Overview

This patch reworks the handling of POSIX CPU timers, including the
ITIMER_PROF, ITIMER_VIRT timers and rlimit handling.  It was put together
with the help of Roland McGrath, the owner and original writer of this code.

The problem we ran into, and the reason for this rework, has to do with using
a profiling timer in a process with a large number of threads.  It appears
that the performance of the old implementation of run_posix_cpu_timers() was
at least O(n*3) (where "n" is the number of threads in a process) or worse.
Everything is fine with an increasing number of threads until the time taken
for that routine to run becomes the same as or greater than the tick time, at
which point things degrade rather quickly.

This patch fixes bug 9906, "Weird hang with NPTL and SIGPROF."

Code Changes

This rework corrects the implementation of run_posix_cpu_timers() to make it
run in constant time for a particular machine.  (Performance may vary between
one machine and another depending upon whether the kernel is built as single-
or multiprocessor and, in the latter case, depending upon the number of
running processors.)  To do this, at each tick we now update fields in
signal_struct as well as task_struct.  The run_posix_cpu_timers() function
uses those fields to make its decisions.

We define a new structure, "task_cputime," to contain user, system and
scheduler times and use these in appropriate places:

struct task_cputime {
	cputime_t utime;
	cputime_t stime;
	unsigned long long sum_exec_runtime;
};

This is included in the structure "thread_group_cputime," which is a new
substructure of signal_struct and which varies for uniprocessor versus
multiprocessor kernels.  For uniprocessor kernels, it uses "task_cputime" as
a simple substructure, while for multiprocessor kernels it is a pointer:

struct thread_group_cputime {
	struct task_cputime totals;
};

struct thread_group_cputime {
	struct task_cputime *totals;
};

We also add a new task_cputime substructure directly to signal_struct, to
cache the earliest expiration of process-wide timers, and task_cputime also
replaces the it_*_expires fields of task_struct (used for earliest expiration
of thread timers).  The "thread_group_cputime" structure contains process-wide
timers that are updated via account_user_time() and friends.  In the non-SMP
case the structure is a simple aggregator; unfortunately in the SMP case that
simplicity was not achievable due to cache-line contention between CPUs (in
one measured case performance was actually _worse_ on a 16-cpu system than
the same test on a 4-cpu system, due to this contention).  For SMP, the
thread_group_cputime counters are maintained as a per-cpu structure allocated
using alloc_percpu().  The timer functions update only the timer field in
the structure corresponding to the running CPU, obtained using per_cpu_ptr().

We define a set of inline functions in sched.h that we use to maintain the
thread_group_cputime structure and hide the differences between UP and SMP
implementations from the rest of the kernel.  The thread_group_cputime_init()
function initializes the thread_group_cputime structure for the given task.
The thread_group_cputime_alloc() is a no-op for UP; for SMP it calls the
out-of-line function thread_group_cputime_alloc_smp() to allocate and fill
in the per-cpu structures and fields.  The thread_group_cputime_free()
function, also a no-op for UP, in SMP frees the per-cpu structures.  The
thread_group_cputime_clone_thread() function (also a UP no-op) for SMP calls
thread_group_cputime_alloc() if the per-cpu structures haven't yet been
allocated.  The thread_group_cputime() function fills the task_cputime
structure it is passed with the contents of the thread_group_cputime fields;
in UP it's that simple but in SMP it must also safely check that tsk->signal
is non-NULL (if it is it just uses the appropriate fields of task_struct) and,
if so, sums the per-cpu values for each online CPU.  Finally, the three
functions account_group_user_time(), account_group_system_time() and
account_group_exec_runtime() are used by timer functions to update the
respective fields of the thread_group_cputime structure.

Non-SMP operation is trivial and will not be mentioned further.

The per-cpu structure is always allocated when a task creates its first new
thread, via a call to thread_group_cputime_clone_thread() from copy_signal().
It is freed at process exit via a call to thread_group_cputime_free() from
cleanup_signal().

