Adding parallel tests for optional collective chunk APIs
Description:
Three new APIs
"H5Pset_dxpl_mpio_chunk_opt_ratio
H5Pset_dxpl_mpio_chunk_opt_num
H5Pset_dxpl_mpio_chunk_opt"
for optional optimization choices from users
have been added to the libraries.
This check-in adds six tests to verify the funcationality and correctedness
of these APIs.
These tests need to be verified with 3 or more processors and with MPI-IO driver only.
Solution:
Using H5Pinsert, H5Pget, H5Pset to verify that the library indeed goes into the branch we hope for.
Using H5_HAVE_INSTRUMENT macro to isolate these changes so that it won't affect or be misused by the application.
Platforms tested:
h5committest(shanti still refused to be connected)
Parallel tests on heping somehow are skipped. Manually testing at heping. Have checked
1,2,3,4,5 processes.
Misc. update:
Adding parallel tests for optional collective chunk APIs
Description:
Three new APIs
"H5Pset_dxpl_mpio_chunk_opt_ratio
H5Pset_dxpl_mpio_chunk_opt_num
H5Pset_dxpl_mpio_chunk_opt"
for optional optimization choices from users
have been added to the libraries.
This check-in adds six tests to verify the funcationality and correctedness
of these APIs.
These tests need to be verified with 3 or more processors and with MPI-IO driver only.
Solution:
Using H5Pinsert, H5Pget, H5Pset to verify that the library indeed goes into the branch we hope for.
Using H5_HAVE_INSTRUMENT macro to isolate these changes so that it won't affect or be misused by the application.
Platforms tested:
h5committest(shanti still refused to be connected)
Parallel tests on heping somehow are skipped. Manually testing at heping. Have checked
1,2,3,4,5 processes.
Misc. update:
configure refactoring
Description:
Migrated a couple of Fortran macros to standard autoconf macros
Solution:
HDF5 had a couple of macros, AC_TRY_FLINK and AC_CHECK_FLIB, which were
adaptions of standard autoconf macros before those macros supported
F90.
Now Autoconf seems to have better support for f90, so its standard macros
AC_TRY_LINK and AC_CHECK_LIB can do this job (and move the burden of
supporting those macros to autoconf).
Platforms tested:
heping, modi4, copper
bug fixes
Description:
h5dump/h5ls were not displaying long doubles correctly
Solution:
1) the print datatype functions were incorrectly testing for the valid return value from H5Tequal,
(TRUE), causing the display of an incorrect name of a dataype in error cases from H5Tequal
2) h5tools_print_str did not have a case for native long double
3) added a file generator for a long double dataset
4) added one script test for the long double data (commented , some sytems don't have a native long double match, and the output differs)
5) added a vms file and h5dump script test
Platforms tested:
linux 32, 64
solaris
AIX
Misc. update:
Maintenance on Windows
Description:
update batch file according to Pedro's bug fix (h5dump was not printing attributes for the named datatype cases)
Solution:
Platforms tested:
Windows XP
Misc. update:
Code checkpoint
Description:
Check in fractal heap code to add basic support for skipping direct blocks
when an object is too large to fit in one.
Platforms tested:
FreeBSD 4.11 (sleipnir)
Linux 2.4 (chicago)
Code checkpoint
Description:
Checkpoint fractal heap code at a stable point in development. It appears
that adding objects to heaps and reading them back out is now working
properly (as long as there are no gaps in the heap and no standalone objects
are created).
Platforms tested:
FreeBSD 4.11 (sleipnir)
Linux 2.4 (chicago)
Linux 2.4 (mir) w/64-bit & FORTRAN & C++
Solaris 9 (shanti) w/64-bit
bug fix
Description:
h5repack was not dealing with family files
Solution:
use the toolslib function h5tools_open to open the file instead of H5Fopen in h5repack
Platforms tested:
linux
solaris
AIX
Misc. update:
bug fix
Description:
regenerate a binary file. the previous one was not being open on the scale offeset filter dataset, probably due to a solved bug with the corresponding binary file not being updated
Solution:
Platforms tested:
linux
Misc. update:
Code optimization
Description:
Tweak file free space allocation routines to optimize them a bit.
Platforms tested:
FreeBSD 4.11 (sleipnir)
Too minor to require h5committest
Removed -march flag
Description:
The -march optimization flag is bad in some circumstances
Solution:
Prevented HDF5 from adding the -march archicture-specific optimization
flag, so that users could have one version of HDF5 used by
several different machines.
