Trac September 2012

trac@lists.einsteintoolkit.org

1 participants
182 discussions

[Einstein Toolkit] #517: archive CactusExternal
by Einstein Toolkit 23 Oct '12

23 Oct '12

#517: archive CactusExternal -------------------------+-------------------------------------------------- Reporter: knarf | Owner: Type: enhancement | Status: new Priority: minor | Milestone: Component: Cactus | Version: Keywords: | -------------------------+-------------------------------------------------- We should move the CactusExternal thorns to archive. They cause trouble and confusion where they are right now. This should happen for the next Cactus release. -- Ticket URL: <https://trac.einsteintoolkit.org/ticket/517> Einstein Toolkit <http://einsteintoolkit.org> The Einstein Toolkit

1 6

[Einstein Toolkit] #176: Test parameter files without running them
by Einstein Toolkit 23 Oct '12

23 Oct '12

#176: Test parameter files without running them -------------------------+-------------------------------------------------- Reporter: hinder | Owner: Type: enhancement | Status: new Priority: minor | Milestone: Component: Cactus | Version: Keywords: | -------------------------+-------------------------------------------------- It would be useful to be able to check that the parameters of a run are OK before running it. Since there can be long queue times on supercomputers, it would be good to do this on a local machine which might not have as much memory as is required for the full run. Currently, running the job would cause memory exhaustion on the smaller machine. One solution would be an additional command-line argument to Cactus --exit-after-paramcheck which stops the run cleanly after the PARAMCHECK Cactus bin. This is preferable to a new Cactus parameter, as it would not require the parameter file to be modified. One could imagine this also being potentially used by simfactory automatically when submitting a job (though this would only work if you could run MPI executables on the head node). -- Ticket URL: <https://trac.einsteintoolkit.org/ticket/176> Einstein Toolkit <http://einsteintoolkit.org> The Einstein Toolkit

1 12

[Einstein Toolkit] #773: McLachlan dissipation should be disabled by default
by Einstein Toolkit 23 Oct '12

23 Oct '12

#773: McLachlan dissipation should be disabled by default -------------------------+-------------------------------------------------- Reporter: hinder | Owner: Type: enhancement | Status: new Priority: major | Milestone: Component: Other | Version: Keywords: | -------------------------+-------------------------------------------------- In the current version of McLachlan, the parameter apply_dissipation is set to "always" by default. The corresponding "epsdiss" is set to 0 by default. The dissipation calculation in McLachlan is expensive due to the way that it iterates over variables. Use of the Dissipation thorn or similar is usually much faster. I had not realised that apply_dissipation was "always" by default. I think it should be "never". OK to change? (with announcement to the mailing list) -- Ticket URL: <https://trac.einsteintoolkit.org/ticket/773> Einstein Toolkit <http://einsteintoolkit.org> The Einstein Toolkit

1 13

[Einstein Toolkit] #535: Add autoconf macro to test C99 style variable declarations
by Einstein Toolkit 23 Oct '12

23 Oct '12

#535: Add autoconf macro to test C99 style variable declarations -------------------------+-------------------------------------------------- Reporter: eschnett | Owner: Type: enhancement | Status: new Priority: major | Milestone: Component: Cactus | Version: Keywords: | -------------------------+-------------------------------------------------- Many compilers don't support C99 style variable declarations in their default settings. We should test this, add certain flags automatically if we can, or abort with a clear error message otherwise. -- Ticket URL: <https://trac.einsteintoolkit.org/ticket/535> Einstein Toolkit <http://einsteintoolkit.org> The Einstein Toolkit

1 1

[Einstein Toolkit] #632: Changegroup hook failed when pushing to Carpet repository
by Einstein Toolkit 22 Oct '12

22 Oct '12

#632: Changegroup hook failed when pushing to Carpet repository --------------------+------------------------------------------------------- Reporter: hinder | Owner: eschnett Type: defect | Status: new Priority: minor | Milestone: Component: Carpet | Version: Keywords: | --------------------+------------------------------------------------------- When I push to the Carpet repository, I get the following message: {{{ remote: error: changegroup.cia hook failed: http://cia.vc returned an error: queued. }}} The push succeeds. What is the cause of this error? -- Ticket URL: <https://trac.einsteintoolkit.org/ticket/632> Einstein Toolkit <http://einsteintoolkit.org> The Einstein Toolkit

1 3

[Einstein Toolkit] #359: Add MDB entry for Pandora at LSU
by Einstein Toolkit 22 Oct '12

22 Oct '12

#359: Add MDB entry for Pandora at LSU ------------------------+--------------------------------------------------- Reporter: eschnett | Owner: mthomas Type: task | Status: new Priority: major | Milestone: Component: SimFactory | Version: Keywords: | ------------------------+--------------------------------------------------- -- Ticket URL: <https://trac.einsteintoolkit.org/ticket/359> Einstein Toolkit <http://einsteintoolkit.org> The Einstein Toolkit

