2012/11/26 Steven Rostedt :
> On Fri, 2012-11-23 at 15:21 +0100, Frederic Weisbecker wrote:
>> We have thread_group_cputime() and thread_group_times(). The naming
>> doesn't provide enough information about the difference between
>> these two APIs.
>>
&
2012/11/6 Gilad Ben-Yossef :
> On Sat, Nov 3, 2012 at 6:09 PM, Frederic Weisbecker
> wrote:
>> diff --git a/arch/Kconfig b/arch/Kconfig
>> index 366ec06..3855e06 100644
>> --- a/arch/Kconfig
>> +++ b/arch/Kconfig
>> @@ -300,15 +300,15 @@ config SECCOMP_FILT
2012/11/27 Li Zhong :
> I noticed some warnings complaining about dynticks_nesting value, like
>
> [ 267.545032] [ cut here ]
> [ 267.545032] WARNING: at kernel/rcutree.c:382 rcu_eqs_enter+0xab/0xc0()
> [ 267.545032] Hardware name: Bochs
> [ 267.545032] Modules linked in
2012/11/27 Gleb Natapov :
> On Tue, Nov 27, 2012 at 01:15:25PM +0800, Li Zhong wrote:
>> @@ -247,10 +247,17 @@ do_async_page_fault(struct pt_regs *regs, unsigned
>> long error_code)
>> break;
>> case KVM_PV_REASON_PAGE_NOT_PRESENT:
>> /* page is swapped out by the
2012/11/27 Li Zhong :
> I noticed some warnings complaining about dynticks_nesting value, like
>
> [ 267.545032] [ cut here ]
> [ 267.545032] WARNING: at kernel/rcutree.c:382 rcu_eqs_enter+0xab/0xc0()
> [ 267.545032] Hardware name: Bochs
> [ 267.545032] Modules linked in
2012/11/27 Gleb Natapov :
> For KVM_PV_REASON_PAGE_NOT_PRESENT it behaves like an exception.
Ok.
There seem to be a bug in kvm_async_pf_task_wait(). Using
idle_cpu(cpu) to find out if the current task is the idle task may not
work if there is pending wake up. Me may schedule another task but
when
2012/11/27 Li Zhong :
> @@ -247,10 +247,17 @@ do_async_page_fault(struct pt_regs *regs, unsigned long
> error_code)
> break;
> case KVM_PV_REASON_PAGE_NOT_PRESENT:
> /* page is swapped out by the host. */
> - rcu_irq_enter();
> +
2012/11/27 Paul E. McKenney :
> It is OK to call rcu_irq_exit() without a matching rcu_irq_enter() -only-
> if you have also called rcu_idle_exit() since the last rcu_idle_enter().
> There will be a similar rule for rcu_user_exit().
>
> More generally, it is OK to call rcu_irq_exit() without a matc
2012/11/27 Li Zhong :
> I noticed some warnings complaining about dynticks_nesting value, like
>
> [ 267.545032] [ cut here ]
> [ 267.545032] WARNING: at kernel/rcutree.c:382 rcu_eqs_enter+0xab/0xc0()
> [ 267.545032] Hardware name: Bochs
> [ 267.545032] Modules linked in
2012/11/27 Paul E. McKenney :
> On Tue, Nov 27, 2012 at 04:56:30PM +0100, Frederic Weisbecker wrote:
>> 2012/11/27 Gleb Natapov :
>> > For KVM_PV_REASON_PAGE_NOT_PRESENT it behaves like an exception.
>>
>> Ok.
>> There seem to be a bug in kvm_async_pf_task_wa
2012/11/27 Gleb Natapov :
> On Tue, Nov 27, 2012 at 04:56:30PM +0100, Frederic Weisbecker wrote:
>> 2012/11/27 Gleb Natapov :
>> > For KVM_PV_REASON_PAGE_NOT_PRESENT it behaves like an exception.
>>
>> Ok.
