freebsd-dev/sys/kern/kern_acct.c

555 lines
15 KiB
C
Raw Normal View History

1994-05-24 10:09:53 +00:00
/*-
* Copyright (c) 1982, 1986, 1989, 1993
* The Regents of the University of California. All rights reserved.
* (c) UNIX System Laboratories, Inc.
* All or some portions of this file are derived from material licensed
* to the University of California by American Telephone and Telegraph
* Co. or Unix System Laboratories, Inc. and are reproduced herein with
* the permission of UNIX System Laboratories, Inc.
*
Significant refactoring of the accounting code to improve locking and VFS happiness, as well as correct other bugs: - Replace notion of current and saved accounting credential/vnode with a single credential/vnode and an acct_suspended flag. This simplifies the accounting logic substantially. - Replace acct_mtx with acct_sx, a sleepable lock held exclusively during reconfiguration and space polling, but shared during log entry generation. This avoids holding a mutex over sleepable VFS operations. - Hold the sx lock over the duration of the I/O so that the vnode I/O cannot occur after vnode close, which could occur previously if accounting was disabled as a process exited. - Write the accounting log entry with Giant conditionally acquired based on the file system where the log is stored. Previously, the accounting code relied on the caller acquiring Giant. - Acquire Giant conditionally in the accounting callout based on the file system where the accounting log is stored. Run the callout MPSAFE. - Expose acct_suspended via a read-only sysctl so it is possibly to programmatically determine whether accounting is suspended or not without attempting to parse logs. - Check both acct_vp and acct_suspended lock-free before entering the accounting sx lock in acct(). - When accounting is disabled due to a VBAD vnode (i.e., forceable unmount), generate a log message indicating accounting has been disabled. - Correct a long-standing bug in how free space is calculated and compared to the required space: generate and compare signed results, not unsigned results, or negative free space will cause accounting to not be suspended when required, or worse, incorrectly resumed once negative free space is reached. MFC after: 2 weeks
2005-11-12 10:45:13 +00:00
* Copyright (c) 1994 Christopher G. Demetriou
* Copyright (c) 2005 Robert N. M. Watson
*
1994-05-24 10:09:53 +00:00
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
* 3. All advertising materials mentioning features or use of this software
* must display the following acknowledgement:
* This product includes software developed by the University of
* California, Berkeley and its contributors.
* 4. Neither the name of the University nor the names of its contributors
* may be used to endorse or promote products derived from this software
* without specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
* SUCH DAMAGE.
*
* @(#)kern_acct.c 8.1 (Berkeley) 6/14/93
1994-05-24 10:09:53 +00:00
*/
2003-06-11 00:56:59 +00:00
#include <sys/cdefs.h>
__FBSDID("$FreeBSD$");
#include "opt_mac.h"
1994-05-24 10:09:53 +00:00
#include <sys/param.h>
#include <sys/systm.h>
2006-02-03 16:37:55 +00:00
#include <sys/acct.h>
#include <sys/fcntl.h>
#include <sys/kernel.h>
#include <sys/kthread.h>
#include <sys/lock.h>
1994-05-24 10:09:53 +00:00
#include <sys/mount.h>
2006-02-03 16:37:55 +00:00
#include <sys/mutex.h>
#include <sys/namei.h>
#include <sys/priv.h>
2006-02-03 16:37:55 +00:00
#include <sys/proc.h>
#include <sys/resourcevar.h>
#include <sys/sched.h>
2006-02-03 16:37:55 +00:00
#include <sys/sx.h>
#include <sys/sysctl.h>
#include <sys/sysent.h>
#include <sys/syslog.h>
#include <sys/sysproto.h>
#include <sys/tty.h>
2006-02-03 16:37:55 +00:00
#include <sys/vnode.h>
1994-05-24 10:09:53 +00:00
#include <security/mac/mac_framework.h>
/*
* The routines implemented in this file are described in:
* Leffler, et al.: The Design and Implementation of the 4.3BSD
* UNIX Operating System (Addison Welley, 1989)
* on pages 62-63.
*
* Arguably, to simplify accounting operations, this mechanism should
* be replaced by one in which an accounting log file (similar to /dev/klog)
* is read by a user process, etc. However, that has its own problems.
*/
/*
* Internal accounting functions.
* The former's operation is described in Leffler, et al., and the latter
* was provided by UCB with the 4.4BSD-Lite release
*/
2002-03-19 21:25:46 +00:00
static comp_t encode_comp_t(u_long, u_long);
static void acctwatch(void);
static void acct_thread(void *);
static int acct_disable(struct thread *);
init_main.c subr_autoconf.c: Add support for "interrupt driven configuration hooks". A component of the kernel can register a hook, most likely during auto-configuration, and receive a callback once interrupt services are available. This callback will occur before the root and dump devices are configured, so the configuration task can affect the selection of those two devices or complete any tasks that need to be performed prior to launching init. System boot is posponed so long as a hook is registered. The hook owner is responsible for removing the hook once their task is complete or the system boot can continue. kern_acct.c kern_clock.c kern_exit.c kern_synch.c kern_time.c: Change the interface and implementation for the kernel callout service. The new implemntaion is based on the work of Adam M. Costello and George Varghese, published in a technical report entitled "Redesigning the BSD Callout and Timer Facilities". The interface used in FreeBSD is a little different than the one outlined in the paper. The new function prototypes are: struct callout_handle timeout(void (*func)(void *), void *arg, int ticks); void untimeout(void (*func)(void *), void *arg, struct callout_handle handle); If a client wishes to remove a timeout, it must store the callout_handle returned by timeout and pass it to untimeout. The new implementation gives 0(1) insert and removal of callouts making this interface scale well even for applications that keep 100s of callouts outstanding. See the updated timeout.9 man page for more details.
1997-09-21 22:00:25 +00:00
/*
* Accounting vnode pointer, saved vnode pointer, and flags for each.
Significant refactoring of the accounting code to improve locking and VFS happiness, as well as correct other bugs: - Replace notion of current and saved accounting credential/vnode with a single credential/vnode and an acct_suspended flag. This simplifies the accounting logic substantially. - Replace acct_mtx with acct_sx, a sleepable lock held exclusively during reconfiguration and space polling, but shared during log entry generation. This avoids holding a mutex over sleepable VFS operations. - Hold the sx lock over the duration of the I/O so that the vnode I/O cannot occur after vnode close, which could occur previously if accounting was disabled as a process exited. - Write the accounting log entry with Giant conditionally acquired based on the file system where the log is stored. Previously, the accounting code relied on the caller acquiring Giant. - Acquire Giant conditionally in the accounting callout based on the file system where the accounting log is stored. Run the callout MPSAFE. - Expose acct_suspended via a read-only sysctl so it is possibly to programmatically determine whether accounting is suspended or not without attempting to parse logs. - Check both acct_vp and acct_suspended lock-free before entering the accounting sx lock in acct(). - When accounting is disabled due to a VBAD vnode (i.e., forceable unmount), generate a log message indicating accounting has been disabled. - Correct a long-standing bug in how free space is calculated and compared to the required space: generate and compare signed results, not unsigned results, or negative free space will cause accounting to not be suspended when required, or worse, incorrectly resumed once negative free space is reached. MFC after: 2 weeks
2005-11-12 10:45:13 +00:00
* acct_sx protects against changes to the active vnode and credentials
* while accounting records are being committed to disk.
