I think the question Ali was asking is where in the python is curTick() being called before simulate(). On the C++ side, any call from python is going to come through core_wrap.cc. You might have to use the python debugger and put a breakpoint on the python side of the swig-generated code to figure this out.
Steve On Tue, Jan 22, 2013 at 1:25 PM, Nilay <[email protected]> wrote: > Ali, curTick() is being called from the file python/swig/core_wrap.cc. I > am running the following command: > ./build/X86/gem5.debug ./configs/example/ruby_network_test.py --help > > When I try to run the network test itself, i.e. with out the help option, > I get a segmentation fault, again because of curTick(). This time a ruby > object tries to query for time from its init() function. > > Note that the main event queues are being dynamically allocated when the > simulate() function in src/sim/simulate.cc is called. It seems to me that > the queues need to be created earlier than this, may be before the > SimObjects have been created. > > Any ideas as to where this should be done? > > -- > Nilay > > On Tue, January 22, 2013 2:06 pm, Ali Saidi wrote: > > > > > > Is something calling the curTick() method or are you saying that by > > virtue of wrapping curTick() it's being called somehow. I'm not sure how > > the latter could happen. In the case of the prior, any idea what's > > calling it? > > > > Ali > > > > On 22.01.2013 14:56, Nilay wrote: > > > >> I am trying to > > allocate the main event queue(s) dynamically, but I am > >> running into a > > problem related to curTick(). In the file > >> src/python/swig/core.i, the > > function curTick() has been declared. This > >> leads SWIG to generate code > > that calls curTick(). It seems this function > >> call happens before the > > main event queue been allocated. Since the > >> function curTick() gets the > > tick value from the main event queue's tick > >> value, the call results in > > a segmentation fault. > >> > >> Can any one suggest some way out? > >> > >> -- > >> > > Nilay > >> > >> _______________________________________________ > >> gem5-dev > > mailing list > >> [email protected] > >> > > _______________________________________________ > gem5-dev mailing list > [email protected] > http://m5sim.org/mailman/listinfo/gem5-dev > _______________________________________________ gem5-dev mailing list [email protected] http://m5sim.org/mailman/listinfo/gem5-dev