>> There seem to be a bug in kvm_async_pf_task_wait(). U
2012/11/27 Gleb Natapov :
> diff --git a/arch/x86/kernel/kvm.c b/arch/x86/kernel/kvm.c
> index 4180a87..636800d 100644
> --- a/arch/x86/kernel/kvm.c
> +++ b/arch/x86/kernel/kvm.c
> @@ -113,7 +113,7 @@ void kvm_async_pf_task_wait(u32 token)
> int cpu, idle;
>
> cpu = get_cpu();
> -
2012/11/27 Gleb Natapov :
> On Tue, Nov 27, 2012 at 06:30:32PM +0100, Frederic Weisbecker wrote:
>> 2012/11/27 Gleb Natapov :
>> > diff --git a/arch/x86/kernel/kvm.c b/arch/x86/kernel/kvm.c
>> > index 4180a87..636800d 100644
>> > --- a/arch/x86/kernel/kv
2012/11/27 Gleb Natapov :
> What is the semantics of enter_idle()/exit_idle(), what are they used for?
It's used by drivers/idle/i7300_idle.c for some tracking. I don't know much
the details.
enter_idle() is called right before the CPU is set to lower power mode: hlt()
exit_idle() is called anyt
2012/11/27 Frederic Weisbecker :
> 2012/11/27 Gleb Natapov :
>> What is the semantics of enter_idle()/exit_idle(), what are they used for?
>
> It's used by drivers/idle/i7300_idle.c for some tracking. I don't know much
> the details.
>
> enter_idle() is called ri
2012/11/26 Steven Rostedt :
> OK, let's take a look at the other version now:
>
> void thread_group_times(struct task_struct *p, cputime_t *ut, cputime_t *st)
So this does the same thing than thread_group_cputime(), ie: fetch the
raw cputime stats from the task/signal struct, with a two adjustment
2012/11/28 Li Zhong :
> Thank you all for the review and education.
>
> Below are my current understandings and an update version. Would you
> please help to review it again and give your comments?
>
> Thanks, Zhong
>
> Now it seems to me that it is legal to call rcu_irq_exit/enter() without
> a ma
2012/11/28 Gleb Natapov :
> On Wed, Nov 28, 2012 at 01:55:42PM +0100, Frederic Weisbecker wrote:
>> Yes but if rcu_irq_*() calls are fine to be called there, and I
>> believe they are because exception_enter() exits the user mode, we
>> should start to protect there right n
Hi,
Changes since v1 address Steven Rostedt and Paul Gortmaker reviews:
- More comments to distinguish struct cputime / struct task_cputime [3/4]
- Comment the reasons and the details for cputime adjustment [4/4]
Thanks.
Frederic Weisbecker (4):
cputime: Move thread_group_cputime() to sched
thread_group_cputime() is a general cputime API that is not only
used by posix cpu timer. Let's move this helper to sched code.
Signed-off-by: Frederic Weisbecker
Cc: Ingo Molnar
Cc: Peter Zijlstra
Cc: Thomas Gleixner
Cc: Steven Rostedt
Cc: Paul Gortmaker
---
kernel/posix-cpu-tim
ks in the group and the
previous adjusted snapshot of the whole group from the signal
structure.
Just consolidate the common code that does the adjustment. These
functions just need to fetch the values from the appropriate
source.
Signed-off-by: Frederic Weisbecker
Cc: Ingo Molnar
Cc: Peter Zij
version of thread_group_cputime() that does some stabilization
on the raw cputime values. ie here: scale on top of CFS runtime
stats and bound lower value for monotonicity.
Signed-off-by: Frederic Weisbecker
Cc: Ingo Molnar
Cc: Peter Zijlstra
Cc: Thomas Gleixner
Cc: Steven Rostedt
Cc: Paul Gortmaker
--
The reason for the scaling and monotonicity correction performed
by cputime_adjust() may not be immediately clear to the reviewer.
Add some comments to explain what happens there.
