[PATCH]: additional futex operations

[Gary Stanley]

At 04:51 AM 3/20/2008, Roman Divacky wrote:
>On Thu, Mar 20, 2008 at 08:07:03AM +0100, Alexander Leidinger wrote:
> > Quoting Roman Divacky <rdivacky at FreeBSD.org> (from Wed, 19 Mar 2008
> > 21:45:21 +0100):
> >
> > >
> > >can you guys please test:
> > >
> > >     www.vlakno.cz/~rdivacky/futex_private_pi.patch
> > >
> > >especially if linux-firefox is still broken with this patch.
> >
> > What do you think about rate limiting (only one) the FD case instead
> > of hiding it completely (and using the content of the comment as the
> > message to print with a little bit of "only report if something is
> > obviously broken")? This way we could determine if we need it for
> > linux-backwards compatibility.
>
>I dont think the FD case is used widely and we correctly (now) return
>ENOSYS so no problems should be here.
>
>Also.. if anyone is willing/able to implement the FD backing I think such
>person is skilled enough to see what is the problem even without the printf.
>It can only confuse normal people I think..
>
>I'd let it be as it is

I think it should be commented out and not rate limited. If nothing 
else, just add another #define for FUTEX_DEBUG and enable it if DEBUG
is defined in the kernel.

-G