Re: Chaining socket filters
Re: Chaining socket filters
- Subject: Re: Chaining socket filters
- From: Prokash Sinha <email@hidden>
- Date: Thu, 20 Sep 2018 09:37:13 -0700
As it turned out, it seems like not possible to have more than one socket
filter for same protocol ( like ipv4), second one fails with error at
registration time ). Also not sure if one of them can be global and another on
demand.
I assume it is the same at IP layer, and IF layer! Correct me if I’m wrong .
Also I assume the DLIL_ ( data link layer if ) filtering is not encouraged and
perhaps on the verge of deprecation.
Any input/thought ?
> On Sep 19, 2018, at 1:31 PM, Prokash Sinha <email@hidden> wrote:
>
> Folks,
>
> I’ve a question about the possibility of chaining socket filter. So assume I
> have two socket filters, and I want to register them ( let’s say globally ).
>
> Now I would assume they would be chained along with the existing ones ( if
> any ). Now what is the mechanism of hand-off ( For example, a shared ISR
> vector can have multiple interrupts registered thru a chain. Or like multiple
> exception handling registered in some stack fashion ).
>
> So there are two returns to indicate, if the kernel continue thru the chain
> or not.
>
> Is it possible ? If so, then what would be the return indicator/value so that
> the Network BF module can pass it along the chain. And what would be the
> value to cut short ( of course without error, something like EJUSTRETURN ???
> ).
>
>
> Thanks,
> Prokash
_______________________________________________
Do not post admin requests to the list. They will be ignored.
Darwin-kernel mailing list (email@hidden)
Help/Unsubscribe/Update your Subscription:
This email sent to email@hidden