*/
static int acct_configured;
Significant refactoring of the accounting code to improve locking and VFS happiness, as well as correct other bugs: - Replace notion of current and saved accounting credential/vnode with a single credential/vnode and an acct_suspended flag. This simplifies the accounting logic substantially. - Replace acct_mtx with acct_sx, a sleepable lock held exclusively during reconfiguration and space polling, but shared during log entry generation. This avoids holding a mutex over sleepable VFS operations. - Hold the sx lock over the duration of the I/O so that the vnode I/O cannot occur after vnode close, which could occur previously if accounting was disabled as a process exited. - Write the accounting log entry with Giant conditionally acquired based on the file system where the log is stored. Previously, the accounting code relied on the caller acquiring Giant. - Acquire Giant conditionally in the accounting callout based on the file system where the accounting log is stored. Run the callout MPSAFE. - Expose acct_suspended via a read-only sysctl so it is possibly to programmatically determine whether accounting is suspended or not without attempting to parse logs. - Check both acct_vp and acct_suspended lock-free before entering the accounting sx lock in acct(). - When accounting is disabled due to a VBAD vnode (i.e., forceable unmount), generate a log message indicating accounting has been disabled. - Correct a long-standing bug in how free space is calculated and compared to the required space: generate and compare signed results, not unsigned results, or negative free space will cause accounting to not be suspended when required, or worse, incorrectly resumed once negative free space is reached. MFC after: 2 weeks
2005-11-12 10:45:13 +00:00
static int acct_suspended;
static struct vnode *acct_vp;
static struct ucred *acct_cred;
static int acct_flags;
static struct sx acct_sx;
Significant refactoring of the accounting code to improve locking and VFS happiness, as well as correct other bugs: - Replace notion of current and saved accounting credential/vnode with a single credential/vnode and an acct_suspended flag. This simplifies the accounting logic substantially. - Replace acct_mtx with acct_sx, a sleepable lock held exclusively during reconfiguration and space polling, but shared during log entry generation. This avoids holding a mutex over sleepable VFS operations. - Hold the sx lock over the duration of the I/O so that the vnode I/O cannot occur after vnode close, which could occur previously if accounting was disabled as a process exited. - Write the accounting log entry with Giant conditionally acquired based on the file system where the log is stored. Previously, the accounting code relied on the caller acquiring Giant. - Acquire Giant conditionally in the accounting callout based on the file system where the accounting log is stored. Run the callout MPSAFE. - Expose acct_suspended via a read-only sysctl so it is possibly to programmatically determine whether accounting is suspended or not without attempting to parse logs. - Check both acct_vp and acct_suspended lock-free before entering the accounting sx lock in acct(). - When accounting is disabled due to a VBAD vnode (i.e., forceable unmount), generate a log message indicating accounting has been disabled. - Correct a long-standing bug in how free space is calculated and compared to the required space: generate and compare signed results, not unsigned results, or negative free space will cause accounting to not be suspended when required, or worse, incorrectly resumed once negative free space is reached. MFC after: 2 weeks
2005-11-12 10:45:13 +00:00
SX_SYSINIT(acct, &acct_sx, "acct_sx");
/*
* State of the accounting kthread.
*/
static int acct_state;
#define ACCT_RUNNING 1 /* Accounting kthread is running. */
#define ACCT_EXITREQ 2 /* Accounting kthread should exit. */
/*
* Values associated with enabling and disabling accounting
*/
static int acctsuspend = 2; /* stop accounting when < 2% free space left */
SYSCTL_INT(_kern, OID_AUTO, acct_suspend, CTLFLAG_RW,
&acctsuspend, 0, "percentage of free disk space below which accounting stops");
static int acctresume = 4; /* resume when free space risen to > 4% */
SYSCTL_INT(_kern, OID_AUTO, acct_resume, CTLFLAG_RW,
&acctresume, 0, "percentage of free disk space above which accounting resumes");
static int acctchkfreq = 15; /* frequency (in seconds) to check space */
static int
sysctl_acct_chkfreq(SYSCTL_HANDLER_ARGS)
{
int error, value;
/* Write out the old value. */
error = SYSCTL_OUT(req, &acctchkfreq, sizeof(int));
if (error || req->newptr == NULL)
return (error);
/* Read in and verify the new value. */
error = SYSCTL_IN(req, &value, sizeof(int));
if (error)
return (error);
if (value <= 0)
return (EINVAL);
acctchkfreq = value;
return (0);
}
SYSCTL_PROC(_kern, OID_AUTO, acct_chkfreq, CTLTYPE_INT|CTLFLAG_RW,
&acctchkfreq, 0, sysctl_acct_chkfreq, "I",
"frequency for checking the free space");
SYSCTL_INT(_kern, OID_AUTO, acct_configured, CTLFLAG_RD, &acct_configured, 0,
"Accounting configured or not");
Significant refactoring of the accounting code to improve locking and VFS happiness, as well as correct other bugs: - Replace notion of current and saved accounting credential/vnode with a single credential/vnode and an acct_suspended flag. This simplifies the accounting logic substantially. - Replace acct_mtx with acct_sx, a sleepable lock held exclusively during reconfiguration and space polling, but shared during log entry generation. This avoids holding a mutex over sleepable VFS operations. - Hold the sx lock over the duration of the I/O so that the vnode I/O cannot occur after vnode close, which could occur previously if accounting was disabled as a process exited. - Write the accounting log entry with Giant conditionally acquired based on the file system where the log is stored. Previously, the accounting code relied on the caller acquiring Giant. - Acquire Giant conditionally in the accounting callout based on the file system where the accounting log is stored. Run the callout MPSAFE. - Expose acct_suspended via a read-only sysctl so it is possibly to programmatically determine whether accounting is suspended or not without attempting to parse logs. - Check both acct_vp and acct_suspended lock-free before entering the accounting sx lock in acct(). - When accounting is disabled due to a VBAD vnode (i.e., forceable unmount), generate a log message indicating accounting has been disabled. - Correct a long-standing bug in how free space is calculated and compared to the required space: generate and compare signed results, not unsigned results, or negative free space will cause accounting to not be suspended when required, or worse, incorrectly resumed once negative free space is reached. MFC after: 2 weeks
2005-11-12 10:45:13 +00:00
SYSCTL_INT(_kern, OID_AUTO, acct_suspended, CTLFLAG_RD, &acct_suspended, 0,
"Accounting suspended or not");
/*
* Accounting system call. Written based on the specification and
* previous implementation done by Mark Tinguely.
*
* MPSAFE
*/
int
Significant refactoring of the accounting code to improve locking and VFS happiness, as well as correct other bugs: - Replace notion of current and saved accounting credential/vnode with a single credential/vnode and an acct_suspended flag. This simplifies the accounting logic substantially. - Replace acct_mtx with acct_sx, a sleepable lock held exclusively during reconfiguration and space polling, but shared during log entry generation. This avoids holding a mutex over sleepable VFS operations. - Hold the sx lock over the duration of the I/O so that the vnode I/O cannot occur after vnode close, which could occur previously if accounting was disabled as a process exited. - Write the accounting log entry with Giant conditionally acquired based on the file system where the log is stored. Previously, the accounting code relied on the caller acquiring Giant. - Acquire Giant conditionally in the accounting callout based on the file system where the accounting log is stored. Run the callout MPSAFE. - Expose acct_suspended via a read-only sysctl so it is possibly to programmatically determine whether accounting is suspended or not without attempting to parse logs. - Check both acct_vp and acct_suspended lock-free before entering the accounting sx lock in acct(). - When accounting is disabled due to a VBAD vnode (i.e., forceable unmount), generate a log message indicating accounting has been disabled. - Correct a long-standing bug in how free space is calculated and compared to the required space: generate and compare signed results, not unsigned results, or negative free space will cause accounting to not be suspended when required, or worse, incorrectly resumed once negative free space is reached. MFC after: 2 weeks
2005-11-12 10:45:13 +00:00
acct(struct thread *td, struct acct_args *uap)
1994-05-24 10:09:53 +00:00
{
struct nameidata nd;
int error, flags, vfslocked;
error = priv_check(td, PRIV_ACCT);
if (error)
return (error);
1994-05-24 10:09:53 +00:00
/*
* If accounting is to be started to a file, open that file for
* appending and make sure it's a 'normal'.
