Hi Oleg,
On Mon, Sep 07, 2015 at 07:06:52PM +0200, Oleg Nesterov wrote:
> Sorry for delay,
>
> On 09/02, Boqun Feng wrote:
> >
> > On Tue, Sep 01, 2015 at 06:39:23PM +0200, Oleg Nesterov wrote:
> > > On 09/01, Boqun Feng wrote:
> > > >
> > > > On Tue, Sep 01, 2015 at 11:59:23AM +0200, Oleg
Sorry for delay,
On 09/02, Boqun Feng wrote:
>
> On Tue, Sep 01, 2015 at 06:39:23PM +0200, Oleg Nesterov wrote:
> > On 09/01, Boqun Feng wrote:
> > >
> > > On Tue, Sep 01, 2015 at 11:59:23AM +0200, Oleg Nesterov wrote:
> > > >
> > > > And just in case, wake_up() differs in a sense that it doesn't
Sorry for delay,
On 09/02, Boqun Feng wrote:
>
> On Tue, Sep 01, 2015 at 06:39:23PM +0200, Oleg Nesterov wrote:
> > On 09/01, Boqun Feng wrote:
> > >
> > > On Tue, Sep 01, 2015 at 11:59:23AM +0200, Oleg Nesterov wrote:
> > > >
> > > > And just in case, wake_up() differs in a sense that it doesn't
Hi Oleg,
On Mon, Sep 07, 2015 at 07:06:52PM +0200, Oleg Nesterov wrote:
> Sorry for delay,
>
> On 09/02, Boqun Feng wrote:
> >
> > On Tue, Sep 01, 2015 at 06:39:23PM +0200, Oleg Nesterov wrote:
> > > On 09/01, Boqun Feng wrote:
> > > >
> > > > On Tue, Sep 01, 2015 at 11:59:23AM +0200, Oleg
On Tue, Sep 01, 2015 at 06:39:23PM +0200, Oleg Nesterov wrote:
> On 09/01, Boqun Feng wrote:
> >
> > On Tue, Sep 01, 2015 at 11:59:23AM +0200, Oleg Nesterov wrote:
> > >
> > > And just in case, wake_up() differs in a sense that it doesn't even need
> > > that STORE-LOAD barrier in
On 09/01, Boqun Feng wrote:
>
> On Tue, Sep 01, 2015 at 11:59:23AM +0200, Oleg Nesterov wrote:
> >
> > And just in case, wake_up() differs in a sense that it doesn't even need
> > that STORE-LOAD barrier in try_to_wake_up(), we can rely on
> > wait_queue_head_t->lock. Assuming that wake_up() pairs
On Tue, Sep 01, 2015 at 11:59:23AM +0200, Oleg Nesterov wrote:
> On 09/01, Boqun Feng wrote:
> >
> > But I'm still a little confused at Oleg's words:
> >
> > "What is really important is that we have a barrier before we _read_ the
> > task state."
> >
> > I read is as "What is really important is
On 09/01, Boqun Feng wrote:
>
> But I'm still a little confused at Oleg's words:
>
> "What is really important is that we have a barrier before we _read_ the
> task state."
>
> I read is as "What is really important is that we have a barrier before
> we _read_ the task state and _after_ we write
On 08/31, Paul E. McKenney wrote:
>
> On Mon, Aug 31, 2015 at 08:33:35PM +0200, Oleg Nesterov wrote:
> > On 08/31, Boqun Feng wrote:
> > >
> > > Fair enough, I went too far. How about just a single paragraph saying
> > > that:
> > >
> > > The wake_up(), wait_event() and their friends have proper
On Tue, Sep 01, 2015 at 06:39:23PM +0200, Oleg Nesterov wrote:
> On 09/01, Boqun Feng wrote:
> >
> > On Tue, Sep 01, 2015 at 11:59:23AM +0200, Oleg Nesterov wrote:
> > >
> > > And just in case, wake_up() differs in a sense that it doesn't even need
> > > that STORE-LOAD barrier in
On 09/01, Boqun Feng wrote:
>
> But I'm still a little confused at Oleg's words:
>
> "What is really important is that we have a barrier before we _read_ the
> task state."
>
> I read is as "What is really important is that we have a barrier before
> we _read_ the task state and _after_ we write
On 08/31, Paul E. McKenney wrote:
>
> On Mon, Aug 31, 2015 at 08:33:35PM +0200, Oleg Nesterov wrote:
> > On 08/31, Boqun Feng wrote:
> > >
> > > Fair enough, I went too far. How about just a single paragraph saying
> > > that:
> > >
> > > The wake_up(), wait_event() and their friends have proper
On 09/01, Boqun Feng wrote:
>
> On Tue, Sep 01, 2015 at 11:59:23AM +0200, Oleg Nesterov wrote:
> >
> > And just in case, wake_up() differs in a sense that it doesn't even need
> > that STORE-LOAD barrier in try_to_wake_up(), we can rely on
> > wait_queue_head_t->lock. Assuming that wake_up() pairs
On Tue, Sep 01, 2015 at 11:59:23AM +0200, Oleg Nesterov wrote:
> On 09/01, Boqun Feng wrote:
> >
> > But I'm still a little confused at Oleg's words:
> >
> > "What is really important is that we have a barrier before we _read_ the
> > task state."
