RE: [OE-core] [PATCH 1/3] libgcc: Fix standalone target builds with usrmerge distro feature

[Peter Kjellerstedt <peter.kjellerstedt@axis.com>]

> -----Original Message-----
> From: Khem Raj <raj.khem@gmail.com>
> Sent: den 24 juli 2022 15:52
> To: Peter Kjellerstedt <peter.kjellerstedt@axis.com>
> Cc: openembedded-core@lists.openembedded.org
> Subject: Re: [OE-core] [PATCH 1/3] libgcc: Fix standalone target builds
> with usrmerge distro feature
> 
> Hi Peter
> 
> Thanks for the review.
> 
> On Sun, Jul 24, 2022 at 7:47 AM Peter Kjellerstedt
> <peter.kjellerstedt@axis.com> wrote:
> >
> > > -----Original Message-----
> > > From: openembedded-core@lists.openembedded.org <openembedded-
> core@lists.openembedded.org> On Behalf Of Khem Raj
> > > Sent: den 23 juli 2022 16:54
> > > To: openembedded-core@lists.openembedded.org
> > > Cc: Khem Raj <raj.khem@gmail.com>
> > > Subject: [OE-core] [PATCH 1/3] libgcc: Fix standalone target builds
> with usrmerge distro feature
> > >
> > > Ignore the rmdir cmd if using usrmerge distro feature since the
> > > intention is to delete /lib or /lib64 but not libdir under /usr and
> > > base_libdir = libdir when usrmerge is enabled in distro
> > >
> > > Signed-off-by: Khem Raj <raj.khem@gmail.com>
> > > ---
> > >  meta/recipes-devtools/gcc/libgcc-common.inc | 8 ++++++--
> > >  1 file changed, 6 insertions(+), 2 deletions(-)
> > >
> > > diff --git a/meta/recipes-devtools/gcc/libgcc-common.inc
> b/meta/recipes-devtools/gcc/libgcc-common.inc
> > > index 148c9f85a7c..9d4fd41f975 100644
> > > --- a/meta/recipes-devtools/gcc/libgcc-common.inc
> > > +++ b/meta/recipes-devtools/gcc/libgcc-common.inc
> > > @@ -44,10 +44,14 @@ do_install () {
> > >  }
> > >
> > >  do_install:append:libc-baremetal () {
> > > -     rmdir ${D}${base_libdir}
> > > +     if
> ${@bb.utils.contains('DISTRO_FEATURES','usrmerge','false','true',d)}; then
> >
> > Since the goal is to avoid deleting ${libdir} in case it happens to
> match
> > ${base_libdir}, I recommend to change the tests to instead be:
> >
> >         if [ "${base_libdir}" != "${libdir}" ]; then
> >
> 
> I don't see much benefit of doing this since our usecase is guarded by
> distro feature and I found it more readable

We already have corresponding tests in other places where we want to 
avoid removing ${libdir} if it happens to match ${base_libdir}, e.g., 
in the glibc-package.inc file.

The benefit I see with comparing the two variables rather than relying 
on the distro feature is that the test isn't really related to usrmerge, 
it only happens to coincide. I.e., if I would manually configure my 
system so that ${base_libdir} == ${libdir} without setting usrmerge, 
the test here would still do the right thing (things would probably 
break in other places though, but that is a different story).

> having said that I am fine to change it as well, other option would be
> to use rmdir --ignore-fail-on-non-empty and avoid
> the check but that will make it generic and we might not see some
> unwanted files sneaking into base_libdir so erroring out
> is a good check.
> 
> > > +             rmdir ${D}${base_libdir}
> > > +     fi
> > >  }
> > >  do_install:append:libc-newlib () {
> > > -     rmdir ${D}${base_libdir}
> > > +     if ${@bb.utils.contains('DISTRO_FEATURES','usrmerge','false','true',d)}; then
> > > +             rmdir ${D}${base_libdir}
> > > +     fi
> > >  }
> > >
> > >  # No rpm package is actually created but -dev depends on it, avoid dnf error
> > > --
> > > 2.37.1
> >
> > //Peter

//Peter