On Thu, May 07, 2015 at 02:14:39PM +0200, David Hildenbrand wrote:
> Thanks :), well just to make sure I got your opinion on this correctly:
>
> 1. You think that 2 counters is the way to go for now
ack
> 2. You agree that we can't replace preempt_disable()+pagefault_disable() with
>
* David Hildenbrand wrote:
> @Ingo, do you have a strong feeling against this whole
> patchset/idea?
No objections, sounds good to me now.
Thanks,
Ingo
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
> On Thu, May 07, 2015 at 01:40:30PM +0200, David Hildenbrand wrote:
> > But anyhow, opinions seem to differ how to best handle that whole stuff.
> >
> > I think a separate counter just makes sense, as we are dealing with two
> > different concepts and we don't want to lose the preempt_disable =^
On Thu, May 07, 2015 at 01:40:30PM +0200, David Hildenbrand wrote:
> But anyhow, opinions seem to differ how to best handle that whole stuff.
>
> I think a separate counter just makes sense, as we are dealing with two
> different concepts and we don't want to lose the preempt_disable =^ NOP
> for
On Thu, May 07, 2015 at 01:40:30PM +0200, David Hildenbrand wrote:
> I think a separate counter just makes sense, as we are dealing with two
> different concepts and we don't want to lose the preempt_disable =^ NOP
> for !CONFIG_PREEMPT.
Right, let me try and get my head on straight -- I'm so
On Thu, May 07, 2015 at 01:23:35PM +0200, David Hildenbrand wrote:
> > On Thu, May 07, 2015 at 12:50:53PM +0200, David Hildenbrand wrote:
> > > Just to make sure we have a common understanding (as written in my cover
> > > letter):
> > >
> > > Your suggestion won't work with !CONFIG_PREEMPT
> On Thu, May 07, 2015 at 12:50:53PM +0200, David Hildenbrand wrote:
> > Just to make sure we have a common understanding (as written in my cover
> > letter):
> >
> > Your suggestion won't work with !CONFIG_PREEMPT (!CONFIG_PREEMPT_COUNT). If
> > there is no preempt counter, in_atomic() won't
>
> * David Hildenbrand wrote:
>
> > > On Thu, May 07, 2015 at 12:50:53PM +0200, David Hildenbrand wrote:
> > > > Just to make sure we have a common understanding (as written in my cover
> > > > letter):
> > > >
> > > > Your suggestion won't work with !CONFIG_PREEMPT
> > > >
* David Hildenbrand wrote:
> > On Thu, May 07, 2015 at 12:50:53PM +0200, David Hildenbrand wrote:
> > > Just to make sure we have a common understanding (as written in my cover
> > > letter):
> > >
> > > Your suggestion won't work with !CONFIG_PREEMPT (!CONFIG_PREEMPT_COUNT).
> > > If
> > >
> On Thu, May 07, 2015 at 12:50:53PM +0200, David Hildenbrand wrote:
> > Just to make sure we have a common understanding (as written in my cover
> > letter):
> >
> > Your suggestion won't work with !CONFIG_PREEMPT (!CONFIG_PREEMPT_COUNT). If
> > there is no preempt counter, in_atomic() won't
On Thu, May 07, 2015 at 12:50:53PM +0200, David Hildenbrand wrote:
> Just to make sure we have a common understanding (as written in my cover
> letter):
>
> Your suggestion won't work with !CONFIG_PREEMPT (!CONFIG_PREEMPT_COUNT). If
> there is no preempt counter, in_atomic() won't work.
But
* David Hildenbrand wrote:
> > AFAICR we did this to avoid having to do both:
> >
> > preempt_disable();
> > pagefault_disable();
> >
> > in a fair number of places -- just like this patch-set does, this is
> > touching two cachelines where one would have been enough.
> >
> > Also,
> On Wed, May 06, 2015 at 07:50:25PM +0200, David Hildenbrand wrote:
> > +/*
> > + * Is the pagefault handler disabled? If so, user access methods will not
> > sleep.
> > + */
> > +#define pagefault_disabled() (current->pagefault_disabled != 0)
>
> So -RT has:
>
> static inline bool
On Wed, May 06, 2015 at 07:50:25PM +0200, David Hildenbrand wrote:
> +/*
> + * Is the pagefault handler disabled? If so, user access methods will not
> sleep.
> + */
> +#define pagefault_disabled() (current->pagefault_disabled != 0)
So -RT has:
static inline bool pagefault_disabled(void)
{
On Thu, May 07, 2015 at 12:50:53PM +0200, David Hildenbrand wrote:
Just to make sure we have a common understanding (as written in my cover
letter):
Your suggestion won't work with !CONFIG_PREEMPT (!CONFIG_PREEMPT_COUNT). If
there is no preempt counter, in_atomic() won't work.
