Hi, all,
Thanks for your patient and kind support. Now I fully understand this

piece of code.
And during the course of making the patch, I find the following code

(comments) fragment in include/linux/wait.h:
/*

* These are the old interfaces to sleep waiting for an event.

* They are racy.  DO NOT use them, use the wait_event* interfaces above.

* We plan to remove these interfaces during 2.7.

*/

extern void FASTCALL(sleep_on(wait_queue_head_t *q));

extern long FASTCALL(sleep_on_timeout(wait_queue_head_t *q,

      signed long timeout));

extern void FASTCALL(interruptible_sleep_on(wait_queue_head_t *q));

extern long FASTCALL(interruptible_sleep_on_timeout(wait_queue_head_t *

                                 signed long timeout));
Since the comments to warn kernel developers have been there, it is

acceptable for me and therefore no need to create another patch.
Thanks everyone again.

Feng,Dong
2006/12/25, Randy Dunlap :

quoted text
> On Sun, 24 Dec 2006 23:55:29 +0800 Dong Feng wrote:

>

> > 2006/12/24, Arjan van de Ven :

> > >

> > > hi,

> > >

> > > first of all your patch is word-wrapped so won't apply ;)

> > >

> > > hmm. There are many failure modes, it may just be simpler to say that

> > > sleep_on() is inherently racey... maybe give an example but don't make

> > > it look this limited.. there are many ways in which it's broken ;_)

> > >

> > > (sleep_on_interruptible is sometimes used correctly in that the timeout

> > > will save you; it even is sometimes used instead of msleep(), and it's

> > > use should be replaced by msleep() instead ... )

> > >

> > >

> > > there is wait_event() and family of API's that is there to resolve

> > > this....

> > >

> >

> > Hi, thanks a lot for the answer and comments. I re-create the patch

> > and please you and others review it.

> >

> >

> > --- linux-2.6.19/kernel/sched.c       2006-11-30 05:57:37.000000000 +0800

> > +++ patch/sched.c     2006-12-24 23:36:15.000000000 +0800

> > @@ -3823,6 +3823,9 @@

> >  }

> >  EXPORT_SYMBOL(interruptible_sleep_on);

> >

> > +/*

> > + *  The function has been obsoleted. See sleep_on().

> > + */

> >  long fastcall __sched

> >  interruptible_sleep_on_timeout(wait_queue_head_t *q, long timeout)

> >  {

> > @@ -3838,6 +3841,14 @@

> >  }

> >  EXPORT_SYMBOL(interruptible_sleep_on_timeout);

> >

> > +/*

> > + *  Function sleep_on() has been obsoleted. Some legacy drivers

> > + *  still use it but they shall be fixed. The function is inherently

> > + *  racey.

>

> Spell it "racy", please.

>

> > + *

> > + *  Use function wait_event() or the appropriate family functions

> > + *  instead.

>

> Can someone provide more examples of the "appropriate" families

> of functions to use here?

>

> > + */

> >  void fastcall __sched sleep_on(wait_queue_head_t *q)

> >  {

> >       SLEEP_ON_VAR

>

> Would using __deprecated be too noisy in builds?

>

> ---

> ~Randy

>

_______________________________________________

Kernel-mentors mailing list

Kernel-mentors@selenic.com

http://selenic.com/mailman/listinfo/kernel-mentors