Re: gdb with root

From: Arthur E. Snyder <snyder_at_slac.stanford.edu>
Date: Wed, 12 Jul 2006 15:47:19 -0700 (PDT)


Thanks Rene. The trick is to load shared library before attaching gdb.

-Art S.

A.E. Snyder, Group EC                        \!c*p?/
SLAC Mail Stop #95                          ((.   .))
Box 4349                                        |
Stanford, Ca, USA, 94309                      '\|/`
e-mail:snyder_at_slac.stanford.edu                 o
phone:650-926-2701                              _
http://www.slac.stanford.edu/~snyder          BaBar
FAX:650-926-2657                          Collaboration



On Wed, 12 Jul 2006, Rene Brun wrote:

> Art,
>
> In general I use "gdb root.exe"
> However to attach gdb to a running executable, I see a nice page by Axel
> describing
> how to do it. See:
> http://www.muenster.de/~naumana/rootgdb.html
>
> see also: http://www.phenix.bnl.gov/software/tutorials/debuggingTips.html
>
> Rene Brun
>
>
> Arthur E. Snyder wrote:
> > I'm trying to run gdb with root but can't see the symbols in my
> > shared library.
> >
> > I use the following procedure:
> >
> > Start a root session. In another window on the same machine run the
> > following macro:
> >
> > #!/bin/sh
> > line=`ps -wwfu $USER | grep root.exe | grep -v grep | tail -1`
> > if [ "$line" == "" ] ; then
> > echo "NO ROOT session running"
> > exit 1
> > fi
> > set $line
> > exec gdb $8 $2
> >
> > This brings up root in one window and gdb in the other. Typing "cont" in
> > the gdb window sets root session runing. That much works.
> >
> > I've compiled my shared library with -g and -O0. I've checked that
> > my symbols are actually in the library.
> >
> > Despite this I can't see the sysmbols associated with the shared library
> > routines even after I've executed them once to make sure they've been
> > loaded.
> >
> > Is there some trick I'm missing?
> >
> > -Art S.
> >
> > A.E. Snyder, Group EC \!c*p?/
> > SLAC Mail Stop #95 ((. .))
> > Box 4349 |
> > Stanford, Ca, USA, 94309 '\|/`
> > e-mail:snyder_at_slac.stanford.edu o
> > phone:650-926-2701 _
> > http://www.slac.stanford.edu/~snyder BaBar
> > FAX:650-926-2657 Collaboration
> >
> >
> >
> > On Tue, 4 Jul 2006, Rene Brun wrote:
> >
> >
> >> Art,
> >>
> >> I have translated the function in CERNLIB to C++. See attachment.
> >> I just tested that it compiles and executes with some random input numbers.
> >> Could you test it and let me know if it is what you want?
> >>
> >> Rene Brun
> >>
> >> Arthur E. Snyder wrote:
> >>
> >>> Hi again Rene,
> >>>
> >>> I see D-functions are not in mathlib yet though lots of other good stuff
> >>> is. The so called beta term is available in cernlib (called RDJMNB) from
> >>> the MATHLIB library. Is it possible to access this from ROOT?
> >>>
> >>> -Art S.
> >>>
> >>> A.E. Snyder, Group EC \!c*p?/
> >>> SLAC Mail Stop #95 ((. .))
> >>> Box 4349 |
> >>> Stanford, Ca, USA, 94309 '\|/`
> >>> e-mail:snyder_at_slac.stanford.edu o
> >>> phone:650-926-2701 _
> >>> http://www.slac.stanford.edu/~snyder BaBar
> >>> FAX:650-926-2657 Collaboration
> >>>
> >>>
> >>>
> >>> On Mon, 3 Jul 2006, Arthur E. Snyder wrote:
> >>>
> >>>
> >>>
> >>>> Hi Rene,
> >>>>
> >>>> I'm not looking for Breit-Wigner but for Wigner's so called D-function
> >>>> (D(alpha,beta,gamma)) which are functions of anlges used in partial wave
> >>>> analysis based on the Jacob and Wick formalism.
> >>>>
> >>>> The article I refer to proposes somekind of mathlib to collect in C++ form
> >>>> a lot of mathmatical functions available dispersed among a lot libraries
> >>>> of functions like cernlib, clhep, napl, and so forth. The D-functions are
> >>>> just mentioned in one of the tables .. google hunted them out .. with a
> >>>> search for "Wigner D-function C++".
> >>>>
> >>>> Anyway I need the D-function in a form I can access from ROOT. I may just
> >>>> have to code the few lower order ones I need myself, but I'd rather have a
> >>>> nice (well debugged) set from some package or class library.
> >>>>
> >>>> -Art S.
> >>>>
> >>>> A.E. Snyder, Group EC \!c*p?/
> >>>> SLAC Mail Stop #95 ((. .))
> >>>> Box 4349 |
> >>>> Stanford, Ca, USA, 94309 '\|/`
> >>>> e-mail:snyder_at_slac.stanford.edu o
> >>>> phone:650-926-2701 _
> >>>> http://www.slac.stanford.edu/~snyder BaBar
> >>>> FAX:650-926-2657 Collaboration
> >>>>
> >>>>
> >>>>
> >>>> On Mon, 3 Jul 2006, Rene Brun wrote:
> >>>>
> >>>>
> >>>>
> >>>>> Art,
> >>>>>
> >>>>> I am curious to know the talk/article you are referring to ::)
> >>>>> In ROOT we have the Breit-Wigner distributtion. See
> >>>>> http://root.cern.ch/root/htmldoc/src/TMath.cxx.html#TMath:BreitWigner
> >>>>> http://root.cern.ch/root/htmldoc/src/TRandom.cxx.html#TRandom:BreitWigner
> >>>>>
> >>>>> see also:
> >>>>> http://root.cern.ch/root/roottalk/roottalk05/0954.html
> >>>>>
> >>>>> Rene Brun
> >>>>>
> >>>>> Snyder, Arthur E. wrote:
> >>>>>
> >>>>>
> >>>>>> Does anybody know where to find a implentation of Wigner D-functions that I can use in ROOT?
> >>>>>>
> >>>>>> I found a talk/article by Rene where they are mentioned but the implentation seems to be in
> >>>>>> the future.
> >>>>>>
> >>>>>> -Art S.
> >>>>>>
> >>>>>>
> >>>>>> -----Original Message-----
> >>>>>> From: owner-roottalk_at_pcroot.cern.ch on behalf of Fine, Valeri
> >>>>>> Sent: Sun 7/2/2006 6:56 PM
> >>>>>> To: nasim; roottalk_at_pcroot.cern.ch
> >>>>>> Subject: RE: [ROOT] geting the values stored in the histograms.
> >>>>>>
> >>>>>>
> >>>>>> You need either
> >>>>>> http://root.cern.ch/root/htmldoc/TH1.html#TH1:GetBinLowEdge
> >>>>>> or
> >>>>>> http://root.cern.ch/root/htmldoc/TH1.html#TH1:GetBinCenter
> >>>>>>
> >>>>>> Valeri
> >>>>>>
> >>>>>>
> >>>>>>
> >>>>>>> There are two ways:
> >>>>>>>
> >>>>>>> 1. Each concrete histogram class is derived from TArray class
> >>>>>>> http://root.cern.ch/root/htmldoc/TH1F.html
> >>>>>>> This means one can treat each histogram object as a plain array of
> >>>>>>> "float" number and use it as such. (For example method "[int]} returns
> >>>>>>> what you want)
> >>>>>>>
> >>>>>>> TH1F *histogram = new TH1F( some parameters);
> >>>>>>> TH1F &bins = *histogram;
> >>>>>>> for (int i=0; i< 100; i++)
> >>>>>>> printf(" Bin %i contains the value %f\n", i, bins[i]);
> >>>>>>>
> >>>>>>> 2. One can use the methods like TH1::GetBin<XXX>
> >>>>>>> , where XXX could be "Center", "Content", "Error" etc
> >>>>>>> see doc please
> >>>>>>> http://root.cern.ch/root/htmldoc/TH1.html#TH1:GetBinContent
> >>>>>>>
> >>>>>>>
> >>>>>>>
> >>>>>>>
> >>>>>>
> >>>>>>> Hope this helps.
> >>>>>>>
> >>>>>>> ----
> >>>>>>> Best regards
> >>>>>>> Valeri
> >>>>>>>
> >>>>>>>
> >>>>>>>
> >>>>>>>
> >>>>>>>
> >>>>>>>> -----Original Message-----
> >>>>>>>> From: owner-roottalk_at_pcroot.cern.ch
> >>>>>>>>
> >>>>>>>>
> >>>>>>>>
> >>>>>>> [mailto:owner-roottalk_at_pcroot.cern.ch] On
> >>>>>>>
> >>>>>>>
> >>>>>>>
> >>>>>>>> Behalf Of nasim
> >>>>>>>> Sent: Sunday, July 02, 2006 7:43 PM
> >>>>>>>> To: roottalk_at_pcroot.cern.ch
> >>>>>>>> Subject: [ROOT] geting the values stored in the histograms.
> >>>>>>>>
> >>>>>>>>
> >>>>>>>> Hi,
> >>>>>>>> If we have one dimensional histogram with 100 entries.
> >>>>>>>> Is it possible to extract the value of its x axis for each entry?
> >>>>>>>>
> >>>>>>>> Any help would be appreciated,
> >>>>>>>>
> >>>>>>>> Nasim
> >>>>>>>>
> >>>>>>>>
> >>>>>>>> --
> >>>>>>>>
> >>>>>>>>
> >>>>>>>>
> >>>>>>
> >>>>>>
> >>>>>>
> >>>>>>
> >>>>>>
> >>
>
>
Received on Thu Jul 13 2006 - 00:47:33 MEST

This archive was generated by hypermail 2.2.0 : Mon Jan 01 2007 - 16:31:59 MET