From owner-freebsd-questions Mon Mar 13 14:03:33 1995 Return-Path: questions-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id OAA00150 for questions-outgoing; Mon, 13 Mar 1995 14:03:33 -0800 Received: from bach.seattleu.edu (root@bach.seattleu.edu [199.237.224.11]) by freefall.cdrom.com (8.6.10/8.6.6) with SMTP id NAA00247 for ; Mon, 13 Mar 1995 13:57:51 -0800 Received: from opus.seattleu.edu.seattleu.edu by bach.seattleu.edu (4.1/SMI-4.1) id AA16365; Mon, 13 Mar 95 13:58:38 PST Received: by opus.seattleu.edu.seattleu.edu (4.1/SMI-4.1) id AA07480; Mon, 13 Mar 95 13:59:45 PST Date: Mon, 13 Mar 1995 13:48:18 -0800 (PST) From: Jack Valko Subject: i/o interrupts To: freebsd-questions@FreeBSD.org Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: questions-owner@FreeBSD.org Precedence: bulk We're having a perplexing problem with I/O and FreeBSD 2.0 (not the February release). We have the FreeBSD box (486, 33mhz, 8megs RAM) and an I/O board sending a continuous stream of data to COM1 (/dev/tty00). It seems that interrupts are building in a buffer if they are not handled fast enough. Instead of the buffer overflowing or interrupts being lost, the tty interfaces lock, but the system continues to function (for example, httpd still runs). As soon as we kill the program(s) that are sending and receiving data thru tty00, or we disconnect the external I/O source the ttys free up. We use fopen("/dev/tty00", "r+") to move data to and from the port. Any ideas?? ============================================================================== Jack Valko valko@seattleu.edu ==============================================================================