ICE with 11.1.059

ICE with 11.1.059

Just came across this ICE, which happened both with 11.1.059 and 11.1.046. Not sure what I did that triggered it... I was in the process of changing a bunch of dummy arguments from automatic to allocatable deferred arrays. The full command was:
[shell]ifort -I. -Isrc/ -Isrc2/ -fpp -warn all -fpp -check all -gen-interfaces -assume realloc_lhs -static -pg -warn notruncated_source -debug full -traceback -O0 -g -FR -c -o lmain.o /gusr/alr99/cluster/workspace/frealix/src/lmain.f90[/shell] Not sure what info might be needed... I attach the lmain.f90 file (sorry, it's pretty ugly in there!)

AttachmentSize
Downloadapplication/octet-stream lmain.f9043.21 KB
18 posts / 0 new
Last post
For more complete information about compiler optimizations, see our Optimization Notice.

Quoting - Ronald W. Green (Intel)

bug report DPD200139107 - fixed in 11.1 Update 4 due in December.

We're in luck - there were several bug reports against -realloc_lhs that were entered in the past months. There is a fix coming out in a few weeks in the "Update 4" release of the compiler (exact date TBD - but it should in just about 2-3 weeks barring any unforseen delays).

I have tested your code with an engineering build version of this compiler and have confirmed it fixes your bug too.

I will alert you when this fixed compiler is ready for download.

ron

Many thanks, Ron. I do need -realloc_lhs there, although I'd be hard pressed to remember why right now. :)

Doh! Forgot to copy the output from ifort... here goes...

