On Tue, 12 Apr 2022 13:49:21 +0200
Nico Boehr <nrb@xxxxxxxxxxxxx> wrote:
> On Mon, 2022-04-11 at 14:49 +0200, Claudio Imbrenda wrote:
> [...]
> > > diff --git a/s390x/selftest-migration.c b/s390x/selftest-
> > > migration.c
> [...]
> > > +int main(void)
> > > +{
> > > + /* don't say migrate here otherwise we will migrate right
> > > away */
> > > + report_prefix_push("selftest migration");
> > > +
> > > + /* ask migrate_cmd to migrate (it listens for 'migrate') */
> > > + puts("Please migrate me\n");
> > > +
> > > + /* wait for migration to finish, we will read a newline */
> > > + (void)getchar();
> >
> > how hard would it be to actually check that you got the newline?
>
> It would be simple. I decided for ignoring what we actually read
> because that's what ARM and PPC do.
oh, then it's fine as it is
>
> But I am also OK checking we really read a newline. What would you
> suggest to do if we read something that's not a newline? Read again
> until we actually do get a newline?
I was more thinking that it's a failure, but see the comment above
