Assertion failed: ...

[geha]

Hello,

when I start the debugger (idb as well as idbc) and load my program, I get the following message:

Assertion failed: "(*i)->getThreadId() != ThreadId()" src/appl/procs/addrspace.C:2819
This is an unexpected condition and may indicate the presence of a defect.
If you wish to report this, please include the stack trace that follows.
/opt/intel/Compiler/11.0/074/bin/intel64/iidb(_ZN15IDBAssertFailed3runEPKcS1_j+0xe) [0x5d3ca4]
/opt/intel/Compiler/11.0/074/bin/intel64/iidb(_ZN14DTLU_namespace12assertFailedEPKcS1_j+0x24) [0xb8210c]
/opt/intel/Compiler/11.0/074/bin/intel64/iidb(_ZN12AddrSpaceDef9getThreadERK8ThreadId+0x7b) [0x675ec3]
/opt/intel/Compiler/11.0/074/bin/intel64/iidb(_ZN10ProcessDef9getThreadERK8ThreadId+0x17) [0x68d56b]
/opt/intel/Compiler/11.0/074/bin/intel64/iidb(_ZN10ControlDef20forceThreadExistenceEP17LightweightWorkerR8ThreadIdRP6Thread+0x24d) [0xab1843]
/opt/intel/Compiler/11.0/074/bin/intel64/iidb(_ZN10ControlDef36getPrstatusLWPForcingThreadExistenceEv+0x4c) [0xaa5d1a]
/opt/intel/Compiler/11.0/074/bin/intel64/iidb(_ZN10ControlDef13getStoppedLWPEv+0x54) [0xab19a4]
/opt/intel/Compiler/11.0/074/bin/intel64/iidb(_ZN12AddrSpaceDef22getLatestStoppedThreadEv+0x64) [0x696caa]
/opt/intel/Compiler/11.0/074/bin/intel64/iidb(_ZN12AddrSpaceDef23setThreadDebugInterfaceE15EGIThreadFlavorb+0x104) [0x673be4]
/opt/intel/Compiler/11.0/074/bin/intel64/iidb(_ZN12AddrSpaceDef10noteLoadedEPK6FileIdP7Program+0x1f8) [0x6731c2]
/opt/intel/Compiler/11.0/074/bin/intel64/iidb(_ZN10ProcessDef10noteExecedEPKc+0x66) [0x68acce]
/opt/intel/Compiler/11.0/074/bin/intel64/iidb(_ZN12ControlDefUx10noteExecedEv+0x27) [0xa9e479]
/opt/intel/Compiler/11.0/074/bin/intel64/iidb(_ZN18ControlDefUxPtrace21startProcess_postExecEPK6FileId+0x1af) [0xaadd0d]
/opt/intel/Compiler/11.0/074/bin/intel64/iidb(_ZN12ControlDefUx12startProcessEPKcRK7ArgListPK11EnvironmentS1_S1_S1_Ri+0x2c1) [0xa9bb9b]
/opt/intel/Compiler/11.0/074/bin/intel64/iidb(_ZN14ProcessFactory22createLocalLiveProcessEP3JobP7ArgListPcS4_S4_22IsOrIsNotADummyProcess+0x15e) [0x683b06]
/opt/intel/Compiler/11.0/074/bin/intel64/iidb(_ZN3Job22createLocalLiveProcessEP7ArgListPcS2_S2_22IsOrIsNotADummyProcess+0x2b) [0x6839a3]
/opt/intel/Compiler/11.0/074/bin/intel64/iidb(_ZN11SessionBase17startLocalProcessEP3JobP7ArgList22IsOrIsNotADummyProcess+0x9b) [0x660675]
/opt/intel/Compiler/11.0/074/bin/intel64/iidb(_Z17StartDummyProcessP3Job+0x5c) [0x7c6510]
/opt/intel/Compiler/11.0/074/bin/intel64/iidb(_ZN7CmdLoad5do_itER19CmdExecutionContextRN10BaseForCmd9CmdResultE+0x1fd) [0x7c4c5b]
/opt/intel/Compiler/11.0/074/bin/intel64/iidb(_ZN10BaseForCmd7executeEb+0xcce) [0x818ed0]
/opt/intel/Compiler/11.0/074/bin/intel64/iidb(_ZN3Cmd20executeAsInternalCmdEv+0x23) [0x7bf1f9]
/opt/intel/Compiler/11.0/074/bin/intel64/iidb(_ZN12InitialState10executeCmdEv+0x44) [0x8b1896]
/opt/intel/Compiler/11.0/074/bin/intel64/iidb(_Z15decladebug_initiPPKcS1_+0x80e) [0x5d1de2]
/opt/intel/Compiler/11.0/074/bin/intel64/iidb(_Z7idbMainiPPKcS1_+0x199) [0x5cec81]
/opt/intel/Compiler/11.0/074/bin/intel64/iidb(main+0x3c) [0x5cda40]
/lib/libc.so.6(__libc_start_main+0xf4) [0x7f3fb96b31c4]
/opt/intel/Compiler/11.0/074/bin/intel64/iidb(_ZStlsISt11char_traitsIcEERSt13basic_ostreamIcT_ES5_c+0x72) [0x5cd96a]

