Bug fix
Description:
Fix "double root open" bug and apply same fix to datasets & named datatypes.
Also fix bug in symbol table entry lookup code that would "transfer" an
object from one file handle to another file handle when the two file handles
were opened on the same actual file.
Platforms tested:
FreeBSD 4.11 (sleipnir)
Too minor to require h5committest
Bug fix.
Description:
mpich2 using f95 did not support function 'unlink'.
Removed the unlink call for now. Need a better solution.
Platforms tested:
heping using mpich and mpich2 versions of mpif90.
Misc. update:
Code cleanup
Description:
Clean up & optimize (a little) the recent "hyperslab rebuilding" code
w/Kent.
Platforms tested:
FreeBSD 4.11 (sleipnir)
Too minor to require h5committest
bug fix.
Description:
The complex derived datatype test assumed that the fill value would be 0. This is not
the case on all systems.
Solution:
Modified the test to check against a known value in the outbuf array, instead of the fill value.
Platforms tested:
heping and MCR.
Misc. update:
bug fix.
Description:
Fixed typo in a comment. The word "file" was supposed to be "fill." The explanation of
how the complex derived datatype test works is much clearer now.
Solution:
Platforms tested:
minor change.
Misc. update:
bug fix.
Description:
When a parallel test script test fails, make would continue because the
way it was setup inside a for loop. Fixed it by issuing an exit 1 inside
the loop.
There was also a typo error in the newer command comparision that it
must be $${chkname} in order to be valid. Also, the test script itself
was not checked in the newer lists. All fixed.
Platforms tested:
h5committested and also hand tested in heping pp mode.
Bug fix
Description:
In production builds, H5MM_malloc was undefined in H5AC.c
Solution:
Include H5MMprivate.h in H5AC.c, which defines H5MM_malloc() in
both debug and production modes.
Platforms tested:
h5committested, serial production build on Heping
Misc. update:
revision.
Description:
The copyright notice for Makefile(in) were fixed at the beginning of the files.
But they are now generated by automake and autoconf such that Gnu jams its
own copyright block at the beginning.
Solution:
Changed the rule such that the HDF5 copyright notice can be somewhere else
in the file as long as they are together as one paragraph.
Also added recognition of Makefile.am as a possible MAKEFILE type.
Platforms tested:
Tested by hand in heping.
Repair synchronization bug in the metadata cache in PHDF5
Also repair numerous other bugs that surfaced in testing the
bug fix.
Description:
While operations modifying metadata must be collective, we allow
independant reads. This allows metadata caches on different processes
to adjust to different sizes, and to place the entries on their dirty
lists in different orders. Since only process 0 actually writes
metadata to disk (all other processes thought they did, but the writes
were discarded on the theory that they had to be collective), this made
it possible for another process to modify metadata, flush it, and then
read it back in in its original form (pre-modification) form. The
possibilities for file corruption should be obvious.
Solution:
Make the policy that only process 0 can write to file explicit, and
visible to the metadata caches. Thus only process 0 may flush dirty
entries -- all other caches must retain dirty entries until they are
informed by process 0 that the entries are clean.
Synchronization is handled by counting the bytes of dirty cache entries
created, and then synching up between the caches whenever the sum
exceeds an (eventually user specified) limit. Dirty metadata creation
is consistent across all processes because all operations modifying
metadata must be collective.
This change uncovered may bugs which are repaired in this checkin.
It also required modification of H5HL and H5O to allocate file space
on insertion rather than on flush from cache.
Platforms tested:
H5committest, heping(parallel & serial)
Misc. update:
Code refinement
Description:
Make the re-build hyperslab selection code shorter to be maintained easily
Solution:
Platforms tested:
Sol 2.9, linux 2.4, Aix 5.1
Misc. update:
Feature.
Description:
Added the copyright checking--most settings are temporary. Need
cleanup or fixes when most copyrights notices are completed.
Platforms tested:
Tested by hand.
Maintenance on Windows
Description:
Modify Batch file due to the changes of output of H5diff tool
Solution:
Platforms tested:
MSVS 6.0 on Windows XP
Misc. update:
Code cleanup
Description:
Removed Fortran type sizes from config/* files.
Solution:
These lines weren't doing anything anymore; they've been replaced by automatic
type generation. This is just housekeeping to remove them from
the config/* files.
Platforms tested:
mir, sol, modi4
A bug fix
Description:
MPI_Status_IGNORE is treated as a NULL pointer for mpich 1.2.4 or similar MPI packages.
It caused segmentation fault for MPI derived datatype test.
Solution:
Define MPI_STATUS status,
and pass &status into MPI_File_read and MPI_File_write.
Platforms tested:
too trivial to test.
