[Users] Issues compiling ETK on Blue Waters

Haas, Roland rhaas at illinois.edu
Thu Aug 15 07:32:31 CDT 2019


Hello Federico,

> Can anyone give me some hint?
I can compile the ET_Proca release of the ET one BW using the
release option list.

this change is likely due to a change in the BW programming environment
(which may have slipped proper announcement channels).

The error you describe (incorrect number of arguments to
CCTK_LOOP3STROFF_NORMAL) is most likely due to mixing differing versions of the codes.

Eg the argument list of CCTK_LOOP3STROFF_NORMAL increased from 23 to 24
in e896df97

https://bitbucket.org/cactuscode/cactus/commits/e896df97e3afb6e2f90588dc8edb92d4b0a2a885

I cannot give a definitive answer as to why different different
versions may be used. One guess would be that something confused make.
This can happen if you mix using simfactory's "sync" command to copy
source code to a cluster with editing / updating / downloading code
directly on the cluster.

In that case you can try and see if a full fresh recompile using, in
order of severity of wiping the configuration

make sim-clean
make sim-realclean
rm -rf configs/sim

before a fresh

simfactory/bin/sim build --thornlist thornlists/einsteintoolkit.th

is used.

Note that currently "master" does not build on BW due to a failure to
properly detect the F90 compiler though that failure happens during the
CST phase.

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.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 833 bytes
Desc: OpenPGP digital signature
Url : http://lists.einsteintoolkit.org/pipermail/users/attachments/20190815/5dbc09a0/attachment.bin 


More information about the Users mailing list