Looks like a bug, if you can track it down that would be great. Probably something about the prefetcher not setting a thread_id on the request that something is trying to use or perhaps the reverse.
Ali On Oct 5, 2011, at 5:08 AM, Mahmood Naderan wrote: > Hi ali, > I found that whenever I use the following cache spec, I get that error > > class L2Cache(BaseCache): > size = '2MB' > latency = '10ns' > block_size = 64 > assoc = 8 > mshrs = 20 > tgts_per_mshr = 12 > prefetch_policy = 'tagged' > > If I remove the prefetch_policy, then there is no problem. > > Do you have any idea about that? > thanks > -- > // Naderan *Mahmood; > > > On Fri, Sep 16, 2011 at 9:29 AM, Mahmood Naderan <[email protected]> wrote: > (gdb) backtrace > #0 0x00007ffff63f9a75 in raise () from /lib/libc.so.6 > #1 0x00007ffff63fd5c0 in abort () from /lib/libc.so.6 > #2 0x00007ffff63f2941 in __assert_fail () from /lib/libc.so.6 > #3 0x000000000052f97c in Request::threadId (this=0x29a62e0, > pkt=<value optimized out>, > time=29033589792000) at build/ALPHA_FS/mem/request.hh:436 > #4 BasePrefetcher::notify (this=0x29a62e0, pkt=<value optimized out>, > time=29033589792000) at build/ALPHA_FS/mem/cache/prefetch/base.cc:239 > #5 0x00000000004fd534 in Cache<LRU>::timingAccess(Packet*) () > #6 0x00000000004fdc99 in Cache<LRU>::CpuSidePort::recvTiming(Packet*) () > #7 0x0000000000483675 in Port::sendTiming (this=0x2a05b90, pkt=0x3911f10) > at build/ALPHA_FS/mem/port.hh:186 > #8 Bus::recvTiming (this=0x2a05b90, pkt=0x3911f10) at > build/ALPHA_FS/mem/bus.cc:265 > #9 0x00000000004fbbb2 in Cache<LRU>::MemSidePort::sendPacket() () > #10 0x000000000048ccc5 in Port::sendRetry (this=0x2a05b90, id=28060) > at build/ALPHA_FS/mem/port.hh:212 > #11 Bus::recvRetry (this=0x2a05b90, id=28060) at build/ALPHA_FS/mem/bus.cc:305 > #12 0x0000000000663844 in EventQueue::serviceOne (this=<value optimized out>) > at build/ALPHA_FS/sim/eventq.cc:204 > #13 0x00000000006ad272 in simulate (num_cycles=9223372036854775807) > at build/ALPHA_FS/sim/simulate.cc:73 > #14 0x000000000043994d in _wrap_simulate__SWIG_0 (self=<value optimized out>, > args=0x1de5310) at build/ALPHA_FS/python/swig/event_wrap.cc:4534 > #15 _wrap_simulate (self=<value optimized out>, args=0x1de5310) > at build/ALPHA_FS/python/swig/event_wrap.cc:4584 > #16 0x00007ffff73fe313 in PyEval_EvalFrameEx () from > /usr/lib/libpython2.6.so.1.0 > #17 0x00007ffff73ffd60 in PyEval_EvalCodeEx () from > /usr/lib/libpython2.6.so.1.0 > #18 0x00007ffff73fe0ab in PyEval_EvalFrameEx () from > /usr/lib/libpython2.6.so.1.0 > #19 0x00007ffff73fe928 in PyEval_EvalFrameEx () from > /usr/lib/libpython2.6.so.1.0 > #20 0x00007ffff73ffd60 in PyEval_EvalCodeEx () from > /usr/lib/libpython2.6.so.1.0 > > > On 9/16/11, Ali Saidi <[email protected]> wrote: > > You should run gem5 in gdb, not attach to the simulated guest. e.g. gdb > > --args build/ALPHA_FS/gem5.opt configs/example/fs.py ...... > > > > Ali > > > > On Sep 15, 2011, at 11:47 PM, Mahmood Naderan wrote: > > > >> I can not attach gdb because the error appears quickly. I see this message > >> 0: system.remote_gdb.listener: listening for remote gdb #0 on port 7000 > >> > >> but when I want to attach from another terminal, soon I get error > >> Switch at curTick count:10000 > >> info: Entering event queue @ 29033580035000. Starting simulation... > >> Switched CPUS @ tick 29033580045000 > >> Changing memory mode to timing > >> switching cpus > >> **** REAL SIMULATION **** > >> info: Entering event queue @ 29033580045000. Starting simulation... > >> m5.opt: build/ALPHA_FS/mem/request.hh:436: int Request::threadId() > >> const: Assertion `privateFlags.isSet(VALID_THREAD_ID)' failed. > >> Program aborted at cycle 29033589782000 > >> Aborted > >> > >> > >> One thing that I found is that if I remove "--l2cache", there is no > >> problem with resuming the checkpoint. > >> > >> On 9/15/11, Ali Saidi <[email protected]> wrote: > >>> You'll have to run gdb and see what you can find. > >>> Ali > >>> > >>> > >>> On Thu, 15 Sep 2011 18:38:30 +0430, Mahmood Naderan > >>> <[email protected]> wrote: > >>>> Hi > >>>> I want to resume from a checkpoint in ALPHA_FS mode with this command > >>>> > >>>> build/ALPHA_FS/m5.opt configs/example/fs.py --detailed --caches > >>>> --l2cache -r 1 > >>>> > >>>> but I get this error: > >>>> info: Entering event queue @ 29033580045000. Starting simulation... > >>>> m5.opt: build/ALPHA_FS/mem/request.hh:436: int Request::threadId() > >>>> const: Assertion `privateFlags.isSet(VALID_THREAD_ID)' failed. > >>>> Program aborted at cycle 29033589782000 > >>>> Aborted > >>>> > >>>> Where is the fault? thanks > >>>> _______________________________________________ > >>>> gem5-users mailing list > >>>> [email protected] > >>>> http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users > >>> > >>> _______________________________________________ > >>> gem5-users mailing list > >>> [email protected] > >>> http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users > >>> > >> > > > > >
_______________________________________________ gem5-users mailing list [email protected] http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users