1994-05-24 10:09:53 +00:00
*/
2002-12-14 01:56:26 +00:00
if (uap->path != NULL) {
NDINIT(&nd, LOOKUP, NOFOLLOW | MPSAFE | AUDITVNODE1,
UIO_USERSPACE, uap->path, td);
flags = FWRITE | O_APPEND;
error = vn_open(&nd, &flags, 0, -1);
if (error)
return (error);
vfslocked = NDHASGIANT(&nd);
NDFREE(&nd, NDF_ONLY_PNBUF);
#ifdef MAC
error = mac_check_system_acct(td->td_ucred, nd.ni_vp);
if (error) {
VOP_UNLOCK(nd.ni_vp, 0, td);
vn_close(nd.ni_vp, flags, td->td_ucred, td);
VFS_UNLOCK_GIANT(vfslocked);
return (error);
}
#endif
VOP_UNLOCK(nd.ni_vp, 0, td);
if (nd.ni_vp->v_type != VREG) {
vn_close(nd.ni_vp, flags, td->td_ucred, td);
VFS_UNLOCK_GIANT(vfslocked);
return (EACCES);
}
VFS_UNLOCK_GIANT(vfslocked);
#ifdef MAC
} else {
error = mac_check_system_acct(td->td_ucred, NULL);
if (error)
return (error);
#endif
}
1994-05-24 10:09:53 +00:00
Significant refactoring of the accounting code to improve locking and VFS happiness, as well as correct other bugs: - Replace notion of current and saved accounting credential/vnode with a single credential/vnode and an acct_suspended flag. This simplifies the accounting logic substantially. - Replace acct_mtx with acct_sx, a sleepable lock held exclusively during reconfiguration and space polling, but shared during log entry generation. This avoids holding a mutex over sleepable VFS operations. - Hold the sx lock over the duration of the I/O so that the vnode I/O cannot occur after vnode close, which could occur previously if accounting was disabled as a process exited. - Write the accounting log entry with Giant conditionally acquired based on the file system where the log is stored. Previously, the accounting code relied on the caller acquiring Giant. - Acquire Giant conditionally in the accounting callout based on the file system where the accounting log is stored. Run the callout MPSAFE. - Expose acct_suspended via a read-only sysctl so it is possibly to programmatically determine whether accounting is suspended or not without attempting to parse logs. - Check both acct_vp and acct_suspended lock-free before entering the accounting sx lock in acct(). - When accounting is disabled due to a VBAD vnode (i.e., forceable unmount), generate a log message indicating accounting has been disabled. - Correct a long-standing bug in how free space is calculated and compared to the required space: generate and compare signed results, not unsigned results, or negative free space will cause accounting to not be suspended when required, or worse, incorrectly resumed once negative free space is reached. MFC after: 2 weeks
2005-11-12 10:45:13 +00:00
/*
* Disallow concurrent access to the accounting vnode while we swap
* it out, in order to prevent access after close.
*/
sx_xlock(&acct_sx);
/*
* If accounting was previously enabled, kill the old space-watcher,
Significant refactoring of the accounting code to improve locking and VFS happiness, as well as correct other bugs: - Replace notion of current and saved accounting credential/vnode with a single credential/vnode and an acct_suspended flag. This simplifies the accounting logic substantially. - Replace acct_mtx with acct_sx, a sleepable lock held exclusively during reconfiguration and space polling, but shared during log entry generation. This avoids holding a mutex over sleepable VFS operations. - Hold the sx lock over the duration of the I/O so that the vnode I/O cannot occur after vnode close, which could occur previously if accounting was disabled as a process exited. - Write the accounting log entry with Giant conditionally acquired based on the file system where the log is stored. Previously, the accounting code relied on the caller acquiring Giant. - Acquire Giant conditionally in the accounting callout based on the file system where the accounting log is stored. Run the callout MPSAFE. - Expose acct_suspended via a read-only sysctl so it is possibly to programmatically determine whether accounting is suspended or not without attempting to parse logs. - Check both acct_vp and acct_suspended lock-free before entering the accounting sx lock in acct(). - When accounting is disabled due to a VBAD vnode (i.e., forceable unmount), generate a log message indicating accounting has been disabled. - Correct a long-standing bug in how free space is calculated and compared to the required space: generate and compare signed results, not unsigned results, or negative free space will cause accounting to not be suspended when required, or worse, incorrectly resumed once negative free space is reached. MFC after: 2 weeks
2005-11-12 10:45:13 +00:00
* close the file, and (if no new file was specified, leave). Reset
* the suspended state regardless of whether accounting remains
* enabled.
*/
Significant refactoring of the accounting code to improve locking and VFS happiness, as well as correct other bugs: - Replace notion of current and saved accounting credential/vnode with a single credential/vnode and an acct_suspended flag. This simplifies the accounting logic substantially. - Replace acct_mtx with acct_sx, a sleepable lock held exclusively during reconfiguration and space polling, but shared during log entry generation. This avoids holding a mutex over sleepable VFS operations. - Hold the sx lock over the duration of the I/O so that the vnode I/O cannot occur after vnode close, which could occur previously if accounting was disabled as a process exited. - Write the accounting log entry with Giant conditionally acquired based on the file system where the log is stored. Previously, the accounting code relied on the caller acquiring Giant. - Acquire Giant conditionally in the accounting callout based on the file system where the accounting log is stored. Run the callout MPSAFE. - Expose acct_suspended via a read-only sysctl so it is possibly to programmatically determine whether accounting is suspended or not without attempting to parse logs. - Check both acct_vp and acct_suspended lock-free before entering the accounting sx lock in acct(). - When accounting is disabled due to a VBAD vnode (i.e., forceable unmount), generate a log message indicating accounting has been disabled. - Correct a long-standing bug in how free space is calculated and compared to the required space: generate and compare signed results, not unsigned results, or negative free space will cause accounting to not be suspended when required, or worse, incorrectly resumed once negative free space is reached. MFC after: 2 weeks
2005-11-12 10:45:13 +00:00
acct_suspended = 0;
if (acct_vp != NULL) {
vfslocked = VFS_LOCK_GIANT(acct_vp->v_mount);
error = acct_disable(td);
VFS_UNLOCK_GIANT(vfslocked);
}
2002-12-14 01:56:26 +00:00
if (uap->path == NULL) {
if (acct_state & ACCT_RUNNING) {
acct_state |= ACCT_EXITREQ;
wakeup(&acct_state);
}
Significant refactoring of the accounting code to improve locking and VFS happiness, as well as correct other bugs: - Replace notion of current and saved accounting credential/vnode with a single credential/vnode and an acct_suspended flag. This simplifies the accounting logic substantially. - Replace acct_mtx with acct_sx, a sleepable lock held exclusively during reconfiguration and space polling, but shared during log entry generation. This avoids holding a mutex over sleepable VFS operations. - Hold the sx lock over the duration of the I/O so that the vnode I/O cannot occur after vnode close, which could occur previously if accounting was disabled as a process exited. - Write the accounting log entry with Giant conditionally acquired based on the file system where the log is stored. Previously, the accounting code relied on the caller acquiring Giant. - Acquire Giant conditionally in the accounting callout based on the file system where the accounting log is stored. Run the callout MPSAFE. - Expose acct_suspended via a read-only sysctl so it is possibly to programmatically determine whether accounting is suspended or not without attempting to parse logs. - Check both acct_vp and acct_suspended lock-free before entering the accounting sx lock in acct(). - When accounting is disabled due to a VBAD vnode (i.e., forceable unmount), generate a log message indicating accounting has been disabled. - Correct a long-standing bug in how free space is calculated and compared to the required space: generate and compare signed results, not unsigned results, or negative free space will cause accounting to not be suspended when required, or worse, incorrectly resumed once negative free space is reached. MFC after: 2 weeks
2005-11-12 10:45:13 +00:00
sx_xunlock(&acct_sx);
return (error);
}
1994-05-24 10:09:53 +00:00
/*
* Save the new accounting file vnode, and schedule the new
* free space watcher.
1994-05-24 10:09:53 +00:00
*/
Significant refactoring of the accounting code to improve locking and VFS happiness, as well as correct other bugs: - Replace notion of current and saved accounting credential/vnode with a single credential/vnode and an acct_suspended flag. This simplifies the accounting logic substantially. - Replace acct_mtx with acct_sx, a sleepable lock held exclusively during reconfiguration and space polling, but shared during log entry generation. This avoids holding a mutex over sleepable VFS operations. - Hold the sx lock over the duration of the I/O so that the vnode I/O cannot occur after vnode close, which could occur previously if accounting was disabled as a process exited. - Write the accounting log entry with Giant conditionally acquired based on the file system where the log is stored. Previously, the accounting code relied on the caller acquiring Giant. - Acquire Giant conditionally in the accounting callout based on the file system where the accounting log is stored. Run the callout MPSAFE. - Expose acct_suspended via a read-only sysctl so it is possibly to programmatically determine whether accounting is suspended or not without attempting to parse logs. - Check both acct_vp and acct_suspended lock-free before entering the accounting sx lock in acct(). - When accounting is disabled due to a VBAD vnode (i.e., forceable unmount), generate a log message indicating accounting has been disabled. - Correct a long-standing bug in how free space is calculated and compared to the required space: generate and compare signed results, not unsigned results, or negative free space will cause accounting to not be suspended when required, or worse, incorrectly resumed once negative free space is reached. MFC after: 2 weeks
2005-11-12 10:45:13 +00:00
acct_vp = nd.ni_vp;
acct_cred = crhold(td->td_ucred);
acct_flags = flags;
if (acct_state & ACCT_RUNNING)
acct_state &= ~ACCT_EXITREQ;
else {
/*
* Try to start up an accounting kthread. We may start more
* than one, but if so the extras will commit suicide as
* soon as they start up.
