Re: [PATCH 8/8] selftests: livepatch: functions.sh: Extend check for taint flag kernel message

[Marcos Paulo de Souza <mpdesouza@suse.com>]

On Fri, 2026-03-20 at 14:26 +0100, Petr Mladek wrote:
> On Fri 2026-03-13 17:58:39, Marcos Paulo de Souza wrote:
> > On SLE kernels there is a warning when a livepatch is disabled:
> >   livepatch: attempt to disable live patch test_klp_livepatch,
> > setting
> >   NO_SUPPORT taint flag
> > 
> > Extend lightly the detection of messages when a livepatch is
> > disabled
> > to cover this case as well.
> > 
> > Signed-off-by: Marcos Paulo de Souza <mpdesouza@suse.com>
> > ---
> >  tools/testing/selftests/livepatch/functions.sh | 2 +-
> >  1 file changed, 1 insertion(+), 1 deletion(-)
> > 
> > diff --git a/tools/testing/selftests/livepatch/functions.sh
> > b/tools/testing/selftests/livepatch/functions.sh
> > index 781346d6e94e0..73a1d4e6acaeb 100644
> > --- a/tools/testing/selftests/livepatch/functions.sh
> > +++ b/tools/testing/selftests/livepatch/functions.sh
> > @@ -324,7 +324,7 @@ function check_result {
> >  	# - filter out dmesg timestamp prefixes
> >  	result=$(dmesg | awk -v last_dmesg="$LAST_DMESG" 'p; $0 ==
> > last_dmesg { p=1 }' | \
> >  		 grep -e 'livepatch:' -e 'test_klp' | \
> > -		 grep -v '\(tainting\|taints\) kernel' | \
> > +		 grep -v '\(tainting\|taints\|taint\)
> > \(kernel\|flag\)' | \
> >  		 sed 's/^\[[ 0-9.]*\] //' | \
> >  		 sed 's/^\[[ ]*[CT][0-9]*\] //')
> 
> With the upstream maintainer hat on:
> 
> I am afraid that we could not take this. It is needed only because
> of another out-of-tree patch. It does not describe the upstream
> behavior. It might even hide problems.
> 
> We should maintain a SUSE-specific patch against the selftests
> as a counter-part for the patch adding the tainting.
> 
> Or we could try to upstream the patch which adds the tainting.
> Well, we might want to limit the tainting only to livepatches
> with callbacks or shadow variables. IMHO, only these features
> are source of potential problems.

TBH I wasn't expecting this patch to be merged, but I send it anyway
since the change wasn't big. But I agree, we can drop this one from the
patchset.

> 
> Best Rerards,
> Petr