[shell][alr99@kiev debug_intel]$ ifort -I. -Isrc/ -Isrc2/ -fpp -warn all -fpp -check all -gen-interfaces -assume realloc_lhs -static -pg -warn notruncated_source -debug full -traceback -O0 -g -FR -c -o lmain.o /gusr/alr99/cluster/workspace/frealix/src/lmain.f90
/gusr/alr99/cluster/workspace/frealix/src/lmain.f90(195): remark #7712: This variable has not been used.   [IS]
        INTEGER ::      I, ID, IDEBUG, IFLIP, IMASZE, IS, ITF, IXC, IYC,                        &
------------------------------------------------------^
/gusr/alr99/cluster/workspace/frealix/src/lmain.f90(35): remark #7712: This variable has not been used.   [RBINS]
                                        C3DV,C3DF,C3DS,RBINS,RBINN,NSET,PRESA,                          &
-------------------------------------------------------^
/gusr/alr99/cluster/workspace/frealix/src/lmain.f90(35): remark #7712: This variable has not been used.   [RBINN]
                                        C3DV,C3DF,C3DS,RBINS,RBINN,NSET,PRESA,                          &
-------------------------------------------------------------^
/gusr/alr99/cluster/workspace/frealix/src/lmain.f90(44): remark #7712: This variable has not been used.   [WGH]
                                        IFIRST,IOPROJ,IEWALD,WGH,                                                       &
-------------------------------------------------------------^
/gusr/alr99/cluster/workspace/frealix/src/lmain.f90(47): remark #7712: This variable has not been used.   [FINPAT2]
                                        ISTART,FINPAT1,FINPAT2,ASYM,VX,ILAST,                           &
-------------------------------------------------------^
/gusr/alr99/cluster/workspace/frealix/src/lmain.f90(48): remark #7712: This variable has not been used.   [NNSET]
                                        NNSET,CDATE,CTIME,CZONE,DTVAL,SINCLUT,MBUF,                     &
----------------------------------------^
/gusr/alr99/cluster/workspace/frealix/src/lmain.f90(177): remark #7712: This variable has not been used.   [TESTNM]
        CHARACTER(LEN=20)       ::      TESTNM
----------------------------------------^
/gusr/alr99/cluster/workspace/frealix/src/lmain.f90(178): remark #7712: This variable has not been used.   [IB1]
        INTEGER IB1,IB2,JC,NSAMH
----------------^
/gusr/alr99/cluster/workspace/frealix/src/lmain.f90(178): remark #7712: This variable has not been used.   [IB2]
        INTEGER IB1,IB2,JC,NSAMH
--------------------^
/gusr/alr99/cluster/workspace/frealix/src/lmain.f90(178): remark #7712: This variable has not been used.   [JC]
        INTEGER IB1,IB2,JC,NSAMH
------------------------^
/gusr/alr99/cluster/workspace/frealix/src/lmain.f90(178): remark #7712: This variable has not been used.   [NSAMH]
        INTEGER IB1,IB2,JC,NSAMH
---------------------------^
/gusr/alr99/cluster/workspace/frealix/src/lmain.f90(189): remark #7712: This variable has not been used.   [CC3M]
                        WLN,AAOLD(NNP),AIPOLD(NNP),CC,CANG,CC3M,                                                &
-----------------------------------------------------------^
/gusr/alr99/cluster/workspace/frealix/src/lmain.f90(191): remark #7712: This variable has not been used.   [XM]
                        THETAMID,STHETA,THETAK,THETATR,XM,YM,XXK,YYK, TEMP
-------------------------------------------------------^
/gusr/alr99/cluster/workspace/frealix/src/lmain.f90(191): remark #7712: This variable has not been used.   [YM]
                        THETAMID,STHETA,THETAK,THETATR,XM,YM,XXK,YYK, TEMP
----------------------------------------------------------^
/gusr/alr99/cluster/workspace/frealix/src/lmain.f90(195): remark #7712: This variable has not been used.   [ID]
        INTEGER ::      I, ID, IDEBUG, IFLIP, IMASZE, IS, ITF, IXC, IYC,                        &
---------------------------^
/gusr/alr99/cluster/workspace/frealix/src/lmain.f90(195): remark #7712: This variable has not been used.   [IFLIP]
        INTEGER ::      I, ID, IDEBUG, IFLIP, IMASZE, IS, ITF, IXC, IYC,                        &
---------------------------------------^
/gusr/alr99/cluster/workspace/frealix/src/lmain.f90(195): remark #7712: This variable has not been used.   [IMASZE]
        INTEGER ::      I, ID, IDEBUG, IFLIP, IMASZE, IS, ITF, IXC, IYC,                        &
----------------------------------------------^
/gusr/alr99/cluster/workspace/frealix/src/lmain.f90(196): remark #7712: This variable has not been used.   [JJ]
                                J, JJ, KK, L, LASTF, LL, M, MAXR1, MAXR2, MAXRES, MM,           &
-----------------------------------^
/gusr/alr99/cluster/workspace/frealix/src/lmain.f90(196): remark #7712: This variable has not been used.   [L]
                                J, JJ, KK, L, LASTF, LL, M, MAXR1, MAXR2, MAXRES, MM,           &
-------------------------------------------^
/gusr/alr99/cluster/workspace/frealix/src/lmain.f90(196): remark #7712: This variable has not been used.   [LASTF]
                                J, JJ, KK, L, LASTF, LL, M, MAXR1, MAXR2, MAXRES, MM,           &
----------------------------------------------^
/gusr/alr99/cluster/workspace/frealix/src/lmain.f90(196): remark #7712: This variable has not been used.   [LL]
                                J, JJ, KK, L, LASTF, LL, M, MAXR1, MAXR2, MAXRES, MM,           &
-----------------------------------------------------^
/gusr/alr99/cluster/workspace/frealix/src/lmain.f90(196): remark #7712: This variable has not been used.   [MAXRES]
                                J, JJ, KK, L, LASTF, LL, M, MAXR1, MAXR2, MAXRES, MM,           &
--------------------------------------------------------------------------^
/gusr/alr99/cluster/workspace/frealix/src/lmain.f90(196): remark #7712: This variable has not been used.   [MM]
                                J, JJ, KK, L, LASTF, LL, M, MAXR1, MAXR2, MAXRES, MM,           &
----------------------------------------------------------------------------------^
/gusr/alr99/cluster/workspace/frealix/src/lmain.f90(197): remark #7712: This variable has not been used.   [N1]
                                N1,N2,N3,NSEG
--------------------------------^
/gusr/alr99/cluster/workspace/frealix/src/lmain.f90(197): remark #7712: This variable has not been used.   [N2]
                                N1,N2,N3,NSEG
-----------------------------------^
/gusr/alr99/cluster/workspace/frealix/src/lmain.f90(204): remark #7712: This variable has not been used.   [CCIMG]
        REAL    ::      CCOEF, CCIMG,CCLIN
-------------------------------^
/gusr/alr99/cluster/workspace/frealix/src/lmain.f90(204): remark #7712: This variable has not been used.   [CCLIN]
        REAL    ::      CCOEF, CCIMG,CCLIN
-------------------------------------^
0_0

: catastrophic error: **Internal compiler error: internal abort** 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 /gusr/alr99/cluster/workspace/frealix/src/lmain.f90 (code 3)[/shell]

OK, thanks to gfortran, I found the cause of the ICE: on line 132 of the file, scalar variables were being declared as ALLOCATABLE (variables DPRESS,OCCS,OALL).

gfortran gave the helpful message: Error: Scalar object 'dpress' at (1) may not be ALLOCATABLE

Guess I should have checked with gfortran before posting here. Sorry about the verbose report !

We would need the source code for your modules to investigate this.

We support this feature of Fortran 2003 (allocatable scalars), gfortran does not. It should work.

Steve - Intel Developer Support

Quoting - tim18
We would need the source code for your modules to investigate this.

Attached are the sources for the two modules USEd in the subroutine.

I am happy to upload more if required.

Attachments: 

AttachmentSize
Downloadapplication/octet-stream global_parameters.f90680 bytes
Downloadapplication/octet-stream iof2.f903.35 KB

Quoting - A. Rohou

Attached are the sources for the two modules USEd in the subroutine.

According to ifort -diag-enable sc :
lmain.f90(280): error #12144: "KLAST" is possibly uninitialized
lmain.f90(341): error #12084: interface is needed or it is incorrect
lmain.f90(914): error #12144: "SIG2" is possibly uninitialized
lmain.f90(960): error #12144: "DSHX" is possibly uninitialized
lmain.f90(961): error #12144: "DSHY" is possibly uninitialized

and, according to gfortran:
lmain.f90:391.16:

WRITE(*,'(A,F6.2,X,F6.2)') '**debug(LMAIN): before TRACE: AIP, PSI_TR(1
1
Warning: Extension: X descriptor requires leading space count at (1)
lmain.f90:394.10:

WRITE(*,'(A,F6.2,X,F6.2)') '**debug(LMAIN): after TRACE: AIP, PSI_TR(1): ',
1
Warning: Extension: X descriptor requires leading space count at (1)

and the warning about klast is confirmed, along with tons of unused variable and tab concerns.

I suppose implicit none should allay concerns about the unused variable warnings.

Thank tim18,

I'm in the process of tidying up these files - I appreciate the pointers. I was not even aware of the diag-enable option.

Quoting - A. Rohou

Attached are the sources for the two modules USEd in the subroutine.

I am happy to upload more if required.

OK, this is interesting - I have tried to reproduce this with the files you posted, with the 11.1.059 and 11.1.046, on BOTH Intel 64 and IA-32 (64 bit and 32 bit systems). I cannot reproduce the Internal compiler error, although my error messages match your EXACTLY.

I have seen something like this just once before but could not reproduce it: I noticed your source files appear to be on a cluster global file system. Would this perhaps be a Panasas file system, or Lustre, or IBM GPFS, PVFS, etc? I had colleagues at Sandia Nat'l Labs that were getting a strange compiler error like this when they compiled on their parallel file system (I think it was Panasas) but I could not reproduce it since we don't use any of these internally.

If this is the case, try moving the source to local disk and let me know what you see. Here is my output with 11.1.059 which is the same as 11.1.046:

rwgreen@spdr65:~/quad/rwgreen/forums/70115> ifort -I. -fpp -warn all -fpp -check all -gen-interfaces -assume realloc_lhs -static -pg -warn notruncated_source -debug full -traceback -O0 -g -FR -c  lmain.f90
lmain.f90(58): error #7002: Error in opening the compiled module file.  Check INCLUDE paths.   [GLOBAL_PARAMETERS]
	USE GLOBAL_PARAMETERS
------------^
lmain.f90(59): error #7002: Error in opening the compiled module file.  Check INCLUDE paths.   [MOD_IOF]
	USE MOD_IOF
------------^
lmain.f90(363): error #6404: This name does not have a type, and must have an explicit type.   [INPUT_RECDIM]
								INPUT_RECDIM,BOXDIM,	&
----------------------------------------------------------------^
lmain.f90(363): error #6404: This name does not have a type, and must have an explicit type.   [BOXDIM]
								INPUT_RECDIM,BOXDIM,	&
-----------------------------------------------------------------------------^
lmain.f90(1083): error #6404: This name does not have a type, and must have an explicit type.   [RECDIM]
			THETATR=WL(NSET)/(PSIZE*RECDIM)
------------------------------------------------^
lmain.f90(195): remark #7712: This variable has not been used.   [IS]
	INTEGER :: 	I, ID, IDEBUG, IFLIP, IMASZE, IS, ITF, IXC, IYC,			&
------------------------------------------------------^
lmain.f90(35): remark #7712: This variable has not been used.   [RBINS]
					C3DV,C3DF,C3DS,RBINS,RBINN,NSET,PRESA,				&
-------------------------------------------------------^
lmain.f90(35): remark #7712: This variable has not been used.   [RBINN]
					C3DV,C3DF,C3DS,RBINS,RBINN,NSET,PRESA,				&
-------------------------------------------------------------^
lmain.f90(44): remark #7712: This variable has not been used.   [WGH]
					IFIRST,IOPROJ,IEWALD,WGH,							&
-------------------------------------------------------------^
lmain.f90(47): remark #7712: This variable has not been used.   [FINPAT2]
					ISTART,FINPAT1,FINPAT2,ASYM,VX,ILAST,				&
-------------------------------------------------------^
lmain.f90(48): remark #7712: This variable has not been used.   [NNSET]
					NNSET,CDATE,CTIME,CZONE,DTVAL,SINCLUT,MBUF,			&
----------------------------------------^
lmain.f90(177): remark #7712: This variable has not been used.   [TESTNM]
	CHARACTER(LEN=20)	::	TESTNM
----------------------------------------^
lmain.f90(178): remark #7712: This variable has not been used.   [IB1]
	INTEGER IB1,IB2,JC,NSAMH
----------------^
lmain.f90(178): remark #7712: This variable has not been used.   [IB2]
	INTEGER IB1,IB2,JC,NSAMH
--------------------^
lmain.f90(178): remark #7712: This variable has not been used.   [JC]
	INTEGER IB1,IB2,JC,NSAMH
------------------------^
lmain.f90(178): remark #7712: This variable has not been used.   [NSAMH]
	INTEGER IB1,IB2,JC,NSAMH
---------------------------^
lmain.f90(189): remark #7712: This variable has not been used.   [CC3M]
			WLN,AAOLD(NNP),AIPOLD(NNP),CC,CANG,CC3M,						&
-----------------------------------------------------------^
lmain.f90(191): remark #7712: This variable has not been used.   [XM]
			THETAMID,STHETA,THETAK,THETATR,XM,YM,XXK,YYK, TEMP
-------------------------------------------------------^
lmain.f90(191): remark #7712: This variable has not been used.   [YM]
			THETAMID,STHETA,THETAK,THETATR,XM,YM,XXK,YYK, TEMP
----------------------------------------------------------^
lmain.f90(195): remark #7712: This variable has not been used.   [ID]
	INTEGER :: 	I, ID, IDEBUG, IFLIP, IMASZE, IS, ITF, IXC, IYC,			&
---------------------------^
lmain.f90(195): remark #7712: This variable has not been used.   [IFLIP]
	INTEGER :: 	I, ID, IDEBUG, IFLIP, IMASZE, IS, ITF, IXC, IYC,			&
---------------------------------------^
lmain.f90(195): remark #7712: This variable has not been used.   [IMASZE]
	INTEGER :: 	I, ID, IDEBUG, IFLIP, IMASZE, IS, ITF, IXC, IYC,			&
----------------------------------------------^
lmain.f90(196): remark #7712: This variable has not been used.   [JJ]
				J, JJ, KK, L, LASTF, LL, M, MAXR1, MAXR2, MAXRES, MM,		&
-----------------------------------^
lmain.f90(196): remark #7712: This variable has not been used.   [L]
				J, JJ, KK, L, LASTF, LL, M, MAXR1, MAXR2, MAXRES, MM,		&
-------------------------------------------^
lmain.f90(196): remark #7712: This variable has not been used.   [LASTF]
				J, JJ, KK, L, LASTF, LL, M, MAXR1, MAXR2, MAXRES, MM,		&
----------------------------------------------^
lmain.f90(196): remark #7712: This variable has not been used.   [LL]
				J, JJ, KK, L, LASTF, LL, M, MAXR1, MAXR2, MAXRES, MM,		&
-----------------------------------------------------^
lmain.f90(196): remark #7712: This variable has not been used.   [MAXRES]
				J, JJ, KK, L, LASTF, LL, M, MAXR1, MAXR2, MAXRES, MM,		&
--------------------------------------------------------------------------^
lmain.f90(196): remark #7712: This variable has not been used.   [MM]
				J, JJ, KK, L, LASTF, LL, M, MAXR1, MAXR2, MAXRES, MM,		&
----------------------------------------------------------------------------------^
lmain.f90(197): remark #7712: This variable has not been used.   [N1]
				N1,N2,N3,NSEG
--------------------------------^
lmain.f90(197): remark #7712: This variable has not been used.   [N2]
				N1,N2,N3,NSEG
-----------------------------------^
lmain.f90(204): remark #7712: This variable has not been used.   [CCIMG]
	REAL	::	CCOEF, CCIMG,CCLIN
-------------------------------^
lmain.f90(204): remark #7712: This variable has not been used.   [CCLIN]
	REAL	::	CCOEF, CCIMG,CCLIN
-------------------------------------^
compilation aborted for lmain.f90 (code 1)
rwgreen@spdr65:~/quad/rwgreen/forums/70115> 

