Re: regarding timeout/untimeout kernel functions
From: pradeep reddy punnam (prady_p_at_yahoo.com)
Date: 07/23/04
- Previous message: Chris Pressey: "Re: "Next Generation" kernel configuration?"
- In reply to: Joseph M Link: "Re: regarding timeout/untimeout kernel functions"
- Next in thread: Dan Nelson: "Re: regarding timeout/untimeout kernel functions"
- Reply: Dan Nelson: "Re: regarding timeout/untimeout kernel functions"
- Reply: Joseph M Link: "Re: regarding timeout/untimeout kernel functions"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Date: Thu, 22 Jul 2004 19:23:24 -0700 (PDT) To: Joseph M Link <freebsd@joelink.net>
HI joseph ,
i thought of threading with select before , but i belive that if the number of timers to be checked increases the number of the threads to be maintained increses,so the process may become very hevy. what do u think.
i think ultimatley i am going to use the above thing.
but in the process of my search i came across the timeout kernel function implemenation
but i can not use that ( which i belive very efficient implementation of timers ), which user can not able to use it , so i just want to discuss it .
thanks
- pradeep
Joseph M Link <freebsd@joelink.net> wrote:
If you're willing to take some precautions, you could run the timer code
with select/usleep in a separate thread. However, since the callbacks
would originate from that thread, you would need mutexes to protect any
data that the function accesses that could also be accessed by the
normal program flow.
Joe
pradeep reddy punnam wrote:
> HI all,
> i am working on a project , where i came across a situation where i need to execute a function when a timer expires ,exactly similar to functionality of the timeout() kernel function but i need this in userland(application), and the execution of the function is time sensitive, it should be run immediately when timer expires.
>
> i can't be using poll or select for timer becuse those will block the process untill the timer expires.for me the proess should not be blocked.
> and i also thought of taking the service of the timeout function by writing a system call and using signaling mechanism but i think this will become expensive when the number of timers to be checked increeses.
>
> i read the kern_timeout.c code that is very good implentation.with very less expensive.
> but i think user unable to enjoy that service.
>
> i will thankful if somebody can tell if there is any such a service or way provided by os( that i overlooked).
>
> thanks,
>
> -Pradeep
>
>
>
>
>
>
> ---------------------------------
> Do you Yahoo!?
> Vote for the stars of Yahoo!'s next ad campaign!
> _______________________________________________
> freebsd-hackers@freebsd.org mailing list
> http://lists.freebsd.org/mailman/listinfo/freebsd-hackers
> To unsubscribe, send any mail to "freebsd-hackers-unsubscribe@freebsd.org"
---------------------------------
Do you Yahoo!?
Vote for the stars of Yahoo!'s next ad campaign!
_______________________________________________
freebsd-hackers@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-hackers
To unsubscribe, send any mail to "freebsd-hackers-unsubscribe@freebsd.org"
- Previous message: Chris Pressey: "Re: "Next Generation" kernel configuration?"
- In reply to: Joseph M Link: "Re: regarding timeout/untimeout kernel functions"
- Next in thread: Dan Nelson: "Re: regarding timeout/untimeout kernel functions"
- Reply: Dan Nelson: "Re: regarding timeout/untimeout kernel functions"
- Reply: Joseph M Link: "Re: regarding timeout/untimeout kernel functions"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]