All of lore.kernel.org
 help / color / mirror / Atom feed
* [GIT PULL] timer fixes
@ 2022-08-13 10:25 Ingo Molnar
  2022-08-13 16:25 ` Borislav Petkov
  2022-08-13 21:48 ` pr-tracker-bot
  0 siblings, 2 replies; 69+ messages in thread
From: Ingo Molnar @ 2022-08-13 10:25 UTC (permalink / raw)
  To: Linus Torvalds
  Cc: linux-kernel, Thomas Gleixner, Borislav Petkov, Peter Zijlstra

Linus,

Please pull the latest timers/urgent git tree from:

   git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git timers-urgent-2022-08-13

   # HEAD: e362359ace6f87c201531872486ff295df306d13 posix-cpu-timers: Cleanup CPU timers before freeing them during exec

Misc timer fixes:

 - fix a potential use-after-free bug in posix timers
 - correct a prototype
 - address a build warning

 Thanks,

	Ingo

------------------>
Jiri Slaby (1):
      posix-timers: Make do_clock_gettime() static

Thadeu Lima de Souza Cascardo (1):
      posix-cpu-timers: Cleanup CPU timers before freeing them during exec

Youngmin Nam (1):
      time: Correct the prototype of ns_to_kernel_old_timeval and ns_to_timespec64


 fs/exec.c                 | 3 +++
 include/linux/time64.h    | 2 +-
 kernel/time/posix-stubs.c | 3 ++-
 kernel/time/time.c        | 4 ++--
 4 files changed, 8 insertions(+), 4 deletions(-)

diff --git a/fs/exec.c b/fs/exec.c
index 5fd73915c62c..f793221f4eb6 100644
--- a/fs/exec.c
+++ b/fs/exec.c
@@ -1304,6 +1304,9 @@ int begin_new_exec(struct linux_binprm * bprm)
 	bprm->mm = NULL;
 
 #ifdef CONFIG_POSIX_TIMERS
+	spin_lock_irq(&me->sighand->siglock);
+	posix_cpu_timers_exit(me);
+	spin_unlock_irq(&me->sighand->siglock);
 	exit_itimers(me);
 	flush_itimer_signals();
 #endif
diff --git a/include/linux/time64.h b/include/linux/time64.h
index 2fb8232cff1d..f1bcea8c124a 100644
--- a/include/linux/time64.h
+++ b/include/linux/time64.h
@@ -145,7 +145,7 @@ static inline s64 timespec64_to_ns(const struct timespec64 *ts)
  *
  * Returns the timespec64 representation of the nsec parameter.
  */
-extern struct timespec64 ns_to_timespec64(const s64 nsec);
+extern struct timespec64 ns_to_timespec64(s64 nsec);
 
 /**
  * timespec64_add_ns - Adds nanoseconds to a timespec64
diff --git a/kernel/time/posix-stubs.c b/kernel/time/posix-stubs.c
index fcb3b21d8bdc..90ea5f373e50 100644
--- a/kernel/time/posix-stubs.c
+++ b/kernel/time/posix-stubs.c
@@ -70,7 +70,7 @@ SYSCALL_DEFINE2(clock_settime, const clockid_t, which_clock,
 	return do_sys_settimeofday64(&new_tp, NULL);
 }
 
-int do_clock_gettime(clockid_t which_clock, struct timespec64 *tp)
+static int do_clock_gettime(clockid_t which_clock, struct timespec64 *tp)
 {
 	switch (which_clock) {
 	case CLOCK_REALTIME:
@@ -90,6 +90,7 @@ int do_clock_gettime(clockid_t which_clock, struct timespec64 *tp)
 
 	return 0;
 }
+
 SYSCALL_DEFINE2(clock_gettime, const clockid_t, which_clock,
 		struct __kernel_timespec __user *, tp)
 {
diff --git a/kernel/time/time.c b/kernel/time/time.c
index 29923b20e0e4..526257b3727c 100644
--- a/kernel/time/time.c
+++ b/kernel/time/time.c
@@ -449,7 +449,7 @@ time64_t mktime64(const unsigned int year0, const unsigned int mon0,
 }
 EXPORT_SYMBOL(mktime64);
 
-struct __kernel_old_timeval ns_to_kernel_old_timeval(const s64 nsec)
+struct __kernel_old_timeval ns_to_kernel_old_timeval(s64 nsec)
 {
 	struct timespec64 ts = ns_to_timespec64(nsec);
 	struct __kernel_old_timeval tv;
@@ -503,7 +503,7 @@ EXPORT_SYMBOL(set_normalized_timespec64);
  *
  * Returns the timespec64 representation of the nsec parameter.
  */
-struct timespec64 ns_to_timespec64(const s64 nsec)
+struct timespec64 ns_to_timespec64(s64 nsec)
 {
 	struct timespec64 ts = { 0, 0 };
 	s32 rem;

^ permalink raw reply related	[flat|nested] 69+ messages in thread

* Re: [GIT PULL] timer fixes
  2022-08-13 10:25 [GIT PULL] timer fixes Ingo Molnar
@ 2022-08-13 16:25 ` Borislav Petkov
  2022-08-13 20:27   ` Linus Torvalds
  2022-08-13 21:48 ` pr-tracker-bot
  1 sibling, 1 reply; 69+ messages in thread
From: Borislav Petkov @ 2022-08-13 16:25 UTC (permalink / raw)
  To: Ingo Molnar, Linus Torvalds, Thomas Gleixner
  Cc: linux-kernel, Peter Zijlstra, x86-ml

Might wanna hold off on pulling that:

On Sat, Aug 13, 2022 at 12:25:51PM +0200, Ingo Molnar wrote:
> Linus,
> 
> Please pull the latest timers/urgent git tree from:
> 
>    git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git timers-urgent-2022-08-13
> 
>    # HEAD: e362359ace6f87c201531872486ff295df306d13 posix-cpu-timers: Cleanup CPU timers before freeing them during exec

...

> Thadeu Lima de Souza Cascardo (1):
>       posix-cpu-timers: Cleanup CPU timers before freeing them during exec

...

> diff --git a/fs/exec.c b/fs/exec.c
> index 5fd73915c62c..f793221f4eb6 100644
> --- a/fs/exec.c
> +++ b/fs/exec.c
> @@ -1304,6 +1304,9 @@ int begin_new_exec(struct linux_binprm * bprm)
>  	bprm->mm = NULL;
>  
>  #ifdef CONFIG_POSIX_TIMERS
> +	spin_lock_irq(&me->sighand->siglock);
> +	posix_cpu_timers_exit(me);
> +	spin_unlock_irq(&me->sighand->siglock);

sparse is not happy about this:

https://lore.kernel.org/r/202208140040.MMi4z6Ek-lkp@intel.com

That task_struct.sighand is marked __rcu and thus noderef and sparse
complains:

fs/exec.c:1307:26: sparse: sparse: incorrect type in argument 1 (different address spaces) @@     expected struct spinlock [usertype] *lock
+@@     got struct spinlock [noderef] __rcu * @@

-- 
Regards/Gruss,
    Boris.

https://people.kernel.org/tglx/notes-about-netiquette

^ permalink raw reply	[flat|nested] 69+ messages in thread

* Re: [GIT PULL] timer fixes
  2022-08-13 16:25 ` Borislav Petkov
@ 2022-08-13 20:27   ` Linus Torvalds
  2022-08-14  8:57     ` Ingo Molnar
  2022-08-14 12:41     ` Borislav Petkov
  0 siblings, 2 replies; 69+ messages in thread
From: Linus Torvalds @ 2022-08-13 20:27 UTC (permalink / raw)
  To: Borislav Petkov
  Cc: Ingo Molnar, Thomas Gleixner, linux-kernel, Peter Zijlstra, x86-ml

On Sat, Aug 13, 2022 at 9:25 AM Borislav Petkov <bp@alien8.de> wrote:
>
> That task_struct.sighand is marked __rcu and thus noderef and sparse
> complains:

I think that RCU marking is misleading.

Doing a

        git grep -e '->sighand'

shows that we basically never treat that as some kind of RCU pointer.

Adding a

        grep -i rcu

to the above shows that we have a couple of places that do this
carefully, but they are the exception rather than the rule.

I think the issue is that "current->sighand" is always safe (and that
"me->sighand" is the same thing), and that sighand has RCU-delayed
freeing so that __lock_task_sighand() can safely try to take the lock
of another process' sighand.

And we have no real way to explain to sparse that *some* cases are
fine, others are not and need the sighand lock (after that careful
__lock_task_sighand thing).

              Linus

^ permalink raw reply	[flat|nested] 69+ messages in thread

* Re: [GIT PULL] timer fixes
  2022-08-13 10:25 [GIT PULL] timer fixes Ingo Molnar
  2022-08-13 16:25 ` Borislav Petkov
@ 2022-08-13 21:48 ` pr-tracker-bot
  1 sibling, 0 replies; 69+ messages in thread
From: pr-tracker-bot @ 2022-08-13 21:48 UTC (permalink / raw)
  To: Ingo Molnar
  Cc: Linus Torvalds, linux-kernel, Thomas Gleixner, Borislav Petkov,
	Peter Zijlstra

The pull request you sent on Sat, 13 Aug 2022 12:25:51 +0200:

> git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git timers-urgent-2022-08-13

has been merged into torvalds/linux.git:
https://git.kernel.org/torvalds/c/f6eb0fed6a3957c0b93e3a00c1ffaad84d4ffc31

Thank you!

-- 
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/prtracker.html

^ permalink raw reply	[flat|nested] 69+ messages in thread

* Re: [GIT PULL] timer fixes
  2022-08-13 20:27   ` Linus Torvalds
@ 2022-08-14  8:57     ` Ingo Molnar
  2022-08-14 12:41     ` Borislav Petkov
  1 sibling, 0 replies; 69+ messages in thread
From: Ingo Molnar @ 2022-08-14  8:57 UTC (permalink / raw)
  To: Linus Torvalds
  Cc: Borislav Petkov, Thomas Gleixner, linux-kernel, Peter Zijlstra, x86-ml


* Linus Torvalds <torvalds@linux-foundation.org> wrote:

> I think the issue is that "current->sighand" is always safe (and that 
> "me->sighand" is the same thing), and that sighand has RCU-delayed 
> freeing so that __lock_task_sighand() can safely try to take the lock of 
> another process' sighand.

Yeah - 'me' (== current) here can never go away from under the locking 
context which is 'me' as well.

Thanks,

	Ingo

^ permalink raw reply	[flat|nested] 69+ messages in thread

* Re: [GIT PULL] timer fixes
  2022-08-13 20:27   ` Linus Torvalds
  2022-08-14  8:57     ` Ingo Molnar
@ 2022-08-14 12:41     ` Borislav Petkov
  2022-08-14 17:24       ` Paul E. McKenney
  1 sibling, 1 reply; 69+ messages in thread
From: Borislav Petkov @ 2022-08-14 12:41 UTC (permalink / raw)
  To: Linus Torvalds
  Cc: Ingo Molnar, Thomas Gleixner, linux-kernel, Peter Zijlstra,
	x86-ml, Madhuparna Bhowmik, Paul E. McKenney, Christian Brauner

On Sat, Aug 13, 2022 at 01:27:40PM -0700, Linus Torvalds wrote:
> On Sat, Aug 13, 2022 at 9:25 AM Borislav Petkov <bp@alien8.de> wrote:
> >
> > That task_struct.sighand is marked __rcu and thus noderef and sparse
> > complains:
> 
> I think that RCU marking is misleading.
> 
> Doing a
> 
>         git grep -e '->sighand'
> 
> shows that we basically never treat that as some kind of RCU pointer.
> 
> Adding a
> 
>         grep -i rcu
> 
> to the above shows that we have a couple of places that do this
> carefully, but they are the exception rather than the rule.
> 
> I think the issue is that "current->sighand" is always safe (and that
> "me->sighand" is the same thing), and that sighand has RCU-delayed
> freeing so that __lock_task_sighand() can safely try to take the lock
> of another process' sighand.
> 
> And we have no real way to explain to sparse that *some* cases are
> fine, others are not and need the sighand lock (after that careful
> __lock_task_sighand thing).

Sounds to me like that sparse check was not such a good idea in the
first place. Especially since the 0day bot is probably warning about all
those cases where we try to lock ->sighand.

It was added by

913292c97d75 ("sched.h: Annotate sighand_struct with __rcu")

Lemme add the involved parties to Cc.

Thx.

-- 
Regards/Gruss,
    Boris.

https://people.kernel.org/tglx/notes-about-netiquette

^ permalink raw reply	[flat|nested] 69+ messages in thread

* Re: [GIT PULL] timer fixes
  2022-08-14 12:41     ` Borislav Petkov
@ 2022-08-14 17:24       ` Paul E. McKenney
  2022-08-14 17:42         ` Borislav Petkov
  0 siblings, 1 reply; 69+ messages in thread
From: Paul E. McKenney @ 2022-08-14 17:24 UTC (permalink / raw)
  To: Borislav Petkov
  Cc: Linus Torvalds, Ingo Molnar, Thomas Gleixner, linux-kernel,
	Peter Zijlstra, x86-ml, Madhuparna Bhowmik, Christian Brauner

On Sun, Aug 14, 2022 at 02:41:54PM +0200, Borislav Petkov wrote:
> On Sat, Aug 13, 2022 at 01:27:40PM -0700, Linus Torvalds wrote:
> > On Sat, Aug 13, 2022 at 9:25 AM Borislav Petkov <bp@alien8.de> wrote:
> > >
> > > That task_struct.sighand is marked __rcu and thus noderef and sparse
> > > complains:
> > 
> > I think that RCU marking is misleading.
> > 
> > Doing a
> > 
> >         git grep -e '->sighand'
> > 
> > shows that we basically never treat that as some kind of RCU pointer.
> > 
> > Adding a
> > 
> >         grep -i rcu
> > 
> > to the above shows that we have a couple of places that do this
> > carefully, but they are the exception rather than the rule.
> > 
> > I think the issue is that "current->sighand" is always safe (and that
> > "me->sighand" is the same thing), and that sighand has RCU-delayed
> > freeing so that __lock_task_sighand() can safely try to take the lock
> > of another process' sighand.
> > 
> > And we have no real way to explain to sparse that *some* cases are
> > fine, others are not and need the sighand lock (after that careful
> > __lock_task_sighand thing).
> 
> Sounds to me like that sparse check was not such a good idea in the
> first place. Especially since the 0day bot is probably warning about all
> those cases where we try to lock ->sighand.
> 
> It was added by
> 
> 913292c97d75 ("sched.h: Annotate sighand_struct with __rcu")
> 
> Lemme add the involved parties to Cc.

If it is causing more trouble than it is worth, then I have not objection
to taking a different approach.

							Thanx, Paul

^ permalink raw reply	[flat|nested] 69+ messages in thread

* Re: [GIT PULL] timer fixes
  2022-08-14 17:24       ` Paul E. McKenney
@ 2022-08-14 17:42         ` Borislav Petkov
  0 siblings, 0 replies; 69+ messages in thread
From: Borislav Petkov @ 2022-08-14 17:42 UTC (permalink / raw)
  To: Paul E. McKenney
  Cc: Linus Torvalds, Ingo Molnar, Thomas Gleixner, linux-kernel,
	Peter Zijlstra, x86-ml, Madhuparna Bhowmik, Christian Brauner

On Sun, Aug 14, 2022 at 10:24:45AM -0700, Paul E. McKenney wrote:
> If it is causing more trouble than it is worth, then I have not objection
> to taking a different approach.

Well, pretty much every file I tried in the output of

git grep -e '->sighand'

triggers that sparse warning. For example:

$ make C=1 kernel/signal.o

...

kernel/signal.c:195:31: warning: incorrect type in argument 1 (different address spaces)
kernel/signal.c:195:31:    expected struct spinlock [usertype] *lock
kernel/signal.c:195:31:    got struct spinlock [noderef] __rcu *
kernel/signal.c:198:33: warning: incorrect type in argument 1 (different address spaces)
kernel/signal.c:198:33:    expected struct spinlock [usertype] *lock
kernel/signal.c:198:33:    got struct spinlock [noderef] __rcu *
kernel/signal.c:480:9: warning: incorrect type in argument 1 (different address spaces)
kernel/signal.c:480:9:    expected struct spinlock [usertype] *lock
kernel/signal.c:480:9:    got struct spinlock [noderef] __rcu *
kernel/signal.c:484:34: warning: incorrect type in argument 1 (different address spaces)
kernel/signal.c:484:34:    expected struct spinlock [usertype] *lock
kernel/signal.c:484:34:    got struct spinlock [noderef] __rcu *
kernel/signal.c:517:9: warning: incorrect type in argument 1 (different address spaces)
kernel/signal.c:517:9:    expected struct spinlock [usertype] *lock
kernel/signal.c:517:9:    got struct spinlock [noderef] __rcu *
kernel/signal.c:520:36: warning: incorrect type in argument 1 (different address spaces)
kernel/signal.c:520:36:    expected struct spinlock [usertype] *lock
kernel/signal.c:520:36:    got struct spinlock [noderef] __rcu *
kernel/signal.c:542:53: warning: incorrect type in initializer (different address spaces)
kernel/signal.c:542:53:    expected struct k_sigaction *ka
kernel/signal.c:542:53:    got struct k_sigaction [noderef] __rcu *
./include/uapi/asm-generic/signal-defs.h:83:29: error: multiple address spaces given
kernel/signal.c:698:33: warning: incorrect type in argument 1 (different address spaces)
kernel/signal.c:698:33:    expected struct spinlock [usertype] *lock
kernel/signal.c:698:33:    got struct spinlock [noderef] __rcu *
kernel/signal.c:700:31: warning: incorrect type in argument 1 (different address spaces)
kernel/signal.c:700:31:    expected struct spinlock [usertype] *lock
kernel/signal.c:700:31:    got struct spinlock [noderef] __rcu *
kernel/signal.c:1328:9: warning: incorrect type in argument 1 (different address spaces)
kernel/signal.c:1328:9:    expected struct spinlock [usertype] *lock
kernel/signal.c:1328:9:    got struct spinlock [noderef] __rcu *
kernel/signal.c:1329:16: warning: incorrect type in assignment (different address spaces)
kernel/signal.c:1329:16:    expected struct k_sigaction *action
kernel/signal.c:1329:16:    got struct k_sigaction [noderef] __rcu *
kernel/signal.c:1349:34: warning: incorrect type in argument 1 (different address spaces)
kernel/signal.c:1349:34:    expected struct spinlock [usertype] *lock
kernel/signal.c:1349:34:    got struct spinlock [noderef] __rcu *
kernel/signal.c:1938:36: warning: incorrect type in initializer (different address spaces)
kernel/signal.c:1938:36:    expected struct spinlock [usertype] *lock
kernel/signal.c:1938:36:    got struct spinlock [noderef] __rcu *
kernel/signal.c:2048:44: warning: cast removes address space '__rcu' of expression
kernel/signal.c:2067:65: warning: too many warnings

-- 
Regards/Gruss,
    Boris.

https://people.kernel.org/tglx/notes-about-netiquette

^ permalink raw reply	[flat|nested] 69+ messages in thread

* Re: [GIT PULL] timer fixes
  2023-01-12 14:44 Ingo Molnar
@ 2023-01-12 23:01 ` pr-tracker-bot
  0 siblings, 0 replies; 69+ messages in thread
From: pr-tracker-bot @ 2023-01-12 23:01 UTC (permalink / raw)
  To: Ingo Molnar
  Cc: Linus Torvalds, linux-kernel, Thomas Gleixner, Peter Zijlstra,
	Borislav Petkov, Andrew Morton

The pull request you sent on Thu, 12 Jan 2023 15:44:09 +0100:

> git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git timers-urgent-2023-01-12

has been merged into torvalds/linux.git:
https://git.kernel.org/torvalds/c/772d0e9144df4fdf68b23f2136a512e2e5670e9a

Thank you!

-- 
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/prtracker.html

^ permalink raw reply	[flat|nested] 69+ messages in thread

* [GIT PULL] timer fixes
@ 2023-01-12 14:44 Ingo Molnar
  2023-01-12 23:01 ` pr-tracker-bot
  0 siblings, 1 reply; 69+ messages in thread
From: Ingo Molnar @ 2023-01-12 14:44 UTC (permalink / raw)
  To: Linus Torvalds
  Cc: linux-kernel, Thomas Gleixner, Peter Zijlstra, Borislav Petkov,
	Andrew Morton

Linus,

Please pull the latest timers/urgent git tree from:

   git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git timers-urgent-2023-01-12

   # HEAD: f3cb80804b8295323919e031281768ba3bf5f8da time: Fix various kernel-doc problems

- Fix various DocBook formatting errors in kernel/time/ that
  generated (justified) warnings during a kernel-doc build.

 Thanks,

	Ingo

------------------>
Randy Dunlap (1):
      time: Fix various kernel-doc problems


 kernel/time/tick-oneshot.c | 4 ++--
 kernel/time/time.c         | 8 ++++----
 kernel/time/timekeeping.c  | 8 ++++----
 3 files changed, 10 insertions(+), 10 deletions(-)

diff --git a/kernel/time/tick-oneshot.c b/kernel/time/tick-oneshot.c
index 475ecceda768..5e2c2c26b3cc 100644
--- a/kernel/time/tick-oneshot.c
+++ b/kernel/time/tick-oneshot.c
@@ -18,7 +18,7 @@
 #include "tick-internal.h"
 
 /**
- * tick_program_event
+ * tick_program_event - program the CPU local timer device for the next event
  */
 int tick_program_event(ktime_t expires, int force)
 {
@@ -99,7 +99,7 @@ int tick_switch_to_oneshot(void (*handler)(struct clock_event_device *))
 }
 
 /**
- * tick_check_oneshot_mode - check whether the system is in oneshot mode
+ * tick_oneshot_mode_active - check whether the system is in oneshot mode
  *
  * returns 1 when either nohz or highres are enabled. otherwise 0.
  */
diff --git a/kernel/time/time.c b/kernel/time/time.c
index 526257b3727c..f4198af60fee 100644
--- a/kernel/time/time.c
+++ b/kernel/time/time.c
@@ -462,7 +462,7 @@ struct __kernel_old_timeval ns_to_kernel_old_timeval(s64 nsec)
 EXPORT_SYMBOL(ns_to_kernel_old_timeval);
 
 /**
- * set_normalized_timespec - set timespec sec and nsec parts and normalize
+ * set_normalized_timespec64 - set timespec sec and nsec parts and normalize
  *
  * @ts:		pointer to timespec variable to be set
  * @sec:	seconds to set
@@ -526,7 +526,7 @@ struct timespec64 ns_to_timespec64(s64 nsec)
 EXPORT_SYMBOL(ns_to_timespec64);
 
 /**
- * msecs_to_jiffies: - convert milliseconds to jiffies
+ * __msecs_to_jiffies: - convert milliseconds to jiffies
  * @m:	time in milliseconds
  *
  * conversion is done as follows:
@@ -541,12 +541,12 @@ EXPORT_SYMBOL(ns_to_timespec64);
  *   handling any 32-bit overflows.
  *   for the details see __msecs_to_jiffies()
  *
- * msecs_to_jiffies() checks for the passed in value being a constant
+ * __msecs_to_jiffies() checks for the passed in value being a constant
  * via __builtin_constant_p() allowing gcc to eliminate most of the
  * code, __msecs_to_jiffies() is called if the value passed does not
  * allow constant folding and the actual conversion must be done at
  * runtime.
- * the _msecs_to_jiffies helpers are the HZ dependent conversion
+ * The _msecs_to_jiffies helpers are the HZ dependent conversion
  * routines found in include/linux/jiffies.h
  */
 unsigned long __msecs_to_jiffies(const unsigned int m)
diff --git a/kernel/time/timekeeping.c b/kernel/time/timekeeping.c
index f72b9f1de178..5579ead449f2 100644
--- a/kernel/time/timekeeping.c
+++ b/kernel/time/timekeeping.c
@@ -1590,10 +1590,10 @@ void __weak read_persistent_clock64(struct timespec64 *ts)
 /**
  * read_persistent_wall_and_boot_offset - Read persistent clock, and also offset
  *                                        from the boot.
+ * @wall_time:	  current time as returned by persistent clock
+ * @boot_offset:  offset that is defined as wall_time - boot_time
  *
  * Weak dummy function for arches that do not yet support it.
- * @wall_time:	- current time as returned by persistent clock
- * @boot_offset: - offset that is defined as wall_time - boot_time
  *
  * The default function calculates offset based on the current value of
  * local_clock(). This way architectures that support sched_clock() but don't
@@ -1701,7 +1701,7 @@ static void __timekeeping_inject_sleeptime(struct timekeeper *tk,
 }
 
 #if defined(CONFIG_PM_SLEEP) && defined(CONFIG_RTC_HCTOSYS_DEVICE)
-/**
+/*
  * We have three kinds of time sources to use for sleep time
  * injection, the preference order is:
  * 1) non-stop clocksource
@@ -1722,7 +1722,7 @@ bool timekeeping_rtc_skipresume(void)
 	return !suspend_timing_needed;
 }
 
-/**
+/*
  * 1) can be determined whether to use or not only when doing
  * timekeeping_resume() which is invoked after rtc_suspend(),
  * so we can't skip rtc_suspend() surely if system has 1).

^ permalink raw reply related	[flat|nested] 69+ messages in thread

* Re: [GIT PULL] timer fixes
  2020-12-27  9:26 Ingo Molnar
@ 2020-12-27 17:27 ` pr-tracker-bot
  0 siblings, 0 replies; 69+ messages in thread
From: pr-tracker-bot @ 2020-12-27 17:27 UTC (permalink / raw)
  To: Ingo Molnar
  Cc: Linus Torvalds, linux-kernel, Thomas Gleixner, Peter Zijlstra,
	Andrew Morton

The pull request you sent on Sun, 27 Dec 2020 10:26:58 +0100:

> git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git timers-urgent-2020-12-27

has been merged into torvalds/linux.git:
https://git.kernel.org/torvalds/c/2eeefc60ad70ffb7a5daf9f47aef5b1ebd1f39ad

Thank you!

-- 
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/prtracker.html

^ permalink raw reply	[flat|nested] 69+ messages in thread

* [GIT PULL] timer fixes
@ 2020-12-27  9:26 Ingo Molnar
  2020-12-27 17:27 ` pr-tracker-bot
  0 siblings, 1 reply; 69+ messages in thread
From: Ingo Molnar @ 2020-12-27  9:26 UTC (permalink / raw)
  To: Linus Torvalds
  Cc: linux-kernel, Thomas Gleixner, Peter Zijlstra, Andrew Morton

Linus,

Please pull the latest timers/urgent git tree from:

   git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git timers-urgent-2020-12-27

   # HEAD: f6f5cd840ae782680c5e94048c72420e4e6857f9 timekeeping: Fix spelling mistake in Kconfig "fullfill" -> "fulfill"

Update/fix two CPU sanity checks in the hotplug and the boot code,
and fix a typo in the Kconfig help text.

( Context: the first two commits are the result of an ongoing 
  annotation+review work of (intentional) tick_do_timer_cpu() data 
  races reported by KCSAN, but the annotations aren't fully cooked 
  yet. )

 Thanks,

	Ingo

------------------>
Colin Ian King (1):
      timekeeping: Fix spelling mistake in Kconfig "fullfill" -> "fulfill"

Thomas Gleixner (2):
      tick: Remove pointless cpu valid check in hotplug code
      tick/sched: Remove bogus boot "safety" check


 kernel/time/Kconfig       |  2 +-
 kernel/time/tick-common.c | 10 +++-------
 kernel/time/tick-sched.c  |  7 -------
 3 files changed, 4 insertions(+), 15 deletions(-)

diff --git a/kernel/time/Kconfig b/kernel/time/Kconfig
index a09b1d61df6a..64051f47475c 100644
--- a/kernel/time/Kconfig
+++ b/kernel/time/Kconfig
@@ -141,7 +141,7 @@ config CONTEXT_TRACKING_FORCE
 	  dynticks working.
 
 	  This option stands for testing when an arch implements the
-	  context tracking backend but doesn't yet fullfill all the
+	  context tracking backend but doesn't yet fulfill all the
 	  requirements to make the full dynticks feature working.
 	  Without the full dynticks, there is no way to test the support
 	  for context tracking and the subsystems that rely on it: RCU
diff --git a/kernel/time/tick-common.c b/kernel/time/tick-common.c
index a03764df5366..9d3a22510bab 100644
--- a/kernel/time/tick-common.c
+++ b/kernel/time/tick-common.c
@@ -407,17 +407,13 @@ EXPORT_SYMBOL_GPL(tick_broadcast_oneshot_control);
 /*
  * Transfer the do_timer job away from a dying cpu.
  *
- * Called with interrupts disabled. Not locking required. If
+ * Called with interrupts disabled. No locking required. If
  * tick_do_timer_cpu is owned by this cpu, nothing can change it.
  */
 void tick_handover_do_timer(void)
 {
-	if (tick_do_timer_cpu == smp_processor_id()) {
-		int cpu = cpumask_first(cpu_online_mask);
-
-		tick_do_timer_cpu = (cpu < nr_cpu_ids) ? cpu :
-			TICK_DO_TIMER_NONE;
-	}
+	if (tick_do_timer_cpu == smp_processor_id())
+		tick_do_timer_cpu = cpumask_first(cpu_online_mask);
 }
 
 /*
diff --git a/kernel/time/tick-sched.c b/kernel/time/tick-sched.c
index a9e68936822d..5fbc748f0058 100644
--- a/kernel/time/tick-sched.c
+++ b/kernel/time/tick-sched.c
@@ -991,13 +991,6 @@ static bool can_stop_idle_tick(int cpu, struct tick_sched *ts)
 		 */
 		if (tick_do_timer_cpu == cpu)
 			return false;
-		/*
-		 * Boot safety: make sure the timekeeping duty has been
-		 * assigned before entering dyntick-idle mode,
-		 * tick_do_timer_cpu is TICK_DO_TIMER_BOOT
-		 */
-		if (unlikely(tick_do_timer_cpu == TICK_DO_TIMER_BOOT))
-			return false;
 
 		/* Should not happen for nohz-full */
 		if (WARN_ON_ONCE(tick_do_timer_cpu == TICK_DO_TIMER_NONE))

^ permalink raw reply related	[flat|nested] 69+ messages in thread

* Re: [GIT PULL] timer fixes
  2019-12-17 20:43         ` Linus Torvalds
@ 2019-12-17 20:54           ` Peter Zijlstra
  0 siblings, 0 replies; 69+ messages in thread
From: Peter Zijlstra @ 2019-12-17 20:54 UTC (permalink / raw)
  To: Linus Torvalds
  Cc: Ingo Molnar, Linux Kernel Mailing List, Thomas Gleixner,
	Andrew Morton, Borislav Petkov

On Tue, Dec 17, 2019 at 12:43:46PM -0800, Linus Torvalds wrote:
> On Tue, Dec 17, 2019 at 12:35 PM Peter Zijlstra <peterz@infradead.org> wrote:
> >
> > Pray.. the TSC MSR is still writable from SMM, so BIOS monkeys could
> > still do what they've been doing for decades.
> 
> Sure. And the HPET is unreliable, so the checking causes issues.
> 
> Which one should we worry about?

I'm just waiting for the system that messes up the TSC and doesn't have
HPET/PIT _at_all_ :-/

But yes, I get your point.

> > Also, what consititutes a 'modern' CPU?
> 
> I think anything that has TSC_STABLE set should likely be considered
> more reliable than HPET.
> 
> Or whatever the bit is called. The "doesn't stop in idle" thing.

The doesn't stop on idle thing is CONSTANT_TSC, and that has been set
since Nehalem, and there are a metric ton of systems still failing due
to the described SMM 'feature'.

Perhaps we can go with TSC_KNOWN_FREQ, which relies on CPUID.15h to tell
us the actual TSC frequency, but I'm not sure all modern systems
actually fill out that leaf :/

The systems that completely got rid of the HPET/PIT pretty much must
fill that out, otherwise there's just no way we can recover the TSC
frequency.

I'm going to have to wait for Thomas to chime in though; I only
occasionally poke at this stuff ;-)

^ permalink raw reply	[flat|nested] 69+ messages in thread

* Re: [GIT PULL] timer fixes
  2019-12-17 20:35       ` Peter Zijlstra
@ 2019-12-17 20:43         ` Linus Torvalds
  2019-12-17 20:54           ` Peter Zijlstra
  0 siblings, 1 reply; 69+ messages in thread
From: Linus Torvalds @ 2019-12-17 20:43 UTC (permalink / raw)
  To: Peter Zijlstra
  Cc: Ingo Molnar, Linux Kernel Mailing List, Thomas Gleixner,
	Andrew Morton, Borislav Petkov

On Tue, Dec 17, 2019 at 12:35 PM Peter Zijlstra <peterz@infradead.org> wrote:
>
> Pray.. the TSC MSR is still writable from SMM, so BIOS monkeys could
> still do what they've been doing for decades.

Sure. And the HPET is unreliable, so the checking causes issues.

Which one should we worry about?

> Also, what consititutes a 'modern' CPU?

I think anything that has TSC_STABLE set should likely be considered
more reliable than HPET.

Or whatever the bit is called. The "doesn't stop in idle" thing.

> > The HPET seems to get disabled on all the modern platforms, why do we
> > even have it enabled by default?
>
> These new ones yeah, cuz they wrecked HPET in PC10 :/

That's my point. HPET isn't _used_ by Windows, so it gets no testing,
so it will continue to have bugs.

At least the TSC is used.

I do agree about the whole SMM issue, but it seems less of a pain than
the HPET issue.

             Linus

^ permalink raw reply	[flat|nested] 69+ messages in thread

* Re: [GIT PULL] timer fixes
  2019-12-17 20:16     ` Linus Torvalds
@ 2019-12-17 20:35       ` Peter Zijlstra
  2019-12-17 20:43         ` Linus Torvalds
  0 siblings, 1 reply; 69+ messages in thread
From: Peter Zijlstra @ 2019-12-17 20:35 UTC (permalink / raw)
  To: Linus Torvalds
  Cc: Ingo Molnar, Linux Kernel Mailing List, Thomas Gleixner,
	Andrew Morton, Borislav Petkov

On Tue, Dec 17, 2019 at 12:16:52PM -0800, Linus Torvalds wrote:
> On Tue, Dec 17, 2019 at 11:30 AM Peter Zijlstra <peterz@infradead.org> wrote:
> >
> > What alternatives are there? That is, we normally only use HPET to
> > double check nobody messed up the TSC.
> 
> The thing is HPET seems to be _less_ reliable than the TSC we're
> checking these days.
> 
> If that's the only use-case for HPET, we should just stop doing it.
> 
> > We can't just blindly trust TSC across everything x86.
> 
> No, but we can trust it when it's a modern CPU.

Pray.. the TSC MSR is still writable from SMM, so BIOS monkeys could
still do what they've been doing for decades. Which is try and 'hide'
SMM latency by taking the TSC timestamp on SMM entry and writing the
timestamp back into the TSC MSR on exit.

Yes, ever since TSC_ADJUST we can better recover from it, but
we still need to first detect it went sideways and by then time has been
observed buggered and any recovery is basically too late :/

Granted, this is happening less (at least, I really do hope so).

Also, what consititutes a 'modern' CPU?

> The HPET seems to get disabled on all the modern platforms, why do we
> even have it enabled by default?

These new ones yeah, cuz they wrecked HPET in PC10 :/

> We should do the HPET cross-check only when we know the TSC might be
> unreliable, I suspect.

But how do we know? Ever since Nehalem TSC has basically been good
hardware wise -- there's a few exception on large (>4) socket machines,
but nobody has those anyway.

It has always been the BIOS messing it up. Now, the reason I think it
has gotten better is because Windows is now also relying on TSC (like
we've been doing forever).

Maybe I'm too scarred by too much TSC wreckage over the years...

^ permalink raw reply	[flat|nested] 69+ messages in thread

* Re: [GIT PULL] timer fixes
  2019-12-17 19:30   ` Peter Zijlstra
@ 2019-12-17 20:16     ` Linus Torvalds
  2019-12-17 20:35       ` Peter Zijlstra
  0 siblings, 1 reply; 69+ messages in thread
From: Linus Torvalds @ 2019-12-17 20:16 UTC (permalink / raw)
  To: Peter Zijlstra
  Cc: Ingo Molnar, Linux Kernel Mailing List, Thomas Gleixner,
	Andrew Morton, Borislav Petkov

On Tue, Dec 17, 2019 at 11:30 AM Peter Zijlstra <peterz@infradead.org> wrote:
>
> What alternatives are there? That is, we normally only use HPET to
> double check nobody messed up the TSC.

The thing is HPET seems to be _less_ reliable than the TSC we're
checking these days.

If that's the only use-case for HPET, we should just stop doing it.

> We can't just blindly trust TSC across everything x86.

No, but we can trust it when it's a modern CPU.

The HPET seems to get disabled on all the modern platforms, why do we
even have it enabled by default?

We should do the HPET cross-check only when we know the TSC might be
unreliable, I suspect.

           Linus

^ permalink raw reply	[flat|nested] 69+ messages in thread

* Re: [GIT PULL] timer fixes
  2019-12-17 19:14 ` Linus Torvalds
@ 2019-12-17 19:30   ` Peter Zijlstra
  2019-12-17 20:16     ` Linus Torvalds
  0 siblings, 1 reply; 69+ messages in thread
From: Peter Zijlstra @ 2019-12-17 19:30 UTC (permalink / raw)
  To: Linus Torvalds
  Cc: Ingo Molnar, Linux Kernel Mailing List, Thomas Gleixner,
	Andrew Morton, Borislav Petkov

On Tue, Dec 17, 2019 at 11:14:48AM -0800, Linus Torvalds wrote:
> On Tue, Dec 17, 2019 at 3:55 AM Ingo Molnar <mingo@kernel.org> wrote:
> >
> > Add HPET quirks for the Intel "Coffee Lake H" and "Ice Lake" platforms.
> 
> Is there any reason we don't force-disable HPET by default if there
> are alternatives?

What alternatives are there? That is, we normally only use HPET to
double check nobody messed up the TSC.

Now it (fingers crossed) seems to be getting better in that more recent
firmware is less likely to wreck TSC, but we're still running this on
plenty of old machines where TSC really isn't all that sane.

We can't just blindly trust TSC across everything x86.

^ permalink raw reply	[flat|nested] 69+ messages in thread

* Re: [GIT PULL] timer fixes
  2019-12-17 11:55 Ingo Molnar
  2019-12-17 19:14 ` Linus Torvalds
@ 2019-12-17 19:20 ` pr-tracker-bot
  1 sibling, 0 replies; 69+ messages in thread
From: pr-tracker-bot @ 2019-12-17 19:20 UTC (permalink / raw)
  To: Ingo Molnar
  Cc: Linus Torvalds, linux-kernel, Thomas Gleixner, Peter Zijlstra,
	Andrew Morton, Borislav Petkov

The pull request you sent on Tue, 17 Dec 2019 12:55:47 +0100:

> git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git timers-urgent-for-linus

has been merged into torvalds/linux.git:
https://git.kernel.org/torvalds/c/2abf193275768ac89f5d93eec7bcacb238168151

Thank you!

-- 
Deet-doot-dot, I am a bot.
https://korg.wiki.kernel.org/userdoc/prtracker

^ permalink raw reply	[flat|nested] 69+ messages in thread

* Re: [GIT PULL] timer fixes
  2019-12-17 11:55 Ingo Molnar
@ 2019-12-17 19:14 ` Linus Torvalds
  2019-12-17 19:30   ` Peter Zijlstra
  2019-12-17 19:20 ` pr-tracker-bot
  1 sibling, 1 reply; 69+ messages in thread
From: Linus Torvalds @ 2019-12-17 19:14 UTC (permalink / raw)
  To: Ingo Molnar
  Cc: Linux Kernel Mailing List, Thomas Gleixner, Peter Zijlstra,
	Andrew Morton, Borislav Petkov

On Tue, Dec 17, 2019 at 3:55 AM Ingo Molnar <mingo@kernel.org> wrote:
>
> Add HPET quirks for the Intel "Coffee Lake H" and "Ice Lake" platforms.

Is there any reason we don't force-disable HPET by default if there
are alternatives?

I think that's what Microsoft did, which is why it's so broken on so
many platforms - it's often simply not used.

                   Linus

^ permalink raw reply	[flat|nested] 69+ messages in thread

* [GIT PULL] timer fixes
@ 2019-12-17 11:55 Ingo Molnar
  2019-12-17 19:14 ` Linus Torvalds
  2019-12-17 19:20 ` pr-tracker-bot
  0 siblings, 2 replies; 69+ messages in thread
From: Ingo Molnar @ 2019-12-17 11:55 UTC (permalink / raw)
  To: Linus Torvalds
  Cc: linux-kernel, Thomas Gleixner, Peter Zijlstra, Andrew Morton,
	Borislav Petkov

Linus,

Please pull the latest timers-urgent-for-linus git tree from:

   git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git timers-urgent-for-linus

   # HEAD: e0748539e3d594dd26f0d27a270f14720b22a406 x86/intel: Disable HPET on Intel Ice Lake platforms

Add HPET quirks for the Intel "Coffee Lake H" and "Ice Lake" platforms.

 Thanks,

	Ingo

------------------>
Kai-Heng Feng (2):
      x86/intel: Disable HPET on Intel Coffee Lake H platforms
      x86/intel: Disable HPET on Intel Ice Lake platforms


 arch/x86/kernel/early-quirks.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/arch/x86/kernel/early-quirks.c b/arch/x86/kernel/early-quirks.c
index 4cba91ec8049..2f9ec14be3b1 100644
--- a/arch/x86/kernel/early-quirks.c
+++ b/arch/x86/kernel/early-quirks.c
@@ -710,8 +710,12 @@ static struct chipset early_qrk[] __initdata = {
 	 */
 	{ PCI_VENDOR_ID_INTEL, 0x0f00,
 		PCI_CLASS_BRIDGE_HOST, PCI_ANY_ID, 0, force_disable_hpet},
+	{ PCI_VENDOR_ID_INTEL, 0x3e20,
+		PCI_CLASS_BRIDGE_HOST, PCI_ANY_ID, 0, force_disable_hpet},
 	{ PCI_VENDOR_ID_INTEL, 0x3ec4,
 		PCI_CLASS_BRIDGE_HOST, PCI_ANY_ID, 0, force_disable_hpet},
+	{ PCI_VENDOR_ID_INTEL, 0x8a12,
+		PCI_CLASS_BRIDGE_HOST, PCI_ANY_ID, 0, force_disable_hpet},
 	{ PCI_VENDOR_ID_BROADCOM, 0x4331,
 	  PCI_CLASS_NETWORK_OTHER, PCI_ANY_ID, 0, apple_airport_reset},
 	{}

^ permalink raw reply related	[flat|nested] 69+ messages in thread

* Re: [GIT PULL] timer fixes
  2019-05-16 16:09 Ingo Molnar
@ 2019-05-16 18:20 ` pr-tracker-bot
  0 siblings, 0 replies; 69+ messages in thread
From: pr-tracker-bot @ 2019-05-16 18:20 UTC (permalink / raw)
  To: Ingo Molnar
  Cc: Linus Torvalds, linux-kernel, Thomas Gleixner, Peter Zijlstra,
	Andrew Morton

The pull request you sent on Thu, 16 May 2019 18:09:25 +0200:

> git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git timers-urgent-for-linus

has been merged into torvalds/linux.git:
https://git.kernel.org/torvalds/c/b2c3dda6f8f06d825b9b6099f57b906c774141c0

Thank you!

-- 
Deet-doot-dot, I am a bot.
https://korg.wiki.kernel.org/userdoc/prtracker

^ permalink raw reply	[flat|nested] 69+ messages in thread

* [GIT PULL] timer fixes
@ 2019-05-16 16:09 Ingo Molnar
  2019-05-16 18:20 ` pr-tracker-bot
  0 siblings, 1 reply; 69+ messages in thread
From: Ingo Molnar @ 2019-05-16 16:09 UTC (permalink / raw)
  To: Linus Torvalds
  Cc: linux-kernel, Thomas Gleixner, Peter Zijlstra, Andrew Morton

Linus,

Please pull the latest timers-urgent-for-linus git tree from:

   git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git timers-urgent-for-linus

   # HEAD: fdc6bae940ee9eb869e493990540098b8c0fd6ab ntp: Allow TAI-UTC offset to be set to zero

A TIA adjtimex interface extension, and a POSIX compliance ABI fix for 
timespec64 users.