and my compile line:

ifort -I. -fpp -warn all -fpp -check all -gen-interfaces -assume realloc_lhs -static -pg -warn notruncated_source -debug full -traceback -O0 -g -FR -c lmain.f90

Quoting - Ronald W. Green (Intel)

OK, this is interesting - I have tried to reproduce this with the files you posted, with the 11.1.059 and 11.1.046, on BOTH Intel 64 and IA-32 (64 bit and 32 bit systems). I cannot reproduce the Internal compiler error, although my error messages match your EXACTLY.

I have seen something like this just once before but could not reproduce it: I noticed your source files appear to be on a cluster global file system. Would this perhaps be a Panasas file system, or Lustre, or IBM GPFS, PVFS, etc? I had colleagues at Sandia Nat'l Labs that were getting a strange compiler error like this when they compiled on their parallel file system (I think it was Panasas) but I could not reproduce it since we don't use any of these internally.

If this is the case, try moving the source to local disk and let me know what you see. Here is my output with 11.1.059 which is the same as 11.1.046:

OK I've done as suggested, using the files as posted and in a fresh directory, either on my normal disk (I've asked the sysadmin what type that is - I'll let you know when I find out) or on a local scratch disk, which has an ext3 FS.

Attached are the outputs, but as far as I can tell the same ICE happens in both cases.

Do let me know what other info I can provide to try to get to the bottom of this!

Attachments: 

AttachmentSize
Downloadapplication/octet-stream ice_gusr.log4.62 KB
Downloadapplication/octet-stream ice_scratch.log4.62 KB

Let's try a few more things, since it appears to be something unique to your environment. Do you source the ifortvars.sh environment script directly, or is it part of your login, or is it under MODULE or some other environment setting tool?

Try explicitly sourcing ifortvars.sh explicitly: source /
then what is the output of ifort -V

And I note that you have a couple of -I paths that I don't: ./src and /src2 or something. Try removing those and using exactly my compile line.

