On Mon, 22 Jul 2002 08:35, Matthew Dharm wrote: > It's basically an impossible problem. Only the device, which might have a > physical sensor to detect removal, can make a positive determination. > > Most devices err on the side of caution -- if they are powering up due to a > bus attach, they assume media-change while they were off. Actually, many > assume media-change at attachment, because they're not sure if they are > attached to the same host. > > Most media doesn't have a serial number or any way to uniquely identify it. So we could only validate this at the filesystem layer? Which may not be in use (eg mtools), and for some filesystems, may not have ids :(
> Really, the only safe thing to do is assume media-change on > removal/insertion, given that devices (as a generalized term) are > inconsistant in their ability/willingness to indicate media change. Could you "quirk" this, so that the normal assumption is that unannounced media change did occur, and then make happier assumptions for known good devices? -- http://conf.linux.org.au. 22-25Jan2003. Perth, Australia. Birds in Black. ------------------------------------------------------- This sf.net email is sponsored by:ThinkGeek Welcome to geek heaven. http://thinkgeek.com/sf _______________________________________________ [EMAIL PROTECTED] To unsubscribe, use the last form field at: https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
