Internal compiler error: segmentation violation signal raised

[stdweird]

i am compiling CP2K using latest intel fortran/mpi/mkl and i got the following error message.

: catastrophic error: **Internal compiler error: segmentation violation signal raised** Please report this error along with the circumstances in which it occurred in a Software Problem Report. Note: File and line given may not be explicit cause of this error.

as i'm not sure what a Software Problem Report is, i'm posting it here, hoping for some developers to pick this up.

OS is SL5.3 x86_64 (RHEL5.3 clone)

-bash-3.2$ ifort -V
Intel Fortran Intel 64 Compiler Professional for applications running on Intel 64, Version 11.1 Build 20090630 Package ID: l_cprof_p_11.1.046
Copyright (C) 1985-2009 Intel Corporation. All rights reserved.

i can reproduce the error with minimal commandline (but it needs some parts of the CP2K build to reproduce it)

-bash-3.2$ ifort -free ../../../src/semi_empirical_int_utils.F
: catastrophic error: **Internal compiler error: segmentation violation signal raised** Please report this error along with the circumstances in which it occurred in a Software Problem Report. Note: File and line given may not be explicit cause of this error.

compilation aborted for ../../../src/semi_empirical_int_utils.F (code 3)


to reproduce:
- setup sl5.3 (or RHEL or centos) x86_64
- install ifort 11.1.046
- download http://users.ugent.be/~stdweird/cp2k-20090728.intelerror.tar.bz2 (it's 13MB compressed, and 180MB uncompressed)
- do
tar xjf cp2k-20090728.intelerror.tar.bz2
cd cp2k-20090728/obj/Linux-x86-64-ictce/popt/
ifort -free ../../../src/semi_empirical_int_utils.F

and that should do the trick

[Kevin_D_Intel]

Software Problem Reports are filed through our Intel Premier Support service (here). The Intel Premier Support FAQ offers more info, butreporting in this forum works too.

Appreciate your extra efforts to create the reproducer. I'll have a look at that and post again when I know more.

[Kevin_D_Intel]

It took a bit of time, but I was able to create a smaller reproducing test case for Development to investigate this further. The issue appears to occur in the Front-end in relation to the sophisticated derived types that exist. It occursall optimization levels,therefore, I do not have a work around to offer at this time.

The development compiler branch does not exhibit the error, so there's potentially already a fix available that perhaps can be made available in a future 11.1 update. I will keep the thread updated as I learn more.

(Internal tracking id: DPD200138652)

(Resolution Update on 3/30/2010): This defect is fixed in the Intel Fortran Compiler Professional Edition 11.1 Update 5 (11.1.069 - Linux).

[Lorri_M_Intel]

Kevin is right - it should be available in a future update.


- Lorri