*/
error = kthread_create(acct_thread, NULL, NULL, 0, 0,
"accounting");
if (error) {
vfslocked = VFS_LOCK_GIANT(acct_vp->v_mount);
(void) vn_close(acct_vp, acct_flags, acct_cred, td);
VFS_UNLOCK_GIANT(vfslocked);
crfree(acct_cred);
acct_configured = 0;
acct_vp = NULL;
acct_cred = NULL;
acct_flags = 0;
sx_xunlock(&acct_sx);
log(LOG_NOTICE, "Unable to start accounting thread\n");
return (error);
}
}
acct_configured = 1;
Significant refactoring of the accounting code to improve locking and VFS happiness, as well as correct other bugs: - Replace notion of current and saved accounting credential/vnode with a single credential/vnode and an acct_suspended flag. This simplifies the accounting logic substantially. - Replace acct_mtx with acct_sx, a sleepable lock held exclusively during reconfiguration and space polling, but shared during log entry generation. This avoids holding a mutex over sleepable VFS operations. - Hold the sx lock over the duration of the I/O so that the vnode I/O cannot occur after vnode close, which could occur previously if accounting was disabled as a process exited. - Write the accounting log entry with Giant conditionally acquired based on the file system where the log is stored. Previously, the accounting code relied on the caller acquiring Giant. - Acquire Giant conditionally in the accounting callout based on the file system where the accounting log is stored. Run the callout MPSAFE. - Expose acct_suspended via a read-only sysctl so it is possibly to programmatically determine whether accounting is suspended or not without attempting to parse logs. - Check both acct_vp and acct_suspended lock-free before entering the accounting sx lock in acct(). - When accounting is disabled due to a VBAD vnode (i.e., forceable unmount), generate a log message indicating accounting has been disabled. - Correct a long-standing bug in how free space is calculated and compared to the required space: generate and compare signed results, not unsigned results, or negative free space will cause accounting to not be suspended when required, or worse, incorrectly resumed once negative free space is reached. MFC after: 2 weeks
2005-11-12 10:45:13 +00:00
sx_xunlock(&acct_sx);
log(LOG_NOTICE, "Accounting enabled\n");
return (error);
1994-05-24 10:09:53 +00:00
}
/*
* Disable currently in-progress accounting by closing the vnode, dropping
* our reference to the credential, and clearing the vnode's flags.
*/
static int
acct_disable(struct thread *td)
{
int error;
sx_assert(&acct_sx, SX_XLOCKED);
error = vn_close(acct_vp, acct_flags, acct_cred, td);
crfree(acct_cred);
acct_configured = 0;
acct_vp = NULL;
acct_cred = NULL;
acct_flags = 0;
log(LOG_NOTICE, "Accounting disabled\n");
return (error);
}
1994-05-24 10:09:53 +00:00
/*
* Write out process accounting information, on process exit.
* Data to be written out is specified in Leffler, et al.
* and are enumerated below. (They're also noted in the system
* "acct.h" header file.)
1994-05-24 10:09:53 +00:00
*/
int
Significant refactoring of the accounting code to improve locking and VFS happiness, as well as correct other bugs: - Replace notion of current and saved accounting credential/vnode with a single credential/vnode and an acct_suspended flag. This simplifies the accounting logic substantially. - Replace acct_mtx with acct_sx, a sleepable lock held exclusively during reconfiguration and space polling, but shared during log entry generation. This avoids holding a mutex over sleepable VFS operations. - Hold the sx lock over the duration of the I/O so that the vnode I/O cannot occur after vnode close, which could occur previously if accounting was disabled as a process exited. - Write the accounting log entry with Giant conditionally acquired based on the file system where the log is stored. Previously, the accounting code relied on the caller acquiring Giant. - Acquire Giant conditionally in the accounting callout based on the file system where the accounting log is stored. Run the callout MPSAFE. - Expose acct_suspended via a read-only sysctl so it is possibly to programmatically determine whether accounting is suspended or not without attempting to parse logs. - Check both acct_vp and acct_suspended lock-free before entering the accounting sx lock in acct(). - When accounting is disabled due to a VBAD vnode (i.e., forceable unmount), generate a log message indicating accounting has been disabled. - Correct a long-standing bug in how free space is calculated and compared to the required space: generate and compare signed results, not unsigned results, or negative free space will cause accounting to not be suspended when required, or worse, incorrectly resumed once negative free space is reached. MFC after: 2 weeks
2005-11-12 10:45:13 +00:00
acct_process(struct thread *td)
{
struct acct acct;
struct timeval ut, st, tmp;
Locking for the per-process resource limits structure. - struct plimit includes a mutex to protect a reference count. The plimit structure is treated similarly to struct ucred in that is is always copy on write, so having a reference to a structure is sufficient to read from it without needing a further lock. - The proc lock protects the p_limit pointer and must be held while reading limits from a process to keep the limit structure from changing out from under you while reading from it. - Various global limits that are ints are not protected by a lock since int writes are atomic on all the archs we support and thus a lock wouldn't buy us anything. - All accesses to individual resource limits from a process are abstracted behind a simple lim_rlimit(), lim_max(), and lim_cur() API that return either an rlimit, or the current or max individual limit of the specified resource from a process. - dosetrlimit() was renamed to kern_setrlimit() to match existing style of other similar syscall helper functions. - The alpha OSF/1 compat layer no longer calls getrlimit() and setrlimit() (it didn't used the stackgap when it should have) but uses lim_rlimit() and kern_setrlimit() instead. - The svr4 compat no longer uses the stackgap for resource limits calls, but uses lim_rlimit() and kern_setrlimit() instead. - The ibcs2 compat no longer uses the stackgap for resource limits. It also no longer uses the stackgap for accessing sysctl's for the ibcs2_sysconf() syscall but uses kernel_sysctl() instead. As a result, ibcs2_sysconf() no longer needs Giant. - The p_rlimit macro no longer exists. Submitted by: mtm (mostly, I only did a few cleanups and catchups) Tested on: i386 Compiled on: alpha, amd64
2004-02-04 21:52:57 +00:00
struct plimit *newlim, *oldlim;
struct proc *p;
struct rusage *r;
Significant refactoring of the accounting code to improve locking and VFS happiness, as well as correct other bugs: - Replace notion of current and saved accounting credential/vnode with a single credential/vnode and an acct_suspended flag. This simplifies the accounting logic substantially. - Replace acct_mtx with acct_sx, a sleepable lock held exclusively during reconfiguration and space polling, but shared during log entry generation. This avoids holding a mutex over sleepable VFS operations. - Hold the sx lock over the duration of the I/O so that the vnode I/O cannot occur after vnode close, which could occur previously if accounting was disabled as a process exited. - Write the accounting log entry with Giant conditionally acquired based on the file system where the log is stored. Previously, the accounting code relied on the caller acquiring Giant. - Acquire Giant conditionally in the accounting callout based on the file system where the accounting log is stored. Run the callout MPSAFE. - Expose acct_suspended via a read-only sysctl so it is possibly to programmatically determine whether accounting is suspended or not without attempting to parse logs. - Check both acct_vp and acct_suspended lock-free before entering the accounting sx lock in acct(). - When accounting is disabled due to a VBAD vnode (i.e., forceable unmount), generate a log message indicating accounting has been disabled. - Correct a long-standing bug in how free space is calculated and compared to the required space: generate and compare signed results, not unsigned results, or negative free space will cause accounting to not be suspended when required, or worse, incorrectly resumed once negative free space is reached. MFC after: 2 weeks
2005-11-12 10:45:13 +00:00
int t, ret, vfslocked;
/*
* Lockless check of accounting condition before doing the hard
* work.
