ACK/Cmnt: [PATCH 0/2][F] Build linux-libc-dev for riscv64 from main kernel package

Seth Forshee seth.forshee at canonical.com
Thu Jul 9 16:12:40 UTC 2020


On Thu, Jul 09, 2020 at 10:18:31AM +0200, Stefan Bader wrote:
> On 08.07.20 18:27, Seth Forshee wrote:
> > BugLink: https://bugs.launchpad.net/bugs/1886188
> > 
> > SRU Justification
> > 
> > Impact: Building linux-libc-dev for riscv64 from the linux-riscv package means
> > that it can have a different version from the package on other architectures.
> > This makes it uninstallable for cross building.
> > 
> > Fix: Move the riscv64 linux-libc-dev build back to the main kernel package so
> > that the version will match.
> > 
> > Test Case: See bug description.
> > 
> > Regression Potential: This could cause trouble if the main kernel package
> > version is behind the linux-riscv package version, and may require artificially
> > inflating the version of the linux package. Currently this is not necessary,
> > but it needs to be kept in mind when preparing packages for upload with the
> > fixes applied.
> > 
> > Note that these changes should only be applied along with the
> > corresponding changes for linux-riscv, sent separately.
> > 
> > Thanks,
> > Seth
> > 
> > ---
> > 
> > Seth Forshee (2):
> >   UBUNTU: [Packaging] Produce linux-libc-deb package for riscv64
> >   UBUNTU: [Debian] Disallow building linux-libc-dev from linux-riscv
> > 
> >  debian.master/control.d/linux-libc-dev.stub |  2 +-
> >  debian.master/rules.d/riscv64.mk            | 21 +++++++++++++++++++++
> >  debian/rules.d/2-binary-arch.mk             |  2 +-
> >  3 files changed, 23 insertions(+), 2 deletions(-)
> >  create mode 100644 debian.master/rules.d/riscv64.mk
> > 
> It would have been better to group the change to linux-riscv under this thread
> as well. Since all 3 must be applied together to avoid breakage.
> But anyhow...

Personally I tend to dislike the patch "series" which target multiple
trees. Understood that it would capture the interdependency between
these two sets of patches more clearly than the comment I included in
the cover letter, but too bad that there isn't some nicer way to do it.



More information about the kernel-team mailing list