Re: [hurd,commited] hurd TIOCFLUSH: Cope BSD 4.1 semantic

[Samuel Thibault]

Andreas Schwab, le lun. 01 févr. 2021 19:55:05 +0100, a ecrit:
> On Feb 01 2021, Samuel Thibault wrote:
> 
> > BSD 4.1 did not have an argument for TIOCFLUSH, BSD 4.2 added it. There
> > are still a lot of applications out there that pass a NULL argument to
> > TIOCFLUSH, so we should rather cope with it.
> > ---
> >  sysdeps/mach/hurd/ioctl.c | 4 ++++
> >  1 file changed, 4 insertions(+)
> >
> > diff --git a/sysdeps/mach/hurd/ioctl.c b/sysdeps/mach/hurd/ioctl.c
> > index 662e8ee919..0d5bc176d1 100644
> > --- a/sysdeps/mach/hurd/ioctl.c
> > +++ b/sysdeps/mach/hurd/ioctl.c
> > @@ -113,6 +113,10 @@ __ioctl (int fd, unsigned long int request, ...)
> >       /* We don't want to advance ARG since it will be used to copy out
> >          too if IOC_OUT is also set.  */
> >       void *argptr = arg;
> > +     int zero = 0;
> > +
> > +     if (request == TIOCFLUSH && !arg)
> > +       arg = &zero;
> 
> s/arg/argptr/.

Argl, yes, that was so in my actual test tree (that can actually be
tested within an actual distribution), and not backported to my commit
tree.

Fixed so, thanks for proofreading.

Samuel