lkml.org 
[lkml]   [2014]   [Mar]   [4]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
From
Date
SubjectRe: [PATCH v4 5/6] timerfd: Add support for deferrable timers
On Tue, Mar 4, 2014 at 2:11 PM, Thomas Gleixner <tglx@linutronix.de> wrote:
> On Tue, 4 Mar 2014, Andy Lutomirski wrote:
>
>> On Tue, Mar 4, 2014 at 12:58 PM, Thomas Gleixner <tglx@linutronix.de> wrote:
>> > On Tue, 25 Feb 2014, Andy Lutomirski wrote:
>> >> On the other hand, if you added a fancier version of timerfd_settime
>> >> that could explicitly set the slack value (or, equivalently, the
>> >> earliest and latest allowable times), that could be quite useful.
>> >>
>> >> It's often bugged me that timer slack is per-process.
>> >
>> > That's a totally different issue. There is a world aside of timerfd
>> > timers.
>>
>> This is a patch to add deferrable support *to timerfd*. I'm asking
>
> There is a new patch series which adds deferrable support to all timer
> related interfaces which have a flags field. And that's the only
> sensible solution right now.
>
> We do no add another random special case syscall for timerfd just
> because timerfd is linux specific.

What syscalls? I can think of exactly two timer interfaces that
actually accept a clock id and flags: clock_nanosleep and
timerfd_settime.

>
> No, we want to support that stuff right now with the existing
> interfaces as we have to revisit all of the timer related interfaces
> in the near future anyway due to the Y2038 issue.
>
> And your idea of per thread slack is completely bogus. If we want to
> make the slack value usefull then it needs to be done per timer and
> not per thread/process.

This is *exactly* what I'm suggesting.

>
> But we cannot do that right now as we cannot whip up severl dozen of
> new syscalls just because we want to add slack/deferrable whatever
> properties.

Two syscalls, right?

>
> Once we agree on a solution to the Y2038 issue on 32bit with a unified
> 32/64 bit syscall interface which simply gets rid of the timespec/val
> nonsense and takes a simple u64 nsec value we can add the slack
> property to that without any further inconvenience.
>
> Thanks,
>
> tglx
>



--
Andy Lutomirski
AMA Capital Management, LLC


\
 
 \ /
  Last update: 2014-03-07 10:41    [W:0.048 / U:0.140 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site