[Users] How to use Formaline -- symbol table/file error

[Erik Schnetter]

Bernard

Very strange.

The next step would be to run Cactus in a debug while it is outputting
the source. Maybe that would give a clue?

You can also check out a new version of Cactus and try outputting its
sources, to see whether this works.

-erik


On Tue, Apr 19, 2022 at 6:13 PM Bernard Kelly <physicsbeany at gmail.com> wrote:
>
> Hi Erik.
>
> No. Space / file number has sometimes been an issue on this machine, but not recently, and I tried this with two different executables compiled a week apart. I just compiled a different executable with Formaline, and got the same result.
>
> Bernard
>
> On Tue, 19 Apr 2022 at 13:40, Erik Schnetter <schnetter at gmail.com> wrote:
>>
>> Bernard
>>
>> I don't know why Formaline wouldn't output all thorns. There is no
>> mechanism that would store or output only some of the thorns. Did you
>> run out of disk space just now, so that the other tarballs couldn't be
>> written?
>>
>> -erik
>>
>> On Tue, Apr 19, 2022 at 5:51 AM Bernard Kelly <physicsbeany at gmail.com> wrote:
>> >
>> > Hi Roland. Thanks for the quick response.
>> >
>> > I tried running the executable using a parameter file that only included 'ActiveThorns = "Formaline"', as you suggested. This seemed to work, but the result was a new directory called 'cactus-source' that included only a handful of basic thorns:
>> >
>> > Cactus-source-ADMAnalysis.tar.gz
>> > Cactus-source-ADMBase.tar.gz
>> > Cactus-source-ADMCoupling.tar.gz
>> > Cactus-source-ADMMacros.tar.gz
>> > Cactus-source-ADMMass.tar.gz
>> > Cactus-source-CactusConfig.tar.gz
>> > Cactus-source-Cactus.tar.gz
>> >
>> > Does this mean that Formaline didn't store the other thorns, or that they have to be dug out some other way?
>> >
>> > Bernard
>> >
>> > On Mon, 18 Apr 2022 at 14:26, Roland Haas <rhaas at illinois.edu> wrote:
>> >>
>> >> Hello Bernard,
>> >>
>> >> > Follow-up thought: since gdb is a debugger, does this mean I should have
>> >> > compiled the original executable with debugging symbols? (Something I
>> >> > almost never do for a production executable.)
>> >> Almost certainly, yes. Though maybe not all off them but enough that
>> >> say "nm exe/cactus_sim_ET_2021_05_BC" reports something for
>> >> cactus_source (eg in my case nm exe/cactus_sim | grep
>> >> 'cactus_source$' -> 0000000006fdc1c0 D cactus_source). Gdb needs to be
>> >> able to resolve the "cactus_source" symbol into the actual address.
>> >> Otherwise it will be hard to find out that correct starting point for
>> >> the linked list of tar file fragments.
>> >>
>> >> If you can still run the the executable then running it with a minimal
>> >> thornlist that says
>> >>
>> >> ActiveThorns = "Formaline"
>> >>
>> >> will write out the source code tarball.
>> >>
>> >> Yours,
>> >> Roland
>> >>
>> >> --
>> >> My email is as private as my paper mail. I therefore support encrypting
>> >> and signing email messages. Get my PGP key from http://keys.gnupg.net.
>> >
>> >
>> >
>> > --
>> > ------------------------------------------------------------------
>> > Bernard Kelly -- CRESST Assistant Research Scientist, NASA/GSFC
>> > Gravitational Astrophysics Laboratory --- Code 663
>> >
>> > Phone: +1 (301) 286-7243 *** Fax: +1 (301) 286-2226
>> > Web: http://science.gsfc.nasa.gov/sed/bio/bernard.j.kelly
>> > ORCID: orcid.org/0000-0002-3326-4454
>> > ------------------------------------------------------------------
>> > _______________________________________________
>> > Users mailing list
>> > Users at einsteintoolkit.org
>> > http://lists.einsteintoolkit.org/mailman/listinfo/users
>>
>>
>>
>> --
>> Erik Schnetter <schnetter at gmail.com>
>> http://www.perimeterinstitute.ca/personal/eschnetter/
>
>
>
> --
> ------------------------------------------------------------------
> Bernard Kelly -- CRESST Assistant Research Scientist, NASA/GSFC
> Gravitational Astrophysics Laboratory --- Code 663
>
> Phone: +1 (301) 286-7243 *** Fax: +1 (301) 286-2226
> Web: http://science.gsfc.nasa.gov/sed/bio/bernard.j.kelly
> ORCID: orcid.org/0000-0002-3326-4454
> ------------------------------------------------------------------



-- 
Erik Schnetter <schnetter at gmail.com>
http://www.perimeterinstitute.ca/personal/eschnetter/