kernel-fxtec-pro1x/kernel
Thomas Gleixner d8bb6f4c16 x86: tsc prevent time going backwards
We already catch most of the TSC problems by sanity checks, but there
is a subtle bug which has been in the code forever. This can cause
time jumps in the range of hours.

This was reported in:
     http://lkml.org/lkml/2007/8/23/96
and
     http://lkml.org/lkml/2008/3/31/23

I was able to reproduce the problem with a gettimeofday loop test on a
dual core and a quad core machine which both have sychronized
TSCs. The TSCs seems not to be perfectly in sync though, but the
kernel is not able to detect the slight delta in the sync check. Still
there exists an extremly small window where this delta can be observed
with a real big time jump. So far I was only able to reproduce this
with the vsyscall gettimeofday implementation, but in theory this
might be observable with the syscall based version as well.

CPU 0 updates the clock source variables under xtime/vyscall lock and
CPU1, where the TSC is slighty behind CPU0, is reading the time right
after the seqlock was unlocked.

The clocksource reference data was updated with the TSC from CPU0 and
the value which is read from TSC on CPU1 is less than the reference
data. This results in a huge delta value due to the unsigned
subtraction of the TSC value and the reference value. This algorithm
can not be changed due to the support of wrapping clock sources like
pm timer.

The huge delta is converted to nanoseconds and added to xtime, which
is then observable by the caller. The next gettimeofday call on CPU1
will show the correct time again as now the TSC has advanced above the
reference value.

To prevent this TSC specific wreckage we need to compare the TSC value
against the reference value and return the latter when it is larger
than the actual TSC value.