Signed-off-by: Frederic Weisbecker
Cc: Ingo Molnar
Cc: Peter Zijlstra
Cc: Thomas Gleixner
Cc: Steven Rostedt
Cc
2012/11/29 Li Zhong :
> On Tue, 2012-11-27 at 19:33 +0100, Frederic Weisbecker wrote:
> []
>> -
>> - WARN_ON_ONCE(!current->mm);
>> -
>> - local_irq_save(flags);
>> - rdtp = &__get_cpu_var(rcu_dynticks);
>> - if (!rdtp->ignor
2012/11/29 Li Zhong :
> On Wed, 2012-11-28 at 13:55 +0100, Frederic Weisbecker wrote:
>> > With rcu_user_exit() at the beginning, now rcu_irq_enter() only protects
>> > the cpu idle eqs, but it's not good to call rcu_irq_exit() after the cpu
>> > halt and the p
2012/11/29 Gleb Natapov :
> On Wed, Nov 28, 2012 at 03:25:07PM +0100, Frederic Weisbecker wrote:
>> 2012/11/28 Gleb Natapov :
>> > On Wed, Nov 28, 2012 at 01:55:42PM +0100, Frederic Weisbecker wrote:
>> >> Yes but if rcu_irq_*() calls are fine to be called there
* Comment the adjustement code
Signed-off-by: Frederic Weisbecker
----
Frederic Weisbecker (4):
cputime: Move thread_group_cputime() to sched code
cputime: Rename thread_group_times to thread_group_cputime_adjusted
2013/3/18 Viresh Kumar :
> In order to save power, it would be useful to schedule light weight work on
> cpus
> that aren't IDLE instead of waking up an IDLE one.
>
> By idle cpu (from scheduler's perspective) we mean:
> - Current task is idle task
> - nr_running == 0
> - wake_list is empty
>
> Th
2013/3/18 Viresh Kumar :
> On Mon, Mar 18, 2013 at 9:09 PM, Frederic Weisbecker
> wrote:
>> 2013/3/18 Viresh Kumar :
>
>>> +static inline int sched_select_cpu(unsigned int sd_flags)
>>> +{
>>> + return raw_smp_processor_id();
>>
>>
2013/3/18 Rob Landley :
> On 03/18/2013 11:29:42 AM, Paul E. McKenney wrote:
> And really seems like it's kconfig help text?
It's more exhaustive than a Kconfig help. A Kconfig help text should
have the level of detail that describe the purpose and impact of a
feature, as well as some quick refere
2013/3/18 Rob Landley :
> On 03/18/2013 01:46:32 PM, Frederic Weisbecker wrote:
>> I really think we want to keep all the detailed explanations from
>> Paul's doc. What we need is not a quick reference but a very detailed
>> documentation.
>
>
> It's
2013/3/18 Andrew Morton :
> On Fri, 15 Mar 2013 17:54:02 +0100 Frederic Weisbecker
> wrote:
>
>> Add some initial basic tests on a few posix timers
>> interface such as setitimer() and timer_settime().
>>
>> These simply check that expiration happens in a reaso
2013/3/19 Andrew Morton :
> On Fri, 15 Mar 2013 17:53:59 +0100 Frederic Weisbecker
> wrote:
>
>> The posix cpu timer expiry time is stored in a union of
>> two types: a 64 bits field if we rely on scheduler precise
>> accounting, or a cputime_t if we rely on jiffies.
2013/3/20 David Miller :
> From: Chris Metcalf
> Date: Tue, 19 Mar 2013 17:35:58 -0400
>
>> Previously, if you did an "ifconfig down" or similar on one core, and
>> the kernel had CONFIG_XFRM enabled, every core would be interrupted to
>> check its percpu flow list for items that could be garbage
.
Sorry for the missing credits, I lost a bit track...
Thanks.
---
Frederic Weisbecker (3):
nohz: Basic full dynticks interface
nohz: Assign timekeeping duty to a CPU outside the full dynticks
range
nohz: Wake up full dynticks CPUs when a timer gets enqueued
Documentation/kernel
sted-by: Paul E. McKenney
Signed-off-by: Frederic Weisbecker
Cc: Andrew Morton
Cc: Chris Metcalf
Cc: Christoph Lameter
Cc: Geoff Levand
Cc: Gilad Ben Yossef
Cc: Hakan Akkan
Cc: Ingo Molnar
Cc: Kevin Hilman
Cc: Li Zhong
Cc: Namhyung Kim
Cc: Paul E. McKenney
Cc: Paul Gortmaker
Cc: Peter Zijls
imple approach first.
Signed-off-by: Frederic Weisbecker
Cc: Andrew Morton
Cc: Chris Metcalf
Cc: Christoph Lameter
Cc: Geoff Levand
Cc: Gilad Ben Yossef
Cc: Hakan Akkan
Cc: Ingo Molnar
Cc: Kevin Hilman
Cc: Li Zhong
Cc: Namhyung Kim
Cc: Paul E. McKenney
Cc: Paul Gortmaker
Cc: Peter Zijlstra
need some careful
synchronization though. So deal with such optimization later
and start simple.