All functions that formerly summed utime/stime/sum_sched_runtime values from
from all threads in the thread group now use thread_group_cputime() to
snapshot the values in the thread_group_cputime structure or the values in
the task structure itself if the per-cpu structure hasn't been allocated.

Finally, the code in kernel/posix-cpu-timers.c has changed quite a bit.
The run_posix_cpu_timers() function has been split into a fast path and a
slow path; the former safely checks whether there are any expired thread
timers and, if not, just returns, while the slow path does the heavy lifting.
With the dedicated thread group fields, timers are no longer "rebalanced" and
the process_timer_rebalance() function and related code has gone away.  All
summing loops are gone and all code that used them now uses the
thread_group_cputime() inline.  When process-wide timers are set, the new
task_cputime structure in signal_struct is used to cache the earliest
expiration; this is checked in the fast path.

Performance

The fix appears not to add significant overhead to existing operations.  It
generally performs the same as the current code except in two cases, one in
which it performs slightly worse (Case 5 below) and one in which it performs
very significantly better (Case 2 below).  Overall it's a wash except in those
two cases.

I've since done somewhat more involved testing on a dual-core Opteron system.

Case 1: With no itimer running, for a test with 100,000 threads, the fixed
	kernel took 1428.5 seconds, 513 seconds more than the unfixed system,
	all of which was spent in the system.  There were twice as many
	voluntary context switches with the fix as without it.

Case 2: With an itimer running at .01 second ticks and 4000 threads (the most
	an unmodified kernel can handle), the fixed kernel ran the test in
	eight percent of the time (5.8 seconds as opposed to 70 seconds) and
	had better tick accuracy (.012 seconds per tick as opposed to .023
	seconds per tick).

Case 3: A 4000-thread test with an initial timer tick of .01 second and an
	interval of 10,000 seconds (i.e. a timer that ticks only once) had
	very nearly the same performance in both cases:  6.3 seconds elapsed
	for the fixed kernel versus 5.5 seconds for the unfixed kernel.

With fewer threads (eight in these tests), the Case 1 test ran in essentially
the same time on both the modified and unmodified kernels (5.2 seconds versus
5.8 seconds).  The Case 2 test ran in about the same time as well, 5.9 seconds
versus 5.4 seconds but again with much better tick accuracy, .013 seconds per
tick versus .025 seconds per tick for the unmodified kernel.

Since the fix affected the rlimit code, I also tested soft and hard CPU limits.

Case 4: With a hard CPU limit of 20 seconds and eight threads (and an itimer
	running), the modified kernel was very slightly favored in that while
	it killed the process in 19.997 seconds of CPU time (5.002 seconds of
	wall time), only .003 seconds of that was system time, the rest was
	user time.  The unmodified kernel killed the process in 20.001 seconds
	of CPU (5.014 seconds of wall time) of which .016 seconds was system
	time.  Really, though, the results were too close to call.  The results
	were essentially the same with no itimer running.

Case 5: With a soft limit of 20 seconds and a hard limit of 2000 seconds
	(where the hard limit would never be reached) and an itimer running,
	the modified kernel exhibited worse tick accuracy than the unmodified
	kernel: .050 seconds/tick versus .028 seconds/tick.  Otherwise,
	performance was almost indistinguishable.  With no itimer running this
	test exhibited virtually identical behavior and times in both cases.

In times past I did some limited performance testing.  those results are below.

On a four-cpu Opteron system without this fix, a sixteen-thread test executed
in 3569.991 seconds, of which user was 3568.435s and system was 1.556s.  On
the same system with the fix, user and elapsed time were about the same, but
system time dropped to 0.007 seconds.  Performance with eight, four and one
thread were comparable.  Interestingly, the timer ticks with the fix seemed
more accurate:  The sixteen-thread test with the fix received 149543 ticks
for 0.024 seconds per tick, while the same test without the fix received 58720
for 0.061 seconds per tick.  Both cases were configured for an interval of
0.01 seconds.  Again, the other tests were comparable.  Each thread in this
test computed the primes up to 25,000,000.

