![Alan Stern](/assets/img/avatar_default.png)
The kernel's implementation of notifier chains is unsafe. There is no protection against entries being added to or removed from a chain while the chain is in use. The issues were discussed in this thread: http://marc.theaimsgroup.com/?l=linux-kernel&m=113018709002036&w=2 We noticed that notifier chains in the kernel fall into two basic usage classes: "Blocking" chains are always called from a process context and the callout routines are allowed to sleep; "Atomic" chains can be called from an atomic context and the callout routines are not allowed to sleep. We decided to codify this distinction and make it part of the API. Therefore this set of patches introduces three new, parallel APIs: one for blocking notifiers, one for atomic notifiers, and one for "raw" notifiers (which is really just the old API under a new name). New kinds of data structures are used for the heads of the chains, and new routines are defined for registration, unregistration, and calling a chain. The three APIs are explained in include/linux/notifier.h and their implementation is in kernel/sys.c. With atomic and blocking chains, the implementation guarantees that the chain links will not be corrupted and that chain callers will not get messed up by entries being added or removed. For raw chains the implementation provides no guarantees at all; users of this API must provide their own protections. (The idea was that situations may come up where the assumptions of the atomic and blocking APIs are not appropriate, so it should be possible for users to handle these things in their own way.) There are some limitations, which should not be too hard to live with. For atomic/blocking chains, registration and unregistration must always be done in a process context since the chain is protected by a mutex/rwsem. Also, a callout routine for a non-raw chain must not try to register or unregister entries on its own chain. (This did happen in a couple of places and the code had to be changed to avoid it.) Since atomic chains may be called from within an NMI handler, they cannot use spinlocks for synchronization. Instead we use RCU. The overhead falls almost entirely in the unregister routine, which is okay since unregistration is much less frequent that calling a chain. Here is the list of chains that we adjusted and their classifications. None of them use the raw API, so for the moment it is only a placeholder. ATOMIC CHAINS ------------- arch/i386/kernel/traps.c: i386die_chain arch/ia64/kernel/traps.c: ia64die_chain arch/powerpc/kernel/traps.c: powerpc_die_chain arch/sparc64/kernel/traps.c: sparc64die_chain arch/x86_64/kernel/traps.c: die_chain drivers/char/ipmi/ipmi_si_intf.c: xaction_notifier_list kernel/panic.c: panic_notifier_list kernel/profile.c: task_free_notifier net/bluetooth/hci_core.c: hci_notifier net/ipv4/netfilter/ip_conntrack_core.c: ip_conntrack_chain net/ipv4/netfilter/ip_conntrack_core.c: ip_conntrack_expect_chain net/ipv6/addrconf.c: inet6addr_chain net/netfilter/nf_conntrack_core.c: nf_conntrack_chain net/netfilter/nf_conntrack_core.c: nf_conntrack_expect_chain net/netlink/af_netlink.c: netlink_chain BLOCKING CHAINS --------------- arch/powerpc/platforms/pseries/reconfig.c: pSeries_reconfig_chain arch/s390/kernel/process.c: idle_chain arch/x86_64/kernel/process.c idle_notifier drivers/base/memory.c: memory_chain drivers/cpufreq/cpufreq.c cpufreq_policy_notifier_list drivers/cpufreq/cpufreq.c cpufreq_transition_notifier_list drivers/macintosh/adb.c: adb_client_list drivers/macintosh/via-pmu.c sleep_notifier_list drivers/macintosh/via-pmu68k.c sleep_notifier_list drivers/macintosh/windfarm_core.c wf_client_list drivers/usb/core/notify.c usb_notifier_list drivers/video/fbmem.c fb_notifier_list kernel/cpu.c cpu_chain kernel/module.c module_notify_list kernel/profile.c munmap_notifier kernel/profile.c task_exit_notifier kernel/sys.c reboot_notifier_list net/core/dev.c netdev_chain net/decnet/dn_dev.c: dnaddr_chain net/ipv4/devinet.c: inetaddr_chain It's possible that some of these classifications are wrong. If they are, please let us know or submit a patch to fix them. Note that any chain that gets called very frequently should be atomic, because the rwsem read-locking used for blocking chains is very likely to incur cache misses on SMP systems. (However, if the chain's callout routines may sleep then the chain cannot be atomic.) The patch set was written by Alan Stern and Chandra Seetharaman, incorporating material written by Keith Owens and suggestions from Paul McKenney and Andrew Morton. [jes@sgi.com: restructure the notifier chain initialization macros] Signed-off-by: Alan Stern <stern@rowland.harvard.edu> Signed-off-by: Chandra Seetharaman <sekharan@us.ibm.com> Signed-off-by: Jes Sorensen <jes@sgi.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
272 lines
6.2 KiB
C
272 lines
6.2 KiB
C
/*
|
|
* linux/kernel/panic.c
|
|
*
|
|
* Copyright (C) 1991, 1992 Linus Torvalds
|
|
*/
|
|
|
|
/*
|
|
* This function is used through-out the kernel (including mm and fs)
|
|
* to indicate a major problem.