*/
Significant refactoring of the accounting code to improve locking and VFS happiness, as well as correct other bugs: - Replace notion of current and saved accounting credential/vnode with a single credential/vnode and an acct_suspended flag. This simplifies the accounting logic substantially. - Replace acct_mtx with acct_sx, a sleepable lock held exclusively during reconfiguration and space polling, but shared during log entry generation. This avoids holding a mutex over sleepable VFS operations. - Hold the sx lock over the duration of the I/O so that the vnode I/O cannot occur after vnode close, which could occur previously if accounting was disabled as a process exited. - Write the accounting log entry with Giant conditionally acquired based on the file system where the log is stored. Previously, the accounting code relied on the caller acquiring Giant. - Acquire Giant conditionally in the accounting callout based on the file system where the accounting log is stored. Run the callout MPSAFE. - Expose acct_suspended via a read-only sysctl so it is possibly to programmatically determine whether accounting is suspended or not without attempting to parse logs. - Check both acct_vp and acct_suspended lock-free before entering the accounting sx lock in acct(). - When accounting is disabled due to a VBAD vnode (i.e., forceable unmount), generate a log message indicating accounting has been disabled. - Correct a long-standing bug in how free space is calculated and compared to the required space: generate and compare signed results, not unsigned results, or negative free space will cause accounting to not be suspended when required, or worse, incorrectly resumed once negative free space is reached. MFC after: 2 weeks
2005-11-12 10:45:13 +00:00
if (acct_vp == NULL || acct_suspended)
return (0);
Significant refactoring of the accounting code to improve locking and VFS happiness, as well as correct other bugs: - Replace notion of current and saved accounting credential/vnode with a single credential/vnode and an acct_suspended flag. This simplifies the accounting logic substantially. - Replace acct_mtx with acct_sx, a sleepable lock held exclusively during reconfiguration and space polling, but shared during log entry generation. This avoids holding a mutex over sleepable VFS operations. - Hold the sx lock over the duration of the I/O so that the vnode I/O cannot occur after vnode close, which could occur previously if accounting was disabled as a process exited. - Write the accounting log entry with Giant conditionally acquired based on the file system where the log is stored. Previously, the accounting code relied on the caller acquiring Giant. - Acquire Giant conditionally in the accounting callout based on the file system where the accounting log is stored. Run the callout MPSAFE. - Expose acct_suspended via a read-only sysctl so it is possibly to programmatically determine whether accounting is suspended or not without attempting to parse logs. - Check both acct_vp and acct_suspended lock-free before entering the accounting sx lock in acct(). - When accounting is disabled due to a VBAD vnode (i.e., forceable unmount), generate a log message indicating accounting has been disabled. - Correct a long-standing bug in how free space is calculated and compared to the required space: generate and compare signed results, not unsigned results, or negative free space will cause accounting to not be suspended when required, or worse, incorrectly resumed once negative free space is reached. MFC after: 2 weeks
2005-11-12 10:45:13 +00:00
sx_slock(&acct_sx);
/*
* If accounting isn't enabled, don't bother. Have to check again
* once we own the lock in case we raced with disabling of accounting
* by another thread.
*/
Significant refactoring of the accounting code to improve locking and VFS happiness, as well as correct other bugs: - Replace notion of current and saved accounting credential/vnode with a single credential/vnode and an acct_suspended flag. This simplifies the accounting logic substantially. - Replace acct_mtx with acct_sx, a sleepable lock held exclusively during reconfiguration and space polling, but shared during log entry generation. This avoids holding a mutex over sleepable VFS operations. - Hold the sx lock over the duration of the I/O so that the vnode I/O cannot occur after vnode close, which could occur previously if accounting was disabled as a process exited. - Write the accounting log entry with Giant conditionally acquired based on the file system where the log is stored. Previously, the accounting code relied on the caller acquiring Giant. - Acquire Giant conditionally in the accounting callout based on the file system where the accounting log is stored. Run the callout MPSAFE. - Expose acct_suspended via a read-only sysctl so it is possibly to programmatically determine whether accounting is suspended or not without attempting to parse logs. - Check both acct_vp and acct_suspended lock-free before entering the accounting sx lock in acct(). - When accounting is disabled due to a VBAD vnode (i.e., forceable unmount), generate a log message indicating accounting has been disabled. - Correct a long-standing bug in how free space is calculated and compared to the required space: generate and compare signed results, not unsigned results, or negative free space will cause accounting to not be suspended when required, or worse, incorrectly resumed once negative free space is reached. MFC after: 2 weeks
2005-11-12 10:45:13 +00:00
if (acct_vp == NULL || acct_suspended) {
sx_sunlock(&acct_sx);
return (0);
}
p = td->td_proc;
/*
* Get process accounting information.
*/
PROC_LOCK(p);
/* (1) The name of the command that ran */
bcopy(p->p_comm, acct.ac_comm, sizeof acct.ac_comm);
/* (2) The amount of user and system time that was used */
Rework how we store process times in the kernel such that we always store the raw values including for child process statistics and only compute the system and user timevals on demand. - Fix the various kern_wait() syscall wrappers to only pass in a rusage pointer if they are going to use the result. - Add a kern_getrusage() function for the ABI syscalls to use so that they don't have to play stackgap games to call getrusage(). - Fix the svr4_sys_times() syscall to just call calcru() to calculate the times it needs rather than calling getrusage() twice with associated stackgap, etc. - Add a new rusage_ext structure to store raw time stats such as tick counts for user, system, and interrupt time as well as a bintime of the total runtime. A new p_rux field in struct proc replaces the same inline fields from struct proc (i.e. p_[isu]ticks, p_[isu]u, and p_runtime). A new p_crux field in struct proc contains the "raw" child time usage statistics. ruadd() has been changed to handle adding the associated rusage_ext structures as well as the values in rusage. Effectively, the values in rusage_ext replace the ru_utime and ru_stime values in struct rusage. These two fields in struct rusage are no longer used in the kernel. - calcru() has been split into a static worker function calcru1() that calculates appropriate timevals for user and system time as well as updating the rux_[isu]u fields of a passed in rusage_ext structure. calcru() uses a copy of the process' p_rux structure to compute the timevals after updating the runtime appropriately if any of the threads in that process are currently executing. It also now only locks sched_lock internally while doing the rux_runtime fixup. calcru() now only requires the caller to hold the proc lock and calcru1() only requires the proc lock internally. calcru() also no longer allows callers to ask for an interrupt timeval since none of them actually did. - calcru() now correctly handles threads executing on other CPUs. - A new calccru() function computes the child system and user timevals by calling calcru1() on p_crux. Note that this means that any code that wants child times must now call this function rather than reading from p_cru directly. This function also requires the proc lock. - This finishes the locking for rusage and friends so some of the Giant locks in exit1() and kern_wait() are now gone. - The locking in ttyinfo() has been tweaked so that a shared lock of the proctree lock is used to protect the process group rather than the process group lock. By holding this lock until the end of the function we now ensure that the process/thread that we pick to dump info about will no longer vanish while we are trying to output its info to the console. Submitted by: bde (mostly) MFC after: 1 month
2004-10-05 18:51:11 +00:00
calcru(p, &ut, &st);
acct.ac_utime = encode_comp_t(ut.tv_sec, ut.tv_usec);
acct.ac_stime = encode_comp_t(st.tv_sec, st.tv_usec);
/* (3) The elapsed time the command ran (and its starting time) */
tmp = boottime;
timevaladd(&tmp, &p->p_stats->p_start);
acct.ac_btime = tmp.tv_sec;
microuptime(&tmp);
timevalsub(&tmp, &p->p_stats->p_start);
acct.ac_etime = encode_comp_t(tmp.tv_sec, tmp.tv_usec);
/* (4) The average amount of memory used */
r = &p->p_stats->p_ru;
tmp = ut;
timevaladd(&tmp, &st);
t = tmp.tv_sec * hz + tmp.tv_usec / tick;
if (t)
acct.ac_mem = (r->ru_ixrss + r->ru_idrss + r->ru_isrss) / t;
else
acct.ac_mem = 0;
/* (5) The number of disk I/O operations done */
acct.ac_io = encode_comp_t(r->ru_inblock + r->ru_oublock, 0);
/* (6) The UID and GID of the process */
