Re: [patch 4/4] hpet: factor timer allocate from open

Re: [patch 4/4] hpet: factor timer allocate from open

[Thomas Gleixner]

On Mon, 5 Mar 2012, akpm@linux-foundation.org wrote:

> From: Magnus Lynch <maglyx@gmail.com>
> Subject: hpet: factor timer allocate from open
> 
> The current implementation of the /dev/hpet driver couples opening the
> device with allocating one of the (scarce) timers (aka comparators).  This
> is a limitation in that the main counter may be valuable to applications
> seeking a high-resolution timer who have no use for the interrupt
> generating functionality of the comparators.
> 
> This patch alters the open semantics so that when the device is opened, no
> timer is allocated.  Operations that depend on a timer being in context
> implicitly attempt allocating a timer, to maintain backward compatibility.

AFAICT, there is only fasync which calls hpet_alloc_timer(). All other
functions are broken by this chage.

>  There is also an IOCTL (HPET_ALLOC_TIMER _IO) added so that the
> allocation may be done explicitly.  (I prefer the explicit open then

And that IOCTL does what? Return 0 is not really useful functionality.

Thanks,

	tglx

Re: [patch 4/4] hpet: factor timer allocate from open

[Clemens Ladisch]

Thomas Gleixner wrote:
> On Mon, 5 Mar 2012, akpm@linux-foundation.org wrote:
>> From: Magnus Lynch <maglyx@gmail.com>
>> Subject: hpet: factor timer allocate from open
>>
>> The current implementation of the /dev/hpet driver couples opening the
>> device with allocating one of the (scarce) timers (aka comparators).  This
>> is a limitation in that the main counter may be valuable to applications
>> seeking a high-resolution timer who have no use for the interrupt
>> generating functionality of the comparators.
>>
>> This patch alters the open semantics so that when the device is opened, no
>> timer is allocated.  Operations that depend on a timer being in context
>> implicitly attempt allocating a timer, to maintain backward compatibility.
>
> AFAICT, there is only fasync which calls hpet_alloc_timer(). All other
> functions are broken by this chage.

Indeed.  There was another call in hpet_ioctl_common(), but it appears
that one somehow got dropped.

Magnus, is this the patch that you submitted and have tested?


Regards,
Clemens

Re: [patch 4/4] hpet: factor timer allocate from open

[Andrew Morton]

On Tue, 06 Mar 2012 16:04:21 +0100
Clemens Ladisch <clemens@ladisch.de> wrote:

> Thomas Gleixner wrote:
> > On Mon, 5 Mar 2012, akpm@linux-foundation.org wrote:
> >> From: Magnus Lynch <maglyx@gmail.com>
> >> Subject: hpet: factor timer allocate from open
> >>
> >> The current implementation of the /dev/hpet driver couples opening the
> >> device with allocating one of the (scarce) timers (aka comparators).  This
> >> is a limitation in that the main counter may be valuable to applications
> >> seeking a high-resolution timer who have no use for the interrupt
> >> generating functionality of the comparators.
> >>
> >> This patch alters the open semantics so that when the device is opened, no
> >> timer is allocated.  Operations that depend on a timer being in context
> >> implicitly attempt allocating a timer, to maintain backward compatibility.
> >
> > AFAICT, there is only fasync which calls hpet_alloc_timer(). All other
> > functions are broken by this chage.
> 
> Indeed.  There was another call in hpet_ioctl_common(), but it appears
> that one somehow got dropped.
> 
> Magnus, is this the patch that you submitted and have tested?

I've been carrying this patch since March 2010 - perhaps it rotted.