Martin Paljak wrote:
Hi,
It turnes out that the default build on osx is not very usable on os x
when one starts to plug and unplug the card at 'strange' moments. I got
a lot of asserts on card_lock>=1, For example when removing the card
just when starting a SSL connection, before the pkcs#11 had been
initialized, or for example when removing the card before ssl session
had ended.
I spent some time trying to understand the issue and looking at the
locking code and doing various experiments.
Anyway now i can play in-out with the card while firefox stays open so
it seems to be better than before. Nils, Stef - please have a look and
comment/adjust.
...
Index: pkcs11/pkcs11-global.c
===================================================================
--- pkcs11/pkcs11-global.c (revision 343)
+++ pkcs11/pkcs11-global.c (working copy)
@@ -29,6 +29,9 @@
#endif
#include "sc-pkcs11.h"
+/* Usually we want to support threads however we can (for example, to make FF
happy) */
+#define PKCS11_THREAD_LOCKING 1
perhaps we should remove the PKCS11_THREAD_LOCKING macro completely
+
sc_context_t *context = NULL;
struct sc_pkcs11_pool session_pool;
struct sc_pkcs11_slot virtual_slots[SC_PKCS11_MAX_VIRTUAL_SLOTS];
@@ -114,6 +117,7 @@
static CK_C_INITIALIZE_ARGS_PTR _locking;
static void * _lock = NULL;
#if (defined(HAVE_PTHREAD) || defined(_WIN32)) &&
defined(PKCS11_THREAD_LOCKING)
+#define HAVE_OS_LOCKING
what is this define good for ?
otherwise it looks good
Cheers,
Nils
_______________________________________________
opensc-devel mailing list
opensc-devel@lists.opensc-project.org
http://www.opensc-project.org/mailman/listinfo/opensc-devel
