Takashi Iwai wrote:

>if it's on the blocking mode, the driver should block if it's possible
>to do write in future _by any chance_.
>if it's not on the blocking mode, the driver should return -EAGAIN.
>and, the behavior of poll() depends on the behavior of write().
>if it blocks, then poll blocks, too.

sorry, i beg to differ. poll returning immediately with EAGAIN on a
nonblock fd doesn't make sense to me. 

you call poll because you want to block until a read/write possible
condition, and not to check if read/write is possible now -- you'd
call read()/write() instead.

that's the behaviour other linux fds (file, socket, fifo etc) show
when set to nonblock, and that's what i think one naturally expects.

tim



-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf
_______________________________________________
Alsa-devel mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/alsa-devel

Reply via email to