o Merge contents of struct pcred into struct ucred. Specifically, add the real uid, saved uid, real gid, and saved gid to ucred, as well as the pcred->pc_uidinfo, which was associated with the real uid, only rename it to cr_ruidinfo so as not to conflict with cr_uidinfo, which corresponds to the effective uid. o Remove p_cred from struct proc; add p_ucred to struct proc, replacing original macro that pointed. p->p_ucred to p->p_cred->pc_ucred. o Universally update code so that it makes use of ucred instead of pcred, p->p_ucred instead of p->p_pcred, cr_ruidinfo instead of p_uidinfo, cr_{r,sv}{u,g}id instead of p_*, etc. o Remove pcred0 and its initialization from init_main.c; initialize cr_ruidinfo there. o Restruction many credential modification chunks to always crdup while we figure out locking and optimizations; generally speaking, this means moving to a structure like this: newcred = crdup(oldcred); ... p->p_ucred = newcred; crfree(oldcred); It's not race-free, but better than nothing. There are also races in sys_process.c, all inter-process authorization, fork, exec, and exit. o Remove sigio->sio_ruid since sigio->sio_ucred now contains the ruid; remove comments indicating that the old arrangement was a problem. o Restructure exec1() a little to use newcred/oldcred arrangement, and use improved uid management primitives. o Clean up exit1() so as to do less work in credential cleanup due to pcred removal. o Clean up fork1() so as to do less work in credential cleanup and allocation. o Clean up ktrcanset() to take into account changes, and move to using suser_xxx() instead of performing a direct uid==0 comparision. o Improve commenting in various kern_prot.c credential modification calls to better document current behavior. In a couple of places, current behavior is a little questionable and we need to check POSIX.1 to make sure it's "right". More commenting work still remains to be done. o Update credential management calls, such as crfree(), to take into account new ruidinfo reference. o Modify or add the following uid and gid helper routines: change_euid() change_egid() change_ruid() change_rgid() change_svuid() change_svgid() In each case, the call now acts on a credential not a process, and as such no longer requires more complicated process locking/etc. They now assume the caller will do any necessary allocation of an exclusive credential reference. Each is commented to document its reference requirements. o CANSIGIO() is simplified to require only credentials, not processes and pcreds. o Remove lots of (p_pcred==NULL) checks. o Add an XXX to authorization code in nfs_lock.c, since it's questionable, and needs to be considered carefully. o Simplify posix4 authorization code to require only credentials, not processes and pcreds. Note that this authorization, as well as CANSIGIO(), needs to be updated to use the p_cansignal() and p_cansched() centralized authorization routines, as they currently do not take into account some desirable restrictions that are handled by the centralized routines, as well as being inconsistent with other similar authorization instances. o Update libkvm to take these changes into account. Obtained from: TrustedBSD Project Reviewed by: green, bde, jhb, freebsd-arch, freebsd-audit
2001-05-25 16:59:11 +00:00
acct.ac_uid = p->p_ucred->cr_ruid;
acct.ac_gid = p->p_ucred->cr_rgid;
/* (7) The terminal from which the process was started */
SESS_LOCK(p->p_session);
if ((p->p_flag & P_CONTROLT) && p->p_pgrp->pg_session->s_ttyp)
acct.ac_tty = dev2udev(p->p_pgrp->pg_session->s_ttyp->t_dev);
else
acct.ac_tty = NODEV;
SESS_UNLOCK(p->p_session);
/* (8) The boolean flags that tell how the process terminated, etc. */
acct.ac_flag = p->p_acflag;
PROC_UNLOCK(p);
/*
* Eliminate any file size rlimit.
*/
Locking for the per-process resource limits structure. - struct plimit includes a mutex to protect a reference count. The plimit structure is treated similarly to struct ucred in that is is always copy on write, so having a reference to a structure is sufficient to read from it without needing a further lock. - The proc lock protects the p_limit pointer and must be held while reading limits from a process to keep the limit structure from changing out from under you while reading from it. - Various global limits that are ints are not protected by a lock since int writes are atomic on all the archs we support and thus a lock wouldn't buy us anything. - All accesses to individual resource limits from a process are abstracted behind a simple lim_rlimit(), lim_max(), and lim_cur() API that return either an rlimit, or the current or max individual limit of the specified resource from a process. - dosetrlimit() was renamed to kern_setrlimit() to match existing style of other similar syscall helper functions. - The alpha OSF/1 compat layer no longer calls getrlimit() and setrlimit() (it didn't used the stackgap when it should have) but uses lim_rlimit() and kern_setrlimit() instead. - The svr4 compat no longer uses the stackgap for resource limits calls, but uses lim_rlimit() and kern_setrlimit() instead. - The ibcs2 compat no longer uses the stackgap for resource limits. It also no longer uses the stackgap for accessing sysctl's for the ibcs2_sysconf() syscall but uses kernel_sysctl() instead. As a result, ibcs2_sysconf() no longer needs Giant. - The p_rlimit macro no longer exists. Submitted by: mtm (mostly, I only did a few cleanups and catchups) Tested on: i386 Compiled on: alpha, amd64
2004-02-04 21:52:57 +00:00
newlim = lim_alloc();
PROC_LOCK(p);
oldlim = p->p_limit;
lim_copy(newlim, oldlim);
newlim->pl_rlimit[RLIMIT_FSIZE].rlim_cur = RLIM_INFINITY;
p->p_limit = newlim;
PROC_UNLOCK(p);
lim_free(oldlim);
/*
* Write the accounting information to the file.
*/
Significant refactoring of the accounting code to improve locking and VFS happiness, as well as correct other bugs: - Replace notion of current and saved accounting credential/vnode with a single credential/vnode and an acct_suspended flag. This simplifies the accounting logic substantially. - Replace acct_mtx with acct_sx, a sleepable lock held exclusively during reconfiguration and space polling, but shared during log entry generation. This avoids holding a mutex over sleepable VFS operations. - Hold the sx lock over the duration of the I/O so that the vnode I/O cannot occur after vnode close, which could occur previously if accounting was disabled as a process exited. - Write the accounting log entry with Giant conditionally acquired based on the file system where the log is stored. Previously, the accounting code relied on the caller acquiring Giant. - Acquire Giant conditionally in the accounting callout based on the file system where the accounting log is stored. Run the callout MPSAFE. - Expose acct_suspended via a read-only sysctl so it is possibly to programmatically determine whether accounting is suspended or not without attempting to parse logs. - Check both acct_vp and acct_suspended lock-free before entering the accounting sx lock in acct(). - When accounting is disabled due to a VBAD vnode (i.e., forceable unmount), generate a log message indicating accounting has been disabled. - Correct a long-standing bug in how free space is calculated and compared to the required space: generate and compare signed results, not unsigned results, or negative free space will cause accounting to not be suspended when required, or worse, incorrectly resumed once negative free space is reached. MFC after: 2 weeks
2005-11-12 10:45:13 +00:00
vfslocked = VFS_LOCK_GIANT(acct_vp->v_mount);
VOP_LEASE(acct_vp, td, acct_cred, LEASE_WRITE);
ret = vn_rdwr(UIO_WRITE, acct_vp, (caddr_t)&acct, sizeof (acct),
(off_t)0, UIO_SYSSPACE, IO_APPEND|IO_UNIT, acct_cred, NOCRED,
(int *)0, td);
Significant refactoring of the accounting code to improve locking and VFS happiness, as well as correct other bugs: - Replace notion of current and saved accounting credential/vnode with a single credential/vnode and an acct_suspended flag. This simplifies the accounting logic substantially. - Replace acct_mtx with acct_sx, a sleepable lock held exclusively during reconfiguration and space polling, but shared during log entry generation. This avoids holding a mutex over sleepable VFS operations. - Hold the sx lock over the duration of the I/O so that the vnode I/O cannot occur after vnode close, which could occur previously if accounting was disabled as a process exited. - Write the accounting log entry with Giant conditionally acquired based on the file system where the log is stored. Previously, the accounting code relied on the caller acquiring Giant. - Acquire Giant conditionally in the accounting callout based on the file system where the accounting log is stored. Run the callout MPSAFE. - Expose acct_suspended via a read-only sysctl so it is possibly to programmatically determine whether accounting is suspended or not without attempting to parse logs. - Check both acct_vp and acct_suspended lock-free before entering the accounting sx lock in acct(). - When accounting is disabled due to a VBAD vnode (i.e., forceable unmount), generate a log message indicating accounting has been disabled. - Correct a long-standing bug in how free space is calculated and compared to the required space: generate and compare signed results, not unsigned results, or negative free space will cause accounting to not be suspended when required, or worse, incorrectly resumed once negative free space is reached. MFC after: 2 weeks
2005-11-12 10:45:13 +00:00
VFS_UNLOCK_GIANT(vfslocked);
sx_sunlock(&acct_sx);
return (ret);
}
1994-05-24 10:09:53 +00:00
/*
* Encode_comp_t converts from ticks in seconds and microseconds
* to ticks in 1/AHZ seconds. The encoding is described in
* Leffler, et al., on page 63.
