HDFFV-5919 - GMQS: h5diff - The number of object difference is not consistent between dataset and group/type when attribute(s) have differences
Description:
Object differences are not consistent between dataset and group/datatype when their attribute(s) have differences. This is because attribute(s) differences is not accumulated to group or datatype object’s difference, but accumulated to dataset difference.
To fix, do not accumulate attribute difference to dataset difference. This is referred to h5diff’s default behavior and also past report from users that users were confused by the accumulated behavior. (also can’t figure out only for dataset difference , also hard to spot dataset difference when it has lots of attributes or differences)
This also lead to fix inconsistent format indicating difference between dataset and group/datatype object.
Tested:
jam (linux32-LE), koala (linux64-LE), ostrich (linuxppc64-BE), tejeda (mac32-LE), linew (solaris-BE), Windows (32-LE cmake), cmake (jam)
HDFFV-7942 - h5diff: incorrect result for comparing attribute data with different type size in same class
Description:
This is additional fix due to daily build fail against previous r22684.
Refer to the svn -r22684 log for previous details
Tested:
jam (linux32-LE), koala (linux64-LE), ostrich (linuxppc64-BE), cmake (jam)
Address HDFFV-7942 - h5diff: incorrect result for comparing attribute data with different type size in same class
Description:
When comparing attribute data values with same type class but different size, the result was incorrect. It was due to the size difference and got truncated.
Fixed to match up the smaller type size to big type size like what dataset does.
Tested:
jam (linux32-LE), koala (linux64-LE), ostrich (linuxppc64-BE), tejeda (mac32-LE), linew (solaris-BE), Windows (32-LE cmake), cmake (jam)
Purpose:
HDFFV-8003 - ph5diff (parallel h5diff): daily test failure on ember intermittently during non comparable test file comparison
HDFFV-7755 - parallel h5diff : hanging on koala intermittently during non comparable test file comparison
Description:
non-comparable test intermittently hung on koala and ember, but not on jam. it didn't occur until -np reaches 4 or bigger. it occurred once out of many repeated attempts of the same test.
There was a incorrectly (mistakenly?) duplicated code in MPI section which caused such hang in a certain condition. The test used more processes than other tests, which increased chance to trigger more undone processes, and such process could enter the incorrect code section and wait for wrong pair of send. it explains why it occurred intermittently according to machine condition and using a certain feature.
Removed incorrect code which blocked correct code.
Tested: some manually repeated test performed
jam (linux32-LE), koala (linux64-LE), ostrich (linuxppc64-BE)
Fix for HDFFV-8107 testh5diff will fail if build/test in HDF5 source tree
Description:
This is sub-task for "HDFFV-8105 testh5diff.sh uses the wrong operator (-a) in an if statement."
From the HDFFV-8105's update, h5diff test failed if build&test is performed in HDF5 source tree because 'cp' try to copy test files to self dir.
It's addressed by skipping if cp's src dir and dest dir is same.
Also this applied for all other tools under src/tools dir.
No change to the CMakeLists.txt files because CMake cautions/demands that in-source builds be avoided.
Tested:
jam (linux32-LE), koala (linux64-LE), ostrich (linuxppc64-BE), tejeda (mac32-LE), linew (solaris-BE), some manual tests as well
Description:
The test command in COPY_TESTFILES_TO_TESTDIR used the wrong operator -a which
means and. It is not sufficient to test mere existence as $tstfile could have
been a directory or other non-file things. If the file is not readable, the
$CP command would fail but its return code was not checked.
Solution:
Get rid of the test command and go ahead to $CP it and then check its returned
code and report errors if detected.
Tested; committested. (note that ostrick still failed on errors committed by
other staff).
HDFFV-7602 - GMQS: HDF5 command tools: Provide framework for reusable test files among tools
Stage2 working for h5diff tool.
Description:
Update cmake script to fetch test files from any tools to a test directory
and run test in organized manner.
Also keep the apperance similiar to shell script for better maintainance.
Tested:
CMake (Windows and jam)
The ph5diff (parallel h5diff) can intermittently hang in parallel mode
when comparing the two HDF5 files that contains common name objects with
different object types. (HDFFV-8003)
Skip only the specific test for parallel until the fix has made.
Tested:
jam (linux32-LE), koala (linux64-LE), ostrich (linuxppc64-BE), tejeda (mac32-LE), linew (solaris-BE)
Fix for HDFFV-7998-h5diff: incorrect behavior and output for comparing symbolic dangling links without following-symlinks option
Description:
Fix not to check and display dangling link status without --follow-symlinks option.
Berfor the fix, when comparing symbolic link to symbolic link without follow-symlinks option, h5diff followed to check if those links are dangling or not. It caused to display output incorrectly (dangling link instead of symbolic link).
