On Thu, Nov 09, 2017 at 09:24:08AM -0800, Davidlohr Bueso wrote:
> On Wed, 08 Nov 2017, Boqun Feng wrote:
> > Or worse:
> >
> > * CPU0 CPU1
> > * dlock_list_add() dlock_lists_empty()
> > *smp_mb__bef
On Wed, 08 Nov 2017, Boqun Feng wrote:
Or worse:
* CPU0 CPU1
* dlock_list_add() dlock_lists_empty()
*smp_mb__before_atomic();
*[L] atomic_re
On Tue, Nov 07, 2017 at 10:59:29AM -0700, Andreas Dilger wrote:
> On Nov 7, 2017, at 4:59 AM, Jan Kara wrote:
> > On Mon 06-11-17 10:47:08, Davidlohr Bueso wrote:
> >> + /*
> >> + * Serialize dlist->used_lists such that a 0->1 transition is not
> >> + * missed by another thread checking if an
On Tue, 2017-11-07 at 13:57 -0500, Waiman Long wrote:
> On 11/07/2017 12:59 PM, Andreas Dilger wrote:
> >
> > On Nov 7, 2017, at 4:59 AM, Jan Kara wrote:
> > >
> > > On Mon 06-11-17 10:47:08, Davidlohr Bueso wrote:
> > > >
> > > > + /*
> > > > + * Serialize dlist->used_lists such t
On 11/07/2017 12:59 PM, Andreas Dilger wrote:
> On Nov 7, 2017, at 4:59 AM, Jan Kara wrote:
>> On Mon 06-11-17 10:47:08, Davidlohr Bueso wrote:
>>> + /*
>>> +* Serialize dlist->used_lists such that a 0->1 transition is not
>>> +* missed by another thread checking if any of the dlock list
On Nov 7, 2017, at 4:59 AM, Jan Kara wrote:
> On Mon 06-11-17 10:47:08, Davidlohr Bueso wrote:
>> +/*
>> + * Serialize dlist->used_lists such that a 0->1 transition is not
>> + * missed by another thread checking if any of the dlock lists are
>> + * used.
>> + *
>> + * CPU0
On Mon 06-11-17 10:47:08, Davidlohr Bueso wrote:
> Instead of the current O(N) implementation, at the cost
> of adding an atomic counter, we can convert the call to
> an atomic_read(). The counter only serves for accounting
> empty to non-empty transitions, and vice versa; therefore
> only modified
On 11/06/2017 01:47 PM, Davidlohr Bueso wrote:
> Instead of the current O(N) implementation, at the cost
> of adding an atomic counter, we can convert the call to
> an atomic_read(). The counter only serves for accounting
> empty to non-empty transitions, and vice versa; therefore
> only modified t
Instead of the current O(N) implementation, at the cost
of adding an atomic counter, we can convert the call to
an atomic_read(). The counter only serves for accounting
empty to non-empty transitions, and vice versa; therefore
only modified twice for each of the lists during the
lifetime of the dl
9 matches
Mail list logo
