Date: Mon, 04 Mar 2002 13:47:25 -0800 From: Maksim Yevmenkin <myevmenk@digisle.net> To: Julian Elischer <julian@elischer.org> Cc: freebsd-current@FreeBSD.ORG, archie@FreeBSD.ORG Subject: Re: Netgraph, device drivers and mutexes Message-ID: <3C83EB6D.CFE8016C@digisle.net> References: <Pine.BSF.4.21.0203041114200.26829-100000@InterJet.elischer.org>
next in thread | previous in thread | raw e-mail | index | archive | help
Julian, [...] > > > I hope that this helps you! > > > > yes it did help :) i changed my code and it seems to work just fine. > > i wish i had SMP laptop to test it :) > > Well it aint exactly SMP safe YET, until I make those changes through teh > REST of the system. There are still direct timeout() calls in several > modules that I need to change to follow my own suggestions and there are > many nodes that need to be changed to gain a lock when they first > try insert data into the graph. e.g. ng_tty, ng_ether, speaking of ng_tty... it is clear to me how to inject data into Netgraph in a safe way, but it is not yet clear how Netgraph can inject data into other subsystems. you see, the Bluetooth spec defines several Host (PC) to Host Controller (Bluetooth unit) communication protocols. one of them is UART transport layer (AKA H4). i have implemented H4 line discipline that also a Netgraph node. (i called it ng_sio in my report but it was wrong). it works now, and i can talk to Xircom card, but it should be changed later. any hints? thanks, max To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?3C83EB6D.CFE8016C>