Crash in FSI code when running on hydrogen target

[karuboniru]

input file to reproduce the issue

---

The traceback is:

#0  __pthread_kill_implementation (threadid=<optimized out>, signo=signo@entry=6, no_tid=no_tid@entry=0) at pthread_kill.c:44
#1  0x00007ffff5680793 in __pthread_kill_internal (threadid=<optimized out>, signo=6) at pthread_kill.c:78
#2  0x00007ffff5627d0e in __GI_raise (sig=sig@entry=6) at ../sysdeps/posix/raise.c:26
#3  0x00007ffff560f942 in __GI_abort () at abort.c:79
#4  0x00007ffff560f85e in __assert_fail_base (fmt=0x7ffff57c3cb0 "%s%s%s:%u: %s%sAssertion `%s' failed.\n%n", assertion=assertion@entry=0x220b53 "E>=_mass", file=file@entry=0x21fb38 "src/particle.h", 
    line=line@entry=227, function=function@entry=0x21f680 "void particle::set_energy(double)") at assert.c:94
#5  0x00007ffff561fe47 in __assert_fail (assertion=0x220b53 "E>=_mass", file=0x21fb38 "src/particle.h", line=227, function=0x21f680 "void particle::set_energy(double)") at assert.c:103
#6  0x0000000000507a95 in particle::set_energy (this=0x7fffffffc110, E=<optimized out>) at src/particle.h:220
#7  particle::set_energy (this=0x7fffffffc110, E=<optimized out>) at src/particle.h:220
#8  kaskada::prepare_particles (this=this@entry=0x7fffffffc690) at src/kaskada7.cc:145
#9  0x000000000050824c in kaskada::kaskadaevent (this=this@entry=0x7fffffffc690) at src/kaskada7.cc:41
#10 0x00000000004d1359 in NuWro::finishevent (this=this@entry=0x724c60 <nuwro>, e=0x29ffd90, p=...) at src/nuwro.cc:513
#11 0x00000000004d7e99 in NuWro::real_events (this=this@entry=0x724c60 <nuwro>, p=...) at src/nuwro.cc:841
#12 0x00000000004e3675 in NuWro::main (this=0x724c60 <nuwro>, argc=5, argv=0x7fffffffd588) at src/nuwro.cc:1004
#13 0x000000000051c15a in main (argc=5, argv=0x7fffffffd588) at src/main.cc:34

The problem causing code is following:

nuwro/src/kaskada7.cc

Lines 142 to 145 in c8d29a6

	if(par.nucleus_target) p1.set_fermi(nucl->hyp_BE(p1.r.length(),p1.pdg));
	else p1.set_fermi(0);
	p1.set_energy(p1.E() + p1.his_fermi);

• the code set the binding energy by checking par.nucleus_target which is 2 in the input (which I think should never happen, so the condition to do set_fermi is the code causing the issue)

  gdb> p par.nucleus_target
  $2 = 2
  

• The related sigma has a quite low momentum

  $3 = {
    <vect> = {
      t = 1244.7778345721508,
      x = 39.566419917875351,
      y = 97.297454760823186,
      z = 351.90703922212208
    }, 
    members of particle:
    _mass = 1189.3700000000003,
    r = {
      t = 0,
      x = 0,
      y = 0,
      z = 0
    },
    pdg = 3222,
    ks = 1 '\001',
    orgin = 3 '\003',
    travelled = 0,
    id = 3,
    mother = 0,
    endproc = -1,
    his_fermi = -70,
    primary = false
  }
  

• The substraction of binding energy caused the issue.

---

I am unsure if this should be considered as invalid input or something. But if we want to properly handle the input we might want to

• reset the nucleus_target to 0 when nucleon target is configured;
• add another check like nucleus_p + nucleus_n > 1 to everywhere it applies (or disable FSI for nucleon target).

[jsobczyk]

Hello, From a physics perspective, there is no FSI for a hydrogen target. Jan prof. dr hab. Jan Sobczyk PROFESOR Zakład Fizyki Neutrin, Instytut Fizyki Teoretycznej, Wydział Fizyki i Astronomii KIEROWNIK Zakład Fizyki Neutrin, Instytut Fizyki Teoretycznej, Wydział Fizyki i Astronomii Uniwersytet Wrocławski pl. M. Borna 9 50-204 Wrocław tel. +48 71 375 94 39 [cid:2022_02_uwr_arqus_pocitedostopki_bdb4c7bb-1172-4107-90f8-1b2742af358e.png]<https://uni.wroc.pl>[cid:2022_02_uwr_arqus_pocitedostopki3_e5829564-9e6b-4628-ac3b-6e1d97627e4b.png]<https://uwr.edu.pl/inicjatywa-doskonalosci-uczelnia-badawcza/>[cid:2022_02_uwr_arqus_pocitedostopki4_89c9e33b-31a4-4974-a07d-b817a789344c.png] Wiadomość ta oraz wszelkie załączone do niej pliki są poufne i mogą być prawnie chronione. Jeżeli nie jest Pani/Pan zamierzonym adresatem niniejszej wiadomości, nie może Pani/Pan jej ujawniać, kopiować, dystrybuować ani też w żaden inny sposób udostępniać lub wykorzystywać. O błędnym zaadresowaniu wiadomości proszę niezwłocznie poinformować nadawcę i usunąć wiadomość. pon., 23 wrz 2024 o 16:34 Karuboniru ***@***.******@***.***>> napisał(a): input file to reproduce the issue<https://github.com/user-attachments/files/17098984/i.txt>

…

________________________________ The traceback is: #0 __pthread_kill_implementation (threadid=<optimized out>, ***@***.***=6, ***@***.***=0) at pthread_kill.c:44 #1 0x00007ffff5680793 in __pthread_kill_internal (threadid=<optimized out>, signo=6) at pthread_kill.c:78 #2 0x00007ffff5627d0e in __GI_raise ***@***.***=6) at ../sysdeps/posix/raise.c:26 #3 0x00007ffff560f942 in __GI_abort () at abort.c:79 #4 0x00007ffff560f85e in __assert_fail_base (fmt=0x7ffff57c3cb0 "%s%s%s:%u: %s%sAssertion `%s' failed.\n%n", ***@***.***=0x220b53 "E>=_mass", ***@***.***=0x21fb38 "src/particle.h", ***@***.***=227, ***@***.***=0x21f680 "void particle::set_energy(double)") at assert.c:94 #5 0x00007ffff561fe47 in __assert_fail (assertion=0x220b53 "E>=_mass", file=0x21fb38 "src/particle.h", line=227, function=0x21f680 "void particle::set_energy(double)") at assert.c:103 #6 0x0000000000507a95 in particle::set_energy (this=0x7fffffffc110, E=<optimized out>) at src/particle.h:220 #7 particle::set_energy (this=0x7fffffffc110, E=<optimized out>) at src/particle.h:220 #8 kaskada::prepare_particles ***@***.***=0x7fffffffc690) at src/kaskada7.cc:145 #9 0x000000000050824c in kaskada::kaskadaevent ***@***.***=0x7fffffffc690) at src/kaskada7.cc:41 #10 0x00000000004d1359 in NuWro::finishevent ***@***.***=0x724c60 <nuwro>, e=0x29ffd90, p=...) at src/nuwro.cc:513 #11 0x00000000004d7e99 in NuWro::real_events ***@***.***=0x724c60 <nuwro>, p=...) at src/nuwro.cc:841 #12 0x00000000004e3675 in NuWro::main (this=0x724c60 <nuwro>, argc=5, argv=0x7fffffffd588) at src/nuwro.cc:1004 #13 0x000000000051c15a in main (argc=5, argv=0x7fffffffd588) at src/main.cc:34 The problem causing code is following: https://github.com/NuWro/nuwro/blob/c8d29a6c61e65b847642bab7b29c1683a5853ea6/src/kaskada7.cc#L142-L145 * the code set the binding energy by checking par.nucleus_target which is 2 in the input gdb> p par.nucleus_target $2 = 2 * The related sigma has a quite low momentum $3 = { <vect> = { t = 1244.7778345721508, x = 39.566419917875351, y = 97.297454760823186, z = 351.90703922212208 }, members of particle: _mass = 1189.3700000000003, r = { t = 0, x = 0, y = 0, z = 0 }, pdg = 3222, ks = 1 '\001', orgin = 3 '\003', travelled = 0, id = 3, mother = 0, endproc = -1, his_fermi = -70, primary = false } * The substraction of binding energy caused the issue.

________________________________ I am unsure if this should be considered as invalid input or something. But if we want to properly handle the input we might want to * reset the nucleus_target to 0 when nucleon target is configured; * add another check like nucleus_p + nucleus_n > 1 to everywhere it applies. — Reply to this email directly, view it on GitHub<#33>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/AF3X7PWXQ74O7C2VJ26X72LZYARF3AVCNFSM6AAAAABOWHIOPCVHI2DSMVQWIX3LMV43ASLTON2WKOZSGU2DEOBWGAZTQOI>. You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>

[karuboniru]

Yes, if the code in kaskada7 only does FSI then we should be safe to skip them for nucleon target. (See the PR #34)

While when reading the code, it seems that the actual cascade starts from following line

nuwro/src/kaskada7.cc

Line 66 in cf1d497

while (parts.size () > 0 and nucl->Ar() > 0) // main loop in cascade

and the check of nucl->Ar() > 0 should be able to make the code to skip nucleon target as we wanted. The code before cascade seems to be doing something else like "make sure about the energy balance" (from the text in comment).