Is this a bug or have I done anything wrong?

Regards,

Georg

[maemarcus]

Same issue for C++ compiler 11 I've just get:

Assertion failed: "(*i)->getThreadId() != ThreadId()" src/appl/procs/addrspace.C:2819
This is an unexpected condition and may indicate the presence of a defect.
If you wish to report this, please include the stack trace that follows.
/opt/intel/Compiler/11.0/074/bin/intel64/iidb(_ZN15IDBAssertFailed3runEPKcS1_j+0xe) [0x5d3ca4]
/opt/intel/Compiler/11.0/074/bin/intel64/iidb(_ZN14DTLU_namespace12assertFailedEPKcS1_j+0x24) [0xb8210c]
/opt/intel/Compiler/11.0/074/bin/intel64/iidb(_ZN12AddrSpaceDef9getThreadERK8ThreadId+0x7b) [0x675ec3]
/opt/intel/Compiler/11.0/074/bin/intel64/iidb(_ZN10ProcessDef9getThreadERK8ThreadId+0x17) [0x68d56b]
/opt/intel/Compiler/11.0/074/bin/intel64/iidb(_ZN10ControlDef20forceThreadExistenceEP17LightweightWorkerR8ThreadIdRP6Thread+0x24d) [0xab1843]
/opt/intel/Compiler/11.0/074/bin/intel64/iidb(_ZN10ControlDef36getPrstatusLWPForcingThreadExistenceEv+0x4c) [0xaa5d1a]
/opt/intel/Compiler/11.0/074/bin/intel64/iidb(_ZN10ControlDef13getStoppedLWPEv+0x54) [0xab19a4]
/opt/intel/Compiler/11.0/074/bin/intel64/iidb(_ZN12AddrSpaceDef22getLatestStoppedThreadEv+0x64) [0x696caa]
/opt/intel/Compiler/11.0/074/bin/intel64/iidb(_ZN12AddrSpaceDef23setThreadDebugInterfaceE15EGIThreadFlavorb+0x104) [0x673be4]
/opt/intel/Compiler/11.0/074/bin/intel64/iidb(_ZN12AddrSpaceDef10noteLoadedEPK6FileIdP7Program+0x1f8) [0x6731c2]
/opt/intel/Compiler/11.0/074/bin/intel64/iidb(_ZN10ProcessDef10noteExecedEPKc+0x66) [0x68acce]
/opt/intel/Compiler/11.0/074/bin/intel64/iidb(_ZN12ControlDefUx10noteExecedEv+0x27) [0xa9e479]
/opt/intel/Compiler/11.0/074/bin/intel64/iidb(_ZN18ControlDefUxPtrace21startProcess_postExecEPK6FileId+0x1af) [0xaadd0d]
/opt/intel/Compiler/11.0/074/bin/intel64/iidb(_ZN12ControlDefUx12startProcessEPKcRK7ArgListPK11EnvironmentS1_S1_S1_Ri+0x2c1) [0xa9bb9b]
/opt/intel/Compiler/11.0/074/bin/intel64/iidb(_ZN14ProcessFactory22createLocalLiveProcessEP3JobP7ArgListPcS4_S4_22IsOrIsNotADummyProcess+0x15e) [0x683b06]
/opt/intel/Compiler/11.0/074/bin/intel64/iidb(_ZN3Job22createLocalLiveProcessEP7ArgListPcS2_S2_22IsOrIsNotADummyProcess+0x2b) [0x6839a3]
/opt/intel/Compiler/11.0/074/bin/intel64/iidb(_ZN11SessionBase17startLocalProcessEP3JobP7ArgList22IsOrIsNotADummyProcess+0x9b) [0x660675]
/opt/intel/Compiler/11.0/074/bin/intel64/iidb(_Z17StartDummyProcessP3Job+0x5c) [0x7c6510]
/opt/intel/Compiler/11.0/074/bin/intel64/iidb(_ZN7CmdLoad5do_itER19CmdExecutionContextRN10BaseForCmd9CmdResultE+0x1fd) [0x7c4c5b]
/opt/intel/Compiler/11.0/074/bin/intel64/iidb(_ZN10BaseForCmd7executeEb+0xcce) [0x818ed0]
/opt/intel/Compiler/11.0/074/bin/intel64/iidb [0x5cfcd1]
/opt/intel/Compiler/11.0/074/bin/intel64/iidb(_Z15ProcessCommandsv+0x47) [0x5cf211]
/opt/intel/Compiler/11.0/074/bin/intel64/iidb(_Z7idbMainiPPKcS1_+0x1b0) [0x5cec98]
/opt/intel/Compiler/11.0/074/bin/intel64/iidb(main+0x3c) [0x5cda40]
/lib64/libc.so.6(__libc_start_main+0xe6) [0x373981e546]
/opt/intel/Compiler/11.0/074/bin/intel64/iidb(_ZStlsISt11char_traitsIcEERSt13basic_ostreamIcT_ES5_c+0x72) [0x5cd96a]
Could not start process for
No image loaded ... Recovering ...

[Kevin_D_Intel]

The IDB Development team is investigating this issue now. In at least one other reported case, the issue appears to be related to compiling with openmp.

Do either of you compile with openmp, are you able to confirm your executable is loadable by IDB when *not* compiling with this option?

We apologize for the inconvenience associated with this issue.

(Internal ref. CQ-110688)

[leo_studer]

Quoting - Kevin Davis (Intel)

The IDB Development team is investigating this issue now. In at least one other reported case, the issue appears to be related to compiling with openmp.

Do either of you compile with openmp, are you able to confirm your executable is loadable by IDB when *not* compiling with this option?

We apologize for the inconvenience associated with this issue.

(Internal ref. CQ-110688)

This idb behaviour just appeared for me too. With an -openmp compiled fortran program... It appeared suddenly as I was able to debug the same program beforehand. Turning the -openmp option off does not change that behaviour, I still get the same error message.

Here is the solution: I happened in fact to change my compilation to a -static one. Removing the -static option allows to debug the program with or without -openmp.

I only have 2 days left for my idb evaluation period, so far this does not makes me feel like buying the license !

[TimP]

Quoting - leo.studer

Here is the solution: I happened in fact to change my compilation to a -static one. Removing the -static option allows to debug the program with or without -openmp.

I only have 2 days left for my idb evaluation period, so far this does not makes me feel like buying the license !

I don't find any problem with -static -openmp in my case (SuSE, not even a supported configuration). If you have such a problem in a supported configuration, please submit your problem report on premier.intel.com. Maybe you will get e-mail about any progress. If you do buy a standard license, you are entitled to download all upgrades for a year.

I find it possible to crash idb sometimes, even to the extent of having to resort to the power off button, by attempting to open the Parallel options prematurely. What seems more successful is to run to a breakpoint before attempting Parallel options. To me, this wasn't intuitive, and it seems that idb ought to be made more bullet proof.

idb Intel64 didn't support -openmp until the 11.0/074 update. If you didn't get 074 for the evaluation, you might request a renewal of evaluation when you submit a problem report.

[maemarcus]

Quoting - Kevin Davis (Intel)

The IDB Development team is investigating this issue now. In at least one other reported case, the issue appears to be related to compiling with openmp.

Do either of you compile with openmp, are you able to confirm your executable is loadable by IDB when *not* compiling with this option?

We apologize for the inconvenience associated with this issue.

(Internal ref. CQ-110688)

Hello,Kevin,

Thanks for reply,

I do compile my program with both -openmp and -static for fortran symbols, i.e. linking -lifcore with icpc for cross-language project, and for me turning off -openmp makes idb to work.

[Kevin_D_Intel]

Thank you maemarcus for the update on your situation. We will provide more information in this thread as we learn it.

Leo, thank youfor the additional note about the possible relationship to static either with or without openmp. I notified the IDB developer investigating this issue of that. There have been other reports of assertions occurring with IDB 11.0 and in absence of seeing your actual assertion and trace, I would like to confirm, does your instance occur in src/appl/procs/addrspace.C:2819?

As tim18 suggested, you can share the complete trace with us in a problem report to Intel Premier where we can also discuss renewing the evaluation to assist with you completing the evaluation of the Intel Fortran compiler and debugger.

[leo_studer]

Quoting - Kevin Davis (Intel)

Thank you maemarcus for the update on your situation. We will provide more information in this thread as we learn it.

Leo, thank youfor the additional note about the possible relationship to static either with or without openmp. I notified the IDB developer investigating this issue of that. There have been other reports of assertions occurring with IDB 11.0 and in absence of seeing your actual assertion and trace, I would like to confirm, does your instance occur in src/appl/procs/addrspace.C:2819?

As tim18 suggested, you can share the complete trace with us in a problem report to Intel Premier where we can also discuss renewing the evaluation to assist with you completing the evaluation of the Intel Fortran compiler and debugger.

Here is the corresponding trace with -static enabled (again, does not accurs without -static- and with -openmp) :

