[Mesa-users] error with ./mk on astero module

[Aaron Dotter]

OK, I tried doing exactly what you've done.

MESA_DIR is also hard-coded in the star_job section of the file "inlist".

Aaron

On Tue, Dec 10, 2019 at 7:33 PM Jennifer van Saders <jlvs at hawaii.edu> wrote:

> Dear Aaron,
>
> I have already, that's why I'm so puzzled! In the working directory I have
> copied from astero_gyre, I have both deleted the MESA_DIR in the makefile,
> as well as set MESA_DIR explicitly to the correct path in an attempt to
> resolve it, and neither action has.
>
> Thanks!
> Jen
>
> On Tue, Dec 10, 2019 at 2:22 PM Aaron Dotter <aaron.dotter at gmail.com>
> wrote:
>
>> Hi Jen,
>>
>> Check astero_gyre/make/makefile.  In the test_suite it has MESA_DIR
>> hard-coded to a relative value.  Delete that line and you should be all set.
>>
>> Aaron
>>
>> On Tue, Dec 10, 2019 at 7:18 PM Jennifer van Saders <jlvs at hawaii.edu>
>> wrote:
>>
>>> Hi all,
>>> It turns out I spoke too soon - I can now compile, but there still
>>> appears to be a path issue when running:
>>>
>>>  read inlist_astero_search_controls
>>>  open failed for ../../../data/version_number
>>>  please check that mesa_dir is set correctly
>>>  failed in stardata_init
>>>  star_init ierr           2
>>>
>>> As before, both $MESA_DIR and $MESASDK_ROOT are set (evidenced by echo).
>>> My makefile in make/makefile in my working directory has MESA_DIR set to
>>> the correct path, and the only include statement is: include
>>> $(MESA_DIR)/star/astero/makefile. Again, the module compiles and runs
>>> within the test_suite directory, but not in my copied version. I've poked
>>> around for a good while trying to find where I'm overwriting my mesa_dir
>>> with a '../../../' with no luck.
>>>
>>>
>>> On Tue, Dec 10, 2019 at 10:11 AM Jennifer van Saders <jlvs at hawaii.edu>
>>> wrote:
>>>
>>>> Success, thank you!
>>>>
>>>> Cheers,
>>>> Jen
>>>>
>>>> On Mon, Dec 9, 2019 at 5:43 PM Josiah Schwab <jwschwab at ucsc.edu> wrote:
>>>>
>>>>> Hi Jen,
>>>>>
>>>>> I feel like I'm missing something simple, but it has me stumped!
>>>>>>
>>>>>
>>>>> This happens when the __FILE__ macro gets expanded and as a result the
>>>>> line length exceeds the default compiler line length limit.  That's why it
>>>>> only happens for some directory locations (and macro expansion makes the
>>>>> error confusing, since the line doesn't look too long in the error message).
>>>>>
>>>>> Open up utils/makefile_header and remove the length limits by making
>>>>> edits like:
>>>>>
>>>>> FCfixed = -ffixed-form -ffixed-line-length-none $(FC_fixed_preprocess)
>>>>>
>>>>> FCfree = -ffree-form -ffree-line-length-none $(FC_free_preprocess)
>>>>>
>>>>> I believe should resolve this issue.  Warrick hit this recently too,
>>>>> so we'll make sure it is fixed in the next release.
>>>>>
>>>>> Josiah
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>
>>>>
>>>> --
>>>> Jennifer van Saders
>>>> Assistant Astronomer
>>>> Institute for Astronomy, University of Hawaii
>>>> 2680 Woodlawn Dr, Honolulu, HI 96822
>>>>
>>>
>>>
>>> --
>>> Jennifer van Saders
>>> Assistant Astronomer
>>> Institute for Astronomy, University of Hawaii
>>> 2680 Woodlawn Dr, Honolulu, HI 96822
>>> _______________________________________________
>>> mesa-users at lists.mesastar.org
>>> https://lists.mesastar.org/mailman/listinfo/mesa-users
>>>
>>>
>
> --
> Jennifer van Saders
> Assistant Astronomer
> Institute for Astronomy, University of Hawaii
> 2680 Woodlawn Dr, Honolulu, HI 96822
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.mesastar.org/pipermail/mesa-users/attachments/20191210/d4cc4590/attachment.htm>