[gem5-users] Re: Muti-thread workload bug in new version

[Jason Lowe-Power via gem5-users]

Hi Taiyu,

This seems like a different bug than
https://gem5.atlassian.net/browse/GEM5-332. We should probably open a new
ticket on Jira.

I'm a bit surprised our threads tests didn't catch this. We'll look into
it. @Hoa Nguyen <hoangu...@ucdavis.edu>, can you create a ticket on Jira
and see if you can reproduce this?

Cheers,
Jason

On Sun, Jun 21, 2020 at 5:58 AM Taiyu Zhou via gem5-users <
gem5-users@gem5.org> wrote:

> Hi guys!
> My gem5 version is 96fce476785a834f102ae69a895e661cf08e47cd which clone
> from GitHub.
> I am trying to run a muti-thread program in se mode. But  I encountered a
> bug.
> My program is simple:
> "
> void f1()
> {
>         char a[64];
>         printf("hello\n");
> }
>
> int main()
> {
>     std::thread threads[2]; // t1 is not a thread
>     for(int i=0;i<2;i++)
>         threads[i] = std::thread(f1);
>     for (auto& t: threads) {
>         t.join();
>     }
>
> }
>
> “
>
> My run cmd is
> ./build/X86/gem5.opt configs/example/se.py -c
> /home/ubuntu/taiyu/test_app/thread_t --mem-size=8GB --cpu-type=DerivO3CPU
> --caches --l2cache -n 3”
>
> And the gem5 will report :
> warn: ignoring syscall set_robust_list(...)
> warn: ignoring syscall mprotect(...)
> hello
> hello
> panic: panic condition !clobber occurred: EmulationPageTable::allocate:
> addr 0x7ffff7471000 already mapped
> Memory Usage: 8642200 KBytes
> Program aborted at tick 1411052000
> --- BEGIN LIBC BACKTRACE ---
> ./build/X86/gem5.opt(_Z15print_backtracev+0x28)[0xa977b8]
> ./build/X86/gem5.opt(_Z12abortHandleri+0x46)[0xaa8ab6]
> /lib/x86_64-linux-gnu/libpthread.so.0(+0x11390)[0x7f46d3bad390]
> /lib/x86_64-linux-gnu/libc.so.6(gsignal+0x38)[0x7f46d2552428]
> /lib/x86_64-linux-gnu/libc.so.6(abort+0x16a)[0x7f46d255402a]
> ./build/X86/gem5.opt[0x807e0f]
> ./build/X86/gem5.opt(_ZN18EmulationPageTable3mapEmmlm+0x93c)[0xa230cc]
> ./build/X86/gem5.opt(_ZN8MemState10fixupFaultEm+0xd8)[0xaf1518]
>
> ./build/X86/gem5.opt(_ZN6X86ISA9PageFault6invokeEP13ThreadContextRK14RefCountingPtrI10StaticInstE+0xf8)[0x11d4a58]
>
> ./build/X86/gem5.opt(_ZN13DefaultCommitI9O3CPUImplE10commitHeadERK14RefCountingPtrI13BaseO3DynInstIS0_EEj+0x9a7)[0xc87ba7]
>
> ./build/X86/gem5.opt(_ZN13DefaultCommitI9O3CPUImplE11commitInstsEv+0x637)[0xc88907]
>
> ./build/X86/gem5.opt(_ZN13DefaultCommitI9O3CPUImplE6commitEv+0xc70)[0xc8b070]
> ./build/X86/gem5.opt(_ZN13DefaultCommitI9O3CPUImplE4tickEv+0xc8)[0xc8bd88]
> ./build/X86/gem5.opt(_ZN9FullO3CPUI9O3CPUImplE4tickEv+0x155)[0xc9d935]
> ./build/X86/gem5.opt(_ZN10EventQueue10serviceOneEv+0x9d)[0xa9f84d]
> ./build/X86/gem5.opt(_Z9doSimLoopP10EventQueue+0x7b)[0xac1a6b]
> ./build/X86/gem5.opt(_Z8simulatem+0xc2a)[0xac2a0a]
> ./build/X86/gem5.opt[0x17f8671]
> ./build/X86/gem5.opt[0xb14faa]
>
> /usr/lib/x86_64-linux-gnu/libpython2.7.so.1.0(PyEval_EvalFrameEx+0x7852)[0x7f46d3e6a7b2]
>
> /usr/lib/x86_64-linux-gnu/libpython2.7.so.1.0(PyEval_EvalCodeEx+0x85c)[0x7f46d3fa111c]
>
> /usr/lib/x86_64-linux-gnu/libpython2.7.so.1.0(PyEval_EvalFrameEx+0x6ffd)[0x7f46d3e69f5d]
>
> /usr/lib/x86_64-linux-gnu/libpython2.7.so.1.0(PyEval_EvalCodeEx+0x85c)[0x7f46d3fa111c]
>
> /usr/lib/x86_64-linux-gnu/libpython2.7.so.1.0(PyEval_EvalFrameEx+0x6ffd)[0x7f46d3e69f5d]
>
> /usr/lib/x86_64-linux-gnu/libpython2.7.so.1.0(PyEval_EvalCodeEx+0x85c)[0x7f46d3fa111c]
>
> /usr/lib/x86_64-linux-gnu/libpython2.7.so.1.0(PyEval_EvalFrameEx+0x6ffd)[0x7f46d3e69f5d]
>
> /usr/lib/x86_64-linux-gnu/libpython2.7.so.1.0(PyEval_EvalCodeEx+0x85c)[0x7f46d3fa111c]
>
> /usr/lib/x86_64-linux-gnu/libpython2.7.so.1.0(PyEval_EvalCode+0x19)[0x7f46d3e62de9]
>
> /usr/lib/x86_64-linux-gnu/libpython2.7.so.1.0(PyEval_EvalFrameEx+0x613b)[0x7f46d3e6909b]
>
> /usr/lib/x86_64-linux-gnu/libpython2.7.so.1.0(PyEval_EvalCodeEx+0x85c)[0x7f46d3fa111c]
>
> /usr/lib/x86_64-linux-gnu/libpython2.7.so.1.0(PyEval_EvalFrameEx+0x6ffd)[0x7f46d3e69f5d]
>
> /usr/lib/x86_64-linux-gnu/libpython2.7.so.1.0(PyEval_EvalCodeEx+0x85c)[0x7f46d3fa111c]
> --- END LIBC BACKTRACE —
>
>    This problem will not occur in an older version which I clone from
> GitHub in March. It is work in the older version.
>
> _______________________________________________
> gem5-users mailing list -- gem5-users@gem5.org
> To unsubscribe send an email to gem5-users-le...@gem5.org
> %(web_page_url)slistinfo%(cgiext)s/%(_internal_name)s

_______________________________________________
gem5-users mailing list -- gem5-users@gem5.org
To unsubscribe send an email to gem5-users-le...@gem5.org
%(web_page_url)slistinfo%(cgiext)s/%(_internal_name)s