Hello, Well, I have fixed a bunch of problems with development release 0.6.0. The new release will be 0.6.2 I have cleaned up a few things such as the following: 1) Only allocate enough shared memory to reflect the number of readers - this will change a bit when we start supporting USB readers so that they can be hot pluggable and the resource manager will recognize this. 2) Pthreads replace the old and ugly fork() that was used to receive events. This is a polling mechanism, but it can be lightweight. It calls the IFDStatus function which calls the IFD_Is_Icc_Present function in the driver. If your reader supports interrupts then this would be as simple as either returning PRESENT or ABSENT instead of sending a command to the reader each time this is called. 3) Pthread Mutexes replace my old 'test and set' algorithm for mutual exclusion. This keeps 1 application from infinite looping the resource manager and causing starvation to other threads. 4) Function pointers to dynamic loaded libraries will be mapped at startup instead of at runtime. Before, each time a function was called it mapped the function pointer to the library. Now it will be done once and stored in a structure. The readers in /etc/pcsc.conf instead of /etc/reader.conf are only the ones that are active on the system. Before you could specify more readers even if you weren't using them. The resource manager will test these readers on startup and either exit if there is a problem or continue. In 0.6.0 I noticed that a few memory corruptions would occur on a not so predictable basis. These changes/fixes seem to take care of the problem and also make the resource manager alot more lightweight. The following might not be in the next release but it will be soon. Depending on your settings in /etc/pcsc.conf the power on the card will suspend after a certain time of no connections and the port will be freed after a certain timeout of no connection If you choose this may never happen and the power and port will always be allocated. On another connection the port and power will be reallocated until disconnect + timeout Let me know if you have any ideas .... Best Regards, Dave *************************************************************** Linux Smart Card Developers - M.U.S.C.L.E. (Movement for the Use of Smart Cards in a Linux Environment) http://www.linuxnet.com/smartcard/index.html ***************************************************************