The fix also improved performance when comparing lots of symbolic links without the --follow-symlinks option.
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)
Fix for HDFFV-7644 GMQS: h5diff - incorrect behaviors when comparing HDF5 files with different type of objects (dataset, group, type) as common objects
Description:
Fixed failure for comparing same named object with different object types in comparing groups. (common object comparison)
Prior to the fix, h5diff resulted in error. After the fix, h5diff detects such case as non-comparable and display messages accordingly.
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)
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)
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)
Fix for HDFFV-7835 h5diff: incorrect result for comparing the two same type symlinks as dangling links.
Description:
When two symbolic dangling links are compared with --follow-symlinks option,
the result should be same. It works for comparing two files, but didn't work
for comparing two objects.
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)
Fix for HDFFV-7836 h5diff: displays error stack message for comparing the two dangling symlink with follow-symlinks option
Description:
While ago, to improve performance, skipping same object checking
(h5tools_is_obj_same()) was added.
However the checking function doesn't understand about the dangling link and
caused the issue.
Since handling dangling link code section already implemented, move the
checking function after handling dangling-links to address the problem.
Test was 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)
HDFFV-7770 - h5diff segmentation fault on JPSS files
Description:
Fixed segfault over non-comparable attribute with different dimention or
rank, along with '-c' option to display details.
Tested:
jam (linux32-LE), koala (linux64-LE), heiwa (linuxppc64-BE), tejeda (mac32-LE), linew (solaris-BE), Windows, Cmake
Added a funtion to reset dataset & hyperslab buffer size for h5repack from an
environment variable.
This is performance debugging purpose for now.
Tested:
jam (linux32-LE), koala (linux64-LE), heiwa (linuxppc64-BE), tejeda (mac32-LE), Windows (32-LE), cmake
Skip parallel diff test "-c non_comparables1.h5 non_comparables2.h5" due
to hanging issue occurring on koala (linux64-LE) intermittently.
Plan to put back when resolved.
The test was related to J7693.
Tested:
jam (linux32-LE), koala (linux64-LE)
HDFFV-7693 - h5diff produces different behavior between versions: 1.8.3 and 1.8.6
Description:
There were two issues on this.
One was not displaying all the comparable objects if non-comparable
object/attribute exist and compared before comparables ones. This issue
occurred after 1.8.4 release. This is the issue from user.
The other issue was not displaying all the comparable attributes,
if non-comparable object/attribute exist and compared ahead. This issue
were exist even before 1.8.4 release. So it's possible some attribute
comparison may have not displayed correctly in the past, if non-comparable
data were exist in common object.
Fixed h5diff to display all the comparable object and attribute regardless
of non-comparables.
Tested:
jam (linux32-LE), koala (linux64-LE), heiwa (linuxppc64-BE), tejeda (mac32-LE), linew (solaris-BE), Cmake (jam, Windows)
Ran bin/reconfigure to update the Makefile.in in directories not part of the fortran directory check=in. Updates Makefile.in due to changes made in configure.in for the Fortran 2003 additions.
Tested on all platforms run under daily tests.
Add test cases for HDFFV-7656 - "--delta=something" considers two NaN of the same type are different.
The fix was added (r21105) before but test cases were incorrectly added and missing for cmake script.
Tested:
jam (linux32-LE), koala (linux64-LE), heiwa (linuxppc64-BE), tejeda (mac32-LE), linew (solaris-BE), cmake
Revert the previous makefile change as release snapshot test failed with
MANIFEST check. Don't clean 'testfiles' dir via 'make distclean' for
the case of build&test in source dir.
In regular daily testings, we build&test in seperate dir from source dir,
so no issues occurred.
However for the release snapshot test, it seems that we build&test in
source dir.
The previous change can remove testfiles in source dir if build&test is
done in source dir becasue currently the dir name 'testfiles' is same in
source dir and test dir.
Other task (7602) may change to use different name for the test directory,
so 'make distclean' can do its job without interrupting MANIFEST.
- h5repack: h5repack failed to copy dataset if the layout is changed from chunked with
unlimited dims to contiguous. (PC -- 2011/07/15)
- h5diff: "--delta" option considers two NaN of the same type are different, which is wrong
based on http://www.hdfgroup.org/HDF5/doc/RM/Tools.html#Tools-Diff. (PC -- 2011/07/15)
Purpose:
Remove H5_MPI_SPECIAL_COLLECTIVE_IO_WORKS and
H5_MPI_COMPLEX_DERIVED_DATATYPE_WORKS #defines from source.