Reading symbols from /media/stock/LagaProgs_GNU_Linux/simulations/zarya.meca.iteration.8/zarya_tmp/lagarun...done.
Assertion failed: "(*i)->getThreadId() != ThreadId()" src/appl/procs/addrspace.C:2819
This is an unexpected condition and may indicate the presence of a defect.
If you wish to report this, please include the stack trace that follows.
/opt/intel/Compiler/11.0/074/bin/intel64/iidb(_ZN15IDBAssertFailed3runEPKcS1_j+0xe) [0x5d3ca4]
/opt/intel/Compiler/11.0/074/bin/intel64/iidb(_ZN14DTLU_namespace12assertFailedEPKcS1_j+0x24) [0xb8210c]
/opt/intel/Compiler/11.0/074/bin/intel64/iidb(_ZN12AddrSpaceDef9getThreadERK8ThreadId+0x7b) [0x675ec3]
/opt/intel/Compiler/11.0/074/bin/intel64/iidb(_ZN10ProcessDef9getThreadERK8ThreadId+0x17) [0x68d56b]
/opt/intel/Compiler/11.0/074/bin/intel64/iidb(_ZN10ControlDef20forceThreadExistenceEP17LightweightWorkerR8ThreadIdRP6Thread+0x24d) [0xab1843]
/opt/intel/Compiler/11.0/074/bin/intel64/iidb(_ZN10ControlDef36getPrstatusLWPForcingThreadExistenceEv+0x4c) [0xaa5d1a]
/opt/intel/Compiler/11.0/074/bin/intel64/iidb(_ZN10ControlDef13getStoppedLWPEv+0x54) [0xab19a4]
/opt/intel/Compiler/11.0/074/bin/intel64/iidb(_ZN12AddrSpaceDef22getLatestStoppedThreadEv+0x64) [0x696caa]
/opt/intel/Compiler/11.0/074/bin/intel64/iidb(_ZN12AddrSpaceDef23setThreadDebugInterfaceE15EGIThreadFlavorb+0x104) [0x673be4]
/opt/intel/Compiler/11.0/074/bin/intel64/iidb(_ZN12AddrSpaceDef10noteLoadedEPK6FileIdP7Program+0x1f8) [0x6731c2]
/opt/intel/Compiler/11.0/074/bin/intel64/iidb(_ZN10ProcessDef10noteExecedEPKc+0x66) [0x68acce]
/opt/intel/Compiler/11.0/074/bin/intel64/iidb(_ZN12ControlDefUx10noteExecedEv+0x27) [0xa9e479]
/opt/intel/Compiler/11.0/074/bin/intel64/iidb(_ZN18ControlDefUxPtrace21startProcess_postExecEPK6FileId+0x1af) [0xaadd0d]
/opt/intel/Compiler/11.0/074/bin/intel64/iidb(_ZN12ControlDefUx12startProcessEPKcRK7ArgListPK11EnvironmentS1_S1_S1_Ri+0x2c1) [0xa9bb9b]
/opt/intel/Compiler/11.0/074/bin/intel64/iidb(_ZN14ProcessFactory22createLocalLiveProcessEP3JobP7ArgListPcS4_S4_22IsOrIsNotADummyProcess+0x15e) [0x683b06]
/opt/intel/Compiler/11.0/074/bin/intel64/iidb(_ZN3Job22createLocalLiveProcessEP7ArgListPcS2_S2_22IsOrIsNotADummyProcess+0x2b) [0x6839a3]
/opt/intel/Compiler/11.0/074/bin/intel64/iidb(_ZN11SessionBase17startLocalProcessEP3JobP7ArgList22IsOrIsNotADummyProcess+0x9b) [0x660675]
/opt/intel/Compiler/11.0/074/bin/intel64/iidb(_Z17StartDummyProcessP3Job+0x5c) [0x7c6510]
/opt/intel/Compiler/11.0/074/bin/intel64/iidb(_ZN7CmdLoad5do_itER19CmdExecutionContextRN10BaseForCmd9CmdResultE+0x1fd) [0x7c4c5b]
/opt/intel/Compiler/11.0/074/bin/intel64/iidb(_ZN10BaseForCmd7executeEb+0xcce) [0x818ed0]
/opt/intel/Compiler/11.0/074/bin/intel64/iidb(_ZN12IDBEngineCmd7executeEv+0x1b8) [0x932368]
/opt/intel/Compiler/11.0/074/idb/lib/intel64/libDebuggerServices.so(_ZN4IUDG2DS16DirectiveHandler10processMsgEPNS_15MSGCLASSFACTORY9ClientMsgE+0x160) [0x2aae534edad0]
/opt/intel/Compiler/11.0/074/idb/lib/intel64/libDebuggerServices.so(_ZN4IUDG2DS10Dispatcher11dispatchMsgEPNS_15MSGCLASSFACTORY9ClientMsgE+0x3eb) [0x2aae534eff8f]
/opt/intel/Compiler/11.0/074/idb/lib/intel64/libDebuggerServices.so(_ZN4IUDG2DS12DSPostOffice13flushMsgQueueEv+0x6d) [0x2aae534f463b]
/opt/intel/Compiler/11.0/074/bin/intel64/iidb(_ZN8IudgGlue18processPendingMsgsEb+0x159) [0x91cb7f]
/opt/intel/Compiler/11.0/074/bin/intel64/iidb(vfprintf+0x3ab6) [0x5cf57e]
/opt/intel/Compiler/11.0/074/bin/intel64/iidb(_Z15ProcessCommandsv+0x47) [0x5cf211]
/opt/intel/Compiler/11.0/074/bin/intel64/iidb(_Z7idbMainiPPKcS1_+0x1b0) [0x5cec98]
/opt/intel/Compiler/11.0/074/bin/intel64/iidb(main+0x3c) [0x5cda40]
/lib/libc.so.6(__libc_start_main+0xe6) [0x2aae52877466]
/opt/intel/Compiler/11.0/074/bin/intel64/iidb(_ZStlsISt11char_traitsIcEERSt13basic_ostreamIcT_ES5_c+0x72) [0x5cd96a]
Could not start process for /media/stock/LagaProgs_GNU_Linux/simulations/zarya.meca.iteration.8/zarya_tmp/lagarun
No image loaded ... Recovering ...

And I am using Ubuntu 8.10 which is not a supported OS by the way.

[Kevin_D_Intel]

The IDB development team indicates that on Fedora Core 9 (and other new Linux distros), thread_db can use 0 for the id of a thread, however, IDB insists on a non-zero thread_id, hence the assert.

There is no convenient work-around available, unless one has an alternative platform available to use in the interim, or can proceed with development without using either openmp or static as others reported induced the error.

The issue will be fixed in a future IDB update early next year.