[Bug 1653] MPFR spec URL wrong

[bugzilla-daemon at rtems.org]

https://www.rtems.org/bugzilla/show_bug.cgi?id=1653

--- Comment #3 from Ralf Corsepius <ralf.corsepius at rtems.org> 2010-08-05 00:51:44 CDT ---
(In reply to comment #2)
> (In reply to comment #1)
> > (In reply to comment #0)
> > > MPFR has a new release

> > > http://www.mpfr.org/mpfr-2.4.2/mpfr-2.4.2.tar.bz2
> > OK, will update the URL.
I just changed CVS-HEAD to using
Source60: http://www.mpfr.org/mpfr-%{mpfr_version}/mpfr-%{mpfr_version}.tar.bz2

> > > <side-issue>
> > > 
> > > I have also noticed:
> > > 
> > >  http://www.rtems.org/ftp/pub/rtems/SOURCES/4.11
> > > 
> > > does not contain some of these sources:
> > > 
> > >  http://www.multiprecision.org/mpc/download/mpc-0.8.1.tar.gz
> > >  http://www.mpfr.org/mpfr-current/mpfr-2.4.2.tar.bz2
> > >  ftp://ftp.gnu.org/gnu/gmp/gmp-4.3.2.tar.bz2
> > >  ftp://sources.redhat.com/pub/newlib/newlib-1.18.0.tar.gz
> > >  ftp://ftp.gnu.org/gnu/gcc/gcc-4.5.1/gcc-core-4.5.1.tar.bz2
> > >  ftp://ftp.gnu.org/gnu/gcc/gcc-4.5.1/gcc-g++-4.5.1.tar.bz2
> > > 
> > > plus patches. I am willing to automate the updating of these files based on the
> > > references in the spec files and to remove files after 6 months (or how-ever
> > > long) if not referenced in spec files. Please let me know if it is ok for me to
> > > do this.
> > > 
> > > </side-issues>
> > Not OK with me.
> > 
> > I am populating this directory from the contents of the rpms. Currently this is
> > done manually after a built-spin has completed, which usually takes 2 days and
> > thus occasionally causes me to forget about it.
> 
> Ah yes this is more correct but do you think they would not be the same. We
> should provide MD5 checksums on these files.
Whatfor? The srpms do contain sha256 checksums on them.

> > 
> > I.e. the appropriate step would be to extend the scripts I am using to maintain
> > the rpm repositories to automatically populate/cleanup these directories. So
> > far this hasn't happened.
> 
> I am more than happy for you to handle this. If this directory does not have
> the source I go to the location in the spec file.
The SRPMS directory would be correct.

> The spec files reference some source packages that as far as I can see you do
> not use such as mpfr etc, and I assume this is due to the host having a
> suitable package installed so no need for the source. Is this the case ?
The logic inside of the specs is supposed to "use the version from the distro,
if it is sufficient for GCC and fall back to in-source-treebuilding if it is
not".

In cases "a distro's version is sufficent for GCC" side-effects are:
* gcc is dynamically linked against the library (vs. statically linked in case
of in-sourcetree-building)
* the rpms are smaller (do not include the library referenced)
* the srpm for this distro does not contain the tarball (srpms are smaller).
...

All in all this means, the newer and more current a distro is, the more
distro-provided libraries will be used. Conversely this also means: The older
and outdated a distro is, the more of these in-sourcetree-built tarballs will
be used.

Example:
Old/outdated OS (CentOS 5/RHEL 5):
# rpm -qlp
centos-5-i386-rtems4.11/rtems-4.11-arm-rtems4.11-gcc-4.5.1-3.el5.src.rpm 
gcc-core-4.5.1-rtems4.11-20100726.diff
gcc-core-4.5.1.tar.bz2
gcc-g++-4.5.1.tar.bz2
gmp-4.3.2.tar.bz2
mpc-0.8.1.tar.gz
mpfr-2.4.2.tar.bz2
newlib-1.18.0-rtems4.11-20100804.diff
newlib-1.18.0.tar.gz
rtems-4.11-arm-rtems4.11-gcc.spec

New/current OS (Fedora 13):
# rpm -qlp
fedora-13-i386-rtems4.11/rtems-4.11-arm-rtems4.11-gcc-4.5.1-3.fc13.src.rpm 
gcc-core-4.5.1-rtems4.11-20100726.diff
gcc-core-4.5.1.tar.bz2
gcc-g++-4.5.1.tar.bz2
newlib-1.18.0-rtems4.11-20100804.diff
newlib-1.18.0.tar.gz
rtems-4.11-arm-rtems4.11-gcc.spec

> Please note, I have added one or more of the other sources before you raised
> you  objection. I can remove if you wish.
Pardon, but I do not understand this remark - Could you elaborate?

-- 
Configure bugmail: https://www.rtems.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are watching all bug changes.