I also did a test with a large number of threads, 100,000 threads, which is
impossible without the fix.  In this case each thread computed the primes only
up to 10,000 (to make the runtime manageable).  System time dominated, at
1546.968 seconds out of a total 2176.906 seconds (giving a user time of
629.938s).  It received 147651 ticks for 0.015 seconds per tick, still quite
accurate.  There is obviously no comparable test without the fix.
Signed-off-by: NFrank Mayhar <fmayhar@google.com>
Cc: Roland McGrath <roland@redhat.com>
Cc: Alexey Dobriyan <adobriyan@gmail.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

It's useful to detect which process is killed by RT watchdog.
Signed-off-by: NHiroshi Shimamoto <h-shimamoto@ct.jp.nec.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>
Signed-off-by: NThomas Gleixner <tglx@linutronix.de>

x86 is the only arch right now, which provides an optimized for
div_long_long_rem and it has the downside that one has to be very careful that
the divide doesn't overflow.

The API is a little akward, as the arguments for the unsigned divide are
signed.  The signed version also doesn't handle a negative divisor and
produces worse code on 64bit archs.

There is little incentive to keep this API alive, so this converts the few
users to the new API.
Signed-off-by: NRoman Zippel <zippel@linux-m68k.org>
Cc: Ralf Baechle <ralf@linux-mips.org>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: john stultz <johnstul@us.ibm.com>
Cc: Christoph Lameter <clameter@sgi.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Fix sparse warnings like this:
kernel/posix-cpu-timers.c:1090:25: warning: symbol 't' shadows an earlier one
kernel/posix-cpu-timers.c:1058:21: originally declared here
Signed-off-by: NWANG Cong <xiyou.wangcong@gmail.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NThomas Gleixner <tglx@linutronix.de>

All the functions that need to lookup a task by pid in posix timers obtain
this pid from a user space, and thus this value refers to a task in the same
namespace, as the current task lives in.

So the proper behavior is to call find_task_by_vpid() here.
Signed-off-by: NPavel Emelyanov <xemul@openvz.org>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Remove the curious logic to set it_sched_expires in the future. It useless
because rt.timeout wouldn't be incremented anyway.

Explicity check for RLIM_INFINITY as a test programm that had a 1s soft limit
and a inf hard limit would SIGKILL at 1s. This is because RLIM_INFINITY+d-1
is d-2.
Signed-off-by: NPeter Zijlsta <a.p.zijlstra@chello.nl>
CC: Michal Schmidt <mschmidt@redhat.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Introduce a new rlimit that allows the user to set a runtime timeout on
real-time tasks their slice. Once this limit is exceeded the task will receive
SIGXCPU.

So it measures runtime since the last sleep.

Input and ideas by Thomas Gleixner and Lennart Poettering.
Signed-off-by: NPeter Zijlstra <a.p.zijlstra@chello.nl>
CC: Lennart Poettering <mzxreary@0pointer.de>
CC: Michael Kerrisk <mtk.manpages@googlemail.com>
CC: Ulrich Drepper <drepper@redhat.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

With pid namespaces this field is now dangerous to use explicitly, so hide
it behind the helpers.

Also the pid and pgrp fields o task_struct and signal_struct are to be
deprecated.  Unfortunately this patch cannot be sent right now as this
leads to tons of warnings, so start isolating them, and deprecate later.