Signed-off-by: Frederic Weisbecker
Cc: Andrew Morton
Cc: Chris Metcalf
Cc: Christoph Lameter
Cc: Geoff Levand
Cc: Gilad Ben Yossef
Cc: Hakan Akkan
Cc: Ingo Molnar
Cc: Kevin Hilman
Cc: Li Zhong
From: anish kumar
As no one is using the return value of irq_work_queue function
it is better to just make it void.
Acked-by: Steven Rostedt
Signed-off-by: anish kumar
[fix stale comments, remove now unnecessary __irq_work_queue
intermediate function]
Signed-off-by: Frederic Weisbecker
IRT_CPU_ACCOUNTING except that context location
and cputime snaphots are synchronized between write and read
side such that the latter can safely retrieve the pending tickless
cputime of a task and add it to its latest cputime snapshot to
return the correct result to the user.
Signed-of
just made a new branch
that contains tip:irq/core with nohz/printk-v8 merged inside which deals
with the conflict.
Thanks.
---
Frederic Weisbecker (5):
nohz: Add API to check tick state
irq_work: Don't stop the tick with pending works
irq_work: Make self-IPIs op
2013/2/5 Andrew Morton :
> On Tue, 5 Feb 2013 01:51:18 +0100
> Frederic Weisbecker wrote:
>
>> printk: Wake up klogd using irq_work
>
> That seems reasonable.
>
> I'm wondering if we can now remove the printk_sched() special-case.
> iirc, that was
2013/2/5 Steven Rostedt :
> On Mon, 2013-02-04 at 18:09 -0800, Andrew Morton wrote:
>
>> Maybe there were other deadlock scenarios, dunno. That knowledge
>> appears to be disappearing into the mists of time :(
>
> Discussing this on IRC with Frederic, he brought up that the
> up(&console_sem) can
2013/2/5 Andrew Morton :
> On Mon, 4 Feb 2013 23:17:10 +0100
> Jan Kara wrote:
>
>> A CPU can be caught in console_unlock() for a long time (tens of seconds are
>> reported by our customers) when other CPUs are using printk heavily and
>> serial
>> console makes printing slow. Despite serial con
2013/2/6 Ingo Molnar :
>
> btw., I got a non-trivial conflicts in
> kernel/context_tracking.c, with the following commits
> interacting between the RCU tree (tip:core/rcu) and the
> scheduler tree (tip:sched/core):
>
> 6a61671bb2f3 cputime: Safely read cputime of full dynticks CPUs
> abf917cd91cb
applied in -tip
---
Frederic Weisbecker (26):
nohz: Basic full dynticks interface
nohz: Assign timekeeping duty to a non-full-nohz CPU
nohz: Trace timekeeping update
nohz: Wake up full dynticks CPUs when a timer gets enqueued
rcu: Restart the tick on non-responding full dynticks CPUs
sched: Comme
2013/2/6 Daniel Baluta :
> Obviously this is a typo and could result in memory leaks
> if kzalloc fails on a given cpu.
>
> Signed-off-by: Daniel Baluta
Good catch!
Acked-by: Frederic Weisbecker
> ---
> kernel/events/hw_breakpoint.c |2 +-
> 1 files changed, 1 inser
2013/2/1 Li Zhong :
> This is the syscall slow path hooks for context tracking subsystem,
> corresponding to
> [PATCH] x86: Syscall hooks for userspace RCU extended QS
> commit bf5a3c13b939813d28ce26c01425054c740d6731
>
> TIF_MEMDIE is moved to the second 16-bits (with value 17), as it seems ther
2013/2/7 Li Zhong :
> On Thu, 2013-02-07 at 01:29 +0100, Frederic Weisbecker wrote:
>> In x86-64, schedule_user() and do_notify_resume() can be called before
>> syscall_trace_leave(). As a result we may be entering
>> syscall_trace_leave() in user mode (from a context tracking
2013/2/7 Ingo Molnar :
>
> * Steven Rostedt wrote:
>
>> I'll reply to this as I come up with comments.