And add -dryrun to the end of your compile line (and double check that my compile line is exactly what you use - I added -o lmain like yours but still no error here)

Here is my -dryrun output - yours will be different as we have our compilers in a non-default path:

rwgreen@spdr65:~/quad/rwgreen/forums/70115> ifort -I. -fpp -warn all -fpp -check all -gen-interfaces -assume realloc_lhs -static -pg -warn notruncated_source -debug full -traceback -O0 -g -FR -c -o lmain  lmain.f90 -dryrun
/opt/spdtools/compiler/cpro/Compiler/11.1/046/bin/intel64/fpp  
    -D__INTEL_COMPILER=1110 
    -D_MT 
    -D__ELF__ 
    -D__INTEL_COMPILER_BUILD_DATE=20090630 
    -D__PTRDIFF_TYPE__=long 
    "-D__SIZE_TYPE__=unsigned long" 
    -D__WCHAR_TYPE__=int 
    "-D__WINT_TYPE__=unsigned int" 
    "-D__INTMAX_TYPE__=long int" 
    "-D__UINTMAX_TYPE__=long unsigned int" 
    -D__LONG_MAX__=9223372036854775807L 
    -D__unix__ 
    -D__unix 
    -D__linux__ 
    -D__linux 
    -D__gnu_linux__ 
    -Dunix 
    -Dlinux 
    -D__x86_64 
    -D__x86_64__ 
    -I. 
    -I. 
    -I/opt/spdtools/compiler/cpro/Compiler/11.1/046/include/intel64 
    -I/opt/spdtools/compiler/cpro/Compiler/11.1/046/mkl/include 
    -I/opt/spdtools/compiler/cpro/Compiler/11.1/046/mkl/include 
    -I/opt/spdtools/compiler/cpro/Compiler/11.1/059/mkl/include 
    -I/opt/spdtools/compiler/cpro/Compiler/11.1/059/mkl/include 
    -I/opt/spdtools/compiler/cpro/Compiler/11.0/069/mkl/include 
    -I/opt/spdtools/compiler/cpro/Compiler/11.0/069/mkl/include 
    -I/opt/spdtools/compiler/cpro/Compiler/11.1/038/mkl/include 
    -I/opt/spdtools/compiler/cpro/Compiler/11.1/038/mkl/include 
    -I/opt/spdtools/compiler/cpro/Compiler/11.1/059-dbg/mkl/include 
    -I/opt/spdtools/compiler/cpro/Compiler/11.1/059-dbg/mkl/include 
    -I/opt/spdtools/compiler/cpro/Compiler/11.1/059/mkl/include 
    -I/opt/spdtools/compiler/cpro/Compiler/11.1/059/mkl/include 
    -I/opt/spdtools/compiler/cpro/Compiler/11.1/059-dbg/mkl/include 
    -I/opt/spdtools/compiler/cpro/Compiler/11.1/059-dbg/mkl/include 
    -I/opt/spdtools/compiler/cpro/Compiler/11.1/059/mkl/include 
    -I/opt/spdtools/compiler/cpro/Compiler/11.1/059/mkl/include 
    -I/opt/spdtools/compiler/cpro/Compiler/11.1/046/include/intel64 
    -I/opt/spdtools/compiler/cpro/Compiler/11.1/046/include 
    -I/usr/local/include 
    -I/usr/include 
    -I/usr/lib64/gcc/x86_64-suse-linux/4.1.2/include 
    -free 
    -4Ycpp 
    -4Ncvf 
    -f_com=yes 
    lmain.f90 
    /tmp/ifortb3BXwa.i90

