Date: Thu, 12 Jan 2006 19:33:32 -0500 (EST) From: Andrew Gallatin <gallatin@cs.duke.edu> To: Nate Lawson <nate@root.org> Cc: cvs-src@FreeBSD.org, Scott Long <scottl@samsco.org>, src-committers@FreeBSD.org, cvs-all@FreeBSD.org Subject: Re: cvs commit: src/sys/dev/em if_em.c if_em.h Message-ID: <17350.62812.122727.808320@grasshopper.cs.duke.edu> In-Reply-To: <43C6E2EB.7040902@root.org> References: <200601110030.k0B0UPOx009098@repoman.freebsd.org> <20060112152119.A6776@grasshopper.cs.duke.edu> <43C6C4EA.20303@samsco.org> <17350.53992.494972.787933@grasshopper.cs.duke.edu> <43C6DAED.3040901@samsco.org> <43C6E2EB.7040902@root.org>
next in thread | previous in thread | raw e-mail | index | archive | help
Nate Lawson writes: > Scott Long wrote: > > Andrew Gallatin wrote: > > > >> Scott Long writes: > >> > However, taskqueues are really just a proof of concept for what I > >> really > >> > want, which is to allow drivers to register both a fast handler and an > >> > ithread handler. For drivers doing this, the ithread would be private > >> > >> Ah, the darwin / MacOSX model. That would be very cool. > > > > > > Yep. Working in IOKit was very interesting, and this is one of the few > > things that transfers well to FreeBSD. C++ does have a certain elagence > > for drivers, but the cost of virtual methods in the fast path of the > > driver and stack is still far too high to justify using it. > > Yeah, that's why newbus is only for configuration but interrupt handlers > are called directly. It's nice to have OO for config and relationships > between devices but just direct calls for the fast path. Amusingly, I was just today trying to figure out how to get at my Darwin driver's if_output() routine so I can cut out the interposing layers of C++ crud and improve transmit performance :) Drew
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?17350.62812.122727.808320>