>>
>> First thing is, don't call it NO_HZ_FULL. A better name would
>> be NO_HZ_CPU. I would like to reserve NO_HZ_FULL when we
>> totally remove jiffies :-)
>
> I don't think we want yet another
2013/2/7 Steven Rostedt :
> I'll reply to this as I come up with comments.
>
> First thing is, don't call it NO_HZ_FULL. A better name would be
> NO_HZ_CPU. I would like to reserve NO_HZ_FULL when we totally remove
> jiffies :-)
I'm not sure we'll ever be able to completely remove the tick, even i
2013/2/7 Steven Rostedt :
> On Thu, 2013-02-07 at 17:25 +0100, Frederic Weisbecker wrote:
>
>> At least for now we seem to agree on CONFIG_NO_HZ_IDLE and keep
>> CONFIG_NO_HZ for compatibility. Are you ok with that? If so I'll send
>> a patch.
>
> I belie
2013/2/7 Steven Rostedt :
> On Thu, 2013-02-07 at 17:41 +0100, Frederic Weisbecker wrote:
>
>> I'm not convinced that "single task" must be a fundamental component
>> of this. It's an implementation detail. We should be able to keep the
>> tick off in t
2013/2/7 Steven Rostedt :
> On Thu, 2013-02-07 at 17:45 +0100, Frederic Weisbecker wrote:
>
>> >
>> > config NO_HZ_TASK
>> > bool "Stop tick on specified CPUs when single task is running"
>> > default n
>> > d
2013/2/8 Stephen Rothwell :
> Hi Frederic,
>
> On Fri, 8 Feb 2013 14:07:49 +1100 Stephen Rothwell
> wrote:
>>
>> This patch has the side effect of changing the default configurations:
>> (This is PowerPC pseries_defconfig before/after this patch)
>>
>> @@ -119,8 +120,8 @@
>> #
>> # CPU/Task tim
2013/2/4 Vincent Guittot :
> On 1 February 2013 19:03, Frederic Weisbecker wrote:
>>> diff --git a/kernel/sched/core.c b/kernel/sched/core.c
>>> index 257002c..fd41924 100644
>>> --- a/kernel/sched/core.c
>>> +++ b/kernel/sched/core.c
>>> @@ -5884,
2013/2/7 Ingo Molnar :
>
> * Frederic Weisbecker wrote:
>
>> 2013/2/7 Ingo Molnar :
>> >
>> > * Steven Rostedt wrote:
>> >
>> >> I'll reply to this as I come up with comments.
>> >>
>> >> First thing is, don&#x
2013/2/7 Christoph Lameter :
> On Thu, 7 Feb 2013, Frederic Weisbecker wrote:
>
>> Not with hrtick.
>
> hrtick? Did we not already try that a couple of years back and it turned
> out that the overhead of constantly reprogramming a timer via the PCI bus
> was causing t
e 17), as it seems there
> is no asm code using it. TIF_NOHZ is added to _TIF_SYCALL_T_OR_A, so it is
> better for it to be in the same 16 bits with others in the group, so in the
> asm code, andi. with this group could work.
>
> Signed-off-by: Li Zhong
Looks good. Thanks.
Acked-by: Fr
2013/2/1 Li Zhong :
> This is the exception hooks for context tracking subsystem, including
> data access, program check, single step, instruction breakpoint, machine
> check,
> alignment, fp unavailable, altivec assist, unknown exception, whose handlers
> might use RCU.
>
> This patch corresponds
2013/3/30 Ingo Molnar :
>
> * Frederic Weisbecker wrote:
>
>> 2013/3/28 Ingo Molnar :
>> >
>> > * Frederic Weisbecker wrote:
>> >
>> >> +config NO_HZ_EXTENDED_ALL
>> >> + bool "Full dynticks system on all CPUs"
&
2013/4/3 Mel Gorman :
> Commit ba6fdda4 (profiling: Remove unused timer hook) removed
> [un]register_timer_hook due to a lack of upstream users and a belief
> that there were no out-of-tree users. However, systemtap uses it and
> with that patch applied, some stap scripts fail with
>
> WARNING: "un
r backward compatibility.
