| Commit message (Collapse) | Author | Age | Files | Lines |
| |
|
|
|
|
|
|
|
|
|
|
|
| |
Description: Created tar ball for testing and changed version number
to 1.6.5-pre3
Solution: ran bin/h5vers script on heping
Platforms tested: N/A
Misc. update:
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Bug fix
Description:
Absoft needs to be built with --disable-shared only on Macintosh.
Solution:
Moved check for --disable-shared from fortran/configure.in to
config/powerpc-apple.
Platforms tested:
mir, heping, pommier
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Bug fix.
Description:
PGI v6.0-5 pgf90 compiler has a bug in Mstandard that caused compiling
of legal logical value to fail.
We also have second thought of imposing -Mstandard and -Mdclchk on users
since these flags are included in h5fc.
Solution:
Removed -Mstandard until we can fix the v6.0-5 compiler error
and moved -Mdclchk to DEBUG_FFLAGS so that it is used during development
mode but not imposed on users for production release.
Platforms tested:
heping: production mode using v6.0-4 and v6.0-5 compilers. Debug mode using
v6.0-4 compiler.
mir: production mode using v6.0-4 and v6.0-5 compilers. Debug mode using
v6.0-5 compiler.
|
|
|
|
|
|
|
|
|
|
|
| |
Update copyrights.
Description:
Solution:
This should address almost all of the "simple" cases in the repository.
There's still work to do, but it's going to require actually thinking about
the files in question instead of just copying & pasting.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Purpose:
Configure feature and cleanup
Description:
Added error when Absoft f95 is used without --disable-shared.
Removed tools/misc/h5import.c and tools/misc/pdb2hdf.c .
Platforms tested:
mir heping
Misc. update:
|
|
|
|
|
|
|
| |
Description:
Add copyright notices.
Platforms tested:
Ran hdf5/chkcopyright successfully.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Appearance change.
Description:
depend1.in:
Change the style of the copyright generation commands to a simplier
style.
Dependencies:
Regenerated the Dependencies file.
Platforms tested:
Tested in heping in serial and parallel modes.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Copyright fix
Description:
Added copyright to C++ and Fortran dependency files in 1.6.
Solution:
copies hdf5/config/depend1.in to hdf5/c++/config and hdf5/fortran/config.
Platforms tested:
heping
Misc. update:
|
|
|
|
|
|
|
|
|
|
| |
Description: Changed the version number to 1.6.5-pre1
Solution: Use bin/h5vers script to change relevant files
Platforms tested: heping
Misc. update:
|
|
|
|
|
|
|
|
|
|
|
|
| |
Copyright fix
Description:
Added copyright to generated Dependency files.
This had the side effect of updating some Dependency files.
Platforms tested:
heping
|
|
|
|
|
|
|
|
|
|
|
|
| |
Description: Added configuration files for HPUX 11.23
Solution:
Platforms tested: HPUX 11.23 with 32-bit and 64-bit for C and C++
32-bit for Fortran
kelgia
Misc. update:
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Purpose: Maintenance/ HPUX11.23 port
Description: Added configuration files for HPUX11.23 and fixed
Fortran H5f90i.h to support HPUX11.23
Solution:
Platforms tested: HPUX 11.23 C, C++ 32 and 64-bit, Fortran 32-bit only
kelgia
Misc. update:
|
|
|
|
|
|
|
|
|
|
|
|
| |
Description: Created a tar ball for testing and changed version
number to 1.6.5-snap3. I didn't change it to "pre"
since we usually do it after "code freeze" announcement.
Solution: Ran bin/h5vers -s 1.6.5-snap3 on heping
Platforms tested: N/A
Misc. update:
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
feature
Description:
Added code to detect what fortran compiler the MPICH kind of mpif90
compiler scripts are actually using. This allows configure to adjust
the C-stub function names according to what the Fortran compiler
wants.
Platforms tested:
Heping serial and PP (with mpich-ifort and mpich2-g95.)
Also hand tested with mpich-pgf90.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
feature
Description:
Ported fortran to use the g95 fortran compiler. Followed the example
of Absoft compiler to set -DH5_G95 to make sure C-stub routines are
compiled with what g95 wants (xyz__).
Platforms tested:
h5committested.
Also tested in heping using intel compilers (C,Fortran,C++).
Unfortunately, mpich2/mpif90 still does not work.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
New feature.
Description:
Added the time command to the make check target to report time usage
of the execute of each test and test scripts. This gives us some idea
how long each test takes and some vague idea it is compute bound or
not.
powerpc-ibm-aix5.x:
Change $RUNPARALLEL default setting to allow it being invoked by the
time command.
Platforms tested:
h5committested.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Feature
Description:
Created H5_CFLAGS, H5_CPPFLAGS, H5_CXXFLAGS, and H5_FFLAGS variables.
These hold flags that are used to build hdf5, but won't be passed on to
h5cc (or h5fc, or h5c++).
Currently there are no flags included; this is just the framework for such
flags.
Platforms tested:
mir, modi4, sleipnir
Misc. update:
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Bug fix.
Description:
Many debug purposes compiler options are set in $CFLAGS even in production
mode. All these are copied to the compile commands (h5cc, h5fc, ...)
and imposed upon users whether they want those debug options or not.
Solution:
Moved most of the debug purpose options to $DEBUG_CFLAGS so that
they are not used (and imposed on users) for production mode.
Platforms tested:
heping.
|
|
|
|
|
|
|
|
|
|
|
| |
Cleanup.