Actually the p->tgid == pid has to be changed to has_group_leader_pid(),
but Oleg pointed out that in case of posix cpu timers this is the same, and
thread_group_leader() is more preferable.
Signed-off-by: NPavel Emelyanov <xemul@openvz.org>
Acked-by: NOleg Nesterov <oleg@tv-sign.ru>
Cc: Sukadev Bhattiprolu <sukadev@us.ibm.com>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

update the posix-cpu-timers code to use CFS's CPU accounting information.
Signed-off-by: NIngo Molnar <mingo@elte.hu>

There are many places in the kernel where the construction like

   foo = list_entry(head->next, struct foo_struct, list);

are used.
The code might look more descriptive and neat if using the macro

   list_first_entry(head, type, member) \
             list_entry((head)->next, type, member)

Here is the macro itself and the examples of its usage in the generic code.
 If it will turn out to be useful, I can prepare the set of patches to
inject in into arch-specific code, drivers, networking, etc.
Signed-off-by: NPavel Emelianov <xemul@openvz.org>
Signed-off-by: NKirill Korotaev <dev@openvz.org>
Cc: Randy Dunlap <randy.dunlap@oracle.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Zach Brown <zach.brown@oracle.com>
Cc: Davide Libenzi <davidel@xmailserver.org>
Cc: John McCutchan <ttb@tentacle.dhs.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: john stultz <johnstul@us.ibm.com>
Cc: Ram Pai <linuxram@us.ibm.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Use RCU to avoid the need to acquire tasklist_lock in the single-threaded
case of clock_gettime().  It still acquires tasklist_lock when for a
(potentially multithreaded) process.  This change allows realtime
applications to frequently monitor CPU consumption of individual tasks, as
requested (and now deployed) by some off-list users.

This has been in Ingo Molnar's -rt patchset since late 2005 with no
problems reported, and tests successfully on 2.6.20-rc6, so I believe that
it is long-since ready for mainline adoption.

[paulmck@linux.vnet.ibm.com: fix exit()/posix_cpu_clock_get() race spotted by Oleg]
Signed-off-by: NPaul E. McKenney <paulmck@linux.vnet.ibm.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: john stultz <johnstul@us.ibm.com>
Cc: Roman Zippel <zippel@linux-m68k.org>
Cc: Oleg Nesterov <oleg@tv-sign.ru>
Signed-off-by: NPaul E. McKenney <paulmck@linux.vnet.ibm.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

The integer divisions in the timer accounting code can round the result
down to 0.  Adding 0 is without effect and the signal delivery stops.

Clamp the division result to minimum 1 to avoid this.

Problem was reported by Seongbae Park <spark@google.com>, who provided
also an inital patch.

Roland sayeth:

  I have had some more time to think about the problem, and to reproduce it
  using Toyo's test case.  For the record, if my understanding of the problem
  is correct, this happens only in one very particular case.  First, the
  expiry time has to be so soon that in cputime_t units (usually 1s/HZ ticks)
  it's < nthreads so the division yields zero.  Second, it only affects each
  thread that is so new that its CPU time accumulation is zero so now+0 is
  still zero and ->it_*_expires winds up staying zero.  For the VIRT and PROF
  clocks when cputime_t is tick granularity (or the SCHED clock on
  configurations where sched_clock's value only advances on clock ticks), this
  is not hard to arrange with new threads starting up and blocking before they
  accumulate a whole tick of CPU time.  That's what happens in Toyo's test
  case.

  Note that in general it is fine for that division to round down to zero,
  and set each thread's expiry time to its "now" time.  The problem only
  arises with thread's whose "now" value is still zero, so that now+0 winds up
  0 and is interpreted as "not set" instead of ">= now".  So it would be a
  sufficient and more precise fix to just use max(ticks, 1) inside the loop
  when setting each it_*_expires value.

  But, it does no harm to round the division up to one and always advance
  every thread's expiry time.  If the thread didn't already fire timers for
  the expiry time of "now", there is no expectation that it will do so before
  the next tick anyway.  So I followed Thomas's patch in lifting the max out
  of the loops.

  This patch also covers the reload cases, which are harder to write a test
  for (and I didn't try).  I've tested it with Toyo's case and it fixes that.

