Trac January 2012

trac@lists.einsteintoolkit.org

1 participants
151 discussions

[Einstein Toolkit] #731: simfactory does not expand @MACHINE@ in the SubmitScript if submitted with option --submitscriipt
by Einstein Toolkit 13 Jan '12

13 Jan '12

#731: simfactory does not expand @MACHINE@ in the SubmitScript if submitted with option --submitscriipt ------------------------+--------------------------------------------------- Reporter: alibeck | Owner: eschnett Type: defect | Status: new Priority: major | Milestone: Component: SimFactory | Version: Keywords: | ------------------------+--------------------------------------------------- In the submitscript template for supermuc there is a line as [snip] @SIMFACTORY@ run @SIMULATION_NAME@ --machine=@MACHINE@ --restart- id=@RESTART_ID@ @FROM_RESTART_COMMAND@ [snip] This line is then expanded during submit e.g. to [snip] /home/hpc/pr32pi/lu32vak/Cactus- Simfact2-NewRepos/Cactus/simfactory/bin/sim run test_queue_40 --machine=supermuc --restart-id=0 [snip] If I add --submitscript <scriptfile> to sim submit, this line looks as follows: [snip] /home/hpc/pr32pi/lu32vak/Cactus- Simfact2-NewRepos/Cactus/simfactory/bin/sim run test_queue_40 --machine=@MACHINE@ --restart-id=0 [snip] The submission command in my case was: [snip] simfactory/bin/sim submit test_queue_40 --parfile=par/static_tov.par --configuration aliremote --procs=40 --submitscript configs/aliremote/SubmitScript --machine=supermuc --walltime 2:00:00 [snip] So even with giving the regular SubmitScript file name to the option --submitscript will cause this error -- Ticket URL: <https://trac.einsteintoolkit.org/ticket/731> Einstein Toolkit <http://einsteintoolkit.org> The Einstein Toolkit

1 5

[Einstein Toolkit] #707: Use single minus to run testsuite, not double
by Einstein Toolkit 12 Jan '12

12 Jan '12

#707: Use single minus to run testsuite, not double ---------------------+------------------------------------------------------ Reporter: sbrandt | Owner: Type: defect | Status: new Priority: minor | Milestone: Cactus_4.1.0 Component: Cactus | Version: Keywords: | ---------------------+------------------------------------------------------ make Cfg-testsuite not make Cfg--testsuite -- Ticket URL: <https://trac.einsteintoolkit.org/ticket/707> Einstein Toolkit <http://einsteintoolkit.org> The Einstein Toolkit

1 1

[Einstein Toolkit] #728: RCSfile appears in lower left corner of UsersGuide and MaintGuide
by Einstein Toolkit 12 Jan '12

12 Jan '12

#728: RCSfile appears in lower left corner of UsersGuide and MaintGuide ---------------------+------------------------------------------------------ Reporter: sbrandt | Owner: Type: defect | Status: new Priority: minor | Milestone: Cactus_4.1.0 Component: Other | Version: Keywords: | ---------------------+------------------------------------------------------ The word "RCSfile" appears in lower left corner of UsersGuide and MaintGuide on most pages. -- Ticket URL: <https://trac.einsteintoolkit.org/ticket/728> Einstein Toolkit <http://einsteintoolkit.org> The Einstein Toolkit

1 1

[Einstein Toolkit] #716: Adding steerable checkpoint directories
by Einstein Toolkit 11 Jan '12

11 Jan '12

#716: Adding steerable checkpoint directories --------------------+------------------------------------------------------- Reporter: ashley | Type: enhancement Status: new | Priority: minor Milestone: | Component: Other Version: | Keywords: --------------------+------------------------------------------------------- Hello! With help from Frank and Erik I wrote a quick patch to enable steerable checkpoint directories in IOUtil. Here is what it looks like: Index: param.ccl =================================================================== --- param.ccl (revision 285) +++ param.ccl (working copy) @@ -167,7 +167,7 @@ { ".+" :: "A valid filename" } "checkpoint.chkpt" -STRING checkpoint_dir "Output directory for checkpoint files" STEERABLE = RECOVER +STRING checkpoint_dir "Output directory for checkpoint files" STEERABLE = ALWAYS { ".+" :: "A valid directory name" } "." Index: CheckpointRecovery.c =================================================================== --- CheckpointRecovery.c (revision 285) +++ CheckpointRecovery.c (working copy) @@ -200,8 +200,8 @@ char *filename; const char *dir, *basename; char iteration_postfix[32], filenumber_postfix[32]; + int ierr; - /* get the right parameters */ dir = basename = NULL; switch (called_from) @@ -209,11 +209,17 @@ case CP_INITIAL_DATA: dir = checkpoint_dir; basename = checkpoint_ID_file; + ierr = CCTK_CreateDirectory(0755, checkpoint_dir); + if(ierr<0) + CCTK_VWarn(0, __LINE__, __FILE__, "Checkpoint directory %s could not be created", checkpoint_dir); break; case CP_EVOLUTION_DATA: dir = checkpoint_dir; basename = checkpoint_file; + ierr = CCTK_CreateDirectory(0755, checkpoint_dir); + if(ierr<0) + CCTK_VWarn(0, __LINE__, __FILE__, "Checkpoint directory %s could not be created", checkpoint_dir); break; case CP_RECOVER_DATA: I'm not 100% sure that this will work out in all cases, but I thought I'd put this here in case it was useful :) -- Ticket URL: <https://trac.einsteintoolkit.org/ticket/716> Einstein Toolkit <http://einsteintoolkit.org> The Einstein Toolkit