> >
> > I read is as "What is really important is
On Tue, Sep 01, 2015 at 11:40:14AM +0800, Boqun Feng wrote:
> Hi Paul,
>
> On Mon, Aug 31, 2015 at 01:37:39PM -0700, Paul E. McKenney wrote:
> > On Mon, Aug 31, 2015 at 08:33:35PM +0200, Oleg Nesterov wrote:
> > > On 08/31, Boqun Feng wrote:
> > > >
> > > > Fair enough, I went too far. How about
Hi Paul,
On Mon, Aug 31, 2015 at 01:37:39PM -0700, Paul E. McKenney wrote:
> On Mon, Aug 31, 2015 at 08:33:35PM +0200, Oleg Nesterov wrote:
> > On 08/31, Boqun Feng wrote:
> > >
> > > Fair enough, I went too far. How about just a single paragraph saying
> > > that:
> > >
> > > The wake_up(),
On Mon, Aug 31, 2015 at 08:33:35PM +0200, Oleg Nesterov wrote:
> On 08/31, Boqun Feng wrote:
> >
> > Fair enough, I went too far. How about just a single paragraph saying
> > that:
> >
> > The wake_up(), wait_event() and their friends have proper barriers in
> > them, but these implicity barriers
On 08/31, Boqun Feng wrote:
>
> Fair enough, I went too far. How about just a single paragraph saying
> that:
>
> The wake_up(), wait_event() and their friends have proper barriers in
> them, but these implicity barriers are only for the correctness for
> sleep and wakeup. So don't rely on these
On 08/31, Boqun Feng wrote:
>
> Fair enough, I went too far. How about just a single paragraph saying
> that:
>
> The wake_up(), wait_event() and their friends have proper barriers in
> them, but these implicity barriers are only for the correctness for
> sleep and wakeup. So don't rely on these
On Mon, Aug 31, 2015 at 08:33:35PM +0200, Oleg Nesterov wrote:
> On 08/31, Boqun Feng wrote:
> >
> > Fair enough, I went too far. How about just a single paragraph saying
> > that:
> >
> > The wake_up(), wait_event() and their friends have proper barriers in
> > them, but these implicity barriers
On Tue, Sep 01, 2015 at 11:40:14AM +0800, Boqun Feng wrote:
> Hi Paul,
>
> On Mon, Aug 31, 2015 at 01:37:39PM -0700, Paul E. McKenney wrote:
> > On Mon, Aug 31, 2015 at 08:33:35PM +0200, Oleg Nesterov wrote:
> > > On 08/31, Boqun Feng wrote:
> > > >
> > > > Fair enough, I went too far. How about
Hi Paul,
On Mon, Aug 31, 2015 at 01:37:39PM -0700, Paul E. McKenney wrote:
> On Mon, Aug 31, 2015 at 08:33:35PM +0200, Oleg Nesterov wrote:
> > On 08/31, Boqun Feng wrote:
> > >
> > > Fair enough, I went too far. How about just a single paragraph saying
> > > that:
> > >
> > > The wake_up(),
Hi Oleg,
On Sat, Aug 29, 2015 at 04:27:07PM +0200, Oleg Nesterov wrote:
> Hello Boqun,
>
...
> > By this, I think you actually means the example below the added text,
> > i.e. the example for "to repeat..", right?
>
> And above. Even
>
> The barrier occurs before the task state is
Hi Oleg,
On Sat, Aug 29, 2015 at 04:27:07PM +0200, Oleg Nesterov wrote:
Hello Boqun,
...
By this, I think you actually means the example below the added text,
i.e. the example for to repeat.., right?
