Re: debuginfo sub-packages for the kernel
Hi Laura,
Sorry for the late reply.
I am currently on vacation with limited internet access.
Back in 1.5 weeks.
On Tue, Aug 15, 2017 at 05:12:46PM -0700, Laura Abbott wrote:
I started playing around with having RPM automatically generate the
debuginfo subpackages per
https://fedoraproject.org/wiki/Changes/SubpackageAndSourceDebuginfo
Thanks for looking at this!
Because the kernel is a continual source of exceptions, the existing
code doesn't handle the case where the debuginfo and file may have
different names.
Yes, the kernel spec is "special" in lots of ways :)
The kernel modules are compressed at the install stage so foo.ko
becomes foo.ko.xz. The debuginfo is generated as foo.ko.debug.
The existing code to split debugfiles.list can't detect this so
those files end up as unpackaged. A similar problem happens with
the vmlinux for the main kernel.
Is this an issue that can reasonably be fixed or worked around to
support the debuginfo subpackages for the kernel? I want to make
sure I'm not digging myself into a hole before I spend more time
on this.
In the new code debug files processing is split partly between
find-debuginfo.sh (based on the files in the buildsubdir) and
rpmbuild post %install (based on the actual %files list). This
causes an issue for the current way the kernel compresses the
kernel modules, which is ran through __modsign_install_post after
the find-debuginfo.sh invocation, but before the %files list is
created.
I think for the compressed ELF files it might be best if rpm handled
this directly. It is somewhat kernel specific, but could be useful
in general. But I am not sure yet what the best way to do that in rpm
would be. Either you would be able to install ELF files compressed an
rpm would just work on compressed ELF files (but this might mean rpm
would have to decompress and recompress them). Or there should be some
way that you can just install the files normally and then indicate in
the %files list that some files should really be compressed on disk
(then they would only be compressed at the very end by rpm).
I'll file a bug report for that against rpm and will look into it when
I return from vacation.
Cheers,
Mark