Present: Roland, Steve, Bhavesh, Peter, Erik
piraha_everywhere: * there are multiple fixes in multiple branches by now * will set to "review" for branches that should be included * currently have code that reports multiple errors from .par files
stampede KNL updates: * compiles ET properly on KNL can run GW150914 * could compile on login node * currently testing speed on KNL nodes
ET workshop * nothing planned yet * possibly attach a one day workshop at the end of a general computational science workshop that Erik is in the early planning stages ofwill
NaN values in GW150914 (and Llama+Symmetry in general) * found by Christian Ott and tracked down to b/c in Llama * Erik will create a ticket
Hydro example code in CactusExamples: * https://bitbucket.org/cactuscode/cactusexamples/branch/eschnett/hydro * is currently functional and can be used to evolve a Gamma-law fluid plus GR * will discuss in detail in one of the next calls
Yours, Roland
On 17 Apr 2017, at 17:26, Roland Haas rhaas@illinois.edu wrote:
NaN values in GW150914 (and Llama+Symmetry in general)
- found by Christian Ott and tracked down to b/c in Llama
- Erik will create a ticket
Ticket is https://trac.einsteintoolkit.org/ticket/2030, and I have described my understanding of the problem (which may only be a subset of what Erik discussed), and provided a possible fix. Please would someone with a case that fails try out the fix and report back?
-- Ian Hinder http://members.aei.mpg.de/ianhin
users@lists.einsteintoolkit.org