Hello, I don't understand. OpenSC PKCS#11 modules *ALWAYS* allocate specific number of virtual slots. The plug&play just link between virtual slot and real slot. So it all works with Firefox 2.X and Firefox 3.X, I checked it now again with Firefox 3.X.
I am not happy with this implementation but this is the simplest one I could find without modifying the entire reader allocation. Anyway, as usual I disagree with NSS developers. Slots can be removed! As long as you don't reuse slot ids. This way if the application calls a function with removed slot id it gets a proper error. You can insert/remove token 4G times... I guess then you can reuse id 0... :) But we don't have this as we have a fixed set of slot number. If we want proper PKCS#11 implementation we will have to someday reimplement the whole slot usage... There is the stateless mode we need to add and true plug&play throughout the whole libopensc. Alon. On 1/19/09, Martin Paljak <mar...@paljak.pri.ee> wrote: > On Mon, Jan 19, 2009 at 3:23 PM, Alon Bar-Lev <alon.bar...@gmail.com> wrote: > > I think the 2.20 should be the default, as the plug&play mode is important. > > Agreed. > > > All 2.11 applications will also work as the number of slots is > > returned correctly and if the application wish to get fewer slots it > > will get only these. > > OpenSC has been 2.11 for all the time, I doubt there is any surprises > here for (PC/SC) users. > > > > So actually I don't think this option should be added... But if it is > > added, I think the default should be in favor of the new > > specification. > > I'm writing a longer follow-up to ticket #29 but you might be > interested in > http://groups.google.com/group/mozilla.dev.tech.crypto/browse_thread/thread/a776fe268d149b57 > as well. > > Basically, v2.20 mode should be the default but there are some kinks > to be ironed out first. > > > > > -- > Martin Paljak > http://martin.paljak.pri.ee > GSM:+3725156495 > _______________________________________________ opensc-devel mailing list opensc-devel@lists.opensc-project.org http://www.opensc-project.org/mailman/listinfo/opensc-devel