Description:
Two advanced parallel functionalities, special collective IO and
complex derived datatypes, are not supported by older
implementations of mpi, and thus our code limits the use of these
features with #ifdefs and has checks in configure to set them (or
not). Unfortunately, configure can't actually run a parallel check
to see if these features are working (nor not) so it resorts to
looking in the config files where they are explicity enabled or
disabled based on versions of mpi, sytems being built on, or for
no documented reason at all (i.e. just set to on or off as some
'default'). Overriding these settings is easy if need be, provided
it is known that it needs to be done to get improved performance,
and oftentimes it is not.
Most new MPI implementations successfully handle the functionality
requested when these #defines are set, and many of the "turn these
features off" cases in the config files are for old (> 5 years)
versions of MPI and retired systems (such as NCSA's tungsten).
Therefore, the decision has been made to remove the support for
these old versions of MPI and systems that cannot handle these
behaviors. The #ifdefs and supporting setup in the config/ files
and configure script has been removed, and the code executed when
these options were not set removed from the source.
In passing, this commit also cleans up some whitespace issues in
both t_mpi.c and H5Dmpio.c. Furthermore, in t_mpi.c, the special
collective IO test was not getting regularly run due to it being
written to work only with four processes (we regularly test with
six, previously with three), and thus it failed when actually run
due to an out of bounds data buffer assignment. It has been
modified to run at any number of processes greater than four, and
the memory problem has been fixed so the test passes.
Tested:
jam, h5committest, ember
Work for HDFFV-7600 - GMQS: h5diff - argument options -d, -p and --use-system-epsilon should be mutually exclusive.
Description:
Fixed h5diff to display instructive error message and exit with 1
when mutually exclusive options (-d, -p and --use-system-epsilon)
are used together.
Tested:
jam (linux32-LE), koala (linux64-LE), heiwa (linuxppc64-BE), tejeda (mac32-LE), linew (solaris-BE), Cmake (jam)
Purpose:
Work for HDFFV-7602 - HDF5 command tools: Provide framework for reusable
test files among tools
Description:
Provide framework to share test files among tools for tools test.
Tested:
jam (linux32-LE), koala (linux64-LE), heiwa (linuxppc64-BE)
Test scripts sometimes need to filter some system-specific messages from the
actual output so that it can match the correct expected output. These filtering
functions, ususally called "STDOUT_FILTER()" and "STDERR_FILTER()" were being
repeated in individual test scripts. This becomes a maintenance problem and
is error prone.
Solution:
Extract the two filter functions code to bin/output_filter.sh and then each
test script sources it in. This allows reuse of coding and is much easier to
maintain and to add new filtering.
Tested:
LLNL Zeus (linux64 cluster) and Dawndev (Blue-Gene cluster), both for serial
mode only.
--This line, and those below, will be ignored--
M tools/misc/testh5mkgrp.sh
M tools/h5dump/testh5dump.sh.in
M tools/h5diff/testh5diff.sh
M tools/h5copy/testh5copy.sh
M tools/h5ls/testh5ls.sh.in
M MANIFEST
A bin/output_filter.sh
HDFFV-5928 - GMQS: h5diff problem and improvement on comparsing the same objects
Description:
Improved performance by eliminating duplicated action for getting object
information in half from the previous fixe when comparing group vs group.
This is addition to the previous commit r20676.
Tested:
jam (linux32-LE), koala (linux64-LE), heiwa (linuxppc64-BE), tejeda (mac32-LE), linew (solaris-BE), cmake
H5Dread() produces different output for invalid enum type values
on systems of different endiannes. On BE systems, it always
emits -1. On LE systems, the invalid value is emitted. This
difference caused the h5diff test to fail since it tests to
see if different invalid enum values are reported as different
(on BE systems they are not since -1 == -1).
This fix removes the "invalid vs. different invalid" test
while we fix the library bug. HDFFV-7527 will be reopened
in JIRA until the full test passes.
Tested on: jam, heiwa, linew
not represent a valid enum value.
Description:
The h5diff code compares enum values by converting them to strings and
then comparing them. When the enum value is out of range and can't be
converted to an enum string representation, the comparison was skipped.
The code now flags differences when one of the two enum values is out
of range (two out of range values are compared in memory). A test has
been added to the tools test script.
This fixes JIRA HDFFV-7527
Tested on: jam, koala, heiwa (h5committest)
- HDFFV-5928 - GMQS: h5diff problem and improvement on comparsing the same objects
Description:
Fixed:
1) adding h5tools_is_obj_same() function to check if two given IDs or paths point to the same object. This function can be very useful for other tools and applications.
2) using h5tools_is_obj_same() at h5diff() and diff() in h5diff.c. If two paths point to the same object, there is no need to check the details of the object since we know there is no difference. The fix will increase the performance by skipping the content comparison. It also fixed the problem of reporting difference for some cases of comparing the same file, e.g. empty files or files with incomparable objects the same file.