Platforms tested:
heping (one of the few platforms that affected by this small change)
Description: cross_read.c is a new program to test reading data created
on an OpenVMS, a little-endian, and a big-endian machine. The three
data files are vms_data.h5, be_data.h5, and le_data.h5. To generate
these data files, run gen_cross.c on these machines and change the file
names.
Platforms tested: fuss and h5committest
Misc. update: MANIFEST
Bug fix
Description:
Previous checkin did a bad thing; 'make clean' failed in example directories.
Solution:
Fixed commence.am so that examples no longer break, and fixed a mistake
in conclude.am.
Platforms tested:
heping (minor makefile change)
Misc. update:
Add tests for optional APIs to support collective chunk IO
Description:
In order to test whether library picks up the user's options,
The number of chunks need to be varied for different processes,
Selection of the number of processes selected in one chunk also
need to be varied.
Solution:
Create two cases,
1. Each chunk only selected by one unique process, this case
library should use independent for collective call.
2. One-third of the processes occupies the top half of the whole domain,
The rest of the processes occupies the lower half of the domain.
The total number of chunk is a fixed number 8.
Platforms tested:
Linux 2.4 with mpich 1.2.6(only)
Since I only checked in the code that handles the selection, haven't added any new tests yet. So it won't affect any platforms.
Misc. update:
bug fix
Description:
percent relative error was done using integer arythmetic; use floating point instead
added the case for unsigned long long integer to float conversion
Solution:
Platforms tested:
linux (32,64)
AIX
solaris
Misc. update:
Bug fix
Description:
make check-clean didn't clean results of example tests
Solution:
Fixed Makefiles so that check-clean recurses into example directories.
Also a little Makefile cleanup.
Platforms tested:
mir, modi4, heping, copper
change the array size of collective chunking features of parallel tests.
Description:
Previously array size for collective optimization tests
including
cchunk1,
cchunk2,
cchunk3,
cchunk4,
ccontw,
ccontr,
cschunkw,
cschunkr,
ccchunkw,
ccchunkr
are fixed,
They are only valid for some good number of processors(1,2,3,4,6,8,12,16,24,32,48 etc).
Recently there are more requests for parallel tests to be valid on some odd number of processes such as 5,7,11,13 etc.
Solution:
I change the array size to be dynamic rather than static. Now the fastest change array size is a function of mpi_size. dim2 = constant *mpi_size. After some tunings, theoretically the above tests should be valid for any number of processors. However, other parallel tests still need to be tuned.
To verify the correctness of these tests, using mpirun -np 5 ./testphdf5 -b cchunk1 at heping.
Platforms tested:
h5committest(shanti is refused to be connected)
at heping, 5 and 7 processes are used to verify the correctness.
Misc. update:
portable issue on windows
Description:
setvbuf on visual studio 2005 needs the size to be greater or equal to 2.
<description of setvbuf from msdn library>
Buffer size in bytes. Allowable range: 2 <= size <=
INT_MAX (2147483647). Internally, the value supplied
for size is rounded down to the nearest multiple of 2.
h5import used linebuffer option of setvbuf for stderr and stdout and the size is
0 by default. This causes core dump with visual 2005 on windows 64-bit.
Solution:
1. Use HDsetvbuf inside h5import.c,
2. Define HDsetvbuf inside windows H5pubconf.h for windows to work around
the core dump of h5import test. This is probably a bug inside visual studio 2005.
Platforms tested:
h5commit(shanti is down)
VS 6.0 on windows XP
VS 8.0 on windows XP-64bit
Misc. update:
portable issue on windows
Description:
setvbuf on visual studio 2005 needs the size to be greater or equal to 2.
<description of setvbuf from msdn library>
Buffer size in bytes. Allowable range: 2 <= size <=
INT_MAX (2147483647). Internally, the value supplied
for size is rounded down to the nearest multiple of 2.
h5import used linebuffer option of setvbuf for stderr and stdout and the size is
0 by default. This causes core dump with visual 2005 on windows 64-bit.
Solution:
1. Use HDsetvbuf inside h5import.c,
2. Define HDsetvbuf inside windows H5pubconf.h for windows to work around
the core dump of h5import test. This is probably a bug inside visual studio 2005.
