> available.
Yep, it is limits.h from newlib.
> Here is the section from the "bit" script which attempts to do
> this:
I did a full rebuilt from snapshot's of egcs, gas, newlib (1.8.0) without using
"one-egcs" (my script doesn't work anymore for gas/egcs snapshots, because of
internationalization and egcs' Makefile probs) or "bit".
I have never used "bit" before, hence I am quite surprised that building rtems
has been working for me all the time until now. I suppose something in rtems must
have changed (presumably posix/base/Makefile, there is a comment mentioning
limits.h somewhere).
> # Regenerate (1) limits.h now that we have that the include_next will
> work,
> # and (2) libgcc.a now that we have a C Library installed
> cd gcc
> rm -f xlimits.h stmp-multilib
> find . -name "*.a" -print | xargs -e rm -f
> gmake CC=gcc CFLAGS="-O2" LANGUAGES="c c++" all install
> check_fatal $? "**** FAILED **** libgcc.a/limits.h rebuild step"
>
> Perhaps this does not work with a newer gcc snapshot. Or you built the
> toolset without bit and left this step out.
I didn't even know about this step.
Ralf.
--
Ralf Corsepius
Forschungsinstitut fuer Anwendungsorientierte Wissensverarbeitung (FAW)
Helmholtzstr. 16, 89089 Ulm, Germany Tel: +49/731/501-8690
mailto:corsepiu at faw.uni-ulm.de FAX: +49/731/501-999
http://www.faw.uni-ulm.de