[OE-core] [PATCH 04/10] lib/oe/package_manager: turn postinst_intercept warnings into failures for nativesdk

[Joshua Watt]

On Fri, 2019-01-25 at 13:13 +0100, Alexander Kanavin wrote:
> There is a similar issue in multilib for target packages (a warning
> because qemu usermode support is missing):
> https://autobuilder.yoctoproject.org/typhoon/#/builders/44/builds/214/steps/7/logs/warnings
> 
> I'm as well not sure how to handle this:
> 1. Keep these two warnings as warnings so users see them in their
> local builds, but
>  add some kind of string pattern to AB so it excludes missing qemu
> support and missing wine support warnings.
> 
> 2. Turn them into notes, which means they will go to the logs only,
> and users are not going to see them ever. On the other hand, the
> postinsts tend to
> create things like font caches and similar, which should not
> generally
> be needed at compile time (which is what SDKs are for).
> 
> I'm leaning towards option two.

I was leaning toward keeping them warnings and then suppressing them in
the cases where we know they don't work and are unnecessary (perhaps
with a variable like POSTINST_INTERCEPT_${PN} = "0" ?)

That way if a user enables a package where it's not going to run and we
haven't previously evaluated it to be OK, they will at least see the
warning. I naively think this might help reduce the number of bugs
filed, or at least make them easier to triage :)


> 
> Alex
> 
> On Fri, 25 Jan 2019 at 00:18, Joshua Watt <jpewhacker at gmail.com>
> wrote:
> > 
> > 
> > On Thu, Jan 24, 2019, 4:43 PM Richard Purdie <
> > richard.purdie at linuxfoundation.org wrote:
> > > On Wed, 2019-01-23 at 17:17 +0100, Alexander Kanavin wrote:
> > > > The few cases where they failed should be now all fixed. The
> > > > only allowed
> > > > exception is when building mingw32 SDKs, as there is currently
> > > > no support for running
> > > > postinst_intercepts through wine.
> > > > 
> > > > Signed-off-by: Alexander Kanavin <alex.kanavin at gmail.com>
> > > > ---
> > > >  meta/lib/oe/package_manager.py | 7 ++++++-
> > > >  1 file changed, 6 insertions(+), 1 deletion(-)
> > > > 
> > > > diff --git a/meta/lib/oe/package_manager.py
> > > > b/meta/lib/oe/package_manager.py
> > > > index 1087144d47f..f26f597d03d 100644
> > > > --- a/meta/lib/oe/package_manager.py
> > > > +++ b/meta/lib/oe/package_manager.py
> > > > @@ -439,6 +439,11 @@ class PackageManager(object,
> > > > metaclass=ABCMeta):
> > > >                  self._postpone_to_first_boot(script_full)
> > > >                  continue
> > > > 
> > > > +            if populate_sdk == 'host' and
> > > > self.d.getVar('SDK_OS') == 'mingw32':
> > > > +                bb.warn("The postinstall intercept hook '%s'
> > > > could not be executed due to missing wine support, details in
> > > > %s/log.do_%s"
> > > > +                                % (script, self.d.getVar('T'),
> > > > self.d.getVar('BB_CURRENTTASK')))
> > > > +                continue
> > > > +
> > > >              bb.note("> Executing %s intercept ..." % script)
> > > > 
> > > >              try:
> > > > @@ -447,7 +452,7 @@ class PackageManager(object,
> > > > metaclass=ABCMeta):
> > > >              except subprocess.CalledProcessError as e:
> > > >                  bb.note("Exit code %d. Output:\n%s" %
> > > > (e.returncode, e.output.decode("utf-8")))
> > > >                  if populate_sdk == 'host':
> > > > -                    bb.warn("The postinstall intercept hook
> > > > '%s' failed, details in %s/log.do_%s" % (script,
> > > > self.d.getVar('T'), self.d.getVar('BB_CURRENTTASK')))
> > > > +                    bb.fatal("The postinstall intercept hook
> > > > '%s' failed, details in %s/log.do_%s" % (script,
> > > > self.d.getVar('T'), self.d.getVar('BB_CURRENTTASK')))
> > > >                  elif populate_sdk == 'target':
> > > >                      if "qemuwrapper: qemu usermode is not
> > > > supported" in e.output.decode("utf-8"):
> > > >                          bb.warn("The postinstall intercept
> > > > hook '%s' could not be executed due to missing qemu usermode
> > > > support, details in %s/log.do_%s"
> > > 
> > > I'm ok with this but could we change the bb.warn to something
> > > less
> > > important like a note?
> > > 
> > > The reason I say this is that I'd like to get the autobuilder to
> > > the
> > > point where its warning free. This warning is the only reason
> > > meta-
> > > mingw shows warnings now. Either we fix them or if we can't, they
> > > probably aren't warnings.
> > > 
> > > What should we do here?
> > > 
> > > (also cc'd Joshua)
> > 
> > I think the one that fails currently is updating the GIO module
> > cache. I'm not sure if this is even necessary for mingw (my guess
> > is no, since this has never worked and no one has complained). We
> > probably just need to disable it in mingw, but when I looked very
> > briefly a while ago, I didn't see a straight forward way of doing
> > that.
> > 
> > See https://bugzilla.yoctoproject.org/show_bug.cgi?id=12962
> > 
> > > Cheers,
> > > 
> > > Richard
> > > 
-- 
Joshua Watt <JPEWhacker at gmail.com>