HDFFV-944: Fortran test problem with HDF5 182, intel 11.0, open-mpi-1.2.8
Description:
Moved libtool post-processing 'ed' snippet that replaces wl=""
with wl="-Wl," when intel compilers are used to the end of the
configure script, after libtool is actually generated. (I imagine
that at some point in time during a libtool version upgrade the
libtool script ceased being generated immediately after the libtool
initialization macro and is generated at the end of configure with
all other files).
Additionally, added a leading 'g' to the replacement invocation to
change all instances of wl="" to wl="-Wl," instead of just the
first that it comes across. (There's a separate one for each
compiler used (C, Fortran, C++)).
Tested:
Reproduced the reported make failure and tested resolution on
NCSA's Forge with OpenMPI built with Intel 12.0.4.
Also, h5committested.
HDFFV-398: h5cc doesn't work with automake
Description:
Fixed compiler wrapper scripts to correctly detect compilation when
-MT preprocessor flag is provided, fixing a bug in which its *.o
argument was added to link_objs but not compile_args.
This previously broke usage of h5cc as the provided compiler in
configure scripts (like with h5edit) as automake may supply the -MT
option to the compiler via the makefiles.
Tested:
Tested using h5edit and h5committested.
update for HDFFV-7837 - h5diff: incorrect behavior with exclude-path option when unique object exist only in one file
Description:
Add the new test output list to test shell script.
This is based on the previous commit r22114.
Tested:
jam (linux32-LE), koala (linux64-LE), ostrich (linuxppc64-BE)
Description:
Closed various HDF5 objects in DataSet::getInMemDataSize and
Attribute::getInMemDataSize to remove some memory leaks.
Platforms tested:
Linux/32 2.6 (jam)
Linux/64 2.6 (amani)
SunOS 5.10 (linew)
when it fails to allocate memory. I didn't add any test to the test suite since there is no good way to test it. But I tested and verified the error stack by hand.
Tested on jam, koala, ostrich.
Propagates the SEC2 driver fixes from HDF5 1.8.8 to the core VFD (mainly concerning the backing store). These fixes also conveniently fixed 7603 as well.
Tested on:
64-bit Windows 7
jam
koala
ostrich
Fix for HDFFV-7837 - h5diff: incorrect behavior with exclude-path option when unique object exist only in one file
Description:
If unique objects exists only in one file and try to exclude
the unique objects with --exclude-path option, h5diff missed
excluding some objects.
Fixed to exclude objects correctly in such case.
Test cases were added and tagged with jira#.
Tested:
jam (linux32-LE), koala (linux64-LE), ostrich (linuxppc64-BE), tejeda (mac32-LE), linew (solaris-BE), Windows (32-LE cmake), Cmake (jam)
Minor code safety issue in test/fheap.c and whitespace in other files.
Tested on:
Mac OSX/64 10.7.3 (amazon) w/debug
(Too minor to require h5committest)
HDFFV-7523 JPSS: h5dump fails to open an attribute when using the -a option to specify an attribute with a slash in the name
Description:
Update command line help page and test generating file.
Previous commits: r22052, r22063
Tested:
jam (linux32-LE), koala (linux64-LE), cmake (jam)
This occurs due to the istore test creating very large files on systems which
do not have POSIX-like sparse file semantics. The large amount of I/O causes
the test to run for a very long period of time.
The fix was to copy the "big" test's sparse file check and only run the
largest sparse file test when POSIX-like sparse file semantics are found.
Tested on:
jam (nfs)
ostrich (nfs)
loyalty (ufs)
64-bit linux VM (ext4)
64-bit OS-X Lion (hfs, detected as not POSIX sparse)
64-bit Windows 7 (NTFS, detected as not POSIX sparse)
The OS-X failure to pass the sparse check is likely an error and will be
entered as a new bug.
The parser mistaked "-" as substraction. I fixed it and also fixed another problem
with some special cases like 100-x and 2/x.
Tested on jam, koala, and ostrich.
Description:
When using the new object header format, it was possible for corruption to occur
if the first object header chunk changed size such that the lenght of the "chunk
0 size" field changed. This only occurred if there were messages that had not
been decoded. The original algorithm that changed the object header chunk size
marked all messages as dirty, causing those that had not been decoded to have
both the raw and native form invalidated. Changed the algorithm to avoid
marking messages dirty and added assertions to catch the case where messages
are dirtied without being decoded (or recently created) first.
Tested: jam, koala, ostrich (h5committest), durandal