Test update:
Updat prvious test cases (171, 172, 530) affected by this fix, so they
still perfrom originally intended testing without bypassing.
Tested:
jam (linux32-LE), koala (linux64-LE), heiwa (linuxppc64-BE), tejeda (mac32-LE), linew (solaris-BE), cmake
General shared library improvements for CYGWIN / AIX
Description:
Shared libraries are disabled on both CYGWIN and AIX due
to inability to build them correctly. Part of the problem
in both of these situations is the lack of the libtool
flag -no-undefined, which tells libtool that all needed
symbols are defined at link time (a requirement on these
systems) and that it's okay to build shared libraries.
Another problem are lack of dependencies between wrapper
libraries and core C HDF5 library.
This patch addresses both of these by fixing configure to
add in -no-undefined flag for libtool during linking and
adds automake dependencies in the Makefile.am files.
After testing, both CYGWIN and AIX now generate shared
libraries, but there are still some test failures in each.
(cache_api, dt_arith, and testerror.sh on CYGWIN, and
fortran tests on AIX).
Even though the shared libraries are not quite perfect,
this is a general improvement to what we had before, so
I'm applying the patch anyways. Note that default behavior
of shared libraries on these systems being disabled has
NOT been changed and requires the use of the
--enable-unsupported to attempt to build them.
We will need to address the test failures in each
architecture prior to formally supporting shared
libraries on each.
Tested:
h5committested & CYGWIN tested (on bangan)
(AIX tested by Albert on bp-login2)
Add "silent make" mode configure option.
Description:
Automake 1.11 has a new option available that allows for a
silent make mode. This functionality needs to be explicitly
enabled in configure.in via the use of the automake macro
AM_SILENT_RULES, which is what this commit is adding.
This introduces a new configure option:
--{en|dis}able-silent-rules
This option is on by default, and simplies compile and link
line outputs when building the library. Disabling this option
will print full "verbose" output (i.e., full compile and
linking lines for each target).
Tested:
This was tested on jam & h5committested
- Revise shared Fortran library disabling scenarios in configure
- Improve configure output summary
Description:
Shared Fortran libraries are not supported on Mac, but were being
disabled by configure in a way that also forced the C libraries
to be static-only. This has been fixed, so now only shared Fortran
is disabled while shared C can remain.
This prompted two additional changes:
1. While working on the check that addresses whether or not
shared Fortran libraries are allowed, removed old and no
longer needed check(s) that disable shared Fortran
libraries with HP, Intel 8, PGI, and Absoft compilers.
(Essentially, Mac is the only situation in which Fortran
shared are disabled by configure.)
2. Having two different states of libraries (i.e. shared C
library with static-only Fortran library) was not apparent
in the configure summary, which labeled all libraries as
either shared and/or static. I've added lines to both the
C++ and Fortran output sections to list shared/static-ness
of these libraries specifically.
Additionally, I've made sure that the new --enable-unsupported
configure option correctly overrides configure if it tries to
disable a shared library.
Tested:
jam, fred, & h5committest
Fixed Bug 2184 - GMQS: h5diff - incorrect calculation code for
--use-system-epsilon option
Description:
Merged from HDF5 1.8 branch r20369.
Fixed h5diff for --use-system-epsilon option: the calculation changed
from ( |a - b| / b ) to ( |a - b| ). This was decided for better
performance and was corrected only in HDF5 trunk, so 1.8 got updated.
Also comments for equal_XXX() function were updated correctly.
Also help page and RM got updated correctly.
Also add test cases for testing the differences w/wo the option.
Tested:
jam (linux32-LE), amani (linux64-LE), heiwa (linuxppc64-BE), tejeda (mac32-LE), linew (solaris-BE), Windows (32-LE)
Fixed CHICAGO: Bug 2121 - h5diff - incorrect and lack of output for the
different set of attributes (different number and names)
Description:
This is related to previous checkin r20294.
Add help page update for verbose option level feature, which displays
useful information about attribute difference with -v1, --verbose=1 -v2
or --verbose=2 option.
Tested:
jam (linux32-LE), amani (linux64-LE), heiwa (linuxppc64-BE), tejeda (mac32-LE), linew (solaris-BE)
Fixed CHICAGO: Bug 2121 - h5diff - incorrect and lack of output for the
different set of attributes (different number and names)
Description:
Previously h5diff compared attributes correctly only when two objects have
the same number of attributes and attribute names are identical.
This fix covers all other cases.
Also didn't display useful information about attribute difference.
This fixes both issues.
