ACK/CMNT: [PATCH 2/2][SRU][E] UBUNTU: [Debian] final-checks -- Get arch list from debian/control

[Seth Forshee]

On Mon, Sep 30, 2019 at 11:13:06AM -0500, Tyler Hicks wrote:
> On 2019-09-30 09:47:45, Seth Forshee wrote:
> > BugLink: https://bugs.launchpad.net/bugs/1845714
> > 
> > Getting the list of architectures from kernelconfig means we
> > can't keep i386 in the list for updating configs. Instead get the
> > list from the control file. This means that the finalchecks
> > target needs to depend on debian/control.
> > 
> > Signed-off-by: Seth Forshee <seth.forshee at canonical.com>
> > ---
> >  debian/rules.d/1-maintainer.mk   | 2 +-
> >  debian/scripts/misc/final-checks | 1 +
> >  2 files changed, 2 insertions(+), 1 deletion(-)
> > 
> > diff --git a/debian/rules.d/1-maintainer.mk b/debian/rules.d/1-maintainer.mk
> > index ae5d5cefc742..828e47bc1330 100644
> > --- a/debian/rules.d/1-maintainer.mk
> > +++ b/debian/rules.d/1-maintainer.mk
> > @@ -126,7 +126,7 @@ autoreconstruct:
> >  		$(DROOT)/scripts/misc/gen-auto-reconstruct $(upstream_tag) $(DEBIAN)/reconstruct $(DROOT)/source/options; \
> >  	fi
> >  
> > -finalchecks:
> > +finalchecks: debian/control
> >  	$(DROOT)/scripts/misc/final-checks "$(DEBIAN)" "$(prev_fullver)"
> >  
> >  diffupstream:
> > diff --git a/debian/scripts/misc/final-checks b/debian/scripts/misc/final-checks
> > index bd204b2bc9e2..3ffdde84f076 100755
> > --- a/debian/scripts/misc/final-checks
> > +++ b/debian/scripts/misc/final-checks
> > @@ -5,6 +5,7 @@ abi="$2"
> >  abi=${abi%~*}
> >  
> >  . "$debian/etc/kernelconfig"
> > +archs=$( awk '/^Architecture:/ { $1=""; for (i=1; i<=NF; i++) { if ($i != "all") { print $i }}}' debian/control | sort -u)
> 
> I think the presence of "any" also needs to be filtered out. Our current
> debian/control doesn't contain that value in any of the "Architecture:"
> lines so it isn't strictly required right now but would be protection
> for future changes that could add an "any".
> 
> With or without that change,
> 
>  Acked-by: Tyler Hicks <tyhicks at canonical.com>

Good point. I'll add that in a later patch, in the interest of getting
kernels uploaded asap.

Thanks,
Seth