From owner-freebsd-current Sun Jun 18 16:59:18 1995 Return-Path: current-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id QAA10473 for current-outgoing; Sun, 18 Jun 1995 16:59:18 -0700 Received: from clinet.fi (root@clinet.fi [193.64.6.1]) by freefall.cdrom.com (8.6.10/8.6.6) with ESMTP id QAA10466 for ; Sun, 18 Jun 1995 16:59:16 -0700 Received: from katiska.clinet.fi (root@katiska.clinet.fi [193.64.6.3]) by clinet.fi (8.6.10/8.6.4) with ESMTP id CAA27179; Mon, 19 Jun 1995 02:59:08 +0300 From: Heikki Suonsivu Received: (hsu@localhost) by katiska.clinet.fi (8.6.11/8.6.4) id CAA03497; Mon, 19 Jun 1995 02:59:08 +0300 Date: Mon, 19 Jun 1995 02:59:08 +0300 Message-Id: <199506182359.CAA03497@katiska.clinet.fi> To: Bruce Evans Cc: hsu@clinet.fi, current@freebsd.org Subject: Re: kern/528: interrupt-level buffer overflows In-Reply-To: <199506180946.TAA22668@godzilla.zeta.org.au> References: <199506180946.TAA22668@godzilla.zeta.org.au> Organization: Clinet Ltd, Espoo, Finland Sender: current-owner@freebsd.org Precedence: bulk Bruce Evans writes: > >Jun 18 05:14:47 pommi /kernel: sio1: 119 more interrupt-level buffer over > >flows (total 3642) > >Jun 18 05:14:47 pommi /kernel: sio1: 119 more interrupt-level buffer overflows ( > >total 3642) > For this to happen, softclock() must sometimes be delayed for a long time > several clock ticks (about 3 clock ticks for 119 characters at 38400 bps > and about (256 - 38) / 38 clock ticks for filling up the 256 character > buffer before that. > >>Fix: > Use hardware handshaking. Its a leased line, no hardware handshake. WIth decent uarts it shouldn't a problem to run even at 115k (It did work with Linux on 386/16 and FreeBSD on 486/66, about 11kbytes per second for ftp transfer). I can think of increasing the serial buffers (256 characters sounds low) and if the problem is interrupt latency, dropping the trigger level on uarts. Any ideas? > I've heard that some systems 20 times as fast as a 386/16 are too slow for > a 38400 bps link. With 16550's? Less than 300 interrupts per second? -- Heikki Suonsivu, T{ysikuu 10 C 83/02210 Espoo/FINLAND, hsu@cs.hut.fi home +358-0-8031121 work -4375209 fax -4555276 riippu SN