Thanks for the Info, Steve.

regards,
Soumyaroop.

On 7/16/09, Steve Reinhardt <[email protected]> wrote:
>
> I generally use cscope to find things in the source.  There's a script
> to build a cscope index in util/cscope-index.py.
>
>
> Steve
>
>
> On Thu, Jul 16, 2009 at 10:28 AM, soumyaroop roy<[email protected]> wrote:
> > Hello Devraj,
> >
> > contextId() is defined in request.hh
> >
> > s...@theoracle:~/work/m5-stable$ grep "contextId()" `find ./src/mem
> -name
> > "*.hh"
> > ./src/mem/cache/blk.hh:            return (contextId ==
> req->contextId());
> > ./src/mem/cache/blk.hh:            : contextId(req->contextId())
> > ./src/mem/request.hh:    contextId() const
> > ./src/mem/physical.hh:            return (contextId == req->contextId());
> > ./src/mem/physical.hh:              contextId(req->contextId())
> >
> > regards,
> > Soumyaroop.
> >
> > On 7/15/09, Devraj Chapagain <[email protected]> wrote:
> >>
> >> Hi Steve,
> >> Thanks lot for your reply.
> >> Probably, i have not updates all the latest one. When i search
> >> the contextId() in my M5 (within  /m5-2.0b6/src/mem/ ), I could not get
> it.
> >>  Is it already defined and assigned by M5 developers or i have to
> myself?
> >> I'm really sorry for asking such minor things again and again.
> >>
> >>
> >>
> >> On Wed, Jul 15, 2009 at 6:20 AM, Steve Reinhardt <[email protected]>
> wrote:
> >>>
> >>>
> >>> On Tue, Jul 14, 2009 at 11:23 PM, Devraj Chapagain
> >>> <[email protected]> wrote:
> >>>>
> >>>> Hi,
> >>>> I am running a multi core multi-thread program on M5 under ALPHA-SE
> mode
> >>>> in SPEC CPU 2006. I can not get context id when tried to trace the
> >>>> parameters of packet. Is core id is context id or not? If not, how can
> we
> >>>> get the context id?
> >>>
> >>> The context ID is in req->contextId().
> >>>
> >>>>
> >>>> The next problem is that when I dump the content of the block in
> packet,
> >>>> in some tick times it has no data value in it.
> >>>> Is it possible having no value in some tick values? if not, what can
> be
> >>>> the error for this?
> >>>
> >>> Some packets naturally do not have data (e.g., read requests as opposed
> >>> to read responses).
> >>>
> >>>>
> >>>> The last problem is when i use isInstFetch() to distinguish the
> content
> >>>> of packet as whether it is data or instruction, it always return 0
> value.
> >>>> So, i can't distinguish whether the fetch content is data or
> instruction.
> >>>
> >>> Are you using the latest version of the code?  There was a bug fix
> within
> >>> the past few months on this.  If you're still having problems, what CPU
> >>> model are you using?  Are you tracing from the very beginning of the
> >>> program?
> >>>
> >>>>
> >>>> Could anyone please provide me hints to figure out these problems.
> >>>>
> >>>>
> >>>> Thanks in advance,
> >>>> devraj
> >>>> _______________________________________________
> >>>> 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
> >>
> >>
> >> _______________________________________________
> >> m5-users mailing list
> >> [email protected]
> >> http://m5sim.org/cgi-bin/mailman/listinfo/m5-users
> >
> >
> >
> > --
> > Soumyaroop Roy
> > Ph.D. Candidate
> > Department of Computer Science and Engineering
> > University of South Florida, Tampa
> > http://www.csee.usf.edu/~sroy
> > _______________________________________________
> > 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
>



-- 
Soumyaroop Roy
Ph.D. Candidate
Department of Computer Science and Engineering
University of South Florida, Tampa
http://www.csee.usf.edu/~sroy
_______________________________________________
m5-users mailing list
[email protected]
http://m5sim.org/cgi-bin/mailman/listinfo/m5-users

Reply via email to