Platforms tested:
h5committest( shanti refuses to be connected)
windows 6.0 at XP
visual 2005 at XP-64bit
Misc. update:
portibility issue to windows
Description:
setvbuf on visual studio 2005 needs the size to be greater or equal to 2.
<description of setvbuf from msdn library>
Buffer size in bytes. Allowable range: 2 <= size <=
INT_MAX (2147483647). Internally, the value supplied
for size is rounded down to the nearest multiple of 2.
h5import used linebuffer option of setvbuf for stderr and stdout and the size is
0 by default. This causes core dump with visual 2005 on windows 64-bit.
Solution:
1. Use HDsetvbuf inside h5import.c,
2. Define HDsetvbuf inside windows H5pubconf.h for windows to work around
the core dump of h5import test. This is probably a bug inside visual studio 2005.
Platforms tested:
h5committest
windows 6.0 at XP
visual 2005 at XP-64bit
Misc. update:
Description: Unix remove function removes only the latest version of a file
on VMS. Some of our tests create multiple versions of the testfiles
and as a result, test programs may be confused, give false negative result,
etc.
Solution: Created HDremove_all function for VMS that removes all versions of the files.
HDremove on VMS is an alias to HDremove_all.
Platforms tested: VMS server and heping (to check that nothing is borken on UNIX side)
Misc. update:
Description: OpenVMS doesn't support denormalized floating-point values.
Disable the denormalized and special values for floating to floating-point
tests.
Platforms tested: OpenVMS - the only concerned machine.
Misc. update:
bug fix
Description:
1) added a more explainative usage message
2) the percent relative error for the integer type (division) was being done using integer arythmetic; use floating point arythmetic instead
3) added a new test for integer percent
Solution:
Platforms tested:
linux (32,64)
AIX
solaris
Misc. update:
Code cleanup
Description:
Update a couple of files that bin/reconfigure rearranged for me... :-)
Platforms tested:
FreeBSD 4.11 (sleipnir)
Too minor to require h5committest
Code cleanup
Description:
Add H5E header to this header, to make certain that the errors in the
"inline" routines are handled correctly.
Platforms tested:
FreeBSD 4.11 (sleipnir)
Linux 2.4 (chicago)
Mac OSX (amazon)
Code update
Description:
Add new minor error, for heap metadata operations.
Platforms tested:
FreeBSD 4.11 (sleipnir)
Linux 2.4 (chicago)
Mac OSX (amazon)
Disable debugging feature
Description:
Disable the "function stack" by default & take out it's dependency on
the "enable-debug" configure flag. The function stack code can slow things
down quite a bit and shouldn't be enabled for the upcoming alpha testing.
Platforms tested:
FreeBSD 4.11 (sleipnir)
Linux 2.4 (chicago)
Mac OSX (amazon)
Bug fix
Description:
The snapshot script was trying to install-doc, even though the docs are
no longer included in the cvs repo.
Solution:
Replaced 'make install install-doc' with 'make install-all', which should
have the same effect (installing the library and examples).
Platforms tested:
very minor change
Code cleanup
Description:
Clean up some compiler warnings (esp. those flagged on Windows builds)
Platforms tested:
FreeBSD 4.11 (sleipnir)
Too minor to require h5committest
bug fix
Description:
the compare check for the datatype sign was not done in the correct place, causing invalid
comparisons to be made
Solution:
put it on the correct place
Platforms tested:
linux 32, 64
AIX
Misc. update:
Code checkpoint
Description:
Update fractal heap code to insert & read heaps up to 64MB in size
(with my current configuration paramaters) and add initial support for
iteratively walking down nested indirect blocks.
Platforms tested:
FreeBSD 4.11 (sleipnir)
Linux 2.4 (chicago)
Solaris 9 (shanti)
Linux 2.4 (mir) w/64-bit
bug fix
Description:
1) the compare flag test was not being put in a correct place, making comparisons attempts that were not supposed to be done
2) some duplicate warnings were being made
Solution:
eliminate the duplicate warnings, put the if compare flag on the correct place
Platforms tested:
linux 32, 64
solaris
Misc. update:
Add more comments
Description:
Add more comments to H5Dmpio.c, which describes a little bit more about collective IO management.
Solution:
Platforms tested:
Only test at heping since only comments were added.
Misc. update:
Code update
Description:
Add basics of routine for reading information back out of a fractal heap.
Platforms tested:
FreeBSD 4.11 (sleipnir)
Too minor to require h5committest