Signed-off-by: Frederic Weisbecker
Cc: Andrew Morton
Cc: Chris Metcalf
Cc: Christoph Lameter
Cc: Geoff Levand
Cc: Gilad Ben Yossef
Cc: Hakan Akkan
Cc: Ingo Molnar
Cc: Kevin Hilman
Cc: Li Zhong
Cc: Namhyung Kim
Cc: Paul E. McKenney
Cc: Paul Gortmaker
Cc: Peter Zijlstra
ing to run.
Signed-off-by: Frederic Weisbecker
Cc: Andrew Morton
Cc: Chris Metcalf
Cc: Christoph Lameter
Cc: Geoff Levand
Cc: Gilad Ben Yossef
Cc: Hakan Akkan
Cc: Ingo Molnar
Cc: Kevin Hilman
Cc: Li Zhong
Cc: Namhyung Kim
Cc: Paul E. McKenney
Cc: Paul Gortmaker
Cc: Peter Zijlstra
kernel/git/frederic/linux-dynticks.git
timers/nohz-v2
Thanks.
---
Frederic Weisbecker (4):
nohz: Unhide full dynticks feature from its dependencies
nohz: Rename CONFIG_NO_HZ to CONFIG_NO_HZ_COMMON
nohz: Pack nohz Kconfig option in a menu of choices
nohz: Print final full dynticks
:
CONFIG_VIRT_CPU_ACCOUNTING_GEN is part of a Kconfig choice
menu and it appears we can't select it from another Kconfig
selection when it's under such layout. So for now this
particular item stays as a passive dependency.
Reported-by: Ingo Molnar
Signed-off-by: Frederic Weisbecker
Cc: Andrew Morton
Cc: Chris M
d-off-by: Frederic Weisbecker
Cc: Andrew Morton
Cc: Chris Metcalf
Cc: Christoph Lameter
Cc: Geoff Levand
Cc: Gilad Ben Yossef
Cc: Hakan Akkan
Cc: Ingo Molnar
Cc: Kevin Hilman
Cc: Li Zhong
Cc: Namhyung Kim
Cc: Paul E. McKenney
Cc: Paul Gortmaker
Cc: Peter Zijlstra
Cc: Steven Rosted
2013/4/3 Frank Ch. Eigler :
> Frederic Weisbecker writes:
>
>> [...]
>> Sometimes I don't mind keeping around code in the kernel for out of
>> tree users, depending on the case. But in this specific matter we have
>> more standard ways to do this kind of hook
2013/4/4 Stanislaw Gruszka :
> This patch series removes cputime scaling from kernel. It can be easily
> done in user space using floating point if we provide sum_exec_runtime,
> what patches 2/4 and 3/4 do. I have procps patch which utilize that:
>
> http://people.redhat.com/sgruszka/procps-use-su
2013/4/3 Frank Ch. Eigler :
> Hi -
>
> On Wed, Apr 03, 2013 at 02:49:53PM +0200, Frederic Weisbecker wrote:
>
>> Sounds good, would you like to propose a version? We are also
>> interested in a timer tick event tracepoint for dynticks debugging.
>
> How about this
2013/4/4 Stanislaw Gruszka :
> On Thu, Apr 04, 2013 at 02:31:42PM +0200, Frederic Weisbecker wrote:
>> I don't know. I'm not convinced userland is the right place to perform
>> this kind of check. The kernel perhaps doesn't give guarantee about
>> utime/stime
2013/4/4 Arnd Bergmann :
> Patch 6d0e025 "posix_cpu_timer: consolidate expiry time type" in linux-next
> introduces a new compiler warning:
>
> kernel/posix-cpu-timers.c: In function 'posix_cpu_timer_schedule':
> kernel/posix-cpu-timers.c:1127:19: warning: 'now' may be used uninitialized
> in this
ted with old sched_domain once it has been initialized.
> But this solution introduces a additionnal latency in the rebuild sequence
> that is called during cpu hotplug.
>
> As suggested by Frederic Weisbecker, another solution is to have the same
> rcu lifecycle for both NOHZ_IDLE and sche
2013/4/4 Frederic Weisbecker :
> 2013/4/3 Vincent Guittot :
>> On my smp platform which is made of 5 cores in 2 clusters, I have the
>> nr_busy_cpu field of sched_group_power struct that is not null when the
>> platform is fully idle. The root cause is:
>> During the boo
ted with old sched_domain once it has been initialized.
> But this solution introduces a additionnal latency in the rebuild sequence
> that is called during cpu hotplug.
