I just came across a discussion from last year where Paul had similar
issues running multi-core configuration with the Atom core model:

http://comments.gmane.org/gmane.comp.emulators.marss86/493

However, in my case, I am able to run multi-core configuration, but
not SMT. Is this a known issue ? I get a segafault as soon as the
simulation mode is entered, even without any workload (just
./start_sim) in this config.

On Sun, Oct 7, 2012 at 11:11 PM, Ankita (Garg) Goel
<[email protected]> wrote:
> Hi,
>
> I'm getting Segmentation Fault when simulating the Atom core model
> with SMT enabled. My machine config is on the lines of ooo_2_th
> machine config :
>
>   atom_2_th:
>     description: Atom core with 2 threads
>     min_contexts: 2
>     cores:
>       - type: atom
>         name_prefix: atom_
>         option:
>             threads: 2
>     caches:
>       - type: l1_32K_mesi
>         name_prefix: L1_I_
>         insts: $NUMCORES # Per core L1-I cache
>         option:
>             private: true
>       - type: l1_32K_mesi
>         name_prefix: L1_D_
>         insts: $NUMCORES # Per core L1-D cache
>         option:
>             private: true
>       - type: l2_512K
>         name_prefix: L2_
>         insts: $NUMCORES # Private L2 config
>         option:
>             private: true
>             last_private: true
>     memory:
>       - type: dram_cont
>         name_prefix: MEM_
>         insts: 1 # Single DRAM controller
>     interconnects:
>       - type: p2p
>         connections:
>           - core_$: I
>             L1_I_$: UPPER
>           - core_$: D
>             L1_D_$: UPPER
>           - L1_I_$: LOWER
>             L2_$: UPPER
>           - L1_D_$: LOWER
>             L2_$: UPPER2
>       - type: split_bus
>         connections:
>           - L2_*: LOWER
>             MEM_0: UPPER
>
> The bt from gdb is as follows:
>
> #0  0x00000000005a91ff in
> Memory::CoherentCache::MESILogic::complete_request (this=<optimized
> out>, queueEntry=0x2cb7498,
>     message=...) at ptlsim/build/cache/mesiLogic.cpp:305
> #1  0x000000000058f946 in
> Memory::CoherentCache::CacheController::complete_request
> (this=0x2cb7230, message=...,
>     queueEntry=0x2cb7498) at ptlsim/build/cache/coherentCache.cpp:389
> #2  0x000000000059135c in
> Memory::CoherentCache::CacheController::handle_lower_interconnect
> (this=0x2cb7230, message=...)
>     at ptlsim/build/cache/coherentCache.cpp:233
> #3  0x00000000005915ab in
> Memory::CoherentCache::CacheController::handle_interconnect_cb
> (this=0x2cb7230, arg=0x2b2cd80)
>     at ptlsim/build/cache/coherentCache.cpp:415
> #4  0x0000000000586ca8 in
> superstl::TFunctor1<Memory::Controller>::operator() (this=<optimized
> out>, arg=<optimized out>)
>     at ptlsim/lib/superstl.h:3950
> #5  0x000000000062768b in superstl::Signal::emit (this=<optimized
> out>, arg=<optimized out>) at ptlsim/build/lib/superstl.cpp:1431
> #6  0x00000000005ad984 in
> Memory::P2PInterconnect::controller_request_cb (this=0x2d15000,
> arg=0x2b2cd30)
>     at ptlsim/build/cache/p2p.cpp:85
> #7  0x00000000005817ce in
> superstl::TFunctor1<Memory::Interconnect>::operator() (this=<optimized
> out>, arg=<optimized out>)
>     at ptlsim/lib/superstl.h:3950
> #8  0x000000000062768b in superstl::Signal::emit (this=<optimized
> out>, arg=<optimized out>) at ptlsim/build/lib/superstl.cpp:1431
> #9  0x00000000005834c6 in
> Memory::CacheController::wait_interconnect_cb (this=0x2cd6f80,
> arg=0x2cd71e8)
>     at ptlsim/build/cache/cacheController.cpp:694
> #10 0x0000000000586c78 in
> superstl::TFunctor1<Memory::CacheController>::operator()
> (this=<optimized out>, arg=<optimized out>)
>     at ptlsim/lib/superstl.h:3950
> #11 0x000000000062768b in superstl::Signal::emit (this=<optimized
> out>, arg=<optimized out>) at ptlsim/build/lib/superstl.cpp:1431
> #12 0x00000000005a545b in execute (this=0x2b308a8) at
> ptlsim/cache/memoryHierarchy.h:110
> #13 Memory::MemoryHierarchy::add_event (this=0x2b2c9f0,
> signal=<optimized out>, delay=<optimized out>, arg=<optimized out>)
>     at ptlsim/build/cache/memoryHierarchy.cpp:312
> #14 0x00000000006a5f7c in marss_add_event (signal=<optimized out>,
> delay=<optimized out>, arg=<optimized out>)
>     at ptlsim/build/sim/machine.cpp:816
> #15 0x0000000000586050 in
> Memory::CacheController::handle_interconnect_cb (this=0x2cd6f80,
> arg=0x2b2cce0)
>     at ptlsim/build/cache/cacheController.cpp:294
> #16 0x0000000000586ca8 in
> superstl::TFunctor1<Memory::Controller>::operator() (this=<optimized
> out>, arg=<optimized out>)
>     at ptlsim/lib/superstl.h:3950
> #17 0x000000000062768b in superstl::Signal::emit (this=<optimized
> out>, arg=<optimized out>) at ptlsim/build/lib/superstl.cpp:1431
> #18 0x00000000005b080e in
> Memory::SplitPhaseBus::BusInterconnect::data_broadcast_completed_cb
> (this=0x2d15520, arg=0x2d15788)
>     at ptlsim/build/cache/splitPhaseBus.cpp:509
> #19 0x00000000005b3676 in
> superstl::TFunctor1<Memory::SplitPhaseBus::BusInterconnect>::operator()
> (this=<optimized out>,
>     arg=<optimized out>) at ptlsim/lib/superstl.h:3950
> #20 0x000000000062768b in superstl::Signal::emit (this=<optimized
> out>, arg=<optimized out>) at ptlsim/build/lib/superstl.cpp:1431
> ---Type <return> to continue, or q <return> to quit---
> #21 0x00000000005a46d3 in execute (this=0x2b306b0) at
> ptlsim/cache/memoryHierarchy.h:110
> #22 Memory::MemoryHierarchy::clock (this=0x2b2c9f0) at
> ptlsim/build/cache/memoryHierarchy.cpp:106
> #23 0x00000000006a2673 in BaseMachine::run (this=0x129a180,
> config=...) at ptlsim/build/sim/machine.cpp:356
> #24 0x00000000006b1ba3 in ptl_simulate () at ptlsim/build/sim/ptlsim.cpp:1398
> #25 0x00000000005792d2 in sim_cpu_exec () at qemu/cpu-exec.c:310
> #26 0x000000000041e079 in main_loop () at qemu/vl.c:1450
> #27 main (argc=8, argv=0x7fffffffe188, envp=<optimized out>) at qemu/vl.c:3189
>
>
> Any ideas on what could be going wrong ? Is this a bug or some issue
> with my machine configuration ? Will greatly appreciate your help on
> debugging this! Please let me know if I should provide more debug
> information.
>
> Thanks for your help!
>
> --
> Regards,
> Ankita



-- 
Regards,
Ankita
Graduate Student
Department of Computer Science
University of Texas at Austin

_______________________________________________
http://www.marss86.org
Marss86-Devel mailing list
[email protected]
https://www.cs.binghamton.edu/mailman/listinfo/marss86-devel

Reply via email to