|
|
*/
|
|
#include <linux/config.h>
|
|
#include <linux/module.h>
|
|
#include <linux/sched.h>
|
|
#include <linux/delay.h>
|
|
#include <linux/reboot.h>
|
|
#include <linux/notifier.h>
|
|
#include <linux/init.h>
|
|
#include <linux/sysrq.h>
|
|
#include <linux/interrupt.h>
|
|
#include <linux/nmi.h>
|
|
#include <linux/kexec.h>
|
|
|
|
int panic_on_oops;
|
|
int tainted;
|
|
static int pause_on_oops;
|
|
static int pause_on_oops_flag;
|
|
static DEFINE_SPINLOCK(pause_on_oops_lock);
|
|
|
|
int panic_timeout;
|
|
EXPORT_SYMBOL(panic_timeout);
|
|
|
|
ATOMIC_NOTIFIER_HEAD(panic_notifier_list);
|
|
|
|
EXPORT_SYMBOL(panic_notifier_list);
|
|
|
|
static int __init panic_setup(char *str)
|
|
{
|
|
panic_timeout = simple_strtoul(str, NULL, 0);
|
|
return 1;
|
|
}
|
|
__setup("panic=", panic_setup);
|
|
|
|
static long no_blink(long time)
|
|
{
|
|
return 0;
|
|
}
|
|
|
|
/* Returns how long it waited in ms */
|
|
long (*panic_blink)(long time);
|
|
EXPORT_SYMBOL(panic_blink);
|
|
|
|
/**
|
|
* panic - halt the system
|
|
* @fmt: The text string to print
|
|
*
|
|
* Display a message, then perform cleanups.
|
|
*
|
|
* This function never returns.
|
|
*/
|
|
|
|
NORET_TYPE void panic(const char * fmt, ...)
|
|
{
|
|
long i;
|
|
static char buf[1024];
|
|
va_list args;
|
|
#if defined(CONFIG_S390)
|
|
unsigned long caller = (unsigned long) __builtin_return_address(0);
|
|
#endif
|
|
|
|
/*
|
|
* It's possible to come here directly from a panic-assertion and not
|
|
* have preempt disabled. Some functions called from here want
|
|
* preempt to be disabled. No point enabling it later though...
|
|
*/
|
|
preempt_disable();
|
|
|
|
bust_spinlocks(1);
|
|
va_start(args, fmt);
|
|
vsnprintf(buf, sizeof(buf), fmt, args);
|
|
va_end(args);
|
|
printk(KERN_EMERG "Kernel panic - not syncing: %s\n",buf);
|
|
bust_spinlocks(0);
|
|
|
|
/*
|
|
* If we have crashed and we have a crash kernel loaded let it handle
|
|
* everything else.
|
|
* Do we want to call this before we try to display a message?
|
|
*/
|
|
crash_kexec(NULL);
|
|
|
|
#ifdef CONFIG_SMP
|
|
/*
|
|
* Note smp_send_stop is the usual smp shutdown function, which
|
|
* unfortunately means it may not be hardened to work in a panic
|
|
* situation.
|
|
*/
|
|
smp_send_stop();
|
|
#endif
|
|
|
|
atomic_notifier_call_chain(&panic_notifier_list, 0, buf);
|
|
|
|
if (!panic_blink)
|
|
panic_blink = no_blink;
|
|
|
|
if (panic_timeout > 0) {
|
|
/*
|
|
* Delay timeout seconds before rebooting the machine.
|
|
* We can't use the "normal" timers since we just panicked..
|
|
*/
|
|
printk(KERN_EMERG "Rebooting in %d seconds..",panic_timeout);
|
|
for (i = 0; i < panic_timeout*1000; ) {
|
|
touch_nmi_watchdog();
|
|
i += panic_blink(i);
|
|
mdelay(1);
|
|
i++;
|
|
}
|
|
/* This will not be a clean reboot, with everything
|
|
* shutting down. But if there is a chance of
|
|
* rebooting the system it will be rebooted.
|
|
*/
|
|
emergency_restart();
|
|
}
|
|
#ifdef __sparc__
|
|
{
|
|
extern int stop_a_enabled;
|
|
/* Make sure the user can actually press Stop-A (L1-A) */
|
|
stop_a_enabled = 1;
|
|
printk(KERN_EMERG "Press Stop-A (L1-A) to return to the boot prom\n");
|
|
}
|
|
#endif
|
|
#if defined(CONFIG_S390)
|
|
disabled_wait(caller);
|
|
#endif
|
|
local_irq_enable();
|
|
for (i = 0;;) {
|
|
touch_softlockup_watchdog();
|
|
i += panic_blink(i);
|
|
mdelay(1);
|
|
i++;
|
|
}
|
|
}
|
|
|
|
EXPORT_SYMBOL(panic);
|
|
|
|
/**
|
|
* print_tainted - return a string to represent the kernel taint state.
