Re: [PATCH] Reinitialize dl_load_write_lock on fork (bug 19282)

[Florian Weimer <fweimer at redhat dot com>]

On 10/05/2017 05:50 PM, Andreas Schwab wrote:
> On Okt 05 2017, Florian Weimer <fweimer@redhat.com> wrote:
>
> > On 10/05/2017 05:39 PM, Andreas Schwab wrote:
> > > On Okt 05 2017, Florian Weimer <fweimer@redhat.com> wrote:
> > >
> > > > On 10/05/2017 05:17 PM, Andreas Schwab wrote:
> > > > > -      /* Reset the lock the dynamic loader uses to protect its data.  */
> > > > > +      /* Reset the locks the dynamic loader uses to protect its data.  */
> > > > >           __rtld_lock_initialize (GL(dl_load_lock));
> > > > > +      __rtld_lock_initialize (GL(dl_load_write_lock));
> > > >
> > > > I'm pretty sure that the child now sees corrupted a dynamic linker state
> > > > instead of hanging.
> > >
> > > That's a good point, though this problem already existed before the
> > > introduction of dl_load_write_lock.
> >
> > Why does this make a difference?
>
> Difference to what?

I assumed you were arguing the patch was somehow okay because we used to 
have the corruption bug.  If not and you are retracting your patch, 
that's fine, too.

Thanks,
Florian