Re: e2_fs.c fix for MNT_LOCAL being defined
- From: <tpgww@xxxxxxxxxxx>
- To: emelfm2@xxxxxxxxxxxxx
- Date: Mon, 23 Feb 2009 20:36:33 +1100
On Sat, 21 Feb 2009 14:00:39 +0100
Marcus von Appen <mva@xxxxxxxxxxxx> wrote:
> On, Sat Feb 21, 2009, tpgww@xxxxxxxxxxx wrote:
>
> > On Fri, 20 Feb 2009 09:28:18 +0100
> > Marcus von Appen <mva@xxxxxxxxxxxx> wrote:
> >
> > > in src/filesystem/e2_fs.c an error sneaked in (current SVN and 0.5.1
> > > upstream package) for the case MNT_LOCAL being defined. A quick fix
> > > (without checking, if the logical behaviour is correct) is attached.
> >
> > I don't suppose you have a moment to investigate whether, for your
> > system, statfs() can be clobbered by an interrupt ?
> >
> >
> > In that case, the preceding 2 lines would be relevant, and we could
> > keep the "else".
>
> If you can describe me a scenario, which might produce such a case, I
> can try it. I'll read up the internal specs for the FreeBSD statfs() to
> see, whether such a case is mentioned there (e.g. exclusive locks or so).
I don't know exactly what can generate interrupts on my system, much less yours.
The online bsd information that I found made no reference to EINTR, but I'm not
sure if that lack is always correlated with defined MNT_LOCAL. ATM I _think_
it's so, so maybe best to just go ahead and apply your patch ...
Regards
Tom
--
Users can unsubscribe from the list by sending email to
emelfm2-request@xxxxxxxxxxxxx with 'unsubscribe' in the subject field or by
logging into the web interface.
Other related posts: