From owner-freebsd-net Tue Jun 20 14:32:41 2000 Delivered-To: freebsd-net@freebsd.org Received: from mail.interware.hu (mail.interware.hu [195.70.32.130]) by hub.freebsd.org (Postfix) with ESMTP id 3F92737BFB6 for ; Tue, 20 Jun 2000 14:32:36 -0700 (PDT) (envelope-from julian@elischer.org) Received: from mogadishu-49.budapest.interware.hu ([195.70.52.113] helo=jules.elischer.org) by mail.interware.hu with smtp (Exim 3.12 #1 (Debian)) id 134Vs9-0005Ra-00; Tue, 20 Jun 2000 23:47:54 +0200 Message-ID: <394FE2CE.1CFBAE39@elischer.org> Date: Tue, 20 Jun 2000 14:31:58 -0700 From: Julian Elischer X-Mailer: Mozilla 3.04Gold (X11; I; FreeBSD 5.0-CURRENT i386) MIME-Version: 1.0 To: Archie Cobbs Cc: freebsd-net@FreeBSD.ORG Subject: Re: Patch review request (ng_ether(4)) References: <200006201905.MAA26374@bubba.whistle.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-net@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Archie Cobbs wrote: > > > You can't use NG_SEND_DATA_RET() in an ethernet driver (on input) > > because you need to use NG_QUEUE_DATA() (splimp, remember) > > My thought is that you would supply a separate hook for > > 'FILTER' operations which you would warn would have to > > be run at splimp. The splnet and splimp graphs can only be joined > > by queueing links (going up). I think this needs work, and there > > is a possibility that we could put transfer characteristics > > as a property on links so that the right thing happens. > > OK, back from vacation :-) > > Here's a new patch addressing the above problem: > > ftp://ftp.whistle.com/pub/archie/misc/NGETHER.patch.3 Not sure if using the stub code in output makes sense if it's not used on the input side.. looks good though I'm not the person to convince.... > > > The FILTER hooks would either have to be attached > > to nodes that use NG_QUEUE_DATA(), or which > > had no connection to the splnet universe. Automatic registration > > of spl levels on hooks might lead to some automatic handling > > of this but > > a/ I think the programmer should know what is going on. > > b/ this may all become moot with the removal of SPLs for > > interrupts. > > I vote for option (b) :-) yeah, hear no evil, see no..... > > -Archie > > ___________________________________________________________________________ > Archie Cobbs * Whistle Communications, Inc. * http://www.whistle.com -- __--_|\ Julian Elischer / \ julian@elischer.org ( OZ ) World tour 2000 )_.---._/ presently in: Budapest v To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message