From owner-freebsd-current Sat Aug 29 18:39:33 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id SAA24149 for freebsd-current-outgoing; Sat, 29 Aug 1998 18:39:33 -0700 (PDT) (envelope-from owner-freebsd-current@FreeBSD.ORG) Received: from skynet.ctr.columbia.edu (skynet.ctr.columbia.edu [128.59.64.70]) by hub.freebsd.org (8.8.8/8.8.8) with SMTP id SAA24144; Sat, 29 Aug 1998 18:39:28 -0700 (PDT) (envelope-from wpaul@skynet.ctr.columbia.edu) Received: (from wpaul@localhost) by skynet.ctr.columbia.edu (8.6.12/8.6.9) id VAA01492; Sat, 29 Aug 1998 21:41:33 -0400 From: Bill Paul Message-Id: <199808300141.VAA01492@skynet.ctr.columbia.edu> Subject: Re: Looking for feedback on xl (3c905/3c905B) driver To: bright@hotjobs.com (Alfred Perlstein) Date: Sat, 29 Aug 1998 21:41:32 -0400 (EDT) Cc: stable@FreeBSD.ORG, current@FreeBSD.ORG In-Reply-To: from "Alfred Perlstein" at Aug 29, 98 05:20:09 pm X-Mailer: ELM [version 2.4 PL24] Content-Type: text Sender: owner-freebsd-current@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG Of all the gin joints in all the towns in all the world, Alfred Perlstein had to walk into mine and say: > works BEAUTIFULLY from a user standpoint, only odd thing is that when i > enable a bpfilter program, trafshow/tcpdump it sorta locks up all my > processes for a second. it's odd in X as i can movei my mouse, but things > like x11amp (linux emulation) lock up for a good 2-3 seconds, then all is > well until the program "detaches" from the bpf: This happens because changing the interface flags (via xl_ioctl(), when setting the IFF_PROMISC flag) causes xl_init() to be invoked, and xl_init() forcibly puts the interface into an initialized state, flushes out the receive and transmit queues and resets the receiver and transmitter. The receive and transmit reset appears to take some time, during which NIC operations are temporarily suspended. This is probably a bit much considering that the only thing that really needs to be done is to set the RX filter bits to allow reception of all frames. I think I inherited this behavior from the fxp driver, which is what I used as a reference when I got started with driver writing. I'll look into changing this. Note that using BPF without setting the IFF_PROMISC flag shouldn't cause any delay. -Bill -- ============================================================================= -Bill Paul (212) 854-6020 | System Manager, Master of Unix-Fu Work: wpaul@ctr.columbia.edu | Center for Telecommunications Research Home: wpaul@skynet.ctr.columbia.edu | Columbia University, New York City ============================================================================= "It is not I who am crazy; it is I who am mad!" - Ren Hoek, "Space Madness" ============================================================================= To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message