Tested:
jam (linux32-LE), amani (linux64-LE), heiwa (linuxppc64-BE), tejeda (mac32-LE), linew (solaris-BE)
Clean up MPI resource leaks in parallel tests, along with a bunch of
compiler warnings.
Tested on:
FreeBSD/32 6.3 (duty) in debug mode
FreeBSD/64 6.3 (liberty) w/C++ & FORTRAN, in debug mode
Linux/32 2.6 (jam) w/PGI compilers, w/default API=1.8.x,
w/C++ & FORTRAN, w/threadsafe, in debug mode
Linux/64-amd64 2.6 (amani) w/Intel compilers, w/default API=1.6.x,
w/C++ & FORTRAN, in production mode
Solaris/32 2.10 (linew) w/deprecated symbols disabled, w/C++ & FORTRAN,
w/szip filter, w/threadsafe, in production mode
Linux/PPC 2.6 (heiwa) w/C++ & FORTRAN, w/threadsafe, in debug mode
Linux/64-amd64 2.6 (abe) w/parallel, w/FORTRAN, in debug mode
Mac OS X/32 10.6.6 (amazon) in debug mode
Mac OS X/32 10.6.6 (amazon) w/C++ & FORTRAN, w/threadsafe,
in production mode
Bug 2089 - GMQS: h5diff segfault on a compound dataset with fixed length +
vlen string type order
Description:
This is fix for the segfault when h5diff compares a compound dataset with
combination of fixed length string types and vlen string types in certain
orders. Optimized vlariable length string handling codes. The fix is
referred from h5dump handling vlen strings.
For testing, several compound datasets were added with various combinations.
Previous failed cases:
- Vlen string, Fixed length string, Vlen string, Fixed length string
- Fixed length string, Fixed length string, Vlen string, Vlen string
- Fixed length string, Vlen string, Fixed length string, Vlen string
Tested:
jam (linux32-LE), amani (linux64-LE), heiwa (linuxppc64-BE), tejeda (mac32-LE), linew (solaris-BE)
Fixed h5diff to handle variable-length strings in a compound dataset
correctly. Also variable-length string array in a compound dataset.
Bug #1989.
Description:
Garbage values were displayed when h5diff compared variable-length
strings (or string array) in a compound type dataset.
Tested:
jam (linux32-LE), amani (linux64-LE), heiwa (linuxppc64-BE), tejeda (Mac32)
Update help page for exclude-path feature.
Description:
Related to "1890: h5diff excluding object for file comparison via command line" checkins. (r19406)
Tested:
jam, amani, heiwa
Fix for group comparison with exclude-path case. Use relative path.
Description:
Related to "1890: h5diff excluding object for file comparison via command line" checkins. (r19406)
Tested:
jam, amani and heiwa
Add missing files from the previous checkin.
(Add extra test cases for Bug1975 h5diff - support recursive comparison on group when specified as an object)
Add extra test cases for bug1975 h5diff - support recursive comparison on group when specified as an object.
Description:
Additional tests for combination of group recursive and --follow-symlinks
with multi-linked external links with several files (same name/strucure).
Tested:
jam, amani, heiwa
Add --exclude-path option
Description:
Specified path to an object will be excluded from comparing the two files or two groups. If group is specified all the member objects will be excluded.
Related to "1890: h5diff excluding object for file comparison via command line"
Tested:
jam, amani and heiwa
Fix for Bug1975 h5diff - support recursive comparison on group when specified as an object
Description:
Compare member objects and groups recursively when two files or groups are specified to be compared. Support parallel diff and handling symbolic links accordingly.
Tested:
jam, amani, heiwa
Corrected use/name of source folder aliases.
Duplicated FindMPI.cmake so that non-c++ compiler is found first (recommemded commands did not work).
Tested: local linux with mpich
Problem: some launching command like yod (and sometimes mpirun), do not pass
back the exit code of the executables back to the shell. It always returns 0
if it can launch the executable. This cause tests that expect non-zero exit
code to fail incorrectly.
Short term Solution:
Added the feature that if $RUNSERIAL_NOEXITCODE is defined, do not
test conditions that rely on non-zero exit code.
Tests:
dove-linux using RUNSERIAL=thgyod to see the failures and using
$RUNSERIAL_NOEXITCODE=yes to bypass the errors.
Also h5committested.
Rename '--follow-links' to '--follow-symlinks'
Description:
The '--follow-links' option is to follow symbolic links (soft and external).
Make the name more intuitive and specific to the feature.
Tested:
jam, amani and linew
Enable tools lib to be built as a dll on windows. Added two get/set functions for progname and d_status.
Also add windows import/export declarations to functions.