 Thanks,

	Ingo

------------------>
Arnd Bergmann (1):
      y2038: Make CONFIG_64BIT_TIME unconditional

Miroslav Lichvar (1):
      ntp: Allow TAI-UTC offset to be set to zero


 arch/Kconfig      | 2 +-
 kernel/time/ntp.c | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/arch/Kconfig b/arch/Kconfig
index 33687dddd86a..9092e0ffe4d3 100644
--- a/arch/Kconfig
+++ b/arch/Kconfig
@@ -764,7 +764,7 @@ config COMPAT_OLD_SIGACTION
 	bool
 
 config 64BIT_TIME
-	def_bool ARCH_HAS_64BIT_TIME
+	def_bool y
 	help
 	  This should be selected by all architectures that need to support
 	  new system calls with a 64-bit time_t. This is relevant on all 32-bit
diff --git a/kernel/time/ntp.c b/kernel/time/ntp.c
index 92a90014a925..f43d47c8c3b6 100644
--- a/kernel/time/ntp.c
+++ b/kernel/time/ntp.c
@@ -690,7 +690,7 @@ static inline void process_adjtimex_modes(const struct __kernel_timex *txc,
 		time_constant = max(time_constant, 0l);
 	}
 
-	if (txc->modes & ADJ_TAI && txc->constant > 0)
+	if (txc->modes & ADJ_TAI && txc->constant >= 0)
 		*time_tai = txc->constant;
 
 	if (txc->modes & ADJ_OFFSET)

^ permalink raw reply related	[flat|nested] 69+ messages in thread

* Re: [GIT PULL] timer fixes
  2019-04-20  7:45 Ingo Molnar
@ 2019-04-20 19:25 ` pr-tracker-bot
  0 siblings, 0 replies; 69+ messages in thread
From: pr-tracker-bot @ 2019-04-20 19:25 UTC (permalink / raw)
  To: Ingo Molnar
  Cc: Linus Torvalds, linux-kernel, Thomas Gleixner, Peter Zijlstra,
	Andrew Morton

The pull request you sent on Sat, 20 Apr 2019 09:45:50 +0200:

> git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git timers-urgent-for-linus

has been merged into torvalds/linux.git:
https://git.kernel.org/torvalds/c/e899cc3b3d9813f795c039c0f3c689861d6836da

Thank you!

-- 
Deet-doot-dot, I am a bot.
https://korg.wiki.kernel.org/userdoc/prtracker

^ permalink raw reply	[flat|nested] 69+ messages in thread

* [GIT PULL] timer fixes
@ 2019-04-20  7:45 Ingo Molnar
  2019-04-20 19:25 ` pr-tracker-bot
  0 siblings, 1 reply; 69+ messages in thread
From: Ingo Molnar @ 2019-04-20  7:45 UTC (permalink / raw)
  To: Linus Torvalds
  Cc: linux-kernel, Thomas Gleixner, Peter Zijlstra, Andrew Morton

Linus,

Please pull the latest timers-urgent-for-linus git tree from:

   git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git timers-urgent-for-linus

   # HEAD: 3f2552f7e9c5abef2775c53f7af66532f8bf65bc timers/sched_clock: Prevent generic sched_clock wrap caused by tick_freeze()

Misc clocksource driver fixes, and a sched-clock wrapping fix.

 Thanks,

	Ingo

------------------>
Arnd Bergmann (1):
      clocksource/drivers/npcm: select TIMER_OF

Chang-An Chen (1):
      timers/sched_clock: Prevent generic sched_clock wrap caused by tick_freeze()

Nathan Chancellor (1):
      clocksource/drivers/timer-ti-dm: Remove omap_dm_timer_set_load_start

Neil Armstrong (1):
      clocksource/drivers/oxnas: Fix OX820 compatible

Yangtao Li (1):
      clocksource/drivers/arm_arch_timer: Remove unneeded pr_fmt macro


 drivers/clocksource/Kconfig           |  1 +
 drivers/clocksource/arm_arch_timer.c  |  5 +----
 drivers/clocksource/timer-oxnas-rps.c |  2 +-
 drivers/clocksource/timer-ti-dm.c     | 28 ----------------------------
 kernel/time/sched_clock.c             |  4 ++--
 kernel/time/tick-common.c             |  2 ++
 kernel/time/timekeeping.h             |  7 +++++++
 7 files changed, 14 insertions(+), 35 deletions(-)

diff --git a/drivers/clocksource/Kconfig b/drivers/clocksource/Kconfig
index 171502a356aa..4b3d143f0f8a 100644
--- a/drivers/clocksource/Kconfig
+++ b/drivers/clocksource/Kconfig
@@ -145,6 +145,7 @@ config VT8500_TIMER
 config NPCM7XX_TIMER
 	bool "NPCM7xx timer driver" if COMPILE_TEST
 	depends on HAS_IOMEM
+	select TIMER_OF
 	select CLKSRC_MMIO
 	help
 	  Enable 24-bit TIMER0 and TIMER1 counters in the NPCM7xx architecture,
diff --git a/drivers/clocksource/arm_arch_timer.c b/drivers/clocksource/arm_arch_timer.c
index aa4ec53281ce..ea373cfbcecb 100644
--- a/drivers/clocksource/arm_arch_timer.c
+++ b/drivers/clocksource/arm_arch_timer.c
@@ -9,7 +9,7 @@
  * published by the Free Software Foundation.
  */
 
-#define pr_fmt(fmt)	"arm_arch_timer: " fmt
+#define pr_fmt(fmt) 	"arch_timer: " fmt
 
 #include <linux/init.h>
 #include <linux/kernel.h>
@@ -33,9 +33,6 @@
 
 #include <clocksource/arm_arch_timer.h>
 
-#undef pr_fmt
-#define pr_fmt(fmt) "arch_timer: " fmt
-
 #define CNTTIDR		0x08
 #define CNTTIDR_VIRT(n)	(BIT(1) << ((n) * 4))
 
diff --git a/drivers/clocksource/timer-oxnas-rps.c b/drivers/clocksource/timer-oxnas-rps.c
index eed6feff8b5f..30c6f4ce672b 100644
--- a/drivers/clocksource/timer-oxnas-rps.c
+++ b/drivers/clocksource/timer-oxnas-rps.c
@@ -296,4 +296,4 @@ static int __init oxnas_rps_timer_init(struct device_node *np)
 TIMER_OF_DECLARE(ox810se_rps,
 		       "oxsemi,ox810se-rps-timer", oxnas_rps_timer_init);
 TIMER_OF_DECLARE(ox820_rps,
-		       "oxsemi,ox820se-rps-timer", oxnas_rps_timer_init);
+		       "oxsemi,ox820-rps-timer", oxnas_rps_timer_init);
diff --git a/drivers/clocksource/timer-ti-dm.c b/drivers/clocksource/timer-ti-dm.c
index 3352da6ed61f..ee8ec5a8cb16 100644
--- a/drivers/clocksource/timer-ti-dm.c
+++ b/drivers/clocksource/timer-ti-dm.c
@@ -585,34 +585,6 @@ static int omap_dm_timer_set_load(struct omap_dm_timer *timer, int autoreload,
 	return 0;
 }
 
-/* Optimized set_load which removes costly spin wait in timer_start */
-static int omap_dm_timer_set_load_start(struct omap_dm_timer *timer,
-					int autoreload, unsigned int load)
-{
-	u32 l;
-
-	if (unlikely(!timer))
-		return -EINVAL;
-
-	omap_dm_timer_enable(timer);
-
-	l = omap_dm_timer_read_reg(timer, OMAP_TIMER_CTRL_REG);
-	if (autoreload) {
-		l |= OMAP_TIMER_CTRL_AR;
-		omap_dm_timer_write_reg(timer, OMAP_TIMER_LOAD_REG, load);
-	} else {
-		l &= ~OMAP_TIMER_CTRL_AR;
-	}
-	l |= OMAP_TIMER_CTRL_ST;
-
-	__omap_dm_timer_load_start(timer, l, load, timer->posted);
-
-	/* Save the context */
-	timer->context.tclr = l;
-	timer->context.tldr = load;
-	timer->context.tcrr = load;
-	return 0;
-}
 static int omap_dm_timer_set_match(struct omap_dm_timer *timer, int enable,
 				   unsigned int match)
 {
diff --git a/kernel/time/sched_clock.c b/kernel/time/sched_clock.c
index 094b82ca95e5..930113b9799a 100644
--- a/kernel/time/sched_clock.c
+++ b/kernel/time/sched_clock.c
@@ -272,7 +272,7 @@ static u64 notrace suspended_sched_clock_read(void)
 	return cd.read_data[seq & 1].epoch_cyc;
 }
 
-static int sched_clock_suspend(void)
+int sched_clock_suspend(void)
 {
 	struct clock_read_data *rd = &cd.read_data[0];
 
@@ -283,7 +283,7 @@ static int sched_clock_suspend(void)
 	return 0;
 }
 
-static void sched_clock_resume(void)
+void sched_clock_resume(void)
 {
 	struct clock_read_data *rd = &cd.read_data[0];
 
diff --git a/kernel/time/tick-common.c b/kernel/time/tick-common.c
index 529143b4c8d2..df401463a191 100644
--- a/kernel/time/tick-common.c
+++ b/kernel/time/tick-common.c
@@ -487,6 +487,7 @@ void tick_freeze(void)
 		trace_suspend_resume(TPS("timekeeping_freeze"),
 				     smp_processor_id(), true);
 		system_state = SYSTEM_SUSPEND;
+		sched_clock_suspend();
 		timekeeping_suspend();
 	} else {
 		tick_suspend_local();
@@ -510,6 +511,7 @@ void tick_unfreeze(void)
 
 	if (tick_freeze_depth == num_online_cpus()) {
 		timekeeping_resume();
+		sched_clock_resume();
 		system_state = SYSTEM_RUNNING;
 		trace_suspend_resume(TPS("timekeeping_freeze"),
 				     smp_processor_id(), false);
diff --git a/kernel/time/timekeeping.h b/kernel/time/timekeeping.h
index 7a9b4eb7a1d5..141ab3ab0354 100644
--- a/kernel/time/timekeeping.h
+++ b/kernel/time/timekeeping.h
@@ -14,6 +14,13 @@ extern u64 timekeeping_max_deferment(void);
 extern void timekeeping_warp_clock(void);
 extern int timekeeping_suspend(void);
 extern void timekeeping_resume(void);
+#ifdef CONFIG_GENERIC_SCHED_CLOCK
+extern int sched_clock_suspend(void);
+extern void sched_clock_resume(void);
+#else
+static inline int sched_clock_suspend(void) { return 0; }
+static inline void sched_clock_resume(void) { }
+#endif
 
 extern void do_timer(unsigned long ticks);
 extern void update_wall_time(void);

^ permalink raw reply related	[flat|nested] 69+ messages in thread

* [GIT PULL] timer fixes
@ 2018-07-13 20:01 Ingo Molnar
  0 siblings, 0 replies; 69+ messages in thread
From: Ingo Molnar @ 2018-07-13 20:01 UTC (permalink / raw)
  To: Linus Torvalds
  Cc: linux-kernel, Thomas Gleixner, Peter Zijlstra, Andrew Morton

Linus,

Please pull the latest timers-urgent-for-linus git tree from:

   git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git timers-urgent-for-linus

   # HEAD: 5e18e412973d6bb1804de1d4d30a891c774b006e clocksource: arm_arch_timer: Set arch_mem_timer cpumask to cpu_possible_mask

A clocksource driver fix and a revert.

 Thanks,

	Ingo

------------------>
Sudeep Holla (2):
      Revert "tick: Prefer a lower rating device only if it's CPU local device"
      clocksource: arm_arch_timer: Set arch_mem_timer cpumask to cpu_possible_mask


 drivers/clocksource/arm_arch_timer.c | 2 +-
 kernel/time/tick-common.c            | 3 +--
 2 files changed, 2 insertions(+), 3 deletions(-)

diff --git a/drivers/clocksource/arm_arch_timer.c b/drivers/clocksource/arm_arch_timer.c
index 57cb2f00fc07..d8c7f5750cdb 100644
--- a/drivers/clocksource/arm_arch_timer.c
+++ b/drivers/clocksource/arm_arch_timer.c
@@ -735,7 +735,7 @@ static void __arch_timer_setup(unsigned type,
 		clk->features |= CLOCK_EVT_FEAT_DYNIRQ;
 		clk->name = "arch_mem_timer";
 		clk->rating = 400;
-		clk->cpumask = cpu_all_mask;
+		clk->cpumask = cpu_possible_mask;
 		if (arch_timer_mem_use_virtual) {
 			clk->set_state_shutdown = arch_timer_shutdown_virt_mem;
 			clk->set_state_oneshot_stopped = arch_timer_shutdown_virt_mem;
diff --git a/kernel/time/tick-common.c b/kernel/time/tick-common.c
index b7005dd21ec1..14de3727b18e 100644
--- a/kernel/time/tick-common.c
+++ b/kernel/time/tick-common.c
@@ -277,8 +277,7 @@ static bool tick_check_preferred(struct clock_event_device *curdev,
 	 */
 	return !curdev ||
 		newdev->rating > curdev->rating ||
-	       (!cpumask_equal(curdev->cpumask, newdev->cpumask) &&
-	        !tick_check_percpu(curdev, newdev, smp_processor_id()));
+	       !cpumask_equal(curdev->cpumask, newdev->cpumask);
 }
 
 /*

^ permalink raw reply related	[flat|nested] 69+ messages in thread

* [GIT PULL] timer fixes
@ 2017-03-07 20:36 Ingo Molnar
  0 siblings, 0 replies; 69+ messages in thread
From: Ingo Molnar @ 2017-03-07 20:36 UTC (permalink / raw)
  To: Linus Torvalds
  Cc: linux-kernel, Thomas Gleixner, Peter Zijlstra, Andrew Morton

Linus,

Please pull the latest timers-urgent-for-linus git tree from:

   git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git timers-urgent-for-linus

   # HEAD: fa3aa7a54fe6d3abf128f13cd4bbd40eaa48fed2 jiffies: Revert bogus conversion of NSEC_PER_SEC to TICK_NSEC

This includes a fix for lockups caused by incorrect nsecs related cleanup, and a 
capabilities check fix for timerfd.

 Thanks,

	Ingo

------------------>
Frederic Weisbecker (1):
      jiffies: Revert bogus conversion of NSEC_PER_SEC to TICK_NSEC

Stephen Smalley (1):
      timerfd: Only check CAP_WAKE_ALARM when it is needed


 fs/timerfd.c          | 8 ++++----
 kernel/time/jiffies.c | 2 +-
 2 files changed, 5 insertions(+), 5 deletions(-)

diff --git a/fs/timerfd.c b/fs/timerfd.c
index 384fa759a563..c543cdb5f8ed 100644
--- a/fs/timerfd.c
+++ b/fs/timerfd.c
@@ -400,9 +400,9 @@ SYSCALL_DEFINE2(timerfd_create, int, clockid, int, flags)
 	     clockid != CLOCK_BOOTTIME_ALARM))
 		return -EINVAL;
 
-	if (!capable(CAP_WAKE_ALARM) &&
-	    (clockid == CLOCK_REALTIME_ALARM ||
-	     clockid == CLOCK_BOOTTIME_ALARM))
+	if ((clockid == CLOCK_REALTIME_ALARM ||
+	     clockid == CLOCK_BOOTTIME_ALARM) &&
+	    !capable(CAP_WAKE_ALARM))
 		return -EPERM;
 
 	ctx = kzalloc(sizeof(*ctx), GFP_KERNEL);
@@ -449,7 +449,7 @@ static int do_timerfd_settime(int ufd, int flags,
 		return ret;
 	ctx = f.file->private_data;
 
-	if (!capable(CAP_WAKE_ALARM) && isalarm(ctx)) {
+	if (isalarm(ctx) && !capable(CAP_WAKE_ALARM)) {
 		fdput(f);
 		return -EPERM;
 	}
diff --git a/kernel/time/jiffies.c b/kernel/time/jiffies.c
index 7906b3f0c41a..497719127bf9 100644
--- a/kernel/time/jiffies.c
+++ b/kernel/time/jiffies.c
@@ -125,7 +125,7 @@ int register_refined_jiffies(long cycles_per_second)
 	shift_hz += cycles_per_tick/2;
 	do_div(shift_hz, cycles_per_tick);
 	/* Calculate nsec_per_tick using shift_hz */
-	nsec_per_tick = (u64)TICK_NSEC << 8;
+	nsec_per_tick = (u64)NSEC_PER_SEC << 8;
 	nsec_per_tick += (u32)shift_hz/2;
 	do_div(nsec_per_tick, (u32)shift_hz);
 

^ permalink raw reply related	[flat|nested] 69+ messages in thread

* [GIT PULL] timer fixes
@ 2016-10-28  8:38 Ingo Molnar
  0 siblings, 0 replies; 69+ messages in thread
From: Ingo Molnar @ 2016-10-28  8:38 UTC (permalink / raw)
  To: Linus Torvalds
  Cc: linux-kernel, Thomas Gleixner, Peter Zijlstra, Andrew Morton

Linus,

Please pull the latest timers-urgent-for-linus git tree from:

   git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git timers-urgent-for-linus

   # HEAD: 6bad6bccf2d717f652d37e63cf261eaa23466009 timers: Prevent base clock corruption when forwarding

Fix four timer locking races: two were noticed by you while reviewing the code 
while chasing for a corruption bug, and two from fixing spurious USB timeouts.

 Thanks,

	Ingo

------------------>
Thomas Gleixner (4):
      timers: Plug locking race vs. timer migration
      timers: Lock base for same bucket optimization
      timers: Prevent base clock rewind when forwarding clock
      timers: Prevent base clock corruption when forwarding


 kernel/time/timer.c | 74 +++++++++++++++++++++++++++++++----------------------
 1 file changed, 44 insertions(+), 30 deletions(-)

diff --git a/kernel/time/timer.c b/kernel/time/timer.c
index 2d47980a1bc4..c611c47de884 100644
--- a/kernel/time/timer.c
+++ b/kernel/time/timer.c
@@ -878,7 +878,7 @@ static inline struct timer_base *get_timer_base(u32 tflags)
 
 #ifdef CONFIG_NO_HZ_COMMON
 static inline struct timer_base *
-__get_target_base(struct timer_base *base, unsigned tflags)
+get_target_base(struct timer_base *base, unsigned tflags)
 {
 #ifdef CONFIG_SMP
 	if ((tflags & TIMER_PINNED) || !base->migration_enabled)
@@ -891,25 +891,27 @@ __get_target_base(struct timer_base *base, unsigned tflags)
 
 static inline void forward_timer_base(struct timer_base *base)
 {
+	unsigned long jnow = READ_ONCE(jiffies);
+
 	/*
 	 * We only forward the base when it's idle and we have a delta between
 	 * base clock and jiffies.
 	 */
-	if (!base->is_idle || (long) (jiffies - base->clk) < 2)
+	if (!base->is_idle || (long) (jnow - base->clk) < 2)
 		return;
 
 	/*
 	 * If the next expiry value is > jiffies, then we fast forward to
 	 * jiffies otherwise we forward to the next expiry value.
 	 */
-	if (time_after(base->next_expiry, jiffies))
-		base->clk = jiffies;
+	if (time_after(base->next_expiry, jnow))
+		base->clk = jnow;
 	else
 		base->clk = base->next_expiry;
 }
 #else
 static inline struct timer_base *
-__get_target_base(struct timer_base *base, unsigned tflags)
+get_target_base(struct timer_base *base, unsigned tflags)
 {
 	return get_timer_this_cpu_base(tflags);
 }
@@ -917,14 +919,6 @@ __get_target_base(struct timer_base *base, unsigned tflags)
 static inline void forward_timer_base(struct timer_base *base) { }
 #endif
 
-static inline struct timer_base *
-get_target_base(struct timer_base *base, unsigned tflags)
-{
-	struct timer_base *target = __get_target_base(base, tflags);
-
-	forward_timer_base(target);
-	return target;
-}
 
 /*
  * We are using hashed locking: Holding per_cpu(timer_bases[x]).lock means
@@ -943,7 +937,14 @@ static struct timer_base *lock_timer_base(struct timer_list *timer,
 {
 	for (;;) {
 		struct timer_base *base;
-		u32 tf = timer->flags;
+		u32 tf;
+
+		/*
+		 * We need to use READ_ONCE() here, otherwise the compiler
+		 * might re-read @tf between the check for TIMER_MIGRATING
+		 * and spin_lock().
+		 */
+		tf = READ_ONCE(timer->flags);
 
 		if (!(tf & TIMER_MIGRATING)) {
 			base = get_timer_base(tf);
@@ -964,6 +965,8 @@ __mod_timer(struct timer_list *timer, unsigned long expires, bool pending_only)
 	unsigned long clk = 0, flags;
 	int ret = 0;
 
+	BUG_ON(!timer->function);
+
 	/*
 	 * This is a common optimization triggered by the networking code - if
 	 * the timer is re-modified to have the same timeout or ends up in the
@@ -972,13 +975,16 @@ __mod_timer(struct timer_list *timer, unsigned long expires, bool pending_only)
 	if (timer_pending(timer)) {
 		if (timer->expires == expires)
 			return 1;
+
 		/*
-		 * Take the current timer_jiffies of base, but without holding
-		 * the lock!
+		 * We lock timer base and calculate the bucket index right
+		 * here. If the timer ends up in the same bucket, then we
+		 * just update the expiry time and avoid the whole
+		 * dequeue/enqueue dance.
 		 */
-		base = get_timer_base(timer->flags);
-		clk = base->clk;
+		base = lock_timer_base(timer, &flags);
 
+		clk = base->clk;
 		idx = calc_wheel_index(expires, clk);
 
 		/*
@@ -988,14 +994,14 @@ __mod_timer(struct timer_list *timer, unsigned long expires, bool pending_only)
 		 */
 		if (idx == timer_get_idx(timer)) {
 			timer->expires = expires;
-			return 1;
+			ret = 1;
+			goto out_unlock;
 		}
+	} else {
+		base = lock_timer_base(timer, &flags);
 	}
 
 	timer_stats_timer_set_start_info(timer);
-	BUG_ON(!timer->function);
-
-	base = lock_timer_base(timer, &flags);
 
 	ret = detach_if_pending(timer, base, false);
 	if (!ret && pending_only)
@@ -1025,12 +1031,16 @@ __mod_timer(struct timer_list *timer, unsigned long expires, bool pending_only)
 		}
 	}
 
+	/* Try to forward a stale timer base clock */
+	forward_timer_base(base);
+
 	timer->expires = expires;
 	/*
 	 * If 'idx' was calculated above and the base time did not advance
-	 * between calculating 'idx' and taking the lock, only enqueue_timer()
-	 * and trigger_dyntick_cpu() is required. Otherwise we need to
-	 * (re)calculate the wheel index via internal_add_timer().
+	 * between calculating 'idx' and possibly switching the base, only
+	 * enqueue_timer() and trigger_dyntick_cpu() is required. Otherwise
+	 * we need to (re)calculate the wheel index via
+	 * internal_add_timer().
 	 */
 	if (idx != UINT_MAX && clk == base->clk) {
 		enqueue_timer(base, timer, idx);
@@ -1510,12 +1520,16 @@ u64 get_next_timer_interrupt(unsigned long basej, u64 basem)
 	is_max_delta = (nextevt == base->clk + NEXT_TIMER_MAX_DELTA);
 	base->next_expiry = nextevt;
 	/*
-	 * We have a fresh next event. Check whether we can forward the base:
+	 * We have a fresh next event. Check whether we can forward the
+	 * base. We can only do that when @basej is past base->clk
+	 * otherwise we might rewind base->clk.
 	 */
-	if (time_after(nextevt, jiffies))
-		base->clk = jiffies;
-	else if (time_after(nextevt, base->clk))
-		base->clk = nextevt;
+	if (time_after(basej, base->clk)) {
+		if (time_after(nextevt, basej))
+			base->clk = basej;
+		else if (time_after(nextevt, base->clk))
+			base->clk = nextevt;
+	}
 
 	if (time_before_eq(nextevt, basej)) {
 		expires = basem;

^ permalink raw reply related	[flat|nested] 69+ messages in thread

* [GIT PULL] timer fixes
@ 2016-08-18 20:46 Ingo Molnar
  0 siblings, 0 replies; 69+ messages in thread
From: Ingo Molnar @ 2016-08-18 20:46 UTC (permalink / raw)
  To: Linus Torvalds
  Cc: linux-kernel, Thomas Gleixner, Peter Zijlstra, Andrew Morton

Linus,

Please pull the latest timers-urgent-for-linus git tree from:

   git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git timers-urgent-for-linus

   # HEAD: be5769e2061ac40b32daa83e28e1c4aac7133511 clocksource/drivers/mips-gic-timer: Make gic_clocksource_of_init() return int

Three clocksource driver fixes.

 Thanks,

	Ingo

------------------>
Arnd Bergmann (1):
      clocksource/drivers/kona: Fix get_counter() error handling

Gregory CLEMENT (1):
      clocksource/drivers/time-armada-370-xp: Fix the clock reference

Paul Gortmaker (1):
      clocksource/drivers/mips-gic-timer: Make gic_clocksource_of_init() return int


 drivers/clocksource/bcm_kona_timer.c     | 16 ++++++++++------
 drivers/clocksource/mips-gic-timer.c     |  2 +-
 drivers/clocksource/time-armada-370-xp.c |  1 -
 3 files changed, 11 insertions(+), 8 deletions(-)

diff --git a/drivers/clocksource/bcm_kona_timer.c b/drivers/clocksource/bcm_kona_timer.c
index 7e3fd375a627..92f6e4deee74 100644
--- a/drivers/clocksource/bcm_kona_timer.c
+++ b/drivers/clocksource/bcm_kona_timer.c
@@ -66,10 +66,10 @@ static void kona_timer_disable_and_clear(void __iomem *base)
 
 }
 
-static void
+static int
 kona_timer_get_counter(void __iomem *timer_base, uint32_t *msw, uint32_t *lsw)
 {
-	int loop_limit = 4;
+	int loop_limit = 3;
 
 	/*
 	 * Read 64-bit free running counter
@@ -83,18 +83,19 @@ kona_timer_get_counter(void __iomem *timer_base, uint32_t *msw, uint32_t *lsw)
 	 *      if new hi-word is equal to previously read hi-word then stop.
 	 */
 
-	while (--loop_limit) {
+	do {
 		*msw = readl(timer_base + KONA_GPTIMER_STCHI_OFFSET);
 		*lsw = readl(timer_base + KONA_GPTIMER_STCLO_OFFSET);
 		if (*msw == readl(timer_base + KONA_GPTIMER_STCHI_OFFSET))
 			break;
-	}
+	} while (--loop_limit);
 	if (!loop_limit) {
 		pr_err("bcm_kona_timer: getting counter failed.\n");
 		pr_err(" Timer will be impacted\n");
+		return -ETIMEDOUT;
 	}
 
-	return;
+	return 0;
 }
 
 static int kona_timer_set_next_event(unsigned long clc,
@@ -112,8 +113,11 @@ static int kona_timer_set_next_event(unsigned long clc,
 
 	uint32_t lsw, msw;
 	uint32_t reg;
+	int ret;
 
-	kona_timer_get_counter(timers.tmr_regs, &msw, &lsw);
+	ret = kona_timer_get_counter(timers.tmr_regs, &msw, &lsw);
+	if (ret)
+		return ret;
 
 	/* Load the "next" event tick value */
 	writel(lsw + clc, timers.tmr_regs + KONA_GPTIMER_STCM0_OFFSET);
diff --git a/drivers/clocksource/mips-gic-timer.c b/drivers/clocksource/mips-gic-timer.c
index d91e8725917c..b4b3ab5a11ad 100644
--- a/drivers/clocksource/mips-gic-timer.c
+++ b/drivers/clocksource/mips-gic-timer.c
@@ -164,7 +164,7 @@ void __init gic_clocksource_init(unsigned int frequency)
 	gic_start_count();
 }
 
-static void __init gic_clocksource_of_init(struct device_node *node)
+static int __init gic_clocksource_of_init(struct device_node *node)
 {
 	struct clk *clk;
 	int ret;
diff --git a/drivers/clocksource/time-armada-370-xp.c b/drivers/clocksource/time-armada-370-xp.c
index 719b478d136e..3c39e6f45971 100644
--- a/drivers/clocksource/time-armada-370-xp.c
+++ b/drivers/clocksource/time-armada-370-xp.c
@@ -338,7 +338,6 @@ static int __init armada_xp_timer_init(struct device_node *np)
 	struct clk *clk = of_clk_get_by_name(np, "fixed");
 	int ret;
 
-	clk = of_clk_get(np, 0);
 	if (IS_ERR(clk)) {
 		pr_err("Failed to get clock");
 		return PTR_ERR(clk);

^ permalink raw reply related	[flat|nested] 69+ messages in thread

* [GIT PULL] timer fixes
@ 2016-08-12 19:43 Ingo Molnar
  0 siblings, 0 replies; 69+ messages in thread
From: Ingo Molnar @ 2016-08-12 19:43 UTC (permalink / raw)
  To: Linus Torvalds
  Cc: linux-kernel, Thomas Gleixner, Peter Zijlstra, Andrew Morton

Linus,

Please pull the latest timers-urgent-for-linus git tree from:

   git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git timers-urgent-for-linus

   # HEAD: 22cc1ca3c5469cf17e149be232817b9223afa5e4 x86/hpet: Fix /dev/rtc breakage caused by RTC cleanup

Misc fixes: a /dev/rtc regression fix, two APIC timer period calibration fixes, an 
ARM clocksource driver fix and a NOHZ power use regression fix.

 Thanks,

	Ingo

------------------>
Arnd Bergmann (1):
      x86/hpet: Fix /dev/rtc breakage caused by RTC cleanup

Chris Metcalf (1):
      timers: Fix get_next_timer_interrupt() computation

Marc Zyngier (1):
      clocksource/arm_arch_timer: Force per-CPU interrupt to be level-triggered

Nicolai Stange (2):
      x86/timers/apic: Fix imprecise timer interrupts by eliminating TSC clockevents frequency roundoff error
      x86/timers/apic: Inform TSC deadline clockevent device about recalibration


 arch/x86/include/asm/apic.h          |  2 ++
 arch/x86/kernel/apic/apic.c          | 28 ++++++++++++++++++++++++++--
 arch/x86/kernel/hpet.c               |  2 +-
 arch/x86/kernel/tsc.c                |  4 ++++
 drivers/clocksource/arm_arch_timer.c | 26 +++++++++++++++++++++++---
 kernel/time/timer.c                  |  5 ++++-
 6 files changed, 60 insertions(+), 7 deletions(-)

diff --git a/arch/x86/include/asm/apic.h b/arch/x86/include/asm/apic.h
index f5befd4945f2..124357773ffa 100644
--- a/arch/x86/include/asm/apic.h
+++ b/arch/x86/include/asm/apic.h
@@ -135,6 +135,7 @@ extern void init_apic_mappings(void);
 void register_lapic_address(unsigned long address);
 extern void setup_boot_APIC_clock(void);
 extern void setup_secondary_APIC_clock(void);
+extern void lapic_update_tsc_freq(void);
 extern int APIC_init_uniprocessor(void);
 
 #ifdef CONFIG_X86_64
@@ -170,6 +171,7 @@ static inline void init_apic_mappings(void) { }
 static inline void disable_local_APIC(void) { }
 # define setup_boot_APIC_clock x86_init_noop
 # define setup_secondary_APIC_clock x86_init_noop
+static inline void lapic_update_tsc_freq(void) { }
 #endif /* !CONFIG_X86_LOCAL_APIC */
 
 #ifdef CONFIG_X86_X2APIC
diff --git a/arch/x86/kernel/apic/apic.c b/arch/x86/kernel/apic/apic.c
index 20abd912f0e4..cea4fc19e844 100644
--- a/arch/x86/kernel/apic/apic.c
+++ b/arch/x86/kernel/apic/apic.c
@@ -313,7 +313,7 @@ int lapic_get_maxlvt(void)
 
 /* Clock divisor */
 #define APIC_DIVISOR 16
-#define TSC_DIVISOR  32
+#define TSC_DIVISOR  8
 
 /*
  * This function sets up the local APIC timer, with a timeout of
@@ -565,13 +565,37 @@ static void setup_APIC_timer(void)
 				    CLOCK_EVT_FEAT_DUMMY);
 		levt->set_next_event = lapic_next_deadline;
 		clockevents_config_and_register(levt,
-						(tsc_khz / TSC_DIVISOR) * 1000,
+						tsc_khz * (1000 / TSC_DIVISOR),
 						0xF, ~0UL);
 	} else
 		clockevents_register_device(levt);
 }
 
 /*
+ * Install the updated TSC frequency from recalibration at the TSC
+ * deadline clockevent devices.
+ */
+static void __lapic_update_tsc_freq(void *info)
+{
+	struct clock_event_device *levt = this_cpu_ptr(&lapic_events);
+
+	if (!this_cpu_has(X86_FEATURE_TSC_DEADLINE_TIMER))
+		return;
+
+	clockevents_update_freq(levt, tsc_khz * (1000 / TSC_DIVISOR));
+}
+
+void lapic_update_tsc_freq(void)
+{
+	/*
+	 * The clockevent device's ->mult and ->shift can both be
+	 * changed. In order to avoid races, schedule the frequency
+	 * update code on each CPU.
+	 */
+	on_each_cpu(__lapic_update_tsc_freq, NULL, 0);
+}
+
+/*
  * In this functions we calibrate APIC bus clocks to the external timer.
  *
  * We want to do the calibration only once since we want to have local timer
diff --git a/arch/x86/kernel/hpet.c b/arch/x86/kernel/hpet.c
index ed16e58658a4..c6dfd801df97 100644
--- a/arch/x86/kernel/hpet.c
+++ b/arch/x86/kernel/hpet.c
@@ -1242,7 +1242,7 @@ irqreturn_t hpet_rtc_interrupt(int irq, void *dev_id)
 	memset(&curr_time, 0, sizeof(struct rtc_time));
 
 	if (hpet_rtc_flags & (RTC_UIE | RTC_AIE))
-		mc146818_set_time(&curr_time);
+		mc146818_get_time(&curr_time);
 
 	if (hpet_rtc_flags & RTC_UIE &&
 	    curr_time.tm_sec != hpet_prev_update_sec) {
diff --git a/arch/x86/kernel/tsc.c b/arch/x86/kernel/tsc.c
index 1ef87e887051..78b9cb5a26af 100644
--- a/arch/x86/kernel/tsc.c
+++ b/arch/x86/kernel/tsc.c
@@ -22,6 +22,7 @@
 #include <asm/nmi.h>
 #include <asm/x86_init.h>
 #include <asm/geode.h>
+#include <asm/apic.h>
 
 unsigned int __read_mostly cpu_khz;	/* TSC clocks / usec, not used here */
 EXPORT_SYMBOL(cpu_khz);
@@ -1249,6 +1250,9 @@ static void tsc_refine_calibration_work(struct work_struct *work)
 		(unsigned long)tsc_khz / 1000,
 		(unsigned long)tsc_khz % 1000);
 
+	/* Inform the TSC deadline clockevent devices about the recalibration */
+	lapic_update_tsc_freq();
+
 out:
 	if (boot_cpu_has(X86_FEATURE_ART))
 		art_related_clocksource = &clocksource_tsc;
diff --git a/drivers/clocksource/arm_arch_timer.c b/drivers/clocksource/arm_arch_timer.c
index 28bce3f4f81d..57700541f951 100644
--- a/drivers/clocksource/arm_arch_timer.c
+++ b/drivers/clocksource/arm_arch_timer.c
@@ -8,6 +8,9 @@
  * it under the terms of the GNU General Public License version 2 as
  * published by the Free Software Foundation.
  */
+
+#define pr_fmt(fmt)	"arm_arch_timer: " fmt
+
 #include <linux/init.h>
 #include <linux/kernel.h>
 #include <linux/device.h>
@@ -370,16 +373,33 @@ static bool arch_timer_has_nonsecure_ppi(void)
 		arch_timer_ppi[PHYS_NONSECURE_PPI]);
 }
 
+static u32 check_ppi_trigger(int irq)
+{
+	u32 flags = irq_get_trigger_type(irq);
+
+	if (flags != IRQF_TRIGGER_HIGH && flags != IRQF_TRIGGER_LOW) {
+		pr_warn("WARNING: Invalid trigger for IRQ%d, assuming level low\n", irq);
+		pr_warn("WARNING: Please fix your firmware\n");
+		flags = IRQF_TRIGGER_LOW;
+	}
+
+	return flags;
+}
+
 static int arch_timer_starting_cpu(unsigned int cpu)
 {
 	struct clock_event_device *clk = this_cpu_ptr(arch_timer_evt);
+	u32 flags;
 
 	__arch_timer_setup(ARCH_CP15_TIMER, clk);
 
-	enable_percpu_irq(arch_timer_ppi[arch_timer_uses_ppi], 0);
+	flags = check_ppi_trigger(arch_timer_ppi[arch_timer_uses_ppi]);
+	enable_percpu_irq(arch_timer_ppi[arch_timer_uses_ppi], flags);
 
-	if (arch_timer_has_nonsecure_ppi())
-		enable_percpu_irq(arch_timer_ppi[PHYS_NONSECURE_PPI], 0);
+	if (arch_timer_has_nonsecure_ppi()) {
+		flags = check_ppi_trigger(arch_timer_ppi[PHYS_NONSECURE_PPI]);
+		enable_percpu_irq(arch_timer_ppi[PHYS_NONSECURE_PPI], flags);
+	}
 
 	arch_counter_set_user_access();
 	if (evtstrm_enable)
diff --git a/kernel/time/timer.c b/kernel/time/timer.c
index 555670a5143c..32bf6f75a8fe 100644
--- a/kernel/time/timer.c
+++ b/kernel/time/timer.c
@@ -1496,6 +1496,7 @@ u64 get_next_timer_interrupt(unsigned long basej, u64 basem)
 	struct timer_base *base = this_cpu_ptr(&timer_bases[BASE_STD]);
 	u64 expires = KTIME_MAX;
 	unsigned long nextevt;
+	bool is_max_delta;
 
 	/*
 	 * Pretend that there is no timer pending if the cpu is offline.
@@ -1506,6 +1507,7 @@ u64 get_next_timer_interrupt(unsigned long basej, u64 basem)
 
 	spin_lock(&base->lock);
 	nextevt = __next_timer_interrupt(base);
+	is_max_delta = (nextevt == base->clk + NEXT_TIMER_MAX_DELTA);
 	base->next_expiry = nextevt;
 	/*
 	 * We have a fresh next event. Check whether we can forward the base:
@@ -1519,7 +1521,8 @@ u64 get_next_timer_interrupt(unsigned long basej, u64 basem)
 		expires = basem;
 		base->is_idle = false;
 	} else {
-		expires = basem + (nextevt - basej) * TICK_NSEC;
+		if (!is_max_delta)
+			expires = basem + (nextevt - basej) * TICK_NSEC;
 		/*
 		 * If we expect to sleep more than a tick, mark the base idle:
 		 */

^ permalink raw reply related	[flat|nested] 69+ messages in thread

* [GIT PULL] timer fixes
@ 2016-01-14 10:12 Ingo Molnar
  0 siblings, 0 replies; 69+ messages in thread
From: Ingo Molnar @ 2016-01-14 10:12 UTC (permalink / raw)
  To: Linus Torvalds
  Cc: linux-kernel, Thomas Gleixner, Peter Zijlstra, Andrew Morton

Linus,

Please pull the latest timers-urgent-for-linus git tree from:

   git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git timers-urgent-for-linus

   # HEAD: 1f16f116b01c110db20ab808562c8b8bc3ee3d6e Merge branches 'clockevents/4.4-fixes' and 'clockevents/4.5-fixes' of http://git.linaro.org/people/daniel.lezcano/linux into timers/urgent

Three clocksource driver fixes.

 Thanks,

	Ingo

------------------>
Daniel Lezcano (1):
      clocksource/drivers/fsl_ftm_timer: Fix CLKSRC_MMIO dependency

Richard Weinberger (1):
      clocksource/drivers: Fix dependencies for !HAS_IOMEM archs

Roman Volkov (1):
      clocksource/drivers/vt8500: Increase the minimum delta

Thomas Gleixner (1):
      Merge branches 'clockevents/4.4-fixes' and 'clockevents/4.5-fixes' of http://git.linaro.org/people/daniel.lezcano/linux into timers/urgent

 drivers/clocksource/Kconfig        | 13 ++++++++-----
 drivers/clocksource/vt8500_timer.c |  6 ++++--
 2 files changed, 12 insertions(+), 7 deletions(-)

^ permalink raw reply	[flat|nested] 69+ messages in thread

* [GIT PULL] timer fixes
@ 2015-10-03 10:20 Ingo Molnar
  0 siblings, 0 replies; 69+ messages in thread
From: Ingo Molnar @ 2015-10-03 10:20 UTC (permalink / raw)
  To: Linus Torvalds
  Cc: linux-kernel, Thomas Gleixner, John Stultz, Peter Zijlstra,
	Andrew Morton

Linus,

Please pull the latest timers-urgent-for-linus git tree from:

   git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git timers-urgent-for-linus

   # HEAD: 67dfae0cd72fec5cd158b6e5fb1647b7dbe0834c clocksource: Fix abs() usage w/ 64bit values

An abs64() fix in the watchdog driver, and two clocksource driver NO_IRQ 
assumption fixes.

 Thanks,

	Ingo

------------------>
Daniel Lezcano (2):
      clocksource/drivers/rockchip: Fix bad NO_IRQ usage
      clocksource/drivers/keystone: Fix bad NO_IRQ usage

John Stultz (1):
      clocksource: Fix abs() usage w/ 64bit values


 drivers/clocksource/rockchip_timer.c | 2 +-
 drivers/clocksource/timer-keystone.c | 2 +-
 kernel/time/clocksource.c            | 2 +-
 3 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/clocksource/rockchip_timer.c b/drivers/clocksource/rockchip_timer.c
index bb2c2b050964..d3c1742ded1a 100644
--- a/drivers/clocksource/rockchip_timer.c
+++ b/drivers/clocksource/rockchip_timer.c
@@ -148,7 +148,7 @@ static void __init rk_timer_init(struct device_node *np)
 	bc_timer.freq = clk_get_rate(timer_clk);
 
 	irq = irq_of_parse_and_map(np, 0);
-	if (irq == NO_IRQ) {
+	if (!irq) {
 		pr_err("Failed to map interrupts for '%s'\n", TIMER_NAME);
 		return;
 	}
diff --git a/drivers/clocksource/timer-keystone.c b/drivers/clocksource/timer-keystone.c
index edacf3902e10..1cea08cf603e 100644
--- a/drivers/clocksource/timer-keystone.c
+++ b/drivers/clocksource/timer-keystone.c
@@ -152,7 +152,7 @@ static void __init keystone_timer_init(struct device_node *np)
 	int irq, error;
 
 	irq  = irq_of_parse_and_map(np, 0);
-	if (irq == NO_IRQ) {
+	if (!irq) {
 		pr_err("%s: failed to map interrupts\n", __func__);
 		return;
 	}
diff --git a/kernel/time/clocksource.c b/kernel/time/clocksource.c
index 841b72f720e8..3a38775b50c2 100644
--- a/kernel/time/clocksource.c
+++ b/kernel/time/clocksource.c
@@ -217,7 +217,7 @@ static void clocksource_watchdog(unsigned long data)
 			continue;
 
 		/* Check the deviation from the watchdog clocksource. */
-		if ((abs(cs_nsec - wd_nsec) > WATCHDOG_THRESHOLD)) {
+		if (abs64(cs_nsec - wd_nsec) > WATCHDOG_THRESHOLD) {
 			pr_warn("timekeeping watchdog: Marking clocksource '%s' as unstable because the skew is too large:\n",
 				cs->name);
 			pr_warn("                      '%s' wd_now: %llx wd_last: %llx mask: %llx\n",

^ permalink raw reply related	[flat|nested] 69+ messages in thread

* [GIT PULL] timer fixes
@ 2015-09-17  8:13 Ingo Molnar
  0 siblings, 0 replies; 69+ messages in thread
From: Ingo Molnar @ 2015-09-17  8:13 UTC (permalink / raw)
  To: Linus Torvalds
  Cc: linux-kernel, Thomas Gleixner, Peter Zijlstra, Andrew Morton

Linus,

Please pull the latest timers-urgent-for-linus git tree from:

   git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git timers-urgent-for-linus

   # HEAD: eef7635a22f6b144206b5ca2f1398f637acffc4d clockevents: Remove unused set_mode() callback

An fix for an abs()/abs64() bug that caused too slow NTP convergence on 32-bit 
kernels, plus a removal of an obsolete clockevents driver facility after all users 
got converted during the merge window.

 Thanks,

	Ingo

------------------>
John Stultz (1):
      time: Fix timekeeping_freqadjust()'s incorrect use of abs() instead of abs64()

Viresh Kumar (1):
      clockevents: Remove unused set_mode() callback


 include/linux/clockchips.h | 29 +++++--------------------
 kernel/time/clockevents.c  | 42 +-----------------------------------
 kernel/time/tick-common.c  |  1 -
 kernel/time/timekeeping.c  |  2 +-
 kernel/time/timer_list.c   | 54 +++++++++++++++++++++-------------------------
 5 files changed, 31 insertions(+), 97 deletions(-)

diff --git a/include/linux/clockchips.h b/include/linux/clockchips.h
index 31ce435981fe..bdcf358dfce2 100644
--- a/include/linux/clockchips.h
+++ b/include/linux/clockchips.h
@@ -18,15 +18,6 @@
 struct clock_event_device;
 struct module;
 
-/* Clock event mode commands for legacy ->set_mode(): OBSOLETE */
-enum clock_event_mode {
-	CLOCK_EVT_MODE_UNUSED,
-	CLOCK_EVT_MODE_SHUTDOWN,
-	CLOCK_EVT_MODE_PERIODIC,
-	CLOCK_EVT_MODE_ONESHOT,
-	CLOCK_EVT_MODE_RESUME,
-};
-
 /*
  * Possible states of a clock event device.
  *
@@ -86,16 +77,14 @@ enum clock_event_state {
  * @min_delta_ns:	minimum delta value in ns
  * @mult:		nanosecond to cycles multiplier
  * @shift:		nanoseconds to cycles divisor (power of two)
- * @mode:		operating mode, relevant only to ->set_mode(), OBSOLETE
  * @state_use_accessors:current state of the device, assigned by the core code
  * @features:		features
  * @retries:		number of forced programming retries
- * @set_mode:		legacy set mode function, only for modes <= CLOCK_EVT_MODE_RESUME.
- * @set_state_periodic:	switch state to periodic, if !set_mode
- * @set_state_oneshot:	switch state to oneshot, if !set_mode
- * @set_state_oneshot_stopped: switch state to oneshot_stopped, if !set_mode
- * @set_state_shutdown:	switch state to shutdown, if !set_mode
- * @tick_resume:	resume clkevt device, if !set_mode
+ * @set_state_periodic:	switch state to periodic
+ * @set_state_oneshot:	switch state to oneshot
+ * @set_state_oneshot_stopped: switch state to oneshot_stopped
+ * @set_state_shutdown:	switch state to shutdown
+ * @tick_resume:	resume clkevt device
  * @broadcast:		function to broadcast events
  * @min_delta_ticks:	minimum delta value in ticks stored for reconfiguration
  * @max_delta_ticks:	maximum delta value in ticks stored for reconfiguration
@@ -116,18 +105,10 @@ struct clock_event_device {
 	u64			min_delta_ns;
 	u32			mult;
 	u32			shift;
-	enum clock_event_mode	mode;
 	enum clock_event_state	state_use_accessors;
 	unsigned int		features;
 	unsigned long		retries;
 
-	/*
-	 * State transition callback(s): Only one of the two groups should be
-	 * defined:
-	 * - set_mode(), only for modes <= CLOCK_EVT_MODE_RESUME.
-	 * - set_state_{shutdown|periodic|oneshot|oneshot_stopped}(), tick_resume().
-	 */
-	void			(*set_mode)(enum clock_event_mode mode, struct clock_event_device *);
 	int			(*set_state_periodic)(struct clock_event_device *);
 	int			(*set_state_oneshot)(struct clock_event_device *);
 	int			(*set_state_oneshot_stopped)(struct clock_event_device *);
diff --git a/kernel/time/clockevents.c b/kernel/time/clockevents.c
index 50eb107f1198..a9b76a40319e 100644
--- a/kernel/time/clockevents.c
+++ b/kernel/time/clockevents.c
@@ -97,20 +97,6 @@ EXPORT_SYMBOL_GPL(clockevent_delta2ns);
 static int __clockevents_switch_state(struct clock_event_device *dev,
 				      enum clock_event_state state)
 {
-	/* Transition with legacy set_mode() callback */
-	if (dev->set_mode) {
-		/* Legacy callback doesn't support new modes */
-		if (state > CLOCK_EVT_STATE_ONESHOT)
-			return -ENOSYS;
-		/*
-		 * 'clock_event_state' and 'clock_event_mode' have 1-to-1
-		 * mapping until *_ONESHOT, and so a simple cast will work.
-		 */
-		dev->set_mode((enum clock_event_mode)state, dev);
-		dev->mode = (enum clock_event_mode)state;
-		return 0;
-	}
-
 	if (dev->features & CLOCK_EVT_FEAT_DUMMY)
 		return 0;
 
@@ -204,12 +190,8 @@ int clockevents_tick_resume(struct clock_event_device *dev)
 {
 	int ret = 0;
 
-	if (dev->set_mode) {
-		dev->set_mode(CLOCK_EVT_MODE_RESUME, dev);
-		dev->mode = CLOCK_EVT_MODE_RESUME;
-	} else if (dev->tick_resume) {
+	if (dev->tick_resume)
 		ret = dev->tick_resume(dev);
-	}
 
 	return ret;
 }
@@ -460,26 +442,6 @@ int clockevents_unbind_device(struct clock_event_device *ced, int cpu)
 }
 EXPORT_SYMBOL_GPL(clockevents_unbind_device);
 
-/* Sanity check of state transition callbacks */
-static int clockevents_sanity_check(struct clock_event_device *dev)
-{
-	/* Legacy set_mode() callback */
-	if (dev->set_mode) {
-		/* We shouldn't be supporting new modes now */
-		WARN_ON(dev->set_state_periodic || dev->set_state_oneshot ||
-			dev->set_state_shutdown || dev->tick_resume ||
-			dev->set_state_oneshot_stopped);
-
-		BUG_ON(dev->mode != CLOCK_EVT_MODE_UNUSED);
-		return 0;
-	}
-
-	if (dev->features & CLOCK_EVT_FEAT_DUMMY)
-		return 0;
-
-	return 0;
-}
-
 /**
  * clockevents_register_device - register a clock event device
  * @dev:	device to register
@@ -488,8 +450,6 @@ void clockevents_register_device(struct clock_event_device *dev)
 {
 	unsigned long flags;
 
-	BUG_ON(clockevents_sanity_check(dev));
-
 	/* Initialize state to DETACHED */
 	clockevent_set_state(dev, CLOCK_EVT_STATE_DETACHED);
 
diff --git a/kernel/time/tick-common.c b/kernel/time/tick-common.c
index d11c55b6ab7d..4fcd99e12aa0 100644
--- a/kernel/time/tick-common.c
+++ b/kernel/time/tick-common.c
@@ -398,7 +398,6 @@ void tick_shutdown(unsigned int cpu)
 		 * the set mode function!
 		 */
 		clockevent_set_state(dev, CLOCK_EVT_STATE_DETACHED);
-		dev->mode = CLOCK_EVT_MODE_UNUSED;
 		clockevents_exchange_device(dev, NULL);
 		dev->event_handler = clockevents_handle_noop;
 		td->evtdev = NULL;
diff --git a/kernel/time/timekeeping.c b/kernel/time/timekeeping.c
index f6ee2e6b6f5d..3739ac6aa473 100644
--- a/kernel/time/timekeeping.c
+++ b/kernel/time/timekeeping.c
@@ -1614,7 +1614,7 @@ static __always_inline void timekeeping_freqadjust(struct timekeeper *tk,
 	negative = (tick_error < 0);
 
 	/* Sort out the magnitude of the correction */
-	tick_error = abs(tick_error);
+	tick_error = abs64(tick_error);
 	for (adj = 0; tick_error > interval; adj++)
 		tick_error >>= 1;
 
diff --git a/kernel/time/timer_list.c b/kernel/time/timer_list.c
index 129c96033e46..f75e35b60149 100644
--- a/kernel/time/timer_list.c
+++ b/kernel/time/timer_list.c
@@ -225,7 +225,7 @@ print_tickdevice(struct seq_file *m, struct tick_device *td, int cpu)
 		   (unsigned long long) dev->min_delta_ns);
 	SEQ_printf(m, " mult:           %u\n", dev->mult);
 	SEQ_printf(m, " shift:          %u\n", dev->shift);
-	SEQ_printf(m, " mode:           %d\n", dev->mode);
+	SEQ_printf(m, " mode:           %d\n", clockevent_get_state(dev));
 	SEQ_printf(m, " next_event:     %Ld nsecs\n",
 		   (unsigned long long) ktime_to_ns(dev->next_event));
 
@@ -233,40 +233,34 @@ print_tickdevice(struct seq_file *m, struct tick_device *td, int cpu)
 	print_name_offset(m, dev->set_next_event);
 	SEQ_printf(m, "\n");
 
-	if (dev->set_mode) {
-		SEQ_printf(m, " set_mode:       ");
-		print_name_offset(m, dev->set_mode);
+	if (dev->set_state_shutdown) {
+		SEQ_printf(m, " shutdown: ");
+		print_name_offset(m, dev->set_state_shutdown);
 		SEQ_printf(m, "\n");
-	} else {
-		if (dev->set_state_shutdown) {
-			SEQ_printf(m, " shutdown: ");
-			print_name_offset(m, dev->set_state_shutdown);
-			SEQ_printf(m, "\n");
-		}
+	}
 
-		if (dev->set_state_periodic) {
-			SEQ_printf(m, " periodic: ");
-			print_name_offset(m, dev->set_state_periodic);
-			SEQ_printf(m, "\n");
-		}
+	if (dev->set_state_periodic) {
+		SEQ_printf(m, " periodic: ");
+		print_name_offset(m, dev->set_state_periodic);
+		SEQ_printf(m, "\n");
+	}
 
-		if (dev->set_state_oneshot) {
-			SEQ_printf(m, " oneshot:  ");
-			print_name_offset(m, dev->set_state_oneshot);
-			SEQ_printf(m, "\n");
-		}
+	if (dev->set_state_oneshot) {
+		SEQ_printf(m, " oneshot:  ");
+		print_name_offset(m, dev->set_state_oneshot);
+		SEQ_printf(m, "\n");
+	}
 
-		if (dev->set_state_oneshot_stopped) {
-			SEQ_printf(m, " oneshot stopped: ");
-			print_name_offset(m, dev->set_state_oneshot_stopped);
-			SEQ_printf(m, "\n");
-		}
+	if (dev->set_state_oneshot_stopped) {
+		SEQ_printf(m, " oneshot stopped: ");
+		print_name_offset(m, dev->set_state_oneshot_stopped);
+		SEQ_printf(m, "\n");
+	}
 
-		if (dev->tick_resume) {
-			SEQ_printf(m, " resume:   ");
-			print_name_offset(m, dev->tick_resume);
-			SEQ_printf(m, "\n");
-		}
+	if (dev->tick_resume) {
+		SEQ_printf(m, " resume:   ");
+		print_name_offset(m, dev->tick_resume);
+		SEQ_printf(m, "\n");
 	}
 
 	SEQ_printf(m, " event_handler:  ");

^ permalink raw reply related	[flat|nested] 69+ messages in thread

* [GIT PULL] timer fixes
@ 2015-03-28 13:50 Ingo Molnar
  0 siblings, 0 replies; 69+ messages in thread
From: Ingo Molnar @ 2015-03-28 13:50 UTC (permalink / raw)
  To: Linus Torvalds
  Cc: linux-kernel, Peter Zijlstra, Thomas Gleixner, Paul E. McKenney,
	Andrew Morton

Linus,

Please pull the latest timers-urgent-for-linus git tree from:

   git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git timers-urgent-for-linus

   # HEAD: 6e206020324c50a95486f6b279a53512febed92d clocksource/drivers/sun5i: Fix cpufreq interaction with sched_clock()

Two clocksource driver fixes, and an idle loop RCU warning fix.

 Thanks,

	Ingo

------------------>
Maxime Ripard (1):
      clocksource/drivers/sun5i: Fix cpufreq interaction with sched_clock()

Preeti U Murthy (1):
      timers/tick/broadcast-hrtimer: Fix suspicious RCU usage in idle loop

Richard Weinberger (1):
      clocksource/drivers: Fix various !CONFIG_HAS_IOMEM build errors


 drivers/clocksource/Kconfig          |  3 +++
 drivers/clocksource/timer-sun5i.c    |  7 -------
 kernel/time/tick-broadcast-hrtimer.c | 11 +++++++++--
 3 files changed, 12 insertions(+), 9 deletions(-)

diff --git a/drivers/clocksource/Kconfig b/drivers/clocksource/Kconfig
index 68161f7a07d6..a0b036ccb118 100644
--- a/drivers/clocksource/Kconfig
+++ b/drivers/clocksource/Kconfig
@@ -192,6 +192,7 @@ config SYS_SUPPORTS_EM_STI
 config SH_TIMER_CMT
 	bool "Renesas CMT timer driver" if COMPILE_TEST
 	depends on GENERIC_CLOCKEVENTS
+	depends on HAS_IOMEM
 	default SYS_SUPPORTS_SH_CMT
 	help
 	  This enables build of a clocksource and clockevent driver for
@@ -201,6 +202,7 @@ config SH_TIMER_CMT
 config SH_TIMER_MTU2
 	bool "Renesas MTU2 timer driver" if COMPILE_TEST
 	depends on GENERIC_CLOCKEVENTS
+	depends on HAS_IOMEM
 	default SYS_SUPPORTS_SH_MTU2
 	help
 	  This enables build of a clockevent driver for the Multi-Function
@@ -210,6 +212,7 @@ config SH_TIMER_MTU2
 config SH_TIMER_TMU
 	bool "Renesas TMU timer driver" if COMPILE_TEST
 	depends on GENERIC_CLOCKEVENTS
+	depends on HAS_IOMEM
 	default SYS_SUPPORTS_SH_TMU
 	help
 	  This enables build of a clocksource and clockevent driver for
diff --git a/drivers/clocksource/timer-sun5i.c b/drivers/clocksource/timer-sun5i.c
index 5dcbf90b8015..58597fbcc046 100644
--- a/drivers/clocksource/timer-sun5i.c
+++ b/drivers/clocksource/timer-sun5i.c
@@ -17,7 +17,6 @@
 #include <linux/irq.h>
 #include <linux/irqreturn.h>
 #include <linux/reset.h>
-#include <linux/sched_clock.h>
 #include <linux/of.h>
 #include <linux/of_address.h>
 #include <linux/of_irq.h>
@@ -137,11 +136,6 @@ static struct irqaction sun5i_timer_irq = {
 	.dev_id = &sun5i_clockevent,
 };
 
-static u64 sun5i_timer_sched_read(void)
-{
-	return ~readl(timer_base + TIMER_CNTVAL_LO_REG(1));
-}
-
 static void __init sun5i_timer_init(struct device_node *node)
 {
 	struct reset_control *rstc;
@@ -172,7 +166,6 @@ static void __init sun5i_timer_init(struct device_node *node)
 	writel(TIMER_CTL_ENABLE | TIMER_CTL_RELOAD,
 	       timer_base + TIMER_CTL_REG(1));
 
-	sched_clock_register(sun5i_timer_sched_read, 32, rate);
 	clocksource_mmio_init(timer_base + TIMER_CNTVAL_LO_REG(1), node->name,
 			      rate, 340, 32, clocksource_mmio_readl_down);
 
diff --git a/kernel/time/tick-broadcast-hrtimer.c b/kernel/time/tick-broadcast-hrtimer.c
index eb682d5c697c..6aac4beedbbe 100644
--- a/kernel/time/tick-broadcast-hrtimer.c
+++ b/kernel/time/tick-broadcast-hrtimer.c
@@ -49,6 +49,7 @@ static void bc_set_mode(enum clock_event_mode mode,
  */
 static int bc_set_next(ktime_t expires, struct clock_event_device *bc)
 {
+	int bc_moved;
 	/*
 	 * We try to cancel the timer first. If the callback is on
 	 * flight on some other cpu then we let it handle it. If we
@@ -60,9 +61,15 @@ static int bc_set_next(ktime_t expires, struct clock_event_device *bc)
 	 * restart the timer because we are in the callback, but we
 	 * can set the expiry time and let the callback return
 	 * HRTIMER_RESTART.
+	 *
+	 * Since we are in the idle loop at this point and because
+	 * hrtimer_{start/cancel} functions call into tracing,
+	 * calls to these functions must be bound within RCU_NONIDLE.
 	 */
-	if (hrtimer_try_to_cancel(&bctimer) >= 0) {
-		hrtimer_start(&bctimer, expires, HRTIMER_MODE_ABS_PINNED);
+	RCU_NONIDLE(bc_moved = (hrtimer_try_to_cancel(&bctimer) >= 0) ?
+		!hrtimer_start(&bctimer, expires, HRTIMER_MODE_ABS_PINNED) :
+			0);
+	if (bc_moved) {
 		/* Bind the "device" to the cpu */
 		bc->bound_on = smp_processor_id();
 	} else if (bc->bound_on == smp_processor_id()) {

^ permalink raw reply related	[flat|nested] 69+ messages in thread

* [GIT PULL] timer fixes
@ 2015-03-17 16:50 Ingo Molnar
  0 siblings, 0 replies; 69+ messages in thread
From: Ingo Molnar @ 2015-03-17 16:50 UTC (permalink / raw)
  To: Linus Torvalds
  Cc: linux-kernel, Peter Zijlstra, Thomas Gleixner, Andrew Morton,
	Daniel Lezcano

Linus,

Please pull the latest timers-urgent-for-linus git tree from:

   git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git timers-urgent-for-linus

   # HEAD: 9ab2b7f3acfa5490f36c9e5774a2bfecd01943db Merge branch 'clockevents/4.0-rc2' of http://git.linaro.org/people/daniel.lezcano/linux into timers/urgent

Two clocksource driver fixes.

 Thanks,

	Ingo

------------------>
Yongbae Park (2):
      clocksource: efm32: Fix a NULL pointer dereference
      clockevents: sun5i: Fix setup_irq init sequence


 drivers/clocksource/time-efm32.c  | 4 ++--
 drivers/clocksource/timer-sun5i.c | 8 ++++----
 2 files changed, 6 insertions(+), 6 deletions(-)

diff --git a/drivers/clocksource/time-efm32.c b/drivers/clocksource/time-efm32.c
index bba62f9deefb..ec57ba2bbd87 100644
--- a/drivers/clocksource/time-efm32.c
+++ b/drivers/clocksource/time-efm32.c
@@ -225,12 +225,12 @@ static int __init efm32_clockevent_init(struct device_node *np)
 	clock_event_ddata.base = base;
 	clock_event_ddata.periodic_top = DIV_ROUND_CLOSEST(rate, 1024 * HZ);
 
-	setup_irq(irq, &efm32_clock_event_irq);
-
 	clockevents_config_and_register(&clock_event_ddata.evtdev,
 					DIV_ROUND_CLOSEST(rate, 1024),
 					0xf, 0xffff);
 
+	setup_irq(irq, &efm32_clock_event_irq);
+
 	return 0;
 
 err_get_irq:
diff --git a/drivers/clocksource/timer-sun5i.c b/drivers/clocksource/timer-sun5i.c
index 02268448dc85..5dcbf90b8015 100644
--- a/drivers/clocksource/timer-sun5i.c
+++ b/drivers/clocksource/timer-sun5i.c
@@ -178,10 +178,6 @@ static void __init sun5i_timer_init(struct device_node *node)
 
 	ticks_per_jiffy = DIV_ROUND_UP(rate, HZ);
 
-	ret = setup_irq(irq, &sun5i_timer_irq);
-	if (ret)
-		pr_warn("failed to setup irq %d\n", irq);
-
 	/* Enable timer0 interrupt */
 	val = readl(timer_base + TIMER_IRQ_EN_REG);
 	writel(val | TIMER_IRQ_EN(0), timer_base + TIMER_IRQ_EN_REG);
@@ -191,6 +187,10 @@ static void __init sun5i_timer_init(struct device_node *node)
 
 	clockevents_config_and_register(&sun5i_clockevent, rate,
 					TIMER_SYNC_TICKS, 0xffffffff);
+
+	ret = setup_irq(irq, &sun5i_timer_irq);
+	if (ret)
+		pr_warn("failed to setup irq %d\n", irq);
 }
 CLOCKSOURCE_OF_DECLARE(sun5i_a13, "allwinner,sun5i-a13-hstimer",
 		       sun5i_timer_init);

^ permalink raw reply related	[flat|nested] 69+ messages in thread

* [GIT PULL] timer fixes
@ 2015-03-01 17:06 Ingo Molnar
  0 siblings, 0 replies; 69+ messages in thread
From: Ingo Molnar @ 2015-03-01 17:06 UTC (permalink / raw)
  To: Linus Torvalds
  Cc: linux-kernel, Thomas Gleixner, Peter Zijlstra, Andrew Morton

Linus,

Please pull the latest timers-urgent-for-linus git tree from:

   git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git timers-urgent-for-linus

   # HEAD: e3eff6fe7d8a877d94f5dbb8fea46b4f8ca4ce5a Merge branch 'clockevents/4.0-rc1' of git://git.linaro.org/people/daniel.lezcano/linux into timers/urgent

Three clockevents/clocksource driver fixes.

 Thanks,

	Ingo

------------------>
Daniel Lezcano (1):
      clockevents: asm9260: Fix compilation error with sparc/sparc64 allyesconfig

Matthias Brugger (1):
      clocksource: mtk: Fix race conditions in probe code

Robert Jarzmik (1):
      clocksource: pxa: Fix section mismatch


 arch/arm/mach-asm9260/Kconfig   |  2 ++
 drivers/clocksource/Kconfig     | 16 +++++-----------
 drivers/clocksource/mtk_timer.c |  9 +++++----
 drivers/clocksource/pxa_timer.c |  2 +-
 4 files changed, 13 insertions(+), 16 deletions(-)

diff --git a/arch/arm/mach-asm9260/Kconfig b/arch/arm/mach-asm9260/Kconfig
index 8423be76080e..52241207a82a 100644
--- a/arch/arm/mach-asm9260/Kconfig
+++ b/arch/arm/mach-asm9260/Kconfig
@@ -2,5 +2,7 @@ config MACH_ASM9260
 	bool "Alphascale ASM9260"
 	depends on ARCH_MULTI_V5
 	select CPU_ARM926T
+	select ASM9260_TIMER
+	select GENERIC_CLOCKEVENTS
 	help
 	  Support for Alphascale ASM9260 based platform.
diff --git a/drivers/clocksource/Kconfig b/drivers/clocksource/Kconfig
index 1c2506f68122..68161f7a07d6 100644
--- a/drivers/clocksource/Kconfig
+++ b/drivers/clocksource/Kconfig
@@ -63,6 +63,11 @@ config VT8500_TIMER
 config CADENCE_TTC_TIMER
 	bool
 
+config ASM9260_TIMER
+	bool
+	select CLKSRC_MMIO
+	select CLKSRC_OF
+
 config CLKSRC_NOMADIK_MTU
 	bool
 	depends on (ARCH_NOMADIK || ARCH_U8500)
@@ -245,15 +250,4 @@ config CLKSRC_PXA
 	help
 	  This enables OST0 support available on PXA and SA-11x0
 	  platforms.
-
-config ASM9260_TIMER
-	bool "Alphascale ASM9260 timer driver"
-	depends on GENERIC_CLOCKEVENTS
-	select CLKSRC_MMIO
-	select CLKSRC_OF
-	default y if MACH_ASM9260
-	help
-	  This enables build of a clocksource and clockevent driver for
-	  the 32-bit System Timer hardware available on a Alphascale ASM9260.
-
 endmenu
diff --git a/drivers/clocksource/mtk_timer.c b/drivers/clocksource/mtk_timer.c
index 32a3d25795d3..68ab42356d0e 100644
--- a/drivers/clocksource/mtk_timer.c
+++ b/drivers/clocksource/mtk_timer.c
@@ -224,6 +224,8 @@ static void __init mtk_timer_init(struct device_node *node)
 	}
 	rate = clk_get_rate(clk);
 
+	mtk_timer_global_reset(evt);
+
 	if (request_irq(evt->dev.irq, mtk_timer_interrupt,
 			IRQF_TIMER | IRQF_IRQPOLL, "mtk_timer", evt)) {
 		pr_warn("failed to setup irq %d\n", evt->dev.irq);
@@ -232,8 +234,6 @@ static void __init mtk_timer_init(struct device_node *node)
 
 	evt->ticks_per_jiffy = DIV_ROUND_UP(rate, HZ);
 
-	mtk_timer_global_reset(evt);
-
 	/* Configure clock source */
 	mtk_timer_setup(evt, GPT_CLK_SRC, TIMER_CTRL_OP_FREERUN);
 	clocksource_mmio_init(evt->gpt_base + TIMER_CNT_REG(GPT_CLK_SRC),
@@ -241,10 +241,11 @@ static void __init mtk_timer_init(struct device_node *node)
 
 	/* Configure clock event */
 	mtk_timer_setup(evt, GPT_CLK_EVT, TIMER_CTRL_OP_REPEAT);
-	mtk_timer_enable_irq(evt, GPT_CLK_EVT);
-
 	clockevents_config_and_register(&evt->dev, rate, 0x3,
 					0xffffffff);
+
+	mtk_timer_enable_irq(evt, GPT_CLK_EVT);
+
 	return;
 
 err_clk_disable:
diff --git a/drivers/clocksource/pxa_timer.c b/drivers/clocksource/pxa_timer.c
index 941f3f344e08..d9438af2bbd6 100644
--- a/drivers/clocksource/pxa_timer.c
+++ b/drivers/clocksource/pxa_timer.c
@@ -163,7 +163,7 @@ static struct irqaction pxa_ost0_irq = {
 	.dev_id		= &ckevt_pxa_osmr0,
 };
 
-static void pxa_timer_common_init(int irq, unsigned long clock_tick_rate)
+static void __init pxa_timer_common_init(int irq, unsigned long clock_tick_rate)
 {
 	timer_writel(0, OIER);
 	timer_writel(OSSR_M0 | OSSR_M1 | OSSR_M2 | OSSR_M3, OSSR);

^ permalink raw reply related	[flat|nested] 69+ messages in thread

* [GIT PULL] timer fixes
@ 2013-12-17 13:52 Ingo Molnar
  0 siblings, 0 replies; 69+ messages in thread
From: Ingo Molnar @ 2013-12-17 13:52 UTC (permalink / raw)
  To: Linus Torvalds
  Cc: linux-kernel, Thomas Gleixner, Peter Zijlstra, H. Peter Anvin,
	Andrew Morton

Linus,

Please pull the latest timers-urgent-for-linus git tree from:

   git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git timers-urgent-for-linus

   # HEAD: 5d5119a476f77b57381ddd66d4f7034a5447480b Merge branch 'clockevents/fixes' of git://git.linaro.org/people/daniel.lezcano/linux into timers/urgent

Five self-contained fixlets:

 - fix clocksource driver build bug
 - fix two sched_clock() bugs triggering on specific hardware
 - fix devicetree enumeration bug affecting specific hardware
 - fix irq handler registration race resulting in boot crash
 - fix device node refcount bug

 Thanks,

	Ingo

------------------>
Axel Lin (1):
      clocksource: time-efm32: Select CLKSRC_MMIO

Dinh Nguyen (2):
      clocksource: dw_apb_timer_of: Fix read_sched_clock
      clocksource: dw_apb_timer_of: Fix support for dts binding "snps,dw-apb-timer"

Ezequiel Garcia (1):
      clocksource: armada-370-xp: Register sched_clock after the counter reset

Marc Zyngier (1):
      clocksource: sunxi: Stop timer from ticking before enabling interrupts

Thierry Reding (1):
      clocksource: clksrc-of: Do not drop unheld reference on device node


 drivers/clocksource/Kconfig              |  1 +
 drivers/clocksource/clksrc-of.c          |  1 -
 drivers/clocksource/dw_apb_timer_of.c    |  7 ++++---
 drivers/clocksource/sun4i_timer.c        |  3 +++
 drivers/clocksource/time-armada-370-xp.c | 10 +++++-----
 5 files changed, 13 insertions(+), 9 deletions(-)

diff --git a/drivers/clocksource/Kconfig b/drivers/clocksource/Kconfig
index 5c07a56..634c4d6 100644
--- a/drivers/clocksource/Kconfig
+++ b/drivers/clocksource/Kconfig
@@ -75,6 +75,7 @@ config CLKSRC_DBX500_PRCMU_SCHED_CLOCK
 config CLKSRC_EFM32
 	bool "Clocksource for Energy Micro's EFM32 SoCs" if !ARCH_EFM32
 	depends on OF && ARM && (ARCH_EFM32 || COMPILE_TEST)
+	select CLKSRC_MMIO
 	default ARCH_EFM32
 	help
 	  Support to use the timers of EFM32 SoCs as clock source and clock
diff --git a/drivers/clocksource/clksrc-of.c b/drivers/clocksource/clksrc-of.c
index 35639cf4..b9ddd9e 100644
--- a/drivers/clocksource/clksrc-of.c
+++ b/drivers/clocksource/clksrc-of.c
@@ -35,6 +35,5 @@ void __init clocksource_of_init(void)
 
 		init_func = match->data;
 		init_func(np);
-		of_node_put(np);
 	}
 }
diff --git a/drivers/clocksource/dw_apb_timer_of.c b/drivers/clocksource/dw_apb_timer_of.c
index 45ba8ae..2a2ea27 100644
--- a/drivers/clocksource/dw_apb_timer_of.c
+++ b/drivers/clocksource/dw_apb_timer_of.c
@@ -108,12 +108,11 @@ static void __init add_clocksource(struct device_node *source_timer)
 
 static u64 read_sched_clock(void)
 {
-	return __raw_readl(sched_io_base);
+	return ~__raw_readl(sched_io_base);
 }
 
 static const struct of_device_id sptimer_ids[] __initconst = {
 	{ .compatible = "picochip,pc3x2-rtc" },
-	{ .compatible = "snps,dw-apb-timer-sp" },
 	{ /* Sentinel */ },
 };
 
@@ -151,4 +150,6 @@ static void __init dw_apb_timer_init(struct device_node *timer)
 	num_called++;
 }
 CLOCKSOURCE_OF_DECLARE(pc3x2_timer, "picochip,pc3x2-timer", dw_apb_timer_init);
-CLOCKSOURCE_OF_DECLARE(apb_timer, "snps,dw-apb-timer-osc", dw_apb_timer_init);
+CLOCKSOURCE_OF_DECLARE(apb_timer_osc, "snps,dw-apb-timer-osc", dw_apb_timer_init);
+CLOCKSOURCE_OF_DECLARE(apb_timer_sp, "snps,dw-apb-timer-sp", dw_apb_timer_init);
+CLOCKSOURCE_OF_DECLARE(apb_timer, "snps,dw-apb-timer", dw_apb_timer_init);
diff --git a/drivers/clocksource/sun4i_timer.c b/drivers/clocksource/sun4i_timer.c
index 2fb4695..a4f6119 100644
--- a/drivers/clocksource/sun4i_timer.c
+++ b/drivers/clocksource/sun4i_timer.c
@@ -179,6 +179,9 @@ static void __init sun4i_timer_init(struct device_node *node)
 	writel(TIMER_CTL_CLK_SRC(TIMER_CTL_CLK_SRC_OSC24M),
 	       timer_base + TIMER_CTL_REG(0));
 
+	/* Make sure timer is stopped before playing with interrupts */
+	sun4i_clkevt_time_stop(0);
+
 	ret = setup_irq(irq, &sun4i_timer_irq);
 	if (ret)
 		pr_warn("failed to setup irq %d\n", irq);
diff --git a/drivers/clocksource/time-armada-370-xp.c b/drivers/clocksource/time-armada-370-xp.c
index d8e47e5..4e7f680 100644
--- a/drivers/clocksource/time-armada-370-xp.c
+++ b/drivers/clocksource/time-armada-370-xp.c
@@ -256,11 +256,6 @@ static void __init armada_370_xp_timer_common_init(struct device_node *np)
 	ticks_per_jiffy = (timer_clk + HZ / 2) / HZ;
 
 	/*
-	 * Set scale and timer for sched_clock.
-	 */
-	sched_clock_register(armada_370_xp_read_sched_clock, 32, timer_clk);
-
-	/*
 	 * Setup free-running clocksource timer (interrupts
 	 * disabled).
 	 */
@@ -270,6 +265,11 @@ static void __init armada_370_xp_timer_common_init(struct device_node *np)
 	timer_ctrl_clrset(0, TIMER0_EN | TIMER0_RELOAD_EN |
 			     TIMER0_DIV(TIMER_DIVIDER_SHIFT));
 
+	/*
+	 * Set scale and timer for sched_clock.
+	 */
+	sched_clock_register(armada_370_xp_read_sched_clock, 32, timer_clk);
+
 	clocksource_mmio_init(timer_base + TIMER0_VAL_OFF,
 			      "armada_370_xp_clocksource",
 			      timer_clk, 300, 32, clocksource_mmio_readl_down);

^ permalink raw reply related	[flat|nested] 69+ messages in thread

* [GIT PULL] timer fixes
@ 2013-09-28 18:19 Ingo Molnar
  0 siblings, 0 replies; 69+ messages in thread
From: Ingo Molnar @ 2013-09-28 18:19 UTC (permalink / raw)
  To: Linus Torvalds
  Cc: linux-kernel, Thomas Gleixner, Peter Zijlstra, Andrew Morton,
	Daniel Lezcano

Linus,

Please pull the latest timers-urgent-for-linus git tree from:

   git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git timers-urgent-for-linus

   HEAD: 2199a5574b6d94b9ca26c6345356f45ec60fef8b clocksource: em_sti: Set cpu_possible_mask to fix SMP broadcast

A handful of ARM clocksource/clockevent driver fixes.

Thanks,

	Ingo

------------------>
Jean Pihet (1):
      arm: clocksource: mvebu: Use the main timer as clock source from DT

Magnus Damm (1):
      clocksource: em_sti: Set cpu_possible_mask to fix SMP broadcast

Sebastian Hesselbarth (1):
      clocksource: of: Respect device tree node status

Tomasz Figa (1):
      clocksource: exynos_mct: Set IRQ affinity when the CPU goes online


 drivers/clocksource/Kconfig      |  1 +
 drivers/clocksource/clksrc-of.c  |  3 +++
 drivers/clocksource/em_sti.c     |  2 +-
 drivers/clocksource/exynos_mct.c | 10 +++++++++-
 4 files changed, 14 insertions(+), 2 deletions(-)

diff --git a/drivers/clocksource/Kconfig b/drivers/clocksource/Kconfig
index 41c6946..971d796 100644
--- a/drivers/clocksource/Kconfig
+++ b/drivers/clocksource/Kconfig
@@ -26,6 +26,7 @@ config DW_APB_TIMER_OF
 
 config ARMADA_370_XP_TIMER
 	bool
+	select CLKSRC_OF
 
 config ORION_TIMER
 	select CLKSRC_OF
diff --git a/drivers/clocksource/clksrc-of.c b/drivers/clocksource/clksrc-of.c
index 37f5325..b9ddd9e 100644
--- a/drivers/clocksource/clksrc-of.c
+++ b/drivers/clocksource/clksrc-of.c
@@ -30,6 +30,9 @@ void __init clocksource_of_init(void)
 	clocksource_of_init_fn init_func;
 
 	for_each_matching_node_and_match(np, __clksrc_of_table, &match) {
+		if (!of_device_is_available(np))
+			continue;
+
 		init_func = match->data;
 		init_func(np);
 	}
diff --git a/drivers/clocksource/em_sti.c b/drivers/clocksource/em_sti.c
index b9c81b7..3a5909c 100644
--- a/drivers/clocksource/em_sti.c
+++ b/drivers/clocksource/em_sti.c
@@ -301,7 +301,7 @@ static void em_sti_register_clockevent(struct em_sti_priv *p)
 	ced->name = dev_name(&p->pdev->dev);
 	ced->features = CLOCK_EVT_FEAT_ONESHOT;
 	ced->rating = 200;
-	ced->cpumask = cpumask_of(0);
+	ced->cpumask = cpu_possible_mask;
 	ced->set_next_event = em_sti_clock_event_next;
 	ced->set_mode = em_sti_clock_event_mode;
 
diff --git a/drivers/clocksource/exynos_mct.c b/drivers/clocksource/exynos_mct.c
index 5b34768..62b0de6 100644
--- a/drivers/clocksource/exynos_mct.c
+++ b/drivers/clocksource/exynos_mct.c
@@ -428,7 +428,6 @@ static int exynos4_local_timer_setup(struct clock_event_device *evt)
 				evt->irq);
 			return -EIO;
 		}
-		irq_set_affinity(evt->irq, cpumask_of(cpu));
 	} else {
 		enable_percpu_irq(mct_irqs[MCT_L0_IRQ], 0);
 	}
@@ -449,6 +448,7 @@ static int exynos4_mct_cpu_notify(struct notifier_block *self,
 					   unsigned long action, void *hcpu)
 {
 	struct mct_clock_event_device *mevt;
+	unsigned int cpu;
 
 	/*
 	 * Grab cpu pointer in each case to avoid spurious
@@ -459,6 +459,12 @@ static int exynos4_mct_cpu_notify(struct notifier_block *self,
 		mevt = this_cpu_ptr(&percpu_mct_tick);
 		exynos4_local_timer_setup(&mevt->evt);
 		break;
+	case CPU_ONLINE:
+		cpu = (unsigned long)hcpu;
+		if (mct_int_type == MCT_INT_SPI)
+			irq_set_affinity(mct_irqs[MCT_L0_IRQ + cpu],
+						cpumask_of(cpu));
+		break;
 	case CPU_DYING:
 		mevt = this_cpu_ptr(&percpu_mct_tick);
 		exynos4_local_timer_stop(&mevt->evt);
@@ -500,6 +506,8 @@ static void __init exynos4_timer_resources(struct device_node *np, void __iomem
 					 &percpu_mct_tick);
 		WARN(err, "MCT: can't request IRQ %d (%d)\n",
 		     mct_irqs[MCT_L0_IRQ], err);
+	} else {
+		irq_set_affinity(mct_irqs[MCT_L0_IRQ], cpumask_of(0));
 	}
 
 	err = register_cpu_notifier(&exynos4_mct_cpu_nb);

^ permalink raw reply related	[flat|nested] 69+ messages in thread

* [GIT PULL] timer fixes
@ 2013-08-19  9:41 Ingo Molnar
  0 siblings, 0 replies; 69+ messages in thread
From: Ingo Molnar @ 2013-08-19  9:41 UTC (permalink / raw)
  To: Linus Torvalds
  Cc: linux-kernel, Thomas Gleixner, Peter Zijlstra, Andrew Morton

Linus,

Please pull the latest timers-urgent-for-linus git tree from:

   git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git timers-urgent-for-linus

   HEAD: ae920eb24277e4a174a3ca575ce42b98b18a6748 Merge branch 'fortglx/3.11/time' of git://git.linaro.org/people/jstultz/linux into timers/urgent

Three small fixlets.

 Thanks,

	Ingo

------------------>
Baruch Siach (1):
      sched_clock: Fix integer overflow

Li Zhong (1):
      nohz: fix compile warning in tick_nohz_init()

Steven Rostedt (1):
      nohz: Do not warn about unstable tsc unless user uses nohz_full


 kernel/time/sched_clock.c | 2 +-
 kernel/time/tick-sched.c  | 5 ++---
 2 files changed, 3 insertions(+), 4 deletions(-)

diff --git a/kernel/time/sched_clock.c b/kernel/time/sched_clock.c
index a326f27..0b479a6 100644
--- a/kernel/time/sched_clock.c
+++ b/kernel/time/sched_clock.c
@@ -121,7 +121,7 @@ void __init setup_sched_clock(u32 (*read)(void), int bits, unsigned long rate)
 	BUG_ON(bits > 32);
 	WARN_ON(!irqs_disabled());
 	read_sched_clock = read;
-	sched_clock_mask = (1 << bits) - 1;
+	sched_clock_mask = (1ULL << bits) - 1;
 	cd.rate = rate;
 
 	/* calculate the mult/shift to convert counter ticks to ns. */
diff --git a/kernel/time/tick-sched.c b/kernel/time/tick-sched.c
index e80183f..9563c74 100644
--- a/kernel/time/tick-sched.c
+++ b/kernel/time/tick-sched.c
@@ -182,7 +182,8 @@ static bool can_stop_full_tick(void)
 		 * Don't allow the user to think they can get
 		 * full NO_HZ with this machine.
 		 */
-		WARN_ONCE(1, "NO_HZ FULL will not work with unstable sched clock");
+		WARN_ONCE(have_nohz_full_mask,
+			  "NO_HZ FULL will not work with unstable sched clock");
 		return false;
 	}
 #endif
@@ -343,8 +344,6 @@ static int tick_nohz_init_all(void)
 
 void __init tick_nohz_init(void)
 {
-	int cpu;
-
 	if (!have_nohz_full_mask) {
 		if (tick_nohz_init_all() < 0)
 			return;

^ permalink raw reply related	[flat|nested] 69+ messages in thread

* [GIT PULL] timer fixes
@ 2013-02-26 11:32 Ingo Molnar
  0 siblings, 0 replies; 69+ messages in thread
From: Ingo Molnar @ 2013-02-26 11:32 UTC (permalink / raw)
  To: Linus Torvalds
  Cc: linux-kernel, Thomas Gleixner, Peter Zijlstra, Andrew Morton

Linus,

Please pull the latest timers-urgent-for-linus git tree from:

   git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git timers-urgent-for-linus

   HEAD: 00f4e13c4634b10f6a16b26a980b22a53dfa6bb5 clocksource : Nomadik-mtu : fix missing irq initialization

 Thanks,

	Ingo

------------------>
Daniel Lezcano (1):
      clocksource : Nomadik-mtu : fix missing irq initialization

Tejun Heo (1):
      posix-timer: Don't call idr_find() with out-of-range ID


 drivers/clocksource/nomadik-mtu.c | 1 +
 kernel/posix-timers.c             | 7 +++++++
 2 files changed, 8 insertions(+)

diff --git a/drivers/clocksource/nomadik-mtu.c b/drivers/clocksource/nomadik-mtu.c
index 8914c3c..7cbcaa0 100644
--- a/drivers/clocksource/nomadik-mtu.c
+++ b/drivers/clocksource/nomadik-mtu.c
@@ -226,5 +226,6 @@ void __init nmdk_timer_init(void __iomem *base, int irq)
 	/* Timer 1 is used for events, register irq and clockevents */
 	setup_irq(irq, &nmdk_timer_irq);
 	nmdk_clkevt.cpumask = cpumask_of(0);
+	nmdk_clkevt.irq = irq;
 	clockevents_config_and_register(&nmdk_clkevt, rate, 2, 0xffffffffU);
 }
diff --git a/kernel/posix-timers.c b/kernel/posix-timers.c
index 10349d5..7edfe4b 100644
--- a/kernel/posix-timers.c
+++ b/kernel/posix-timers.c
@@ -639,6 +639,13 @@ static struct k_itimer *__lock_timer(timer_t timer_id, unsigned long *flags)
 {
 	struct k_itimer *timr;
 
+	/*
+	 * timer_t could be any type >= int and we want to make sure any
+	 * @timer_id outside positive int range fails lookup.
+	 */
+	if ((unsigned long long)timer_id > INT_MAX)
+		return NULL;
+
 	rcu_read_lock();
 	timr = idr_find(&posix_timers_id, (int)timer_id);
 	if (timr) {

^ permalink raw reply related	[flat|nested] 69+ messages in thread

* [GIT PULL] timer fixes
@ 2012-09-21 19:12 Ingo Molnar
  0 siblings, 0 replies; 69+ messages in thread
From: Ingo Molnar @ 2012-09-21 19:12 UTC (permalink / raw)
  To: Linus Torvalds
  Cc: linux-kernel, Thomas Gleixner, Peter Zijlstra, Andrew Morton

Linus,

Please pull the latest timers-urgent-for-linus git tree from:

   git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git timers-urgent-for-linus

   HEAD: ec145babe754f9ea1079034a108104b6001e001c time: Fix timeekeping_get_ns overflow on 32bit systems

One more timekeeping fix for v3.6.

 Thanks,

	Ingo

------------------>
John Stultz (1):
      time: Fix timeekeping_get_ns overflow on 32bit systems


 kernel/time/timekeeping.c | 19 ++++++++++++-------
 1 file changed, 12 insertions(+), 7 deletions(-)

diff --git a/kernel/time/timekeeping.c b/kernel/time/timekeeping.c
index 34e5eac..d3b91e7 100644
--- a/kernel/time/timekeeping.c
+++ b/kernel/time/timekeeping.c
@@ -303,10 +303,11 @@ void getnstimeofday(struct timespec *ts)
 		seq = read_seqbegin(&tk->lock);
 
 		ts->tv_sec = tk->xtime_sec;
-		ts->tv_nsec = timekeeping_get_ns(tk);
+		nsecs = timekeeping_get_ns(tk);
 
 	} while (read_seqretry(&tk->lock, seq));
 
+	ts->tv_nsec = 0;
 	timespec_add_ns(ts, nsecs);
 }
 EXPORT_SYMBOL(getnstimeofday);
@@ -345,6 +346,7 @@ void ktime_get_ts(struct timespec *ts)
 {
 	struct timekeeper *tk = &timekeeper;
 	struct timespec tomono;
+	s64 nsec;
 	unsigned int seq;
 
 	WARN_ON(timekeeping_suspended);
@@ -352,13 +354,14 @@ void ktime_get_ts(struct timespec *ts)
 	do {
 		seq = read_seqbegin(&tk->lock);
 		ts->tv_sec = tk->xtime_sec;
-		ts->tv_nsec = timekeeping_get_ns(tk);
+		nsec = timekeeping_get_ns(tk);
 		tomono = tk->wall_to_monotonic;
 
 	} while (read_seqretry(&tk->lock, seq));
 
-	set_normalized_timespec(ts, ts->tv_sec + tomono.tv_sec,
-				ts->tv_nsec + tomono.tv_nsec);
+	ts->tv_sec += tomono.tv_sec;
+	ts->tv_nsec = 0;
+	timespec_add_ns(ts, nsec + tomono.tv_nsec);
 }
 EXPORT_SYMBOL_GPL(ktime_get_ts);
 
@@ -1244,6 +1247,7 @@ void get_monotonic_boottime(struct timespec *ts)
 {
 	struct timekeeper *tk = &timekeeper;
 	struct timespec tomono, sleep;
+	s64 nsec;
 	unsigned int seq;
 
 	WARN_ON(timekeeping_suspended);
@@ -1251,14 +1255,15 @@ void get_monotonic_boottime(struct timespec *ts)
 	do {
 		seq = read_seqbegin(&tk->lock);
 		ts->tv_sec = tk->xtime_sec;
-		ts->tv_nsec = timekeeping_get_ns(tk);
+		nsec = timekeeping_get_ns(tk);
 		tomono = tk->wall_to_monotonic;
 		sleep = tk->total_sleep_time;
 
 	} while (read_seqretry(&tk->lock, seq));
 
-	set_normalized_timespec(ts, ts->tv_sec + tomono.tv_sec + sleep.tv_sec,
-			ts->tv_nsec + tomono.tv_nsec + sleep.tv_nsec);
+	ts->tv_sec += tomono.tv_sec + sleep.tv_sec;
+	ts->tv_nsec = 0;
+	timespec_add_ns(ts, nsec + tomono.tv_nsec + sleep.tv_nsec);
 }
 EXPORT_SYMBOL_GPL(get_monotonic_boottime);
 


^ permalink raw reply related	[flat|nested] 69+ messages in thread

* [GIT PULL] timer fixes
@ 2012-08-03 16:47 Ingo Molnar
  0 siblings, 0 replies; 69+ messages in thread
From: Ingo Molnar @ 2012-08-03 16:47 UTC (permalink / raw)
  To: Linus Torvalds
  Cc: linux-kernel, Thomas Gleixner, Peter Zijlstra, Andrew Morton

Linus,

Please pull the latest timers-urgent-for-linus git tree from:

   git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git timers-urgent-for-linus

   HEAD: 4e250fdde9be50581c7dd5fed88c9b9960615314 time: Remove all direct references to timekeeper

One regression fix, and a couple of cleanups that clean up the 
code flow in areas that had high-profile bugs recently.

 Thanks,

	Ingo

------------------>
Catalin Marinas (1):
      time/jiffies: Allow CLOCK_TICK_RATE to be undefined

John Stultz (5):
      time: Fix casting issue in tk_set_xtime and tk_xtime_add
      time/jiffies: Rename ACTHZ to SHIFTED_HZ
      time: Clean up stray newlines
      time: Clean up offs_real/wall_to_mono and offs_boot/total_sleep_time updates
      time: Remove all direct references to timekeeper


 include/linux/jiffies.h   |   29 ++--
 include/linux/timex.h     |    2 +-
 kernel/time/jiffies.c     |    2 +-
 kernel/time/ntp.c         |    2 +-
 kernel/time/timekeeping.c |  376 ++++++++++++++++++++++++---------------------
 5 files changed, 227 insertions(+), 184 deletions(-)

diff --git a/include/linux/jiffies.h b/include/linux/jiffies.h
index 265e2c3..8268054 100644
--- a/include/linux/jiffies.h
+++ b/include/linux/jiffies.h
@@ -39,9 +39,6 @@
 # error Invalid value of HZ.
 #endif
 
-/* LATCH is used in the interval timer and ftape setup. */
-#define LATCH  ((CLOCK_TICK_RATE + HZ/2) / HZ)	/* For divider */
-
 /* Suppose we want to divide two numbers NOM and DEN: NOM/DEN, then we can
  * improve accuracy by shifting LSH bits, hence calculating:
  *     (NOM << LSH) / DEN
@@ -54,18 +51,30 @@
 #define SH_DIV(NOM,DEN,LSH) (   (((NOM) / (DEN)) << (LSH))              \
                              + ((((NOM) % (DEN)) << (LSH)) + (DEN) / 2) / (DEN))
 
-/* HZ is the requested value. ACTHZ is actual HZ ("<< 8" is for accuracy) */
-#define ACTHZ (SH_DIV (CLOCK_TICK_RATE, LATCH, 8))
+#ifdef CLOCK_TICK_RATE
+/* LATCH is used in the interval timer and ftape setup. */
+# define LATCH ((CLOCK_TICK_RATE + HZ/2) / HZ)	/* For divider */
+
+/*
+ * HZ is the requested value. However the CLOCK_TICK_RATE may not allow
+ * for exactly HZ. So SHIFTED_HZ is high res HZ ("<< 8" is for accuracy)
+ */
+# define SHIFTED_HZ (SH_DIV(CLOCK_TICK_RATE, LATCH, 8))
+#else
+# define SHIFTED_HZ (HZ << 8)
+#endif
 
-/* TICK_NSEC is the time between ticks in nsec assuming real ACTHZ */
-#define TICK_NSEC (SH_DIV (1000000UL * 1000, ACTHZ, 8))
+/* TICK_NSEC is the time between ticks in nsec assuming SHIFTED_HZ */
+#define TICK_NSEC (SH_DIV(1000000UL * 1000, SHIFTED_HZ, 8))
 
 /* TICK_USEC is the time between ticks in usec assuming fake USER_HZ */
 #define TICK_USEC ((1000000UL + USER_HZ/2) / USER_HZ)
 
-/* TICK_USEC_TO_NSEC is the time between ticks in nsec assuming real ACTHZ and	*/
-/* a value TUSEC for TICK_USEC (can be set bij adjtimex)		*/
-#define TICK_USEC_TO_NSEC(TUSEC) (SH_DIV (TUSEC * USER_HZ * 1000, ACTHZ, 8))
+/*
+ * TICK_USEC_TO_NSEC is the time between ticks in nsec assuming SHIFTED_HZ and
+ * a value TUSEC for TICK_USEC (can be set bij adjtimex)
+ */
+#define TICK_USEC_TO_NSEC(TUSEC) (SH_DIV(TUSEC * USER_HZ * 1000, SHIFTED_HZ, 8))
 
 /* some arch's have a small-data section that can be accessed register-relative
  * but that can only take up to, say, 4-byte variables. jiffies being part of
diff --git a/include/linux/timex.h b/include/linux/timex.h
index 99bc88b..7c5ceb2 100644
--- a/include/linux/timex.h
+++ b/include/linux/timex.h
@@ -232,7 +232,7 @@ struct timex {
  * estimated error = NTP dispersion.
  */
 extern unsigned long tick_usec;		/* USER_HZ period (usec) */
-extern unsigned long tick_nsec;		/* ACTHZ          period (nsec) */
+extern unsigned long tick_nsec;		/* SHIFTED_HZ period (nsec) */
 
 extern void ntp_init(void);
 extern void ntp_clear(void);
diff --git a/kernel/time/jiffies.c b/kernel/time/jiffies.c
index a470154..46da053 100644
--- a/kernel/time/jiffies.c
+++ b/kernel/time/jiffies.c
@@ -37,7 +37,7 @@
  * requested HZ value. It is also not recommended
  * for "tick-less" systems.
  */
-#define NSEC_PER_JIFFY	((u32)((((u64)NSEC_PER_SEC)<<8)/ACTHZ))
+#define NSEC_PER_JIFFY	((u32)((((u64)NSEC_PER_SEC)<<8)/SHIFTED_HZ))
 
 /* Since jiffies uses a simple NSEC_PER_JIFFY multiplier
  * conversion, the .shift value could be zero. However
diff --git a/kernel/time/ntp.c b/kernel/time/ntp.c
index b7fbadc..24174b4 100644
--- a/kernel/time/ntp.c
+++ b/kernel/time/ntp.c
@@ -28,7 +28,7 @@ DEFINE_SPINLOCK(ntp_lock);
 /* USER_HZ period (usecs): */
 unsigned long			tick_usec = TICK_USEC;
 
-/* ACTHZ period (nsecs): */
+/* SHIFTED_HZ period (nsecs): */
 unsigned long			tick_nsec;
 
 static u64			tick_length;
diff --git a/kernel/time/timekeeping.c b/kernel/time/timekeeping.c
index f045cc5..2988bc8 100644
--- a/kernel/time/timekeeping.c
+++ b/kernel/time/timekeeping.c
@@ -65,14 +65,14 @@ struct timekeeper {
 	 * used instead.
 	 */
 	struct timespec		wall_to_monotonic;
-	/* time spent in suspend */
-	struct timespec		total_sleep_time;
-	/* The raw monotonic time for the CLOCK_MONOTONIC_RAW posix clock. */
-	struct timespec		raw_time;
 	/* Offset clock monotonic -> clock realtime */
 	ktime_t			offs_real;
+	/* time spent in suspend */
+	struct timespec		total_sleep_time;
 	/* Offset clock monotonic -> clock boottime */
 	ktime_t			offs_boot;
+	/* The raw monotonic time for the CLOCK_MONOTONIC_RAW posix clock. */
+	struct timespec		raw_time;
 	/* Seqlock for all timekeeper values */
 	seqlock_t		lock;
 };
@@ -108,13 +108,38 @@ static struct timespec tk_xtime(struct timekeeper *tk)
 static void tk_set_xtime(struct timekeeper *tk, const struct timespec *ts)
 {
 	tk->xtime_sec = ts->tv_sec;
-	tk->xtime_nsec = ts->tv_nsec << tk->shift;
+	tk->xtime_nsec = (u64)ts->tv_nsec << tk->shift;
 }
 
 static void tk_xtime_add(struct timekeeper *tk, const struct timespec *ts)
 {
 	tk->xtime_sec += ts->tv_sec;
-	tk->xtime_nsec += ts->tv_nsec << tk->shift;
+	tk->xtime_nsec += (u64)ts->tv_nsec << tk->shift;
+}
+
+static void tk_set_wall_to_mono(struct timekeeper *tk, struct timespec wtm)
+{
+	struct timespec tmp;
+
+	/*
+	 * Verify consistency of: offset_real = -wall_to_monotonic
+	 * before modifying anything
+	 */
+	set_normalized_timespec(&tmp, -tk->wall_to_monotonic.tv_sec,
+					-tk->wall_to_monotonic.tv_nsec);
+	WARN_ON_ONCE(tk->offs_real.tv64 != timespec_to_ktime(tmp).tv64);
+	tk->wall_to_monotonic = wtm;
+	set_normalized_timespec(&tmp, -wtm.tv_sec, -wtm.tv_nsec);
+	tk->offs_real = timespec_to_ktime(tmp);
+}
+
+static void tk_set_sleep_time(struct timekeeper *tk, struct timespec t)
+{
+	/* Verify consistency before modifying */
+	WARN_ON_ONCE(tk->offs_boot.tv64 != timespec_to_ktime(tk->total_sleep_time).tv64);
+
+	tk->total_sleep_time	= t;
+	tk->offs_boot		= timespec_to_ktime(t);
 }
 
 /**
@@ -217,14 +242,6 @@ static inline s64 timekeeping_get_ns_raw(struct timekeeper *tk)
 	return nsec + arch_gettimeoffset();
 }
 
-static void update_rt_offset(struct timekeeper *tk)
-{
-	struct timespec tmp, *wtm = &tk->wall_to_monotonic;
-
-	set_normalized_timespec(&tmp, -wtm->tv_sec, -wtm->tv_nsec);
-	tk->offs_real = timespec_to_ktime(tmp);
-}
-
 /* must hold write on timekeeper.lock */
 static void timekeeping_update(struct timekeeper *tk, bool clearntp)
 {
@@ -234,12 +251,10 @@ static void timekeeping_update(struct timekeeper *tk, bool clearntp)
 		tk->ntp_error = 0;
 		ntp_clear();
 	}
-	update_rt_offset(tk);
 	xt = tk_xtime(tk);
 	update_vsyscall(&xt, &tk->wall_to_monotonic, tk->clock, tk->mult);
 }
 
-
 /**
  * timekeeping_forward_now - update clock to the current time
  *
@@ -277,18 +292,19 @@ static void timekeeping_forward_now(struct timekeeper *tk)
  */
 void getnstimeofday(struct timespec *ts)
 {
+	struct timekeeper *tk = &timekeeper;
 	unsigned long seq;
 	s64 nsecs = 0;
 
 	WARN_ON(timekeeping_suspended);
 
 	do {
-		seq = read_seqbegin(&timekeeper.lock);
+		seq = read_seqbegin(&tk->lock);
 
-		ts->tv_sec = timekeeper.xtime_sec;
-		ts->tv_nsec = timekeeping_get_ns(&timekeeper);
+		ts->tv_sec = tk->xtime_sec;
+		ts->tv_nsec = timekeeping_get_ns(tk);
 
-	} while (read_seqretry(&timekeeper.lock, seq));
+	} while (read_seqretry(&tk->lock, seq));
 
 	timespec_add_ns(ts, nsecs);
 }
@@ -296,19 +312,18 @@ EXPORT_SYMBOL(getnstimeofday);
 
 ktime_t ktime_get(void)
 {
+	struct timekeeper *tk = &timekeeper;
 	unsigned int seq;
 	s64 secs, nsecs;
 
 	WARN_ON(timekeeping_suspended);
 
 	do {
-		seq = read_seqbegin(&timekeeper.lock);
-		secs = timekeeper.xtime_sec +
-				timekeeper.wall_to_monotonic.tv_sec;
-		nsecs = timekeeping_get_ns(&timekeeper) +
-				timekeeper.wall_to_monotonic.tv_nsec;
+		seq = read_seqbegin(&tk->lock);
+		secs = tk->xtime_sec + tk->wall_to_monotonic.tv_sec;
+		nsecs = timekeeping_get_ns(tk) + tk->wall_to_monotonic.tv_nsec;
 
-	} while (read_seqretry(&timekeeper.lock, seq));
+	} while (read_seqretry(&tk->lock, seq));
 	/*
 	 * Use ktime_set/ktime_add_ns to create a proper ktime on
 	 * 32-bit architectures without CONFIG_KTIME_SCALAR.
@@ -327,18 +342,19 @@ EXPORT_SYMBOL_GPL(ktime_get);
  */
 void ktime_get_ts(struct timespec *ts)
 {
+	struct timekeeper *tk = &timekeeper;
 	struct timespec tomono;
 	unsigned int seq;
 
 	WARN_ON(timekeeping_suspended);
 
 	do {
-		seq = read_seqbegin(&timekeeper.lock);
-		ts->tv_sec = timekeeper.xtime_sec;
-		ts->tv_nsec = timekeeping_get_ns(&timekeeper);
-		tomono = timekeeper.wall_to_monotonic;
+		seq = read_seqbegin(&tk->lock);
+		ts->tv_sec = tk->xtime_sec;
+		ts->tv_nsec = timekeeping_get_ns(tk);
+		tomono = tk->wall_to_monotonic;
 
-	} while (read_seqretry(&timekeeper.lock, seq));
+	} while (read_seqretry(&tk->lock, seq));
 
 	set_normalized_timespec(ts, ts->tv_sec + tomono.tv_sec,
 				ts->tv_nsec + tomono.tv_nsec);
@@ -358,22 +374,23 @@ EXPORT_SYMBOL_GPL(ktime_get_ts);
  */
 void getnstime_raw_and_real(struct timespec *ts_raw, struct timespec *ts_real)
 {
+	struct timekeeper *tk = &timekeeper;
 	unsigned long seq;
 	s64 nsecs_raw, nsecs_real;
 
 	WARN_ON_ONCE(timekeeping_suspended);
 
 	do {
-		seq = read_seqbegin(&timekeeper.lock);
+		seq = read_seqbegin(&tk->lock);
 
-		*ts_raw = timekeeper.raw_time;
-		ts_real->tv_sec = timekeeper.xtime_sec;
+		*ts_raw = tk->raw_time;
+		ts_real->tv_sec = tk->xtime_sec;
 		ts_real->tv_nsec = 0;
 
-		nsecs_raw = timekeeping_get_ns_raw(&timekeeper);
-		nsecs_real = timekeeping_get_ns(&timekeeper);
+		nsecs_raw = timekeeping_get_ns_raw(tk);
+		nsecs_real = timekeeping_get_ns(tk);
 
-	} while (read_seqretry(&timekeeper.lock, seq));
+	} while (read_seqretry(&tk->lock, seq));
 
 	timespec_add_ns(ts_raw, nsecs_raw);
 	timespec_add_ns(ts_real, nsecs_real);
@@ -406,28 +423,28 @@ EXPORT_SYMBOL(do_gettimeofday);
  */
 int do_settimeofday(const struct timespec *tv)
 {
+	struct timekeeper *tk = &timekeeper;
 	struct timespec ts_delta, xt;
 	unsigned long flags;
 
 	if ((unsigned long)tv->tv_nsec >= NSEC_PER_SEC)
 		return -EINVAL;
 
-	write_seqlock_irqsave(&timekeeper.lock, flags);
+	write_seqlock_irqsave(&tk->lock, flags);
 
-	timekeeping_forward_now(&timekeeper);
+	timekeeping_forward_now(tk);
 
-	xt = tk_xtime(&timekeeper);
+	xt = tk_xtime(tk);
 	ts_delta.tv_sec = tv->tv_sec - xt.tv_sec;
 	ts_delta.tv_nsec = tv->tv_nsec - xt.tv_nsec;
 
-	timekeeper.wall_to_monotonic =
-			timespec_sub(timekeeper.wall_to_monotonic, ts_delta);
+	tk_set_wall_to_mono(tk, timespec_sub(tk->wall_to_monotonic, ts_delta));
 
-	tk_set_xtime(&timekeeper, tv);
+	tk_set_xtime(tk, tv);
 
-	timekeeping_update(&timekeeper, true);
+	timekeeping_update(tk, true);
 
-	write_sequnlock_irqrestore(&timekeeper.lock, flags);
+	write_sequnlock_irqrestore(&tk->lock, flags);
 
 	/* signal hrtimers about time change */
 	clock_was_set();
@@ -436,7 +453,6 @@ int do_settimeofday(const struct timespec *tv)
 }
 EXPORT_SYMBOL(do_settimeofday);
 
-
 /**
  * timekeeping_inject_offset - Adds or subtracts from the current time.
  * @tv:		pointer to the timespec variable containing the offset
@@ -445,23 +461,23 @@ EXPORT_SYMBOL(do_settimeofday);
  */
 int timekeeping_inject_offset(struct timespec *ts)
 {
+	struct timekeeper *tk = &timekeeper;
 	unsigned long flags;
 
 	if ((unsigned long)ts->tv_nsec >= NSEC_PER_SEC)
 		return -EINVAL;
 
-	write_seqlock_irqsave(&timekeeper.lock, flags);
+	write_seqlock_irqsave(&tk->lock, flags);
 
-	timekeeping_forward_now(&timekeeper);
+	timekeeping_forward_now(tk);
 
 
-	tk_xtime_add(&timekeeper, ts);
-	timekeeper.wall_to_monotonic =
-				timespec_sub(timekeeper.wall_to_monotonic, *ts);
+	tk_xtime_add(tk, ts);
+	tk_set_wall_to_mono(tk, timespec_sub(tk->wall_to_monotonic, *ts));
 
-	timekeeping_update(&timekeeper, true);
+	timekeeping_update(tk, true);
 
-	write_sequnlock_irqrestore(&timekeeper.lock, flags);
+	write_sequnlock_irqrestore(&tk->lock, flags);
 
 	/* signal hrtimers about time change */
 	clock_was_set();
@@ -477,23 +493,24 @@ EXPORT_SYMBOL(timekeeping_inject_offset);
  */
 static int change_clocksource(void *data)
 {
+	struct timekeeper *tk = &timekeeper;
 	struct clocksource *new, *old;
 	unsigned long flags;
 
 	new = (struct clocksource *) data;
 
-	write_seqlock_irqsave(&timekeeper.lock, flags);
+	write_seqlock_irqsave(&tk->lock, flags);
 
-	timekeeping_forward_now(&timekeeper);
+	timekeeping_forward_now(tk);
 	if (!new->enable || new->enable(new) == 0) {
-		old = timekeeper.clock;
-		tk_setup_internals(&timekeeper, new);
+		old = tk->clock;
+		tk_setup_internals(tk, new);
 		if (old->disable)
 			old->disable(old);
 	}
-	timekeeping_update(&timekeeper, true);
+	timekeeping_update(tk, true);
 
-	write_sequnlock_irqrestore(&timekeeper.lock, flags);
+	write_sequnlock_irqrestore(&tk->lock, flags);
 
 	return 0;
 }
@@ -507,7 +524,9 @@ static int change_clocksource(void *data)
  */
 void timekeeping_notify(struct clocksource *clock)
 {
-	if (timekeeper.clock == clock)
+	struct timekeeper *tk = &timekeeper;
+
+	if (tk->clock == clock)
 		return;
 	stop_machine(change_clocksource, clock, NULL);
 	tick_clock_notify();
@@ -536,35 +555,36 @@ EXPORT_SYMBOL_GPL(ktime_get_real);
  */
 void getrawmonotonic(struct timespec *ts)
 {
+	struct timekeeper *tk = &timekeeper;
 	unsigned long seq;
 	s64 nsecs;
 
 	do {
-		seq = read_seqbegin(&timekeeper.lock);
-		nsecs = timekeeping_get_ns_raw(&timekeeper);
-		*ts = timekeeper.raw_time;
+		seq = read_seqbegin(&tk->lock);
+		nsecs = timekeeping_get_ns_raw(tk);
+		*ts = tk->raw_time;
 
-	} while (read_seqretry(&timekeeper.lock, seq));
+	} while (read_seqretry(&tk->lock, seq));
 
 	timespec_add_ns(ts, nsecs);
 }
 EXPORT_SYMBOL(getrawmonotonic);
 
-
 /**
  * timekeeping_valid_for_hres - Check if timekeeping is suitable for hres
  */
 int timekeeping_valid_for_hres(void)
 {
+	struct timekeeper *tk = &timekeeper;
 	unsigned long seq;
 	int ret;
 
 	do {
-		seq = read_seqbegin(&timekeeper.lock);
+		seq = read_seqbegin(&tk->lock);
 
-		ret = timekeeper.clock->flags & CLOCK_SOURCE_VALID_FOR_HRES;
+		ret = tk->clock->flags & CLOCK_SOURCE_VALID_FOR_HRES;
 
-	} while (read_seqretry(&timekeeper.lock, seq));
+	} while (read_seqretry(&tk->lock, seq));
 
 	return ret;
 }
@@ -574,15 +594,16 @@ int timekeeping_valid_for_hres(void)
  */
 u64 timekeeping_max_deferment(void)
 {
+	struct timekeeper *tk = &timekeeper;
 	unsigned long seq;
 	u64 ret;
 
 	do {
-		seq = read_seqbegin(&timekeeper.lock);
+		seq = read_seqbegin(&tk->lock);
 
-		ret = timekeeper.clock->max_idle_ns;
+		ret = tk->clock->max_idle_ns;
 
-	} while (read_seqretry(&timekeeper.lock, seq));
+	} while (read_seqretry(&tk->lock, seq));
 
 	return ret;
 }
@@ -622,46 +643,43 @@ void __attribute__((weak)) read_boot_clock(struct timespec *ts)
  */
 void __init timekeeping_init(void)
 {
+	struct timekeeper *tk = &timekeeper;
 	struct clocksource *clock;
 	unsigned long flags;
-	struct timespec now, boot;
+	struct timespec now, boot, tmp;
 
 	read_persistent_clock(&now);
 	read_boot_clock(&boot);
 
-	seqlock_init(&timekeeper.lock);
+	seqlock_init(&tk->lock);
 
 	ntp_init();
 
-	write_seqlock_irqsave(&timekeeper.lock, flags);
+	write_seqlock_irqsave(&tk->lock, flags);
 	clock = clocksource_default_clock();
 	if (clock->enable)
 		clock->enable(clock);
-	tk_setup_internals(&timekeeper, clock);
+	tk_setup_internals(tk, clock);
 
-	tk_set_xtime(&timekeeper, &now);
-	timekeeper.raw_time.tv_sec = 0;
-	timekeeper.raw_time.tv_nsec = 0;
+	tk_set_xtime(tk, &now);
+	tk->raw_time.tv_sec = 0;
+	tk->raw_time.tv_nsec = 0;
 	if (boot.tv_sec == 0 && boot.tv_nsec == 0)
-		boot = tk_xtime(&timekeeper);
-
-	set_normalized_timespec(&timekeeper.wall_to_monotonic,
-				-boot.tv_sec, -boot.tv_nsec);
-	update_rt_offset(&timekeeper);
-	timekeeper.total_sleep_time.tv_sec = 0;
-	timekeeper.total_sleep_time.tv_nsec = 0;
-	write_sequnlock_irqrestore(&timekeeper.lock, flags);
+		boot = tk_xtime(tk);
+
+	set_normalized_timespec(&tmp, -boot.tv_sec, -boot.tv_nsec);
+	tk_set_wall_to_mono(tk, tmp);
+
+	tmp.tv_sec = 0;
+	tmp.tv_nsec = 0;
+	tk_set_sleep_time(tk, tmp);
+
+	write_sequnlock_irqrestore(&tk->lock, flags);
 }
 
 /* time in seconds when suspend began */
 static struct timespec timekeeping_suspend_time;
 
-static void update_sleep_time(struct timespec t)
-{
-	timekeeper.total_sleep_time = t;
-	timekeeper.offs_boot = timespec_to_ktime(t);
-}
-
 /**
  * __timekeeping_inject_sleeptime - Internal function to add sleep interval
  * @delta: pointer to a timespec delta value
@@ -677,13 +695,11 @@ static void __timekeeping_inject_sleeptime(struct timekeeper *tk,
 					"sleep delta value!\n");
 		return;
 	}
-
 	tk_xtime_add(tk, delta);
-	tk->wall_to_monotonic = timespec_sub(tk->wall_to_monotonic, *delta);
-	update_sleep_time(timespec_add(tk->total_sleep_time, *delta));
+	tk_set_wall_to_mono(tk, timespec_sub(tk->wall_to_monotonic, *delta));
+	tk_set_sleep_time(tk, timespec_add(tk->total_sleep_time, *delta));
 }
 
-
 /**
  * timekeeping_inject_sleeptime - Adds suspend interval to timeekeeping values
  * @delta: pointer to a timespec delta value
@@ -696,6 +712,7 @@ static void __timekeeping_inject_sleeptime(struct timekeeper *tk,
  */
 void timekeeping_inject_sleeptime(struct timespec *delta)
 {
+	struct timekeeper *tk = &timekeeper;
 	unsigned long flags;
 	struct timespec ts;
 
@@ -704,21 +721,20 @@ void timekeeping_inject_sleeptime(struct timespec *delta)
 	if (!(ts.tv_sec == 0 && ts.tv_nsec == 0))
 		return;
 
-	write_seqlock_irqsave(&timekeeper.lock, flags);
+	write_seqlock_irqsave(&tk->lock, flags);
 
-	timekeeping_forward_now(&timekeeper);
+	timekeeping_forward_now(tk);
 
-	__timekeeping_inject_sleeptime(&timekeeper, delta);
+	__timekeeping_inject_sleeptime(tk, delta);
 
-	timekeeping_update(&timekeeper, true);
+	timekeeping_update(tk, true);
 
-	write_sequnlock_irqrestore(&timekeeper.lock, flags);
+	write_sequnlock_irqrestore(&tk->lock, flags);
 
 	/* signal hrtimers about time change */
 	clock_was_set();
 }
 
-
 /**
  * timekeeping_resume - Resumes the generic timekeeping subsystem.
  *
@@ -728,6 +744,7 @@ void timekeeping_inject_sleeptime(struct timespec *delta)
  */
 static void timekeeping_resume(void)
 {
+	struct timekeeper *tk = &timekeeper;
 	unsigned long flags;
 	struct timespec ts;
 
@@ -735,18 +752,18 @@ static void timekeeping_resume(void)
 
 	clocksource_resume();
 
-	write_seqlock_irqsave(&timekeeper.lock, flags);
+	write_seqlock_irqsave(&tk->lock, flags);
 
 	if (timespec_compare(&ts, &timekeeping_suspend_time) > 0) {
 		ts = timespec_sub(ts, timekeeping_suspend_time);
-		__timekeeping_inject_sleeptime(&timekeeper, &ts);
+		__timekeeping_inject_sleeptime(tk, &ts);
 	}
 	/* re-base the last cycle value */
-	timekeeper.clock->cycle_last = timekeeper.clock->read(timekeeper.clock);
-	timekeeper.ntp_error = 0;
+	tk->clock->cycle_last = tk->clock->read(tk->clock);
+	tk->ntp_error = 0;
 	timekeeping_suspended = 0;
-	timekeeping_update(&timekeeper, false);
-	write_sequnlock_irqrestore(&timekeeper.lock, flags);
+	timekeeping_update(tk, false);
+	write_sequnlock_irqrestore(&tk->lock, flags);
 
 	touch_softlockup_watchdog();
 
@@ -758,14 +775,15 @@ static void timekeeping_resume(void)
 
 static int timekeeping_suspend(void)
 {
+	struct timekeeper *tk = &timekeeper;
 	unsigned long flags;
 	struct timespec		delta, delta_delta;
 	static struct timespec	old_delta;
 
 	read_persistent_clock(&timekeeping_suspend_time);
 
-	write_seqlock_irqsave(&timekeeper.lock, flags);
-	timekeeping_forward_now(&timekeeper);
+	write_seqlock_irqsave(&tk->lock, flags);
+	timekeeping_forward_now(tk);
 	timekeeping_suspended = 1;
 
 	/*
@@ -774,7 +792,7 @@ static int timekeeping_suspend(void)
 	 * try to compensate so the difference in system time
 	 * and persistent_clock time stays close to constant.
 	 */
-	delta = timespec_sub(tk_xtime(&timekeeper), timekeeping_suspend_time);
+	delta = timespec_sub(tk_xtime(tk), timekeeping_suspend_time);
 	delta_delta = timespec_sub(delta, old_delta);
 	if (abs(delta_delta.tv_sec)  >= 2) {
 		/*
@@ -787,7 +805,7 @@ static int timekeeping_suspend(void)
 		timekeeping_suspend_time =
 			timespec_add(timekeeping_suspend_time, delta_delta);
 	}
-	write_sequnlock_irqrestore(&timekeeper.lock, flags);
+	write_sequnlock_irqrestore(&tk->lock, flags);
 
 	clockevents_notify(CLOCK_EVT_NOTIFY_SUSPEND, NULL);
 	clocksource_suspend();
@@ -898,7 +916,7 @@ static void timekeeping_adjust(struct timekeeper *tk, s64 offset)
 		 * the error. This causes the likely below to be unlikely.
 		 *
 		 * The proper fix is to avoid rounding up by using
-		 * the high precision timekeeper.xtime_nsec instead of
+		 * the high precision tk->xtime_nsec instead of
 		 * xtime.tv_nsec everywhere. Fixing this will take some
 		 * time.
 		 */
@@ -1003,7 +1021,6 @@ static void timekeeping_adjust(struct timekeeper *tk, s64 offset)
 
 }
 
-
 /**
  * accumulate_nsecs_to_secs - Accumulates nsecs into secs
  *
@@ -1024,15 +1041,21 @@ static inline void accumulate_nsecs_to_secs(struct timekeeper *tk)
 
 		/* Figure out if its a leap sec and apply if needed */
 		leap = second_overflow(tk->xtime_sec);
-		tk->xtime_sec += leap;
-		tk->wall_to_monotonic.tv_sec -= leap;
-		if (leap)
-			clock_was_set_delayed();
+		if (unlikely(leap)) {
+			struct timespec ts;
+
+			tk->xtime_sec += leap;
 
+			ts.tv_sec = leap;
+			ts.tv_nsec = 0;
+			tk_set_wall_to_mono(tk,
+				timespec_sub(tk->wall_to_monotonic, ts));
+
+			clock_was_set_delayed();
+		}
 	}
 }
 
-
 /**
  * logarithmic_accumulation - shifted accumulation of cycles
  *
@@ -1076,7 +1099,6 @@ static cycle_t logarithmic_accumulation(struct timekeeper *tk, cycle_t offset,
 	return offset;
 }
 
-
 /**
  * update_wall_time - Uses the current clocksource to increment the wall time
  *
@@ -1084,21 +1106,22 @@ static cycle_t logarithmic_accumulation(struct timekeeper *tk, cycle_t offset,
 static void update_wall_time(void)
 {
 	struct clocksource *clock;
+	struct timekeeper *tk = &timekeeper;
 	cycle_t offset;
 	int shift = 0, maxshift;
 	unsigned long flags;
 	s64 remainder;
 
-	write_seqlock_irqsave(&timekeeper.lock, flags);
+	write_seqlock_irqsave(&tk->lock, flags);
 
 	/* Make sure we're fully resumed: */
 	if (unlikely(timekeeping_suspended))
 		goto out;
 
-	clock = timekeeper.clock;
+	clock = tk->clock;
 
 #ifdef CONFIG_ARCH_USES_GETTIMEOFFSET
-	offset = timekeeper.cycle_interval;
+	offset = tk->cycle_interval;
 #else
 	offset = (clock->read(clock) - clock->cycle_last) & clock->mask;
 #endif
@@ -1111,19 +1134,19 @@ static void update_wall_time(void)
 	 * chunk in one go, and then try to consume the next smaller
 	 * doubled multiple.
 	 */
-	shift = ilog2(offset) - ilog2(timekeeper.cycle_interval);
+	shift = ilog2(offset) - ilog2(tk->cycle_interval);
 	shift = max(0, shift);
 	/* Bound shift to one less than what overflows tick_length */
 	maxshift = (64 - (ilog2(ntp_tick_length())+1)) - 1;
 	shift = min(shift, maxshift);
-	while (offset >= timekeeper.cycle_interval) {
-		offset = logarithmic_accumulation(&timekeeper, offset, shift);
-		if(offset < timekeeper.cycle_interval<<shift)
+	while (offset >= tk->cycle_interval) {
+		offset = logarithmic_accumulation(tk, offset, shift);
+		if (offset < tk->cycle_interval<<shift)
 			shift--;
 	}
 
 	/* correct the clock when NTP error is too big */
-	timekeeping_adjust(&timekeeper, offset);
+	timekeeping_adjust(tk, offset);
 
 
 	/*
@@ -1135,21 +1158,21 @@ static void update_wall_time(void)
 	* the vsyscall implementations are converted to use xtime_nsec
 	* (shifted nanoseconds), this can be killed.
 	*/
-	remainder = timekeeper.xtime_nsec & ((1 << timekeeper.shift) - 1);
-	timekeeper.xtime_nsec -= remainder;
-	timekeeper.xtime_nsec += 1 << timekeeper.shift;
-	timekeeper.ntp_error += remainder << timekeeper.ntp_error_shift;
+	remainder = tk->xtime_nsec & ((1 << tk->shift) - 1);
+	tk->xtime_nsec -= remainder;
+	tk->xtime_nsec += 1 << tk->shift;
+	tk->ntp_error += remainder << tk->ntp_error_shift;
 
 	/*
 	 * Finally, make sure that after the rounding
 	 * xtime_nsec isn't larger than NSEC_PER_SEC
 	 */
-	accumulate_nsecs_to_secs(&timekeeper);
+	accumulate_nsecs_to_secs(tk);
 
-	timekeeping_update(&timekeeper, false);
+	timekeeping_update(tk, false);
 
 out:
-	write_sequnlock_irqrestore(&timekeeper.lock, flags);
+	write_sequnlock_irqrestore(&tk->lock, flags);
 
 }
 
@@ -1166,18 +1189,18 @@ out:
  */
 void getboottime(struct timespec *ts)
 {
+	struct timekeeper *tk = &timekeeper;
 	struct timespec boottime = {
-		.tv_sec = timekeeper.wall_to_monotonic.tv_sec +
-				timekeeper.total_sleep_time.tv_sec,
-		.tv_nsec = timekeeper.wall_to_monotonic.tv_nsec +
-				timekeeper.total_sleep_time.tv_nsec
+		.tv_sec = tk->wall_to_monotonic.tv_sec +
+				tk->total_sleep_time.tv_sec,
+		.tv_nsec = tk->wall_to_monotonic.tv_nsec +
+				tk->total_sleep_time.tv_nsec
 	};
 
 	set_normalized_timespec(ts, -boottime.tv_sec, -boottime.tv_nsec);
 }
 EXPORT_SYMBOL_GPL(getboottime);
 
-
 /**
  * get_monotonic_boottime - Returns monotonic time since boot
  * @ts:		pointer to the timespec to be set
@@ -1189,19 +1212,20 @@ EXPORT_SYMBOL_GPL(getboottime);
  */
 void get_monotonic_boottime(struct timespec *ts)
 {
+	struct timekeeper *tk = &timekeeper;
 	struct timespec tomono, sleep;
 	unsigned int seq;
 
 	WARN_ON(timekeeping_suspended);
 
 	do {
-		seq = read_seqbegin(&timekeeper.lock);
-		ts->tv_sec = timekeeper.xtime_sec;
-		ts->tv_nsec = timekeeping_get_ns(&timekeeper);
-		tomono = timekeeper.wall_to_monotonic;
-		sleep = timekeeper.total_sleep_time;
+		seq = read_seqbegin(&tk->lock);
+		ts->tv_sec = tk->xtime_sec;
+		ts->tv_nsec = timekeeping_get_ns(tk);
+		tomono = tk->wall_to_monotonic;
+		sleep = tk->total_sleep_time;
 
-	} while (read_seqretry(&timekeeper.lock, seq));
+	} while (read_seqretry(&tk->lock, seq));
 
 	set_normalized_timespec(ts, ts->tv_sec + tomono.tv_sec + sleep.tv_sec,
 			ts->tv_nsec + tomono.tv_nsec + sleep.tv_nsec);
@@ -1231,31 +1255,38 @@ EXPORT_SYMBOL_GPL(ktime_get_boottime);
  */
 void monotonic_to_bootbased(struct timespec *ts)
 {
-	*ts = timespec_add(*ts, timekeeper.total_sleep_time);
+	struct timekeeper *tk = &timekeeper;
+
+	*ts = timespec_add(*ts, tk->total_sleep_time);
 }
 EXPORT_SYMBOL_GPL(monotonic_to_bootbased);
 
 unsigned long get_seconds(void)
 {
-	return timekeeper.xtime_sec;
+	struct timekeeper *tk = &timekeeper;
+
+	return tk->xtime_sec;
 }
 EXPORT_SYMBOL(get_seconds);
 
 struct timespec __current_kernel_time(void)
 {
-	return tk_xtime(&timekeeper);
+	struct timekeeper *tk = &timekeeper;
+
+	return tk_xtime(tk);
 }
 
 struct timespec current_kernel_time(void)
 {
+	struct timekeeper *tk = &timekeeper;
 	struct timespec now;
 	unsigned long seq;
 
 	do {
-		seq = read_seqbegin(&timekeeper.lock);
+		seq = read_seqbegin(&tk->lock);
 
-		now = tk_xtime(&timekeeper);
-	} while (read_seqretry(&timekeeper.lock, seq));
+		now = tk_xtime(tk);
+	} while (read_seqretry(&tk->lock, seq));
 
 	return now;
 }
@@ -1263,15 +1294,16 @@ EXPORT_SYMBOL(current_kernel_time);
 
 struct timespec get_monotonic_coarse(void)
 {
+	struct timekeeper *tk = &timekeeper;
 	struct timespec now, mono;
 	unsigned long seq;
 
 	do {
-		seq = read_seqbegin(&timekeeper.lock);
+		seq = read_seqbegin(&tk->lock);
 
-		now = tk_xtime(&timekeeper);
-		mono = timekeeper.wall_to_monotonic;
-	} while (read_seqretry(&timekeeper.lock, seq));
+		now = tk_xtime(tk);
+		mono = tk->wall_to_monotonic;
+	} while (read_seqretry(&tk->lock, seq));
 
 	set_normalized_timespec(&now, now.tv_sec + mono.tv_sec,
 				now.tv_nsec + mono.tv_nsec);
@@ -1300,14 +1332,15 @@ void do_timer(unsigned long ticks)
 void get_xtime_and_monotonic_and_sleep_offset(struct timespec *xtim,
 				struct timespec *wtom, struct timespec *sleep)
 {
+	struct timekeeper *tk = &timekeeper;
 	unsigned long seq;
 
 	do {
-		seq = read_seqbegin(&timekeeper.lock);
-		*xtim = tk_xtime(&timekeeper);
-		*wtom = timekeeper.wall_to_monotonic;
-		*sleep = timekeeper.total_sleep_time;
-	} while (read_seqretry(&timekeeper.lock, seq));
+		seq = read_seqbegin(&tk->lock);
+		*xtim = tk_xtime(tk);
+		*wtom = tk->wall_to_monotonic;
+		*sleep = tk->total_sleep_time;
+	} while (read_seqretry(&tk->lock, seq));
 }
 
 #ifdef CONFIG_HIGH_RES_TIMERS
@@ -1321,19 +1354,20 @@ void get_xtime_and_monotonic_and_sleep_offset(struct timespec *xtim,
  */
 ktime_t ktime_get_update_offsets(ktime_t *offs_real, ktime_t *offs_boot)
 {
+	struct timekeeper *tk = &timekeeper;
 	ktime_t now;
 	unsigned int seq;
 	u64 secs, nsecs;
 
 	do {
-		seq = read_seqbegin(&timekeeper.lock);
+		seq = read_seqbegin(&tk->lock);
 
-		secs = timekeeper.xtime_sec;
-		nsecs = timekeeping_get_ns(&timekeeper);
+		secs = tk->xtime_sec;
+		nsecs = timekeeping_get_ns(tk);
 
-		*offs_real = timekeeper.offs_real;
-		*offs_boot = timekeeper.offs_boot;
-	} while (read_seqretry(&timekeeper.lock, seq));
+		*offs_real = tk->offs_real;
+		*offs_boot = tk->offs_boot;
+	} while (read_seqretry(&tk->lock, seq));
 
 	now = ktime_add_ns(ktime_set(secs, 0), nsecs);
 	now = ktime_sub(now, *offs_real);
@@ -1346,19 +1380,19 @@ ktime_t ktime_get_update_offsets(ktime_t *offs_real, ktime_t *offs_boot)
  */
 ktime_t ktime_get_monotonic_offset(void)
 {
+	struct timekeeper *tk = &timekeeper;
 	unsigned long seq;
 	struct timespec wtom;
 
 	do {
-		seq = read_seqbegin(&timekeeper.lock);
-		wtom = timekeeper.wall_to_monotonic;
-	} while (read_seqretry(&timekeeper.lock, seq));
+		seq = read_seqbegin(&tk->lock);
+		wtom = tk->wall_to_monotonic;
+	} while (read_seqretry(&tk->lock, seq));
 
 	return timespec_to_ktime(wtom);
 }
 EXPORT_SYMBOL_GPL(ktime_get_monotonic_offset);
 
-
 /**
  * xtime_update() - advances the timekeeping infrastructure
  * @ticks:	number of ticks, that have elapsed since the last call.

^ permalink raw reply related	[flat|nested] 69+ messages in thread

* [GIT PULL] timer fixes
@ 2011-12-20 19:25 Ingo Molnar
  0 siblings, 0 replies; 69+ messages in thread
From: Ingo Molnar @ 2011-12-20 19:25 UTC (permalink / raw)
  To: Linus Torvalds
  Cc: linux-kernel, Thomas Gleixner, Peter Zijlstra, Andrew Morton

Linus,

Please pull the latest timers-urgent-for-linus git tree from:

   git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git timers-urgent-for-linus

 Thanks,

	Ingo

------------------>
John Stultz (1):
      rtc: m41t80: Workaround broken alarm functionality

Kusanagi Kouichi (1):
      time/clocksource: Fix kernel-doc warnings

NeilBrown (1):
      rtc: Expire alarms after the time is set.


 drivers/rtc/interface.c     |    6 ++++++
 drivers/rtc/rtc-m41t80.c    |    9 +++++++++
 include/linux/clocksource.h |   11 ++++++++---
 kernel/time/clocksource.c   |   12 +++++++++---
 4 files changed, 32 insertions(+), 6 deletions(-)

diff --git a/drivers/rtc/interface.c b/drivers/rtc/interface.c
index fa4d9f3..3bcc7cf 100644
--- a/drivers/rtc/interface.c
+++ b/drivers/rtc/interface.c
@@ -73,6 +73,8 @@ int rtc_set_time(struct rtc_device *rtc, struct rtc_time *tm)
 		err = -EINVAL;
 
 	mutex_unlock(&rtc->ops_lock);
+	/* A timer might have just expired */
+	schedule_work(&rtc->irqwork);
 	return err;
 }
 EXPORT_SYMBOL_GPL(rtc_set_time);
@@ -112,6 +114,8 @@ int rtc_set_mmss(struct rtc_device *rtc, unsigned long secs)
 		err = -EINVAL;
 
 	mutex_unlock(&rtc->ops_lock);
+	/* A timer might have just expired */
+	schedule_work(&rtc->irqwork);
 
 	return err;
 }
@@ -403,6 +407,8 @@ int rtc_initialize_alarm(struct rtc_device *rtc, struct rtc_wkalrm *alarm)
 		timerqueue_add(&rtc->timerqueue, &rtc->aie_timer.node);
 	}
 	mutex_unlock(&rtc->ops_lock);
+	/* maybe that was in the past.*/
+	schedule_work(&rtc->irqwork);
 	return err;
 }
 EXPORT_SYMBOL_GPL(rtc_initialize_alarm);
diff --git a/drivers/rtc/rtc-m41t80.c b/drivers/rtc/rtc-m41t80.c
index eda128f..64aedd8 100644
--- a/drivers/rtc/rtc-m41t80.c
+++ b/drivers/rtc/rtc-m41t80.c
@@ -357,10 +357,19 @@ static int m41t80_rtc_read_alarm(struct device *dev, struct rtc_wkalrm *t)
 static struct rtc_class_ops m41t80_rtc_ops = {
 	.read_time = m41t80_rtc_read_time,
 	.set_time = m41t80_rtc_set_time,
+	/*
+	 * XXX - m41t80 alarm functionality is reported broken.
+	 * until it is fixed, don't register alarm functions.
+	 *
 	.read_alarm = m41t80_rtc_read_alarm,
 	.set_alarm = m41t80_rtc_set_alarm,
+	*/
 	.proc = m41t80_rtc_proc,
+	/*
+	 * See above comment on broken alarm
+	 *
 	.alarm_irq_enable = m41t80_rtc_alarm_irq_enable,
+	*/
 };
 
 #if defined(CONFIG_RTC_INTF_SYSFS) || defined(CONFIG_RTC_INTF_SYSFS_MODULE)
diff --git a/include/linux/clocksource.h b/include/linux/clocksource.h
index c86c940..081147d 100644
--- a/include/linux/clocksource.h
+++ b/include/linux/clocksource.h
@@ -71,7 +71,7 @@ struct timecounter {
 
 /**
  * cyclecounter_cyc2ns - converts cycle counter cycles to nanoseconds
- * @tc:		Pointer to cycle counter.
+ * @cc:		Pointer to cycle counter.
  * @cycles:	Cycles
  *
  * XXX - This could use some mult_lxl_ll() asm optimization. Same code
@@ -114,7 +114,7 @@ extern u64 timecounter_read(struct timecounter *tc);
  *                        time base as values returned by
  *                        timecounter_read()
  * @tc:		Pointer to time counter.
- * @cycle:	a value returned by tc->cc->read()
+ * @cycle_tstamp:	a value returned by tc->cc->read()
  *
  * Cycle counts that are converted correctly as long as they
  * fall into the interval [-1/2 max cycle count, +1/2 max cycle count],
@@ -156,11 +156,12 @@ extern u64 timecounter_cyc2time(struct timecounter *tc,
  * @mult:		cycle to nanosecond multiplier
  * @shift:		cycle to nanosecond divisor (power of two)
  * @max_idle_ns:	max idle time permitted by the clocksource (nsecs)
- * @maxadj		maximum adjustment value to mult (~11%)
+ * @maxadj:		maximum adjustment value to mult (~11%)
  * @flags:		flags describing special properties
  * @archdata:		arch-specific data
  * @suspend:		suspend function for the clocksource, if necessary
  * @resume:		resume function for the clocksource, if necessary
+ * @cycle_last:		most recent cycle counter value seen by ::read()
  */
 struct clocksource {
 	/*
@@ -187,6 +188,7 @@ struct clocksource {
 	void (*suspend)(struct clocksource *cs);
 	void (*resume)(struct clocksource *cs);
 
+	/* private: */
 #ifdef CONFIG_CLOCKSOURCE_WATCHDOG
 	/* Watchdog related data, used by the framework */
 	struct list_head wd_list;
@@ -261,6 +263,9 @@ static inline u32 clocksource_hz2mult(u32 hz, u32 shift_constant)
 
 /**
  * clocksource_cyc2ns - converts clocksource cycles to nanoseconds
+ * @cycles:	cycles
+ * @mult:	cycle to nanosecond multiplier
+ * @shift:	cycle to nanosecond divisor (power of two)
  *
  * Converts cycles to nanoseconds, using the given mult and shift.
  *
diff --git a/kernel/time/clocksource.c b/kernel/time/clocksource.c
index da2f760..d3ad022 100644
--- a/kernel/time/clocksource.c
+++ b/kernel/time/clocksource.c
@@ -647,7 +647,7 @@ static void clocksource_enqueue(struct clocksource *cs)
 
 /**
  * __clocksource_updatefreq_scale - Used update clocksource with new freq
- * @t:		clocksource to be registered
+ * @cs:		clocksource to be registered
  * @scale:	Scale factor multiplied against freq to get clocksource hz
  * @freq:	clocksource frequency (cycles per second) divided by scale
  *
@@ -699,7 +699,7 @@ EXPORT_SYMBOL_GPL(__clocksource_updatefreq_scale);
 
 /**
  * __clocksource_register_scale - Used to install new clocksources
- * @t:		clocksource to be registered
+ * @cs:		clocksource to be registered
  * @scale:	Scale factor multiplied against freq to get clocksource hz
  * @freq:	clocksource frequency (cycles per second) divided by scale
  *
@@ -727,7 +727,7 @@ EXPORT_SYMBOL_GPL(__clocksource_register_scale);
 
 /**
  * clocksource_register - Used to install new clocksources
- * @t:		clocksource to be registered
+ * @cs:		clocksource to be registered
  *
  * Returns -EBUSY if registration fails, zero otherwise.
  */
@@ -761,6 +761,8 @@ static void __clocksource_change_rating(struct clocksource *cs, int rating)
 
 /**
  * clocksource_change_rating - Change the rating of a registered clocksource
+ * @cs:		clocksource to be changed
+ * @rating:	new rating
  */
 void clocksource_change_rating(struct clocksource *cs, int rating)
 {
@@ -772,6 +774,7 @@ EXPORT_SYMBOL(clocksource_change_rating);
 
 /**
  * clocksource_unregister - remove a registered clocksource
+ * @cs:	clocksource to be unregistered
  */
 void clocksource_unregister(struct clocksource *cs)
 {
@@ -787,6 +790,7 @@ EXPORT_SYMBOL(clocksource_unregister);
 /**
  * sysfs_show_current_clocksources - sysfs interface for current clocksource
  * @dev:	unused
+ * @attr:	unused
  * @buf:	char buffer to be filled with clocksource list
  *
  * Provides sysfs interface for listing current clocksource.
@@ -807,6 +811,7 @@ sysfs_show_current_clocksources(struct sys_device *dev,
 /**
  * sysfs_override_clocksource - interface for manually overriding clocksource
  * @dev:	unused
+ * @attr:	unused
  * @buf:	name of override clocksource
  * @count:	length of buffer
  *
@@ -842,6 +847,7 @@ static ssize_t sysfs_override_clocksource(struct sys_device *dev,
 /**
  * sysfs_show_available_clocksources - sysfs interface for listing clocksource
  * @dev:	unused
+ * @attr:	unused
  * @buf:	char buffer to be filled with clocksource list
  *
  * Provides sysfs interface for listing registered clocksources

^ permalink raw reply related	[flat|nested] 69+ messages in thread

* [GIT PULL] timer fixes
@ 2011-12-05 19:00 Ingo Molnar
  0 siblings, 0 replies; 69+ messages in thread
From: Ingo Molnar @ 2011-12-05 19:00 UTC (permalink / raw)
  To: Linus Torvalds
  Cc: linux-kernel, Thomas Gleixner, Peter Zijlstra, Andrew Morton

Linus,

Please pull the latest timers-urgent-for-linus git tree from:

   git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git timers-urgent-for-linus

 Thanks,

	Ingo

------------------>
Arve Hjønnevåg (1):
      rtc: Fix some bugs that allowed accumulating time drift in suspend/resume

Rabin Vincent (1):
      rtc: Disable the alarm in the hardware

Thomas Gleixner (2):
      tick-broadcast: Stop active broadcast device when replacing it
      clockevents: Set noop handler in clockevents_exchange_device()

Yang Honggang (Joseph) (1):
      clocksource: Fix bug with max_deferment margin calculation


 drivers/rtc/class.c          |   10 ++++----
 drivers/rtc/interface.c      |   44 ++++++++++++++++++++++++++++++++---------
 kernel/time/clockevents.c    |    1 +
 kernel/time/clocksource.c    |    4 +-
 kernel/time/tick-broadcast.c |    2 +-
 5 files changed, 43 insertions(+), 18 deletions(-)

diff --git a/drivers/rtc/class.c b/drivers/rtc/class.c
index 01a7df5..b82a155 100644
--- a/drivers/rtc/class.c
+++ b/drivers/rtc/class.c
@@ -66,7 +66,7 @@ static int rtc_suspend(struct device *dev, pm_message_t mesg)
 	 */
 	delta = timespec_sub(old_system, old_rtc);
 	delta_delta = timespec_sub(delta, old_delta);
-	if (abs(delta_delta.tv_sec)  >= 2) {
+	if (delta_delta.tv_sec < -2 || delta_delta.tv_sec >= 2) {
 		/*
 		 * if delta_delta is too large, assume time correction
 		 * has occured and set old_delta to the current delta.
@@ -100,9 +100,8 @@ static int rtc_resume(struct device *dev)
 	rtc_tm_to_time(&tm, &new_rtc.tv_sec);
 	new_rtc.tv_nsec = 0;
 
-	if (new_rtc.tv_sec <= old_rtc.tv_sec) {
-		if (new_rtc.tv_sec < old_rtc.tv_sec)
-			pr_debug("%s:  time travel!\n", dev_name(&rtc->dev));
+	if (new_rtc.tv_sec < old_rtc.tv_sec) {
+		pr_debug("%s:  time travel!\n", dev_name(&rtc->dev));
 		return 0;
 	}
 
@@ -119,7 +118,8 @@ static int rtc_resume(struct device *dev)
 	sleep_time = timespec_sub(sleep_time,
 			timespec_sub(new_system, old_system));
 
-	timekeeping_inject_sleeptime(&sleep_time);
+	if (sleep_time.tv_sec >= 0)
+		timekeeping_inject_sleeptime(&sleep_time);
 	return 0;
 }
 
diff --git a/drivers/rtc/interface.c b/drivers/rtc/interface.c
index 3195dbd..e3cf2f5 100644
--- a/drivers/rtc/interface.c
+++ b/drivers/rtc/interface.c
@@ -318,6 +318,20 @@ int rtc_read_alarm(struct rtc_device *rtc, struct rtc_wkalrm *alarm)
 }
 EXPORT_SYMBOL_GPL(rtc_read_alarm);
 
+static int ___rtc_set_alarm(struct rtc_device *rtc, struct rtc_wkalrm *alarm)
+{
+	int err;
+
+	if (!rtc->ops)
+		err = -ENODEV;
+	else if (!rtc->ops->set_alarm)
+		err = -EINVAL;
+	else
+		err = rtc->ops->set_alarm(rtc->dev.parent, alarm);
+
+	return err;
+}
+
 static int __rtc_set_alarm(struct rtc_device *rtc, struct rtc_wkalrm *alarm)
 {
 	struct rtc_time tm;
@@ -341,14 +355,7 @@ static int __rtc_set_alarm(struct rtc_device *rtc, struct rtc_wkalrm *alarm)
 	 * over right here, before we set the alarm.
 	 */
 
-	if (!rtc->ops)
-		err = -ENODEV;
-	else if (!rtc->ops->set_alarm)
-		err = -EINVAL;
-	else
-		err = rtc->ops->set_alarm(rtc->dev.parent, alarm);
-
-	return err;
+	return ___rtc_set_alarm(rtc, alarm);
 }
 
 int rtc_set_alarm(struct rtc_device *rtc, struct rtc_wkalrm *alarm)
@@ -762,6 +769,20 @@ static int rtc_timer_enqueue(struct rtc_device *rtc, struct rtc_timer *timer)
 	return 0;
 }
 
+static void rtc_alarm_disable(struct rtc_device *rtc)
+{
+	struct rtc_wkalrm alarm;
+	struct rtc_time tm;
+
+	__rtc_read_time(rtc, &tm);
+
+	alarm.time = rtc_ktime_to_tm(ktime_add(rtc_tm_to_ktime(tm),
+				     ktime_set(300, 0)));
+	alarm.enabled = 0;
+
+	___rtc_set_alarm(rtc, &alarm);
+}
+
 /**
  * rtc_timer_remove - Removes a rtc_timer from the rtc_device timerqueue
  * @rtc rtc device
@@ -783,8 +804,10 @@ static void rtc_timer_remove(struct rtc_device *rtc, struct rtc_timer *timer)
 		struct rtc_wkalrm alarm;
 		int err;
 		next = timerqueue_getnext(&rtc->timerqueue);
-		if (!next)
+		if (!next) {
+			rtc_alarm_disable(rtc);
 			return;
+		}
 		alarm.time = rtc_ktime_to_tm(next->expires);
 		alarm.enabled = 1;
 		err = __rtc_set_alarm(rtc, &alarm);
@@ -846,7 +869,8 @@ again:
 		err = __rtc_set_alarm(rtc, &alarm);
 		if (err == -ETIME)
 			goto again;
-	}
+	} else
+		rtc_alarm_disable(rtc);
 
 	mutex_unlock(&rtc->ops_lock);
 }
diff --git a/kernel/time/clockevents.c b/kernel/time/clockevents.c
index 1ecd6ba..c4eb71c 100644
--- a/kernel/time/clockevents.c
+++ b/kernel/time/clockevents.c
@@ -387,6 +387,7 @@ void clockevents_exchange_device(struct clock_event_device *old,
 	 * released list and do a notify add later.
 	 */
 	if (old) {
+		old->event_handler = clockevents_handle_noop;
 		clockevents_set_mode(old, CLOCK_EVT_MODE_UNUSED);
 		list_del(&old->list);
 		list_add(&old->list, &clockevents_released);
diff --git a/kernel/time/clocksource.c b/kernel/time/clocksource.c
index cfc65e1..da2f760 100644
--- a/kernel/time/clocksource.c
+++ b/kernel/time/clocksource.c
@@ -548,7 +548,7 @@ static u64 clocksource_max_deferment(struct clocksource *cs)
 	 * note a margin of 12.5% is used because this can be computed with
 	 * a shift, versus say 10% which would require division.
 	 */
-	return max_nsecs - (max_nsecs >> 5);
+	return max_nsecs - (max_nsecs >> 3);
 }
 
 #ifndef CONFIG_ARCH_USES_GETTIMEOFFSET
@@ -669,7 +669,7 @@ void __clocksource_updatefreq_scale(struct clocksource *cs, u32 scale, u32 freq)
 	 * ~ 0.06ppm granularity for NTP. We apply the same 12.5%
 	 * margin as we do in clocksource_max_deferment()
 	 */
-	sec = (cs->mask - (cs->mask >> 5));
+	sec = (cs->mask - (cs->mask >> 3));
 	do_div(sec, freq);
 	do_div(sec, scale);
 	if (!sec)
diff --git a/kernel/time/tick-broadcast.c b/kernel/time/tick-broadcast.c
index f954282..fd4a7b1 100644
--- a/kernel/time/tick-broadcast.c
+++ b/kernel/time/tick-broadcast.c
@@ -71,7 +71,7 @@ int tick_check_broadcast_device(struct clock_event_device *dev)
 	     (dev->features & CLOCK_EVT_FEAT_C3STOP))
 		return 0;
 
-	clockevents_exchange_device(NULL, dev);
+	clockevents_exchange_device(tick_broadcast_device.evtdev, dev);
 	tick_broadcast_device.evtdev = dev;
 	if (!cpumask_empty(tick_get_broadcast_mask()))
 		tick_broadcast_start_periodic(dev);

^ permalink raw reply related	[flat|nested] 69+ messages in thread

* [GIT PULL] timer fixes
@ 2011-06-19  8:51 Ingo Molnar
  0 siblings, 0 replies; 69+ messages in thread
From: Ingo Molnar @ 2011-06-19  8:51 UTC (permalink / raw)
  To: Linus Torvalds
  Cc: linux-kernel, Thomas Gleixner, Peter Zijlstra, Andrew Morton

Linus,

Please pull the latest timers-urgent-for-linus git tree from:

   git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip.git timers-urgent-for-linus

 Thanks,

	Ingo

------------------>
Max Asbock (1):
      timerfd: Fix wakeup of processes when timer is cancelled on clock change

Thomas Gleixner (1):
      clocksource: Make watchdog robust vs. interruption


 fs/timerfd.c                |    5 ++++-
 include/linux/clocksource.h |    1 +
 kernel/time/clocksource.c   |   24 +++++++++++++-----------
 3 files changed, 18 insertions(+), 12 deletions(-)

diff --git a/fs/timerfd.c b/fs/timerfd.c
index f67acbd..dffeb37 100644
--- a/fs/timerfd.c
+++ b/fs/timerfd.c
@@ -61,7 +61,9 @@ static enum hrtimer_restart timerfd_tmrproc(struct hrtimer *htmr)
 
 /*
  * Called when the clock was set to cancel the timers in the cancel
- * list.
+ * list. This will wake up processes waiting on these timers. The
+ * wake-up requires ctx->ticks to be non zero, therefore we increment
+ * it before calling wake_up_locked().
  */
 void timerfd_clock_was_set(void)
 {
@@ -76,6 +78,7 @@ void timerfd_clock_was_set(void)
 		spin_lock_irqsave(&ctx->wqh.lock, flags);
 		if (ctx->moffs.tv64 != moffs.tv64) {
 			ctx->moffs.tv64 = KTIME_MAX;
+			ctx->ticks++;
 			wake_up_locked(&ctx->wqh);
 		}
 		spin_unlock_irqrestore(&ctx->wqh.lock, flags);
diff --git a/include/linux/clocksource.h b/include/linux/clocksource.h
index d4646b4..18a1baf 100644
--- a/include/linux/clocksource.h
+++ b/include/linux/clocksource.h
@@ -188,6 +188,7 @@ struct clocksource {
 #ifdef CONFIG_CLOCKSOURCE_WATCHDOG
 	/* Watchdog related data, used by the framework */
 	struct list_head wd_list;
+	cycle_t cs_last;
 	cycle_t wd_last;
 #endif
 } ____cacheline_aligned;
diff --git a/kernel/time/clocksource.c b/kernel/time/clocksource.c
index 1c95fd6..e0980f0 100644
--- a/kernel/time/clocksource.c
+++ b/kernel/time/clocksource.c
@@ -185,7 +185,6 @@ static struct clocksource *watchdog;
 static struct timer_list watchdog_timer;
 static DECLARE_WORK(watchdog_work, clocksource_watchdog_work);
 static DEFINE_SPINLOCK(watchdog_lock);
-static cycle_t watchdog_last;
 static int watchdog_running;
 
 static int clocksource_watchdog_kthread(void *data);
@@ -254,11 +253,6 @@ static void clocksource_watchdog(unsigned long data)
 	if (!watchdog_running)
 		goto out;
 
-	wdnow = watchdog->read(watchdog);
-	wd_nsec = clocksource_cyc2ns((wdnow - watchdog_last) & watchdog->mask,
-				     watchdog->mult, watchdog->shift);
-	watchdog_last = wdnow;
-
 	list_for_each_entry(cs, &watchdog_list, wd_list) {
 
 		/* Clocksource already marked unstable? */
@@ -268,19 +262,28 @@ static void clocksource_watchdog(unsigned long data)
 			continue;
 		}
 
+		local_irq_disable();
 		csnow = cs->read(cs);
+		wdnow = watchdog->read(watchdog);
+		local_irq_enable();
 
 		/* Clocksource initialized ? */
 		if (!(cs->flags & CLOCK_SOURCE_WATCHDOG)) {
 			cs->flags |= CLOCK_SOURCE_WATCHDOG;
-			cs->wd_last = csnow;
+			cs->wd_last = wdnow;
+			cs->cs_last = csnow;
 			continue;
 		}
 
-		/* Check the deviation from the watchdog clocksource. */
-		cs_nsec = clocksource_cyc2ns((csnow - cs->wd_last) &
+		wd_nsec = clocksource_cyc2ns((wdnow - cs->wd_last) & watchdog->mask,
+					     watchdog->mult, watchdog->shift);
+
+		cs_nsec = clocksource_cyc2ns((csnow - cs->cs_last) &
 					     cs->mask, cs->mult, cs->shift);
-		cs->wd_last = csnow;
+		cs->cs_last = csnow;
+		cs->wd_last = wdnow;
+
+		/* Check the deviation from the watchdog clocksource. */
 		if (abs(cs_nsec - wd_nsec) > WATCHDOG_THRESHOLD) {
 			clocksource_unstable(cs, cs_nsec - wd_nsec);
 			continue;
@@ -318,7 +321,6 @@ static inline void clocksource_start_watchdog(void)
 		return;
 	init_timer(&watchdog_timer);
 	watchdog_timer.function = clocksource_watchdog;
-	watchdog_last = watchdog->read(watchdog);
 	watchdog_timer.expires = jiffies + WATCHDOG_INTERVAL;
 	add_timer_on(&watchdog_timer, cpumask_first(cpu_online_mask));
 	watchdog_running = 1;

^ permalink raw reply related	[flat|nested] 69+ messages in thread

* [GIT PULL] timer fixes
@ 2011-06-13  9:55 Ingo Molnar
  0 siblings, 0 replies; 69+ messages in thread
From: Ingo Molnar @ 2011-06-13  9:55 UTC (permalink / raw)
  To: Linus Torvalds
  Cc: linux-kernel, Thomas Gleixner, Peter Zijlstra, Andrew Morton

Linus,

Please pull the latest timers-urgent-for-linus git tree from:

   git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip.git timers-urgent-for-linus

 Thanks,

	Ingo

------------------>
Dan Carpenter (2):
      ptp: Return -EFAULT on copy_to_user() errors
      ptp: Fix some locking bugs in ptp_read()

John Stultz (1):
      rtc: Fix ioctl error path return

Mark Brown (1):
      rtc: Staticize non-exported __rtc_set_alarm()


 drivers/ptp/ptp_chardev.c |   11 ++++++-----
 drivers/rtc/interface.c   |    2 +-
 drivers/rtc/rtc-dev.c     |    3 ++-
 3 files changed, 9 insertions(+), 7 deletions(-)

diff --git a/drivers/ptp/ptp_chardev.c b/drivers/ptp/ptp_chardev.c
index a8d03ae..e7f301da2 100644
--- a/drivers/ptp/ptp_chardev.c
+++ b/drivers/ptp/ptp_chardev.c
@@ -46,7 +46,8 @@ long ptp_ioctl(struct posix_clock *pc, unsigned int cmd, unsigned long arg)
 		caps.n_ext_ts = ptp->info->n_ext_ts;
 		caps.n_per_out = ptp->info->n_per_out;
 		caps.pps = ptp->info->pps;
-		err = copy_to_user((void __user *)arg, &caps, sizeof(caps));
+		if (copy_to_user((void __user *)arg, &caps, sizeof(caps)))
+			err = -EFAULT;
 		break;
 
 	case PTP_EXTTS_REQUEST:
@@ -129,8 +130,10 @@ ssize_t ptp_read(struct posix_clock *pc,
 		return -ERESTARTSYS;
 	}
 
-	if (ptp->defunct)
+	if (ptp->defunct) {
+		mutex_unlock(&ptp->tsevq_mux);
 		return -ENODEV;
+	}
 
 	spin_lock_irqsave(&queue->lock, flags);
 
@@ -150,10 +153,8 @@ ssize_t ptp_read(struct posix_clock *pc,
 
 	mutex_unlock(&ptp->tsevq_mux);
 
-	if (copy_to_user(buf, event, cnt)) {
-		mutex_unlock(&ptp->tsevq_mux);
+	if (copy_to_user(buf, event, cnt))
 		return -EFAULT;
-	}
 
 	return cnt;
 }
diff --git a/drivers/rtc/interface.c b/drivers/rtc/interface.c
index ef6316a..df68618 100644
--- a/drivers/rtc/interface.c
+++ b/drivers/rtc/interface.c
@@ -318,7 +318,7 @@ int rtc_read_alarm(struct rtc_device *rtc, struct rtc_wkalrm *alarm)
 }
 EXPORT_SYMBOL_GPL(rtc_read_alarm);
 
-int __rtc_set_alarm(struct rtc_device *rtc, struct rtc_wkalrm *alarm)
+static int __rtc_set_alarm(struct rtc_device *rtc, struct rtc_wkalrm *alarm)
 {
 	struct rtc_time tm;
 	long now, scheduled;
diff --git a/drivers/rtc/rtc-dev.c b/drivers/rtc/rtc-dev.c
index d0e06ed..cace6d3 100644
--- a/drivers/rtc/rtc-dev.c
+++ b/drivers/rtc/rtc-dev.c
@@ -421,7 +421,8 @@ static long rtc_dev_ioctl(struct file *file,
 			err = ops->ioctl(rtc->dev.parent, cmd, arg);
 			if (err == -ENOIOCTLCMD)
 				err = -ENOTTY;
-		}
+		} else
+			err = -ENOTTY;
 		break;
 	}
 

^ permalink raw reply related	[flat|nested] 69+ messages in thread

* [GIT pull] timer fixes
@ 2011-06-06 19:10 Thomas Gleixner
  0 siblings, 0 replies; 69+ messages in thread
From: Thomas Gleixner @ 2011-06-06 19:10 UTC (permalink / raw)
  To: Linus Torvalds; +Cc: Andrew Morton, LKML, Ingo Molnar

Linus,

Please pull the latest timers-fixes-for-linus git tree from:

   git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip.git timers-fixes-for-linus

Thanks,

	tglx

------------------>
Sebastian Andrzej Siewior (1):
      timers: Consider slack value in mod_timer()

Thomas Gleixner (1):
      clockevents: Handle empty cpumask gracefully


 kernel/time/clockevents.c |    5 ++++-
 kernel/timer.c            |   15 +++++++--------
 2 files changed, 11 insertions(+), 9 deletions(-)

diff --git a/kernel/time/clockevents.c b/kernel/time/clockevents.c
index c027d4f..e4c699d 100644
--- a/kernel/time/clockevents.c
+++ b/kernel/time/clockevents.c
@@ -182,7 +182,10 @@ void clockevents_register_device(struct clock_event_device *dev)
 	unsigned long flags;
 
 	BUG_ON(dev->mode != CLOCK_EVT_MODE_UNUSED);
-	BUG_ON(!dev->cpumask);
+	if (!dev->cpumask) {
+		WARN_ON(num_possible_cpus() > 1);
+		dev->cpumask = cpumask_of(smp_processor_id());
+	}
 
 	raw_spin_lock_irqsave(&clockevents_lock, flags);
 
diff --git a/kernel/timer.c b/kernel/timer.c
index fd61986..8cff361 100644
--- a/kernel/timer.c
+++ b/kernel/timer.c
@@ -749,16 +749,15 @@ unsigned long apply_slack(struct timer_list *timer, unsigned long expires)
 	unsigned long expires_limit, mask;
 	int bit;
 
-	expires_limit = expires;
-
 	if (timer->slack >= 0) {
 		expires_limit = expires + timer->slack;
 	} else {
-		unsigned long now = jiffies;
+		long delta = expires - jiffies;
+
+		if (delta < 256)
+			return expires;
 
-		/* No slack, if already expired else auto slack 0.4% */
-		if (time_after(expires, now))
-			expires_limit = expires + (expires - now)/256;
+		expires_limit = expires + delta / 256;
 	}
 	mask = expires ^ expires_limit;
 	if (mask == 0)
@@ -795,6 +794,8 @@ unsigned long apply_slack(struct timer_list *timer, unsigned long expires)
  */
 int mod_timer(struct timer_list *timer, unsigned long expires)
 {
+	expires = apply_slack(timer, expires);
+
 	/*
 	 * This is a common optimization triggered by the
 	 * networking code - if the timer is re-modified
@@ -803,8 +804,6 @@ int mod_timer(struct timer_list *timer, unsigned long expires)
 	if (timer_pending(timer) && timer->expires == expires)
 		return 1;
 
-	expires = apply_slack(timer, expires);
-
 	return __mod_timer(timer, expires, false, TIMER_NOT_PINNED);
 }
 EXPORT_SYMBOL(mod_timer);

^ permalink raw reply related	[flat|nested] 69+ messages in thread

* [GIT pull] timer fixes
@ 2011-05-20 17:52 Thomas Gleixner
  0 siblings, 0 replies; 69+ messages in thread
From: Thomas Gleixner @ 2011-05-20 17:52 UTC (permalink / raw)
  To: Linus Torvalds; +Cc: Andrew Morton, LKML, Ingo Molnar

Linus,

Please pull the latest timers-fixes-for-linus git tree from:

   git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip.git timers-fixes-for-linus

Thanks,

	tglx

------------------>
Thomas Gleixner (2):
      clockevents/source: Use u64 to make 32bit happy
      MAINTAINERS: Add drivers/clocksource to TIMEKEEPING


 MAINTAINERS               |    1 +
 kernel/time/clockevents.c |    2 +-
 kernel/time/clocksource.c |    2 +-
 3 files changed, 3 insertions(+), 2 deletions(-)

diff --git a/MAINTAINERS b/MAINTAINERS
index 8df8d2d..e20c974 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -5431,6 +5431,7 @@ F:	include/linux/timex.h
 F:	kernel/time/clocksource.c
 F:	kernel/time/time*.c
 F:	kernel/time/ntp.c
+F:	drivers/clocksource
 
 TLG2300 VIDEO4LINUX-2 DRIVER
 M:	Huang Shijie <shijie8@gmail.com>
diff --git a/kernel/time/clockevents.c b/kernel/time/clockevents.c
index 22a9da9..c027d4f 100644
--- a/kernel/time/clockevents.c
+++ b/kernel/time/clockevents.c
@@ -197,7 +197,7 @@ EXPORT_SYMBOL_GPL(clockevents_register_device);
 static void clockevents_config(struct clock_event_device *dev,
 			       u32 freq)
 {
-	unsigned long sec;
+	u64 sec;
 
 	if (!(dev->features & CLOCK_EVT_FEAT_ONESHOT))
 		return;
diff --git a/kernel/time/clocksource.c b/kernel/time/clocksource.c
index d9d5f8c..1c95fd6 100644
--- a/kernel/time/clocksource.c
+++ b/kernel/time/clocksource.c
@@ -639,7 +639,7 @@ static void clocksource_enqueue(struct clocksource *cs)
  */
 void __clocksource_updatefreq_scale(struct clocksource *cs, u32 scale, u32 freq)
 {
-	unsigned long sec;
+	u64 sec;
 
 	/*
 	 * Calc the maximum number of seconds which we can run before

^ permalink raw reply related	[flat|nested] 69+ messages in thread

* [GIT PULL] timer fixes
@ 2011-04-19 16:03 Ingo Molnar
  0 siblings, 0 replies; 69+ messages in thread
From: Ingo Molnar @ 2011-04-19 16:03 UTC (permalink / raw)
  To: Linus Torvalds; +Cc: linux-kernel, Thomas Gleixner, Andrew Morton

Linus,

Please pull the latest timer-fixes-for-linus git tree from:

   git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip.git timer-fixes-for-linus

The posix clocks change fixes a pretty bad latency regression introduced in 
this cycle. (In hindsight this information should have been in the changelog)

 Thanks,

	Ingo

------------------>
Axel Lin (1):
      RTC: rtc-omap: Fix a leak of the IRQ during init failure

Richard Cochran (1):
      posix clocks: Replace mutex with reader/writer semaphore


 drivers/rtc/rtc-omap.c      |    2 +-
 include/linux/posix-clock.h |    5 +++--
 kernel/time/posix-clock.c   |   24 +++++++++---------------
 3 files changed, 13 insertions(+), 18 deletions(-)

diff --git a/drivers/rtc/rtc-omap.c b/drivers/rtc/rtc-omap.c
index de0dd7b..bcae8dd 100644
--- a/drivers/rtc/rtc-omap.c
+++ b/drivers/rtc/rtc-omap.c
@@ -394,7 +394,7 @@ static int __init omap_rtc_probe(struct platform_device *pdev)
 	return 0;
 
 fail2:
-	free_irq(omap_rtc_timer, NULL);
+	free_irq(omap_rtc_timer, rtc);
 fail1:
 	rtc_device_unregister(rtc);
 fail0:
diff --git a/include/linux/posix-clock.h b/include/linux/posix-clock.h
index 369e19d..7f1183d 100644
--- a/include/linux/posix-clock.h
+++ b/include/linux/posix-clock.h
@@ -24,6 +24,7 @@
 #include <linux/fs.h>
 #include <linux/poll.h>
 #include <linux/posix-timers.h>
+#include <linux/rwsem.h>
 
 struct posix_clock;
 
@@ -104,7 +105,7 @@ struct posix_clock_operations {
  * @ops:     Functional interface to the clock
  * @cdev:    Character device instance for this clock
  * @kref:    Reference count.
- * @mutex:   Protects the 'zombie' field from concurrent access.
+ * @rwsem:   Protects the 'zombie' field from concurrent access.
  * @zombie:  If 'zombie' is true, then the hardware has disappeared.
  * @release: A function to free the structure when the reference count reaches
  *           zero. May be NULL if structure is statically allocated.
@@ -117,7 +118,7 @@ struct posix_clock {
 	struct posix_clock_operations ops;
 	struct cdev cdev;
 	struct kref kref;
-	struct mutex mutex;
+	struct rw_semaphore rwsem;
 	bool zombie;
 	void (*release)(struct posix_clock *clk);
 };
diff --git a/kernel/time/posix-clock.c b/kernel/time/posix-clock.c
index 25028dd..c340ca6 100644
--- a/kernel/time/posix-clock.c
+++ b/kernel/time/posix-clock.c
@@ -19,7 +19,6 @@
  */
 #include <linux/device.h>
 #include <linux/file.h>
-#include <linux/mutex.h>
 #include <linux/posix-clock.h>
 #include <linux/slab.h>
 #include <linux/syscalls.h>
@@ -34,19 +33,19 @@ static struct posix_clock *get_posix_clock(struct file *fp)
 {
 	struct posix_clock *clk = fp->private_data;
 
-	mutex_lock(&clk->mutex);
+	down_read(&clk->rwsem);
 
 	if (!clk->zombie)
 		return clk;
 
-	mutex_unlock(&clk->mutex);
+	up_read(&clk->rwsem);
 
 	return NULL;
 }
 
 static void put_posix_clock(struct posix_clock *clk)
 {
-	mutex_unlock(&clk->mutex);
+	up_read(&clk->rwsem);
 }
 
 static ssize_t posix_clock_read(struct file *fp, char __user *buf,
@@ -156,7 +155,7 @@ static int posix_clock_open(struct inode *inode, struct file *fp)
 	struct posix_clock *clk =
 		container_of(inode->i_cdev, struct posix_clock, cdev);
 
-	mutex_lock(&clk->mutex);
+	down_read(&clk->rwsem);
 
 	if (clk->zombie) {
 		err = -ENODEV;
@@ -172,7 +171,7 @@ static int posix_clock_open(struct inode *inode, struct file *fp)
 		fp->private_data = clk;
 	}
 out:
-	mutex_unlock(&clk->mutex);
+	up_read(&clk->rwsem);
 	return err;
 }
 
@@ -211,25 +210,20 @@ int posix_clock_register(struct posix_clock *clk, dev_t devid)
 	int err;
 
 	kref_init(&clk->kref);
-	mutex_init(&clk->mutex);
+	init_rwsem(&clk->rwsem);
 
 	cdev_init(&clk->cdev, &posix_clock_file_operations);
 	clk->cdev.owner = clk->ops.owner;
 	err = cdev_add(&clk->cdev, devid, 1);
-	if (err)
-		goto no_cdev;
 
 	return err;
-no_cdev:
-	mutex_destroy(&clk->mutex);
-	return err;
 }
 EXPORT_SYMBOL_GPL(posix_clock_register);
 
 static void delete_clock(struct kref *kref)
 {
 	struct posix_clock *clk = container_of(kref, struct posix_clock, kref);
-	mutex_destroy(&clk->mutex);
+
 	if (clk->release)
 		clk->release(clk);
 }
@@ -238,9 +232,9 @@ void posix_clock_unregister(struct posix_clock *clk)
 {
 	cdev_del(&clk->cdev);
 
-	mutex_lock(&clk->mutex);
+	down_write(&clk->rwsem);
 	clk->zombie = true;
-	mutex_unlock(&clk->mutex);
+	up_write(&clk->rwsem);
 
 	kref_put(&clk->kref, delete_clock);
 }

^ permalink raw reply related	[flat|nested] 69+ messages in thread

* [GIT PULL] timer fixes
@ 2011-04-16 10:09 Ingo Molnar
  0 siblings, 0 replies; 69+ messages in thread
From: Ingo Molnar @ 2011-04-16 10:09 UTC (permalink / raw)
  To: Linus Torvalds; +Cc: linux-kernel, Thomas Gleixner, Andrew Morton

Linus,

Please pull the latest timer-fixes-for-linus git tree from:

   git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip.git timer-fixes-for-linus

 Thanks,

	Ingo

------------------>
Feng Tang (1):
      RTC: rtc-mrst: follow on to the change of rtc_device_register()

John Stultz (1):
      RTC: Fix early irqs caused by calling rtc_set_alarm too early

Mike Frysinger (1):
      RTC: add missing "return 0" in new alarm func for rtc-bfin.c

Vasily Khoruzhick (1):
      RTC: Fix s3c compile error due to missing s3c_rtc_setpie


 drivers/rtc/class.c     |    2 +-
 drivers/rtc/interface.c |   26 ++++++++++++++++++++++++++
 drivers/rtc/rtc-bfin.c  |    2 ++
 drivers/rtc/rtc-s3c.c   |    2 --
 include/linux/rtc.h     |    2 ++
 5 files changed, 31 insertions(+), 3 deletions(-)

diff --git a/drivers/rtc/class.c b/drivers/rtc/class.c
index 09b4437..3901386 100644
--- a/drivers/rtc/class.c
+++ b/drivers/rtc/class.c
@@ -171,7 +171,7 @@ struct rtc_device *rtc_device_register(const char *name, struct device *dev,
 	err = __rtc_read_alarm(rtc, &alrm);
 
 	if (!err && !rtc_valid_tm(&alrm.time))
-		rtc_set_alarm(rtc, &alrm);
+		rtc_initialize_alarm(rtc, &alrm);
 
 	strlcpy(rtc->name, name, RTC_DEVICE_NAME_SIZE);
 	dev_set_name(&rtc->dev, "rtc%d", id);
diff --git a/drivers/rtc/interface.c b/drivers/rtc/interface.c
index 23719f0..ef6316a 100644
--- a/drivers/rtc/interface.c
+++ b/drivers/rtc/interface.c
@@ -375,6 +375,32 @@ int rtc_set_alarm(struct rtc_device *rtc, struct rtc_wkalrm *alarm)
 }
 EXPORT_SYMBOL_GPL(rtc_set_alarm);
 
+/* Called once per device from rtc_device_register */
+int rtc_initialize_alarm(struct rtc_device *rtc, struct rtc_wkalrm *alarm)
+{
+	int err;
+
+	err = rtc_valid_tm(&alarm->time);
+	if (err != 0)
+		return err;
+
+	err = mutex_lock_interruptible(&rtc->ops_lock);
+	if (err)
+		return err;
+
+	rtc->aie_timer.node.expires = rtc_tm_to_ktime(alarm->time);
+	rtc->aie_timer.period = ktime_set(0, 0);
+	if (alarm->enabled) {
+		rtc->aie_timer.enabled = 1;
+		timerqueue_add(&rtc->timerqueue, &rtc->aie_timer.node);
+	}
+	mutex_unlock(&rtc->ops_lock);
+	return err;
+}
+EXPORT_SYMBOL_GPL(rtc_initialize_alarm);
+
+
+
 int rtc_alarm_irq_enable(struct rtc_device *rtc, unsigned int enabled)
 {
 	int err = mutex_lock_interruptible(&rtc->ops_lock);
diff --git a/drivers/rtc/rtc-bfin.c b/drivers/rtc/rtc-bfin.c
index a0fc4cf..90d8662 100644
--- a/drivers/rtc/rtc-bfin.c
+++ b/drivers/rtc/rtc-bfin.c
@@ -250,6 +250,8 @@ static int bfin_rtc_alarm_irq_enable(struct device *dev, unsigned int enabled)
 		bfin_rtc_int_set_alarm(rtc);
 	else
 		bfin_rtc_int_clear(~(RTC_ISTAT_ALARM | RTC_ISTAT_ALARM_DAY));
+
+	return 0;
 }
 
 static int bfin_rtc_read_time(struct device *dev, struct rtc_time *tm)
diff --git a/drivers/rtc/rtc-s3c.c b/drivers/rtc/rtc-s3c.c
index 7149649..b3466c4 100644
--- a/drivers/rtc/rtc-s3c.c
+++ b/drivers/rtc/rtc-s3c.c
@@ -336,7 +336,6 @@ static void s3c_rtc_release(struct device *dev)
 
 	/* do not clear AIE here, it may be needed for wake */
 
-	s3c_rtc_setpie(dev, 0);
 	free_irq(s3c_rtc_alarmno, rtc_dev);
 	free_irq(s3c_rtc_tickno, rtc_dev);
 }
@@ -408,7 +407,6 @@ static int __devexit s3c_rtc_remove(struct platform_device *dev)
 	platform_set_drvdata(dev, NULL);
 	rtc_device_unregister(rtc);
 
-	s3c_rtc_setpie(&dev->dev, 0);
 	s3c_rtc_setaie(&dev->dev, 0);
 
 	clk_disable(rtc_clk);
diff --git a/include/linux/rtc.h b/include/linux/rtc.h
index 2ca7e8a..877ece4 100644
--- a/include/linux/rtc.h
+++ b/include/linux/rtc.h
@@ -228,6 +228,8 @@ extern int rtc_read_alarm(struct rtc_device *rtc,
 			struct rtc_wkalrm *alrm);
 extern int rtc_set_alarm(struct rtc_device *rtc,
 				struct rtc_wkalrm *alrm);
+extern int rtc_initialize_alarm(struct rtc_device *rtc,
+				struct rtc_wkalrm *alrm);
 extern void rtc_update_irq(struct rtc_device *rtc,
 			unsigned long num, unsigned long events);
 

^ permalink raw reply related	[flat|nested] 69+ messages in thread

* [GIT PULL] timer fixes
@ 2011-04-07 17:42 Ingo Molnar
  0 siblings, 0 replies; 69+ messages in thread
From: Ingo Molnar @ 2011-04-07 17:42 UTC (permalink / raw)
  To: Linus Torvalds
  Cc: linux-kernel, Thomas Gleixner, Alan Cox, H. Peter Anvin, Andrew Morton

Linus,

Please pull the latest timers-fixes-for-linus git tree from:

   git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip.git timers-fixes-for-linus

 Thanks,

	Ingo

------------------>
Feng Tang (2):
      rtc, x86/mrst/vrtc: Fix boot crash in rtc_read_alarm()
      x86/mrst/vrtc: Fix boot crash in mrst_rtc_init()


 arch/x86/platform/mrst/vrtc.c |    4 +++-
 drivers/rtc/rtc-mrst.c        |    7 ++++---
 2 files changed, 7 insertions(+), 4 deletions(-)

diff --git a/arch/x86/platform/mrst/vrtc.c b/arch/x86/platform/mrst/vrtc.c
index 04cf645..73d70d6 100644
--- a/arch/x86/platform/mrst/vrtc.c
+++ b/arch/x86/platform/mrst/vrtc.c
@@ -100,9 +100,11 @@ int vrtc_set_mmss(unsigned long nowtime)
 
 void __init mrst_rtc_init(void)
 {
-	unsigned long vrtc_paddr = sfi_mrtc_array[0].phys_addr;
+	unsigned long vrtc_paddr;
 
 	sfi_table_parse(SFI_SIG_MRTC, NULL, NULL, sfi_parse_mrtc);
+
+	vrtc_paddr = sfi_mrtc_array[0].phys_addr;
 	if (!sfi_mrtc_num || !vrtc_paddr)
 		return;
 
diff --git a/drivers/rtc/rtc-mrst.c b/drivers/rtc/rtc-mrst.c
index 332a2c4..b2f0968 100644
--- a/drivers/rtc/rtc-mrst.c
+++ b/drivers/rtc/rtc-mrst.c
@@ -342,6 +342,8 @@ vrtc_mrst_do_probe(struct device *dev, struct resource *iomem, int rtc_irq)
 
 	mrst_rtc.irq = rtc_irq;
 	mrst_rtc.iomem = iomem;
+	mrst_rtc.dev = dev;
+	dev_set_drvdata(dev, &mrst_rtc);
 
 	mrst_rtc.rtc = rtc_device_register(driver_name, dev,
 				&mrst_rtc_ops, THIS_MODULE);
@@ -350,8 +352,6 @@ vrtc_mrst_do_probe(struct device *dev, struct resource *iomem, int rtc_irq)
 		goto cleanup0;
 	}
 
-	mrst_rtc.dev = dev;
-	dev_set_drvdata(dev, &mrst_rtc);
 	rename_region(iomem, dev_name(&mrst_rtc.rtc->dev));
 
 	spin_lock_irq(&rtc_lock);
@@ -376,9 +376,10 @@ vrtc_mrst_do_probe(struct device *dev, struct resource *iomem, int rtc_irq)
 	return 0;
 
 cleanup1:
-	mrst_rtc.dev = NULL;
 	rtc_device_unregister(mrst_rtc.rtc);
 cleanup0:
+	dev_set_drvdata(dev, NULL);
+	mrst_rtc.dev = NULL;
 	release_region(iomem->start, iomem->end + 1 - iomem->start);
 	dev_err(dev, "rtc-mrst: unable to initialise\n");
 	return retval;

^ permalink raw reply related	[flat|nested] 69+ messages in thread

* [GIT PULL] timer fixes
@ 2011-02-06 12:06 Ingo Molnar
  0 siblings, 0 replies; 69+ messages in thread
From: Ingo Molnar @ 2011-02-06 12:06 UTC (permalink / raw)
  To: Linus Torvalds
  Cc: linux-kernel, Thomas Gleixner, Peter Zijlstra, Andrew Morton

Linus,

Please pull the latest timers-fixes-for-linus git tree from:

   git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip.git timers-fixes-for-linus

 Thanks,

	Ingo

------------------>
Marcelo Roberto Jimenez (1):
      RTC: Prevents a division by zero in kernel code.

Peter Zijlstra (1):
      lockdep, timer: Fix del_timer_sync() annotation


 drivers/rtc/class.c     |    1 +
 drivers/rtc/interface.c |    3 +++
 kernel/timer.c          |    6 +++++-
 3 files changed, 9 insertions(+), 1 deletions(-)

diff --git a/drivers/rtc/class.c b/drivers/rtc/class.c
index 9583cbc..c404b61 100644
--- a/drivers/rtc/class.c
+++ b/drivers/rtc/class.c
@@ -143,6 +143,7 @@ struct rtc_device *rtc_device_register(const char *name, struct device *dev,
 	rtc->id = id;
 	rtc->ops = ops;
 	rtc->owner = owner;
+	rtc->irq_freq = 1;
 	rtc->max_user_freq = 64;
 	rtc->dev.parent = dev;
 	rtc->dev.class = rtc_class;
diff --git a/drivers/rtc/interface.c b/drivers/rtc/interface.c
index 925006d..a0c0196 100644
--- a/drivers/rtc/interface.c
+++ b/drivers/rtc/interface.c
@@ -464,6 +464,9 @@ int rtc_irq_set_freq(struct rtc_device *rtc, struct rtc_task *task, int freq)
 	int err = 0;
 	unsigned long flags;
 
+	if (freq <= 0)
+		return -EINVAL;
+
 	spin_lock_irqsave(&rtc->irq_task_lock, flags);
 	if (rtc->irq_task != NULL && task == NULL)
 		err = -EBUSY;
diff --git a/kernel/timer.c b/kernel/timer.c
index 43ca993..d53ce66 100644
--- a/kernel/timer.c
+++ b/kernel/timer.c
@@ -969,10 +969,14 @@ EXPORT_SYMBOL(try_to_del_timer_sync);
 int del_timer_sync(struct timer_list *timer)
 {
 #ifdef CONFIG_LOCKDEP
+	unsigned long flags;
+
+	raw_local_irq_save(flags);
 	local_bh_disable();
 	lock_map_acquire(&timer->lockdep_map);
 	lock_map_release(&timer->lockdep_map);
-	local_bh_enable();
+	_local_bh_enable();
+	raw_local_irq_restore(flags);
 #endif
 	/*
 	 * don't use it in hardirq context, because it

^ permalink raw reply related	[flat|nested] 69+ messages in thread

* [GIT PULL] timer fixes
@ 2011-01-24 13:24 Ingo Molnar
  0 siblings, 0 replies; 69+ messages in thread
From: Ingo Molnar @ 2011-01-24 13:24 UTC (permalink / raw)
  To: Linus Torvalds
  Cc: linux-kernel, Thomas Gleixner, Peter Zijlstra, Andrew Morton

Linus,

Please pull the latest timers-fixes-for-linus git tree from:

   git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip.git timers-fixes-for-linus

 Thanks,

	Ingo

------------------>
John Stultz (3):
      rtc: Cleanup removed UIE emulation declaration
      RTC: Propagate error handling via rtc_timer_enqueue properly
      RTC: Properly handle rtc_read_alarm error propagation and fix bug

Konrad Rzeszutek Wilk (1):
      acpi_pm: Clear pmtmr_ioport if acpi_pm initialization fails

Stephen Boyd (1):
      hrtimers: Notify hrtimer users of switches to NOHZ mode

Uwe Kleine-König (1):
      RTC: Remove Kconfig symbol for UIE emulation


 drivers/clocksource/acpi_pm.c |    6 +++-
 drivers/rtc/Kconfig           |   12 --------
 drivers/rtc/interface.c       |   61 +++++++++++++++++++++++++----------------
 include/linux/rtc.h           |    4 ---
 kernel/time/tick-sched.c      |    7 +++--
 5 files changed, 46 insertions(+), 44 deletions(-)

diff --git a/drivers/clocksource/acpi_pm.c b/drivers/clocksource/acpi_pm.c
index cfb0f52..effe797 100644
--- a/drivers/clocksource/acpi_pm.c
+++ b/drivers/clocksource/acpi_pm.c
@@ -202,17 +202,21 @@ static int __init init_acpi_pm_clocksource(void)
 			printk(KERN_INFO "PM-Timer had inconsistent results:"
 			       " 0x%#llx, 0x%#llx - aborting.\n",
 			       value1, value2);
+			pmtmr_ioport = 0;
 			return -EINVAL;
 		}
 		if (i == ACPI_PM_READ_CHECKS) {
 			printk(KERN_INFO "PM-Timer failed consistency check "
 			       " (0x%#llx) - aborting.\n", value1);
+			pmtmr_ioport = 0;
 			return -ENODEV;
 		}
 	}
 
-	if (verify_pmtmr_rate() != 0)
+	if (verify_pmtmr_rate() != 0){
+		pmtmr_ioport = 0;
 		return -ENODEV;
+	}
 
 	return clocksource_register_hz(&clocksource_acpi_pm,
 						PMTMR_TICKS_PER_SEC);
diff --git a/drivers/rtc/Kconfig b/drivers/rtc/Kconfig
index 4941cad..cdd9719 100644
--- a/drivers/rtc/Kconfig
+++ b/drivers/rtc/Kconfig
@@ -97,18 +97,6 @@ config RTC_INTF_DEV
 
 	  If unsure, say Y.
 
-config RTC_INTF_DEV_UIE_EMUL
-	bool "RTC UIE emulation on dev interface"
-	depends on RTC_INTF_DEV
-	help
-	  Provides an emulation for RTC_UIE if the underlying rtc chip
-	  driver does not expose RTC_UIE ioctls. Those requests generate
-	  once-per-second update interrupts, used for synchronization.
-
-	  The emulation code will read the time from the hardware
-	  clock several times per second, please enable this option
-	  only if you know that you really need it.
-
 config RTC_DRV_TEST
 	tristate "Test driver/device"
 	help
diff --git a/drivers/rtc/interface.c b/drivers/rtc/interface.c
index 90384b9..925006d 100644
--- a/drivers/rtc/interface.c
+++ b/drivers/rtc/interface.c
@@ -16,6 +16,9 @@
 #include <linux/log2.h>
 #include <linux/workqueue.h>
 
+static int rtc_timer_enqueue(struct rtc_device *rtc, struct rtc_timer *timer);
+static void rtc_timer_remove(struct rtc_device *rtc, struct rtc_timer *timer);
+
 static int __rtc_read_time(struct rtc_device *rtc, struct rtc_time *tm)
 {
 	int err;
@@ -120,12 +123,18 @@ int rtc_read_alarm(struct rtc_device *rtc, struct rtc_wkalrm *alarm)
 	err = mutex_lock_interruptible(&rtc->ops_lock);
 	if (err)
 		return err;
-	alarm->enabled = rtc->aie_timer.enabled;
-	if (alarm->enabled)
+	if (rtc->ops == NULL)
+		err = -ENODEV;
+	else if (!rtc->ops->read_alarm)
+		err = -EINVAL;
+	else {
+		memset(alarm, 0, sizeof(struct rtc_wkalrm));
+		alarm->enabled = rtc->aie_timer.enabled;
 		alarm->time = rtc_ktime_to_tm(rtc->aie_timer.node.expires);
+	}
 	mutex_unlock(&rtc->ops_lock);
 
-	return 0;
+	return err;
 }
 EXPORT_SYMBOL_GPL(rtc_read_alarm);
 
@@ -175,16 +184,14 @@ int rtc_set_alarm(struct rtc_device *rtc, struct rtc_wkalrm *alarm)
 		return err;
 	if (rtc->aie_timer.enabled) {
 		rtc_timer_remove(rtc, &rtc->aie_timer);
-		rtc->aie_timer.enabled = 0;
 	}
 	rtc->aie_timer.node.expires = rtc_tm_to_ktime(alarm->time);
 	rtc->aie_timer.period = ktime_set(0, 0);
 	if (alarm->enabled) {
-		rtc->aie_timer.enabled = 1;
-		rtc_timer_enqueue(rtc, &rtc->aie_timer);
+		err = rtc_timer_enqueue(rtc, &rtc->aie_timer);
 	}
 	mutex_unlock(&rtc->ops_lock);
-	return 0;
+	return err;
 }
 EXPORT_SYMBOL_GPL(rtc_set_alarm);
 
@@ -195,15 +202,15 @@ int rtc_alarm_irq_enable(struct rtc_device *rtc, unsigned int enabled)
 		return err;
 
 	if (rtc->aie_timer.enabled != enabled) {
-		if (enabled) {
-			rtc->aie_timer.enabled = 1;
-			rtc_timer_enqueue(rtc, &rtc->aie_timer);
-		} else {
+		if (enabled)
+			err = rtc_timer_enqueue(rtc, &rtc->aie_timer);
+		else
 			rtc_timer_remove(rtc, &rtc->aie_timer);
-			rtc->aie_timer.enabled = 0;
-		}
 	}
 
+	if (err)
+		return err;
+
 	if (!rtc->ops)
 		err = -ENODEV;
 	else if (!rtc->ops->alarm_irq_enable)
@@ -235,12 +242,9 @@ int rtc_update_irq_enable(struct rtc_device *rtc, unsigned int enabled)
 		now = rtc_tm_to_ktime(tm);
 		rtc->uie_rtctimer.node.expires = ktime_add(now, onesec);
 		rtc->uie_rtctimer.period = ktime_set(1, 0);
-		rtc->uie_rtctimer.enabled = 1;
-		rtc_timer_enqueue(rtc, &rtc->uie_rtctimer);
-	} else {
+		err = rtc_timer_enqueue(rtc, &rtc->uie_rtctimer);
+	} else
 		rtc_timer_remove(rtc, &rtc->uie_rtctimer);
-		rtc->uie_rtctimer.enabled = 0;
-	}
 
 out:
 	mutex_unlock(&rtc->ops_lock);
@@ -488,10 +492,13 @@ EXPORT_SYMBOL_GPL(rtc_irq_set_freq);
  * Enqueues a timer onto the rtc devices timerqueue and sets
  * the next alarm event appropriately.
  *
+ * Sets the enabled bit on the added timer.
+ *
  * Must hold ops_lock for proper serialization of timerqueue
  */
-void rtc_timer_enqueue(struct rtc_device *rtc, struct rtc_timer *timer)
+static int rtc_timer_enqueue(struct rtc_device *rtc, struct rtc_timer *timer)
 {
+	timer->enabled = 1;
 	timerqueue_add(&rtc->timerqueue, &timer->node);
 	if (&timer->node == timerqueue_getnext(&rtc->timerqueue)) {
 		struct rtc_wkalrm alarm;
@@ -501,7 +508,13 @@ void rtc_timer_enqueue(struct rtc_device *rtc, struct rtc_timer *timer)
 		err = __rtc_set_alarm(rtc, &alarm);
 		if (err == -ETIME)
 			schedule_work(&rtc->irqwork);
+		else if (err) {
+			timerqueue_del(&rtc->timerqueue, &timer->node);
+			timer->enabled = 0;
+			return err;
+		}
 	}
+	return 0;
 }
 
 /**
@@ -512,13 +525,15 @@ void rtc_timer_enqueue(struct rtc_device *rtc, struct rtc_timer *timer)
  * Removes a timer onto the rtc devices timerqueue and sets
  * the next alarm event appropriately.
  *
+ * Clears the enabled bit on the removed timer.
+ *
  * Must hold ops_lock for proper serialization of timerqueue
  */
-void rtc_timer_remove(struct rtc_device *rtc, struct rtc_timer *timer)
+static void rtc_timer_remove(struct rtc_device *rtc, struct rtc_timer *timer)
 {
 	struct timerqueue_node *next = timerqueue_getnext(&rtc->timerqueue);
 	timerqueue_del(&rtc->timerqueue, &timer->node);
-
+	timer->enabled = 0;
 	if (next == &timer->node) {
 		struct rtc_wkalrm alarm;
 		int err;
@@ -626,8 +641,7 @@ int rtc_timer_start(struct rtc_device *rtc, struct rtc_timer* timer,
 	timer->node.expires = expires;
 	timer->period = period;
 
-	timer->enabled = 1;
-	rtc_timer_enqueue(rtc, timer);
+	ret = rtc_timer_enqueue(rtc, timer);
 
 	mutex_unlock(&rtc->ops_lock);
 	return ret;
@@ -645,7 +659,6 @@ int rtc_timer_cancel(struct rtc_device *rtc, struct rtc_timer* timer)
 	mutex_lock(&rtc->ops_lock);
 	if (timer->enabled)
 		rtc_timer_remove(rtc, timer);
-	timer->enabled = 0;
 	mutex_unlock(&rtc->ops_lock);
 	return ret;
 }
diff --git a/include/linux/rtc.h b/include/linux/rtc.h
index 3c995b4..a0b639f 100644
--- a/include/linux/rtc.h
+++ b/include/linux/rtc.h
@@ -235,8 +235,6 @@ extern int rtc_irq_set_freq(struct rtc_device *rtc,
 				struct rtc_task *task, int freq);
 extern int rtc_update_irq_enable(struct rtc_device *rtc, unsigned int enabled);
 extern int rtc_alarm_irq_enable(struct rtc_device *rtc, unsigned int enabled);
-extern int rtc_dev_update_irq_enable_emul(struct rtc_device *rtc,
-						unsigned int enabled);
 
 void rtc_aie_update_irq(void *private);
 void rtc_uie_update_irq(void *private);
@@ -246,8 +244,6 @@ int rtc_register(rtc_task_t *task);
 int rtc_unregister(rtc_task_t *task);
 int rtc_control(rtc_task_t *t, unsigned int cmd, unsigned long arg);
 
-void rtc_timer_enqueue(struct rtc_device *rtc, struct rtc_timer *timer);
-void rtc_timer_remove(struct rtc_device *rtc, struct rtc_timer *timer);
 void rtc_timer_init(struct rtc_timer *timer, void (*f)(void* p), void* data);
 int rtc_timer_start(struct rtc_device *rtc, struct rtc_timer* timer,
 			ktime_t expires, ktime_t period);
diff --git a/kernel/time/tick-sched.c b/kernel/time/tick-sched.c
index 3e216e0..c55ea24 100644
--- a/kernel/time/tick-sched.c
+++ b/kernel/time/tick-sched.c
@@ -642,8 +642,7 @@ static void tick_nohz_switch_to_nohz(void)
 	}
 	local_irq_enable();
 
-	printk(KERN_INFO "Switched to NOHz mode on CPU #%d\n",
-	       smp_processor_id());
+	printk(KERN_INFO "Switched to NOHz mode on CPU #%d\n", smp_processor_id());
 }
 
 /*
@@ -795,8 +794,10 @@ void tick_setup_sched_timer(void)
 	}
 
 #ifdef CONFIG_NO_HZ
-	if (tick_nohz_enabled)
+	if (tick_nohz_enabled) {
 		ts->nohz_mode = NOHZ_MODE_HIGHRES;
+		printk(KERN_INFO "Switched to NOHz mode on CPU #%d\n", smp_processor_id());
+	}
 #endif
 }
 #endif /* HIGH_RES_TIMERS */

^ permalink raw reply related	[flat|nested] 69+ messages in thread

* [GIT PULL] timer fixes
@ 2010-11-26 13:40 Ingo Molnar
  0 siblings, 0 replies; 69+ messages in thread
From: Ingo Molnar @ 2010-11-26 13:40 UTC (permalink / raw)
  To: Linus Torvalds
  Cc: linux-kernel, Thomas Gleixner, Peter Zijlstra, Andrew Morton

Linus,

Please pull the latest timers-fixes-for-linus git tree from:

   git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip.git timers-fixes-for-linus

 Thanks,

	Ingo

------------------>
Sergey Senozhatsky (1):
      posix-cpu-timers: Rcu_read_lock/unlock protect find_task_by_vpid call


 kernel/posix-cpu-timers.c |   12 ++++++------
 1 files changed, 6 insertions(+), 6 deletions(-)

diff --git a/kernel/posix-cpu-timers.c b/kernel/posix-cpu-timers.c
index 6842eeb..05bb717 100644
--- a/kernel/posix-cpu-timers.c
+++ b/kernel/posix-cpu-timers.c
@@ -37,13 +37,13 @@ static int check_clock(const clockid_t which_clock)
 	if (pid == 0)
 		return 0;
 
-	read_lock(&tasklist_lock);
+	rcu_read_lock();
 	p = find_task_by_vpid(pid);
 	if (!p || !(CPUCLOCK_PERTHREAD(which_clock) ?
-		   same_thread_group(p, current) : thread_group_leader(p))) {
+		   same_thread_group(p, current) : has_group_leader_pid(p))) {
 		error = -EINVAL;
 	}
-	read_unlock(&tasklist_lock);
+	rcu_read_unlock();
 
 	return error;
 }
@@ -390,7 +390,7 @@ int posix_cpu_timer_create(struct k_itimer *new_timer)
 
 	INIT_LIST_HEAD(&new_timer->it.cpu.entry);
 
-	read_lock(&tasklist_lock);
+	rcu_read_lock();
 	if (CPUCLOCK_PERTHREAD(new_timer->it_clock)) {
 		if (pid == 0) {
 			p = current;
@@ -404,7 +404,7 @@ int posix_cpu_timer_create(struct k_itimer *new_timer)
 			p = current->group_leader;
 		} else {
 			p = find_task_by_vpid(pid);
-			if (p && !thread_group_leader(p))
+			if (p && !has_group_leader_pid(p))
 				p = NULL;
 		}
 	}
@@ -414,7 +414,7 @@ int posix_cpu_timer_create(struct k_itimer *new_timer)
 	} else {
 		ret = -EINVAL;
 	}
-	read_unlock(&tasklist_lock);
+	rcu_read_unlock();
 
 	return ret;
 }

^ permalink raw reply related	[flat|nested] 69+ messages in thread

* [GIT PULL] timer fixes
@ 2010-03-26 15:34 Ingo Molnar
  0 siblings, 0 replies; 69+ messages in thread
From: Ingo Molnar @ 2010-03-26 15:34 UTC (permalink / raw)
  To: Linus Torvalds
  Cc: linux-kernel, Thomas Gleixner, Peter Zijlstra, Andrew Morton

Linus,

Please pull the latest timers-fixes-for-linus git tree from:

   git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip.git timers-fixes-for-linus

 Thanks,

	Ingo

------------------>
Andrew Morton (1):
      timer stats: Fix del_timer_sync() and try_to_del_timer_sync()

John Stultz (1):
      time: Fix accumulation bug triggered by long delay.

Stanislaw Gruszka (1):
      posix-cpu-timers: Reset expire cache when no timer is running

Thomas Gleixner (1):
      clockevents: Sanitize min_delta_ns adjustment and prevent overflows


 include/linux/clockchips.h |    2 +
 kernel/posix-cpu-timers.c  |   10 ++++++--
 kernel/time/tick-oneshot.c |   52 +++++++++++++++++++++++++++++++++----------
 kernel/time/timekeeping.c  |    3 +-
 kernel/time/timer_list.c   |    3 +-
 kernel/timer.c             |    1 +
 6 files changed, 54 insertions(+), 17 deletions(-)

diff --git a/include/linux/clockchips.h b/include/linux/clockchips.h
index 0cf725b..fc53492 100644
--- a/include/linux/clockchips.h
+++ b/include/linux/clockchips.h
@@ -73,6 +73,7 @@ enum clock_event_nofitiers {
  * @list:		list head for the management code
  * @mode:		operating mode assigned by the management code
  * @next_event:		local storage for the next event in oneshot mode
+ * @retries:		number of forced programming retries
  */
 struct clock_event_device {
 	const char		*name;
@@ -93,6 +94,7 @@ struct clock_event_device {
 	struct list_head	list;
 	enum clock_event_mode	mode;
 	ktime_t			next_event;
+	unsigned long		retries;
 };
 
 /*
diff --git a/kernel/posix-cpu-timers.c b/kernel/posix-cpu-timers.c
index 438ff45..edec25a 100644
--- a/kernel/posix-cpu-timers.c
+++ b/kernel/posix-cpu-timers.c
@@ -1060,9 +1060,9 @@ static void check_thread_timers(struct task_struct *tsk,
 	}
 }
 
-static void stop_process_timers(struct task_struct *tsk)
+static void stop_process_timers(struct signal_struct *sig)
 {
-	struct thread_group_cputimer *cputimer = &tsk->signal->cputimer;
+	struct thread_group_cputimer *cputimer = &sig->cputimer;
 	unsigned long flags;
 
 	if (!cputimer->running)
@@ -1071,6 +1071,10 @@ static void stop_process_timers(struct task_struct *tsk)
 	spin_lock_irqsave(&cputimer->lock, flags);
 	cputimer->running = 0;
 	spin_unlock_irqrestore(&cputimer->lock, flags);
+
+	sig->cputime_expires.prof_exp = cputime_zero;
+	sig->cputime_expires.virt_exp = cputime_zero;
+	sig->cputime_expires.sched_exp = 0;
 }
 
 static u32 onecputick;
@@ -1131,7 +1135,7 @@ static void check_process_timers(struct task_struct *tsk,
 	    list_empty(&timers[CPUCLOCK_VIRT]) &&
 	    cputime_eq(sig->it[CPUCLOCK_VIRT].expires, cputime_zero) &&
 	    list_empty(&timers[CPUCLOCK_SCHED])) {
-		stop_process_timers(tsk);
+		stop_process_timers(sig);
 		return;
 	}
 
diff --git a/kernel/time/tick-oneshot.c b/kernel/time/tick-oneshot.c
index 0a8a213..aada0e5 100644
--- a/kernel/time/tick-oneshot.c
+++ b/kernel/time/tick-oneshot.c
@@ -22,6 +22,29 @@
 
 #include "tick-internal.h"
 
+/* Limit min_delta to a jiffie */
+#define MIN_DELTA_LIMIT		(NSEC_PER_SEC / HZ)
+
+static int tick_increase_min_delta(struct clock_event_device *dev)
+{
+	/* Nothing to do if we already reached the limit */
+	if (dev->min_delta_ns >= MIN_DELTA_LIMIT)
+		return -ETIME;
+
+	if (dev->min_delta_ns < 5000)
+		dev->min_delta_ns = 5000;
+	else
+		dev->min_delta_ns += dev->min_delta_ns >> 1;
+
+	if (dev->min_delta_ns > MIN_DELTA_LIMIT)
+		dev->min_delta_ns = MIN_DELTA_LIMIT;
+
+	printk(KERN_WARNING "CE: %s increased min_delta_ns to %llu nsec\n",
+	       dev->name ? dev->name : "?",
+	       (unsigned long long) dev->min_delta_ns);
+	return 0;
+}
+
 /**
  * tick_program_event internal worker function
  */
@@ -37,23 +60,28 @@ int tick_dev_program_event(struct clock_event_device *dev, ktime_t expires,
 		if (!ret || !force)
 			return ret;
 
+		dev->retries++;
 		/*
-		 * We tried 2 times to program the device with the given
-		 * min_delta_ns. If that's not working then we double it
+		 * We tried 3 times to program the device with the given
+		 * min_delta_ns. If that's not working then we increase it
 		 * and emit a warning.
 		 */
 		if (++i > 2) {
 			/* Increase the min. delta and try again */
-			if (!dev->min_delta_ns)
-				dev->min_delta_ns = 5000;
-			else
-				dev->min_delta_ns += dev->min_delta_ns >> 1;
-
-			printk(KERN_WARNING
-			       "CE: %s increasing min_delta_ns to %llu nsec\n",
-			       dev->name ? dev->name : "?",
-			       (unsigned long long) dev->min_delta_ns << 1);
-
+			if (tick_increase_min_delta(dev)) {
+				/*
+				 * Get out of the loop if min_delta_ns
+				 * hit the limit already. That's
+				 * better than staying here forever.
+				 *
+				 * We clear next_event so we have a
+				 * chance that the box survives.
+				 */
+				printk(KERN_WARNING
+				       "CE: Reprogramming failure. Giving up\n");
+				dev->next_event.tv64 = KTIME_MAX;
+				return -ETIME;
+			}
 			i = 0;
 		}
 
diff --git a/kernel/time/timekeeping.c b/kernel/time/timekeeping.c
index 1673637..39f6177 100644
--- a/kernel/time/timekeeping.c
+++ b/kernel/time/timekeeping.c
@@ -818,7 +818,8 @@ void update_wall_time(void)
 	shift = min(shift, maxshift);
 	while (offset >= timekeeper.cycle_interval) {
 		offset = logarithmic_accumulation(offset, shift);
-		shift--;
+		if(offset < timekeeper.cycle_interval<<shift)
+			shift--;
 	}
 
 	/* correct the clock when NTP error is too big */
diff --git a/kernel/time/timer_list.c b/kernel/time/timer_list.c
index bdfb8dd..1a4a7dd 100644
--- a/kernel/time/timer_list.c
+++ b/kernel/time/timer_list.c
@@ -228,6 +228,7 @@ print_tickdevice(struct seq_file *m, struct tick_device *td, int cpu)
 	SEQ_printf(m, " event_handler:  ");
 	print_name_offset(m, dev->event_handler);
 	SEQ_printf(m, "\n");
+	SEQ_printf(m, " retries:        %lu\n", dev->retries);
 }
 
 static void timer_list_show_tickdevices(struct seq_file *m)
@@ -257,7 +258,7 @@ static int timer_list_show(struct seq_file *m, void *v)
 	u64 now = ktime_to_ns(ktime_get());
 	int cpu;
 
-	SEQ_printf(m, "Timer List Version: v0.5\n");
+	SEQ_printf(m, "Timer List Version: v0.6\n");
 	SEQ_printf(m, "HRTIMER_MAX_CLOCK_BASES: %d\n", HRTIMER_MAX_CLOCK_BASES);
 	SEQ_printf(m, "now at %Ld nsecs\n", (unsigned long long)now);
 
diff --git a/kernel/timer.c b/kernel/timer.c
index c61a794..fc965ea 100644
--- a/kernel/timer.c
+++ b/kernel/timer.c
@@ -880,6 +880,7 @@ int try_to_del_timer_sync(struct timer_list *timer)
 	if (base->running_timer == timer)
 		goto out;
 
+	timer_stats_timer_clear_start_info(timer);
 	ret = 0;
 	if (timer_pending(timer)) {
 		detach_timer(timer, 1);

^ permalink raw reply related	[flat|nested] 69+ messages in thread

* [GIT PULL] timer fixes
@ 2009-12-18 18:49 Ingo Molnar
  0 siblings, 0 replies; 69+ messages in thread
From: Ingo Molnar @ 2009-12-18 18:49 UTC (permalink / raw)
  To: Linus Torvalds
  Cc: linux-kernel, Thomas Gleixner, Peter Zijlstra, Andrew Morton

Linus,

Please pull the latest timers-fixes-for-linus git tree from:

   git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip.git timers-fixes-for-linus

 Thanks,

	Ingo

------------------>
Simon Horman (1):
      timers: Remove duplicate setting of new_base in __mod_timer()

Thomas Gleixner (1):
      clockevents: Prevent clockevent_devices list corruption on cpu hotplug


 kernel/time/clockevents.c |   18 +++++++++++++++---
 kernel/timer.c            |    2 --
 2 files changed, 15 insertions(+), 5 deletions(-)

diff --git a/kernel/time/clockevents.c b/kernel/time/clockevents.c
index 20a8920..91db2e3 100644
--- a/kernel/time/clockevents.c
+++ b/kernel/time/clockevents.c
@@ -238,8 +238,9 @@ void clockevents_exchange_device(struct clock_event_device *old,
  */
 void clockevents_notify(unsigned long reason, void *arg)
 {
-	struct list_head *node, *tmp;
+	struct clock_event_device *dev, *tmp;
 	unsigned long flags;
+	int cpu;
 
 	spin_lock_irqsave(&clockevents_lock, flags);
 	clockevents_do_notify(reason, arg);
@@ -250,8 +251,19 @@ void clockevents_notify(unsigned long reason, void *arg)
 		 * Unregister the clock event devices which were
 		 * released from the users in the notify chain.
 		 */
-		list_for_each_safe(node, tmp, &clockevents_released)
-			list_del(node);
+		list_for_each_entry_safe(dev, tmp, &clockevents_released, list)
+			list_del(&dev->list);
+		/*
+		 * Now check whether the CPU has left unused per cpu devices
+		 */
+		cpu = *((int *)arg);
+		list_for_each_entry_safe(dev, tmp, &clockevent_devices, list) {
+			if (cpumask_test_cpu(cpu, dev->cpumask) &&
+			    cpumask_weight(dev->cpumask) == 1) {
+				BUG_ON(dev->mode != CLOCK_EVT_MODE_UNUSED);
+				list_del(&dev->list);
+			}
+		}
 		break;
 	default:
 		break;
diff --git a/kernel/timer.c b/kernel/timer.c
index 5db5a8d..15533b7 100644
--- a/kernel/timer.c
+++ b/kernel/timer.c
@@ -656,8 +656,6 @@ __mod_timer(struct timer_list *timer, unsigned long expires,
 
 	debug_activate(timer, expires);
 
-	new_base = __get_cpu_var(tvec_bases);
-
 	cpu = smp_processor_id();
 
 #if defined(CONFIG_NO_HZ) && defined(CONFIG_SMP)

^ permalink raw reply related	[flat|nested] 69+ messages in thread

* [GIT PULL] timer fixes
@ 2009-10-08 18:35 Ingo Molnar
  0 siblings, 0 replies; 69+ messages in thread
From: Ingo Molnar @ 2009-10-08 18:35 UTC (permalink / raw)
  To: Linus Torvalds
  Cc: linux-kernel, Thomas Gleixner, Peter Zijlstra, Arjan van de Ven

Linus,

Please pull the latest timers-fixes-for-linus git tree from:

   git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip.git timers-fixes-for-linus

 Thanks,

	Ingo

------------------>
Arjan van de Ven (1):
      x86, timers: Check for pending timers after (device) interrupts

Eero Nurkkala (1):
      NOHZ: update idle state also when NOHZ is inactive


 arch/x86/kernel/irq.c    |    2 ++
 arch/x86/kernel/smp.c    |    1 +
 kernel/time/tick-sched.c |    9 +++++++--
 3 files changed, 10 insertions(+), 2 deletions(-)

diff --git a/arch/x86/kernel/irq.c b/arch/x86/kernel/irq.c
index 74656d1..3912061 100644
--- a/arch/x86/kernel/irq.c
+++ b/arch/x86/kernel/irq.c
@@ -244,6 +244,7 @@ unsigned int __irq_entry do_IRQ(struct pt_regs *regs)
 				__func__, smp_processor_id(), vector, irq);
 	}
 
+	run_local_timers();
 	irq_exit();
 
 	set_irq_regs(old_regs);
@@ -268,6 +269,7 @@ void smp_generic_interrupt(struct pt_regs *regs)
 	if (generic_interrupt_extension)
 		generic_interrupt_extension();
 
+	run_local_timers();
 	irq_exit();
 
 	set_irq_regs(old_regs);
diff --git a/arch/x86/kernel/smp.c b/arch/x86/kernel/smp.c
index ec1de97..d915d95 100644
--- a/arch/x86/kernel/smp.c
+++ b/arch/x86/kernel/smp.c
@@ -198,6 +198,7 @@ void smp_reschedule_interrupt(struct pt_regs *regs)
 {
 	ack_APIC_irq();
 	inc_irq_stat(irq_resched_count);
+	run_local_timers();
 	/*
 	 * KVM uses this interrupt to force a cpu out of guest mode
 	 */
diff --git a/kernel/time/tick-sched.c b/kernel/time/tick-sched.c
index e0f59a2..89aed59 100644
--- a/kernel/time/tick-sched.c
+++ b/kernel/time/tick-sched.c
@@ -231,6 +231,13 @@ void tick_nohz_stop_sched_tick(int inidle)
 	if (!inidle && !ts->inidle)
 		goto end;
 
+	/*
+	 * Set ts->inidle unconditionally. Even if the system did not
+	 * switch to NOHZ mode the cpu frequency governers rely on the
+	 * update of the idle time accounting in tick_nohz_start_idle().
+	 */
+	ts->inidle = 1;
+
 	now = tick_nohz_start_idle(ts);
 
 	/*
@@ -248,8 +255,6 @@ void tick_nohz_stop_sched_tick(int inidle)
 	if (unlikely(ts->nohz_mode == NOHZ_MODE_INACTIVE))
 		goto end;
 
-	ts->inidle = 1;
-
 	if (need_resched())
 		goto end;
 

^ permalink raw reply related	[flat|nested] 69+ messages in thread

* [GIT PULL] timer fixes
@ 2009-08-25 18:04 Ingo Molnar
  0 siblings, 0 replies; 69+ messages in thread
From: Ingo Molnar @ 2009-08-25 18:04 UTC (permalink / raw)
  To: Linus Torvalds; +Cc: linux-kernel, Thomas Gleixner, Peter Zijlstra

Linus,

Please pull the latest timers-fixes-for-linus git tree from:

   git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip.git timers-fixes-for-linus

 Thanks,

	Ingo

------------------>
Amerigo Wang (1):
      timers: Drop write permission on /proc/timer_list

Suresh Siddha (1):
      clockevent: Prevent dead lock on clockevents_lock


 arch/x86/kernel/process.c     |    6 +-----
 drivers/acpi/processor_idle.c |    6 ++++--
 kernel/time/clockevents.c     |   16 ++++++++++------
 kernel/time/tick-broadcast.c  |    7 +++----
 kernel/time/timer_list.c      |    2 +-
 5 files changed, 19 insertions(+), 18 deletions(-)

diff --git a/arch/x86/kernel/process.c b/arch/x86/kernel/process.c
index 994dd6a..071166a 100644
--- a/arch/x86/kernel/process.c
+++ b/arch/x86/kernel/process.c
@@ -519,16 +519,12 @@ static void c1e_idle(void)
 		if (!cpumask_test_cpu(cpu, c1e_mask)) {
 			cpumask_set_cpu(cpu, c1e_mask);
 			/*
-			 * Force broadcast so ACPI can not interfere. Needs
-			 * to run with interrupts enabled as it uses
-			 * smp_function_call.
+			 * Force broadcast so ACPI can not interfere.
 			 */
-			local_irq_enable();
 			clockevents_notify(CLOCK_EVT_NOTIFY_BROADCAST_FORCE,
 					   &cpu);
 			printk(KERN_INFO "Switch to broadcast mode on CPU%d\n",
 			       cpu);
-			local_irq_disable();
 		}
 		clockevents_notify(CLOCK_EVT_NOTIFY_BROADCAST_ENTER, &cpu);
 
diff --git a/drivers/acpi/processor_idle.c b/drivers/acpi/processor_idle.c
index 0efa59e..66393d5 100644
--- a/drivers/acpi/processor_idle.c
+++ b/drivers/acpi/processor_idle.c
@@ -162,8 +162,9 @@ static void lapic_timer_check_state(int state, struct acpi_processor *pr,
 		pr->power.timer_broadcast_on_state = state;
 }
 
-static void lapic_timer_propagate_broadcast(struct acpi_processor *pr)
+static void lapic_timer_propagate_broadcast(void *arg)
 {
+	struct acpi_processor *pr = (struct acpi_processor *) arg;
 	unsigned long reason;
 
 	reason = pr->power.timer_broadcast_on_state < INT_MAX ?
@@ -635,7 +636,8 @@ static int acpi_processor_power_verify(struct acpi_processor *pr)
 		working++;
 	}
 
-	lapic_timer_propagate_broadcast(pr);
+	smp_call_function_single(pr->id, lapic_timer_propagate_broadcast,
+				 pr, 1);
 
 	return (working);
 }
diff --git a/kernel/time/clockevents.c b/kernel/time/clockevents.c
index a6dcd67..620b58a 100644
--- a/kernel/time/clockevents.c
+++ b/kernel/time/clockevents.c
@@ -137,11 +137,12 @@ int clockevents_program_event(struct clock_event_device *dev, ktime_t expires,
  */
 int clockevents_register_notifier(struct notifier_block *nb)
 {
+	unsigned long flags;
 	int ret;
 
-	spin_lock(&clockevents_lock);
+	spin_lock_irqsave(&clockevents_lock, flags);
 	ret = raw_notifier_chain_register(&clockevents_chain, nb);
-	spin_unlock(&clockevents_lock);
+	spin_unlock_irqrestore(&clockevents_lock, flags);
 
 	return ret;
 }
@@ -178,16 +179,18 @@ static void clockevents_notify_released(void)
  */
 void clockevents_register_device(struct clock_event_device *dev)
 {
+	unsigned long flags;
+
 	BUG_ON(dev->mode != CLOCK_EVT_MODE_UNUSED);
 	BUG_ON(!dev->cpumask);
 
-	spin_lock(&clockevents_lock);
+	spin_lock_irqsave(&clockevents_lock, flags);
 
 	list_add(&dev->list, &clockevent_devices);
 	clockevents_do_notify(CLOCK_EVT_NOTIFY_ADD, dev);
 	clockevents_notify_released();
 
-	spin_unlock(&clockevents_lock);
+	spin_unlock_irqrestore(&clockevents_lock, flags);
 }
 EXPORT_SYMBOL_GPL(clockevents_register_device);
 
@@ -235,8 +238,9 @@ void clockevents_exchange_device(struct clock_event_device *old,
 void clockevents_notify(unsigned long reason, void *arg)
 {
 	struct list_head *node, *tmp;
+	unsigned long flags;
 
-	spin_lock(&clockevents_lock);
+	spin_lock_irqsave(&clockevents_lock, flags);
 	clockevents_do_notify(reason, arg);
 
 	switch (reason) {
@@ -251,7 +255,7 @@ void clockevents_notify(unsigned long reason, void *arg)
 	default:
 		break;
 	}
-	spin_unlock(&clockevents_lock);
+	spin_unlock_irqrestore(&clockevents_lock, flags);
 }
 EXPORT_SYMBOL_GPL(clockevents_notify);
 #endif
diff --git a/kernel/time/tick-broadcast.c b/kernel/time/tick-broadcast.c
index 877dbed..c2ec250 100644
--- a/kernel/time/tick-broadcast.c
+++ b/kernel/time/tick-broadcast.c
@@ -205,11 +205,11 @@ static void tick_handle_periodic_broadcast(struct clock_event_device *dev)
  * Powerstate information: The system enters/leaves a state, where
  * affected devices might stop
  */
-static void tick_do_broadcast_on_off(void *why)
+static void tick_do_broadcast_on_off(unsigned long *reason)
 {
 	struct clock_event_device *bc, *dev;
 	struct tick_device *td;
-	unsigned long flags, *reason = why;
+	unsigned long flags;
 	int cpu, bc_stopped;
 
 	spin_lock_irqsave(&tick_broadcast_lock, flags);
@@ -276,8 +276,7 @@ void tick_broadcast_on_off(unsigned long reason, int *oncpu)
 		printk(KERN_ERR "tick-broadcast: ignoring broadcast for "
 		       "offline CPU #%d\n", *oncpu);
 	else
-		smp_call_function_single(*oncpu, tick_do_broadcast_on_off,
-					 &reason, 1);
+		tick_do_broadcast_on_off(&reason);
 }
 
 /*
diff --git a/kernel/time/timer_list.c b/kernel/time/timer_list.c
index a999b92..fddd69d 100644
--- a/kernel/time/timer_list.c
+++ b/kernel/time/timer_list.c
@@ -286,7 +286,7 @@ static int __init init_timer_list_procfs(void)
 {
 	struct proc_dir_entry *pe;
 
-	pe = proc_create("timer_list", 0644, NULL, &timer_list_fops);
+	pe = proc_create("timer_list", 0444, NULL, &timer_list_fops);
 	if (!pe)
 		return -ENOMEM;
 	return 0;

^ permalink raw reply related	[flat|nested] 69+ messages in thread

* [GIT PULL] timer fixes
@ 2009-06-26 19:03 Ingo Molnar
  0 siblings, 0 replies; 69+ messages in thread
From: Ingo Molnar @ 2009-06-26 19:03 UTC (permalink / raw)
  To: Linus Torvalds; +Cc: linux-kernel, Thomas Gleixner

Linus,

Please pull the latest timers-fixes-for-linus git tree from:

   git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip.git timers-fixes-for-linus

The timer stats performance fix seemed significant (and simple) 
enough for a post-rc1 merge as it's enabled in a number of distros 
by default - but can leave it out if it's too much.

 Thanks,

	Ingo

------------------>
Arun R Bharadwaj (1):
      timers: Fix timer_migration interface which accepts any number as input

Heiko Carstens (1):
      timer stats: Optimize by adding quick check to avoid function calls


 include/linux/hrtimer.h   |    5 +++++
 include/linux/timer.h     |    4 ++++
 kernel/sysctl.c           |    5 ++++-
 kernel/time/timer_stats.c |   16 ++++++++--------
 kernel/timer.c            |    2 ++
 5 files changed, 23 insertions(+), 9 deletions(-)

diff --git a/include/linux/hrtimer.h b/include/linux/hrtimer.h
index 7400900..54648e6 100644
--- a/include/linux/hrtimer.h
+++ b/include/linux/hrtimer.h
@@ -21,6 +21,7 @@
 #include <linux/list.h>
 #include <linux/wait.h>
 #include <linux/percpu.h>
+#include <linux/timer.h>
 
 
 struct hrtimer_clock_base;
@@ -447,6 +448,8 @@ extern void timer_stats_update_stats(void *timer, pid_t pid, void *startf,
 
 static inline void timer_stats_account_hrtimer(struct hrtimer *timer)
 {
+	if (likely(!timer->start_pid))
+		return;
 	timer_stats_update_stats(timer, timer->start_pid, timer->start_site,
 				 timer->function, timer->start_comm, 0);
 }
@@ -456,6 +459,8 @@ extern void __timer_stats_hrtimer_set_start_info(struct hrtimer *timer,
 
 static inline void timer_stats_hrtimer_set_start_info(struct hrtimer *timer)
 {
+	if (likely(!timer_stats_active))
+		return;
 	__timer_stats_hrtimer_set_start_info(timer, __builtin_return_address(0));
 }
 
diff --git a/include/linux/timer.h b/include/linux/timer.h
index ccf882e..be62ec2 100644
--- a/include/linux/timer.h
+++ b/include/linux/timer.h
@@ -190,6 +190,8 @@ extern unsigned long get_next_timer_interrupt(unsigned long now);
  */
 #ifdef CONFIG_TIMER_STATS
 
+extern int timer_stats_active;
+
 #define TIMER_STATS_FLAG_DEFERRABLE	0x1
 
 extern void init_timer_stats(void);
@@ -203,6 +205,8 @@ extern void __timer_stats_timer_set_start_info(struct timer_list *timer,
 
 static inline void timer_stats_timer_set_start_info(struct timer_list *timer)
 {
+	if (likely(!timer_stats_active))
+		return;
 	__timer_stats_timer_set_start_info(timer, __builtin_return_address(0));
 }
 
diff --git a/kernel/sysctl.c b/kernel/sysctl.c
index 62e4ff9..c428ba1 100644
--- a/kernel/sysctl.c
+++ b/kernel/sysctl.c
@@ -335,7 +335,10 @@ static struct ctl_table kern_table[] = {
 		.data		= &sysctl_timer_migration,
 		.maxlen		= sizeof(unsigned int),
 		.mode		= 0644,
-		.proc_handler	= &proc_dointvec,
+		.proc_handler	= &proc_dointvec_minmax,
+		.strategy	= &sysctl_intvec,
+		.extra1		= &zero,
+		.extra2		= &one,
 	},
 #endif
 	{
diff --git a/kernel/time/timer_stats.c b/kernel/time/timer_stats.c
index c994530..4cde8b9 100644
--- a/kernel/time/timer_stats.c
+++ b/kernel/time/timer_stats.c
@@ -96,7 +96,7 @@ static DEFINE_MUTEX(show_mutex);
 /*
  * Collection status, active/inactive:
  */
-static int __read_mostly active;
+int __read_mostly timer_stats_active;
 
 /*
  * Beginning/end timestamps of measurement:
@@ -242,7 +242,7 @@ void timer_stats_update_stats(void *timer, pid_t pid, void *startf,
 	struct entry *entry, input;
 	unsigned long flags;
 
-	if (likely(!active))
+	if (likely(!timer_stats_active))
 		return;
 
 	lock = &per_cpu(lookup_lock, raw_smp_processor_id());
@@ -254,7 +254,7 @@ void timer_stats_update_stats(void *timer, pid_t pid, void *startf,
 	input.timer_flag = timer_flag;
 
 	spin_lock_irqsave(lock, flags);
-	if (!active)
+	if (!timer_stats_active)
 		goto out_unlock;
 
 	entry = tstat_lookup(&input, comm);
@@ -290,7 +290,7 @@ static int tstats_show(struct seq_file *m, void *v)
 	/*
 	 * If still active then calculate up to now:
 	 */
-	if (active)
+	if (timer_stats_active)
 		time_stop = ktime_get();
 
 	time = ktime_sub(time_stop, time_start);
@@ -368,18 +368,18 @@ static ssize_t tstats_write(struct file *file, const char __user *buf,
 	mutex_lock(&show_mutex);
 	switch (ctl[0]) {
 	case '0':
-		if (active) {
-			active = 0;
+		if (timer_stats_active) {
+			timer_stats_active = 0;
 			time_stop = ktime_get();
 			sync_access();
 		}
 		break;
 	case '1':
-		if (!active) {
+		if (!timer_stats_active) {
 			reset_entries();
 			time_start = ktime_get();
 			smp_mb();
-			active = 1;
+			timer_stats_active = 1;
 		}
 		break;
 	default:
diff --git a/kernel/timer.c b/kernel/timer.c
index 54d3912..0b36b9e 100644
--- a/kernel/timer.c
+++ b/kernel/timer.c
@@ -380,6 +380,8 @@ static void timer_stats_account_timer(struct timer_list *timer)
 {
 	unsigned int flag = 0;
 
+	if (likely(!timer->start_site))
+		return;
 	if (unlikely(tbase_get_deferrable(timer->base)))
 		flag |= TIMER_STATS_FLAG_DEFERRABLE;
 

^ permalink raw reply related	[flat|nested] 69+ messages in thread

* [GIT PULL] timer fixes
@ 2009-04-09 15:43 Ingo Molnar
  0 siblings, 0 replies; 69+ messages in thread
From: Ingo Molnar @ 2009-04-09 15:43 UTC (permalink / raw)
  To: Linus Torvalds; +Cc: linux-kernel, Andrew Morton, Thomas Gleixner

Linus,

Please pull the latest timers-fixes-for-linus git tree from:

   git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip.git timers-fixes-for-linus

 Thanks,

	Ingo

------------------>
Oleg Nesterov (2):
      posix-timers: fix RLIMIT_CPU && fork()
      posix-timers: fix RLIMIT_CPU && setitimer(CPUCLOCK_PROF)

Randy Dunlap (1):
      timers: add missing kernel-doc


 kernel/fork.c             |   13 +++++++++----
 kernel/posix-cpu-timers.c |    2 +-
 kernel/timer.c            |    7 +++++--
 3 files changed, 15 insertions(+), 7 deletions(-)

diff --git a/kernel/fork.c b/kernel/fork.c
index 4854c2c..9b51a1b 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -808,6 +808,12 @@ static void posix_cpu_timers_init_group(struct signal_struct *sig)
 	sig->cputime_expires.virt_exp = cputime_zero;
 	sig->cputime_expires.sched_exp = 0;
 
+	if (sig->rlim[RLIMIT_CPU].rlim_cur != RLIM_INFINITY) {
+		sig->cputime_expires.prof_exp =
+			secs_to_cputime(sig->rlim[RLIMIT_CPU].rlim_cur);
+		sig->cputimer.running = 1;
+	}
+
 	/* The timer lists. */
 	INIT_LIST_HEAD(&sig->cpu_timers[0]);
 	INIT_LIST_HEAD(&sig->cpu_timers[1]);
@@ -823,11 +829,8 @@ static int copy_signal(unsigned long clone_flags, struct task_struct *tsk)
 		atomic_inc(&current->signal->live);
 		return 0;
 	}
-	sig = kmem_cache_alloc(signal_cachep, GFP_KERNEL);
-
-	if (sig)
-		posix_cpu_timers_init_group(sig);
 
+	sig = kmem_cache_alloc(signal_cachep, GFP_KERNEL);
 	tsk->signal = sig;
 	if (!sig)
 		return -ENOMEM;
@@ -865,6 +868,8 @@ static int copy_signal(unsigned long clone_flags, struct task_struct *tsk)
 	memcpy(sig->rlim, current->signal->rlim, sizeof sig->rlim);
 	task_unlock(current->group_leader);
 
+	posix_cpu_timers_init_group(sig);
+
 	acct_init_pacct(&sig->pacct);
 
 	tty_audit_fork(sig);
diff --git a/kernel/posix-cpu-timers.c b/kernel/posix-cpu-timers.c
index 8e5d9a6..bb53185 100644
--- a/kernel/posix-cpu-timers.c
+++ b/kernel/posix-cpu-timers.c
@@ -18,7 +18,7 @@ void update_rlimit_cpu(unsigned long rlim_new)
 
 	cputime = secs_to_cputime(rlim_new);
 	if (cputime_eq(current->signal->it_prof_expires, cputime_zero) ||
-	    cputime_lt(current->signal->it_prof_expires, cputime)) {
+	    cputime_gt(current->signal->it_prof_expires, cputime)) {
 		spin_lock_irq(&current->sighand->siglock);
 		set_process_cpu_timer(current, CPUCLOCK_PROF, &cputime, NULL);
 		spin_unlock_irq(&current->sighand->siglock);
diff --git a/kernel/timer.c b/kernel/timer.c
index 9b77fc9..3af9a0b 100644
--- a/kernel/timer.c
+++ b/kernel/timer.c
@@ -524,10 +524,13 @@ static void __init_timer(struct timer_list *timer)
 }
 
 /**
- * init_timer - initialize a timer.
+ * init_timer_key - initialize a timer
  * @timer: the timer to be initialized
+ * @name: name of the timer
+ * @key: lockdep class key of the fake lock used for tracking timer
+ *       sync lock dependencies
  *
- * init_timer() must be done to a timer prior calling *any* of the
+ * init_timer_key() must be done to a timer prior calling *any* of the
  * other timer functions.
  */
 void init_timer(struct timer_list *timer)

^ permalink raw reply related	[flat|nested] 69+ messages in thread

* [git pull] timer fixes
@ 2009-01-30 23:07 Ingo Molnar
  0 siblings, 0 replies; 69+ messages in thread
From: Ingo Molnar @ 2009-01-30 23:07 UTC (permalink / raw)
  To: Linus Torvalds; +Cc: linux-kernel, Andrew Morton, Thomas Gleixner

Linus,

Please pull the latest timers-fixes-for-linus git tree from:

   git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip.git timers-fixes-for-linus

Special mention for this commit:

      hrtimers: increase clock min delta threshold while interrupt hanging

We found it useful in the case where a very slow (or very emulated) system 
is overwhelmed with HZ irqs - we get a kernel message and some throttling, 
instead of a hard lockup in essence. If that kind of approach is not 
acceptable we can remove this patch and resend the other fixes.

 Thanks,

	Ingo

------------------>
Frederic Weisbecker (1):
      hrtimers: increase clock min delta threshold while interrupt hanging

Sebastien Dugue (1):
      hrtimers: allow the hot-unplugging of all cpus

Thomas Gleixner (1):
      hrtimer: prevent negative expiry value after clock_was_set()


 include/linux/clockchips.h |    1 +
 kernel/hrtimer.c           |   41 ++++++++++++++++++++++++++++++++++++++++-
 kernel/time/tick-common.c  |   26 +++++++++++++++++++-------
 3 files changed, 60 insertions(+), 8 deletions(-)

diff --git a/include/linux/clockchips.h b/include/linux/clockchips.h
index cea1536..3a1dbba 100644
--- a/include/linux/clockchips.h
+++ b/include/linux/clockchips.h
@@ -36,6 +36,7 @@ enum clock_event_nofitiers {
 	CLOCK_EVT_NOTIFY_BROADCAST_EXIT,
 	CLOCK_EVT_NOTIFY_SUSPEND,
 	CLOCK_EVT_NOTIFY_RESUME,
+	CLOCK_EVT_NOTIFY_CPU_DYING,
 	CLOCK_EVT_NOTIFY_CPU_DEAD,
 };
 
diff --git a/kernel/hrtimer.c b/kernel/hrtimer.c
index f33afb0..f394d2a 100644
--- a/kernel/hrtimer.c
+++ b/kernel/hrtimer.c
@@ -501,6 +501,13 @@ static void hrtimer_force_reprogram(struct hrtimer_cpu_base *cpu_base)
 			continue;
 		timer = rb_entry(base->first, struct hrtimer, node);
 		expires = ktime_sub(hrtimer_get_expires(timer), base->offset);
+		/*
+		 * clock_was_set() has changed base->offset so the
+		 * result might be negative. Fix it up to prevent a
+		 * false positive in clockevents_program_event()
+		 */
+		if (expires.tv64 < 0)
+			expires.tv64 = 0;
 		if (expires.tv64 < cpu_base->expires_next.tv64)
 			cpu_base->expires_next = expires;
 	}
@@ -1158,6 +1165,29 @@ static void __run_hrtimer(struct hrtimer *timer)
 
 #ifdef CONFIG_HIGH_RES_TIMERS
 
+static int force_clock_reprogram;
+
+/*
+ * After 5 iteration's attempts, we consider that hrtimer_interrupt()
+ * is hanging, which could happen with something that slows the interrupt
+ * such as the tracing. Then we force the clock reprogramming for each future
+ * hrtimer interrupts to avoid infinite loops and use the min_delta_ns
+ * threshold that we will overwrite.
+ * The next tick event will be scheduled to 3 times we currently spend on
+ * hrtimer_interrupt(). This gives a good compromise, the cpus will spend
+ * 1/4 of their time to process the hrtimer interrupts. This is enough to
+ * let it running without serious starvation.
+ */
+
+static inline void
+hrtimer_interrupt_hanging(struct clock_event_device *dev,
+			ktime_t try_time)
+{
+	force_clock_reprogram = 1;
+	dev->min_delta_ns = (unsigned long)try_time.tv64 * 3;
+	printk(KERN_WARNING "hrtimer: interrupt too slow, "
+		"forcing clock min delta to %lu ns\n", dev->min_delta_ns);
+}
 /*
  * High resolution timer interrupt
  * Called with interrupts disabled
@@ -1167,6 +1197,7 @@ void hrtimer_interrupt(struct clock_event_device *dev)
 	struct hrtimer_cpu_base *cpu_base = &__get_cpu_var(hrtimer_bases);
 	struct hrtimer_clock_base *base;
 	ktime_t expires_next, now;
+	int nr_retries = 0;
 	int i;
 
 	BUG_ON(!cpu_base->hres_active);
@@ -1174,6 +1205,10 @@ void hrtimer_interrupt(struct clock_event_device *dev)
 	dev->next_event.tv64 = KTIME_MAX;
 
  retry:
+	/* 5 retries is enough to notice a hang */
+	if (!(++nr_retries % 5))
+		hrtimer_interrupt_hanging(dev, ktime_sub(ktime_get(), now));
+
 	now = ktime_get();
 
 	expires_next.tv64 = KTIME_MAX;
@@ -1226,7 +1261,7 @@ void hrtimer_interrupt(struct clock_event_device *dev)
 
 	/* Reprogramming necessary ? */
 	if (expires_next.tv64 != KTIME_MAX) {
-		if (tick_program_event(expires_next, 0))
+		if (tick_program_event(expires_next, force_clock_reprogram))
 			goto retry;
 	}
 }
@@ -1580,6 +1615,10 @@ static int __cpuinit hrtimer_cpu_notify(struct notifier_block *self,
 		break;
 
 #ifdef CONFIG_HOTPLUG_CPU
+	case CPU_DYING:
+	case CPU_DYING_FROZEN:
+		clockevents_notify(CLOCK_EVT_NOTIFY_CPU_DYING, &scpu);
+		break;
 	case CPU_DEAD:
 	case CPU_DEAD_FROZEN:
 	{
diff --git a/kernel/time/tick-common.c b/kernel/time/tick-common.c
index 63e05d4..21a5ca8 100644
--- a/kernel/time/tick-common.c
+++ b/kernel/time/tick-common.c
@@ -274,6 +274,21 @@ out_bc:
 }
 
 /*
+ * Transfer the do_timer job away from a dying cpu.
+ *
+ * Called with interrupts disabled.
+ */
+static void tick_handover_do_timer(int *cpup)
+{
+	if (*cpup == tick_do_timer_cpu) {
+		int cpu = cpumask_first(cpu_online_mask);
+
+		tick_do_timer_cpu = (cpu < nr_cpu_ids) ? cpu :
+			TICK_DO_TIMER_NONE;
+	}
+}
+
+/*
  * Shutdown an event device on a given cpu:
  *
  * This is called on a life CPU, when a CPU is dead. So we cannot
@@ -297,13 +312,6 @@ static void tick_shutdown(unsigned int *cpup)
 		clockevents_exchange_device(dev, NULL);
 		td->evtdev = NULL;
 	}
-	/* Transfer the do_timer job away from this cpu */
-	if (*cpup == tick_do_timer_cpu) {
-		int cpu = cpumask_first(cpu_online_mask);
-
-		tick_do_timer_cpu = (cpu < nr_cpu_ids) ? cpu :
-			TICK_DO_TIMER_NONE;
-	}
 	spin_unlock_irqrestore(&tick_device_lock, flags);
 }
 
@@ -357,6 +365,10 @@ static int tick_notify(struct notifier_block *nb, unsigned long reason,
 		tick_broadcast_oneshot_control(reason);
 		break;
 
+	case CLOCK_EVT_NOTIFY_CPU_DYING:
+		tick_handover_do_timer(dev);
+		break;
+
 	case CLOCK_EVT_NOTIFY_CPU_DEAD:
 		tick_shutdown_broadcast_oneshot(dev);
 		tick_shutdown_broadcast(dev);

^ permalink raw reply related	[flat|nested] 69+ messages in thread

* [git pull] timer fixes
@ 2009-01-26 17:21 Ingo Molnar
  0 siblings, 0 replies; 69+ messages in thread
From: Ingo Molnar @ 2009-01-26 17:21 UTC (permalink / raw)
  To: Linus Torvalds
  Cc: linux-kernel, Andrew Morton, Thomas Gleixner, Peter Zijlstra

Linus,

Please pull the latest timers-fixes-for-linus git tree from:

   git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip.git timers-fixes-for-linus

 Thanks,

	Ingo

------------------>
Jaswinder Singh Rajput (1):
      time-sched.c: tick_nohz_update_jiffies should be static

Peter Zijlstra (3):
      itimers: remove the per-cpu-ish-ness
      locking, hpet: annotate false positive warning
      hrtimers: fix inconsistent lock state on resume in hres_timers_resume

Steven Noonan (1):
      kernel/fork.c: unused variable 'ret'


 arch/x86/kernel/hpet.c    |    2 +-
 include/linux/init_task.h |    6 ++++
 include/linux/sched.h     |   29 +++++++++++-------
 include/linux/workqueue.h |    6 ++++
 kernel/fork.c             |   16 ++++------
 kernel/hrtimer.c          |    4 ++-
 kernel/posix-cpu-timers.c |   70 ---------------------------------------------
 kernel/sched_stats.h      |   33 +++++++++-----------
 kernel/time/tick-sched.c  |    2 +-
 9 files changed, 57 insertions(+), 111 deletions(-)

diff --git a/arch/x86/kernel/hpet.c b/arch/x86/kernel/hpet.c
index cd759ad..bb2e0f0 100644
--- a/arch/x86/kernel/hpet.c
+++ b/arch/x86/kernel/hpet.c
@@ -628,7 +628,7 @@ static int hpet_cpuhp_notify(struct notifier_block *n,
 
 	switch (action & 0xf) {
 	case CPU_ONLINE:
-		INIT_DELAYED_WORK(&work.work, hpet_work);
+		INIT_DELAYED_WORK_ON_STACK(&work.work, hpet_work);
 		init_completion(&work.complete);
 		/* FIXME: add schedule_work_on() */
 		schedule_delayed_work_on(cpu, &work.work, 0);
diff --git a/include/linux/init_task.h b/include/linux/init_task.h
index 2f3c2d4..ea0ea1a 100644
--- a/include/linux/init_task.h
+++ b/include/linux/init_task.h
@@ -48,6 +48,12 @@ extern struct fs_struct init_fs;
 	.posix_timers	 = LIST_HEAD_INIT(sig.posix_timers),		\
 	.cpu_timers	= INIT_CPU_TIMERS(sig.cpu_timers),		\
 	.rlim		= INIT_RLIMITS,					\
+	.cputime	= { .totals = {					\
+		.utime = cputime_zero,					\
+		.stime = cputime_zero,					\
+		.sum_exec_runtime = 0,					\
+		.lock = __SPIN_LOCK_UNLOCKED(sig.cputime.totals.lock),	\
+	}, },								\
 }
 
 extern struct nsproxy init_nsproxy;
diff --git a/include/linux/sched.h b/include/linux/sched.h
index 4cae9b8..c20943e 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -450,6 +450,7 @@ struct task_cputime {
 	cputime_t utime;
 	cputime_t stime;
 	unsigned long long sum_exec_runtime;
+	spinlock_t lock;
 };
 /* Alternate field names when used to cache expirations. */
 #define prof_exp	stime
@@ -465,7 +466,7 @@ struct task_cputime {
  * used for thread group CPU clock calculations.
  */
 struct thread_group_cputime {
-	struct task_cputime *totals;
+	struct task_cputime totals;
 };
 
 /*
@@ -2180,24 +2181,30 @@ static inline int spin_needbreak(spinlock_t *lock)
  * Thread group CPU time accounting.
  */
 
-extern int thread_group_cputime_alloc(struct task_struct *);
-extern void thread_group_cputime(struct task_struct *, struct task_cputime *);
-
-static inline void thread_group_cputime_init(struct signal_struct *sig)
+static inline
+void thread_group_cputime(struct task_struct *tsk, struct task_cputime *times)
 {
-	sig->cputime.totals = NULL;
+	struct task_cputime *totals = &tsk->signal->cputime.totals;
+	unsigned long flags;
+
+	spin_lock_irqsave(&totals->lock, flags);
+	*times = *totals;
+	spin_unlock_irqrestore(&totals->lock, flags);
 }
 
-static inline int thread_group_cputime_clone_thread(struct task_struct *curr)
+static inline void thread_group_cputime_init(struct signal_struct *sig)
 {
-	if (curr->signal->cputime.totals)
-		return 0;
-	return thread_group_cputime_alloc(curr);
+	sig->cputime.totals = (struct task_cputime){
+		.utime = cputime_zero,
+		.stime = cputime_zero,
+		.sum_exec_runtime = 0,
+	};
+
+	spin_lock_init(&sig->cputime.totals.lock);
 }
 
 static inline void thread_group_cputime_free(struct signal_struct *sig)
 {
-	free_percpu(sig->cputime.totals);
 }
 
 /*
diff --git a/include/linux/workqueue.h b/include/linux/workqueue.h
index b362911..47151c8 100644
--- a/include/linux/workqueue.h
+++ b/include/linux/workqueue.h
@@ -118,6 +118,12 @@ struct execute_work {
 		init_timer(&(_work)->timer);			\
 	} while (0)
 
+#define INIT_DELAYED_WORK_ON_STACK(_work, _func)		\
+	do {							\
+		INIT_WORK(&(_work)->work, (_func));		\
+		init_timer_on_stack(&(_work)->timer);		\
+	} while (0)
+
 #define INIT_DELAYED_WORK_DEFERRABLE(_work, _func)			\
 	do {							\
 		INIT_WORK(&(_work)->work, (_func));		\
diff --git a/kernel/fork.c b/kernel/fork.c
index 1d68f12..81da4aa 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -817,17 +817,17 @@ static void posix_cpu_timers_init_group(struct signal_struct *sig)
 static int copy_signal(unsigned long clone_flags, struct task_struct *tsk)
 {
 	struct signal_struct *sig;
-	int ret;
 
 	if (clone_flags & CLONE_THREAD) {
-		ret = thread_group_cputime_clone_thread(current);
-		if (likely(!ret)) {
-			atomic_inc(&current->signal->count);
-			atomic_inc(&current->signal->live);
-		}
-		return ret;
+		atomic_inc(&current->signal->count);
+		atomic_inc(&current->signal->live);
+		return 0;
 	}
 	sig = kmem_cache_alloc(signal_cachep, GFP_KERNEL);
+
+	if (sig)
+		posix_cpu_timers_init_group(sig);
+
 	tsk->signal = sig;
 	if (!sig)
 		return -ENOMEM;
@@ -864,8 +864,6 @@ static int copy_signal(unsigned long clone_flags, struct task_struct *tsk)
 	memcpy(sig->rlim, current->signal->rlim, sizeof sig->rlim);
 	task_unlock(current->group_leader);
 
-	posix_cpu_timers_init_group(sig);
-
 	acct_init_pacct(&sig->pacct);
 
 	tty_audit_fork(sig);
diff --git a/kernel/hrtimer.c b/kernel/hrtimer.c
index 1455b76..cb83c6d 100644
--- a/kernel/hrtimer.c
+++ b/kernel/hrtimer.c
@@ -614,7 +614,9 @@ void clock_was_set(void)
  */
 void hres_timers_resume(void)
 {
-	/* Retrigger the CPU local events: */
+	WARN_ONCE(!irqs_disabled(),
+		  KERN_INFO "hres_timers_resume() called with IRQs enabled!");
+
 	retrigger_next_event(NULL);
 }
 
diff --git a/kernel/posix-cpu-timers.c b/kernel/posix-cpu-timers.c
index 157de3a..fa07da9 100644
--- a/kernel/posix-cpu-timers.c
+++ b/kernel/posix-cpu-timers.c
@@ -10,76 +10,6 @@
 #include <linux/kernel_stat.h>
 
 /*
- * Allocate the thread_group_cputime structure appropriately and fill in the
- * current values of the fields.  Called from copy_signal() via
- * thread_group_cputime_clone_thread() when adding a second or subsequent
- * thread to a thread group.  Assumes interrupts are enabled when called.
- */
-int thread_group_cputime_alloc(struct task_struct *tsk)
-{
-	struct signal_struct *sig = tsk->signal;
-	struct task_cputime *cputime;
-
-	/*
-	 * If we have multiple threads and we don't already have a
-	 * per-CPU task_cputime struct (checked in the caller), allocate
-	 * one and fill it in with the times accumulated so far.  We may
-	 * race with another thread so recheck after we pick up the sighand
-	 * lock.
-	 */
-	cputime = alloc_percpu(struct task_cputime);
-	if (cputime == NULL)
-		return -ENOMEM;
-	spin_lock_irq(&tsk->sighand->siglock);
-	if (sig->cputime.totals) {
-		spin_unlock_irq(&tsk->sighand->siglock);
-		free_percpu(cputime);
-		return 0;
-	}
-	sig->cputime.totals = cputime;
-	cputime = per_cpu_ptr(sig->cputime.totals, smp_processor_id());
-	cputime->utime = tsk->utime;
-	cputime->stime = tsk->stime;
-	cputime->sum_exec_runtime = tsk->se.sum_exec_runtime;
-	spin_unlock_irq(&tsk->sighand->siglock);
-	return 0;
-}
-
-/**
- * thread_group_cputime - Sum the thread group time fields across all CPUs.
- *
- * @tsk:	The task we use to identify the thread group.
- * @times:	task_cputime structure in which we return the summed fields.
- *
- * Walk the list of CPUs to sum the per-CPU time fields in the thread group
- * time structure.
- */
-void thread_group_cputime(
-	struct task_struct *tsk,
-	struct task_cputime *times)
-{
-	struct task_cputime *totals, *tot;
-	int i;
-
-	totals = tsk->signal->cputime.totals;
-	if (!totals) {
-		times->utime = tsk->utime;
-		times->stime = tsk->stime;
-		times->sum_exec_runtime = tsk->se.sum_exec_runtime;
-		return;
-	}
-
-	times->stime = times->utime = cputime_zero;
-	times->sum_exec_runtime = 0;
-	for_each_possible_cpu(i) {
-		tot = per_cpu_ptr(totals, i);
-		times->utime = cputime_add(times->utime, tot->utime);
-		times->stime = cputime_add(times->stime, tot->stime);
-		times->sum_exec_runtime += tot->sum_exec_runtime;
-	}
-}
-
-/*
  * Called after updating RLIMIT_CPU to set timer expiration if necessary.
  */
 void update_rlimit_cpu(unsigned long rlim_new)
diff --git a/kernel/sched_stats.h b/kernel/sched_stats.h
index f2773b5..8ab0cef 100644
--- a/kernel/sched_stats.h
+++ b/kernel/sched_stats.h
@@ -296,6 +296,7 @@ sched_info_switch(struct task_struct *prev, struct task_struct *next)
 static inline void account_group_user_time(struct task_struct *tsk,
 					   cputime_t cputime)
 {
+	struct task_cputime *times;
 	struct signal_struct *sig;
 
 	/* tsk == current, ensure it is safe to use ->signal */
@@ -303,13 +304,11 @@ static inline void account_group_user_time(struct task_struct *tsk,
 		return;
 
 	sig = tsk->signal;
-	if (sig->cputime.totals) {
-		struct task_cputime *times;
+	times = &sig->cputime.totals;
 
-		times = per_cpu_ptr(sig->cputime.totals, get_cpu());
-		times->utime = cputime_add(times->utime, cputime);
-		put_cpu_no_resched();
-	}
+	spin_lock(&times->lock);
+	times->utime = cputime_add(times->utime, cputime);
+	spin_unlock(&times->lock);
 }
 
 /**
@@ -325,6 +324,7 @@ static inline void account_group_user_time(struct task_struct *tsk,
 static inline void account_group_system_time(struct task_struct *tsk,
 					     cputime_t cputime)
 {
+	struct task_cputime *times;
 	struct signal_struct *sig;
 
 	/* tsk == current, ensure it is safe to use ->signal */
@@ -332,13 +332,11 @@ static inline void account_group_system_time(struct task_struct *tsk,
 		return;
 
 	sig = tsk->signal;
-	if (sig->cputime.totals) {
-		struct task_cputime *times;
+	times = &sig->cputime.totals;
 
-		times = per_cpu_ptr(sig->cputime.totals, get_cpu());
-		times->stime = cputime_add(times->stime, cputime);
-		put_cpu_no_resched();
-	}
+	spin_lock(&times->lock);
+	times->stime = cputime_add(times->stime, cputime);
+	spin_unlock(&times->lock);
 }
 
 /**
@@ -354,6 +352,7 @@ static inline void account_group_system_time(struct task_struct *tsk,
 static inline void account_group_exec_runtime(struct task_struct *tsk,
 					      unsigned long long ns)
 {
+	struct task_cputime *times;
 	struct signal_struct *sig;
 
 	sig = tsk->signal;
@@ -362,11 +361,9 @@ static inline void account_group_exec_runtime(struct task_struct *tsk,
 	if (unlikely(!sig))
 		return;
 
-	if (sig->cputime.totals) {
-		struct task_cputime *times;
+	times = &sig->cputime.totals;
 
-		times = per_cpu_ptr(sig->cputime.totals, get_cpu());
-		times->sum_exec_runtime += ns;
-		put_cpu_no_resched();
-	}
+	spin_lock(&times->lock);
+	times->sum_exec_runtime += ns;
+	spin_unlock(&times->lock);
 }
diff --git a/kernel/time/tick-sched.c b/kernel/time/tick-sched.c
index 1b6c05b..d3f1ef4 100644
--- a/kernel/time/tick-sched.c
+++ b/kernel/time/tick-sched.c
@@ -134,7 +134,7 @@ __setup("nohz=", setup_tick_nohz);
  * value. We do this unconditionally on any cpu, as we don't know whether the
  * cpu, which has the update task assigned is in a long sleep.
  */
-void tick_nohz_update_jiffies(void)
+static void tick_nohz_update_jiffies(void)
 {
 	int cpu = smp_processor_id();
 	struct tick_sched *ts = &per_cpu(tick_cpu_sched, cpu);

^ permalink raw reply related	[flat|nested] 69+ messages in thread

* [git pull] timer fixes
@ 2009-01-06 16:18 Ingo Molnar
  0 siblings, 0 replies; 69+ messages in thread
From: Ingo Molnar @ 2009-01-06 16:18 UTC (permalink / raw)
  To: Linus Torvalds
  Cc: linux-kernel, Andrew Morton, Thomas Gleixner, Peter Zijlstra

Linus,

Please pull the latest timers-fixes-for-linus git tree from:

   git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip.git timers-fixes-for-linus

 Thanks,

	Ingo

------------------>
Ingo Molnar (1):
      hrtimer: splitout peek ahead functionality, fix

Peter Zijlstra (1):
      hrtimer: fix recursion deadlock by re-introducing the softirq

Thomas Gleixner (4):
      hrtimer: splitout peek ahead functionality
      hrtimer: fix HOTPLUG_CPU=n compile warning
      hrtimer: simplify hotplug migration
      hrtimer: fixup comments


 include/linux/interrupt.h |    3 +-
 kernel/hrtimer.c          |  142 ++++++++++++++++++++++-----------------------
 2 files changed, 72 insertions(+), 73 deletions(-)

diff --git a/include/linux/interrupt.h b/include/linux/interrupt.h
index 0702c4d..2062833 100644
--- a/include/linux/interrupt.h
+++ b/include/linux/interrupt.h
@@ -253,7 +253,8 @@ enum
 	BLOCK_SOFTIRQ,
 	TASKLET_SOFTIRQ,
 	SCHED_SOFTIRQ,
-	RCU_SOFTIRQ, 	/* Preferable RCU should always be the last softirq */
+	HRTIMER_SOFTIRQ,
+	RCU_SOFTIRQ,	/* Preferable RCU should always be the last softirq */
 
 	NR_SOFTIRQS
 };
diff --git a/kernel/hrtimer.c b/kernel/hrtimer.c
index eb2bfef..1455b76 100644
--- a/kernel/hrtimer.c
+++ b/kernel/hrtimer.c
@@ -634,7 +634,6 @@ static inline void hrtimer_init_timer_hres(struct hrtimer *timer)
 {
 }
 
-static void __run_hrtimer(struct hrtimer *timer);
 
 /*
  * When High resolution timers are active, try to reprogram. Note, that in case
@@ -646,13 +645,9 @@ static inline int hrtimer_enqueue_reprogram(struct hrtimer *timer,
 					    struct hrtimer_clock_base *base)
 {
 	if (base->cpu_base->hres_active && hrtimer_reprogram(timer, base)) {
-		/*
-		 * XXX: recursion check?
-		 * hrtimer_forward() should round up with timer granularity
-		 * so that we never get into inf recursion here,
-		 * it doesn't do that though
-		 */
-		__run_hrtimer(timer);
+		spin_unlock(&base->cpu_base->lock);
+		raise_softirq_irqoff(HRTIMER_SOFTIRQ);
+		spin_lock(&base->cpu_base->lock);
 		return 1;
 	}
 	return 0;
@@ -705,11 +700,6 @@ static inline int hrtimer_enqueue_reprogram(struct hrtimer *timer,
 }
 static inline void hrtimer_init_hres(struct hrtimer_cpu_base *base) { }
 static inline void hrtimer_init_timer_hres(struct hrtimer *timer) { }
-static inline int hrtimer_reprogram(struct hrtimer *timer,
-				    struct hrtimer_clock_base *base)
-{
-	return 0;
-}
 
 #endif /* CONFIG_HIGH_RES_TIMERS */
 
@@ -780,9 +770,11 @@ EXPORT_SYMBOL_GPL(hrtimer_forward);
  *
  * The timer is inserted in expiry order. Insertion into the
  * red black tree is O(log(n)). Must hold the base lock.
+ *
+ * Returns 1 when the new timer is the leftmost timer in the tree.
  */
-static void enqueue_hrtimer(struct hrtimer *timer,
-			    struct hrtimer_clock_base *base, int reprogram)
+static int enqueue_hrtimer(struct hrtimer *timer,
+			   struct hrtimer_clock_base *base)
 {
 	struct rb_node **link = &base->active.rb_node;
 	struct rb_node *parent = NULL;
@@ -814,20 +806,8 @@ static void enqueue_hrtimer(struct hrtimer *timer,
 	 * Insert the timer to the rbtree and check whether it
 	 * replaces the first pending timer
 	 */
-	if (leftmost) {
-		/*
-		 * Reprogram the clock event device. When the timer is already
-		 * expired hrtimer_enqueue_reprogram has either called the
-		 * callback or added it to the pending list and raised the
-		 * softirq.
-		 *
-		 * This is a NOP for !HIGHRES
-		 */
-		if (reprogram && hrtimer_enqueue_reprogram(timer, base))
-			return;
-
+	if (leftmost)
 		base->first = &timer->node;
-	}
 
 	rb_link_node(&timer->node, parent, link);
 	rb_insert_color(&timer->node, &base->active);
@@ -836,6 +816,8 @@ static void enqueue_hrtimer(struct hrtimer *timer,
 	 * state of a possibly running callback.
 	 */
 	timer->state |= HRTIMER_STATE_ENQUEUED;
+
+	return leftmost;
 }
 
 /*
@@ -912,7 +894,7 @@ hrtimer_start_range_ns(struct hrtimer *timer, ktime_t tim, unsigned long delta_n
 {
 	struct hrtimer_clock_base *base, *new_base;
 	unsigned long flags;
-	int ret;
+	int ret, leftmost;
 
 	base = lock_hrtimer_base(timer, &flags);
 
@@ -940,12 +922,16 @@ hrtimer_start_range_ns(struct hrtimer *timer, ktime_t tim, unsigned long delta_n
 
 	timer_stats_hrtimer_set_start_info(timer);
 
+	leftmost = enqueue_hrtimer(timer, new_base);
+
 	/*
 	 * Only allow reprogramming if the new base is on this CPU.
 	 * (it might still be on another CPU if the timer was pending)
+	 *
+	 * XXX send_remote_softirq() ?
 	 */
-	enqueue_hrtimer(timer, new_base,
-			new_base->cpu_base == &__get_cpu_var(hrtimer_bases));
+	if (leftmost && new_base->cpu_base == &__get_cpu_var(hrtimer_bases))
+		hrtimer_enqueue_reprogram(timer, new_base);
 
 	unlock_hrtimer_base(timer, &flags);
 
@@ -1157,13 +1143,13 @@ static void __run_hrtimer(struct hrtimer *timer)
 	spin_lock(&cpu_base->lock);
 
 	/*
-	 * Note: We clear the CALLBACK bit after enqueue_hrtimer to avoid
-	 * reprogramming of the event hardware. This happens at the end of this
-	 * function anyway.
+	 * Note: We clear the CALLBACK bit after enqueue_hrtimer and
+	 * we do not reprogramm the event hardware. Happens either in
+	 * hrtimer_start_range_ns() or in hrtimer_interrupt()
 	 */
 	if (restart != HRTIMER_NORESTART) {
 		BUG_ON(timer->state != HRTIMER_STATE_CALLBACK);
-		enqueue_hrtimer(timer, base, 0);
+		enqueue_hrtimer(timer, base);
 	}
 	timer->state &= ~HRTIMER_STATE_CALLBACK;
 }
@@ -1243,6 +1229,22 @@ void hrtimer_interrupt(struct clock_event_device *dev)
 	}
 }
 
+/*
+ * local version of hrtimer_peek_ahead_timers() called with interrupts
+ * disabled.
+ */
+static void __hrtimer_peek_ahead_timers(void)
+{
+	struct tick_device *td;
+
+	if (!hrtimer_hres_active())
+		return;
+
+	td = &__get_cpu_var(tick_cpu_device);
+	if (td && td->evtdev)
+		hrtimer_interrupt(td->evtdev);
+}
+
 /**
  * hrtimer_peek_ahead_timers -- run soft-expired timers now
  *
@@ -1254,20 +1256,23 @@ void hrtimer_interrupt(struct clock_event_device *dev)
  */
 void hrtimer_peek_ahead_timers(void)
 {
-	struct tick_device *td;
 	unsigned long flags;
 
-	if (!hrtimer_hres_active())
-		return;
-
 	local_irq_save(flags);
-	td = &__get_cpu_var(tick_cpu_device);
-	if (td && td->evtdev)
-		hrtimer_interrupt(td->evtdev);
+	__hrtimer_peek_ahead_timers();
 	local_irq_restore(flags);
 }
 
-#endif	/* CONFIG_HIGH_RES_TIMERS */
+static void run_hrtimer_softirq(struct softirq_action *h)
+{
+	hrtimer_peek_ahead_timers();
+}
+
+#else /* CONFIG_HIGH_RES_TIMERS */
+
+static inline void __hrtimer_peek_ahead_timers(void) { }
+
+#endif	/* !CONFIG_HIGH_RES_TIMERS */
 
 /*
  * Called from timer softirq every jiffy, expire hrtimers:
@@ -1513,39 +1518,36 @@ static void migrate_hrtimer_list(struct hrtimer_clock_base *old_base,
 		__remove_hrtimer(timer, old_base, HRTIMER_STATE_MIGRATE, 0);
 		timer->base = new_base;
 		/*
-		 * Enqueue the timers on the new cpu, but do not reprogram 
-		 * the timer as that would enable a deadlock between
-		 * hrtimer_enqueue_reprogramm() running the timer and us still
-		 * holding a nested base lock.
-		 *
-		 * Instead we tickle the hrtimer interrupt after the migration
-		 * is done, which will run all expired timers and re-programm
-		 * the timer device.
+		 * Enqueue the timers on the new cpu. This does not
+		 * reprogram the event device in case the timer
+		 * expires before the earliest on this CPU, but we run
+		 * hrtimer_interrupt after we migrated everything to
+		 * sort out already expired timers and reprogram the
+		 * event device.
 		 */
-		enqueue_hrtimer(timer, new_base, 0);
+		enqueue_hrtimer(timer, new_base);
 
 		/* Clear the migration state bit */
 		timer->state &= ~HRTIMER_STATE_MIGRATE;
 	}
 }
 
-static int migrate_hrtimers(int scpu)
+static void migrate_hrtimers(int scpu)
 {
 	struct hrtimer_cpu_base *old_base, *new_base;
-	int dcpu, i;
+	int i;
 
 	BUG_ON(cpu_online(scpu));
-	old_base = &per_cpu(hrtimer_bases, scpu);
-	new_base = &get_cpu_var(hrtimer_bases);
-
-	dcpu = smp_processor_id();
-
 	tick_cancel_sched_timer(scpu);
+
+	local_irq_disable();
+	old_base = &per_cpu(hrtimer_bases, scpu);
+	new_base = &__get_cpu_var(hrtimer_bases);
 	/*
 	 * The caller is globally serialized and nobody else
 	 * takes two locks at once, deadlock is not possible.
 	 */
-	spin_lock_irq(&new_base->lock);
+	spin_lock(&new_base->lock);
 	spin_lock_nested(&old_base->lock, SINGLE_DEPTH_NESTING);
 
 	for (i = 0; i < HRTIMER_MAX_CLOCK_BASES; i++) {
@@ -1554,15 +1556,11 @@ static int migrate_hrtimers(int scpu)
 	}
 
 	spin_unlock(&old_base->lock);
-	spin_unlock_irq(&new_base->lock);
-	put_cpu_var(hrtimer_bases);
+	spin_unlock(&new_base->lock);
 
-	return dcpu;
-}
-
-static void tickle_timers(void *arg)
-{
-	hrtimer_peek_ahead_timers();
+	/* Check, if we got expired work to do */
+	__hrtimer_peek_ahead_timers();
+	local_irq_enable();
 }
 
 #endif /* CONFIG_HOTPLUG_CPU */
@@ -1583,11 +1581,8 @@ static int __cpuinit hrtimer_cpu_notify(struct notifier_block *self,
 	case CPU_DEAD:
 	case CPU_DEAD_FROZEN:
 	{
-		int dcpu;
-
 		clockevents_notify(CLOCK_EVT_NOTIFY_CPU_DEAD, &scpu);
-		dcpu = migrate_hrtimers(scpu);
-		smp_call_function_single(dcpu, tickle_timers, NULL, 0);
+		migrate_hrtimers(scpu);
 		break;
 	}
 #endif
@@ -1608,6 +1603,9 @@ void __init hrtimers_init(void)
 	hrtimer_cpu_notify(&hrtimers_nb, (unsigned long)CPU_UP_PREPARE,
 			  (void *)(long)smp_processor_id());
 	register_cpu_notifier(&hrtimers_nb);
+#ifdef CONFIG_HIGH_RES_TIMERS
+	open_softirq(HRTIMER_SOFTIRQ, run_hrtimer_softirq);
+#endif
 }
 
 /**

^ permalink raw reply related	[flat|nested] 69+ messages in thread

* [git pull] timer fixes
@ 2008-12-04 20:52 Ingo Molnar
  0 siblings, 0 replies; 69+ messages in thread
From: Ingo Molnar @ 2008-12-04 20:52 UTC (permalink / raw)
  To: Linus Torvalds; +Cc: linux-kernel, Andrew Morton, Thomas Gleixner

Linus,

Please pull the latest timers-fixes-for-linus git tree from:

   git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip.git timers-fixes-for-linus

 Thanks,

	Ingo

------------------>
John Stultz (1):
      time: catch xtime_nsec underflows and fix them

Petr Tesarik (1):
      posix-cpu-timers: fix clock_gettime with CLOCK_PROCESS_CPUTIME_ID


 kernel/posix-cpu-timers.c |    2 +-
 kernel/time/timekeeping.c |   22 ++++++++++++++++++++++
 2 files changed, 23 insertions(+), 1 deletions(-)

diff --git a/kernel/posix-cpu-timers.c b/kernel/posix-cpu-timers.c
index 895337b..4e5288a 100644
--- a/kernel/posix-cpu-timers.c
+++ b/kernel/posix-cpu-timers.c
@@ -311,7 +311,7 @@ static int cpu_clock_sample_group(const clockid_t which_clock,
 	struct task_cputime cputime;
 
 	thread_group_cputime(p, &cputime);
-	switch (which_clock) {
+	switch (CPUCLOCK_WHICH(which_clock)) {
 	default:
 		return -EINVAL;
 	case CPUCLOCK_PROF:
diff --git a/kernel/time/timekeeping.c b/kernel/time/timekeeping.c
index e7acfb4..fa05e88 100644
--- a/kernel/time/timekeeping.c
+++ b/kernel/time/timekeeping.c
@@ -518,6 +518,28 @@ void update_wall_time(void)
 	/* correct the clock when NTP error is too big */
 	clocksource_adjust(offset);
 
+	/*
+	 * Since in the loop above, we accumulate any amount of time
+	 * in xtime_nsec over a second into xtime.tv_sec, its possible for
+	 * xtime_nsec to be fairly small after the loop. Further, if we're
+	 * slightly speeding the clocksource up in clocksource_adjust(),
+	 * its possible the required corrective factor to xtime_nsec could
+	 * cause it to underflow.
+	 *
+	 * Now, we cannot simply roll the accumulated second back, since
+	 * the NTP subsystem has been notified via second_overflow. So
+	 * instead we push xtime_nsec forward by the amount we underflowed,
+	 * and add that amount into the error.
+	 *
+	 * We'll correct this error next time through this function, when
+	 * xtime_nsec is not as small.
+	 */
+	if (unlikely((s64)clock->xtime_nsec < 0)) {
+		s64 neg = -(s64)clock->xtime_nsec;
+		clock->xtime_nsec = 0;
+		clock->error += neg << (NTP_SCALE_SHIFT - clock->shift);
+	}
+
 	/* store full nanoseconds into xtime after rounding it up and
 	 * add the remainder to the error difference.
 	 */

^ permalink raw reply related	[flat|nested] 69+ messages in thread

* [git pull] timer fixes
@ 2008-11-11 18:47 Ingo Molnar
  0 siblings, 0 replies; 69+ messages in thread
From: Ingo Molnar @ 2008-11-11 18:47 UTC (permalink / raw)
  To: Linus Torvalds; +Cc: linux-kernel, Andrew Morton, Thomas Gleixner

Linus,

Please pull the latest timers-fixes-for-linus git tree from:

   git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip.git timers-fixes-for-linus

 Thanks,

	Ingo

------------------>
Gautham R Shenoy (1):
      timers: handle HRTIMER_CB_IRQSAFE_UNLOCKED correctly from softirq context

Matt Fleming (3):
      x86: HPET: convert WARN_ON to WARN_ON_ONCE
      x86: HPET: read from HPET_Tn_CMP() not HPET_T0_CMP
      x86: HPET: enter hpet_interrupt_handler with interrupts disabled

Thomas Gleixner (2):
      irq: call __irq_enter() before calling the tick_idle_check
      nohz: disable tick_nohz_kick_tick() for now


 arch/x86/kernel/hpet.c   |    4 ++--
 kernel/hrtimer.c         |   17 ++++++++++++++++-
 kernel/softirq.c         |    7 ++++---
 kernel/time/tick-sched.c |    4 ++++
 4 files changed, 26 insertions(+), 6 deletions(-)

diff --git a/arch/x86/kernel/hpet.c b/arch/x86/kernel/hpet.c
index 77017e8..067d8de 100644
--- a/arch/x86/kernel/hpet.c
+++ b/arch/x86/kernel/hpet.c
@@ -322,7 +322,7 @@ static int hpet_next_event(unsigned long delta,
 	 * what we wrote hit the chip before we compare it to the
 	 * counter.
 	 */
-	WARN_ON((u32)hpet_readl(HPET_T0_CMP) != cnt);
+	WARN_ON_ONCE((u32)hpet_readl(HPET_Tn_CMP(timer)) != cnt);
 
 	return (s32)((u32)hpet_readl(HPET_COUNTER) - cnt) >= 0 ? -ETIME : 0;
 }
@@ -445,7 +445,7 @@ static int hpet_setup_irq(struct hpet_dev *dev)
 {
 
 	if (request_irq(dev->irq, hpet_interrupt_handler,
-			IRQF_SHARED|IRQF_NOBALANCING, dev->name, dev))
+			IRQF_DISABLED|IRQF_NOBALANCING, dev->name, dev))
 		return -1;
 
 	disable_irq(dev->irq);
diff --git a/kernel/hrtimer.c b/kernel/hrtimer.c
index 2b465df..95d3949 100644
--- a/kernel/hrtimer.c
+++ b/kernel/hrtimer.c
@@ -1209,6 +1209,7 @@ static void run_hrtimer_pending(struct hrtimer_cpu_base *cpu_base)
 		enum hrtimer_restart (*fn)(struct hrtimer *);
 		struct hrtimer *timer;
 		int restart;
+		int emulate_hardirq_ctx = 0;
 
 		timer = list_entry(cpu_base->cb_pending.next,
 				   struct hrtimer, cb_entry);
@@ -1217,10 +1218,24 @@ static void run_hrtimer_pending(struct hrtimer_cpu_base *cpu_base)
 		timer_stats_account_hrtimer(timer);
 
 		fn = timer->function;
+		/*
+		 * A timer might have been added to the cb_pending list
+		 * when it was migrated during a cpu-offline operation.
+		 * Emulate hardirq context for such timers.
+		 */
+		if (timer->cb_mode == HRTIMER_CB_IRQSAFE_PERCPU ||
+		    timer->cb_mode == HRTIMER_CB_IRQSAFE_UNLOCKED)
+			emulate_hardirq_ctx = 1;
+
 		__remove_hrtimer(timer, timer->base, HRTIMER_STATE_CALLBACK, 0);
 		spin_unlock_irq(&cpu_base->lock);
 
-		restart = fn(timer);
+		if (unlikely(emulate_hardirq_ctx)) {
+			local_irq_disable();
+			restart = fn(timer);
+			local_irq_enable();
+		} else
+			restart = fn(timer);
 
 		spin_lock_irq(&cpu_base->lock);
 
diff --git a/kernel/softirq.c b/kernel/softirq.c
index 7110dae..e7c69a7 100644
--- a/kernel/softirq.c
+++ b/kernel/softirq.c
@@ -269,10 +269,11 @@ void irq_enter(void)
 {
 	int cpu = smp_processor_id();
 
-	if (idle_cpu(cpu) && !in_interrupt())
+	if (idle_cpu(cpu) && !in_interrupt()) {
+		__irq_enter();
 		tick_check_idle(cpu);
-
-	__irq_enter();
+	} else
+		__irq_enter();
 }
 
 #ifdef __ARCH_IRQ_EXIT_IRQS_DISABLED
diff --git a/kernel/time/tick-sched.c b/kernel/time/tick-sched.c
index 5bbb104..342fc9c 100644
--- a/kernel/time/tick-sched.c
+++ b/kernel/time/tick-sched.c
@@ -568,6 +568,9 @@ static void tick_nohz_switch_to_nohz(void)
  */
 static void tick_nohz_kick_tick(int cpu)
 {
+#if 0
+	/* Switch back to 2.6.27 behaviour */
+
 	struct tick_sched *ts = &per_cpu(tick_cpu_sched, cpu);
 	ktime_t delta, now;
 
@@ -584,6 +587,7 @@ static void tick_nohz_kick_tick(int cpu)
 		return;
 
 	tick_nohz_restart(ts, now);
+#endif
 }
 
 #else

^ permalink raw reply related	[flat|nested] 69+ messages in thread

* [git pull] timer fixes
@ 2008-09-23 19:41 Ingo Molnar
  0 siblings, 0 replies; 69+ messages in thread
From: Ingo Molnar @ 2008-09-23 19:41 UTC (permalink / raw)
  To: Linus Torvalds
  Cc: linux-kernel, Thomas Gleixner, H. Peter Anvin, Andrew Morton

Linus,

Please pull the latest timers-fixes-for-linus git tree from:

   git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip.git timers-fixes-for-linus

A category of late fixes that solves the problems around regression 
http://bugzilla.kernel.org/show_bug.cgi?id=11611 . Tested by Rafael, and 
tested through on a range of laptops we have access to.

Thanks,

	Ingo

------------------>
Andreas Herrmann (1):
      x86: c1e_idle: don't mark TSC unstable if CPU has invariant TSC

Ingo Molnar (1):
      timers: fix build error in !oneshot case

Thomas Gleixner (6):
      clockevents: prevent cpu online to interfere with nohz
      x86: prevent stale state of c1e_mask across CPU offline/online
      clockevents: prevent stale tick_next_period for onlining CPUs
      clockevents: check broadcast device not tick device
      clockevents: prevent mode mismatch on cpu online
      x86: prevent C-states hang on AMD C1E enabled machines


 arch/x86/kernel/process.c    |   17 ++++++++++++-----
 arch/x86/kernel/process_32.c |    1 +
 arch/x86/kernel/process_64.c |    2 ++
 include/asm-x86/acpi.h       |    2 ++
 include/asm-x86/cpufeature.h |    1 +
 include/asm-x86/idle.h       |    2 ++
 kernel/time/tick-broadcast.c |   12 ++++++++++--
 kernel/time/tick-common.c    |   10 ++++++----
 kernel/time/tick-internal.h  |    7 +++++++
 kernel/time/tick-sched.c     |   11 +++++++----
 10 files changed, 50 insertions(+), 15 deletions(-)

diff --git a/arch/x86/kernel/process.c b/arch/x86/kernel/process.c
index 7fc4d5b..876e918 100644
--- a/arch/x86/kernel/process.c
+++ b/arch/x86/kernel/process.c
@@ -246,6 +246,14 @@ static int __cpuinit check_c1e_idle(const struct cpuinfo_x86 *c)
 	return 1;
 }
 
+static cpumask_t c1e_mask = CPU_MASK_NONE;
+static int c1e_detected;
+
+void c1e_remove_cpu(int cpu)
+{
+	cpu_clear(cpu, c1e_mask);
+}
+
 /*
  * C1E aware idle routine. We check for C1E active in the interrupt
  * pending message MSR. If we detect C1E, then we handle it the same
@@ -253,9 +261,6 @@ static int __cpuinit check_c1e_idle(const struct cpuinfo_x86 *c)
  */
 static void c1e_idle(void)
 {
-	static cpumask_t c1e_mask = CPU_MASK_NONE;
-	static int c1e_detected;
-
 	if (need_resched())
 		return;
 
@@ -265,8 +270,10 @@ static void c1e_idle(void)
 		rdmsr(MSR_K8_INT_PENDING_MSG, lo, hi);
 		if (lo & K8_INTP_C1E_ACTIVE_MASK) {
 			c1e_detected = 1;
-			mark_tsc_unstable("TSC halt in C1E");
-			printk(KERN_INFO "System has C1E enabled\n");
+			if (!boot_cpu_has(X86_FEATURE_CONSTANT_TSC))
+				mark_tsc_unstable("TSC halt in AMD C1E");
+			printk(KERN_INFO "System has AMD C1E enabled\n");
+			set_cpu_cap(&boot_cpu_data, X86_FEATURE_AMDC1E);
 		}
 	}
 
diff --git a/arch/x86/kernel/process_32.c b/arch/x86/kernel/process_32.c
index 3b7a1dd..4b3cfdf 100644
--- a/arch/x86/kernel/process_32.c
+++ b/arch/x86/kernel/process_32.c
@@ -88,6 +88,7 @@ static void cpu_exit_clear(void)
 	cpu_clear(cpu, cpu_callin_map);
 
 	numa_remove_cpu(cpu);
+	c1e_remove_cpu(cpu);
 }
 
 /* We don't actually take CPU down, just spin without interrupts. */
diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c
index 71553b6..e12e0e4 100644
--- a/arch/x86/kernel/process_64.c
+++ b/arch/x86/kernel/process_64.c
@@ -93,6 +93,8 @@ DECLARE_PER_CPU(int, cpu_state);
 static inline void play_dead(void)
 {
 	idle_task_exit();
+	c1e_remove_cpu(raw_smp_processor_id());
+
 	mb();
 	/* Ack it */
 	__get_cpu_var(cpu_state) = CPU_DEAD;
diff --git a/include/asm-x86/acpi.h b/include/asm-x86/acpi.h
index 635d764..35d1743 100644
--- a/include/asm-x86/acpi.h
+++ b/include/asm-x86/acpi.h
@@ -140,6 +140,8 @@ static inline unsigned int acpi_processor_cstate_check(unsigned int max_cstate)
 	    boot_cpu_data.x86_model <= 0x05 &&
 	    boot_cpu_data.x86_mask < 0x0A)
 		return 1;
+	else if (boot_cpu_has(X86_FEATURE_AMDC1E))
+		return 1;
 	else
 		return max_cstate;
 }
diff --git a/include/asm-x86/cpufeature.h b/include/asm-x86/cpufeature.h
index 9489283..cfcfb0a 100644
--- a/include/asm-x86/cpufeature.h
+++ b/include/asm-x86/cpufeature.h
@@ -81,6 +81,7 @@
 #define X86_FEATURE_LFENCE_RDTSC (3*32+18) /* Lfence synchronizes RDTSC */
 #define X86_FEATURE_11AP	(3*32+19) /* Bad local APIC aka 11AP */
 #define X86_FEATURE_NOPL	(3*32+20) /* The NOPL (0F 1F) instructions */
+#define X86_FEATURE_AMDC1E	(3*32+21) /* AMD C1E detected */
 
 /* Intel-defined CPU features, CPUID level 0x00000001 (ecx), word 4 */
 #define X86_FEATURE_XMM3	(4*32+ 0) /* Streaming SIMD Extensions-3 */
diff --git a/include/asm-x86/idle.h b/include/asm-x86/idle.h
index d240e5b..cbb6491 100644
--- a/include/asm-x86/idle.h
+++ b/include/asm-x86/idle.h
@@ -10,4 +10,6 @@ void idle_notifier_register(struct notifier_block *n);
 void enter_idle(void);
 void exit_idle(void);
 
+void c1e_remove_cpu(int cpu);
+
 #endif
diff --git a/kernel/time/tick-broadcast.c b/kernel/time/tick-broadcast.c
index f1f3eee..bd70345 100644
--- a/kernel/time/tick-broadcast.c
+++ b/kernel/time/tick-broadcast.c
@@ -235,7 +235,7 @@ static void tick_do_broadcast_on_off(void *why)
 	case CLOCK_EVT_NOTIFY_BROADCAST_FORCE:
 		if (!cpu_isset(cpu, tick_broadcast_mask)) {
 			cpu_set(cpu, tick_broadcast_mask);
-			if (td->mode == TICKDEV_MODE_PERIODIC)
+			if (bc->mode == TICKDEV_MODE_PERIODIC)
 				clockevents_shutdown(dev);
 		}
 		if (*reason == CLOCK_EVT_NOTIFY_BROADCAST_FORCE)
@@ -245,7 +245,7 @@ static void tick_do_broadcast_on_off(void *why)
 		if (!tick_broadcast_force &&
 		    cpu_isset(cpu, tick_broadcast_mask)) {
 			cpu_clear(cpu, tick_broadcast_mask);
-			if (td->mode == TICKDEV_MODE_PERIODIC)
+			if (bc->mode == TICKDEV_MODE_PERIODIC)
 				tick_setup_periodic(dev, 0);
 		}
 		break;
@@ -575,4 +575,12 @@ void tick_shutdown_broadcast_oneshot(unsigned int *cpup)
 	spin_unlock_irqrestore(&tick_broadcast_lock, flags);
 }
 
+/*
+ * Check, whether the broadcast device is in one shot mode
+ */
+int tick_broadcast_oneshot_active(void)
+{
+	return tick_broadcast_device.mode == TICKDEV_MODE_ONESHOT;
+}
+
 #endif
diff --git a/kernel/time/tick-common.c b/kernel/time/tick-common.c
index 019315e..df12434 100644
--- a/kernel/time/tick-common.c
+++ b/kernel/time/tick-common.c
@@ -33,7 +33,7 @@ DEFINE_PER_CPU(struct tick_device, tick_cpu_device);
  */
 ktime_t tick_next_period;
 ktime_t tick_period;
-int tick_do_timer_cpu __read_mostly = -1;
+int tick_do_timer_cpu __read_mostly = TICK_DO_TIMER_BOOT;
 DEFINE_SPINLOCK(tick_device_lock);
 
 /*
@@ -109,7 +109,8 @@ void tick_setup_periodic(struct clock_event_device *dev, int broadcast)
 	if (!tick_device_is_functional(dev))
 		return;
 
-	if (dev->features & CLOCK_EVT_FEAT_PERIODIC) {
+	if ((dev->features & CLOCK_EVT_FEAT_PERIODIC) &&
+	    !tick_broadcast_oneshot_active()) {
 		clockevents_set_mode(dev, CLOCK_EVT_MODE_PERIODIC);
 	} else {
 		unsigned long seq;
@@ -148,7 +149,7 @@ static void tick_setup_device(struct tick_device *td,
 		 * If no cpu took the do_timer update, assign it to
 		 * this cpu:
 		 */
-		if (tick_do_timer_cpu == -1) {
+		if (tick_do_timer_cpu == TICK_DO_TIMER_BOOT) {
 			tick_do_timer_cpu = cpu;
 			tick_next_period = ktime_get();
 			tick_period = ktime_set(0, NSEC_PER_SEC / HZ);
@@ -300,7 +301,8 @@ static void tick_shutdown(unsigned int *cpup)
 	if (*cpup == tick_do_timer_cpu) {
 		int cpu = first_cpu(cpu_online_map);
 
-		tick_do_timer_cpu = (cpu != NR_CPUS) ? cpu : -1;
+		tick_do_timer_cpu = (cpu != NR_CPUS) ? cpu :
+			TICK_DO_TIMER_NONE;
 	}
 	spin_unlock_irqrestore(&tick_device_lock, flags);
 }
diff --git a/kernel/time/tick-internal.h b/kernel/time/tick-internal.h
index 6e9db97..4692487 100644
--- a/kernel/time/tick-internal.h
+++ b/kernel/time/tick-internal.h
@@ -1,6 +1,10 @@
 /*
  * tick internal variable and functions used by low/high res code
  */
+
+#define TICK_DO_TIMER_NONE	-1
+#define TICK_DO_TIMER_BOOT	-2
+
 DECLARE_PER_CPU(struct tick_device, tick_cpu_device);
 extern spinlock_t tick_device_lock;
 extern ktime_t tick_next_period;
@@ -31,6 +35,7 @@ extern void tick_broadcast_oneshot_control(unsigned long reason);
 extern void tick_broadcast_switch_to_oneshot(void);
 extern void tick_shutdown_broadcast_oneshot(unsigned int *cpup);
 extern int tick_resume_broadcast_oneshot(struct clock_event_device *bc);
+extern int tick_broadcast_oneshot_active(void);
 # else /* BROADCAST */
 static inline void tick_broadcast_setup_oneshot(struct clock_event_device *bc)
 {
@@ -39,6 +44,7 @@ static inline void tick_broadcast_setup_oneshot(struct clock_event_device *bc)
 static inline void tick_broadcast_oneshot_control(unsigned long reason) { }
 static inline void tick_broadcast_switch_to_oneshot(void) { }
 static inline void tick_shutdown_broadcast_oneshot(unsigned int *cpup) { }
+static inline int tick_broadcast_oneshot_active(void) { return 0; }
 # endif /* !BROADCAST */
 
 #else /* !ONESHOT */
@@ -68,6 +74,7 @@ static inline int tick_resume_broadcast_oneshot(struct clock_event_device *bc)
 {
 	return 0;
 }
+static inline int tick_broadcast_oneshot_active(void) { return 0; }
 #endif /* !TICK_ONESHOT */
 
 /*
diff --git a/kernel/time/tick-sched.c b/kernel/time/tick-sched.c
index a87b046..39019b3 100644
--- a/kernel/time/tick-sched.c
+++ b/kernel/time/tick-sched.c
@@ -75,6 +75,9 @@ static void tick_do_update_jiffies64(ktime_t now)
 							   incr * ticks);
 		}
 		do_timer(++ticks);
+
+		/* Keep the tick_next_period variable up to date */
+		tick_next_period = ktime_add(last_jiffies_update, tick_period);
 	}
 	write_sequnlock(&xtime_lock);
 }
@@ -221,7 +224,7 @@ void tick_nohz_stop_sched_tick(int inidle)
 	 */
 	if (unlikely(!cpu_online(cpu))) {
 		if (cpu == tick_do_timer_cpu)
-			tick_do_timer_cpu = -1;
+			tick_do_timer_cpu = TICK_DO_TIMER_NONE;
 	}
 
 	if (unlikely(ts->nohz_mode == NOHZ_MODE_INACTIVE))
@@ -303,7 +306,7 @@ void tick_nohz_stop_sched_tick(int inidle)
 		 * invoked.
 		 */
 		if (cpu == tick_do_timer_cpu)
-			tick_do_timer_cpu = -1;
+			tick_do_timer_cpu = TICK_DO_TIMER_NONE;
 
 		ts->idle_sleeps++;
 
@@ -468,7 +471,7 @@ static void tick_nohz_handler(struct clock_event_device *dev)
 	 * this duty, then the jiffies update is still serialized by
 	 * xtime_lock.
 	 */
-	if (unlikely(tick_do_timer_cpu == -1))
+	if (unlikely(tick_do_timer_cpu == TICK_DO_TIMER_NONE))
 		tick_do_timer_cpu = cpu;
 
 	/* Check, if the jiffies need an update */
@@ -570,7 +573,7 @@ static enum hrtimer_restart tick_sched_timer(struct hrtimer *timer)
 	 * this duty, then the jiffies update is still serialized by
 	 * xtime_lock.
 	 */
-	if (unlikely(tick_do_timer_cpu == -1))
+	if (unlikely(tick_do_timer_cpu == TICK_DO_TIMER_NONE))
 		tick_do_timer_cpu = cpu;
 #endif
 

^ permalink raw reply related	[flat|nested] 69+ messages in thread

* [git pull] timer fixes
@ 2008-09-17  9:59 Ingo Molnar
  0 siblings, 0 replies; 69+ messages in thread
From: Ingo Molnar @ 2008-09-17  9:59 UTC (permalink / raw)
  To: Linus Torvalds; +Cc: linux-kernel, Andrew Morton, Thomas Gleixner

Linus,

Please pull the latest timers-fixes-for-linus git tree from:

   git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip.git timers-fixes-for-linus

 Thanks,

	Ingo

------------------>
Dominik Brodowski (1):
      clocksource, acpi_pm.c: fix check for monotonicity

Thomas Gleixner (2):
      clockevents: remove WARN_ON which was used to gather information
      clockevents: make device shutdown robust


 drivers/clocksource/acpi_pm.c |   21 ++++++++++-----------
 kernel/time/clockevents.c     |   12 +++++++++++-
 kernel/time/tick-broadcast.c  |    9 ++++-----
 kernel/time/tick-common.c     |    4 ++--
 kernel/time/tick-internal.h   |    2 ++
 5 files changed, 29 insertions(+), 19 deletions(-)

diff --git a/drivers/clocksource/acpi_pm.c b/drivers/clocksource/acpi_pm.c
index 4eee533..71d2ac4 100644
--- a/drivers/clocksource/acpi_pm.c
+++ b/drivers/clocksource/acpi_pm.c
@@ -178,11 +178,13 @@ static int verify_pmtmr_rate(void)
 
 /* Number of monotonicity checks to perform during initialization */
 #define ACPI_PM_MONOTONICITY_CHECKS 10
+/* Number of reads we try to get two different values */
+#define ACPI_PM_READ_CHECKS 10000
 
 static int __init init_acpi_pm_clocksource(void)
 {
 	cycle_t value1, value2;
-	unsigned int i, j, good = 0;
+	unsigned int i, j = 0;
 
 	if (!pmtmr_ioport)
 		return -ENODEV;
@@ -192,29 +194,26 @@ static int __init init_acpi_pm_clocksource(void)
 
 	/* "verify" this timing source: */
 	for (j = 0; j < ACPI_PM_MONOTONICITY_CHECKS; j++) {
+		udelay(100 * j);
 		value1 = clocksource_acpi_pm.read();
-		for (i = 0; i < 10000; i++) {
+		for (i = 0; i < ACPI_PM_READ_CHECKS; i++) {
 			value2 = clocksource_acpi_pm.read();
 			if (value2 == value1)
 				continue;
 			if (value2 > value1)
-				good++;
 				break;
 			if ((value2 < value1) && ((value2) < 0xFFF))
-				good++;
 				break;
 			printk(KERN_INFO "PM-Timer had inconsistent results:"
 			       " 0x%#llx, 0x%#llx - aborting.\n",
 			       value1, value2);
 			return -EINVAL;
 		}
-		udelay(300 * i);
-	}
-
-	if (good != ACPI_PM_MONOTONICITY_CHECKS) {
-		printk(KERN_INFO "PM-Timer failed consistency check "
-		       " (0x%#llx) - aborting.\n", value1);
-		return -ENODEV;
+		if (i == ACPI_PM_READ_CHECKS) {
+			printk(KERN_INFO "PM-Timer failed consistency check "
+			       " (0x%#llx) - aborting.\n", value1);
+			return -ENODEV;
+		}
 	}
 
 	if (verify_pmtmr_rate() != 0)
diff --git a/kernel/time/clockevents.c b/kernel/time/clockevents.c
index 1876b52..f8d9680 100644
--- a/kernel/time/clockevents.c
+++ b/kernel/time/clockevents.c
@@ -72,6 +72,16 @@ void clockevents_set_mode(struct clock_event_device *dev,
 }
 
 /**
+ * clockevents_shutdown - shutdown the device and clear next_event
+ * @dev:	device to shutdown
+ */
+void clockevents_shutdown(struct clock_event_device *dev)
+{
+	clockevents_set_mode(dev, CLOCK_EVT_MODE_SHUTDOWN);
+	dev->next_event.tv64 = KTIME_MAX;
+}
+
+/**
  * clockevents_program_event - Reprogram the clock event device.
  * @expires:	absolute expiry time (monotonic clock)
  *
@@ -206,7 +216,7 @@ void clockevents_exchange_device(struct clock_event_device *old,
 
 	if (new) {
 		BUG_ON(new->mode != CLOCK_EVT_MODE_UNUSED);
-		clockevents_set_mode(new, CLOCK_EVT_MODE_SHUTDOWN);
+		clockevents_shutdown(new);
 	}
 	local_irq_restore(flags);
 }
diff --git a/kernel/time/tick-broadcast.c b/kernel/time/tick-broadcast.c
index 2f5a382..f1f3eee 100644
--- a/kernel/time/tick-broadcast.c
+++ b/kernel/time/tick-broadcast.c
@@ -236,8 +236,7 @@ static void tick_do_broadcast_on_off(void *why)
 		if (!cpu_isset(cpu, tick_broadcast_mask)) {
 			cpu_set(cpu, tick_broadcast_mask);
 			if (td->mode == TICKDEV_MODE_PERIODIC)
-				clockevents_set_mode(dev,
-						     CLOCK_EVT_MODE_SHUTDOWN);
+				clockevents_shutdown(dev);
 		}
 		if (*reason == CLOCK_EVT_NOTIFY_BROADCAST_FORCE)
 			tick_broadcast_force = 1;
@@ -254,7 +253,7 @@ static void tick_do_broadcast_on_off(void *why)
 
 	if (cpus_empty(tick_broadcast_mask)) {
 		if (!bc_stopped)
-			clockevents_set_mode(bc, CLOCK_EVT_MODE_SHUTDOWN);
+			clockevents_shutdown(bc);
 	} else if (bc_stopped) {
 		if (tick_broadcast_device.mode == TICKDEV_MODE_PERIODIC)
 			tick_broadcast_start_periodic(bc);
@@ -306,7 +305,7 @@ void tick_shutdown_broadcast(unsigned int *cpup)
 
 	if (tick_broadcast_device.mode == TICKDEV_MODE_PERIODIC) {
 		if (bc && cpus_empty(tick_broadcast_mask))
-			clockevents_set_mode(bc, CLOCK_EVT_MODE_SHUTDOWN);
+			clockevents_shutdown(bc);
 	}
 
 	spin_unlock_irqrestore(&tick_broadcast_lock, flags);
@@ -321,7 +320,7 @@ void tick_suspend_broadcast(void)
 
 	bc = tick_broadcast_device.evtdev;
 	if (bc)
-		clockevents_set_mode(bc, CLOCK_EVT_MODE_SHUTDOWN);
+		clockevents_shutdown(bc);
 
 	spin_unlock_irqrestore(&tick_broadcast_lock, flags);
 }
diff --git a/kernel/time/tick-common.c b/kernel/time/tick-common.c
index c477719..019315e 100644
--- a/kernel/time/tick-common.c
+++ b/kernel/time/tick-common.c
@@ -249,7 +249,7 @@ static int tick_check_new_device(struct clock_event_device *newdev)
 	 * not give it back to the clockevents layer !
 	 */
 	if (tick_is_broadcast_device(curdev)) {
-		clockevents_set_mode(curdev, CLOCK_EVT_MODE_SHUTDOWN);
+		clockevents_shutdown(curdev);
 		curdev = NULL;
 	}
 	clockevents_exchange_device(curdev, newdev);
@@ -311,7 +311,7 @@ static void tick_suspend(void)
 	unsigned long flags;
 
 	spin_lock_irqsave(&tick_device_lock, flags);
-	clockevents_set_mode(td->evtdev, CLOCK_EVT_MODE_SHUTDOWN);
+	clockevents_shutdown(td->evtdev);
 	spin_unlock_irqrestore(&tick_device_lock, flags);
 }
 
diff --git a/kernel/time/tick-internal.h b/kernel/time/tick-internal.h
index 0ffc291..6e9db97 100644
--- a/kernel/time/tick-internal.h
+++ b/kernel/time/tick-internal.h
@@ -10,6 +10,8 @@ extern int tick_do_timer_cpu __read_mostly;
 extern void tick_setup_periodic(struct clock_event_device *dev, int broadcast);
 extern void tick_handle_periodic(struct clock_event_device *dev);
 
+extern void clockevents_shutdown(struct clock_event_device *dev);
+
 /*
  * NO_HZ / high resolution timer shared code
  */

^ permalink raw reply related	[flat|nested] 69+ messages in thread

* [git pull] timer fixes
@ 2008-08-25 17:52 Ingo Molnar
  0 siblings, 0 replies; 69+ messages in thread
From: Ingo Molnar @ 2008-08-25 17:52 UTC (permalink / raw)
  To: Linus Torvalds; +Cc: linux-kernel, Thomas Gleixner, Andrew Morton

Linus,

Please pull the latest timers-fixes-for-linus git tree from:

   git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip.git timers-fixes-for-linus

Thanks,

	Ingo

------------------>
Ingo Molnar (1):
      rtc: fix deadlock


 include/asm-generic/rtc.h |   12 ++++--------
 1 files changed, 4 insertions(+), 8 deletions(-)

diff --git a/include/asm-generic/rtc.h b/include/asm-generic/rtc.h
index be4af00..71ef3f0 100644
--- a/include/asm-generic/rtc.h
+++ b/include/asm-generic/rtc.h
@@ -15,6 +15,7 @@
 #include <linux/mc146818rtc.h>
 #include <linux/rtc.h>
 #include <linux/bcd.h>
+#include <linux/delay.h>
 
 #define RTC_PIE 0x40		/* periodic interrupt enable */
 #define RTC_AIE 0x20		/* alarm interrupt enable */
@@ -43,7 +44,6 @@ static inline unsigned char rtc_is_updating(void)
 
 static inline unsigned int get_rtc_time(struct rtc_time *time)
 {
-	unsigned long uip_watchdog = jiffies;
 	unsigned char ctrl;
 	unsigned long flags;
 
@@ -53,19 +53,15 @@ static inline unsigned int get_rtc_time(struct rtc_time *time)
 
 	/*
 	 * read RTC once any update in progress is done. The update
-	 * can take just over 2ms. We wait 10 to 20ms. There is no need to
+	 * can take just over 2ms. We wait 20ms. There is no need to
 	 * to poll-wait (up to 1s - eeccch) for the falling edge of RTC_UIP.
 	 * If you need to know *exactly* when a second has started, enable
 	 * periodic update complete interrupts, (via ioctl) and then 
 	 * immediately read /dev/rtc which will block until you get the IRQ.
 	 * Once the read clears, read the RTC time (again via ioctl). Easy.
 	 */
-
-	if (rtc_is_updating() != 0)
-		while (jiffies - uip_watchdog < 2*HZ/100) {
-			barrier();
-			cpu_relax();
-		}
+	if (rtc_is_updating())
+		mdelay(20);
 
 	/*
 	 * Only the values that we read from the RTC are set. We leave

^ permalink raw reply related	[flat|nested] 69+ messages in thread

* [git pull] timer fixes
@ 2008-08-22 12:24 Ingo Molnar
  0 siblings, 0 replies; 69+ messages in thread
From: Ingo Molnar @ 2008-08-22 12:24 UTC (permalink / raw)
  To: Linus Torvalds; +Cc: linux-kernel, Andrew Morton, Thomas Gleixner

Linus,

Please pull the latest timers-fixes-for-linus git tree from:

   git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip.git timers-fixes-for-linus

Thanks,

	Ingo

------------------>
Miao Xie (1):
      nohz: fix wrong event handler after online an offlined cpu


 include/linux/tick.h     |    5 ++++-
 kernel/time/tick-sched.c |    6 +++++-
 2 files changed, 9 insertions(+), 2 deletions(-)

diff --git a/include/linux/tick.h b/include/linux/tick.h
index d3c0269..8cf8cfe 100644
--- a/include/linux/tick.h
+++ b/include/linux/tick.h
@@ -74,10 +74,13 @@ extern struct tick_device *tick_get_device(int cpu);
 extern int tick_init_highres(void);
 extern int tick_program_event(ktime_t expires, int force);
 extern void tick_setup_sched_timer(void);
+# endif
+
+# if defined CONFIG_NO_HZ || defined CONFIG_HIGH_RES_TIMERS
 extern void tick_cancel_sched_timer(int cpu);
 # else
 static inline void tick_cancel_sched_timer(int cpu) { }
-# endif /* HIGHRES */
+# endif
 
 # ifdef CONFIG_GENERIC_CLOCKEVENTS_BROADCAST
 extern struct tick_device *tick_get_broadcast_device(void);
diff --git a/kernel/time/tick-sched.c b/kernel/time/tick-sched.c
index f5da526..7a46bde 100644
--- a/kernel/time/tick-sched.c
+++ b/kernel/time/tick-sched.c
@@ -643,17 +643,21 @@ void tick_setup_sched_timer(void)
 		ts->nohz_mode = NOHZ_MODE_HIGHRES;
 #endif
 }
+#endif /* HIGH_RES_TIMERS */
 
+#if defined CONFIG_NO_HZ || defined CONFIG_HIGH_RES_TIMERS
 void tick_cancel_sched_timer(int cpu)
 {
 	struct tick_sched *ts = &per_cpu(tick_cpu_sched, cpu);
 
+# ifdef CONFIG_HIGH_RES_TIMERS
 	if (ts->sched_timer.base)
 		hrtimer_cancel(&ts->sched_timer);
+# endif
 
 	ts->nohz_mode = NOHZ_MODE_INACTIVE;
 }
-#endif /* HIGH_RES_TIMERS */
+#endif
 
 /**
  * Async notification about clocksource changes

^ permalink raw reply related	[flat|nested] 69+ messages in thread

* [git pull] timer fixes
@ 2008-07-24 15:18 Ingo Molnar
  0 siblings, 0 replies; 69+ messages in thread
From: Ingo Molnar @ 2008-07-24 15:18 UTC (permalink / raw)
  To: Linus Torvalds; +Cc: linux-kernel, Thomas Gleixner, Andrew Morton

Linus,

Please pull the latest timer fixes git tree from:

   git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip.git timers-fixes-for-linus

Thanks,

	Ingo

------------------>
Thomas Gleixner (2):
      nohz: prevent tick stop outside of the idle loop
      nohz: adjust tick_nohz_stop_sched_tick() call of s390 as well


 arch/arm/kernel/process.c              |    2 +-
 arch/avr32/kernel/process.c            |    2 +-
 arch/blackfin/kernel/process.c         |    2 +-
 arch/mips/kernel/process.c             |    2 +-
 arch/powerpc/kernel/idle.c             |    2 +-
 arch/powerpc/platforms/iseries/setup.c |    4 ++--
 arch/s390/kernel/process.c             |    2 +-
 arch/sh/kernel/process_32.c            |    2 +-
 arch/sparc64/kernel/process.c          |    2 +-
 arch/um/kernel/process.c               |    2 +-
 arch/x86/kernel/process_32.c           |    2 +-
 arch/x86/kernel/process_64.c           |    2 +-
 include/linux/tick.h                   |    5 +++--
 kernel/softirq.c                       |    2 +-
 kernel/time/tick-sched.c               |   12 ++++++++++--
 15 files changed, 27 insertions(+), 18 deletions(-)

diff --git a/arch/arm/kernel/process.c b/arch/arm/kernel/process.c
index 199b368..89bfded 100644
--- a/arch/arm/kernel/process.c
+++ b/arch/arm/kernel/process.c
@@ -162,7 +162,7 @@ void cpu_idle(void)
 		if (!idle)
 			idle = default_idle;
 		leds_event(led_idle_start);
-		tick_nohz_stop_sched_tick();
+		tick_nohz_stop_sched_tick(1);
 		while (!need_resched())
 			idle();
 		leds_event(led_idle_end);
diff --git a/arch/avr32/kernel/process.c b/arch/avr32/kernel/process.c
index 6cf9df1..ff820a9 100644
--- a/arch/avr32/kernel/process.c
+++ b/arch/avr32/kernel/process.c
@@ -31,7 +31,7 @@ void cpu_idle(void)
 {
 	/* endless idle loop with no priority at all */
 	while (1) {
-		tick_nohz_stop_sched_tick();
+		tick_nohz_stop_sched_tick(1);
 		while (!need_resched())
 			cpu_idle_sleep();
 		tick_nohz_restart_sched_tick();
diff --git a/arch/blackfin/kernel/process.c b/arch/blackfin/kernel/process.c
index 53c2cd2..77800dd 100644
--- a/arch/blackfin/kernel/process.c
+++ b/arch/blackfin/kernel/process.c
@@ -105,7 +105,7 @@ void cpu_idle(void)
 #endif
 		if (!idle)
 			idle = default_idle;
-		tick_nohz_stop_sched_tick();
+		tick_nohz_stop_sched_tick(1);
 		while (!need_resched())
 			idle();
 		tick_nohz_restart_sched_tick();
diff --git a/arch/mips/kernel/process.c b/arch/mips/kernel/process.c
index c06f5b5..b16facd 100644
--- a/arch/mips/kernel/process.c
+++ b/arch/mips/kernel/process.c
@@ -53,7 +53,7 @@ void __noreturn cpu_idle(void)
 {
 	/* endless idle loop with no priority at all */
 	while (1) {
-		tick_nohz_stop_sched_tick();
+		tick_nohz_stop_sched_tick(1);
 		while (!need_resched()) {
 #ifdef CONFIG_SMTC_IDLE_HOOK_DEBUG
 			extern void smtc_idle_loop_hook(void);
diff --git a/arch/powerpc/kernel/idle.c b/arch/powerpc/kernel/idle.c
index c3cf0e8..d308a9f 100644
--- a/arch/powerpc/kernel/idle.c
+++ b/arch/powerpc/kernel/idle.c
@@ -60,7 +60,7 @@ void cpu_idle(void)
 
 	set_thread_flag(TIF_POLLING_NRFLAG);
 	while (1) {
-		tick_nohz_stop_sched_tick();
+		tick_nohz_stop_sched_tick(1);
 		while (!need_resched() && !cpu_should_die()) {
 			ppc64_runlatch_off();
 
diff --git a/arch/powerpc/platforms/iseries/setup.c b/arch/powerpc/platforms/iseries/setup.c
index b721207..70b688c 100644
--- a/arch/powerpc/platforms/iseries/setup.c
+++ b/arch/powerpc/platforms/iseries/setup.c
@@ -561,7 +561,7 @@ static void yield_shared_processor(void)
 static void iseries_shared_idle(void)
 {
 	while (1) {
-		tick_nohz_stop_sched_tick();
+		tick_nohz_stop_sched_tick(1);
 		while (!need_resched() && !hvlpevent_is_pending()) {
 			local_irq_disable();
 			ppc64_runlatch_off();
@@ -591,7 +591,7 @@ static void iseries_dedicated_idle(void)
 	set_thread_flag(TIF_POLLING_NRFLAG);
 
 	while (1) {
-		tick_nohz_stop_sched_tick();
+		tick_nohz_stop_sched_tick(1);
 		if (!need_resched()) {
 			while (!need_resched()) {
 				ppc64_runlatch_off();
diff --git a/arch/s390/kernel/process.c b/arch/s390/kernel/process.c
index 85defd0..9839767 100644
--- a/arch/s390/kernel/process.c
+++ b/arch/s390/kernel/process.c
@@ -142,7 +142,7 @@ static void default_idle(void)
 void cpu_idle(void)
 {
 	for (;;) {
-		tick_nohz_stop_sched_tick();
+		tick_nohz_stop_sched_tick(1);
 		while (!need_resched())
 			default_idle();
 		tick_nohz_restart_sched_tick();
diff --git a/arch/sh/kernel/process_32.c b/arch/sh/kernel/process_32.c
index b98e37a..921892c 100644
--- a/arch/sh/kernel/process_32.c
+++ b/arch/sh/kernel/process_32.c
@@ -86,7 +86,7 @@ void cpu_idle(void)
 		if (!idle)
 			idle = default_idle;
 
-		tick_nohz_stop_sched_tick();
+		tick_nohz_stop_sched_tick(1);
 		while (!need_resched())
 			idle();
 		tick_nohz_restart_sched_tick();
diff --git a/arch/sparc64/kernel/process.c b/arch/sparc64/kernel/process.c
index 2084f81..0798928 100644
--- a/arch/sparc64/kernel/process.c
+++ b/arch/sparc64/kernel/process.c
@@ -97,7 +97,7 @@ void cpu_idle(void)
 	set_thread_flag(TIF_POLLING_NRFLAG);
 
 	while(1) {
-		tick_nohz_stop_sched_tick();
+		tick_nohz_stop_sched_tick(1);
 
 		while (!need_resched() && !cpu_is_offline(cpu))
 			sparc64_yield(cpu);
diff --git a/arch/um/kernel/process.c b/arch/um/kernel/process.c
index 83603cf..a1c6d07 100644
--- a/arch/um/kernel/process.c
+++ b/arch/um/kernel/process.c
@@ -243,7 +243,7 @@ void default_idle(void)
 		if (need_resched())
 			schedule();
 
-		tick_nohz_stop_sched_tick();
+		tick_nohz_stop_sched_tick(1);
 		nsecs = disable_timer();
 		idle_sleep(nsecs);
 		tick_nohz_restart_sched_tick();
diff --git a/arch/x86/kernel/process_32.c b/arch/x86/kernel/process_32.c
index 0c3927a..53bc653 100644
--- a/arch/x86/kernel/process_32.c
+++ b/arch/x86/kernel/process_32.c
@@ -128,7 +128,7 @@ void cpu_idle(void)
 
 	/* endless idle loop with no priority at all */
 	while (1) {
-		tick_nohz_stop_sched_tick();
+		tick_nohz_stop_sched_tick(1);
 		while (!need_resched()) {
 
 			check_pgt_cache();
diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c
index a8e5362..9a10c18 100644
--- a/arch/x86/kernel/process_64.c
+++ b/arch/x86/kernel/process_64.c
@@ -120,7 +120,7 @@ void cpu_idle(void)
 	current_thread_info()->status |= TS_POLLING;
 	/* endless idle loop with no priority at all */
 	while (1) {
-		tick_nohz_stop_sched_tick();
+		tick_nohz_stop_sched_tick(1);
 		while (!need_resched()) {
 
 			rmb();
diff --git a/include/linux/tick.h b/include/linux/tick.h
index a881c65..d3c0269 100644
--- a/include/linux/tick.h
+++ b/include/linux/tick.h
@@ -49,6 +49,7 @@ struct tick_sched {
 	unsigned long			check_clocks;
 	enum tick_nohz_mode		nohz_mode;
 	ktime_t				idle_tick;
+	int				inidle;
 	int				tick_stopped;
 	unsigned long			idle_jiffies;
 	unsigned long			idle_calls;
@@ -105,14 +106,14 @@ static inline int tick_check_oneshot_change(int allow_nohz) { return 0; }
 #endif /* !CONFIG_GENERIC_CLOCKEVENTS */
 
 # ifdef CONFIG_NO_HZ
-extern void tick_nohz_stop_sched_tick(void);
+extern void tick_nohz_stop_sched_tick(int inidle);
 extern void tick_nohz_restart_sched_tick(void);
 extern void tick_nohz_update_jiffies(void);
 extern ktime_t tick_nohz_get_sleep_length(void);
 extern void tick_nohz_stop_idle(int cpu);
 extern u64 get_cpu_idle_time_us(int cpu, u64 *last_update_time);
 # else
-static inline void tick_nohz_stop_sched_tick(void) { }
+static inline void tick_nohz_stop_sched_tick(int inidle) { }
 static inline void tick_nohz_restart_sched_tick(void) { }
 static inline void tick_nohz_update_jiffies(void) { }
 static inline ktime_t tick_nohz_get_sleep_length(void)
diff --git a/kernel/softirq.c b/kernel/softirq.c
index 81e2fe0..f6b03d5 100644
--- a/kernel/softirq.c
+++ b/kernel/softirq.c
@@ -286,7 +286,7 @@ void irq_exit(void)
 #ifdef CONFIG_NO_HZ
 	/* Make sure that timer wheel updates are propagated */
 	if (!in_interrupt() && idle_cpu(smp_processor_id()) && !need_resched())
-		tick_nohz_stop_sched_tick();
+		tick_nohz_stop_sched_tick(0);
 	rcu_irq_exit();
 #endif
 	preempt_enable_no_resched();
diff --git a/kernel/time/tick-sched.c b/kernel/time/tick-sched.c
index beef7cc..a5c26d2 100644
--- a/kernel/time/tick-sched.c
+++ b/kernel/time/tick-sched.c
@@ -195,7 +195,7 @@ u64 get_cpu_idle_time_us(int cpu, u64 *last_update_time)
  * Called either from the idle loop or from irq_exit() when an idle period was
  * just interrupted by an interrupt which did not cause a reschedule.
  */
-void tick_nohz_stop_sched_tick(void)
+void tick_nohz_stop_sched_tick(int inidle)
 {
 	unsigned long seq, last_jiffies, next_jiffies, delta_jiffies, flags;
 	struct tick_sched *ts;
@@ -224,6 +224,11 @@ void tick_nohz_stop_sched_tick(void)
 	if (unlikely(ts->nohz_mode == NOHZ_MODE_INACTIVE))
 		goto end;
 
+	if (!inidle && !ts->inidle)
+		goto end;
+
+	ts->inidle = 1;
+
 	if (need_resched())
 		goto end;
 
@@ -373,11 +378,14 @@ void tick_nohz_restart_sched_tick(void)
 	local_irq_disable();
 	tick_nohz_stop_idle(cpu);
 
-	if (!ts->tick_stopped) {
+	if (!ts->inidle || !ts->tick_stopped) {
+		ts->inidle = 0;
 		local_irq_enable();
 		return;
 	}
 
+	ts->inidle = 0;
+
 	rcu_exit_nohz();
 
 	/* Update jiffies first */

^ permalink raw reply related	[flat|nested] 69+ messages in thread

end of thread, other threads:[~2023-01-12 23:01 UTC | newest]

Thread overview: 69+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2022-08-13 10:25 [GIT PULL] timer fixes Ingo Molnar
2022-08-13 16:25 ` Borislav Petkov
2022-08-13 20:27   ` Linus Torvalds
2022-08-14  8:57     ` Ingo Molnar
2022-08-14 12:41     ` Borislav Petkov
2022-08-14 17:24       ` Paul E. McKenney
2022-08-14 17:42         ` Borislav Petkov
2022-08-13 21:48 ` pr-tracker-bot
  -- strict thread matches above, loose matches on Subject: below --
2023-01-12 14:44 Ingo Molnar
2023-01-12 23:01 ` pr-tracker-bot
2020-12-27  9:26 Ingo Molnar
2020-12-27 17:27 ` pr-tracker-bot
2019-12-17 11:55 Ingo Molnar
2019-12-17 19:14 ` Linus Torvalds
2019-12-17 19:30   ` Peter Zijlstra
2019-12-17 20:16     ` Linus Torvalds
2019-12-17 20:35       ` Peter Zijlstra
2019-12-17 20:43         ` Linus Torvalds
2019-12-17 20:54           ` Peter Zijlstra
2019-12-17 19:20 ` pr-tracker-bot
2019-05-16 16:09 Ingo Molnar
2019-05-16 18:20 ` pr-tracker-bot
2019-04-20  7:45 Ingo Molnar
2019-04-20 19:25 ` pr-tracker-bot
2018-07-13 20:01 Ingo Molnar
2017-03-07 20:36 Ingo Molnar
2016-10-28  8:38 Ingo Molnar
2016-08-18 20:46 Ingo Molnar
2016-08-12 19:43 Ingo Molnar
2016-01-14 10:12 Ingo Molnar
2015-10-03 10:20 Ingo Molnar
2015-09-17  8:13 Ingo Molnar
2015-03-28 13:50 Ingo Molnar
2015-03-17 16:50 Ingo Molnar
2015-03-01 17:06 Ingo Molnar
2013-12-17 13:52 Ingo Molnar
2013-09-28 18:19 Ingo Molnar
2013-08-19  9:41 Ingo Molnar
2013-02-26 11:32 Ingo Molnar
2012-09-21 19:12 Ingo Molnar
2012-08-03 16:47 Ingo Molnar
2011-12-20 19:25 Ingo Molnar
2011-12-05 19:00 Ingo Molnar
2011-06-19  8:51 Ingo Molnar
2011-06-13  9:55 Ingo Molnar
2011-06-06 19:10 [GIT pull] " Thomas Gleixner
2011-05-20 17:52 Thomas Gleixner
2011-04-19 16:03 [GIT PULL] " Ingo Molnar
2011-04-16 10:09 Ingo Molnar
2011-04-07 17:42 Ingo Molnar
2011-02-06 12:06 Ingo Molnar
2011-01-24 13:24 Ingo Molnar
2010-11-26 13:40 Ingo Molnar
2010-03-26 15:34 Ingo Molnar
2009-12-18 18:49 Ingo Molnar
2009-10-08 18:35 Ingo Molnar
2009-08-25 18:04 Ingo Molnar
2009-06-26 19:03 Ingo Molnar
2009-04-09 15:43 Ingo Molnar
2009-01-30 23:07 [git pull] " Ingo Molnar
2009-01-26 17:21 Ingo Molnar
2009-01-06 16:18 Ingo Molnar
2008-12-04 20:52 Ingo Molnar
2008-11-11 18:47 Ingo Molnar
2008-09-23 19:41 Ingo Molnar
2008-09-17  9:59 Ingo Molnar
2008-08-25 17:52 Ingo Molnar
2008-08-22 12:24 Ingo Molnar
2008-07-24 15:18 Ingo Molnar

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.