Merge branch 'scott/testharness-update1' of https://bitbucket.org/petsc/petsc into scott/testharness-update1

gtest: unify test and normal compilation

Do not remove the generated makefile snippets

Previous implementation of cleantest removed the generated files per
standard makefile practice (one should always clean up); however, the
way that make

Do not remove the generated makefile snippets

Previous implementation of cleantest removed the generated files per
standard makefile practice (one should always clean up); however, the
way that make works means that generated make files cannot be reloaded.
See:
http://stackoverflow.com/questions/554742/reevaluate-makefile-variables
(and note that the second implementation does not work).
Here we just do not remove those generated files similar to how the
"files" file generated by gmakegen.py is not removed.
An alternative would be to have cleantest regenerate the file after
removal.

show more ...

Merge branch 'scott/testharness-update1' of https://bitbucket.org/petsc/petsc into scott/testharness-update1

Bug fix for the fortran. Simplify verbosity.

Making the requirements for fortran consistent with an earlier change in
how self.getLanguage denotes the language.

Also simplified the verbosity.

Test verbosity improved, new implicit rules

Verbosity now handled for tests in the same way as the main library
compilation by using the testquiet command.

Executables now have implicit rules in gm

Test verbosity improved, new implicit rules

Verbosity now handled for tests in the same way as the main library
compilation by using the testquiet command.

Executables now have implicit rules in gmakefile instead of being
explicit in the testfiles file.

Some prototype code is in place for doing implicit rules for running
the scripts, but it is not working.

show more ...

gtest: make V=0 to suppress all commands

gtest: label test executables as intermediate

Need full path of testcompile.log; do not print ok or skip cases it just makes the logs unreadable

Time: .25 hours

fix 6ac365ae7778936b9db95785ab404faf4e929077

> /bin/sh: CC: command not found
> /bin/sh: CC: command not found

Handle logging to make test compilation quiet.

It is quiet, perhaps too quiet...

Do not copy over test_harness.sh and report_tests

Invoke report_tests.py with the counts dir specified.
test_harness.sh sourced using PETSC_DIR

gtest: use $(PYTHON) to run (python2) script

Revert previous functionality

Satish fixed circular dependency -- it was not what I thought it was.

Updates based on team review

o report_tests.sh replaced by report_tests.py
o Example file conversion does not have spaces after comment character
per PETSc standards, and it is now parsed correctly

build: attempt to compile (.NOTPARALLEL) fortran sources early - so that there is some overlap with other compiles

fix bsd test wrt bash. Also remove circular dependency, and add 'cleantest' target

make[2]: Circular report_tests <- report_tests dependency dropped.

Improve dependency chain for parallel operation

report_tests has to be last so has to depend on the tests actually run.
This reorders that to make it easier.

Fixes from pull request comments

This commit addresses issues found during the code review:
- gmakefile had the wrong dependency of library on objects
- gmakefile: $(generatedtestargs) now depends o

Fixes from pull request comments

This commit addresses issues found during the code review:
- gmakefile had the wrong dependency of library on objects
- gmakefile: $(generatedtestargs) now depends on $(generatedtest)
instead of invoking gmakegentest.py as well
- TESTTARGETS should be calculated as needed ("=" instead of ":=")
- Determining requirements based on arguments is fraught with peril, but
is now more robust
- Added support for finding the yaml input file so that it is
automatically added to localrunfiles. This allows for more robust testing.
- Bug fix for localrunfiles: Shell script test harness now copies
runfiles to local directories.

show more ...

Bugs and features per discussion on bitbucket

1 convertExamples now handles multiple filters correctly
2 There was a bug in writing out the TODO block that was caught in
implementing filter
3 Ver

Bugs and features per discussion on bitbucket

1 convertExamples now handles multiple filters correctly
2 There was a bug in writing out the TODO block that was caught in
implementing filter
3 Verbosity level changed to be more silent by default, but to enable
V=1 to make more verbose
4 Skip changed from 'not ok' to 'ok'
5 localrunfiles feature implemented:
This copies over input file if the description language contains localrunfiles.
6 filter_output implemented:
This has a new diff line that allows filtering of the output before
diffing.
7 Preliminary logging enabled. This uses a simple tee similar to the
build. The logging is done in the tests directory however.

show more ...

Merge branch 'barry/fortran-update'

Introduce new test harness and script to convert from the old makefile approach to the new test harness

Updated and simplified Fortran interface

Reduced from 4 distinct approaches to using PETSc from Fortran to 1
No more .h90 files (Fortran 90 functionality is now always available)
Parameters and func

Updated and simplified Fortran interface

Reduced from 4 distinct approaches to using PETSc from Fortran to 1
No more .h90 files (Fortran 90 functionality is now always available)
Parameters and function prototypes are now provide through Fortran modules, not includes;
only #define are provided in the petscXXXdef.h files
PETSc objects can be declared as either type(tXXX) or XXX. For example type(tVec) or Vec
Moved now private Fortran include files out of public include directory since uses don't use them;
they are only used to generate the Fortran modules
Easier to support and test; we can now easily add more function prototypes
Also had to rework generation of manual pages to consistently use MANSEC and SUBMANSEC
since these variables also handle how the Fortran modules are generated.

Time: 57 hours

show more ...

Merge branch 'master' into tisaac/dmplex-feature-real-to-ref

I need updated PetscFE interfaces in what I'm going to do

* master: (603 commits)
parmetis: fix typo in testsuite
Added PETSC_HAVE_A

Merge branch 'master' into tisaac/dmplex-feature-real-to-ref

I need updated PetscFE interfaces in what I'm going to do

* master: (603 commits)
parmetis: fix typo in testsuite
Added PETSC_HAVE_ATTRIBUTEALIGNED flag to makefile
PCSetUp_SVD() used incorrect matrix for matrix size in parallel, also need to skip LAPACK call if matrix size is zero
strumpack: update to latest version 1.0.4
Remove valgrind warning about MPI_Allreduce() with uninitialized data
sed for test output that has atol convergence with quad precision
updated ex1_nest tests
fixed incorrect key value
fix: bad merge a69d34a
test: move runex78 to TESTEXAMPLES_C_NOTSINGLE
build: fix warnings
sys: update test to produce same result - irrespective of debug/opt build - and presence of ~/.petscrc
sys: src/sys/examples/tests/ex30.c -> src/sys/examples/tests/ex32.c to avoid conflict in master
Sed output, since quad precision reduces below atol on these tests
remove automatically generated matrix names from test output files since can be different
Fixes for MatMPIXXXSetPreallocation(), they were not reentrant, since they kept the information used for matrix-vector products
GAMG: fix PCView_GAMG() format - and update output/ex56_0.out
superlu_dist: 64-bit-indices fix https://github.com/xiaoyeli/superlu_dist/issues/4
Add SNESSetDivergenceTolerance / -snes_divergence_tolerance.
fixed unit read in Plex
...

show more ...

Merge branch 'stefano_zampini/feature-convert-matnest-matis' into stefano_zampini/feature-pcbddc-saddlepoint

I have to merge it here, since there's an API addition for MatIS in stefano_zampini/featu

Merge branch 'stefano_zampini/feature-convert-matnest-matis' into stefano_zampini/feature-pcbddc-saddlepoint

I have to merge it here, since there's an API addition for MatIS in stefano_zampini/feature-pcbddc-saddlepoint
that prevents me to smoothly merge stefano_zampini/feature-convert-matnest-matis into next.

This will also facilitate merging stefano_zampini/feature-pcbddc-saddlepoint into master

Conflicts:
src/docs/website/documentation/changes/dev.html

show more ...