On Sun, Aug 07, 2005 at 01:50:22PM +0200, Thomas Graf ([EMAIL PROTECTED]) wrote:
> * Evgeniy Polyakov <[EMAIL PROTECTED]> 2005-08-07 15:19
> > On Sun, Aug 07, 2005 at 01:12:19PM +0200, Thomas Graf ([EMAIL PROTECTED])
> > wrote:
> > > I don't get this, you introduce a new option which basically
> > > changes the subscription schema from a bitmask to a single
> > > identifier. I don't see any code which would prevent the
> > > message delivery if the bitmask actually matches an identifier.
> >
> > Since it is new option, it is supposed that callers/users know what it
> > does, i.e. if one sets this, it already knows that groups will be
> > concidered not as bitmask, but as group number.
>
> So you basically assume that all sockets of a specific netlink
> family either have this option enabled or disabled? Wouldn't it
> be easier to just enforce this on the family level rather than
> on a per socket basis?
I even thought that one new family can be created for unicat-only
messages, while others can use it in it's own way, that's why it is
socket option.
> Anyways, your idea of having 32bit identifiers rather than a
> bitmask makes sense to me, however I think we should not lose
> the ability to do multicasting for event notifications so we
> might need a slightly more complex solution. I guess Patrick
> is taking care of this though.
Sure.
--
Evgeniy Polyakov
-
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to [EMAIL PROTECTED]
More majordomo info at http://vger.kernel.org/majordomo-info.html