>
> As suggested by Frederic Weisbecker, another solution is to have the same
> rcu lifecycle for both NOHZ_IDLE and sche
2013/4/4 Dave Hansen :
> With the 3.9-rcs (and probably much earlier) I'm seeing some weird top
> output where the cpu time "spent" is millions of hours:
>
> 445 root 20 0 000 S0 0.0 5124095h kworker/45:1
> 404 root 20 0 000 S0 0.0 5124095h kworker/
2013/1/29 Steven Rostedt :
> On Mon, 2013-01-28 at 20:03 +0100, Frederic Weisbecker wrote:
>> Ingo,
>>
>> Please pull the new full dynticks cputime accounting code that
>> can be found at:
>>
>> git://git.kernel.org/pub/scm/linux/kernel/git/frederic/linux-d
2013/1/28 Christoph Lameter :
> On Mon, 28 Jan 2013, Frederic Weisbecker wrote:
>
>> 2013/1/28 Christoph Lameter :
>> > On Mon, 28 Jan 2013, Frederic Weisbecker wrote:
>> >
>> >> My last concern is the dependency on CONFIG_64BIT. We rely on cputim
2013/1/29 Steven Rostedt :
> On Mon, 2013-01-28 at 19:02 -0500, Steven Rostedt wrote:
>
>> I've ran these through some basic tests and I don't see any issues. But
>> why did you drop the 'comment' patch that you had in v6?
>>
>> Acked-by: Steven Rostedt
>>
>
> I'm sorry but I need to un-Ack. Due t
2013/1/29 Steven Rostedt :
> On Tue, 2013-01-29 at 03:39 +0100, Frederic Weisbecker wrote:
>
>> Please send me your kernelspin so I can test as well.
>
> http://rostedt.homelinux.com/private/kernelspin.c
I can't reproduce your issue. I'm getting almost the same result
2013/1/29 Steven Rostedt :
> On Tue, 2013-01-29 at 16:12 +0100, Frederic Weisbecker wrote:
>
>> I have set some cputime tracepoint and looked at the stacktrace. They
>> are sorted by weight, which is the amount of cputime accounted. I
>> can't find anything w
2013/1/29 Steven Rostedt :
> On Mon, 2013-01-28 at 21:36 -0500, Steven Rostedt wrote:
>
>> [root@bxtest ~]# time /work/c/kernelspin 10
>>
>> real 0m10.001s
>> user 0m1.114s
>> sys 0m8.886s
>>
>>
>> But after enabling the force context_tracking I get this:
>>
>> [root@bxtest ~]# time /work/c/ke
2013/1/28 Sedat Dilek :
> On Mon, Jan 28, 2013 at 8:04 PM, Frederic Weisbecker
> wrote:
>> While remotely reading the cputime of a task running in a
>> full dynticks CPU, the values stored in utime/stime fields
>> of struct task_struct may be stale. Its values may be tho
2013/1/31 Sedat Dilek :
> On Thu, Jan 31, 2013 at 1:38 AM, Frederic Weisbecker
> wrote:
>> 2013/1/28 Sedat Dilek :
>>> On Mon, Jan 28, 2013 at 8:04 PM, Frederic Weisbecker
>>> wrote:
>>>> While remotely reading the cputime of a task running in a
&g
2013/1/31 Sedat Dilek :
> I still dunno all parts of dynticks and their correlation.
> AFAICS they seem to be independent, can't say if it is "complete" with
> dynticks-cputime-only.
Dynticks cputime and dynticks printk are both standalone and
independant. In mainline, both printk and cputime acco
2013/1/29 Vincent Guittot :
> On my smp platform which is made of 5 cores in 2 clusters,I have the
> nr_busy_cpu field of sched_group_power struct that is not null when the
> platform is fully idle. The root cause seems to be:
> During the boot sequence, some CPUs reach the idle loop and set their
2013/2/2 anish kumar :
> On Sat, 2013-01-26 at 17:24 +0100, Frederic Weisbecker wrote:
>> 2012/11/3 anish kumar :
>> > From: anish kumar
>> >
>> > As no one is using the return value of irq_work_queue function
>> > it is better to just make it void.