And above. Even
The barrier occurs before the task state is cleared
is not
Hello Boqun,
On 08/29, Boqun Feng wrote:
> Hi Oleg
>
> On Fri, Aug 28, 2015 at 06:06:37PM +0200, Oleg Nesterov wrote:
> > On 08/28, Michal Hocko wrote:
> > >
> > > On Thu 27-08-15 20:26:54, Oleg Nesterov wrote:
> > > > On 08/27, Michal Hocko wrote:
> > > > >
> > > > > ---
Hi Oleg
On Fri, Aug 28, 2015 at 06:06:37PM +0200, Oleg Nesterov wrote:
> On 08/28, Michal Hocko wrote:
> >
> > On Thu 27-08-15 20:26:54, Oleg Nesterov wrote:
> > > On 08/27, Michal Hocko wrote:
> > > >
> > > > --- a/Documentation/memory-barriers.txt
> > > > +++ b/Documentation/memory-barriers.txt
Hi Oleg
On Fri, Aug 28, 2015 at 06:06:37PM +0200, Oleg Nesterov wrote:
On 08/28, Michal Hocko wrote:
On Thu 27-08-15 20:26:54, Oleg Nesterov wrote:
On 08/27, Michal Hocko wrote:
--- a/Documentation/memory-barriers.txt
+++ b/Documentation/memory-barriers.txt
@@ -2031,6
Hello Boqun,
On 08/29, Boqun Feng wrote:
Hi Oleg
On Fri, Aug 28, 2015 at 06:06:37PM +0200, Oleg Nesterov wrote:
On 08/28, Michal Hocko wrote:
On Thu 27-08-15 20:26:54, Oleg Nesterov wrote:
On 08/27, Michal Hocko wrote:
--- a/Documentation/memory-barriers.txt
+++
On 08/28, Michal Hocko wrote:
>
> On Thu 27-08-15 20:26:54, Oleg Nesterov wrote:
> > On 08/27, Michal Hocko wrote:
> > >
> > > --- a/Documentation/memory-barriers.txt
> > > +++ b/Documentation/memory-barriers.txt
> > > @@ -2031,6 +2031,9 @@ something up. The barrier occurs before the task
> > >
On Thu 27-08-15 20:26:54, Oleg Nesterov wrote:
> On 08/27, Michal Hocko wrote:
> >
> > --- a/Documentation/memory-barriers.txt
> > +++ b/Documentation/memory-barriers.txt
> > @@ -2031,6 +2031,9 @@ something up. The barrier occurs before the task
> > state is cleared, and so sits
> >
On Thu 27-08-15 20:26:54, Oleg Nesterov wrote:
On 08/27, Michal Hocko wrote:
--- a/Documentation/memory-barriers.txt
+++ b/Documentation/memory-barriers.txt
@@ -2031,6 +2031,9 @@ something up. The barrier occurs before the task
state is cleared, and so sits
general barrier
On 08/28, Michal Hocko wrote:
On Thu 27-08-15 20:26:54, Oleg Nesterov wrote:
On 08/27, Michal Hocko wrote:
--- a/Documentation/memory-barriers.txt
+++ b/Documentation/memory-barriers.txt
@@ -2031,6 +2031,9 @@ something up. The barrier occurs before the task
state is cleared,
On 08/27, Michal Hocko wrote:
>
> --- a/Documentation/memory-barriers.txt
> +++ b/Documentation/memory-barriers.txt
> @@ -2031,6 +2031,9 @@ something up. The barrier occurs before the task state
> is cleared, and so sits
>STORE current->state
> LOAD event_indicated
On Thu 27-08-15 14:43:34, Peter Zijlstra wrote:
> On Thu, Aug 27, 2015 at 02:27:27PM +0200, Michal Hocko wrote:
> > Hi,
> > I have just stumbled over the comment above wake_up_process which
> > claims:
> > "
> > * It may be assumed that this function implies a write memory barrier
> > before
> >
On Thu, Aug 27, 2015 at 02:27:27PM +0200, Michal Hocko wrote:
> Hi,
> I have just stumbled over the comment above wake_up_process which
> claims:
> "
> * It may be assumed that this function implies a write memory barrier before
> * changing the task state if and only if any tasks are woken up.
Hi,
I have just stumbled over the comment above wake_up_process which
claims:
"
* It may be assumed that this function implies a write memory barrier before
* changing the task state if and only if any tasks are woken up.
"
but try_to_wake_up does smp_mb__before_spinlock and did smp_wmb
since
Hi,
I have just stumbled over the comment above wake_up_process which
claims:
* It may be assumed that this function implies a write memory barrier before
* changing the task state if and only if any tasks are woken up.
but try_to_wake_up does smp_mb__before_spinlock and did smp_wmb
since
On Thu 27-08-15 14:43:34, Peter Zijlstra wrote:
On Thu, Aug 27, 2015 at 02:27:27PM +0200, Michal Hocko wrote:
Hi,
I have just stumbled over the comment above wake_up_process which
claims:
* It may be assumed that this function implies a write memory barrier
before
* changing the
On Thu, Aug 27, 2015 at 02:27:27PM +0200, Michal Hocko wrote:
Hi,
I have just stumbled over the comment above wake_up_process which
claims:
* It may be assumed that this function implies a write memory barrier before
* changing the task state if and only if any tasks are woken up.
On 08/27, Michal Hocko wrote:
--- a/Documentation/memory-barriers.txt
+++ b/Documentation/memory-barriers.txt
@@ -2031,6 +2031,9 @@ something up. The barrier occurs before the task state
is cleared, and so sits
general barrier STORE current-state
LOAD
40 matches
Mail list logo
