Re: [ns] Timestamp trouble

["Marcelo Lipas Augusto" <lipas@lcs.poli.usp.br>]

Hi, Felix.

I tried to use the already existent UDP timestamp to do the same job you
want. Strangely enough, it seems all generated packtets have a zero
timestamp (what obviously won't work). I guess if someone knows how to fix
this problem, you won't need to write any code, you will have only to use
the existing code.
**Anyone know what's wrong with the current UDP timestamp??**

Thank you!

--Lipas


----- Original Message -----
From: "Andrea Josi" <mr_josi@yahoo.it>
To: <felix.ko@mchp.siemens.de>
Cc: <ns-users@ISI.EDU>
Sent: Wednesday, August 29, 2001 6:58 AM
Subject: [ns] Timestamp trouble


> Dear Felix Ko,
>
> i am encountering the same large simulation memory and
> time problems you quoted in
>
> http://www.isi.edu/nsnam/archive/ns-users/webarch/current/msg01836.html
>
> I need to calculate the packets' end to end delay of a
> multicast traffic. I have tried to apply the same
> solution you suggested in that
> mail,  but i am encountering this problem:
>
> i have added a field and a corresponding method in the
> common header, "my_ts_"  and  "my_timestamp()"
>
> struct hdr_cmn {
> ..............
> .............
>  double ts_;  // timestamp: for q-delay measurement
>  double my_ts_;   <---   this is mine
> .....
> ............
> inline double& timestamp() { return (ts_); }
> inline double& my_timestamp() { return (my_ts_); }
> <-- this is mine
>
> I've also added a line in udp.cc in the places where
> the original timestamp is used
>
> hdr_cmn::access(p)->timestamp() =
>       (u_int32_t)(SAMPLERATE*local_time);
>
> hdr_cmn::access(p)->my_timestamp() =
> local_time;<--this is mine
>
> I have implemented a receiver similar to
> Agent/LossMonitor; in the recv function i have put
> these lines
>
> hdr_rtp* p = hdr_rtp::access(pkt);
> start_time_ = hdr_cmn::access(pkt)->my_timestamp();
>
> The simulation reports an error, which should concern
> multicast forwarding; if i comment the lines i wrote
> before, i get no error !
> It seems like udp timestamp is used to identify
> something ...?!!!
> What was your solution ? (i don't really care to make
> mine work, what i really need is a way to get the
> packets departure times!)
> Thanks for your help,
>
>                 Andrea Josi.
>
>
> ns: _o29 new-group 2 -2147483648 0 wrong-iif: can't
> read "protocols_(0)": no such element in array
>     while executing
> "$protocols_($iface) upcall $code $source $group
> $iface"
>     (procedure "_o30" line 9)
>     (mrtObject upcall line 9)
>     invoked from within
> "$mrtObject_ upcall $code $src $group $iface"
>     (procedure "_o24" line 3)
>     (Node new-group line 3)
>     invoked from within
> "$node_ new-group $src $group $iface $code"
>     (procedure "_o29" line 3)
>     (Classifier/Multicast new-group line 3)
>     invoked from within
> "_o29 new-group 2 -2147483648 0 wrong-iif"
>
> ______________________________________________________________________
> Do You Yahoo!?
> Il tuo indirizzo gratis e per sempre @yahoo.it su http://mail.yahoo.it
>
>