Updated error_mesg() and warn_mesg() to remove progname argument and use get functions
Tested:
Windows, linux
Improve configure's large-file support control.
Description:
Modified configure to now attempt to add defines necessary for
supporting largefiles on all systems, instead of solely on linux. This
is in response to user requests to enable largefile support on Solaris
by default, as well as to give extra control on AIX (instead of just
jamming the necessary flag into the config files).
The old --enable-linux-lfs flag has been removed in favor of the
--enable-largefile flag (enabled by default), which can be used on all
platforms.
On systems where large files cannot be supported in this manner,
configure will report as such.
Tested:
h5committest
AIX (NCSA's blue_print machine)
duty, liberty, and linew.
Autotool Upgrade / Libtool Bug Fix
Description:
Updated autotools referenced in bin/reconfigure as follows:
Autoconf 2.64 --> Autoconf 2.65
Automake 1.11 --> Automake 1.11.1
Libtool 2.2.6a --> Libtool 2.2.6b-mcg
The referenced libtool version is a custom version of 2.2.6b. It
has been tweaked to fix a bug in libtool that occurs
when using PGI 10.0 compilers. A check incorrectly categorizes
the C++ compiler as version 1.0 instead of 10.0, and the link
line is subsequently set up incorrectly and fails to compile.
A patch has been made available and will be included in the next
release of libtool, but in the meantime I've applied the patch to a custom
installation as indicated above. This bin/reconfigure now references
the custom installation, and the resulting configure script will correctly
categorize the PGI 10.0 C++ compiler.
Ran bin/reconfigure to regenerate configure and makefiles.
Tested:
jam and amani with PGI 10.0 compilers.
h5committest
bugzilla 1754: h5diff: support comparing through links.
Description:
Fix incorrect (or hanging) behavior in parallel mode when specifying
invalid combination of command options.
(ex: -v and -q , --no-dangling-links without --follow-links)
Add relate test case
Update h5diffgentest.c due to add test files this and previous time
Note:
svn #18266 (prior to this)
svn #18164 (original check-in)
Tested:
h5committest (jam, amani and linew)
bugzilla 1754: h5diff: support comparing through links.
(original check-in svn revision #18164)
Description:
fix the hang issue in parallel mode when compare external-link.
add --no-dangling-links option.
add test cases (#450-#459) relate to the new option.
improve test script to check exit code.
update --help relate to the new options.
correct some indentations.
Tested:
h5committest (jam, amani and linew)
They are external-link test cases with following link option.
The related code was checked in as svn #18164 and #18165 to support follwing
links for h5diff.
The test cases will be added back when find solution for parallel mode.
Tested on Jam
Take care of test case fail for h5repack from checkin (r18164)
Description:
Previous checkin didn't allow h5diff to return succeed when a link's
target object doesn't exist. Changed it to succeed with warning.
(Need to discuss related feature)
Tested on Jam
Add test cases for h5diff check-in (svn revision #18164).
Description:
This test cases are for testing comparing through various link combinations
relate to bugzilla report 1754.
Also improved the test script to handle external link cases and
simplified to use single line instead of double for each test cases.
Tested on Jam
Add a feature to compare through links. Relate to bugzilla report 1754.
Description:
Currently, h5diff command only compares the name(path) of target object
not the actual data. With this feature, h5diff will go through the link(s)
and figure out the actual object at the end of the link and compare data
if exist. Internally we have hard-link, soft-link, external-link (as part
of user_defined-link). This feature will provide a user transparent
experience when comparing link objects as the user can specify any of
those 3 type of links to be compared in any combination.
Tested on Jam.
Additional fix relted to the fix of bug1672.
Description:
The fix of bug1672 caused some changes of output which required modifying
some test cases. After some discussion, restoring the changes of the test
cases was decided. After many experiments for the solution, this fix was
made so the code which improved performance can stay.
Tested on Jam.
The mixed use of RUNTEST (original) and RUNTESTS (new) caused
confusion. E.g., the timings in test/ was still using the old
$RUNTEST. It made more sense to use $RUNTEST which is used
by the dejagnu feature of automake. So, I changed all
$RUNTEST or $RUNTESTS to $RUNEXEC.
config/commence.am & config/conclude.am are the two files
that got changes. Also fixed an error in test/Makefile.am.
The rest are changed by bin/reconfigure.
Tested: h5committested.
Fix Bug1672 - Display garbage value on LE machine for BE data.
Description:
Casuing by calling diff_dataset twice when -r or no option was given.
Change to call it once which fix the problem.
It also improves the performance. (diffing time in half)
According further test, it also occurred on BE machine as well and it
seems occruing only with the BE data attached to Bugzilla.
Don't know how the file was created. Anyway this fix will prevent
from diffing with garbage values in similiar potential case.