I pondered to mark the TSC unstable when the readout is smaller than
the reference value, but this would render an otherwise good and fast
clocksource unusable without a real good reason.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-04-19 19:19:55 +02:00
..
irq genirq: do not leave interupts enabled on free_irq 2008-02-19 10:43:58 +01:00
power Merge branches 'release' and 'doc' into release 2008-03-13 01:59:53 -04:00
time x86: tsc prevent time going backwards 2008-04-19 19:19:55 +02:00
.gitignore Update kernel/.gitignore with new auto-generated files 2008-02-09 23:27:01 -08:00
acct.c bsd_acct: using task_struct->tgid is not right in pid-namespaces 2008-03-24 19:22:20 -07:00
audit.c Audit: internally use the new LSM audit hooks 2008-04-19 09:52:37 +10:00
audit.h SELinux: use new audit hooks, remove redundant exports 2008-04-19 09:53:46 +10:00
audit_tree.c Introduce path_put() 2008-02-14 21:13:33 -08:00
auditfilter.c Audit: Final renamings and cleanup 2008-04-19 09:59:43 +10:00
auditsc.c Audit: Final renamings and cleanup 2008-04-19 09:59:43 +10:00
backtracetest.c x86: add a simple backtrace test module 2008-01-30 13:33:08 +01:00
capability.c Add 64-bit capability support to the kernel 2008-02-05 09:44:20 -08:00
cgroup.c cgroup: fix a race condition in manipulating tsk->cg_list 2008-04-18 08:17:57 -07:00
cgroup_debug.c Task Control Groups: simple task cgroup debug info subsystem 2007-10-19 11:53:36 -07:00
compat.c hrtimer: use nanosleep specific restart_block fields 2008-04-17 12:22:30 +02:00
configs.c use simple_read_from_buffer in kernel/ 2007-05-09 12:30:49 -07:00
cpu.c cpu: fix section mismatch warnings for enable_nonboot_cpus 2008-02-08 09:22:41 -08:00
cpuset.c cpusets: fix obsolete comment 2008-03-05 17:53:33 -08:00
delayacct.c Add scaled time to taskstats based process accounting 2007-10-18 14:37:28 -07:00
dma.c whitespace fixes: DMA channel allocator 2007-10-18 14:37:24 -07:00
exec_domain.c whitespace fixes: execution domains 2007-10-18 14:37:26 -07:00
exit.c asmlinkage_protect replaces prevent_tail_call 2008-04-10 17:28:26 -07:00
extable.c module: Don't report discarded init pages as kernel text. 2008-01-29 17:13:18 +11:00
fork.c memcgroup: fix spurious EBUSY on memory cgroup removal 2008-03-28 14:45:21 -07:00
futex.c NULL noise: fs/*, mm/*, kernel/* 2008-03-30 14:18:41 -07:00
futex_compat.c futex_compat __user annotation 2008-03-30 14:18:41 -07:00
hrtimer.c hrtimers: simplify lockdep handling 2008-04-17 12:22:31 +02:00
itimer.c ITIMER_REAL: convert to use struct pid 2008-02-08 09:22:29 -08:00
kallsyms.c remove support for un-needed _extratext section 2008-02-06 10:41:01 -08:00
Kconfig.hz sched: high-res preemption tick 2008-01-25 21:08:29 +01:00
Kconfig.preempt rcu: move PREEMPT_RCU config option back under PREEMPT 2008-03-10 18:01:20 -07:00
kexec.c vmcoreinfo: add "VMCOREINFO_" to all the call for vmcoreinfo_append_str() 2008-02-07 08:42:25 -08:00
kfifo.c is_power_of_2: kernel/kfifo.c 2007-07-16 09:05:50 -07:00
kgdb.c kgdb: always use icache flush for sw breakpoints 2008-04-17 20:05:43 +02:00
kmod.c Dont touch fs_struct in usermodehelper 2008-02-14 21:13:32 -08:00
kprobes.c kprobes: fix a null pointer bug in register_kretprobe() 2008-03-04 16:35:19 -08:00
ksysfs.c Kobject: convert remaining kobject_unregister() to kobject_put() 2008-01-24 20:40:40 -08:00
kthread.c sched: fix, always create kernel threads with normal priority 2008-01-25 21:08:33 +01:00
latencytop.c sched: latencytop support 2008-01-25 21:08:34 +01:00
lockdep.c Subject: lockdep: include all lock classes in all_lock_classes 2008-02-25 23:03:02 +01:00
lockdep_internals.h [PATCH] lockdep: more chains 2006-12-07 08:39:43 -08:00
lockdep_proc.c lockdep: Avoid /proc/lockdep & lock_stat infinite output 2007-10-11 22:11:11 +02:00
Makefile Merge git://git.kernel.org/pub/scm/linux/kernel/git/mingo/linux-2.6-kgdb 2008-04-18 08:37:01 -07:00
marker.c markers: use synchronize_sched() 2008-04-02 15:28:19 -07:00
module.c modules: warn about suspicious return values from module's ->init() hook 2008-03-10 18:01:20 -07:00
mutex-debug.c kernel: remove fastcall in kernel/* 2008-02-08 09:22:31 -08:00
mutex-debug.h
mutex.c kernel: remove fastcall in kernel/* 2008-02-08 09:22:31 -08:00
mutex.h
notifier.c kernel/notifier.c should #include <linux/reboot.h> 2008-02-06 10:41:02 -08:00
ns_cgroup.c cgroups: implement namespace tracking subsystem 2007-10-19 11:53:37 -07:00
nsproxy.c namespaces: move the IPC namespace under IPC_NS option 2008-02-08 09:22:23 -08:00
