NetBSD Problem Report #37093

From martin@duskware.de  Tue Oct  9 13:01:50 2007
Return-Path: <martin@duskware.de>
Received: from mail.netbsd.org (mail.netbsd.org [204.152.190.11])
	by narn.NetBSD.org (Postfix) with ESMTP id 70CFE63B8D8
	for <gnats-bugs@gnats.netbsd.org>; Tue,  9 Oct 2007 13:01:50 +0000 (UTC)
Message-Id: <20071009130126.7282E63B8D8@narn.NetBSD.org>
Date: Tue,  9 Oct 2007 13:01:26 +0000 (UTC)
From: ad@netbsd.org
Reply-To: ad@netbsd.org
To: netbsd-bugs-owner@NetBSD.org
Subject: itimer problems
X-Send-Pr-Version: www-1.0

>Number:         37093
>Category:       kern
>Synopsis:       itimer problems
>Confidential:   no
>Severity:       serious
>Priority:       high
>Responsible:    ad
>State:          closed
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Tue Oct 09 13:05:01 +0000 2007
>Closed-Date:    Mon Apr 21 00:22:18 +0000 2008
>Last-Modified:  Mon Apr 21 00:22:18 +0000 2008
>Originator:     Andrew Doran
>Release:        NetBSD 4.99.32
>Organization:
The NetBSD Project
>Environment:
N/A
>Description:
itimers are fired from hardclock(). There are a few problems:

- itimer code run in process context has scant splsched() protection

- hardclock() is called w/o kernel_lock held and the code is not MP
  safe.

- itimerfire() may try to send signals with ksiginfo, which will
  use ksiginfo_pool and thus the VM system above IPL_VM.

>How-To-Repeat:
Code inspection.

>Fix:
- Defer sending signals to a soft interrupt.
- Spinlock the itimers.

>Release-Note:

>Audit-Trail:

Responsible-Changed-From-To: kern-bug-people->ad
Responsible-Changed-By: ad@netbsd.org
Responsible-Changed-When: Wed, 17 Oct 2007 17:36:56 +0000
Responsible-Changed-Why:
Mine.


From: Andrew Doran <ad@netbsd.org>
To: gnats-bugs@gnats.NetBSD.org
Cc: 
Subject: PR/37093 CVS commit: src/sys
Date: Mon, 21 Apr 2008 00:13:47 +0000 (UTC)

 Module Name:	src
 Committed By:	ad
 Date:		Mon Apr 21 00:13:46 UTC 2008

 Modified Files:
 	src/sys/compat/linux/common: linux_misc_notalpha.c
 	src/sys/kern: init_main.c kern_clock.c kern_time.c syscalls.master
 	src/sys/sys: timevar.h

 Log Message:
 timer fixes for PR 37093:

 - Fix serious concurrency problems, making the code MT and MP safe in
   the process.
 - Don't allocate memory or inspect process state from hardclock().


 To generate a diff of this commit:
 cvs rdiff -r1.100 -r1.101 src/sys/compat/linux/common/linux_misc_notalpha.c
 cvs rdiff -r1.349 -r1.350 src/sys/kern/init_main.c
 cvs rdiff -r1.119 -r1.120 src/sys/kern/kern_clock.c
 cvs rdiff -r1.141 -r1.142 src/sys/kern/kern_time.c
 cvs rdiff -r1.196 -r1.197 src/sys/kern/syscalls.master
 cvs rdiff -r1.20 -r1.21 src/sys/sys/timevar.h

 Please note that diffs are not public domain; they are subject to the
 copyright notices on the relevant files.

State-Changed-From-To: open->closed
State-Changed-By: ad@NetBSD.org
State-Changed-When: Mon, 21 Apr 2008 00:22:18 +0000
State-Changed-Why:
fixed


>Unformatted:

NetBSD Home
NetBSD PR Database Search

(Contact us) $NetBSD: query-full-pr,v 1.51 2026/08/10 02:28:17 riastradh Exp $
$NetBSD: gnats_config.sh,v 1.10 2026/05/13 22:00:09 riastradh Exp $
Copyright © 1994-2026 The NetBSD Foundation, Inc. ALL RIGHTS RESERVED.