&g
2012/12/3 Vincent Guittot :
> With the coupled cpuidle driver (but probably also with other drivers),
> a CPU loops in a temporary safe state while waiting for other CPUs of its
> cluster to be ready to enter the coupled C-state. If an IRQ or a softirq
> occurs, the CPU will stay in this internal l
2013/1/24 Xiaotian Feng :
> On Thu, Jan 10, 2013 at 2:33 AM, Frederic Weisbecker
> wrote:
>> --- a/kernel/sched/cputime.c
>> +++ b/kernel/sched/cputime.c
>> @@ -509,11 +509,11 @@ EXPORT_SYMBOL_GPL(vtime_account);
>> # define nsecs_to_cputime(__nsecs) nsecs_
2013/1/25 Vincent Guittot :
> This sequence is not the right one
>
>> I'm going to look for the saved trace to check the sequence above
>
> I haven't been able to reproduce the bug that this patch was supposed to
> solved. The patch 2 and 3 seem enough to fix the nr_busy_cpus field. I will
> contin
2013/1/25 Vincent Guittot :
>
> Le 25 janv. 2013 13:00, "Frederic Weisbecker" a écrit :
>
>
>>
>> 2013/1/25 Vincent Guittot :
>> > This sequence is not the right one
>> >
>> >> I'm going to look for the saved trace to check
2013/1/26 Paul E. McKenney :
> Right... Replied to the wrong email. :-/
>
> I queued "context_tracking: Add comments on interface and internals".
Sounds perfect, thanks :)
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel
2012/11/16 liguang :
> Signed-off-by: liguang
> ---
> kernel/sched/cputime.c |2 +-
> 1 files changed, 1 insertions(+), 1 deletions(-)
>
> diff --git a/kernel/sched/cputime.c b/kernel/sched/cputime.c
> index 81b763b..d2c24c1 100644
> --- a/kernel/sched/cputime.c
> +++ b/kernel/sched/cputime.c
2013/1/26 Joe Perches :
> On Sat, 2013-01-26 at 01:45 +0100, Frederic Weisbecker wrote:
>> > diff --git a/kernel/sched/cputime.c b/kernel/sched/cputime.c
> []
>> > @@ -282,7 +282,7 @@ static __always_inline bool
>> > steal_account_process_tick(void)
&g
ld considerably reduce the chances of an overflow
on most workloads.
This is largely inspired by a patch from Stanislaw Gruszka:
http://lkml.kernel.org/r/20130107113144.ga7...@redhat.com
Inspired-by: Stanislaw Gruszka
Reported-by: Stanislaw Gruszka
Acked-by: Stanislaw Gruszka
Signed-off-by: Frederic
2012/11/3 anish kumar :
> From: anish kumar
>
> As no one is using the return value of irq_work_queue function
> it is better to just make it void.
>
> Acked-by: Steven Rostedt
> Signed-off-by: anish kumar
> ---
> kernel/irq_work.c |5 ++---
> 1 files changed, 2 insertions(+), 3 deletions(-
he correct result to the user.
Signed-off-by: Frederic Weisbecker
---
Changes since v6:
* Fix build error with kvm modules (Thanks Sedat Dilek and
Wu Fenguang)
* Fix mistyped header inclusion in ia64
* Fix more missing symbols for kvm
---
Frederic Weisbecker (8):
context_tracking: Export con
Export the context state: whether we run in user / kernel
from the context tracking subsystem point of view.
This is going to be used by the generic virtual cputime
accounting subsystem that is needed to implement the full
dynticks.
Signed-off-by: Frederic Weisbecker
Cc: Andrew Morton
Cc: Ingo
ch between
tick and vtime based accounting anytime in order to minimize the
overhead associated to user hooks.
Signed-off-by: Frederic Weisbecker
Cc: Andrew Morton
Cc: Ingo Molnar
Cc: Li Zhong
Cc: Namhyung Kim
Cc: Paul E. McKenney
Cc: Paul Gortmaker
Cc: Peter Zijlstra
Cc: Steven Rostedt
t.
To fix this, flush the cputime of the dynticks CPUs on
kernel <-> user transition and record the time / context
where we did this. Then on top of this snapshot and the current
time, perform the fixup on the reader side from task_times()
accessors.
Signed-off-by: Frederic Weisbecker
Cc: A
1 - 100 of 4755 matches
Mail list logo
