Le 25/08/2017 à 09:44, Benjamin Herrenschmidt a écrit :
On Fri, 2017-08-25 at 06:53 +0200, Frederic Barrat wrote:
Le 24/08/2017 à 20:47, Benjamin Herrenschmidt a écrit :
On Thu, 2017-08-24 at 18:40 +0200, Frederic Barrat wrote:
The decrementing part is giving me troubles, and I think it ma
On Fri, 2017-08-25 at 06:53 +0200, Frederic Barrat wrote:
>
> Le 24/08/2017 à 20:47, Benjamin Herrenschmidt a écrit :
> > On Thu, 2017-08-24 at 18:40 +0200, Frederic Barrat wrote:
> > >
> > > The decrementing part is giving me troubles, and I think it makes sense:
> > > if I decrement the counter
Le 24/08/2017 à 20:47, Benjamin Herrenschmidt a écrit :
On Thu, 2017-08-24 at 18:40 +0200, Frederic Barrat wrote:
The decrementing part is giving me troubles, and I think it makes sense:
if I decrement the counter when detaching the context from the capi
card, then the next TLBIs for the memo
On Thu, 2017-08-24 at 18:40 +0200, Frederic Barrat wrote:
>
> The decrementing part is giving me troubles, and I think it makes sense:
> if I decrement the counter when detaching the context from the capi
> card, then the next TLBIs for the memory context may be back to local.
Yes, you need to
Le 21/08/2017 à 19:35, Benjamin Herrenschmidt a écrit :
On Mon, 2017-08-21 at 19:27 +0200, Frederic Barrat wrote:
Hi Ben,
Le 24/07/2017 à 06:28, Benjamin Herrenschmidt a écrit :
Instead of comparing the whole CPU mask every time, let's
keep a counter of how many bits are set in the mask. Thu
it easier.
> Arguably what happens on those accelerators is pretty close to an active
> cpu.
>
> Once it is merged, I'm going to have to backport your patch (and an
> update to mine) to the p9-supporting distros. From a quick look, your
> patch, i.e."[PATCH 5/6] pow
nce it is merged, I'm going to have to backport your patch (and an
update to mine) to the p9-supporting distros. From a quick look, your
patch, i.e."[PATCH 5/6] powerpc/mm: Optimize detection of thread local
mm's" is completely independent from the rest of the series, right?
Fred
Frederic Barrat writes:
> Hi Ben,
>
> Le 24/07/2017 à 06:28, Benjamin Herrenschmidt a écrit :
>> Instead of comparing the whole CPU mask every time, let's
>> keep a counter of how many bits are set in the mask. Thus
>> testing for a local mm only requires testing if that counter
>> is 1 and the c
On Mon, 2017-08-21 at 19:27 +0200, Frederic Barrat wrote:
> Hi Ben,
>
> Le 24/07/2017 à 06:28, Benjamin Herrenschmidt a écrit :
> > Instead of comparing the whole CPU mask every time, let's
> > keep a counter of how many bits are set in the mask. Thus
> > testing for a local mm only requires testi
Hi Ben,
Le 24/07/2017 à 06:28, Benjamin Herrenschmidt a écrit :
Instead of comparing the whole CPU mask every time, let's
keep a counter of how many bits are set in the mask. Thus
testing for a local mm only requires testing if that counter
is 1 and the current CPU bit is set in the mask.
I'm
On Fri, 2017-08-04 at 14:06 +0200, Frederic Barrat wrote:
> > +#ifdef CONFIG_PPC_BOOK3S_64
> > +static inline int mm_is_thread_local(struct mm_struct *mm)
> > +{
> > + if (atomic_read(&mm->context.active_cpus) > 1)
> > + return false;
> > + return cpumask_test_cpu(smp_processor_
Le 24/07/2017 à 06:28, Benjamin Herrenschmidt a écrit :
Instead of comparing the whole CPU mask every time, let's
keep a counter of how many bits are set in the mask. Thus
testing for a local mm only requires testing if that counter
is 1 and the current CPU bit is set in the mask.
Signed-off-b
Nicholas Piggin writes:
> On Mon, 24 Jul 2017 23:46:44 +1000
> Michael Ellerman wrote:
>
>> Nicholas Piggin writes:
>>
>> > On Mon, 24 Jul 2017 14:28:02 +1000
>> > Benjamin Herrenschmidt wrote:
>> >
>> >> Instead of comparing the whole CPU mask every time, let's
>> >> keep a counter of how
On Tue, 25 Jul 2017 11:03:45 +1000
Benjamin Herrenschmidt wrote:
> On Tue, 2017-07-25 at 10:44 +1000, Nicholas Piggin wrote:
> > The two variants are just cleaner versions of the two variants you
> > already introduced.
> >
> > static inline bool mm_activate_cpu(struct mm_struct *mm)
> > {
> >
On Tue, 2017-07-25 at 10:44 +1000, Nicholas Piggin wrote:
> The two variants are just cleaner versions of the two variants you
> already introduced.
>
> static inline bool mm_activate_cpu(struct mm_struct *mm)
> {
> if (!cpumask_test_cpu(smp_processor_id(), mm_cpumask(next))) {
> cpuma
On Tue, 25 Jul 2017 06:58:46 +1000
Benjamin Herrenschmidt wrote:
> On Mon, 2017-07-24 at 21:25 +1000, Nicholas Piggin wrote:
> > > +#ifdef CONFIG_PPC_BOOK3S_64
> > > +static inline void inc_mm_active_cpus(struct mm_struct *mm)
> > > +{
> > > + atomic_inc(&mm->context.active_cpus);
> > > +}
>
On Mon, 24 Jul 2017 23:46:44 +1000
Michael Ellerman wrote:
> Nicholas Piggin writes:
>
> > On Mon, 24 Jul 2017 14:28:02 +1000
> > Benjamin Herrenschmidt wrote:
> >
> >> Instead of comparing the whole CPU mask every time, let's
> >> keep a counter of how many bits are set in the mask. Thus
>
On Mon, 2017-07-24 at 21:25 +1000, Nicholas Piggin wrote:
> > +#ifdef CONFIG_PPC_BOOK3S_64
> > +static inline void inc_mm_active_cpus(struct mm_struct *mm)
> > +{
> > + atomic_inc(&mm->context.active_cpus);
> > +}
> > +#else
> > +static inline void inc_mm_active_cpus(struct mm_struct *mm) { }
>
Nicholas Piggin writes:
> On Mon, 24 Jul 2017 14:28:02 +1000
> Benjamin Herrenschmidt wrote:
>
>> Instead of comparing the whole CPU mask every time, let's
>> keep a counter of how many bits are set in the mask. Thus
>> testing for a local mm only requires testing if that counter
>> is 1 and the
On Mon, 24 Jul 2017 14:28:02 +1000
Benjamin Herrenschmidt wrote:
> Instead of comparing the whole CPU mask every time, let's
> keep a counter of how many bits are set in the mask. Thus
> testing for a local mm only requires testing if that counter
> is 1 and the current CPU bit is set in the mask
Instead of comparing the whole CPU mask every time, let's
keep a counter of how many bits are set in the mask. Thus
testing for a local mm only requires testing if that counter
is 1 and the current CPU bit is set in the mask.
Signed-off-by: Benjamin Herrenschmidt
---
arch/powerpc/include/asm/boo
21 matches
Mail list logo
