Re: kqueue

On Thu, Aug 24, 2000 at 01:09:57PM -0400, Uri Guttman <[email protected]> wrote:
> please quote the docs that show this. according to my solaris 7 docs,
> you have to call aio_wait to get an aio_result back. this can be one of

I am talking about SIGIO and friends, not the posix aio_*-functions. aio_*
might be implemented using sigio, but does not need to, this is the reason
why the only (portable) way to do aio using aio_* functions is, well,
using the aio_* functions.

> there is no method for the signal mechanism that i see that tell you
> which handle is available to read/write.

Hmm, maybe this is a limitation in solaris (I am not 100% sure), but
hp-ux, freebsd and linux all deliver the necessary event info with the
signal (and, if you use a realtime signal, it will utilize the internal
kernel event queue so no events get lost).

If solaris doesn't support siginfo it indeed lacks a very important part
of the posix realtime extensions, which I would find disturbing.

> also show me the exact man pages in linux which has similar features as
> solaris aio support. i have not found any. 

I don't think there are manual pages so far (at leats I don't have them
installed, but I haven't updated my manpages since ages), but the libc
documentation is far more extensive on that topic, anyway, so you might
consult that one.

-- 
      -----==-                                             |
      ----==-- _                                           |
      ---==---(_)__  __ ____  __       Marc Lehmann      +--
      --==---/ / _ \/ // /\ \/ /       [email protected] |e|
      -=====/_/_//_/\_,_/ /_/\_\       XX11-RIPE         --+
    The choice of a GNU generation                       |
                                                         |