2010/1/14 Martin Paljak <mar...@paljak.pri.ee>: > Hello, > > Please help to review and improve a patch that does the following (or take > note of some thing that could be picked out as separate patches): > > Changes: > - Reader subsystem/core: > - removal of slot abstraction. Multislot readers, if found in the wild, can > be accessed as separate readers. > - remove "action" parameter from sc_disconnect_card(). It was not used. > - change the primary APIs from "int reader" to "sc_reader_t *reader". > Readers are kept in a list instead of an array. > - add sc_ctx_get_reader_by_name to find readers by name. > - pcsc: implement SCardCancel > - change the API and name of sc_wait_for_slot_event to sc_wait_for_event > and add sc_cancel. Only usable with PC/SC and don't follow the generic > "framework function" logic. > - pcsc: handle more PC/SC return codes. > - pcsc: add support for "PnP Notification" (hotplugged reader) to > pcsc_wait_for_event > - tools: > - all: make most tools use util_connect_card and allow fetching the reader > by name as well > - pkcs11-tool: test hotplugging and events > (C_GetSlotList/C_WaitForSlotEvent); a slot_id is mandatory for --test > - pkcs11-tool: implement CKA_ALWAYS_AUTHENTICATE and CKU_CONTEXT_SPECIFIC > login > - pkcs11-tool: unwraping needs a RW session. > - pkcs11-tool: properly call C_Finalize when a fatal error happens. > (Otherwise OSX pcsc locks up) > - pkcs11: > - slots, sessions and objects are kept as lists. > - change the way slots, cards and readers are managed. > - re-implement C_WaitForSlotEvent(/C_Finalize) as written in PCKS#11 v2.20, > canceling pending blocking calls. > - implement a "virtual hotplug slot" with a floating slot id to keep NSS > working with C_WaitForSlotEvent with a new reader. > NSS does not call C_GetSlotList(NULL) to re-fetch the list of available > slots if C_WaitForSlotEvent returns an event in an already known slot ID. > By changing the ID of a slot whenever a reader attached NSS/Firefox can > be tricked into recognizing new readers when waiting for events with > C_WaitForSlotEvent. > - change (possibly break something) sc_to_cryptoki_error() to not have > side-effects > - Implement CKU_CONTEXT_SPECIFIC in C_Login to implement > CKA_ALWAYS_AUTHENTICATE (keys with user consent) > > Applies to current trunk. I'll pick some small isolated things related to > pkcs#11(-tool) out one by one, but the listification and de-slotification > could IMHO go in one chunk.
I have no objection to the changes. I can't review your patch file. It is too big (9744 lines). Please commit your changes one by one. I imagine that is what you planned already. Bye -- Dr. Ludovic Rousseau _______________________________________________ opensc-devel mailing list opensc-devel@lists.opensc-project.org http://www.opensc-project.org/mailman/listinfo/opensc-devel
