A simple, but not 100% accurate approach, would be to consider as stack
addr the highest virtual addressess, because usually stack addr are much
higher than heap addr.
Print the addreses of memory located in the heap and stack and you will
understand my point.

--
Eduardo Henrique Molina da Cruz
PhD student
Parallel and Distributed Processing Group
Federal University of Rio Grande do Sul (UFRGS)
Em 31/10/2013 18:38, "Curtis Dunham" <[email protected]> escreveu:

> Correct.  Although you could build heuristics into MARSS.  From each
> thread's perspective, its stack is most likely the memory range in
> /proc/*/maps that esp/rsp points into.
>
> On Thu, Oct 31, 2013 at 06:26:13PM -0200, Eduardo Cruz wrote:
> > I dont think MARSS has knowledge whether a memory access is in a heap or
> > stack addr, sice this is a abstraction provided by the language/os.
> > But you can instrument your application to provide MARSS that
> information.
> > Em 31/10/2013 18:12, "Qiuyun Wang" <[email protected]> escreveu:
> >
> > > Hello all,
> > >
> > > I am trying to separate stack and heap memory access from other memory
> > > access in Marss and DRAMSim for some benchmarks.
> > >
> > > Do anyone know is there a way to trace those information in ptlsim or
> kvm?
> > > I am not sure which part of the code should I look at.
> > >
> > > Thanks in advance for your reply.
> > >
> > > Best,
> > > --Qiuyun
> > >
> > >
> > > --
> > > ***********************
> > > Qiuyun Wang
> > > Hudson 213,
> > > Electrical and Computer Engineering Department,
> > > Duke University
> > > M.S. University Paris-Sud XI
> > >
> > > _______________________________________________
> > > http://www.marss86.org
> > > Marss86-Devel mailing list
> > > [email protected]
> > > https://www.cs.binghamton.edu/mailman/listinfo/marss86-devel
> > >
> > >
>
> > _______________________________________________
> > http://www.marss86.org
> > Marss86-Devel mailing list
> > [email protected]
> > https://www.cs.binghamton.edu/mailman/listinfo/marss86-devel
>
>
_______________________________________________
http://www.marss86.org
Marss86-Devel mailing list
[email protected]
https://www.cs.binghamton.edu/mailman/listinfo/marss86-devel

Reply via email to