|
|
*
|
|
* 'P' - Proprietary module has been loaded.
|
|
* 'F' - Module has been forcibly loaded.
|
|
* 'S' - SMP with CPUs not designed for SMP.
|
|
* 'R' - User forced a module unload.
|
|
* 'M' - Machine had a machine check experience.
|
|
* 'B' - System has hit bad_page.
|
|
*
|
|
* The string is overwritten by the next call to print_taint().
|
|
*/
|
|
|
|
const char *print_tainted(void)
|
|
{
|
|
static char buf[20];
|
|
if (tainted) {
|
|
snprintf(buf, sizeof(buf), "Tainted: %c%c%c%c%c%c",
|
|
tainted & TAINT_PROPRIETARY_MODULE ? 'P' : 'G',
|
|
tainted & TAINT_FORCED_MODULE ? 'F' : ' ',
|
|
tainted & TAINT_UNSAFE_SMP ? 'S' : ' ',
|
|
tainted & TAINT_FORCED_RMMOD ? 'R' : ' ',
|
|
tainted & TAINT_MACHINE_CHECK ? 'M' : ' ',
|
|
tainted & TAINT_BAD_PAGE ? 'B' : ' ');
|
|
}
|
|
else
|
|
snprintf(buf, sizeof(buf), "Not tainted");
|
|
return(buf);
|
|
}
|
|
|
|
void add_taint(unsigned flag)
|
|
{
|
|
tainted |= flag;
|
|
}
|
|
EXPORT_SYMBOL(add_taint);
|
|
|
|
static int __init pause_on_oops_setup(char *str)
|
|
{
|
|
pause_on_oops = simple_strtoul(str, NULL, 0);
|
|
return 1;
|
|
}
|
|
__setup("pause_on_oops=", pause_on_oops_setup);
|
|
|
|
static void spin_msec(int msecs)
|
|
{
|
|
int i;
|
|
|
|
for (i = 0; i < msecs; i++) {
|
|
touch_nmi_watchdog();
|
|
mdelay(1);
|
|
}
|
|
}
|
|
|
|
/*
|
|
* It just happens that oops_enter() and oops_exit() are identically
|
|
* implemented...
|
|
*/
|
|
static void do_oops_enter_exit(void)
|
|
{
|
|
unsigned long flags;
|
|
static int spin_counter;
|
|
|
|
if (!pause_on_oops)
|
|
return;
|
|
|
|
spin_lock_irqsave(&pause_on_oops_lock, flags);
|
|
if (pause_on_oops_flag == 0) {
|
|
/* This CPU may now print the oops message */
|
|
pause_on_oops_flag = 1;
|
|
} else {
|
|
/* We need to stall this CPU */
|
|
if (!spin_counter) {
|
|
/* This CPU gets to do the counting */
|
|
spin_counter = pause_on_oops;
|
|
do {
|
|
spin_unlock(&pause_on_oops_lock);
|
|
spin_msec(MSEC_PER_SEC);
|
|
spin_lock(&pause_on_oops_lock);
|
|
} while (--spin_counter);
|
|
pause_on_oops_flag = 0;
|
|
} else {
|
|
/* This CPU waits for a different one */
|
|
while (spin_counter) {
|
|
spin_unlock(&pause_on_oops_lock);
|
|
spin_msec(1);
|
|
spin_lock(&pause_on_oops_lock);
|
|
}
|
|
}
|
|
}
|
|
spin_unlock_irqrestore(&pause_on_oops_lock, flags);
|
|
}
|
|
|
|
/*
|
|
* Return true if the calling CPU is allowed to print oops-related info. This
|
|
* is a bit racy..
|
|
*/
|
|
int oops_may_print(void)
|
|
{
|
|
return pause_on_oops_flag == 0;
|
|
}
|
|
|
|
/*
|
|
* Called when the architecture enters its oops handler, before it prints
|
|
* anything. If this is the first CPU to oops, and it's oopsing the first time
|
|
* then let it proceed.
|
|
*
|
|
* This is all enabled by the pause_on_oops kernel boot option. We do all this
|
|
* to ensure that oopses don't scroll off the screen. It has the side-effect
|
|
* of preventing later-oopsing CPUs from mucking up the display, too.
|
|
*
|
|
* It turns out that the CPU which is allowed to print ends up pausing for the
|
|
* right duration, whereas all the other CPUs pause for twice as long: once in
|
|
* oops_enter(), once in oops_exit().
|
|
*/
|
|
void oops_enter(void)
|
|
{
|
|
do_oops_enter_exit();
|
|
}
|
|
|
|
/*
|
|
* Called when the architecture exits its oops handler, after printing
|
|
* everything.
|
|
*/
|
|
void oops_exit(void)
|
|
{
|
|
do_oops_enter_exit();
|
|
}
|