Mike, The $srandom functionality was folded into the process class. Thus, the examples showing process::self.srandom() are correct; that is the correct mechanism provided for seeding a process. The system task $srandom was removed from the standard. Nonetheless, section 11.9 seems to be missing the prototype of the srandom method. This appears to have been an editorial mistake. The correct prototype of the process class should be: class process; enum state { FINISHED, RUNNING, static function process self(); function state status(); function void kill(); task await(); function void suspend(); task resume(); function void srandom( int seed ); endclass Thanks for catching this omission. Arturo -----Original Message----- From: owner-sv-ec@eda.org [mailto:owner-sv-ec@eda.org] On Behalf Of Michael Burns Sent: Wednesday, February 22, 2006 3:14 PM To: sv-ec@eda.org Cc: Architecture Testbench-TBARDS Subject: [sv-ec] srandom for thread seeding Hi folks, I've been comparing the Accelera SV3.1a standard and IEEE 1800, and it's not clear what happened to srandom(). 3.1a provides both $srandom and process::self().srandom() for seeding a thread. IEEE 1800 appears to have removed $srandom(). process::self().srandom() is referred to in a few places in the 1800 text, but appears to have been removed from section 11.9 defining the process built-in class. What's the status of thread seeding in 1800? Thanks, Mike Burns FreescaleReceived on Wed Feb 22 23:12:04 2006
This archive was generated by hypermail 2.1.8 : Wed Feb 22 2006 - 23:13:55 PST