Is it OK to sleep when interrupts are disabled ?
The following is the implementation of interruptible_sleep_on in 2.6.10
Here we see that interrupts are disabled before schedule is called and
enabled back only when the control comes back. I wanted to know the
reason for doing this. And also whether sleeping with interrupts disabled
has any negetive impact.
2872 #define SLEEP_ON_VAR \
2873 unsigned long flags; \
2874 wait_queue_t wait; \
2875 init_waitqueue_entry(&wait, current);
2876
2877 #define SLEEP_ON_HEAD \
2878 spin_lock_irqsave(&q->lock,flags); \
2879 __add_wait_queue(q, &wait); \
2880 spin_unlock(&q->lock);
2881
2882 #define SLEEP_ON_TAIL \
2883 spin_lock_irq(&q->lock); \
2884 __remove_wait_queue(q, &wait); \
2885 spin_unlock_irqrestore(&q->lock, flags);
2886
2887 void fastcall __sched interruptible_sleep_on(wait_queue_head_t *q)
2888 {
2889 SLEEP_ON_VAR
2890
2891 current->state = TASK_INTERRUPTIBLE;
2892
2893 SLEEP_ON_HEAD
2894 schedule();
2895 SLEEP_ON_TAIL
2896 }
Gopu.
--
___________________________________________________________
Sign-up for Ads Free at Mail.com
http://promo.mail.com/adsfreejump.htm
--
Kernelnewbies: Help each other learn about the Linux kernel.
Archive: http://mail.nl.linux.org/kernelnewbies/
FAQ: http://kernelnewbies.org/faq/
