sherpa is hosted by Hepforge, IPPP Durham
close Warning: Can't synchronize with repository "(default)" (/hepforge/svn/sherpa does not appear to be a Subversion repository.). Look in the Trac log for more information.
Modify ↓

Opened 14 years ago

Closed 13 years ago

Last modified 10 years ago

#104 closed defect (fixed)

Multithreading makes ttbar + 2jets integration crash

Reported by: Frank Siegert Owned by: Stefan Hoeche
Priority: minor Milestone:
Component: Unknown Version: 1.2.2
Keywords: Cc:

Description

When integrating ttbar + 2 jets with a multithreaded Sherpa, it crashes: 

Process_Group::CalculateTotalXSec(): Calculate xs for '2_8__j__j__t[W+[lepton__neutrino]__b]__tb[W-[Q__Q]__bb]__j__j' (Comix)
Starting the calculation. Lean back and enjoy ... .
METS_Scale_Setter::CalculateScale(): No CSS history for '2_8__db__cb__t[W+[nu_e__e+]__b]__tb[W-[s__cb]__bb]__sb__ub' in >0.04% of calls. Set \hat{s}.
METS_Scale_Setter::CalculateScale(): No CSS history for '2_8__d__ub__t[W+[nu_e__e+]__b]__tb[W-[s__cb]__bb]__s__cb' in >0.01% of calls. Set \hat{s}.
METS_Scale_Setter::CalculateScale(): No CSS history for '2_8__d__sb__t[W+[nu_e__e+]__b]__tb[W-[s__cb]__bb]__u__cb' in >0.01% of calls. Set \hat{s}.
METS_Scale_Setter::CalculateScale(): No CSS history for '2_8__d__sb__t[W+[nu_e__e+]__b]__tb[W-[d__ub]__bb]__u__cb' in >0.01% of calls. Set \hat{s}.
METS_Scale_Setter::CalculateScale(): No CSS history for '2_8__db__cb__t[W+[nu_e__e+]__b]__tb[W-[d__ub]__bb]__sb__ub' in >0.01% of calls. Set \hat{s}.

Exception_Handler::SignalHandler: Signal (11) caught. 
Exception_Handler::GenerateStackTrace(..): Generating stack trace 
{
  0x2af8497cf194  in 'COMIX::PS_Channel::PropWeight(COMIX::Current_Base const*, unsigned long const&, double const&, double const&, double const&)' (PS_Channel.C:368)
  0x2af8497cf194  in 'COMIX::PS_Channel::PropWeight(COMIX::Current_Base const*, unsigned long const&, double const&, double const&, double const&)' (PS_Channel.C:368)
  0x2af8497cf962  in 'COMIX::PS_Channel::GenerateWeight(COMIX::Current_Base*, COMIX::Current_Base*, COMIX::Current_Base*, unsigned long&)' (PS_Channel.C:778)
  0x2af8497cf962  in 'COMIX::PS_Channel::GenerateWeight(COMIX::Current_Base*, COMIX::Current_Base*, COMIX::Current_Base*, unsigned long&)' (PS_Channel.C:778)
  0x2af8497cffdd  in 'COMIX::PS_Channel::GenerateWeight(COMIX::PS_Current*)' (PS_Vertex.H:36)
  0x2af8497cffdd  in 'COMIX::PS_Channel::GenerateWeight(COMIX::PS_Current*)' (PS_Vertex.H:36)
  0x2af8497d0ac9  in 'COMIX::PS_Channel::TGenerateWeight(void*)' (PS_Channel.C:918)
  0x2af8497d0ac9  in 'COMIX::PS_Channel::TGenerateWeight(void*)' (PS_Channel.C:918)
  0x2af8561e9c2d  in 'clone' 
}   Cannot continue.

The run card was simply the following: 

(run){
  WRITE_MAPPING_FILE=3
  MASSIVE[15]=1

  MASS[6]=172.5
  MASS[23]=91.1876
  MASS[24]=80.403
  WIDTH[23]=2.4952
  WIDTH[24]=2.141
  BEAM_1=2212
  BEAM_2=2212
  BEAM_ENERGY_1=3500
  BEAM_ENERGY_2=3500
  FRAGMENTATION=Off
  EVENTS=5000
}(run)

(processes){
  Process : 93 93 ->  6[a] -6[b] 93{2}
  DecayOS : 6[a] -> 24[c] 5
  DecayOS : -6[b] -> -24[d] -5
  #DecayOS : 24[c] -> 90 91
  #DecayOS : -24[d] -> 94 94
  DecayOS : 24[c] -> -11 12
  DecayOS : -24[d] -> 13 -14
  Order_EW 6
  CKKW sqr(20/E_CMS)
  End process
}(processes)

(me){
  ME_SIGNAL_GENERATOR = Comix
}(me)

Attachments (1)

ps_threading_1.patch​ (3.8 KB) - added by Stefan Hoeche 13 years ago.

Download all attachments as: .zip

Change History (7)

comment:1 Changed 13 years ago by Frank Siegert

Milestone: rel-1.2.3 → rel-1.3.0

I still see the crash during integration of ttbar + 2 jets (the + 1 jet process works) with trunk r16987 with multithreading. The eror message is unfortunately completely messed up (reproducedly): 

Process_Group::CalculateTotalXSec(): Calculate xs for '2_8__j__j__t[W+[lepton__neutrino]__b]__tb[W-[lepton__neutrino]__bb]__j__j' (Comix)
Starting the calculation. Lean back and enjoy ... .
    CS_Cluster_Definitions::CoreScale(): Momentum not conserved.
      \sum p = (0.00335007,9.37887e-14,3.59222e-14,-0.00335007) in
      (0x112e3000): 2 -> 2 {
        \mu_r = 0, \mu_f = 0, \mu = 1175.27
      k_T = 1070.72, z = 0, phi = 0, kin = 0
        oew = 0, oqcd = 2, nlo = 0, new = (<no entry>)
        (0,2,3,4,5,6,7)          db (-1.07642e-08,-0,-0,-1.07642e-08) 5.96795e-16 (0,2) [0|0,0] <-> (1)
             (1)           b (-1659.63,-0,-0,1659.63) 0 (2,0) [0|0,0]
             (8)           d (878.062,-0.00312453,-0.00283575,-878.062) 7.80616e-06 (1,0) [0|0,0]
             (9)          bb (781.575,0.00312453,0.00283575,-781.575) 5.89815e-06 (0,1) [0|0,0]
  } 
    CS_Cluster_Definitions::CoreScale(): Momentum not conserved.
      \sum p = (0.00335007,9.37887e-14,3.59222e-14,-0.00335007) in
      (0xdcb44a8): 2 -> 2 {
        \mu_r = 0, \mu_f = 0, \mu = 1175.27
        k_T = 1070.72, z = 0, phi = 0, kin = 0
        oew = 0, oqcd = 2, nlo = 0, new = (<no entry>)
        (0,2,3,4,5,6,7)          db (-1.07642e-08,-0,-0,-1.07642e-08) 5.96795e-16 (0,2) [0|0,0] <-> (1)
                 (1)           b (-1659.63,-0,-0,1659.63) 0 (2,0) [0|0,0]
                 (8)           d (878.062,-0.00312453,-0.00283575,-878.062) 7.80616e-06 (1,0) [0|0,0]
             (9)          bb (781.575,0.00312453,0.00283575,-781.575) 5.89815e-06 (0,1) [0|0,0]
      }
  Exception_Handler::SignalHandler: 
Signal
[0xm(ception_Handler::SignalHandler1:1 Sign[l0 m0m1[1)m[31mm 11c)mu[ht. 
34m caught.
      Exception_Handler::GenerateStackTrace(..): Generatin[ 1smtEaxck cterapcet i0nm_andler::Ge[n1emr
      {a
teStackTrace(..): Generating stack trace
    {
      4294960384 in   <unknown function>
4294960384 in   <unknown function>

Changed 13 years ago by Stefan Hoeche

Attachment: ps_threading_1.patch​ added

comment:2 Changed 13 years ago by Stefan Hoeche

Hi Frank,

can you try the attached patch to see whether it solves the problem? Thanks

Stefan

comment:3 Changed 13 years ago by Stefan Hoeche

Resolution: → fixed
Status: new → closed

This problem seems to be fixed by r17013. Please reopen the ticket if necessary.

comment:4 Changed 13 years ago by Frank Siegert

Resolution: fixed
Status: closed → reopened

It seems somewhat better, but unfortunately not quite working yet: 

Process_Group::CalculateTotalXSec(): Calculate xs for '2_9__j__j__t[W+[lepton__neutrino]__b]__tb[W-[lepton__neutrino]__bb]__j__j__j' (Comix)
Starting the calculation. Lean back and enjoy ... .
0.430638 pb +- ( 0.0405835 pb = 9.42403 % ) 50000 ( 105754 -> 47.2 % )
full optimization:  ( 1h 57m 3s(3h 18m 12s) elapsed / 2d 16h 22m 46s(4d 13h 21s) left )
Channel_Basics::SqLam argument -0.0691094 <0 in Channel_Basics::sqlam()
s;s1;s2: 29758.4;3.82219e-05;29756.2
Channel_Basics::SqLam argument -0.299555 <0 in Channel_Basics::sqlam()
s;s1;s2: 29758.6;29756.2;4.9e-05
Channel_Basics::SqLam argument nan <0 in Channel_Basics::sqlam()
s;s1;s2: nan;0;6464.64
Channel_Basics::SqLam argument nan <0 in Channel_Basics::sqlam()
s;s1;s2: nan;0;0
  CS_Cluster_Definitions::CoreScale(): Momentum not conserved.
  \sum p = (-0.00791556,5.01199e-07,-2.49379e-07,0.00791556) in
(0x90b17c8): 2 -> 2 {
    \mu_r = 0, \mu_f = 0, \mu = 2.85982
    k_T = 2.78618, z = 0, phi = 0, kin = 0
    oew = 0, oqcd = 0, nlo = 0, new = (<no entry>)
    decs = { (2,3)[W+|2,1] (2,3,4)[t|2,1] (5,6)[W-|2,1] (5,6,7)[tb|2,1] }
  (0,2,3,4,5,6,7,9)           u (-0.000104909,-0,-0,-0.000104909) 4.76952e-12 (2,0) [0|0,0] <-> (1)
           (1)          db (-2221.14,-0,-0,2221.14) 0.00187591 (0,2) [0|0,0]
           (8)           s (861.368,-0.416647,-0.218337,-861.368) -1.41533e-05 (1,0) [0|0,0]
          (10)          cb (1359.76,0.416647,0.218337,-1359.76) 0.00700001 (0,1) [0|0,0]
} 
  CS_Cluster_Definitions::CoreScale(): Momentum not conserved.
  \sum p = (-0.00791556,5.01199e-07,-2.49379e-07,0.00791556) in
  (0xe5830a50): 2 -> 2 {
  \mu_r = 0, \mu_f = 0, \mu = 2.85982
  k_T = 2.78618, z = 0, phi = 0, kin = 0
  oew = 0, oqcd = 0, nlo = 0, new = (<no entry>)
  decs = { (2,3)[W+|2,1] (2,3,4)[t|2,1] (5,6)[W-|2,1] (5,6,7)[tb|2,1] }
  (0,2,3,4,5,6,7,9)           c (-0.000104909,-0,-0,-0.000104909) 4.76952e-12 (2,0) [0|0,0] <-> (1)
           (1)          ub (-2221.14,-0,-0,2221.14) 0.00187591 (0,2) [0|0,0]
           (8)           d (861.368,-0.416647,-0.218337,-861.368) -1.41533e-05 (3,0) [0|0,0]
            (10)          sb (1359.76,0.416647,0.218337,-1359.76) 0.00700001 (0,3) [0|0,0]
  }
0.466837 pb +- ( 0.0331166 pb = 7.09381 % ) 100000 ( 175062 -> 72.1 % )
full optimization:  ( 3h 57m 52s(6h 53m 29s) elapsed / 2d 15h 26m(4d 14h 15m 44s) left )
0.478108 pb +- ( 0.0253872 pb = 5.30992 % ) 150000 ( 236755 -> 81 % )
full optimization:  ( 5h 53m 20s(10h 36m 23s) elapsed / 2d 12h 51m 9s(4d 13h 35m 56s) left )

    Exception_Handler::SignalHandler: Signal (11) caught.
    Exception_Handler::GenerateStackTrace(..): Generating stack trace
    {
   4294960384 in   <unknown function>
  0xf7a9cb12      in 'COMIX::PS_Channel::PropWeight(COMIX::Current_Base const*, unsigned int const&, double const&, double const&, double const&)' (PS_Channel.C:363)
  0xf7a9d3df      in 'COMIX::PS_Channel::GenerateWeight(COMIX::Current_Base*, COMIX::Current_Base*, COMIX::Current_Base*, unsigned int&)' (PS_Channel.C:798)
  0xf7a9d7f5      in 'COMIX::PS_Channel::GenerateWeight(COMIX::PS_Current*)' (PS_Vertex.H:36)
  0xf7a9e375      in 'COMIX::PS_Channel::TGenerateWeight(void*)' (PS_Channel.C:938)
  0x8baf6e        in 'clone'
}   Cannot continue.

comment:5 Changed 13 years ago by Stefan Hoeche

Resolution: → fixed
Status: reopened → closed

This is fixed with r17039.

comment:6 Changed 10 years ago by Stefan Hoeche

Milestone: rel-1.3.0

Milestone rel-1.3.0 deleted

Modify Ticket

Change Properties
Action
as closed The owner will remain Stefan Hoeche.

Add Comment

E-mail address and name can be saved in the Preferences.

Attachments ↑ Description ↑
 
Note: See TracTickets for help on using tickets.