But there
* David Hildenbrand d...@linux.vnet.ibm.com wrote:
AFAICR we did this to avoid having to do both:
preempt_disable();
pagefault_disable();
in a fair number of places -- just like this patch-set does, this is
touching two cachelines where one would have been enough.
* David Hildenbrand d...@linux.vnet.ibm.com wrote:
On Thu, May 07, 2015 at 12:50:53PM +0200, David Hildenbrand wrote:
Just to make sure we have a common understanding (as written in my cover
letter):
Your suggestion won't work with !CONFIG_PREEMPT (!CONFIG_PREEMPT_COUNT).
If
* David Hildenbrand d...@linux.vnet.ibm.com wrote:
On Thu, May 07, 2015 at 12:50:53PM +0200, David Hildenbrand wrote:
Just to make sure we have a common understanding (as written in my cover
letter):
Your suggestion won't work with !CONFIG_PREEMPT
On Thu, May 07, 2015 at 12:50:53PM +0200, David Hildenbrand wrote:
Just to make sure we have a common understanding (as written in my cover
letter):
Your suggestion won't work with !CONFIG_PREEMPT (!CONFIG_PREEMPT_COUNT). If
there is no preempt counter, in_atomic() won't work.
But
On Thu, May 07, 2015 at 01:40:30PM +0200, David Hildenbrand wrote:
I think a separate counter just makes sense, as we are dealing with two
different concepts and we don't want to lose the preempt_disable =^ NOP
for !CONFIG_PREEMPT.
Right, let me try and get my head on straight -- I'm so used
On Thu, May 07, 2015 at 01:23:35PM +0200, David Hildenbrand wrote:
On Thu, May 07, 2015 at 12:50:53PM +0200, David Hildenbrand wrote:
Just to make sure we have a common understanding (as written in my cover
letter):
Your suggestion won't work with !CONFIG_PREEMPT
On Thu, May 07, 2015 at 02:14:39PM +0200, David Hildenbrand wrote:
Thanks :), well just to make sure I got your opinion on this correctly:
1. You think that 2 counters is the way to go for now
ack
2. You agree that we can't replace preempt_disable()+pagefault_disable() with
On Thu, May 07, 2015 at 01:40:30PM +0200, David Hildenbrand wrote:
But anyhow, opinions seem to differ how to best handle that whole stuff.
I think a separate counter just makes sense, as we are dealing with two
different concepts and we don't want to lose the preempt_disable =^ NOP
for
On Wed, May 06, 2015 at 07:50:25PM +0200, David Hildenbrand wrote:
+/*
+ * Is the pagefault handler disabled? If so, user access methods will not
sleep.
+ */
+#define pagefault_disabled() (current-pagefault_disabled != 0)
So -RT has:
static inline bool pagefault_disabled(void)
{
On Wed, May 06, 2015 at 07:50:25PM +0200, David Hildenbrand wrote:
+/*
+ * Is the pagefault handler disabled? If so, user access methods will not
sleep.
+ */
+#define pagefault_disabled() (current-pagefault_disabled != 0)
So -RT has:
static inline bool pagefault_disabled(void)
{
On Thu, May 07, 2015 at 12:50:53PM +0200, David Hildenbrand wrote:
Just to make sure we have a common understanding (as written in my cover
letter):
Your suggestion won't work with !CONFIG_PREEMPT (!CONFIG_PREEMPT_COUNT). If
there is no preempt counter, in_atomic() won't work.
But
On Thu, May 07, 2015 at 01:40:30PM +0200, David Hildenbrand wrote:
But anyhow, opinions seem to differ how to best handle that whole stuff.
I think a separate counter just makes sense, as we are dealing with two
different concepts and we don't want to lose the preempt_disable =^ NOP
* David Hildenbrand d...@linux.vnet.ibm.com wrote:
@Ingo, do you have a strong feeling against this whole
patchset/idea?
No objections, sounds good to me now.
Thanks,
Ingo
--
To unsubscribe from this list: send the line unsubscribe linux-kernel in
the body of a message to
Until now, pagefault_disable()/pagefault_enabled() used the preempt
count to track whether in an environment with pagefaults disabled (can
be queried via in_atomic()).
This patch introduces a separate counter in task_struct to count the
level of pagefault_disable() calls. We'll keep manipulating
Until now, pagefault_disable()/pagefault_enabled() used the preempt
count to track whether in an environment with pagefaults disabled (can
be queried via in_atomic()).
This patch introduces a separate counter in task_struct to count the
level of pagefault_disable() calls. We'll keep manipulating
30 matches
Mail list logo