I th On Jul 8, 2013, at 2:21 PM, Mona Jalal <[email protected]> wrote: > Hi, > I am using hooks and have modified them in a way to just run the ROI in code > for X86. But unfortunately m5_exit(0) in the end of ROI_end in hooks.c > doesn't work though I have provided its assembly code and continues to run > to the end of the benchmark. I was wondering if you might be aware of any > solution to this: > x86_m5.S http://paste.ubuntu.com/5856273/ > hooks.c http://paste.ubuntu.com/5856278/ > part of the benchmark which includes the ROI: backprop_kernel.c > http://paste.ubuntu.com/5856280/ > > > Here's my system.terminal which shows that simulation continued even after > the m5_exit(0) put in the end of the rodinia_roi_end in hooks.c: > > after m5_checkpoint(0,0) > [HOOKS] Entering ROI > [HOOKS] Leaving ROI > Training done > [HOOKS] Terminating ... > Simulation is successfully finished! > > Also what are the alternatives to using m5_exit(0) in the end of ROI_end() > function in hooks.c ? > > Best regards, > Mona Jalal. > > > > _______________________________________________ > gem5-users mailing list > [email protected] > http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users >
I think that is the best way, and it certainly should work. I think parsec compiles the hooks as a library, are you sure you got that updated on the disk image? Does the m5_checkpoint() call work? Ali _______________________________________________ gem5-users mailing list [email protected] http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users