/opt/spdtools/compiler/cpro/Compiler/11.1/046/bin/intel64/fortcom  
    -D__INTEL_COMPILER=1110 
    -D_MT 
    -D__ELF__ 
    -D__INTEL_COMPILER_BUILD_DATE=20090630 
    -D__unix__ 
    -D__unix 
    -D__linux__ 
    -D__linux 
    -D__gnu_linux__ 
    -Dunix 
    -Dlinux 
    -D__x86_64 
    -D__x86_64__ 
    -mGLOB_pack_sort_init_list 
    -I. 
    -I. 
    -I/opt/spdtools/compiler/cpro/Compiler/11.1/046/include/intel64 
    -I/opt/spdtools/compiler/cpro/Compiler/11.1/046/mkl/include 
    -I/opt/spdtools/compiler/cpro/Compiler/11.1/046/mkl/include 
    -I/opt/spdtools/compiler/cpro/Compiler/11.1/059/mkl/include 
    -I/opt/spdtools/compiler/cpro/Compiler/11.1/059/mkl/include 
    -I/opt/spdtools/compiler/cpro/Compiler/11.0/069/mkl/include 
    -I/opt/spdtools/compiler/cpro/Compiler/11.0/069/mkl/include 
    -I/opt/spdtools/compiler/cpro/Compiler/11.1/038/mkl/include 
    -I/opt/spdtools/compiler/cpro/Compiler/11.1/038/mkl/include 
    -I/opt/spdtools/compiler/cpro/Compiler/11.1/059-dbg/mkl/include 
    -I/opt/spdtools/compiler/cpro/Compiler/11.1/059-dbg/mkl/include 
    -I/opt/spdtools/compiler/cpro/Compiler/11.1/059/mkl/include 
    -I/opt/spdtools/compiler/cpro/Compiler/11.1/059/mkl/include 
    -I/opt/spdtools/compiler/cpro/Compiler/11.1/059-dbg/mkl/include 
    -I/opt/spdtools/compiler/cpro/Compiler/11.1/059-dbg/mkl/include 
    -I/opt/spdtools/compiler/cpro/Compiler/11.1/059/mkl/include 
    -I/opt/spdtools/compiler/cpro/Compiler/11.1/059/mkl/include 
    -I/opt/spdtools/compiler/cpro/Compiler/11.1/046/include/intel64 
    -I/opt/spdtools/compiler/cpro/Compiler/11.1/046/include 
    -I/usr/local/include 
    -I/usr/include 
    -I/usr/lib64/gcc/x86_64-suse-linux/4.1.2/include 
    "-assume realloc_lhs" 
    "-check all" 
    -free 
    "-fp_modbits honor_f32_conversion" 
    "-fp_modbits honor_f64_conversion" 
    -gen_interfaces 
    -O0 
    -g2 
    -traceback 
    "-warn all" 
    "-warn notruncated_source" 
    -mP1OPT_version=11.1-intel64 
    -mGLOB_diag_file=lmain.diag 
    -mGLOB_source_language=GLOB_SOURCE_LANGUAGE_F90 
    -mGLOB_tune_for_fort 
    -mGLOB_use_fort_dope_vector 
    -mP2OPT_static_promotion 
    -mP1OPT_print_version=FALSE 
    -mP3OPT_use_mspp_call_convention 
    -mCG_use_gas_got_workaround=F 
    -mP2OPT_align_option_used=TRUE 
    "-mGLOB_options_string=-I. -fpp -warn all -fpp -check all -gen-interfaces -assume realloc_lhs -static -pg -warn notruncated_source -debug full -traceback -O0 -g -FR -c -o lmain -dryrun" 
    -mGLOB_cxx_limited_range=FALSE 
    -mGLOB_traceback 
    -mP3OPT_emit_line_numbers 
    -mP2OPT_suppress_unreferenced_locals=FALSE 
    -mGLOB_debug_target=GLOB_DEBUG_TARGET_ALL 
    -mGLOB_debug_format=GLOB_DEBUG_FORMAT_DWARF20 
    -mGLOB_as_output_backup_file_name=/tmp/ifortrWFG9Ras_.s 
    -mGLOB_runtime_check_undefined 
    -mIPOPT_activate 
    -mGLOB_machine_model=GLOB_MACHINE_MODEL_EFI2 
    -mGLOB_maintain_errno 
    -mGLOB_noinline_intrinsics 
    -mGLOB_use_base_pointer 
    -mGLOB_extended_instructions=0x8 
    -mP2OPT_subs_out_of_bound=FALSE 
    -mGLOB_ansi_alias 
    -mP3OPT_call_mcount 
    -mIPOPT_args_in_regs=0 
    -mIPOPT_ninl_limit=0 
    -mIPOPT_ninl_user_level=0 
    -mIPOPT_args_in_regs=0 
    -mIPOPT_args_in_regs=0 
    -mPGOPTI_value_profile_use=T 
    -mP2OPT_align_array_to_cache_line=FALSE 
    -mP2OPT_gen_linux_const_pooling=FALSE 
    -mGLOB_opt_level=0 
    -mIPOPT_obj_output_file_name=lmain 
    "-mGLOB_linker_version=2.16.91.0.5 20051219 (SUSE Linux)" 
    -mP3OPT_asm_target=P3OPT_ASM_TARGET_GAS 
    -mGLOB_obj_output_file=lmain 
    -mGLOB_source_dialect=GLOB_SOURCE_DIALECT_FORTRAN 
    -mP1OPT_source_file_name=lmain.f90 
    /tmp/ifortb3BXwa.i90

