[svn-r15674] Description: Add base support for extensible array "data blocks" to code, tests and h5debug.Tested on:
Mac OS X/32 10.5.4 (amazon) in debug mode
Mac OS X/32 10.5.4 (amazon) w/C++ & FORTRAN, w/threadsafe,
in production mode
FreeBSD/32 6.3 (duty) in debug mode
FreeBSD/64 6.3 (liberty) w/C++ & FORTRAN, in debug mode
Linux/32 2.6 (kagiso) w/PGI compilers, w/C++ & FORTRAN, w/threadsafe,
...
[svn-r15668] Purpose: Add feature requested in bug #1282Description: Adds capability to h5ls to traverse external links when the -r
(recursive) option is given. Changes to the way absolute path names are patched
in h5trav.c. Changes to the way recursive traversal starting from a non-root
group is handled (which also fixes some preexisting issues). Tests added for
these cases.
Tested: kagiso, smirom, linew (h5committest)
[svn-r15665] Bug fix: (ID 1157)Description:
Program would crash, complaining MPI calls were invoked after
MPI_Finalize() has occurred. Previously, H5close() before
MPI_Finalize() would remove the crash.
Solution:
It turned out that two H5 property objects (mpio_pl and acc_tpl)
were not closed before MPI_Finalize(). In the at_exit code,
HDF5 library attempted to close them by releasing the MPI Communicators
in them too. Tha...
[svn-r15661] Purpose: Add szip/zlib link dependencies to Windows earray[dll] projectsDescription:
earray[dll] builds were failing on Windows because of missing link dependencies to szip and zlib libraries (amongst other reasons). This checkin adds those dependencies.
[svn-r15660] Purpose: Fix minor Windows issues in new H5EA interface.Description:
We've run into a few Windows-specific problems with the new H5EA interface. In particular, this checkin makes the following changes:
-- Change shutdown defined in earray.c to shutdown_ea, to avoid conflicting with another Windows-specific system call.
-- Prefix H5EA_get in H5EAprivate.h with H5_DLL, so the function is properly exported in Windows DLL version.
These changes should...
[svn-r15659] Purpose: Fix Windows Fortran DLL definitions fileDescription:
We recently moved the Windows-specific fortran source code into a separate file for specifying DLL exports. However there were a couple definitions missing in the port from 1.8 to the trunk branch. This checkin correctly includes the .def file into our Windows project, and adds the missing definitions to hdf5_fortrandll.def.
Tested:
VS2005 on WinXP
[svn-r15658] Purpose: Replace __func__ usage with FUNC macroDescription:
Some compilers (i.e. Visual Studio on Windows) don't support the C99 keyword __func__. It does, however, have a similar keyword, __FUNCTION__. In H5EApkg.h we define a macro, FUNC, which will pick the right version and use it accordingly. This checkin changes H5EA.c to use FUNC instead of __func__.
Tested:
h5committest (linew smirom kagiso)
[svn-r15656] Purpose: fix bug 1286Description: Added configure test to see if pointer alignment restrictions are enforced (as in dereferencing an unaligned pointer causes an error). Added code in H5Tvlen.c to avoid dereferencing unaligned pointers, conditionally compiled based on the configure test. Added test case in dtypes.c which would previously cause such machines to fail.
Tested: kagiso, smirom, linew (h5committest); l...
[svn-r15650] Cleaned compiler warnings Bug fix: change the options block size variable from int to hsize_t, to match the H5Pset_userblock callTested: windows, linux
[svn-r15639] Purpose: Check for __func__ keyword during configureDescription:
In some of the new H5EA* code, many new C99 features are being used. Most of them are portable to all compilers, but the __func__ keyword in particular isn't supported in Visual Studio on Windows. Instead, Windows defines the __FUNCTION__ keyword, which can be used as a direct substitute.
We now check for the __func__ keyword during configure and define the feature flag H5_HAVE_...
[svn-r15637] Bug 710: partial fix.Temporary removed b2 from the --enable-debug=all list so that this option
will pass. "configure --help" lists the --enable-debug=all and users tend
to use it and end up with compile failures in the b2 debug code. Since this
bug will not be fixed soon, this is a temporary fix to let the configure and
build work.
Tested:
h5committest plus "--enable-debu=all" test in kagiso.
[svn-r15630] Description:In nh5tget_offset_c:
(1) The return value type of H5Tget_offset was set to size_t where
it should be of type int.
(2) Was if offset was equal to 0 it returned the error code of -1 back to
Fortran, this was changed to return an error code of -1 when the
offset value is < 0.
In h5tget_norm_c:
(1) was if the return value of H5Tget_norm = 0 it would return an error
code to Fortran,...
[svn-r15628] Description: Remove trailing whitespace from C/C++ source files, with the following script: foreach f (*.[ch] *.cpp)
sed 's/[[:blank:]]*$//' $f > sed.out && mv sed.out $f
end
Tested on:
Mac OS X/32 10.5.5 (amazon)
No need for h5committest, just whitespace changes...
[svn-r15627] This check-in fixes a bug inside parallel HDF5 testsuite. When enable-debug is turned on, a special macro block H5_HAVE_INSTRUMENTED_LIBRARY inside HDF5 will be executed to check if some collective chunk IO test cases are being run with the correct settings(one link, multiple chunk etc.). However,when complicated derived datatype in some mpi-io packages are not supported, the libra...This check-in fix this problem by providing a second property to avoid the false faiure message.
Tested at abe(NCSA linux cluster) and kagiso.
[svn-r15626] This check-in fixes a bug inside parallel HDF5 testsuite. When enable-debug is turned on, a special macro block H5_HAVE_INSTRUMENTED_LIBRARY inside HDF5 will be executed to check if some collective chunk IO test cases are being run with the correct settings(one link, multiple chunk etc.). However,when complicated derived datatype in some mpi-io packages are not supported, the libra...This check-in fix this problem by providing a second property to avoid the false faiure message.
Tested at abe(NCSA linux cluster) and kagiso.
[svn-r15622] Purpose: Add new source files to Windows projectsDescription:
This checkin adds the following source files to our Windows projects:
-- H5EAiblock.c
-- H5EAstat.c
-- H5EAdbg.c
[svn-r15618] Description: Add additional test and make a few cleanups in extensible array unit test.Tested on:
Mac OS X/32 10.5.4 (amazon) in debug mode
Mac OS X/32 10.5.4 (amazon) w/C++ & FORTRAN, w/threadsafe,
in production mode
FreeBSD/32 6.3 (duty) in debug mode
FreeBSD/64 6.3 (liberty) w/C++ & FORTRAN, in debug mode
Linux/32 2.6 (kagiso) w/PGI compilers, w/C++ & FORTRAN, w/threadsafe,
...
[svn-r15617] Description: Clean up how v2 B-tree (and extensible array) classes are determined in h5debug. Add support for extensible array headers and index blocks to h5debug
Tested on:
Mac OS X/32 10.5.4 (amazon) in debug mode
Mac OS X/32 10.5.4 (amazon) w/C++ & FORTRAN, w/threadsafe,
in production mode
FreeBSD/32 6.3 (duty) in debug mode
FreeBSD/64 6.3 (liberty) w/C++ & FORTRAN, in debug mode
Linux/32 2.6 (kagiso) w/PGI com...
[svn-r15614] Description: More extensible array features: add support for setting/getting elements in the array's index block.Tested on:
Mac OS X/32 10.5.4 (amazon) in debug mode
Mac OS X/32 10.5.4 (amazon) w/C++ & FORTRAN, w/threadsafe,
in production mode
FreeBSD/32 6.3 (duty) in debug mode
FreeBSD/64 6.3 (liberty) w/C++ & FORTRAN, in debug mode
Linux/32 2.6 (kagiso) w/PGI compilers, w/C++ & FORTRAN, w/threadsafe,
...
[svn-r15612] Description: Minor formatting cleanupsTested on:
Mac OS X/32 10.5.4 (amazon) in debug mode
Mac OS X/32 10.5.4 (amazon) w/C++ & FORTRAN, w/threadsafe,
in production mode
FreeBSD/32 6.3 (duty) in debug mode
FreeBSD/64 6.3 (liberty) w/C++ & FORTRAN, in debug mode
Linux/32 2.6 (kagiso) w/PGI compilers, w/C++ & FORTRAN, w/threadsafe,
...
[svn-r15609] Description: Omnibus compiler warning cleanup & some reformatting also.Tested on:
Mac OS X/32 10.5.4 (amazon)
Too minor to require h5committest
[svn-r15604] The MPI compiler(mvapich2-0.9.8p2patched-intel-ofed-1.2) at NCSA's linux cluster(Abe) fails to handle collective IO calls with very complicate MPI derived datatype. All the values of input parameters from HDF5 are correct before passing to MPI-IO collective calls. Sometimes the following errors happen:rank 0 in job 77 honest1.ncsa.uiuc.edu_60945 caused collective abort of all ranks
exit status of rank 0: killed by signal 9
Up to this point, we decide to turn off the use of advanced MPI derived datatype features inside HDF5 so that complicated selection tests are ensured to be handled by HDF5. This check-in reflects this.
[svn-r15588] Purpose: Add new h5repack tests to WindowsDescription:
New h5repack tests were added in revision 15551 and 15557, and also the syntax of one test was changed. This was causing errors on Windows. This checkin should fix those errors.
Tested:
VS2005 on WinXP
[svn-r15585] Purpose: Use H5test_kind_SIZEOF.f90 rather than H5test_kind.f90 in WindowsDescription:
A new version of H5test_kind is safer for platforms that support sizeof. We have replaced H5test_kind.f90 with it on Windows
[svn-r15584] Purpose: Add new extensible array source code and test project for WindowsDescription:
In revision 15549, new source code was added for the extensible array interface. This checkin adds the source code to Windows, along with the new test project, earray.
Note: The new H5EA interface causes compilation errors on Windows, and will need to be fixed in the source.
[svn-r15577] Description:Moved AM_CONDITIONAL([FORTRAN_HAVE_SIZEOF]....) outside of Fortran conditional loop so that the value gets registered regardless of whether Fortran is enabled.
[svn-r15573] Description:Compiles either H5test_kind.f90 or H5test_kind_SIZEOF.f90 depending on the availability of intrinsic Fortran function SIZEOF (indicated by the status of FORTRAN_HAVE_SIZEOF).
[svn-r15572] Desciption:Was not returning the correct fortran types when -i8 and -r8 flag was specified, fixed. This code is now depreciated and only used when SIZEOF function is not available, H5test_kind_SIZEOF.f90 should be used instead.
[svn-r15571] Description:H5test_kind_SIZEOF.f90 is the preferred portable program used for Fortran type detection. It uses the intrinsic Fortran function SIZEOF. This program should be used instead of H5test_kind.f90 if the SIZEOF function is available. Most of the newer Fortran compilers support SIZEOF.
Quincey Koziol
Scot Breitenfeld
Neil Fortner
Private HDF
Albert Cheng
Pedro Vicente Nunes
Scott Wegner
Christian Chilan
Muqun Yang