>I don't know all about mmap, but why does one need to poll. >I would have thought that a callback with info on how many samples it wants >would be a better way.
there is no generalized mechanism for the kernel to call userspace code. POSIX signals are the canonical examples of such a thing, where userspace attaches a handler, and the kernel ensures that if the signal is delivered, the thread returns to user space to execute that code. however, this is a highly specific mechanism with lots of semantics that get in the way of this being a broadly useful system. some unix-like systems have played with allowing kernel->userspace callbacks; none of the APIs has ever taken off. i even implemented one once for Mach - it wasn't easy :) poll(2) is a nice method for this in the absence of such a mechanism. both it and select(2) are moderately broad mechanisms that allow userspace to execute code when certain conditions happen. --p _______________________________________________ Alsa-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/alsa-devel
