Purpose: Maintenance for the 1.6.0 release.
Description: I updated compilers information for Crays and added
h5fc and h5c++ to the new tools sections in REELASE.txt.
I added instructions for how to build HDF5 using Intel compilers
on Windows.
Solution:
Platforms tested:
Misc. update:
Purpose:
Add short copyright notice.
Update release tag line.
Description:
Added short copyright notice as comment in source files;
does not display in browser.
Updated release tag line in footers to read as follows:
Describes HDF5 Release 1.6.0, July 2003
Platforms tested:
IE 5
Purpose:
Add copyright
Update release number
Description:
Added short copyright statement (doc version).
Updated release number and date to Release 1.6.1, July 2003.
Platforms tested:
IE 5
Purpose:
H5Pset/get_hyper_cache are deprecated in R 1..6.0.
Description:
Added notation that H5Pset/get_hyper_cache are deprecated functions
in R1.6.0 and are available only if library is configured for
1.4 compatibility.
Platforms tested:
IE 5
Description:
Add H5_WANT_H5_V1_4_COMPAT notes.
Add or expand "Changed Syntax" and "Deleted Functions" sections.
Edited to refer to Release 1.6.0 (instead of R 1.5 Developoment Branch).
Bring in sections from R 1.4 branch covering changes
from R 1.2.2 through 1.4.5.
Platforms tested:
IE 5
Flag adding for .Net
Description:
Somehow in .Net environment, libc.lib and libcd.lib are used simultenously and that causes
system function redefined. Thereafter, HDF5 build failed.
Solution:
explicitly set an "ignore" flag to ignore libcd.lib for release build.
Platforms tested:
.Net at xp and VC++ 6.0 at windows 2000
Misc. update:
Description: When Quincey and I updated Fortran SZIP flags, we accidentally
renamed one of the C flags H5_SZIP_ALLOW_K13_OPTION_MASK used
by Fortran C-stub routine.
Solution: Fix the typo.
Platforms tested: Linux 2.5 and True64
Misc. update:
Purpose:
Update h5pset_szip_f options mask.
Description:
h5pset_szip_f: Updated list of valid szip option values.
Deleted 2, added 3 and notation "or the sum of any combination".
Added example and note regarding mutually-exclusive options.
Platforms tested:
IE 5
feature change, new feature.
Description:
Had the HDF5_PARAPREFIX default set to /tmp.
Changed it to null so that data files are created in current directory
by default.
Also added a code to print data filename used when debug level > 0.
Platforms tested:
Tested in modi4 parallel.
Misc. update:
Code cleanup
Description:
Add in rest of szip "options mask" macros that were missing. Also made
"raw" options mask set by the library, instead of requiring users to always set
it.
Platforms tested:
FreeBSD 4.8 (sleipnir)
Minor tweaks too small fo h5committest
Purpose:
H5Pset_shuffle -- Add note regarding intended use.
H5Pset_szip -- Correct options mask list. Copy edits.
Description:
H5Pset_shuffle -- Added note that the shuffle filter is designed to be
used with in combination a compression filter.
H5Pset_szip -- Changed 'szip' to 'SZIP' in running text.
Removed RAW_OPTIONS_MASK, LSB_OPTIONS_MASK, and MSB-OPTIONS_MASK
from options maks list as they are not available for user control.
Added 'H5_SZIP_' prefix to all remaining option tags.
Made corresponding changes to running text, removing or revising
mentions of deleted option tags.
Platforms tested:
IE 5
Purpose:
Bug Fix
Description:
LIBTOOLS macro was being used but not defined.
Solution:
Removed the LIBTOOLS macro as it wasn't necessary
Platforms tested:
Modi4 (small fix)
Misc. update:
Bug Fix
Description:
The SUBDIRS macro is now defined as "" in the config/commence.in
file. We shouldn't have definitions of this macro before the
inclusion of config/commence.in.
Solution:
Placed all definitions of SUBDIRS after config/commence.in inclusion.
Platforms tested:
Modi4 (Small fix yet again).
Misc. update:
Code cleanup/Feature removal
Description:
Remove "round robin" writing of metadata in MPI VFDs (and a couple of other
places). This never really worked the way I hoped it would and it causes
problems on certain machines.
Platforms tested:
FreeBSD 4.8 (sleipnir) w/parallel
Too small for triple check.
Purpose:
Bug Fix
Description:
Changes needed to make "pmake" work on Modi4.
Solution:
Needed to add some null macro defines in the commence.in files.
Needed to remove a dependence on LIBH5TEST in the fortran make
since LIBH5TEST isn't used in the fortran stuff.
Platforms tested:
Modi4 (small change)
Misc. update:
Bug fix.
Description:
There was a bug fix for AIX machine when Poe is used to
launch programs, that is in parallel mode. The fix is hostname
dependent.
Solution:
Removed the hostname dependence and let the fix apply to all
AIX machines. But put in a safety net--if $HDF5_AIX_NOPATCH
is set to non-null, it will not use the fix but use the
original code. (Don't expect this being used.)
Platforms tested:
Tested in Copper both with and without $HDF5_AIX_NOPATCH set.
Did not run h5committest since this really applies to AIX
systems only.