[toyoa@mvista.com: fix: min_t -> max_t]
Signed-off-by: NThomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: NRoland McGrath <roland@redhat.com>
Cc: Daniel Walker <dwalker@mvista.com>
Cc: Toyo Abe <toyoa@mvista.com>
Cc: john stultz <johnstul@us.ibm.com>
Cc: Roman Zippel <zippel@linux-m68k.org>
Cc: Seongbae Park <spark@google.com>
Cc: Peter Mattis <pmattis@google.com>
Cc: Rohit Seth <rohitseth@google.com>
Cc: Martin Bligh <mbligh@google.com>
Cc: <stable@kernel.org>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

When a posix_cpu_nsleep() sleep is interrupted by a signal more than twice, it
incorrectly reports the sleep time remaining to the user.  Because
posix_cpu_nsleep() doesn't report back to the user when it's called from
restart function due to the wrong flags handling.

This patch, which applies after previous one, moves the nanosleep() function
from posix_cpu_nsleep() to do_cpu_nanosleep() and cleans up the flags handling
appropriately.
Signed-off-by: NToyo Abe <toyoa@mvista.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Roland McGrath <roland@redhat.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

The clock_nanosleep() function does not return the time remaining when the
sleep is interrupted by a signal.

This patch creates a new call out, compat_clock_nanosleep_restart(), which
handles returning the remaining time after a sleep is interrupted.  This
patch revives clock_nanosleep_restart().  It is now accessed via the new
call out.  The compat_clock_nanosleep_restart() is used for compatibility
access.

Since this is implemented in compatibility mode the normal path is
virtually unaffected - no real performance impact.
Signed-off-by: NToyo Abe <toyoa@mvista.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Roland McGrath <roland@redhat.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

arm_timer() checks PF_EXITING to prevent BUG_ON(->exit_state)
in run_posix_cpu_timers().

However, for some reason it does so only for CPUCLOCK_PERTHREAD
case (which is imho wrong).

Also, this check is not reliable, PF_EXITING could be set on
another cpu without any locks/barriers just after the check,
so it can't prevent from attaching the timer to the exiting
task.

The previous patch makes this check unneeded.
Signed-off-by: NOleg Nesterov <oleg@tv-sign.ru>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

do_exit() clears ->it_##clock##_expires, but nothing prevents
another cpu to attach the timer to exiting process after that.
arm_timer() tries to protect against this race, but the check
is racy.

After exit_notify() does 'write_unlock_irq(&tasklist_lock)' and
before do_exit() calls 'schedule() local timer interrupt can find
tsk->exit_state != 0. If that state was EXIT_DEAD (or another cpu
does sys_wait4) interrupted task has ->signal == NULL.

At this moment exiting task has no pending cpu timers, they were
cleanuped in __exit_signal()->posix_cpu_timers_exit{,_group}(),
so we can just return from irq.

John Stultz recently confirmed this bug, see

	http://marc.theaimsgroup.com/?l=linux-kernel&m=115015841413687Signed-off-by: NOleg Nesterov <oleg@tv-sign.ru>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

If the local timer interrupt happens just after do_exit() sets PF_EXITING
(and before it clears ->it_xxx_expires) run_posix_cpu_timers() will call
check_process_timers() with tasklist_lock + ->siglock held and

	check_process_timers:

		t = tsk;
		do {
			....

			do {
				t = next_thread(t);
			} while (unlikely(t->flags & PF_EXITING));
		} while (t != tsk);

the outer loop will never stop.

Actually, the window is bigger.  Another process can attach the timer
after ->it_xxx_expires was cleared (see the next commit) and the 'if
(PF_EXITING)' check in arm_timer() is racy (see the one after that).
Signed-off-by: NOleg Nesterov <oleg@tv-sign.ru>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Switch clock_nanosleep to use the new nanosleep functions in hrtimer.c
Signed-off-by: NThomas Gleixner <tglx@linutronix.de>
Signed-off-by: NIngo Molnar <mingo@elte.hu>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

add const arguments to the posix-timers.h API functions
Signed-off-by: NThomas Gleixner <tglx@linutronix.de>
Signed-off-by: NIngo Molnar <mingo@elte.hu>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

I've spent the past 3 days digging into a glibc testsuite failure in
current CVS, specifically libc/rt/tst-cputimer1.c The thr1 and thr2
timers fire too early in the second pass of this test.  The second
pass is noteworthy because it makes use of intervals, whereas the
first pass does not.

