Sunitha, You must have broken the coherence protocol in some way with you're change. Comparing a Exec trace before and after your change may give you some idea, depending on if the mp workload diverges or not. Otherwise you're either going to have to create a small program that tests your code and triggers a bug or back-trace from where, "DECthreads bugcheck (version V3.18-037), terminating execution." is printed to find what is triggering that assert and figure out what is going wrong with your change. I don't really know what you're doing but it's likely you're invaliding a block that was never written back to memory or something similar so there is invalid data in the system.
Ali On Feb 27, 2011, at 12:36 PM, sunitha p wrote: > I am invalidating a victim block taken for replacement in L2 in L1. > > Have added a new command Requestinvalidate with - IsRead, IsRequest, > Hasdata, Is Invalidate ---- as attributes... > > As soon as a victim block of L2 is found, a new packet is created with the > RequestInvalidate command and broadcasted along systemtoL2 bus > throughcpusideport.... > > Its working fine with invalidation.....and resumes the work to be done on > victim block... > > but after some ticks am getting the error as > > 0: system.remote_gdb.listener: listening for remote gdb #0 on port 7000 > 0: system.remote_gdb.listener: listening for remote gdb #1 on port 7001 > %DECthreads bugcheck (version V3.18-037), terminating execution. > % Reason: muInit: can't preallocate mutex 119 of 200 > % Running on (732) on COMPAQ Professional Workstation , 1024Mb; 2 CPUs, pid > 100 > warn: ignoring syscall sigaction(4831387048, 6, ...) > For more information see: http://www.m5sim.org/warn/5c5b547f > warn: ignoring syscall sigprocmask(32, 2, ...) > For more information see: http://www.m5sim.org/warn/5c5b547f > fatal: syscall nxm_thread_kill (#-38) unimplemented. > @ cycle 648971000 > [unimplementedFunc:build/ALPHA_SE/sim/syscall_emul.cc, line 82] > Memory Usage: 157824 KBytes > For more information see: http://www.m5sim.org/fatal/be2b42f > > > can anyone help me out as to why and what is this error for...? > > > Thanks & regards > > -- > Sunitha.P > 9092892876 > > _______________________________________________ > m5-users mailing list > [email protected] > http://m5sim.org/cgi-bin/mailman/listinfo/m5-users
_______________________________________________ m5-users mailing list [email protected] http://m5sim.org/cgi-bin/mailman/listinfo/m5-users
