From mboxrd@z Thu Jan 1 00:00:00 1970 From: =?utf-8?B?SsO2cm4=?= Engel Date: Mon, 24 Mar 2008 16:04:30 +0000 Subject: Re: [PATCH] 3/3 maple: update bus driver to support Dreamcast VMU Message-Id: <20080324160429.GG2899@logfs.org> List-Id: References: <1206207805.6324.13.camel@localhost.localdomain> <1206209786.6324.41.camel@localhost.localdomain> <20080324033344.GB15872@linux-sh.org> <20080324144647.GC2899@logfs.org> <1206371178.6283.37.camel@localhost.localdomain> <20080324152952.GF2899@logfs.org> <1206373900.6283.39.camel@localhost.localdomain> In-Reply-To: <1206373900.6283.39.camel@localhost.localdomain> MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: 8bit To: Adrian McMenamin Cc: Andrew Morton , linux-sh , Greg KH , LKML , Paul Mundt , MTD , dwmw2 On Mon, 24 March 2008 15:51:40 +0000, Adrian McMenamin wrote: > On Mon, 2008-03-24 at 16:29 +0100, Jörn Engel wrote: > > On Mon, 24 March 2008 15:06:18 +0000, Adrian McMenamin wrote: > > > > > > I don't think I have any alternative here. The problem is that the bus > > > checks periodically (every HZ ticks but that could be changed) if a > > > device remains connected to the system by queueing a status command. (If > > > the status command is replied to with a "No response" then that > > > indicates to the bus that the device has been removed. > > > > Where does this queueing happen? Somewhere in the kernel where we can > > control it or in hardware? > > Well, the queuing of the status command is under software control. Then we should be fine. I'll try to beat the code into submission. Jörn -- Joern's library part 7: http://www.usenix.org/publications/library/proceedings/neworl/full_papers/mckusick.a