From owner-freebsd-mobile Wed Oct 8 16:22:20 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id QAA21925 for mobile-outgoing; Wed, 8 Oct 1997 16:22:20 -0700 (PDT) (envelope-from owner-freebsd-mobile) Received: from pinot.eecs.harvard.edu (pinot.eecs.harvard.edu [140.247.60.65]) by hub.freebsd.org (8.8.7/8.8.7) with SMTP id QAA21915 for ; Wed, 8 Oct 1997 16:22:14 -0700 (PDT) (envelope-from karp@eecs.harvard.edu) Received: (from karp@localhost) by pinot.eecs.harvard.edu (8.6.12/8.6.12) id TAA24833 for freebsd-mobile@freebsd.org; Wed, 8 Oct 1997 19:21:38 -0400 Date: Wed, 8 Oct 1997 19:21:38 -0400 From: Brad Karp Message-Id: <199710082321.TAA24833@pinot.eecs.harvard.edu> To: freebsd-mobile@freebsd.org Subject: if_ep.c on latest PAO Sender: owner-freebsd-mobile@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Does anyone have PAO-based, mobile-support diffs for if_ep.c that target a version _later_ than that found in 2.2.2-RELEASE? The if_ep.c I wind up with when applying the latest PAO patches to 2.2.2- RELEASE is rather broken with respect to promiscuous mode; specifically, it passes every packet going by on the wire to the inbound IP path, rather than noticing that the destination Ethernet address on most packets going by doesn't merit attention by the local host. This occurs with my 3C589D, with a fairly vanilla PAO kernel. I noticed it because with IP forwarding on in the kernel, it was generating ICMP redirects for all packets going by, because the interface was constantly in promiscuous mode (even in regular operation, no tcpdump, mrouted, or multicast apps running) and the driver was passing all packets to the inbound IP path. Surely, I'm not the first to notice this...though I've scanned the mailing list archives with no leads. A glance at 3.0-current if_ep.c reveals that it at least does something different with respect to promiscuous operation and filtering packets received when in promiscuous mode. Has anyone taken a newer if_ep.c and merged it with the PAO PCMCIA support? Thanks, -Brad, karp@eecs.harvard.edu