Date: Tue, 23 Mar 2004 18:21:44 +0100 From: Bernd Walter <ticso@cicely12.cicely.de> To: "M. Warner Losh" <imp@bsdimp.com> Cc: ticso@cicely.de Subject: Opinions about using USB serial numbers - Patch Message-ID: <20040323172142.GF63452@cicely12.cicely.de> In-Reply-To: <20040322151647.GU63452@cicely12.cicely.de> References: <20040320052007.GC33602@cicely12.cicely.de> <20040320.184336.88475380.imp@bsdimp.com> <20040321071100.GH33602@cicely12.cicely.de> <20040322123048.GP63452@cicely12.cicely.de> <20040322151647.GU63452@cicely12.cicely.de>
next in thread | previous in thread | raw e-mail | index | archive | help
On Mon, Mar 22, 2004 at 04:16:48PM +0100, Bernd Walter wrote: > On Mon, Mar 22, 2004 at 01:30:49PM +0100, Bernd Walter wrote: > > What still puzzles me is why only the nomatch event includes pnpinfo. > > I've found the code in subr_bus and it looks very intentionally to me. > Should I move the serial number into the location string? > Or can we add the pnpinfo to attach/detach messages as well? I just added pnpinfo to attach/detach messages as well and I managed to get devd to create the required symlinks: Processing event '+ubser0 vendor=0x03da product=0x0001 devclass=0x00 devsubclass=0x00 sernum="1080054842" intclass=0xff intsubclass=0x00 at port=3 interface=0 on uhub3' Pushing table device-name=ubser0 vendor=0x03da product=0x0001 devclass=0x00 devsubclass=0x00 sernum=1080054842 intclass=0xff intsubclass=0x00 Processing attach event Testing device-name=ubser0 against ^ed50 Testing device-name=ubser0 against ^ubser[0-9]+ Executing '/etc/devd/ubser-start.sh ubser0 1080054842' Popping table Processing event '-ubser0 vendor=0x03da product=0x0001 devclass=0x00 devsubclass=0x00 sernum="" intclass=0xff intsubclass=0x00 at port=3 interface=0 on uhub3' Pushing table device-name=ubser0 vendor=0x03da product=0x0001 devclass=0x00 devsubclass=0x00 sernum= intclass=0xff intsubclass=0x00 Processing detach event Testing device-name=ubser0 against ^ed50 Testing device-name=ubser0 against ^ubser[0-9]+ Executing '/etc/devd/ubser-stop.sh ubser0 ' Popping table [102]cicely13# ls -al /dev/ubser0.* crw-rw---- 1 uucp dialer 233, 0 Mar 23 17:34 /dev/ubser0.0 crw-rw---- 1 uucp dialer 233, 1 Mar 23 17:34 /dev/ubser0.1 lrwxr-xr-x 1 root wheel 8 Mar 23 17:54 /dev/ubser.1080054842.0 -> ubser0.0 lrwxr-xr-x 1 root wheel 8 Mar 23 17:54 /dev/ubser.1080054842.1 -> ubser0.1 lrwxr-xr-x 1 root wheel 8 Mar 23 17:54 /dev/ubser.1080054842.2 -> ubser0.2 lrwxr-xr-x 1 root wheel 8 Mar 23 17:54 /dev/ubser.1080054842.3 -> ubser0.3 lrwxr-xr-x 1 root wheel 8 Mar 23 17:54 /dev/ubser.1080054842.4 -> ubser0.4 lrwxr-xr-x 1 root wheel 8 Mar 23 17:54 /dev/ubser.1080054842.5 -> ubser0.5 crw-rw---- 1 uucp dialer 233, 2 Mar 23 17:34 /dev/ubser0.2 crw-rw---- 1 uucp dialer 233, 3 Mar 23 17:34 /dev/ubser0.3 crw-rw---- 1 uucp dialer 233, 4 Mar 23 17:34 /dev/ubser0.4 crw-rw---- 1 uucp dialer 233, 5 Mar 23 17:34 /dev/ubser0.5 [103]cicely13# ls -al /dev/ubser0.* ls: No match. Exit 1 The detach message doesn't contain the serial since this is not cached as the other values - however the device-name is enough to rm the links at detach because we know they are pointing to ubser0.* nodes. Since this is an interface level driver I should add the interface number to the symlink name as well - but that's up to the one configuring devd. The patch is at http://www.bwct.de/usb-devd.patch -- B.Walter BWCT http://www.bwct.de ticso@bwct.de info@bwct.de
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20040323172142.GF63452>