From mboxrd@z Thu Jan 1 00:00:00 1970 From: Steven Rostedt Subject: Re: [ANNOUNCE] 3.10.6-rt3 Date: Mon, 19 Aug 2013 21:02:08 -0400 Message-ID: <20130819210208.426e01b3@gandalf.local.home> References: <20130812163413.GI23040@linutronix.de> <520D212F.7030400@localhost> <20130815152210.5ed93696@gandalf.local.home> <520DCE5F.1040402@linutronix.de> <521254A0.6050603@localhost> <20130819202903.692097ea@gandalf.local.home> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Cc: Fernando Lopez-Lezcano , Sebastian Andrzej Siewior , Kent Overstreet , linux-rt-users , LKML , Thomas Gleixner , John Kacur To: Steven Rostedt Return-path: In-Reply-To: <20130819202903.692097ea@gandalf.local.home> Sender: linux-kernel-owner@vger.kernel.org List-Id: linux-rt-users.vger.kernel.org On Mon, 19 Aug 2013 20:29:03 -0400 Steven Rostedt wrote: > On Mon, 19 Aug 2013 10:23:44 -0700 > Fernando Lopez-Lezcano wrote: > > > > > The problem is that bcache is using new semaphore functions which it > > > just introduced which rt does not know about. The comment above their > > > definition says that it is wrong to use them and completion is the > > > right way to do it. > > > So my question is, why don't we use completion but this nasty hack? > > I think I'm going to send them an email about that. > Looking at it more, I can now see why they did what they did. Grumble, we may need to bring back the old "compat" rwsem, or whatever we called it, and just use that instead. And say that this code will have no real time boosting. -- Steve