Hi,
Scott Brozell wrote:
> Hi,
>
> On Thu, 5 Apr 2007, Liang, Lei wrote:
>
>
>> There are the errors we got during the compilation and test:
>>
>> 1. Following the steps below:
>> ./configure compaqf90
>> make serial
>> we get the error:
>> --------------^
>> cc -c -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE -O -std1 -o analyze.o
>> analyze.c
>> cc: Error: analyze.c, line 470: Invalid statement. (badstmt)
>> eigval = (double *) safe_malloc(nelem * sizeof(double)); //
>> for the eigenvalues
>> ---------------------------------------------------------------------^
>>
>
> The C preprocessor on your platform is not stripping C++ style comments.
> These comments do not appear in amber8 or amber9 but in the
> unreleased development version of amber.
> Do not use this list for development code issues.
> I'll fix these comments in cvs.
>
>
We have purchased the academic version of amber9, and it was sent to us
on a CD. Why will this happen? Did we copy an incorrect file from the CD?
>> 2. Then we try to modify the config.h:
>> a. use '-std' instead of '-std1' for the cc compiler
>>
>
> Because ANSI C99 supports C++ style comments,
> but we shall probably try to stay C89 compliant.
>
> http://en.wikipedia.org/wiki/C_(programming_language)#C99
>
>
>> b. add an extra default compile command for '.f90' files to compile
>> $AMBERHOME/src/build_amoeba/amoeba_parm.f90
>>
>
> Send the details of your patch for this to the developers list.
>
We modified the config.h file as following at the very end:
# default rules for Fortran and C compilation:
.f90.o: $< #
this is
$(FPP) $< > _$< # what we
$(FC) -c $(FFLAGS) -o $. _$< # added
.f.o: $<
$(FPP) $< > _$<
$(FC) -c $(FFLAGS) -o $. _$<
.c.o:
$(CC) -c $(CFLAGS) $(CPPFLAGS) -o $. $<
otherwise it has difficulty to recognize the 'amoeba_parm.f90' as the
input file for the c preprocessor.
>
>> Now the compilation is successful.
>> c. run 'make test.serial'
>> we get the following error:
>>
>>
>>> 1. CXML-BLAS-Fatal, ZHEEV - Parameter 8 has illegal value
>>> 2. CXML-BLAS-Fatal, DSYEVX - Parameter 17 has illegal value
>>>
>
> More information is required: which test is failing.
>
I'm sorry, but the error I gave previously was not the first error we
received with the "make test.serial" command. Below is the first error
after we ran 'make test.serial', the CXML error occurred just after this
one for another test, but probably we should resolve this one first.
cd water2_qmmmnmr; ./Run.wnmr
forrtl: error (65): floating invalid
0: __FINI_00_remove_gp_range [0x3ff81a21488]
1: __FINI_00_remove_gp_range [0x3ff81a2a330]
2: __FINI_00_remove_gp_range [0x3ff800d9cf0]
3: __FINI_00_remove_gp_range [0x3ff80279ad4]
4: __FINI_00_remove_gp_range [0x3ff80279084]
5: __FINI_00_remove_gp_range [0x3ff802a7e50]
6: overlap1_ [../obj/overlap.f: 78, 0x120316590]
7: oethccs6_ [../obj/oethccs.f: 418, 0x1202d89c0]
8: diam1_ [../obj/chshift.f: 440, 0x1202c7220]
9: fock_ [../obj/fockqmmm.f: 659, 0x120331ef4]
10: doscf_ [../obj/doscfqmmm.f: 400, 0x1202da540]
11: energy_ [../obj/energyqmmm.f: 110, 0x1202c8244]
12: qm_mm_div_ [../obj/qm_mm_div.f: 167, 0x1202b7940]
13: qm_div_ [_qm_div.f: 243, 0x1202b09d4]
14: force_ [_force.f: 1204, 0x1202b19a4]
15: runmin_ [_runmin.f: 720, 0x120115f14]
16: sander_ [_sander.f: 1488, 0x120106710]
17: multisander_ [_multisander.f: 317, 0x120102d70]
18: main [for_main.c: 203, 0x1202b6d4c]
19: __start [0x1200664d8]
Abort process
./Run.wnmr: Program error
*** Exit 1 (ignored)
> Send the details to the developers list.
>
I am not familiar with the developers list as we have just started to
use Amber9. If you think that these comments would be better directed
to that list could you give me a pointer to it?
Thank you very much!
Lei
> Scott
>
>
>> The information of our system and compiler is:
>> system: OSF1 V5.1 2650 alpha
>> compiler: Compaq Fortran V5.5-2602
>> Compaq Fortran Compiler X5.5-2602-48C8L
>>
>
> -----------------------------------------------------------------------
> The AMBER Mail Reflector
> To post, send mail to amber.scripps.edu
> To unsubscribe, send "unsubscribe amber" to majordomo.scripps.edu
>
-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber.scripps.edu
To unsubscribe, send "unsubscribe amber" to majordomo.scripps.edu
Received on Wed Apr 11 2007 - 06:07:24 PDT
