> Wait a minute. We have both semantics in kernel drivers.
> Ted's random does it the way you suggest it, but it does
> look from the code that it hasn't been that way all the time.
>
> Parport/Printer stuff seems to do it the way I do it,
> and the tty stuff can be configured obviously (using the termios
> stuff).
>
> Is there any argument for or against one behaviour except
> OSS behaviour? Standards? I didn't find anything useful
> in the online unix98 spec.
>
> We can discuss about changing the semantics, the patch proposed
> however looks (as the author also states) like an (IMO rather ugly)
> hack, Ted did it much more cleanly in his random driver.
Joysticks and the upcoming input device suite follows the same
lines - if there is anything, read it, regardless how the buffer
is big, and if there isn't anything, block and wait for something
to arrive.
Vojtech
-
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/