All throughout the posix-cpu-timers.c code, the calculation of the
process sched_time sum is implemented roughly as:

	unsigned long long sum;

	sum = tsk->signal->sched_time;
	t = tsk;
	do {
		sum += t->sched_time;
		t = next_thread(t);
	} while (t != tsk);

In fact this is the exact scheme used by check_process_timers().

In the case of check_process_timers(), current->sched_time has just
been updated (via scheduler_tick(), which is invoked by
update_process_times(), which subsequently invokes
run_posix_cpu_timers()) So there is no special processing necessary
wrt. that.

In other contexts, we have to allot for the fact that tsk->sched_time
might be a bit out of date if we are current.  And the
posix-cpu-timers.c code uses current_sched_time() to deal with that.

Unfortunately it does so in an erroneous and inconsistent manner in
one spot which is what results in the early timer firing.

In cpu_clock_sample_group_locked(), it does this:

		cpu->sched = p->signal->sched_time;
		/* Add in each other live thread.  */
		while ((t = next_thread(t)) != p) {
			cpu->sched += t->sched_time;
		}
		if (p->tgid == current->tgid) {
			/*
			 * We're sampling ourselves, so include the
			 * cycles not yet banked.  We still omit
			 * other threads running on other CPUs,
			 * so the total can always be behind as
			 * much as max(nthreads-1,ncpus) * (NSEC_PER_SEC/HZ).
			 */
			cpu->sched += current_sched_time(current);
		} else {
			cpu->sched += p->sched_time;
		}

The problem is the "p->tgid == current->tgid" test.  If "p" is
not current, and the tgids are the same, we will add the process
t->sched_time twice into cpu->sched and omit "p"'s sched_time
which is very very very wrong.

posix-cpu-timers.c has a helper function, sched_ns(p) which takes care
of this, so my fix is to use that here instead of this special tgid
test.

The fact that current can be one of the sub-threads of "p" points out
that we could make things a little bit more accurate, perhaps by using
sched_ns() on every thread we process in these loops.  It also points
out that we don't use the most accurate value for threads in the group
actively running other cpus (and this is mentioned in the comment).

But that is a future enhancement, and this fix here definitely makes
sense.
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

fix 32bit overflow in timespec_to_sample()
Signed-off-by: NOleg Nesterov <oleg@tv-sign.ru>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

!unlikely(expr) hurts my brain.   likely(!expr) is more straightforward.
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

This change corrects an omission in posix_cpu_timer_schedule, so that it
correctly propagates the overrun calculation to where it will get reported
to the user.
Signed-off-by: NRoland McGrath <roland@redhat.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

This just makes sure that a thread's expiry times can't get reset after
it clears them in do_exit.

This is what allowed us to re-introduce the stricter BUG_ON() check in
a362f463.
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

This reverts commit 3de463c7.

Roland has another patch that allows us to leave the BUG_ON() in place
by just making sure that the condition it tests for really is always
true.

That goes in next.
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

There's a silly off-by-one error in the code that updates the expiration
of posix CPU timers, causing them to not be properly updated when they
hit exactly on their expiration time (which should be the normal case).

This causes them to then fire immediately again, and only _then_ get
properly updated.
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Pointed out by Oleg Nesterov, who has been walking over the code
forwards and backwards.
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

This might be harmless, but looks like a race from code inspection (I
was unable to trigger it).  I must admit, I don't understand why we
can't return TIMER_RETRY after 'spin_unlock(&p->sighand->siglock)'
without doing bump_cpu_timer(), but this is what original code does.

posix_cpu_timer_set:

	read_lock(&tasklist_lock);

	spin_lock(&p->sighand->siglock);
	list_del_init(&timer->it.cpu.entry);
	spin_unlock(&p->sighand->siglock);

We are probaly deleting the timer from run_posix_cpu_timers's 'firing'
local list_head while run_posix_cpu_timers() does list_for_each_safe.

