From owner-freebsd-smp Sat Apr 4 22:35:41 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id WAA01862 for freebsd-smp-outgoing; Sat, 4 Apr 1998 22:35:41 -0800 (PST) (envelope-from owner-freebsd-smp@FreeBSD.ORG) Received: from spinner.netplex.com.au (spinner.netplex.com.au [202.12.86.3]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id WAA01857 for ; Sat, 4 Apr 1998 22:35:36 -0800 (PST) (envelope-from peter@netplex.com.au) Received: from spinner.netplex.com.au (localhost [127.0.0.1]) by spinner.netplex.com.au (8.8.8/8.8.8/Spinner) with ESMTP id OAA06176 for ; Sun, 5 Apr 1998 14:35:31 +0800 (WST) (envelope-from peter@spinner.netplex.com.au) Message-Id: <199804050635.OAA06176@spinner.netplex.com.au> X-Mailer: exmh version 2.0.2 2/24/98 To: smp@FreeBSD.ORG Subject: beware! -current under SMP is "not looking good". Date: Sun, 05 Apr 1998 14:35:30 +0800 From: Peter Wemm Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org The latest time commit has busted -current under SMP. I'm not sure of the exact circumstances, but it causes a hang when ntpdate is run. The process sits in the "running" state but doesn't gain any cpu time. Trying to break into DDB causes a deadlock. The code just prior to this was also broken. select() wasn't working properly, although there is evidence that something _else_ is causing it and select is just the scapegoat. It looks like interruptable sleeps are not being interrupted by signals somehow. (sigalrm and siginfo at least were being blocked for the duration of the select()). Other funny things were happening too, eg: cron will not run jobs on my system when built from an april 2 kernel. Cheers, -Peter To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message