Present: Roland, Vassilios, Christian, Erik, Yosef
provide overriding con2prim for GRHydro: * presented the two competing implementation: ** using aliased functions which lets GRHydro control exactly when con2prim is called and guarantees that only one is present in a run ** using a schedule group is more Cactus like, GRHydro controls where con2prim is called by scheduling the group on its own, there is however no way to strictly guarantee that only one con2prim thorn is active * currently the first approach is used in production and the second (scheduled group) approach has a test implementation * will go ahead and implement the group method and check how well it works in production. Consider using hydrobase::evoution_method type parameter to advertise/select which con2prim is to be used. This is "Cactus like" but often not ahered to eg by the initial data thorns that use a similar mechanism.
* the idea is to have problem adapted con2prim routines that are independent on GRHydro so that one can update GRHydro independently * do not want to use parameters and many different routines in GRHydro to keep the code small
comet file corruption: * CarpetIOASCII output files can contain corrupted lines of output in the middle of a component * happens only on the scratch file system, not on $HOME and only for more than 1 MPI ranks * happens for both the current release version of ET (ET_PayonGaposhkin 2016_11) as well as old (years old) versions of Carpet * workaound is to force flushing after each line by using std::eol instead of "\n". This may be slow since flushing can be a very slow operation (milliseconds if it flushes all the way to the physical disk) * a workaround may be to write output into a string stream first then dump its rdbuf() to file * will engage SDSC support to see if they have suggestions on how to avoid this issue without having to first write to a local file, then copy to scratch once the run finishes
Yours, Roland