1994-05-24 10:09:53 +00:00
*/
#define MANTSIZE 13 /* 13 bit mantissa. */
#define EXPSIZE 3 /* Base 8 (3 bit) exponent. */
#define MAXFRACT ((1 << MANTSIZE) - 1) /* Maximum fractional value. */
static comp_t
Significant refactoring of the accounting code to improve locking and VFS happiness, as well as correct other bugs: - Replace notion of current and saved accounting credential/vnode with a single credential/vnode and an acct_suspended flag. This simplifies the accounting logic substantially. - Replace acct_mtx with acct_sx, a sleepable lock held exclusively during reconfiguration and space polling, but shared during log entry generation. This avoids holding a mutex over sleepable VFS operations. - Hold the sx lock over the duration of the I/O so that the vnode I/O cannot occur after vnode close, which could occur previously if accounting was disabled as a process exited. - Write the accounting log entry with Giant conditionally acquired based on the file system where the log is stored. Previously, the accounting code relied on the caller acquiring Giant. - Acquire Giant conditionally in the accounting callout based on the file system where the accounting log is stored. Run the callout MPSAFE. - Expose acct_suspended via a read-only sysctl so it is possibly to programmatically determine whether accounting is suspended or not without attempting to parse logs. - Check both acct_vp and acct_suspended lock-free before entering the accounting sx lock in acct(). - When accounting is disabled due to a VBAD vnode (i.e., forceable unmount), generate a log message indicating accounting has been disabled. - Correct a long-standing bug in how free space is calculated and compared to the required space: generate and compare signed results, not unsigned results, or negative free space will cause accounting to not be suspended when required, or worse, incorrectly resumed once negative free space is reached. MFC after: 2 weeks
2005-11-12 10:45:13 +00:00
encode_comp_t(u_long s, u_long us)
{
int exp, rnd;
exp = 0;
rnd = 0;
s *= AHZ;
s += us / (1000000 / AHZ); /* Maximize precision. */
while (s > MAXFRACT) {
rnd = s & (1 << (EXPSIZE - 1)); /* Round up? */
s >>= EXPSIZE; /* Base 8 exponent == 3 bit shift. */
exp++;
}
/* If we need to round up, do it (and handle overflow correctly). */
if (rnd && (++s > MAXFRACT)) {
s >>= EXPSIZE;
exp++;
}
/* Clean it up and polish it off. */
exp <<= MANTSIZE; /* Shift the exponent into place */
exp += s; /* and add on the mantissa. */
return (exp);
}
1994-05-24 10:09:53 +00:00
/*
2002-05-16 21:28:32 +00:00
* Periodically check the filesystem to see if accounting
* should be turned on or off. Beware the case where the vnode
* has been vgone()'d out from underneath us, e.g. when the file
* system containing the accounting file has been forcibly unmounted.
1994-05-24 10:09:53 +00:00
*/
/* ARGSUSED */
static void
acctwatch(void)
1994-05-24 10:09:53 +00:00
{
struct statfs sb;
Significant refactoring of the accounting code to improve locking and VFS happiness, as well as correct other bugs: - Replace notion of current and saved accounting credential/vnode with a single credential/vnode and an acct_suspended flag. This simplifies the accounting logic substantially. - Replace acct_mtx with acct_sx, a sleepable lock held exclusively during reconfiguration and space polling, but shared during log entry generation. This avoids holding a mutex over sleepable VFS operations. - Hold the sx lock over the duration of the I/O so that the vnode I/O cannot occur after vnode close, which could occur previously if accounting was disabled as a process exited. - Write the accounting log entry with Giant conditionally acquired based on the file system where the log is stored. Previously, the accounting code relied on the caller acquiring Giant. - Acquire Giant conditionally in the accounting callout based on the file system where the accounting log is stored. Run the callout MPSAFE. - Expose acct_suspended via a read-only sysctl so it is possibly to programmatically determine whether accounting is suspended or not without attempting to parse logs. - Check both acct_vp and acct_suspended lock-free before entering the accounting sx lock in acct(). - When accounting is disabled due to a VBAD vnode (i.e., forceable unmount), generate a log message indicating accounting has been disabled. - Correct a long-standing bug in how free space is calculated and compared to the required space: generate and compare signed results, not unsigned results, or negative free space will cause accounting to not be suspended when required, or worse, incorrectly resumed once negative free space is reached. MFC after: 2 weeks
2005-11-12 10:45:13 +00:00
int vfslocked;
sx_assert(&acct_sx, SX_XLOCKED);
/*
* If accounting was disabled before our kthread was scheduled,
* then acct_vp might be NULL. If so, just ask our kthread to
* exit and return.
*/
if (acct_vp == NULL) {
acct_state |= ACCT_EXITREQ;
return;
}
/*
* If our vnode is no longer valid, tear it down and signal the
* accounting thread to die.
*/
Significant refactoring of the accounting code to improve locking and VFS happiness, as well as correct other bugs: - Replace notion of current and saved accounting credential/vnode with a single credential/vnode and an acct_suspended flag. This simplifies the accounting logic substantially. - Replace acct_mtx with acct_sx, a sleepable lock held exclusively during reconfiguration and space polling, but shared during log entry generation. This avoids holding a mutex over sleepable VFS operations. - Hold the sx lock over the duration of the I/O so that the vnode I/O cannot occur after vnode close, which could occur previously if accounting was disabled as a process exited. - Write the accounting log entry with Giant conditionally acquired based on the file system where the log is stored. Previously, the accounting code relied on the caller acquiring Giant. - Acquire Giant conditionally in the accounting callout based on the file system where the accounting log is stored. Run the callout MPSAFE. - Expose acct_suspended via a read-only sysctl so it is possibly to programmatically determine whether accounting is suspended or not without attempting to parse logs. - Check both acct_vp and acct_suspended lock-free before entering the accounting sx lock in acct(). - When accounting is disabled due to a VBAD vnode (i.e., forceable unmount), generate a log message indicating accounting has been disabled. - Correct a long-standing bug in how free space is calculated and compared to the required space: generate and compare signed results, not unsigned results, or negative free space will cause accounting to not be suspended when required, or worse, incorrectly resumed once negative free space is reached. MFC after: 2 weeks
2005-11-12 10:45:13 +00:00
vfslocked = VFS_LOCK_GIANT(acct_vp->v_mount);
if (acct_vp->v_type == VBAD) {
(void) acct_disable(NULL);
Significant refactoring of the accounting code to improve locking and VFS happiness, as well as correct other bugs: - Replace notion of current and saved accounting credential/vnode with a single credential/vnode and an acct_suspended flag. This simplifies the accounting logic substantially. - Replace acct_mtx with acct_sx, a sleepable lock held exclusively during reconfiguration and space polling, but shared during log entry generation. This avoids holding a mutex over sleepable VFS operations. - Hold the sx lock over the duration of the I/O so that the vnode I/O cannot occur after vnode close, which could occur previously if accounting was disabled as a process exited. - Write the accounting log entry with Giant conditionally acquired based on the file system where the log is stored. Previously, the accounting code relied on the caller acquiring Giant. - Acquire Giant conditionally in the accounting callout based on the file system where the accounting log is stored. Run the callout MPSAFE. - Expose acct_suspended via a read-only sysctl so it is possibly to programmatically determine whether accounting is suspended or not without attempting to parse logs. - Check both acct_vp and acct_suspended lock-free before entering the accounting sx lock in acct(). - When accounting is disabled due to a VBAD vnode (i.e., forceable unmount), generate a log message indicating accounting has been disabled. - Correct a long-standing bug in how free space is calculated and compared to the required space: generate and compare signed results, not unsigned results, or negative free space will cause accounting to not be suspended when required, or worse, incorrectly resumed once negative free space is reached. MFC after: 2 weeks
2005-11-12 10:45:13 +00:00
VFS_UNLOCK_GIANT(vfslocked);
acct_state |= ACCT_EXITREQ;
Significant refactoring of the accounting code to improve locking and VFS happiness, as well as correct other bugs: - Replace notion of current and saved accounting credential/vnode with a single credential/vnode and an acct_suspended flag. This simplifies the accounting logic substantially. - Replace acct_mtx with acct_sx, a sleepable lock held exclusively during reconfiguration and space polling, but shared during log entry generation. This avoids holding a mutex over sleepable VFS operations. - Hold the sx lock over the duration of the I/O so that the vnode I/O cannot occur after vnode close, which could occur previously if accounting was disabled as a process exited. - Write the accounting log entry with Giant conditionally acquired based on the file system where the log is stored. Previously, the accounting code relied on the caller acquiring Giant. - Acquire Giant conditionally in the accounting callout based on the file system where the accounting log is stored. Run the callout MPSAFE. - Expose acct_suspended via a read-only sysctl so it is possibly to programmatically determine whether accounting is suspended or not without attempting to parse logs. - Check both acct_vp and acct_suspended lock-free before entering the accounting sx lock in acct(). - When accounting is disabled due to a VBAD vnode (i.e., forceable unmount), generate a log message indicating accounting has been disabled. - Correct a long-standing bug in how free space is calculated and compared to the required space: generate and compare signed results, not unsigned results, or negative free space will cause accounting to not be suspended when required, or worse, incorrectly resumed once negative free space is reached. MFC after: 2 weeks
2005-11-12 10:45:13 +00:00
return;
}
/*
Significant refactoring of the accounting code to improve locking and VFS happiness, as well as correct other bugs: - Replace notion of current and saved accounting credential/vnode with a single credential/vnode and an acct_suspended flag. This simplifies the accounting logic substantially. - Replace acct_mtx with acct_sx, a sleepable lock held exclusively during reconfiguration and space polling, but shared during log entry generation. This avoids holding a mutex over sleepable VFS operations. - Hold the sx lock over the duration of the I/O so that the vnode I/O cannot occur after vnode close, which could occur previously if accounting was disabled as a process exited. - Write the accounting log entry with Giant conditionally acquired based on the file system where the log is stored. Previously, the accounting code relied on the caller acquiring Giant. - Acquire Giant conditionally in the accounting callout based on the file system where the accounting log is stored. Run the callout MPSAFE. - Expose acct_suspended via a read-only sysctl so it is possibly to programmatically determine whether accounting is suspended or not without attempting to parse logs. - Check both acct_vp and acct_suspended lock-free before entering the accounting sx lock in acct(). - When accounting is disabled due to a VBAD vnode (i.e., forceable unmount), generate a log message indicating accounting has been disabled. - Correct a long-standing bug in how free space is calculated and compared to the required space: generate and compare signed results, not unsigned results, or negative free space will cause accounting to not be suspended when required, or worse, incorrectly resumed once negative free space is reached. MFC after: 2 weeks
2005-11-12 10:45:13 +00:00
* Stopping here is better than continuing, maybe it will be VBAD
* next time around.