rm /tmp/ifortyXGztHlibgcc

rm /tmp/ifortm2AQZVgnudirs

rm /tmp/ifortb3BXwa.i90

rm /tmp/ifortYkBf4oarg

rm /tmp/ifortGq8ABDgas

rm /tmp/ifortrWFG9Ras_.s

rm /tmp/ifortN9oNH6ldashv

rm /tmp/ifortmDZ3glarg

rwgreen@spdr65:~/quad/rwgreen/forums/70115> 

OK, I did as you recommended and ran this script on the local ext3 disk:

[shell]#!/bin/csh
echo 'Some system info... uname -a'
uname -a
echo 'Some system info... pwd'
pwd
echo 'Cleaning up...'
rm -f *mod *.o
echo 'Sourcing ifortvars...'
source /gusr/alr99/cluster/intel/Compiler/11.1/059/bin/ifortvars.csh intel64
echo 'ifort -V'
ifort -V
echo 'Compiling modules...'
ifort -I. -fpp -warn all -fpp -check all -gen-interfaces -assume realloc_lhs -static -pg -warn notruncated_source -debug full -traceback -O0 -g -FR -c  global_parameters.f90
ifort -I. -fpp -warn all -fpp -check all -gen-interfaces -assume realloc_lhs -static -pg -warn notruncated_source -debug full -traceback -O0 -g -FR -c  iof2.f90
echo 'Compiling lmain...'
ifort -I. -fpp -warn all -fpp -check all -gen-interfaces -assume realloc_lhs -static -pg -warn notruncated_source -debug full -traceback -O0 -g -FR -c  lmain.f90
echo 'Dry run...'
ifort -I. -fpp -warn all -fpp -check all -gen-interfaces -assume realloc_lhs -static -pg -warn notruncated_source -debug full -traceback -O0 -g -FR -c  lmain.f90 -dryrun[/shell]

The output is attached. Let me know what to try next.

Attachments: 

AttachmentSize
Downloadtext/x-csh ice_debug.csh998 bytes
Downloadapplication/octet-stream ice_debug.log9.23 KB

Aha, got the error now. I had not be pre-compiling global_parameters.f90 and iof2.f90 prior to compiling lmain.

I'll investigate further and get a bug report going.

ron

Quoting - Ronald W. Green (Intel)

Aha, got the error now. I had not be pre-compiling global_parameters.f90 and iof2.f90 prior to compiling lmain.

I'll investigate further and get a bug report going.

ron

It's definitely something with -realloc_lhs. Do you absolutely need that functionality?

I've eliminated iof2.f90, it's not part of the problem. I'm shaving down to a slim testcase, but as you know, lmain is pretty large. I'll get it submitted once I get the testcase slimmed down.

ron

Quoting - Ronald W. Green (Intel)

It's definitely something with -realloc_lhs. Do you absolutely need that functionality?

I've eliminated iof2.f90, it's not part of the problem. I'm shaving down to a slim testcase, but as you know, lmain is pretty large. I'll get it submitted once I get the testcase slimmed down.

ron

bug report DPD200139107 - fixed in 11.1 Update 4 due in December.

We're in luck - there were several bug reports against -realloc_lhs that were entered in the past months. There is a fix coming out in a few weeks in the "Update 4" release of the compiler (exact date TBD - but it should in just about 2-3 weeks barring any unforseen delays).

I have tested your code with an engineering build version of this compiler and have confirmed it fixes your bug too.

I will alert you when this fixed compiler is ready for download.

ron

this was fixed in December 2009, 11.1 Update 4  version 11.1.064

closing

Leave a Comment

Please sign in to add a comment. Not a member? Join today