Re: schedule_timeout() semantics/usage?

Karsten Keil (kkeil@suse.de)
Thu, 21 Oct 1999 09:54:31 +0200

On Thu, Oct 21, 1999 at 02:34:30AM +0200, Andrea Arcangeli wrote:
> On Wed, 20 Oct 1999, David Hinds wrote:
>
> >Look at net/irda/*.c [..]
>
> /* Make sure nobody tries to transmit during the speed change */
> while (irda_lock((void *) &self->netdev.tbusy) == FALSE) {
> WARNING(__FUNCTION__ "(), device locked!\n");
> current->state = TASK_INTERRUPTIBLE;
> schedule_timeout(MSECS_TO_JIFFIES(10));
>
> if (n++ > 10) {
> WARNING(__FUNCTION__ "(), breaking loop!\n");
> break;
> }
> }
>
> The above looks a bug. It should be using
> __set_current_state(TASK_UNINTERRUPTIBLE).
>
> > [..] or isdn/hisax/*.c, for example.
>
> Yes, it should be using UNINTERRUPTIBLE as well.
>
> This is a case where most people say "Oh well..." ;).

For hisax not true, or I missunderstand it.
In HiSax, the schedule_timeout are use allways on init stage (during insmod)
for 3 purposes:

1.) To do delay (value is not important) for asserting reset on isdn chips.

2.) To wait for an IRQ to test the IRQ path (I send a command to the isdn
chip which 'll cause an IRQ, then I wait with schedule_timeout().

3.) Timer test, some cards have a onboard timer which e.g 8ms, so I enable
this timer IRQ and wait with schedule_timeout(), after that I test how
many time the interrupt function was called.

Is this a right use of
current->state = TASK_INTERRUPTIBLE;
schedule_timeout(delay);

, what is a better way to do point 2 or 3, 1. can be also a simple delay loop.

Karsten

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