Tested on:
linux32 (jam) , linux64 (almani), solaris (linew)
CFLAGS overhaul
Description:
Modified the way configure handles CFLAGS.
(note: all changes also apply to FCFLAGS, CXXFLAGS, CPPFLAGS, and LDFLAGS).
1. The configure process will now always preserve a user's CFLAGS
environment variable setup. Any additional flags necessary for compilation
added at configure time will be passed into the Makefiles as AM_CFLAGS,
which is an automake construct to be used in addition to CFLAGS.
This will allow a user to have the final say, as CFLAGS will always appear
later in the compile line than AM_CFLAGS. Additionally, setting CFLAGS
during make will no longer completely erase all flags set by configure,
since they're maintained in AM_CFLAGS.
2. Additionally, where possible, flags previously being assigned directly
into CFLAGS (and thus propagating into h5cc) have now been redirected into
H5_CFLAGS, so they're used ONLY for compiling hdf5, and not embedded into
the h5cc wrapper script as well.
*Note that H5_CFLAGS ultimately is assigned into AM_CFLAGS for use in the
Makefiles. Complete description of changes and build process will be
included in a Configure Document that Elena and I are working on.
3. Removed unsupported config files. This includes:
config/dec-osf*
config/hpux11.00
config/irix5.x
config/powerpc-ibm-aix4.x
config/rs6000-ibm-aix5.x
config/unicos*
4. Modified configure summary to display additional values. Specifically,
appropriate AM_* variables are being shown, as well as H5_FCFLAGS and
H5_CXXFLAGS, which were for some reason not already present.
Tested:
- H5committest
- Tested on all THG / NCSA machines, using several combinations of the more
prominent configure options (c++, fortran, szip, threadsafe, parallel,
et cetera). (Thanks to Quincey for rysnc testing setup!)
- With regards to new automated testing, anything *necessary* for
compilation will be caught by the daily tests as it stands now. (i.e.,
if LDFLAGS is not properly set when szip is used, linking will fail).
Additionally, with regards to which flags get into h5cc, if any
*necessary* flags have been improperly removed, then daily tests should
fail during make installcheck. Additional machine-specific desired
behaviors and/or checks may have to be set up separately within the
daily tests, so this is something to work on.
The ph5diff tests printout incorrectly reporting they are running
h5diff when it actually is running ph5diff.
Fixed.
Tested: Jam (parallel), linew (serial)
No H5committest since it is a shell script and Jam tests both
serial and parallel h5diff for Linux while Linew tests Big
endian platform.
" Use "--use-system-epsilon" for system EPSILON
" Use "-p" or "-d" for whatever user's choice of epsilon
" Use "-p 0" or "-d 0" for strict equality (same as default)
Updating autotools
Description:
Installed new autotools and used them to reconfigure HDF5.
- Automake upgraded to 1.11
- Autoconf upgraded to 2.64
- bin/reconfigure script edited to use new versions (on jam), and
run to generate new configure script and Makefile.in's.
- configure.in script edited to add "_cv_" to all AC_CACHE_VAL strings
(in order to comply with new autoconf standard).
Tested:
Tested on machines jam, smirom, liberty, linew.
Tested w/ features c++, fortran, parallel.
Tested w/ compilers gcc, pgcc, icc.
Further testing via Daily Tests should catch any other outliers. Upon
passing DT's, I'll propogate the new tools into 1.8, hdf4, et cetera.
Fix BZ #1583
Description:
The --enable-static-exec flag was broken. Configure was adding the
-all-static linker flag to an environment variable which was subsequently
never used. (looks like it was used in 1.6 Makefile.ins, but never
moved to 1.8's Makefile.ams when we added automake support).
To fix, I've added this environment varibale to the link lines of all the
tools. Now, when --enable-static-exec is invoked, it will correctly
generate statically linked executables in the installed bin directory.
Tested:
I've tested manually to ensure that the generated execs are static when
the flag is used and dynamic when not. I've also tested when szip is
used, as this was a use case mentioned in the bug report.
Also: ran full tests with make check on jam and smirom for good measure.
this error
home/hdftest/snapshots-hdf5/current/tools/lib/h5tools_ref.c:126: undefined reference to `d_status'
happens because the original developer of h5tools_ref.c declared these 2 variables in that source file
extern char *progname;
extern int d_status;
this means that all programs that use this source file (typically all tools) will have to define these 2 variables, even if they are *not* used.
examples
const char *progname = "h5jam";
int d_status = EXIT_SUCCESS;
"progname" is usually used to print the program name in the usage, and "d_status" is just a return value of the tool, usually.