1 3

[Einstein Toolkit] #412: Split Appendices
by Einstein Toolkit 22 Oct '12

22 Oct '12

#412: Split Appendices -------------------------+-------------------------------------------------- Reporter: eschnett | Owner: Type: enhancement | Status: new Priority: minor | Milestone: Component: Cactus | Version: Keywords: | -------------------------+-------------------------------------------------- Currently, the appendices of the users' guide are repeated into the reference manual. This is somewhat confusing, and causes problems because the appendices cannot easily reference other sections (e.g. "see page 15" doesn't make sense since one doesn't know in which document this will be read). I suggest to split the appendices and have some of them in the users' guide while moving others into the reference manual, so that each appendix is included in only one document -- Ticket URL: <https://trac.einsteintoolkit.org/ticket/412> Einstein Toolkit <http://einsteintoolkit.org> The Einstein Toolkit

1 1

[Einstein Toolkit] #982: Abort with an error if the argument of --root= is not a directory
by Einstein Toolkit 22 Oct '12

22 Oct '12

#982: Abort with an error if the argument of --root= is not a directory -----------------------------+---------------------------------------------- Reporter: anonymous | Owner: eric9 Type: enhancement | Status: new Priority: minor | Milestone: Component: GetComponents | Version: development version Keywords: error detection | -----------------------------+---------------------------------------------- Currently, when GetComponnents is invoked as GetComponents --root=<dir> and <dir> does not exist, Getcomponents just ignores the option. I would be better if GetComponents aborted with an error if the argument of --root= is not a valid directory. -- Ticket URL: <https://trac.einsteintoolkit.org/ticket/982> Einstein Toolkit <http://einsteintoolkit.org> The Einstein Toolkit

1 5

[Einstein Toolkit] #1043: RotatingSymmetry180::poison_boundaries should not check for NaNs
by Einstein Toolkit 20 Oct '12

20 Oct '12

#1043: RotatingSymmetry180::poison_boundaries should not check for NaNs -----------------------------------+---------------------------------------- Reporter: rhaas | Owner: Type: defect | Status: new Priority: minor | Milestone: Component: EinsteinToolkit thorn | Version: Keywords: McLachlan | -----------------------------------+---------------------------------------- currently poison_boundaries causes RotatingSymmetry180 checks for both the poison value and for nonfinite values in the boundary. It should probably not check for NaN in case there are grid functions that do actually contain nan values near the boundaries (right not I have a run where this is true for ML_BSSN::H where I don't care about the NaNs). RotatingSymmetry's output completely clutters stdout since it list every single nan point it find. If one wants to check for NaN values in the constraints one can always use NaNChecker so no functionality is lost. Also RotatingSymmetry might well see NaN values when combined with say ReflectionSymmetry if the NaNs would be overwritten by say a z-reflection symmetry afterwards (I am not sure what the order in which symmetries are applied is). -- Ticket URL: <https://trac.einsteintoolkit.org/ticket/1043> Einstein Toolkit <http://einsteintoolkit.org> The Einstein Toolkit

1 4

[Einstein Toolkit] #704: Carpet complains about lack of mpi after ExternalLibraries/OpenMPI is built
by Einstein Toolkit 20 Oct '12

20 Oct '12

#704: Carpet complains about lack of mpi after ExternalLibraries/OpenMPI is built ---------------------+------------------------------------------------------ Reporter: bmundim | Owner: Type: defect | Status: new Priority: major | Milestone: Component: Cactus | Version: Keywords: | ---------------------+------------------------------------------------------ I was able to compile ExternalLibraries/OpenMPI successfully, but the Cactus compilation stops afterwards with the message: /home/bruno/tmp/einstein_dev_maxwell/Cactus/arrangements/Carpet/CarpetLib/src/make.configuration.defn:5: *** Configuration error: The Carpet thorns require MPI. Please configure with MPI, or remove the Carpet thorns from the ThornList.. Stop. make: *** [einstein] Error 2 I indicated in my configuration file the following options: OPENMPI_DIR = BUILD OPENMPI_INSTALL_DIR = /home/bruno/local/gcc4.6.1/openmpi-1.5.4 I have then tried to indicate in my config file these extra options after the library was built (and the rest of Cactus compilation stopped): OPENMPI_DIR = /home/bruno/local/gcc4.6.1/openmpi-1.5.4 OPENMPI_INC_DIRS = /home/bruno/local/gcc4.6.1/openmpi-1.5.4/include OPENMPI_LIB_DIRS = /home/bruno/local/gcc4.6.1/openmpi-1.5.4/lib The config-info file does reflect these choices afterwards, and apparently the flag indicating the presence of mpi library, HAVE_MPI, was set correctly at ~/Cactus/configs/einstein/bindings/Configuration/Capabilities: grep -i have_mpi * cctki_MPI.h:#define HAVE_MPI 1 make.MPI.defn:HAVE_MPI = 1 however since I didn't use the old mechanism to tell Cactus about MPI, the ~/Cactus/configs/einstein/config-data/make.extra.defn doesn't have anything indicating the presence of mpi library there. It seems to me a compilation order issue. Somehow HAVE_MPI definition is coming after Carpet compilation, triggering this error then. Does anyone have any idea where I should look at in order to fix this problem? Thanks, Bruno. -- Ticket URL: <https://trac.einsteintoolkit.org/ticket/704> Einstein Toolkit <http://einsteintoolkit.org> The Einstein Toolkit

1 4

← Newer
1
...
10
11
12
13
14
15
16
...
19
Older →

Jump to page:

2026

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

Trac September 2012