One thing that might be happening is that the program isn't working
correctly, possibly because some syscalls are missing, and it's
allocating memory or recursing or otherwise allocating memory long after
it was supposed to stop. On a real machine if a single process uses more
than 2 gigs of memory that's likely, though not necessarily, incorrect
behavior. I'm not sure how off the top of my head, but if you can I
would suggest running this or an x86 equivalent somewhere and checking
what it's actual memory usage is. It won't be exact, but it can tell you
if 2 gigs is reasonable.
One other thing I thought of just now is that SE's page management might
be a little too simple to handle lots of page deallocation and
reallocation with, for instance, a stream of mmaps and munmaps. I'm just
guessing here so I wouldn't spend too much time looking at that unless
you can't think of what else it might be.
Gabe
Meng-Ju Wu wrote:
> Ali, thank you very much.
>
> (1) I modified the src/arch/alpha/linux/process.cc.
>
> SyscallDesc("ftruncate", unimplementedFunc) =>
> SyscallDesc("ftruncate", ftruncateFunc)
>
> Now, the ftruncate() works.
>
> I want to implement other system calls. What the files should I look at?
>
> (2) The memory system configuration is
>
> system = System(cpu = [CPUClass(cpu_id=i) for i in xrange(np)],
> physmem = PhysicalMemory(range=AddrRange("2048MB")),
> membus = Bus(), mem_mode = 'timing')
>
> This is the largest number I can set.
>
> There are 9 spec2006 benchmarks has the problem
>
> " terminate called after throwing an instance of 'std::bad_alloc' "
>
> I tried to use a smaller dataset in 410.bwaves. It finished
> correctly without any problem.
>
> What should I do to avoid this problem or correct it without
> changing the dataset?
>
> Thanks again.
>
> Meng-Ju
> _______________________________________________
> 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
