Ludovic Rousseau wrote: > 2007/11/13, Douglas E. Engert <[EMAIL PROTECTED]>: >> Got the card to work with MIT Kerberos to PKINIT to PKCS#11 >> with Windows AD acting as KDC. >> >> In addition to the thew changes committed by Ludovic yesterday, >> I had to make the attached changes. >> >> flags |= ; >> >> Tell OpenSC that the card could do an RSA operation without >> doing any hash. > > Is this change still needed now that RAW RSA is not needed?
Yes I thing so. Otherwise the code in in pkcs15-sec.c around lines 204-215 return SC_ERROR_INVALID_DATA. That was why I asked about RAW. The card can do RSA with padding without hashing the data. I believe that the SC_ALGORITHM_RSA_HASH_NONE says this. Otherwise opensc would assumed that the card can only do RSA if by hashing then padding on the card first. > >> ret = 0x12; >> >> The decrypt operation needed to expect PKCS1 padding. The >> original code has ret = 13, which produced SW of 69 82 >> I think this might be a bug in the original code? > > committed in revision 3301 I don't have the manuals, but this does work. David, do you have any comments on this? > >> 0x04, "01", SC_PKCS15_CO_FLAG_PRIVATE}, >> >> Change the key reference for thr private key from 0x03 to 0x04. >> This might be the difference in the cards or the way Gemsafe >> formated them. Is there some other DF on the card that could be >> used to determine the key reference? > > David, do you have a regression with this change? It could be differences in the GemSafe versions. Could keep off the ATR and set the card->type, if there is no other way to tell. > > Bye, > -- Douglas E. Engert <[EMAIL PROTECTED]> Argonne National Laboratory 9700 South Cass Avenue Argonne, Illinois 60439 (630) 252-5444 _______________________________________________ opensc-devel mailing list [email protected] http://www.opensc-project.org/mailman/listinfo/opensc-devel
