Open Issues

threads = yes is set in debian/sysdeps/ and debian/sysdeps/, debian/sysdeps/ set to no. But this is only read in debian/rules for deciding some nscd package issue?

debian/sysdeps/'s libc_extra_install for check with GCC configuration.

Could add a toggle to $(stamp)build_% in debian/rules.d/ to skip locale stuff.


IRC, freenode, #hurd, 2013-08-28

<youpi> uh, the i686 profiles have much more progression than i386
<youpi> it seems they don't actually run these
<pinotree> youpi: what do you mean with "we don't run those"?
<pinotree> iirc there are three build profiles done, but there are 4
  regression test files
<youpi> yes, but some failing tests are not run in the three build profiles
<youpi> even if they are built for all of them
<pinotree> not even run? which ones?
<youpi> see for instance test-ifloat.out
<youpi> test-ifloat is built in all profiles, but only run in the libc one
<pinotree> don't have a glibc built tree around atm, sorry :/
<youpi> perhaps because glibc thinks it's not useful to run it again if it
  fails on i386
<youpi> you can check the logs
<pinotree> do you think glibc's build system is that smart? :)
<pinotree> all the builds are done in separate builddirs, so theorically
  they should not touch each other...
<youpi> yes
<youpi> that's why I'm surprised
<pinotree> could it be they get not run in optimized/particular builds?
<pinotree> what about linux/kfreebsd i386?
<youpi> I don't see what  makes them not run
<youpi> or at least be treated particularly by th eMakefile
<youpi> not run on kfreebsd either
<youpi> pinotree: also, most of the tests now working have been marked as
  failing by your patches for 2.17, would it be possible to retry them on
  the box you used at that time?
<pinotree> that's the vm on my machine
<youpi> which kind of vm?
<youpi> kvm?
<pinotree> y
<youpi> they are working here
<youpi> with kvm


Run debian/rules patch to apply patches (instead of having it done during the build). Then you can edit files manually.

Several passes: libc, i686, xen; EGLIBC_PASSES='libc i686', etc.

If building with EGLIBC_PASSES=libc (more specifically, without xen), the libc0.3-dev_extra_pkg_install rule in debian/sysdeps/ will fail. (Same for libc6-dev_extra_pkg_install in debian/sysdeps/, for example.) Why is this special handling only done for xen, but not for i686?

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

Do export LC_ALL=C before building, otherwise the testsuite/make error messages will be different from those stored in the debian/testsuite-checking/expected-results-* files, resulting in a spurious build failure.

Run debian/rules build-arch DEB_BUILD_OPTIONS=parallel=2 [EGLIBC_PASSES=...] to build (or build instead of build-arch to build the arch-independent stuff, too). Can interrupt with C-c during locale stuff or testsuite if only interested in the build tree.

Run fakeroot debian/rules binary DEB_BUILD_OPTIONS=parallel=2 [EGLIBC_PASSES=...] to build Debian packages or binary-arch for just the architecture-dependent ones.

The latter two steps can also be combined as dpkg-buildpackage -R'debian/rules EGLIBC_PASSES=libc' -nc -b -uc. -nc will prevent the clean step which would first try to un-patch, which may conflict if you have done any edits apter applying patches.

If the Debian symbol versioning file is not up to date and the build of Debian packages fails due to this, putting DPKG_GENSYMBOLS_CHECK_LEVEL=0 in the environment ``helps''; see man dpkg-gensymbols.

IRC, freenode, #hurd, 2013-07-01

<braunr> something seems to have changed with regard to patch handling in
  eglibc 2.17
<braunr> pinotree: when i add a patch to series and use dpkg-buildpackage,
  i'm told there are local modifications and the build stops :/
<braunr> any idea what i'm doing wrong ?
<pinotree> which steps do you do?
<braunr> i extract the sources, copy the patch to debian/patches/hurd-i386,
  add the appropriate line to debian/patches/series, call dch -i, then
<pinotree> eglibc is a "3.0 (quilt)" format source package
<pinotree> this means its default patches are in a quilt-style system, and
  they are applied on extraction
<braunr> ok
<braunr> and it can't detect new patches ?
<pinotree> so if you add a new patch to the global serie, you have to push
  it manually
<braunr> i have to revert them all ?
<braunr> ok
<braunr> how do i do that ?
<pinotree> quilt push -a
<braunr> ok
<braunr> thanks
<pinotree> remember to do that before starting the build, since the rest
  assumes the quilt-style patches are fully applied
<bddebian> No push applies them, quilt pop -a reverts them
<pinotree> yeah, and he has to push the new over the dpkg-applied ones
<bddebian> Oh, aye
<braunr> does quilt change series ?
<pinotree> no
<braunr> ok
<pinotree> i mean, some commands do that
<braunr> so i do everything i did, with an additional push, right ?
<pinotree> ok, screw me, i didn't get your question above :P
<braunr> does that change your answer ?
<pinotree> <braunr> does quilt change series ?
<braunr> yes
<pinotree> if you import or create a new patch, it changes series indeed
<braunr> ok
<pinotree> push or pop of patches does not
<braunr> i'm doing it wron
<braunr> g
<pinotree> btw, in a quilt patch stack you can easily import a new patch
  using the import command
<pinotree> so for example you could do
<pinotree> apt-get source eglibc # or get it somehow else
<pinotree> cd eglibc-*
<pinotree> quilt import /location/of/my/patch
<pinotree> quilt push # now your patch is applied
<braunr> ah thanks
<pinotree> dpkg-buildpackage as usual
<braunr> that's what i was looking for
<bddebian> quilt new adds a new entry in series
<pinotree> y
<bddebian> or import, aye
<pinotree> braunr: if you want to learn quilt, a very good doc is its own,
  eg /usr/share/doc/quilt/quilt.txt.gz
* bddebian has never actually used import
<braunr> ok
<pinotree> it is basically a simple stack of patches

<youpi> braunr: yes, patch handling is a bit different
<youpi> the arch-independant patches are applied by dpkg-source -x
<youpi> and the arch-dependent patches are applied during build