Misc. update:
Purpose: Maintenance
Description: I added three new configuration files in c++/config
directory for Cray SV1, T3E and T90IEEE
Solution:
Platforms tested: I used BlankForm in c++/config directory to create
new configuration files for Crays. Only test
on T3E is finally completed successfully. Two other
Crays are still running, and running, and running.
I am checking files in since they will not affect
any other platforms.
Binh-Minh and Bill,
could you please take a look since I made minimum changes
just to make it to work. I am not sure that everything is correct
even if it works.
Thanks!
Misc. update:
Purpose: Maintenance
Description: I added three new configuration files in c++/config
directory for Cray SV1, T3E and T90IEEE
Solution: Modified MANIFEST to add new files.
Platforms tested:
Misc. update:
Purpose: Bug fix
Description: Makefile.in has hardcoded dependencies. H5Zff.lo was
missing for the sequential version. That caused parallel
make to fail, since module HDF5 was built before H5Z.
Solution: Added H5Zff.lo to dependencies list.
Platforms tested: This is an obvious error and also due to the seminar and picnic
schedule I am checking in without testing. I will test it later today.
Misc. update:
Purpose: Small bug fix
Description: On SP when 64 bit mode is used, RUNPARALLEL environment
variable is not set up correctly for Fortran.
Solution: Copy Albert's fix for C library to Fortran
Platforms tested: copper with 64-bit parallle C and Fortran
Misc. update:
Purpose:
Primarily edits from technical review
Description:
H5Dread
H5Dwrite
Clarify library behavior with certain mem_space_id/file_space_id
combinations.
Add comment regarding when space is allocated and fill values written.
H5Dget_space_status -- Correct return value.
Copy edits and formatting fixes.
Platforms tested:
IE 5, Safari
Purpose:
Primarily edits from technical review.
Description:
H5Pget_fill_value -- Corrected 'value' parameter type.
H5Pget_fill_time -- Added H5D_FILL_TIME_IFSET.
H5Pset_filter --Add that filters are applied to chunks in order
in which they were created.
Correct H5Z_FLAG_OPTIONAL description.
Add references to related functions.
Add H5Z_FILTER_SZIP to list of valid filters.
H5Pget_filter -- Add H5Z_FILTER_SZIP to list of possible filters.
H5Pset_szip -- Remove list of publications references.
H5Pset_edc_check -- Reverse notation of default setting.
Copy edits.
Platforms tested:
IE 5, Safari
Code cleanup
Description:
Make the default for the function stack work like the default for the
API tracing: when debug is enabled, the function stack defaults to on and
when debug is disabled, the function stack defaults to off.
Platforms tested:
FreeBSD 4.8 (sleipnir)
h5committest
Bug fix
Description:
Missing returning error for failures occur in C++ examples.
Solution:
Added "return -1;" when an exception is caught.
Also, turned off automatic error printing so that the C++ API
will catch and handle the failures.
Platforms:
SunOS 5.7 (arabica)
Linux 2.4 (eirene)
IRIX 6.5.11 (modi4)
Purpose: Maintenance
Description: Bumped the version number to 1.5.62 after creating
a tar ball for the fourth round of testing.
Solution: run bin/h5vers -s scipt on arabica
Platforms tested: N/A
Misc. update:
Improve test a bit
Description:
Add a small bit of testing for the array field in a compound datatype.
Platforms tested:
FreeBSD 4.8 (sleipnir)
h5committest
Cray SV1 (wind)
Cray T3E (hubble)
Cray T90 (gypsy)
Bug fix
Description:
Alignment of some types (namely 'short's) was still not being correctly
detected on some Crays.
Solution:
Revert part of last change which accidentally got the offset of 'short's
wrong on Crays
Correct check for alignment of 'short's on Crays.
Platforms tested:
FreeBSD 4.8 (sleipnir)
h5committest
Cray SV1 (wind)
Cray T3E (hubble)
Cray T90 (gypsy)
Bug fix
Description:
Aligned data is not being detected by type conversion routines correctly
on some Crays, causing corrupted data during I/O.
Solution:
Add extra checking for strange pointer behavior on Crays.
Platforms tested:
FreeBSD 4.8 (sleipnir)
h5committest
Cray SV1 (wind)
Cray T3E (hubble)
Cray T90 (gypsy)
Code cleanup
Description:
Include roughed-in, but commented out code for performing I/O on time
datatypes.
Platforms tested:
FreeBSD 4.8 (sleipnir)
h5committest
Cray SV1 (wind)
Cray T3E (hubble)
Cray T90 (gypsy)
Code cleanup
Description:
Remove the fill-value option entirely, since the 1.5.x library's default
behavior is now the same as the 1.4.x behavior and also there is no
corresponding code for implementing fill-values with the POSIX or MPI-I/O tests.
Platforms tested:
FreeBSD 4.8 (sleipnir) w/parallel
too small to need h5committest
Update
Description:
Changed the --enable-mpe flag to --with-mpe so that we could
specify the directory information for where the MPE stuff is
kept. (with the =DIR option that we have for --with-* flags)
Solution:
7%
Platforms tested:
Copper (where MPE is...small change no requiring h5committest).
Misc. update:
