Sorry for writing that many emails. Alternatively booting one kernel per cpu with ruby, say 4 cores in a 2 by 2 mesh booting 4 kernels, in FS mode would also serve my purpose. Does gem5 support this (with minor modifications)?
On Thu, Jun 19, 2014 at 9:46 AM, Severin Wischmann <[email protected] > wrote: > Hi all > > I'm trying to run multiple workloads of different size in SE mode. The > simulation gets terminated after the first thread returns an exit event. > In the documentation I could only find a reference to the maximum amount > of instructions that get executed as a user option, but exit() seems to > supersede this global maximum (which is infinity by default). > As far as I understand the code an exit event will asynchronously be > inserted into the event queues of all the other threads, which will cause > them to terminate as well. > > Is there a easy way to circumvent this or do I have to implement what is > described here: http://www.gem5.org/Multiprogrammed_workloads as option 3? > > Best, > Severin >
_______________________________________________ gem5-users mailing list [email protected] http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users
