[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Debian glibc symbol version stuff



Thomas Schwinge, le Thu 20 Oct 2011 11:49:38 +0200, a écrit :
> On Wed, 19 Oct 2011 00:10:52 +0200, Samuel Thibault <sthibault@debian.org> wrote:
> > Thomas Schwinge, le Tue 18 Oct 2011 10:42:00 +0200, a écrit :
> > > Help?  Is my package version ``wrong''?  Or are just some additional
> > > steps needed?
> > 
> > I don't see anything wrong here, I've no idea.
> 
> Hmm, I did a build afresh, and it happens again.  How do you do glibc
> development in this Debian build harness?

Just dpkg-buildpackage with no particular environment.

> Here are my steps.  Please tell me what you're doing differently.
> 
>     $ export LC_ALL=C

Same.

> Otherwise the testsuite/make error messages will be different from those
> stored in the debian/testsuite-checking/expected-results-* files.
> 
>     $ apt-get source eglibc
>     $ cd eglibc-2.13/

Same.

> Apply patches (instead of having it done during the build).  Then you can
> edit files manually.
> 
>     $ debian/rules patch

Same.

> Do the following changes:
> 
>     $ sed '/^ --/q' < debian/changelog
>     eglibc (2.13-21+ts.0) unstable; urgency=low
>     
>       * debian/rules: Build with GCC 4.6.
>       * Revert hurd-i386/submitted-init-first.diff and
>         hurd-i386/unsubmitted-gcc-4.1-init-first.diff, and replace with current
>         t/init-first.c patch.
>       * Apply current hurd-i386/submitted-setresid.diff patch.
>       * sysdeps/mach/hurd/fork.c: Deactivate mach_port_mod_refs calls for
>         thread_refs and sigthread_refs.
>       * debian/sysdeps/hurd-i386.mk: Disable libc0.3-dev_extra_pkg_install.
>     
>      -- Thomas Schwinge <thomas@schwinge.name>  Mon, 17 Oct 2011 12:47:11 +0200

I usually rather use ~0, but it shouldn't matter.

> If building with EGLIBC_PASSES=libc (more specifically, without xen)

I don't use EGLIBC_PASSES, but it shouldn't matter either.

> , the
> libc0.3-dev_extra_pkg_install rule in debian/sysdeps/hurd-i386.mk will
> fail.  (Same for libc6-dev_extra_pkg_install in debian/sysdeps/i386.mk,
> for example.)  Why is this special handling only done for xen, but not
> for i686?

Historically because it's done that way in linux-i386. I don't know the
real reason.

> (Learning from previous pain, I saved a tarball at this point.)
> 
>     $ debian/rules build EGLIBC_PASSES=libc

I usually just use dpkg-buildpackage, but it should be equivalent.

> (I left out -j2 parallel building this time.)
> 
> (Learning from previous pain, I saved a tarball at this point.)
> 
>     $ fakeroot debian/rules binary EGLIBC_PASSES=libc

Same.

Samuel


Reply to: