>----Oorspronkelijk bericht---- >Van: borasahin@xxxxxxxxxxx >Datum : 21/09/2006 14:03 >Aan: "Chris"<chris.boon@xxxxxxxxxx> >Cc: <kernelnewbies@xxxxxxxxxxxx> >Onderw: Re: interruptible_sleep_on() vs wait_event_interruptible() > > >> in my read function i use wait_event_interruptible, when >> readpointer==writepointer in my linked list, and in the interrupt >> routine i wake up the process waiting on the wait queue, after putting >> some data in the linked list. The interrupts occur every 400uS. >> >> my question was: when i use interruptible_sleep_on() instead of >> wake_event_interruptible, is the only reason NOT to use >> interruptible_sleep_on() because of the race i mentioned above. > >I think so... > >> i dont mind then i miss a interrupt and the process isnt awaken. > >But the only user of it or the only type of usage is not yours. Perhaps I dont >understand you wrongly, but from what your sayings, the meaning came out is >that: I dont care race so whether or not interruptible_sleep_on or >wait_event_interruptible be used isnt important for me. >interruptible_sleep_on is depreceated so you should stick to >wait_event_interruptible in any case. I dont care if it is depreceated or not. I just want to know why it is depreceated. Did you have a look in the kernel-2.6.18 tree how many drivers still use interruptible_sleep_on() ? (don't mean to offend you by the way ;) Chris. -- Kernelnewbies: Help each other learn about the Linux kernel. Archive: http://mail.nl.linux.org/kernelnewbies/ FAQ: http://kernelnewbies.org/faq/