I was wondering about removing these 2 declarations from h5tools_ref.c
extern char *progname;
extern int d_status;
and define those variables on each tool if they are only needed (it seems to me that only h5ls and h5dump actually "need" them)
tested: smirom parallel
The file used for input is located in the common source tools for testfiles, in tools/testfiles
Modified the h5repack shell script to read files from this location (h5repack reads its input files from a dedicated testfiles location in h5repack/testfiles)
Changed the h5diff open file call to use h5tools_fopen, so that it can open all file drivers
Tested: windows, linux, solaris
- Updated bin/reconfigure to use latest version of automake (1.10.2).
Re-generated Makefile.in's by running bin/reconfigure.
- Added libtool version numbers to c++, fortran, hl, hl c++, and hl fortran
libraries.
Tested:
jam, liberty, smirom
ISSUE : the tools use the following formula to read by hyperslabs: hyperslab_size[i] = MIN( dim_size[i], H5TOOLS_BUFSIZE / datum_size) where H5TOOLS_BUFSIZE is a constant defined of 1024K. This is OK as long as the datum_size does not exceed 1024K, otherwise we have a hyperslab size of 0 (since 1024K/(greater than 1024K) = 0). This affects h5dump. h5repack, h5diff
SOLUTION: add a check for a 0 size and define as 1 if so.
TEST FOR H5DUMP: Defined a case in the h5dump test generator program of such a type (an array type of doubles with a large array dimension, that was the case the user reported). Since the written file commited in svn would be around 1024K, opted for not writing the data (the part of the code where the hyperslab is defined is executed, since h5dump always reads the files). Defined a macro WRITE_ARRAY to enable such writing if needed. Added a run on the h5dump shell script. Added 2 new files to svn: tools/testfiles/tarray8.ddl, tools/testfiles/tarray8.h5. NOTE: while doing this I thought of adding this dataset case to an existing file, but that would add the large array output to those files (the ddls). The issue is that the file list is increasing.
TEST FOR H5DIFF: for h5diff the check for reading by hyperslabs is H5TOOLS_MALLOCSIZE (128 * H5TOOLS_BUFSIZE) or 128 Mb. This makes it not possible to add such a file to svn, so used the same method as h5dump (only write the dataset if WRITE_ARRAY is defined). As opposed to h5dump, the hyperslab code is NOT executed when the dataset is empty (dataset is not read). Added the new dataset to existing files and shell run (tools/h5diff/testfiles/h5diff_dset1.h5 and tools/h5diff/testfiles/h5diff_dset2.h5 and output in tools/h5diff/testfiles/h5diff_80.txt).
TEST FOR H5REPACK: similar issue as h5diff with the difference that the hyperslab code is run. Added a run to the shell script (with a filter, otherwise the code uses H5Ocopy).
tested: linux (h5commitest failed , apparently it did not detect the code changes in /tools/lib that fix the bug: the error in an assertion in the hyperslab of 0. I am sure that making h5ccomitest --distclean will detect the new code , but don't want to wait more 3 hours :-) )
Bug Fix
Description:
Fixing BZ #1381. The --includedir=DIR configure option, which is used
to specify the installation location of C header files, did not work
correctly as the path was hard-coded in config/commence.am. I'm presuming
this is because an older version of automake didn't know where to put
c header files. In any case, removing this line now defaults the includedir
to the same directory that it is currently hard-coded to, and also fixes
the configure flag to allow customization of this value.
Tested:
jam, liberty
Remove another call to H5E_clear_stack() from within the library.
Clean up lots of compiler warnings.
Tested on:
Mac OS X/32 10.5.6 (amazon)
(followup on other platforms forthcoming)
the way h5ls prints types, it starts searching for NATIVE types first. One solution would be h5ls not to detect these native types, using for example the same print datatype function that h5dump does, that would make the output look the same on all platforms ("32-bit little-endian integer" would be printed instead). Drawback, this "native" information would not be available. Other solution is to have not one but 2 expected outputs and make the shell script detect the endianess and compare with one output or other
tested: h5committest
Solution: for compound types, recursively apply that check
Two new cases are added
1) the compound type has a different number of members. Message printed is
<obj1> has X members <obj2> has Y members
Where X and Y are the number of members of each compound type being compared
2) the compound type has not comparable types (for example a double and an int at the same index)
In this case the message
Comparison not possible: object1 is of class1 and object2 is of class2
Is replaced with
Comparison not possible: object1 has a class1 and object2 has a class2
Modified the test generator program to have these 2 cases
Added a shell run for these 2 cases
Tested: windows, h5committest
-N, --nan Avoid NaNs detection
Note: there is no shell script run for datasets with NaN because the output is non portable (different results and NaN strings for different systems)
Tested: windows, linux
