Directory R_Tester/toplevel/

Lines of Code

Developers

Most Recent Commits

harman 2017-12-21 15:29 Rev.: 57508

Added James and Tony to the gpu_rt email list.
Now compiling all components on gpu_rt.

0 lines of code changed in 2 files:

• R_Tester/toplevel: startTester (new)

harman 2017-11-27 10:06 Rev.: 57344

Added Allen S. to the email list for the gpu_rt.

1 lines of code changed in 1 file:

• R_Tester/toplevel: startTester (+1 -1)

harman 2017-09-29 13:21 Rev.: 57053

*.py: Linux-> all

runSusTests and generateGoldStandards:
- Warn the user if the OS they're using while running the localRT isn't valid for a component.

8 lines of code changed in 2 files:

• R_Tester/toplevel: generateGoldStandards.py (+8 -3)

harman 2017-09-25 17:53 Rev.: 57028

- Developers no longer needs to use 1.1 to signal that mpirun should be used, it's now the default.
- Removed the "-mpi" sus option.

3 lines of code changed in 1 file:

• R_Tester/toplevel: generateGoldStandards.py (+3 -7)

harman 2017-06-23 13:28 Rev.: 56697

export AUTO_UPDATE_SVN_STAMP which enables the passing svn revision to be logged

1 lines of code changed in 1 file:

• R_Tester/toplevel: startTester (+1 -1)

harman 2017-06-23 12:42 Rev.: 56696

Changed all reference to $USER -> $LOGNAME. The environmental variable USER is not defined in the cron
environment, LOGNAME is.

9 lines of code changed in 1 file:

• R_Tester/toplevel: startTester (+9 -8)

harman 2017-06-20 16:11 Rev.: 56673

Removed albion configuration.

Added configuration for gpu_rt@uintah-build.chpc.utah.edu

18 lines of code changed in 1 file:

• R_Tester/toplevel: startTester (+18 -33)

harman 2017-06-06 16:53 Rev.: 56614

Fixed logic to update the svn revision number in the goldstandards dir when a test passes. Only
the main nightly RT script can update it.

1 lines of code changed in 1 file:

• R_Tester/toplevel: startTester (+1)

harman 2017-06-06 12:46 Rev.: 56607

2 space indentation
- added comments
- more descriptive variable name
- added environmental variable (AUTO_UPDATE_PERF_GS) which controls if the performance
gold standards are automatically updated when the run time improve by > %5. This variable is set
to "yes" for the nightly RT and "no" for local regression testing.

20 lines of code changed in 3 files:

• R_Tester/toplevel: startLocalTest (+10 -9), startTester (+10 -14)

dav 2017-02-26 16:05 Rev.: 56252

Continuation of commit r56249 - this time for some various files.

Mostly cosmetic changes, and preparing for removing a hacked out warning when we moved to C++ 11.
These changes have passed the local RT on Baja.

- The biggest update in this commit (the update to remove the c++11 warning hack) is that you will
no longer be able to compare a ProblemSpecP to 0 (you will need to compare to 'nullptr').

- White space, coding standards, indentation, remove tabs.
- Line up some variables to make them easier to read.
- Use of 80+ columns of text (we decided to allow this 10 years ago) and it makes reading longer lines of code on the screen easier.
- Use {} for if/else clauses even if they are only one line long.
- Please put "else" and "else if" on a new line. The added white space makes following the branching much easier.

M CCA/Components/DataArchiver/DataArchiver.h
M CCA/Components/Parent/Switcher.cc

- Remove the outputPS() function as it appears to be dead code (was commented out).

M CCA/Components/Parent/Switcher.h
M CCA/Components/ProblemSpecification/ProblemSpecReader.cc

- More verbose error message.

M CCA/Components/ReduceUda/UdaReducer.cc
M CCA/Components/Solvers/AMR/AMRSolver.cc
M CCA/Components/Solvers/CGSolver.cc
M CCA/Components/Solvers/HypreSolver.cc
M Core/ProblemSpec/ProblemSpec.cc

M R_Tester/helpers/runSusTests.py
M R_Tester/toplevel/generateGoldStandards.py

- Remove the SCI_SIGNALMODE env flag that was being passed to the tester. As far as I can tell, SCI_SIGNALMODE
was removed from Uintah a long time ago when we cut out a lot of SCIRun stuff. Also, in a bizarre twist,
adding "-x SCI_SIGNALMODE" (or in fact, any "-x " var that is over 8 characters long) causes the answers
to a few tests (eg: methane_fire_8patch_petscrad) to fail exact comparison (at e-15 or so).

2 lines of code changed in 1 file:

• R_Tester/toplevel: generateGoldStandards.py (+2 -2)