Re: [RFC] A more general timeout specification

[Kyle Moffett]

On Sep 1, 2005, at 11:18:52, Roman Zippel wrote:
> On Thu, 1 Sep 2005, Joe Korty wrote:
> > On Thu, Sep 01, 2005 at 11:19:51AM +0200, Roman Zippel wrote:
> > > You still didn't explain what's the point in choosing
> > > different clock sources for a _timeout_.
> >
> > Well, if CLOCK_REALTIME is set forward by a minute,
> > timers & timeout specified against that clock will expire
> > a minute earlier than expected.
>
> That just rather suggests that the pthread API is broken as usual.
> (No other possible user was mentioned so far.)

How about a hypothetical time-based event daemon.  I want to run
some jobs every 10 minutes that the system is running (not off or
suspended), I want to run other jobs every hour in real time, and
if one such timer expires while suspended, I want to run it
immediately to catch up.  The first suggests CLOCK_MONOTONIC, and
the second works better with CLOCK_REALTIME.

> So in practice it's easier to advance CLOCK_MONOTONIC/CLOCK_REALTIME
> equally and only apply time jumps to CLOCK_REALTIME.

I thought that's what he said, but maybe I'm just confused :-D.

Cheers,
Kyle Moffett

--
Premature optimization is the root of all evil in programming
  -- C.A.R. Hoare



-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/