panic.c ACPI: Taint kernel on ACPI table override (format corrected) 2008-02-06 22:07:51 -05:00
params.c Add new string functions strict_strto* and convert kernel params to use them 2008-02-08 09:22:41 -08:00
pid.c kernel: remove fastcall in kernel/* 2008-02-08 09:22:31 -08:00
pid_namespace.c namespaces: cleanup the code managed with PID_NS option 2008-02-08 09:22:23 -08:00
pm_qos_params.c pm qos infrastructure and interface 2008-02-05 09:44:22 -08:00
posix-cpu-timers.c posix-timers: fix shadowed variables 2008-04-17 12:22:30 +02:00
posix-timers.c hrtimer: check relative timeouts for overflow 2008-02-14 22:08:30 +01:00
printk.c Fix locking bug in "acquire_console_semaphore_for_printk()" 2008-04-15 13:09:54 -07:00
profile.c Nuke a duplicate include from profile.c 2008-02-08 09:22:34 -08:00
ptrace.c ptrace_check_attach: remove unneeded ->signal != NULL check 2008-02-08 09:22:26 -08:00
rcuclassic.c Preempt-RCU: implementation 2008-01-25 21:08:24 +01:00
rcupdate.c rcupdate: fix comment 2008-02-13 16:21:18 -08:00
rcupreempt.c rcupreempt: remove never-migrates assumption from rcu_process_callbacks() 2008-02-29 20:21:13 +01:00
rcupreempt_trace.c Preempt-RCU: implementation 2008-01-25 21:08:24 +01:00
rcutorture.c cpu-hotplug: replace lock_cpu_hotplug() with get_online_cpus() 2008-01-25 21:08:02 +01:00
relay.c relay: set an spd_release() hook for splice 2008-03-26 12:04:09 +01:00
res_counter.c Memory Resource Controller use strstrip while parsing arguments 2008-03-04 16:35:09 -08:00
resource.c [POWERPC] Add arch-specific walk_memory_remove() for 64-bit powerpc 2008-02-08 19:52:48 +11:00
rtmutex-debug.c Don't operate with pid_t in rtmutex tester 2008-02-08 09:22:41 -08:00
rtmutex-debug.h
rtmutex-tester.c Driver core: change sysdev classes to use dynamic kobject names 2008-01-24 20:40:40 -08:00
rtmutex.c hrtimer: more hrtimer_init_sleeper() fallout. 2008-02-13 15:45:36 +01:00
rtmutex.h
rtmutex_common.h Don't operate with pid_t in rtmutex tester 2008-02-08 09:22:41 -08:00
rwsem.c sched: mark rwsem functions as __sched for wchan/profiling 2007-12-18 15:21:13 +01:00
sched.c NOHZ: reevaluate idle sleep length after add_timer_on() 2008-03-26 08:28:55 +01:00
sched_debug.c sched: improve affine wakeups 2008-03-19 04:27:53 +01:00
sched_fair.c revert "sched: fix fair sleepers" 2008-04-14 14:26:23 +02:00
sched_idletask.c sched: high-res preemption tick 2008-01-25 21:08:29 +01:00
sched_rt.c sched: balance RT task resched only on runqueue 2008-03-07 16:43:00 +01:00
sched_stats.h sched: clean up kernel/sched_stat.h 2007-11-28 15:52:56 +01:00
seccomp.c make seccomp zerocost in schedule 2007-07-16 09:05:50 -07:00
semaphore.c Improve semaphore documentation 2008-04-17 10:43:01 -04:00
signal.c ptrace_signal subroutine 2008-04-18 08:17:57 -07:00
softirq.c rcu: add support for dynamic ticks and preempt rcu 2008-02-29 18:46:50 +01:00
softlockup.c softlockup: fix task state setting 2008-02-29 18:46:53 +01:00
spinlock.c spinlock: lockbreak cleanup 2008-01-30 13:31:20 +01:00
srcu.c make srcu_readers_active() static 2008-02-06 10:41:02 -08:00
stacktrace.c
stop_machine.c stopmachine: semaphore to mutex 2008-02-06 10:41:08 -08:00
sys.c generic, x86: add prctl commands PR_GET_TSC and PR_SET_TSC 2008-04-19 19:19:55 +02:00
sys_ni.c timerfd: new timerfd API 2008-02-05 09:44:07 -08:00
sysctl.c sched: revert load_balance_monitor() changes 2008-03-04 17:54:06 +01:00
sysctl_check.c constify tables in kernel/sysctl_check.c 2008-02-08 09:22:31 -08:00
taskstats.c kernel/taskstats.c: fix bogus nlmsg_free() 2007-11-14 18:45:44 -08:00
test_kprobes.c kprobes: kretprobe user entry-handler 2008-02-06 10:41:11 -08:00
time.c avoid overflows in kernel/time.c 2008-02-08 09:22:39 -08:00
timeconst.pl timeconst.pl: correct reversal of USEC_TO_HZ and HZ_TO_USEC 2008-02-12 14:29:26 -08:00
timer.c timers: simplify lockdep handling 2008-04-17 12:22:31 +02:00
tsacct.c Add scaled time to taskstats based process accounting 2007-10-18 14:37:28 -07:00
uid16.c asmlinkage_protect replaces prevent_tail_call 2008-04-10 17:28:26 -07:00
user.c sched: rt-group: make rt groups scheduling configurable 2008-02-13 15:45:40 +01:00
user_namespace.c namespaces: cleanup the code managed with the USER_NS option 2008-02-08 09:22:23 -08:00
utsname.c Fix UTS corruption during clone(CLONE_NEWUTS) 2007-09-19 11:24:17 -07:00
utsname_sysctl.c Isolate the UTS namespace's domainname and hostname back 2007-11-29 09:24:53 -08:00
wait.c kernel: remove fastcall in kernel/* 2008-02-08 09:22:31 -08:00
workqueue.c timer_list: add annotations to workqueue.c 2008-04-17 12:22:30 +02:00