1 1

[Einstein Toolkit] #725: segfault in a non-MPI Cactus configuration
by Einstein Toolkit 10 Jan '12

10 Jan '12

#725: segfault in a non-MPI Cactus configuration ----------------------+----------------------------------------------------- Reporter: jtao | Owner: Type: defect | Status: new Priority: critical | Milestone: Cactus_4.1.0 Component: Cactus | Version: development version Keywords: PUGH | ----------------------+----------------------------------------------------- David encountered a segfault in a non-MPI Cactus configuration when using the TimerReport thorn. He provides a patch that moves the initialization of pughGH outside of the CCTK_MPI ifdef. -- Ticket URL: <https://trac.einsteintoolkit.org/ticket/725> Einstein Toolkit <http://einsteintoolkit.org> The Einstein Toolkit

1 2

[Einstein Toolkit] #724: segfault in a non-MPI Cactus configuration
by Einstein Toolkit 10 Jan '12

10 Jan '12

#724: segfault in a non-MPI Cactus configuration ----------------------+----------------------------------------------------- Reporter: jtao | Owner: Type: defect | Status: new Priority: critical | Milestone: Cactus_4.1.0 Component: Cactus | Version: development version Keywords: PUGH | ----------------------+----------------------------------------------------- Davidencountered a segfault in a non-MPI Cactus configuration when using the TimerReport thorn. ReductionAvg: Move the initialization of pughGH outside of the CCTK_MPI ifdef. -- Ticket URL: <https://trac.einsteintoolkit.org/ticket/724> Einstein Toolkit <http://einsteintoolkit.org> The Einstein Toolkit

1 1

[Einstein Toolkit] #723: Archived emails to mailing lists are displayed without line-wrapping
by Einstein Toolkit 09 Jan '12

09 Jan '12

#723: Archived emails to mailing lists are displayed without line-wrapping --------------------+------------------------------------------------------- Reporter: hinder | Owner: Type: defect | Status: new Priority: minor | Milestone: Component: Cactus | Version: Keywords: | --------------------+------------------------------------------------------- See http://cactuscode.org/pipermail/developers/2011-November/006258.html This displays without any line wrapping. This only happens with some emails (mine?). I'm not doing anything non-standard, as far as I know. -- Ticket URL: <https://trac.einsteintoolkit.org/ticket/723> Einstein Toolkit <http://einsteintoolkit.org> The Einstein Toolkit

1 6

[Einstein Toolkit] #721: Don't use cfg and run subdirectories in simulations
by Einstein Toolkit 07 Jan '12

07 Jan '12

#721: Don't use cfg and run subdirectories in simulations -------------------------+-------------------------------------------------- Reporter: eschnett | Owner: eschnett Type: enhancement | Status: new Priority: minor | Milestone: Component: SimFactory | Version: Keywords: | -------------------------+-------------------------------------------------- Since the option list, submit script, and run script always have the same names, it is not necessary to put them into subdirectories in the simulations' main SIMFACTORY directory. -- Ticket URL: <https://trac.einsteintoolkit.org/ticket/721> Einstein Toolkit <http://einsteintoolkit.org> The Einstein Toolkit

1 0

[Einstein Toolkit] #717: Simplify and clean up vectorisation
by Einstein Toolkit 06 Jan '12

06 Jan '12

#717: Simplify and clean up vectorisation -------------------------+-------------------------------------------------- Reporter: eschnett | Owner: hinder Type: enhancement | Status: new Priority: major | Milestone: Component: Kranc | Version: Keywords: | -------------------------+-------------------------------------------------- -- Ticket URL: <https://trac.einsteintoolkit.org/ticket/717> Einstein Toolkit <http://einsteintoolkit.org> The Einstein Toolkit

1 3

[Einstein Toolkit] #718: Delete config-info before reconfiguring with an option list
by Einstein Toolkit 05 Jan '12

05 Jan '12

#718: Delete config-info before reconfiguring with an option list ------------------------+--------------------------------------------------- Reporter: eschnett | Owner: eschnett Type: defect | Status: new Priority: major | Milestone: Component: SimFactory | Version: Keywords: | ------------------------+--------------------------------------------------- The file config-info contains information about how the current configuration was configured. When re-configuring with a new option list, this file should be deleted, so that no information from the previous configuration can survive accidentally. -- Ticket URL: <https://trac.einsteintoolkit.org/ticket/718> Einstein Toolkit <http://einsteintoolkit.org> The Einstein Toolkit

1 0

Jump to page:

2026

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

Trac January 2012