Date: Tue, 18 Jan 2005 20:19:34 +0000 From: Steven Smith <sos22@cantab.net> To: Street Chaman <street_chaman@hotmail.com> Cc: sos22@srcf.ucam.org Subject: Re: Kernel mode programming -precisions Message-ID: <20050118201934.GB7786@archibold> In-Reply-To: <BAY15-F6A965C17D8FF205AB06FC998F0@phx.gbl> References: <BAY15-F6A965C17D8FF205AB06FC998F0@phx.gbl>
next in thread | previous in thread | raw e-mail | index | archive | help
--K8nIJk4ghYZn606h Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable > When CPU is waiting on some new data, he cannot preform any other=20 > operation, and freezes data manipultion; that's why I'm looking for a new= =20 > way to in/output the soft... You may want to look at select(2) and poll(2). Those allow you to check whether input is available on a given file descriptor without having to block if it isn't. > I've found out in syscall.master the aio_read/write call. Man page says= =20 > operation is asynchronous, so it returns quite immediatly after enqueuing= =20 > the io thread. Could that be a solution ? Yes, but it sounds a little like overkill for what you're talking about. > Or back to kernel module (by the way: is it possible to write on in > assembly) ? Yes, absolutely, but it's quite an unusual thing to do. It'd be far more common to write most of the module in C with a few little bits of assembly in really critical parts. Steven. --K8nIJk4ghYZn606h Content-Type: application/pgp-signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.6 (FreeBSD) iD8DBQFB7W9WO4S8/gLNrjcRArYYAKCP6+Oe28xWDs7p4rXxFQRDqEfoeQCfXckG o45j9hRkBJgKZKHy8VeNOyk= =4PjD -----END PGP SIGNATURE----- --K8nIJk4ghYZn606h--
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20050118201934.GB7786>