Various bad things can happen, for example we can just delete this timer
so that list_for_each() will not notice it and run_posix_cpu_timers()
will not reset '->firing' flag. In that case,

	....

	if (timer->it.cpu.firing) {
		read_unlock(&tasklist_lock);
		timer->it.cpu.firing = -1;
		return TIMER_RETRY;
	}

sys_timer_settime() goes to 'retry:', calls posix_cpu_timer_set() again,
it returns TIMER_RETRY ...
Signed-off-by: NOleg Nesterov <oleg@tv-sign.ru>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

No need to rebalance when task exited
Signed-off-by: NOleg Nesterov <oleg@tv-sign.ru>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

do_exit() clears ->it_##clock##_expires, but nothing prevents
another cpu to attach the timer to exiting process after that.

After exit_notify() does 'write_unlock_irq(&tasklist_lock)' and
before do_exit() calls 'schedule() local timer interrupt can find
tsk->exit_state != 0. If that state was EXIT_DEAD (or another cpu
does sys_wait4) interrupted task has ->signal == NULL.

At this moment exiting task has no pending cpu timers, they were cleaned
up in __exit_signal()->posix_cpu_timers_exit{,_group}(), so we can just
return from irq.
Signed-off-by: NOleg Nesterov <oleg@tv-sign.ru>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

1. cleanup_timers() sets timer->task = NULL under tasklist + ->sighand locks.
   That means that this code in posix_cpu_timer_del() and posix_cpu_timer_set()

   		lock_timer(timer);
		if (timer->task == NULL)
			return;
		read_lock(tasklist);
		put_task_struct(timer->task)

   is racy. With this patch timer->task modified and accounted only under
   timer->it_lock. Sadly, this means that dead task_struct won't be freed
   until timer deleted or armed.

2. run_posix_cpu_timers() collects expired timers into local list under
   tasklist + ->sighand again. That means that posix_cpu_timer_del()
   should check timer->it.cpu.firing under these locks too.
Signed-off-by: NOleg Nesterov <oleg@tv-sign.ru>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Bursty timers aren't good for anybody, very much including latency for
other programs when we trigger lots of timers in interrupt context.  So
set a random limit, after which we'll handle the rest on the next timer
tick.

Noted by Oleg Nesterov <oleg@tv-sign.ru>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Revert commit e03d13e9, to be replaced
by a much nicer fix from Roland.

Oleg Nesterov reported an SMP deadlock.  If there is a running timer
tracking a different process's CPU time clock when the process owning
the timer exits, we deadlock on tasklist_lock in posix_cpu_timer_del via
exit_itimers.

That code was using tasklist_lock to check for a race with __exit_signal
being called on the timer-target task and clearing its ->signal.
However, there is actually no such race.  __exit_signal will have called
posix_cpu_timers_exit and posix_cpu_timers_exit_group before it does
that.  Those will clear those k_itimer's association with the dying
task, so posix_cpu_timer_del will return early and never reach the code
in question.

In addition, posix_cpu_timer_del called from exit_itimers during execve
or directly from timer_delete in the process owning the timer can race
with an exiting timer-target task to cause a double put on timer-target
task struct.  Make sure we always access cpu_timers lists with sighand
lock held.
Signed-off-by: NRoland McGrath <roland@redhat.com>
Signed-off-by: NChris Wright <chrisw@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Make sure we release the task struct properly when releasing pending
timers.

release_task() does write_lock_irq(&tasklist_lock), so it can't race
with run_posix_cpu_timers() on any cpu.
Signed-off-by: NOleg Nesterov <oleg@tv-sign.ru>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Initial git repository build. I'm not bothering with the full history,
even though we have it. We can create a separate "historical" git
archive of that later if we want to, and in the meantime it's about
3.2GB when imported into git - space that would just make the early
git days unnecessarily complicated, when we don't have a lot of good
infrastructure for it.

Let it rip!