[Users] ET build errors (Leonardo DCGP, CINECA cluster)

Wed Sep 3 05:23:27 CDT 2025

Hi Roland,

Got it.

Here is an update:

I removed all CarpetX-related thorns and tried to compile again.
This time, the build complained about thorns GRHayLHDX, GRHayLIDX, and NewRadX:

CST error 1:
-> Thorn 'GRHayLHDX' requires the capability 'Loop'.
Please add a thorn that provides 'Loop' to your ThornList or remove 'GRHayLHDX' from it !

CST error 2:
-> Thorn 'GRHayLIDX' requires the capability 'Loop'.
Please add a thorn that provides 'Loop' to your ThornList or remove 'GRHayLIDX' from it !

CST error 3:
-> Thorn 'NewRadX' requires the capability 'Loop'.
Please add a thorn that provides 'Loop' to your ThornList or remove 'NewRadX' from it !

Comparing with Bruno's original file, I saw that he had disabled GRHayLHDX and GRHayLIDX.
So I went ahead and disabled these two thorns for starters.

Then the build complained just about the NewRadX thorn (same error message as above).

So I disabled NewRadX too (see updated .ini file attached) and the error about NewRadX went away.

Now the build complains about the CCE_Export thorn.
You may see the error messages in the updated log file attached.

>From what I understand, the first main error seems to be this (see make_updated.log file):
/leonardo/home/userexternal/piosif00/Cactus/arrangements/EinsteinAnalysis/CCE_Export/src/h5_export.cc:8:21: error: 'filesystem' is not a namespace-name; did you mean 'system'?
8 | namespace fs = std::filesystem;
The text around that error message suggests that again a newer C++ dialect option might be required (-std=c++17' or '-std=gnu++17' ).

In my current cfg file, following what was mentioned in the wiki page about configuring a new machine<https://docs.einsteintoolkit.org/et-docs/Configuring_a_new_machine>, I have an older option, namely -std=gnu+11, for CXXFLAGS.

I will experiment with different C standards options and see if something else works.

Questions:

  *
Are GRHayLHDX, GRHayLIDX and NewRadX safe to disable?
     *
For context, my goal is to start with isolated neutron star simulations. Are these thorns necessary for that?
  *
Do we expect newer C/C++ standards, like -std=c++17, to break backwards compatibility, i.e. older code?
  *
Please let me know if I am on the right track, or if you see some additional issues from the log file.

Let me note that I did not start with all of the disabled thorns that Bruno's original ini file<https://bitbucket.org/simfactory/simfactory2/src/a6dff6ecc4a4346c5a73536681486e745d34bb98/mdb/machines/leonardo-DCGP.ini> uses, because I didn't know exactly what was needed or not.
Currently, the disabled thorns in my ini file almost align with those in Bruno's original file.
There are, though, some additional thorns disabled in Bruno's original file, namely: ADIOS2, AMReX, Silo, PAPI and THCExtra/WeakRates.

I do not know if I should disable them too, but since I did not encounter specific errors about them, I let them be.
Let me know if they should go too.

Thanks for the help so far!

Best,
Panayotis

________________________________
From: Roland Haas <rhaas at phas.ubc.ca>
Sent: Tuesday, September 2, 2025 7:05 PM
To: IOSIF PANAGIOTIS <PANAGIOTIS.IOSIF at units.it>
Cc: rhaas at mail.ubc.ca <rhaas at mail.ubc.ca>; Einstein Toolkit Users <users at einsteintoolkit.org>
Subject: Re: [Users] ET build errors (Leonardo DCGP, CINECA cluster)

Hello Panayotis,

oh, oops, for CarpetX (which includes Arith) code you actually need
-std=c++17 .

If you are not planning to develop for CarpetX then you can remove all
CarpetX related thorns from your thornlist.

Note that if you have a specific parameter file that you would like to
run, then you can use the MakeThornList helper script to create a
thornlist with just the thorns required for that parameter file,
potentially speeding up compilation quite a bit.

Something like:

./utils/Scripts/MakeThornList --master thornlists/einsteintoolkit.th
--output thornlists/qc0.th par/qc0-mclachlan.par

if you wanted to run qc0-mclachlan.par and had the full thornlist in
thornlists/einsteintoolkit.th (GetComponents will do that for you).

Then pass qc0.th to simfactory using its --thornlist argument (you
likely also need to pass --reconfig to make sure it actually updates
the thornlist it uses).

Yours,
Roland

> [CAUTION: Non-UBC Email]
>
> Hi Roland,
>
> I amended my cfg file and added the flag -ffixed-line-length-none ,
> in the Fortran 77 and 90 compiler options:
>
>
>   *
> F77FLAGS = -g -march=native -fcray-pointer -ffixed-line-length-none
>   *
> F90FLAGS = -g -march=native -fcray-pointer -ffixed-line-length-none
>
> This step solved the issue with the complaints about the "Exact"
> thorn.
>
>
> The build process continued and stopped at a different point, now
> complaining about the CarpetX/Arith thorn.
>
> I have attached the new make.log file.
>
> I ran the log through ChatGPT, and it suggested that this could be
> caused by C++ standard mismatches. Does this sound reasonable to you?
>
> Currently, the flags for the C and C++ compilers in my config file
> are like this:
>
>   *
> CFLAGS = -g -march=native -std=gnu99
>   *
> CXXFLAGS = -g -march=native -std=gnu++11
>
> I checked Bruno's file, and the respective options there are the
> following:
>
>   *
> CFLAGS = -g -std=gnu99 #-std=c11
>   *
> CXXFLAGS = -g -std=gnu++0x #-std=c++14
>
> However, I should note that in Bruno's version, all CarpetX-related
> thorns are disabled in the ini file.
>
> Again, I am not sure I need CarpetX.
> At this point, I am trying to get the build process to work and run
> some examples on the cluster.
>
>
> Returning to the compiler options, I suspect that CFLAGS can be left
> alone. Do you have any suggestions on how to change CXXFLAGS to
> bypass the CarpetX/Arith complaints?
>
> I checked the different C dialect options
> here<https://gcc.gnu.org/onlinedocs/gcc/C-Dialect-Options.html>, and
> it seems that there are a number of possibilities.
>
> Is there some -std=... option that is new enough for CarpetX and
> old/standard enough so that older C++ code is also OK?
>
>
> Best,
> Panayotis
>
>
>
>
> ________________________________
> From: Roland Haas <rhaas at phas.ubc.ca>
> Sent: Tuesday, September 2, 2025 5:43 PM
> To: IOSIF PANAGIOTIS <PANAGIOTIS.IOSIF at units.it>
> Cc: Einstein Toolkit Users <users at einsteintoolkit.org>
> Subject: Re: [Users] ET build errors (Leonardo DCGP, CINECA cluster)
>
> Hello Panayotis,
>
> Oh, I had not noticed, sorry. Hmm, that is the typical cause for these
> issues.
>
> The only other difference I see is `-ffixed-line-length-none` though
> that would also point to an issue with the Cactus Fortran file
> processing not breaking things into short enough lines.
>
> Yours,
> Roland
>
> > [CAUTION: Non-UBC Email]
> >
> > Hi Roland,
> >
> > In the config file I attached, I do have this line included already:
> >
> > FPPFLAGS = -g -traditional
> >
> > Is there anything else I might try?
> >
> > Best,
> > Panayotis
> > ________________________________
> > From: Roland Haas <rhaas at phas.ubc.ca>
> > Sent: Tuesday, September 2, 2025 5:08 PM
> > To: IOSIF PANAGIOTIS <PANAGIOTIS.IOSIF at units.it>
> > Cc: Einstein Toolkit Users <users at einsteintoolkit.org>
> > Subject: Re: [Users] ET build errors (Leonardo DCGP, CINECA cluster)
> >
> > Hello Panayotis,
> >
> > Thanks for including all the output files.
> >
> > In your options list whenever your set FPP you also must also set
> > FPPFLAGS:
> >
> > FPPFLAGS=-traditional
> >
> > See eg generic.cfg or (very new!) Bruno's leonardo-DCGP.cfg file (in
> > simfactory master).
> >
> > Yours,
> > Roland
> >
> > > [CAUTION: Non-UBC Email]
> > >
> > > Hi Einstein Toolkit users,
> > >
> > > I am reaching out regarding some errors I am encountering during
> > > my attempts to compile ET on the Leonardo, CINECA cluster (DCGP
> > > partition).
> > >
> > > Here are some details on the Leonardo cluster:
> > >
> > >   *
> > > cluster specifics:
> > > https://docs.hpc.cineca.it/hpc/leonardo.html#leonardo-card *
> > > OS: Red Hat Enterprise Linux 8.7
> > >
> > > I attach the 4 configuration files I am using (with many added
> > > comments for guidance and future reference). I adapted them from
> > > the respective config files that Bruno Giacomazzo kindly provided.
> > >
> > > I also attach the make.log output of the command:
> > >
> > > ./simfactory/bin/sim build --verbose --reconfig
> > > --machine=leonardo-dcgp1 --optionlist
> > > simfactory/mdb/optionlists/leonardo-dcgp1.cfg
> > > --allocation=CNHPC_1479290_0 --thornlist
> > > thornlists/einsteintoolkit.th 2>&1 | tee make.log
> > >
> > > From what I understand, the build process complains about syntax
> > > errors in the Exact thorn in these two source files:
> > >
> > >   *
> > > decode_pars.F
> > >   *
> > > Boost_rotation_symmetric.F
> > >
> > > How can I bypass the above problems?
> > >
> > > To be honest, I am not sure that I really need this thorn.
> > > I was simply trying to compile ET using the default thornlist
> > > from a fresh download/installation.
> > >
> > > Is there any other error that the log file is pointing out that I
> > > might be missing?
> > >
> > > Best regards,
> > > Panayotis
> > >
> > >
> > > -------------------
> > > Panagiotis Iosif
> > > postdoctoral researcher
> > > Department of Physics, University of Trieste
> > > Via Alfonso Valerio 2, Trieste 34127
> > > Italy
> > > -------------------
> > >
> >
> >
> > --
> > My email is as private as my paper mail. I therefore support
> > encrypting and signing email messages. Get my PGP key from
> > http://pgp.mit.edu .
>
>
> --
> My email is as private as my paper mail. I therefore support
> encrypting and signing email messages. Get my PGP key from
> http://pgp.mit.edu .

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://pgp.mit.edu .
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.einsteintoolkit.org/pipermail/users/attachments/20250903/078cf604/attachment-0001.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: make_updated.log
Type: text/x-log
Size: 321172 bytes
Desc: make_updated.log
URL: <http://lists.einsteintoolkit.org/pipermail/users/attachments/20250903/078cf604/attachment-0001.bin>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: leonardo-dcgp1_updated.ini
Type: application/octet-stream
Size: 6470 bytes
Desc: leonardo-dcgp1_updated.ini
URL: <http://lists.einsteintoolkit.org/pipermail/users/attachments/20250903/078cf604/attachment-0001.obj>