2003-03-13 23:07:09 +00:00
*/
Significant refactoring of the accounting code to improve locking and VFS happiness, as well as correct other bugs: - Replace notion of current and saved accounting credential/vnode with a single credential/vnode and an acct_suspended flag. This simplifies the accounting logic substantially. - Replace acct_mtx with acct_sx, a sleepable lock held exclusively during reconfiguration and space polling, but shared during log entry generation. This avoids holding a mutex over sleepable VFS operations. - Hold the sx lock over the duration of the I/O so that the vnode I/O cannot occur after vnode close, which could occur previously if accounting was disabled as a process exited. - Write the accounting log entry with Giant conditionally acquired based on the file system where the log is stored. Previously, the accounting code relied on the caller acquiring Giant. - Acquire Giant conditionally in the accounting callout based on the file system where the accounting log is stored. Run the callout MPSAFE. - Expose acct_suspended via a read-only sysctl so it is possibly to programmatically determine whether accounting is suspended or not without attempting to parse logs. - Check both acct_vp and acct_suspended lock-free before entering the accounting sx lock in acct(). - When accounting is disabled due to a VBAD vnode (i.e., forceable unmount), generate a log message indicating accounting has been disabled. - Correct a long-standing bug in how free space is calculated and compared to the required space: generate and compare signed results, not unsigned results, or negative free space will cause accounting to not be suspended when required, or worse, incorrectly resumed once negative free space is reached. MFC after: 2 weeks
2005-11-12 10:45:13 +00:00
if (VFS_STATFS(acct_vp->v_mount, &sb, curthread) < 0) {
VFS_UNLOCK_GIANT(vfslocked);
return;
}
VFS_UNLOCK_GIANT(vfslocked);
if (acct_suspended) {
if (sb.f_bavail > (int64_t)(acctresume * sb.f_blocks /
100)) {
acct_suspended = 0;
1994-05-24 10:09:53 +00:00
log(LOG_NOTICE, "Accounting resumed\n");
}
} else {
Significant refactoring of the accounting code to improve locking and VFS happiness, as well as correct other bugs: - Replace notion of current and saved accounting credential/vnode with a single credential/vnode and an acct_suspended flag. This simplifies the accounting logic substantially. - Replace acct_mtx with acct_sx, a sleepable lock held exclusively during reconfiguration and space polling, but shared during log entry generation. This avoids holding a mutex over sleepable VFS operations. - Hold the sx lock over the duration of the I/O so that the vnode I/O cannot occur after vnode close, which could occur previously if accounting was disabled as a process exited. - Write the accounting log entry with Giant conditionally acquired based on the file system where the log is stored. Previously, the accounting code relied on the caller acquiring Giant. - Acquire Giant conditionally in the accounting callout based on the file system where the accounting log is stored. Run the callout MPSAFE. - Expose acct_suspended via a read-only sysctl so it is possibly to programmatically determine whether accounting is suspended or not without attempting to parse logs. - Check both acct_vp and acct_suspended lock-free before entering the accounting sx lock in acct(). - When accounting is disabled due to a VBAD vnode (i.e., forceable unmount), generate a log message indicating accounting has been disabled. - Correct a long-standing bug in how free space is calculated and compared to the required space: generate and compare signed results, not unsigned results, or negative free space will cause accounting to not be suspended when required, or worse, incorrectly resumed once negative free space is reached. MFC after: 2 weeks
2005-11-12 10:45:13 +00:00
if (sb.f_bavail <= (int64_t)(acctsuspend * sb.f_blocks /
100)) {
acct_suspended = 1;
1994-05-24 10:09:53 +00:00
log(LOG_NOTICE, "Accounting suspended\n");
}
}
}
/*
* The main loop for the dedicated kernel thread that periodically calls
* acctwatch().
*/
static void
acct_thread(void *dummy)
{
u_char pri;
/* This is a low-priority kernel thread. */
pri = PRI_MAX_KERN;
mtx_lock_spin(&sched_lock);
sched_prio(curthread, pri);
mtx_unlock_spin(&sched_lock);
/* If another accounting kthread is already running, just die. */
sx_xlock(&acct_sx);
if (acct_state & ACCT_RUNNING) {
sx_xunlock(&acct_sx);
kthread_exit(0);
}
acct_state |= ACCT_RUNNING;
/* Loop until we are asked to exit. */
while (!(acct_state & ACCT_EXITREQ)) {
/* Perform our periodic checks. */
acctwatch();
/*
* We check this flag again before sleeping since the
* acctwatch() might have shut down accounting and asked us
* to exit.
*/
if (!(acct_state & ACCT_EXITREQ)) {
sx_xunlock(&acct_sx);
tsleep(&acct_state, pri, "-", acctchkfreq * hz);
sx_xlock(&acct_sx);
}
}
/*
* Acknowledge the exit request and shutdown. We clear both the
* exit request and running flags.
*/
acct_state = 0;
Significant refactoring of the accounting code to improve locking and VFS happiness, as well as correct other bugs: - Replace notion of current and saved accounting credential/vnode with a single credential/vnode and an acct_suspended flag. This simplifies the accounting logic substantially. - Replace acct_mtx with acct_sx, a sleepable lock held exclusively during reconfiguration and space polling, but shared during log entry generation. This avoids holding a mutex over sleepable VFS operations. - Hold the sx lock over the duration of the I/O so that the vnode I/O cannot occur after vnode close, which could occur previously if accounting was disabled as a process exited. - Write the accounting log entry with Giant conditionally acquired based on the file system where the log is stored. Previously, the accounting code relied on the caller acquiring Giant. - Acquire Giant conditionally in the accounting callout based on the file system where the accounting log is stored. Run the callout MPSAFE. - Expose acct_suspended via a read-only sysctl so it is possibly to programmatically determine whether accounting is suspended or not without attempting to parse logs. - Check both acct_vp and acct_suspended lock-free before entering the accounting sx lock in acct(). - When accounting is disabled due to a VBAD vnode (i.e., forceable unmount), generate a log message indicating accounting has been disabled. - Correct a long-standing bug in how free space is calculated and compared to the required space: generate and compare signed results, not unsigned results, or negative free space will cause accounting to not be suspended when required, or worse, incorrectly resumed once negative free space is reached. MFC after: 2 weeks
2005-11-12 10:45:13 +00:00
sx_xunlock(&acct_sx);
kthread_exit(0);
1994-05-24 10:09:53 +00:00
}