Re: How to write a proper fork hook code?

[Samuel Thibault]

Svante Signell, le Mon 02 Mar 2015 11:03:22 +0100, a écrit :
> On Mon, 2015-03-02 at 10:57 +0100, Svante Signell wrote:
> > On Mon, 2015-03-02 at 10:38 +0100, Samuel Thibault wrote:
> 
> > > > > But again, I'm still unsure why you believe that locks are being
> > > > > inherited through fork(): before introducing a fork hook to forcibly
> > > > > unlock something, one needs to be absolutely sure that it's actually
> > > > > locked.  Actually it'd even look like a design flaw if two process 
> > > > > could
> > > > > actually hold a lock on the same piece of a file...
> 
> Well, the file record locking patches were mainly written by Neal. So if
> you are so very expert on everything,

I'm not expert on the code that Neal wrote.

> maybe you could help me find where the design flaw is.

That would take time, which I don't have.

> I have not managed to find it so far. The patches
> have been submitted twice, latest (git mail) version starts with:
> http://lists.gnu.org/archive/html/bug-hurd/2015-01/msg00108.html
> and
> http://lists.gnu.org/archive/html/bug-hurd/2015-01/msg00106.html

Yes, and that's still in my mbox, along a lot of other things.  Yes,
that sucks and I'm not happy with that, but days are 24h long only.

Samuel