Description:
Removed redundant option (-Minform,warn) from CFLAGS setting. It is
already the pgcc default.
Platforms tested:
tested in heping.
|
|
|
|
|
|
|
|
|
|
|
| |
Description: Changed version number to 1.6.5-snap0 in anticipation
of the new 1.6.5 release
Solution: Ran bin/h5vers script
Platforms tested: N/A
Misc. update:
|
| |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Code cleanup
Description:
Trim trailing whitespace, which is making 'diff'ing the two branches
difficult.
Solution:
Ran this script in each directory:
foreach f (*.[ch] *.cpp)
sed 's/[[:blank:]]*$//' $f > sed.out && mv sed.out $f
end
Platforms tested:
FreeBSD 4.11 (sleipnir)
Too minor to require h5committest
|
| |
|
| |
|
| |
|
| |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
bug fix.
Description:
test and clean targets did not include $(TEST_PROGS_PARA).
Added it in.
Platforms tested:
Heping pp mode.
Misc. update:
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Cleanup.
Description:
The parallel fortran test was entered under $TEST_PROGS which is for
serial tests. This worked because $RUNTEST is hardset to
$RUNPARALLEL.
Solution:
Moved the parallel test to $TEST_PROGS_PARA and removed the
setting of $RUNTEST.
Platforms tested:
Tested in heping in PP mode only as this is pretty trivial.
Misc. update:
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
feature.
Description:
Added two new make check targets,
check-s: serial tests
check-p: parallel tests.
Check will run check-s and check-p, have the same effect as before.
Added $TEST_SCRIPTS_PARA for parallel test scripts.
Platforms tested:
h5committested".
Also, tested in heping (serial and pp) to verify with previous
test output for consistency.
Tested in modi4 for the special case of pmake (in progress.)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Code cleanup
Description:
On Cray X1, a loop in the Fortran C source confuses the compiler's optimizer
and hangs the build.
Solution:
Documented the fix in the code, and added an #ifdef so that
the optimization is only disabled on Cray X1.
Platforms tested:
mir (fix was already checked into code; this was a minor change)
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Cray X1 port
Description:
Added configuration files (config/nv1-cray) and made some minor changes
so that the 1.6 branch will build and pass tests on Cray X1 machines.
Solution:
hl/test/test_table will never compile if compiler optimizations are
turned on. Optimization in test_table.c is thus disabled when __crayx1
is defined.
Platforms tested:
Cray X1s, eirene, sleipnir
|
|
|
|
|
|
|
|
|
|
|
|
| |
bug fix
Description:
Removed some errorous duplicated code segment.
Platforms tested:
heping PP only. simple change.
Misc. update:
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Bug fix.
Description:
When there were errors in the test, program still just call MPI_FINALIZE
and failed to attempt to exit with some error code so that calling
programs like make be informed of the exceptions.
Solution:
Call MPI_ABORT if error is detected. Though MPI_ABORT does not
guarantee 100% failure report, it has the best chance.
Also made dimension incompatible as a real error.
Platforms tested:
Did not h5committest but tested in heping PP only since this
is a simple fix.
Misc. update:
|
|
|
|
|
|
|
|
|
|
|
|
| |
Description: parameter "driver" to the h5pget_driver_f function did not
have type declaration. copper mpicc didn't catch the problem
but Intel mpicc on tg-login machine did.
Solution: Added type declaration
Platforms tested: None, too obvious and too late :-)
Misc. update:
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Description: h5pget_driver_f function returned information that could not
be interpreted by fortran application
Solution: Defined Fortran global variables that correspond to
the C H5FD_<driver_name> variables at the Fortran library
initialization time.
Platforms tested: heping with PGI compilers, sol 64-bit and copper parallel
Misc. update:
|
| |
|
|
|
|
|
|
|
|
|
|
|
| |
Description: Added missing h5get_member_class_f function
Solution:
Platforms tested: heping; it is a simple change. I will
watch daily tests and take care of failures.
Misc. update:
|
| |
|
|
|
|
|
|
|
|
|
|
|
| |
Description: Edited HISTORY.txt and RELEASE.txt files;
changed version number to hdf5-1.6.4-post0
Solution: ran bin/h5vers on heping
Platforms tested:
Misc. update:
|
|
|
|
|
|
|
|
|
|
|
|
| |
Description: I ran bin/h5vers to change version number to 1.6.4. Several files
were modified by bin/h5vers. I am checking them in before
tagging the release with hdf5-1_6_4 tag
Solution:
Platforms tested:
Misc. update:
|
|
|
|
|
|
|
|
|
|
|
|
| |
Description: Changed suffix to pre3 after creating a tar ball for the
second round of testing
Solution: Ran bin/h5vers on heping; made sure that files were modified
by autoconf 2.59
Platforms tested: heping
Misc. update:
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Bug fix
Description:
HDF5 1.6 was not using the correct version of autoconf.
Solution:
Updated bin/reconfigure script to configure all three configure.in files
using version of autoconf in AFS. Updated bin/h5vers script to use
reconfigure script.
***
CHANGED NAME OF reconfigure.sh TO reconfigure.
***
Used bin/reconfigure to update configure files.
Platforms tested:
mir, eirene, modi4, copper, arabica
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
round of testing
Description: Changed suffix to pre2
Solution: Used bin/h5vers -s command on eirene to modify
the version number (1.6.4-pre2)
Platforms tested: N/A
Misc. update:
|