SIGSEGV signal with h2root version 5.22.00

[quote=“brun”]h2root in version 5.23/02 should work on 64 bits machines

Rene[/quote]

Hi All,
i was seeing precisely the same error in 5.22… i downloaded 5.23/02, ran ‘./configure linuxx8664gcc’ and compiled using ‘make -j2’, all of which worked fine.

root itself runs fine.

  *******************************************
  *                                         *
  *        W E L C O M E  to  R O O T       *
  *                                         *
  *   Version   5.23/02  26 February 2009   *
  *                                         *
  *  You are welcome to visit our Web site  *
  *          http://root.cern.ch            *
  *                                         *
  *******************************************

ROOT 5.23/02 (trunk@27619, Mar 31 2009, 10:58:00 on linuxx8664gcc)

CINT/ROOT C/C++ Interpreter version 5.17.00, Dec 21, 2008

but h2root still crashes…

cat /etc/redhat-release....
    Red Hat Enterprise Linux Server release 5.3 (Tikanga)
gfortran --version:
    GNU Fortran (GCC) 4.1.2 20080704 (Red Hat 4.1.2-44)
gcc --version:
    gcc (GCC) 4.1.2 20080704 (Red Hat 4.1.2-44)

here’s the crash…

h2root d_urqmd_dau200_0.2.hst

 *** Break *** segmentation violation
(no debugging symbols found)
Attaching to program: /proc/29811/exe, process 29811
(no debugging symbols found)...done.
(no debugging symbols found)...done.
(no debugging symbols found)...done.
(no debugging symbols found)...done.
(no debugging symbols found)...done.
(no debugging symbols found)...done.
(no debugging symbols found)...done.
(no debugging symbols found)...done.
(no debugging symbols found)...done.
[Thread debugging using libthread_db enabled]
[New Thread 0x2b6687e7a130 (LWP 29811)]
(no debugging symbols found)...done.
(no debugging symbols found)...done.
(no debugging symbols found)...done.
0x0000003313c99335 in waitpid ()
   from /lib64/libc.so.6
#1  0x0000003313c3c2e1 in do_system () from /lib64/libc.so.6
#2  0x00002b66838c3dec in TUnixSystem::StackTrace ()
   from /home/llope/root_build523/lib/libCore.so
#3  0x00002b66838c131a in TUnixSystem::DispatchSignals ()
   from /home/llope/root_build523/lib/libCore.so
#4  <signal handler called>
#5  0x00002b668720f97e in mzstor_ ()
   from /home/llope/root_build523/lib/libminicern.so
#6  0x00002b668720fd9f in mzpaw_ ()
   from /home/llope/root_build523/lib/libminicern.so
#7  0x00002b66871fbbac in hlimit_ ()
   from /home/llope/root_build523/lib/libminicern.so
#8  0x000000000040449f in main ()
The program is running.  Quit anyway (and detach it)? (y or n) [answered Y; input not from terminal]
Detaching from program: /proc/29811/exe, process 29811

i attached the hbook file that i am trying to convert. it opens & plays fine in paw 2.14/04…

any ideas? many thanks!
bill

EDIT 1:--------------------

i just downloaded and compiled 5.18/00. worked fine.

now h2root runs w/out crashing. but there is still some problem…

i ran this command:

h2root d_urqmd_dau200_0.3.hst d_urqmd_dau200_0.3.root 0 0 1024
 Converting directory //example
TFile**		d_urqmd_dau200_1.root	HBOOK file: d_urqmd_dau200_1.hst converted to ROOT
 TFile*		d_urqmd_dau200_1.root	HBOOK file: d_urqmd_dau200_1.hst converted to ROOT
  KEY: TH1F	h100;1	geant_id
  KEY: TH1F	h110;1	b for all evts (fm) 
  KEY: TH1F	h111;1	b for selected evts (fm)
[...snip...]
  KEY: TProfile	h8031;1	[r]fo vs pt, p  
  KEY: TProfile	h8032;1	[r]fo vs pt, n  
  KEY: TProfile	h8033;1	[r]fo vs pt, d  
  KEY: TProfile	h8041;1	tfo vs pt, p
[...snip...]

i can see non-zero numbers in the stats box, but only the TProfiles show something in the pad when the histogram is drawn (either by clicking in the TBrowser or by e.g. h8031->Draw();

that is, the TH1F’s that h2root made have nonzero stats entries but the histogram seems to be empty…

i attached the root file that h2root (5.18/00) wrote from the rz file that i already attached…

EDIT 2:--------------------

o.k. - i just grabbed h2root.cxx and hntvar2.f from the 5.18/00 tree and wrote a little Makefile and built h2root myself…

now h2root works perfectly. no crashes, and i see all TH1F’s just like they look when viewed in PAW. (?!?)

so while i really have no idea what i fixed, it’s working now…

i’m happy to provide any additional details if anyone thinks that’d be helpful…

cheers
d_urqmd_dau200_0.3.root (124 KB)
d_urqmd_dau200_0.3.hst.zip (25 KB)

Hi Bill,

In fact the fix to h2root was applied the day after the release of 5.23/02.
You must download the SVN trunk to get it working. I checked that it works on your file.

Rene

Hello,

I have been trying to use h2root on the new 64-bit machine at Indiana U to convert .paw files into .root files. The machine has 4 quad-core Xenon 2.93 GHz CPUs.

We are running Linux version 2.6.18-128.1.1.el5xen with Red Hat 4.1.2-42.

Previously, under an earlier version of ROOT 5.22.00a, it would crash immediately. After reading this thread, I updated to ROOT 5.23.04; now the crash comes later in running

[quote]Creating branch:beampz, block:NUGPINFO, fullname:beampz/F, nsub=0, itype=1, isize=4, ielem=1

*** Break *** segmentation violation
Using host libthread_db library “/lib64/libthread_db.so.1”.
Attaching to program: /proc/31350/exe, process 31350
[Thread debugging using libthread_db enabled]
[New Thread 47984261063040 (LWP 31350)]
0x0000003fa6297f85 in waitpid () from /lib64/libc.so.6
#1 0x0000003fa623bf33 in do_system () from /lib64/libc.so.6
#2 0x00002ba42fc8d65b in TUnixSystem::Exec (this=0x122be2b0,
#3 0x00002ba42fc8ca44 in TUnixSystem::StackTrace (this=0x122be2b0)
at core/unix/src/TUnixSystem.cxx:2132
#4 0x00002ba42fc8f99b in TUnixSystem::DispatchSignals (this=0x122be2b0,
sig=kSigSegmentationViolation) at core/unix/src/TUnixSystem.cxx:1099
#5 0x00002ba42fc8fabf in SigHandler (sig=kSigSegmentationViolation)
at core/unix/src/TUnixSystem.cxx:350
#6 0x00002ba42fc86111 in sighandler (sig=11)
at core/unix/src/TUnixSystem.cxx:3368
#7
#8 0x00002ba4339052b2 in hgnt2_ (var1=0x2ba4339402a0, ivoff=0x2ba43394034c,
nvar1=@0x2ba43394034c, idnevt=@0x7fff7b0f8638, ierror=@0x7fff7b0f863c,
var1=1) at misc/minicern/src/hbook.f:1419
#9 0x00002ba433903dca in hgnt1
(idd=@0x7fff7b0f5cdc, blkna1=@0x2ba4339402a0,
var=0x2ba4339402a0, ioffst=0x2ba43394034c, nvar=@0x2ba43394034c,
idnevt=@0x7fff7b0f8638, ierror=@0x7fff7b0f863c, _blkna1=1, var=1)
at misc/minicern/src/hbook.f:1164
#10 0x00002ba433903996 in hgnt
(idn=@0x7fff7b0f5cdc, idnevt=@0x7fff7b0f8638,
ierror=@0x7fff7b0f863c) at misc/minicern/src/hbook.f:1072
#11 0x0000000000402f41 in convert_cwn (id=10) at main/src/h2root.cxx:812
#12 0x0000000000404696 in convert_directory (dir=0x4053a4 “//example”)
at main/src/h2root.cxx:377
#13 0x0000000000404d5f in main (argc=2, argv=0x7fff7b0f9148)
at main/src/h2root.cxx:329[/quote]

“beampz” is the final variable in the .paw file. h2root works fine for these .paw files on our 32-bit machine.

Thank you,

Luke Corwin
Indiana University

[quote=“brun”]Hi Bill,

In fact the fix to h2root was applied the day after the release of 5.23/02.
You must download the SVN trunk to get it working. I checked that it works on your file.

Rene[/quote]

h2 root works fine for me but I get a very similar error in the same spot in g2root. So similar I have to guess it may have the same fix.

I’m using 64 bit gcc-4.4.4 on gentoo (either the ebuild or svn have the same behavior) segfaulting in the same way with or without an input file (a known valid one for that matter).

here’s what gdb says:

GNU DDD 3.3.12 (x86_64-pc-linux-gnu), by Dorothea LReading symbols from /usr/bin/g2root…Reading symbols from /usr/lib64/debug/usr/bin/g2root.debug…(no debugging symbols found)…done.
(no debugging symbols found)…done.
(gdb) run
warning: no loadable sections found in added symbol-file /usr/lib64/debug/usr/lib64/root/libminicern.so.5.28.debug

Program received signal SIGSEGV, Segmentation fault.
0x00007ffff6882732 in mzstor_ () from /usr/lib64/root/libminicern.so.5.28
(gdb) where
#0 0x00007ffff6882732 in mzstor_ () from /usr/lib64/root/libminicern.so.5.28
#1 0x00007ffff6881b6f in mzpaw_ () from /usr/lib64/root/libminicern.so.5.28
#2 0x00007ffff686e1e5 in hlimit_ () from /usr/lib64/root/libminicern.so.5.28
#3 0x0000000000401bd4 in MAIN__ ()
#4 0x000000000040c2ca in main ()
(gdb)

Hi,

Since the issue is not about h2root nor 5.22, could you open a new forum post or report this bug at savannah.cern.ch.

Does the problem still occur if you move to v5.30?

Thanks,
Philippe.