Misc. update:
Improvement.
Description:
The test may hang if there are system failures that some processors
are not working.
Solution:
Added the ALARM calls to limit all tests be done with the default
alarm time. So, even if a process is hanging, the ALARM signal
would terminate the process.
Platforms tested:
tg-login2 of NCSA.
Misc. update:
Maintenance on Windows
Description:
Add in two new project files (h5stat.dsp and h5statdll.dsp)
Solution:
Platforms tested:
Visual Studio .Net on Windows XP
Misc. update:
bug fix
Description:
the MPI_File_get_size returns a different value in one of the tests in Windows
comment the code and not run the test in windows
a ULL suffix on the harcoded return VRY return number is needed on AIX
Solution:
Platforms tested:
Windows
AIX
Misc. update:
new features
Description:
add support for compiling the library and testphdf5 in Windows
Solution:
Platforms tested:
Linux
AIX
Solaris
Windows VC6
Misc. update:
Hyperslab re-build optimization
Description:
Previously only singluar selection can be rebuilt from span-tree application.
This optimization is supposed to rebuild for other regular selection case(such
as count =2, block = 2, stride =3)
Solution:
Platforms tested:
shanti(sol 2.9), heping(Linux 2.4), NCAR SGI(IRIX 6.5)
Misc. update:
h5diff new feature
Description:
a user asked for the message
"Some objects are not comparable"
to be more noticeable
Solution:
--------------------------------
Some objects are not comparable
--------------------------------
Platforms tested:
linux
Misc. update:
Updated release notes
Description:
Mentioned that shared C++ and Fortran libraries are supported in
INSTALL and RELEASE.txt.
Platforms tested:
documentation change only
new features
Description:
added support for the scale/offset filter
there is a new filter symbol 'SOFF'
-f SOFF=<scale_factor,scale_type>
scale_factor = integer
scale_type = 'IN' or 'DS'
Solution:
Platforms tested:
Linux
SunOS
Misc. update:
Feature.
Description:
Added ALARM_ON, ALARM_OFF feature to terminate tests that run more than
the default time limit. So far, the feature is applied in the standard test
frame.
Platforms tested:
heping PP.
Bug fix
Description:
Using the -xarch=v9 flag to specify 64-bit compilation breaks C++ shared
libraries.
Disable shared C++ libraries if this flag is used.
Solution:
Instead of trying to disable compilers which recognize the -xarch flag,
grep for it in CXX, CXXFLAGS, etc.
Platforms tested:
modi4, shanti, sleipnir
Misc. update:
Update to RELEASE and build instructions will be forthcoming once
shared Fortran/C++ libraries are regularly passing the Daily Tests.
Description:
I couldn't get the HDF logo to appear, when the location for the
html files generated by doxygen is arbitrary. This arbitrary
directory doesn't contain the logo image file, which would be
placed in a planned directory.
The logo is removed now, but the image file is kept in the CVS
under hdf5/c++/src, just in case when I can find a solution.
Platforms tested:
Linux 2.4 (heping)
Windows XP
Misc. update:
New feature.
Description:
Command script to build the h5perf tool by standalone mode.
Platforms tested:
Tested by hand in heping using pre-release h5pcc.
Purpose: Typo fix
Description: Declaration of long double variable was after executable
statement. Many compilers choked ;-(
Solution: Declare variable before the first executable statement
Platforms tested: shanti (compilation only), copper (tests passed)
Misc. update:
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.
Bug fix.
Description:
In some racing condition, the process ID file might be removed just
the moment before timekeeper cats its content. When that happens,
the $pid is a blank. Timekeeper was not prepared for this.
Solution:
Added code to verify $pid is some number before proceeding.
Platforms tested:
Tested by hand only.
Description:
Fortran multi_file test failed on SX-6. It turned out that compiler
could not handle (long long HADDR_MAX) * ( float var). The result of this
operation was 0.
Solution:
Introduced long double variable to store the value of HADDR_MAX
and use the new variable in * operation. (Suggested by John)
Platforms tested: SX-6, mir with PGI compilers; a very minor change
Misc. update:
Description:
On VMS dsets test failed screaming about corrupted stack.
Debugging didn't show any obvious problems. Careful examination
of compilation warning messages (something like "function is used as
data in dsets") made me to search for duplicated names.
It turned out that two variables H5Z_SET_LOCAL and H5Z_CAN_APPLY
were confused with H5Z_set_local and H5Z_can_apply functions from
H5Z.c file.
Solution:
Changed variable names to H5Z_SET_LOCAL_TEST and H5Z_CAN_APPLY_TEST.
Platforms tested: Alpha VMS, heping; minor change
Misc. update:
