Description:
The H5TB API makes some improper assumptions about the order of compound
datatype members. Namely, it assumes that members remain in the order in which
they were inserted. Unfortunately, this assumption is inherent in the design of
the interface. The library has been patched so that this assumption holds in
situations relevant to H5TB.
Tested: jam, linew, amani (h5committest)
configure.in. The Pathscale compiler on Sandia's Linux machine misinterprets the values of
unsigned long as negative during the conversion to float, when the first bit of unsigned long
is on.
Tested on jam. Need to test it on Sandia's machine.
Description: The optimized "subset" compound conversion function would
improperly copy the "blank" space at the end of compound types. Modified
H5T_conv_struct_init to detect when the subset type has extra space at the end,
and calculate the size of the data that should be copied into the conversion
buffer for each element. Changes to the functions that implement these
conversions.
Tested: kagiso, linew, smirom (h5committest)
Bring r15609 back from trunk:
Omnibus compiler warning cleanup & some formatting happiness.
Tested on:
Mac OS X/32 10.5.4 (amazon)
Too minor to require h5committest
Bring back revision 15227 from trunk:
Small cleanups to copying datatypes
Tested on:
FreeBSD/32 6.2 (duty) in debug mode
FreeBSD/64 6.2 (liberty) w/C++ & FORTRAN, in debug mode
Linux/32 2.6 (kagiso) w/PGI compilers, w/C++ & FORTRAN, w/threadsafe,
in debug mode
Linux/64-amd64 2.6 (smirom) w/default API=1.6.x, w/C++ & FORTRAN,
in production mode
Linux/64-ia64 2.6 (cobalt) w/Intel compilers, w/C++ & FORTRAN,
in production mode
Solaris/32 2.10 (linew) w/deprecated symbols disabled, w/C++ & FORTRAN,
w/szip filter, in production mode
Mac OS X/32 10.5.3 (amazon) in debug mode
Linux/64-ia64 2.4 (tg-login3) w/parallel, w/FORTRAN, in production mode
Make H5Pget_filter_by_id() API versioned and switch internal usage
to H5Pget_filter_by_id2().
Add simple regression test for H5Pget_filter_by_id1().
Tested on:
FreeBSD/32 6.2 (duty) in debug mode
FreeBSD/64 6.2 (liberty) w/C++ & FORTRAN, in debug mode
Linux/32 2.6 (kagiso) w/PGI compilers, w/C++ & FORTRAN, w/threadsafe,
in debug mode
Linux/64-amd64 2.6 (smirom) w/default API=1.6.x, w/C++ & FORTRAN,
in production mode
Linux/64-ia64 2.6 (cobalt) w/Intel compilers, w/C++ & FORTRAN,
in production mode
Solaris/32 2.10 (linew) w/deprecated symbols disabled, w/C++ & FORTRAN,
w/szip filter, in production mode
Mac OS X/32 10.4.10 (amazon) in debug mode
Fix problem with datatype messages where the version of the format
for a datatype message could depend on the "use the latest format" flag from
the file after it was initially created.
Tested on:
FreeBSD/32 6.2 (duty)
FreeBSD/64 6.2 (liberty)
Linux/32 2.6 (kagiso)
Mac OS X/32 10.4.10 (amazon)
company. The I/O is optimized when the source and destination
members are a subset of each other one way or another, and
the order is the same, and no conversion is needed. For example:
struct source { struct destination {
TYPE1 A; --> TYPE1 A;
TYPE2 B; --> TYPE2 B;
TYPE3 C; --> TYPE3 C;
}; TYPE4 D;
TYPE5 E;
};
or
struct destination { struct source {
TYPE1 A; --> TYPE1 A;
TYPE2 B; --> TYPE2 B;
TYPE3 C; --> TYPE3 C;
}; TYPE4 D;
TYPE5 E;
};
The optimization is simply moving data from the source to the
appropriate places in the buffer and bypass the reading of
the background data and data conversion.
Tested on smirom, liberty, sol, and copper.
Check in "unique, but sharable" optimization to ISOHM code, which
allows object header messages that are only used in one object to remain in
the sole user's header, but migrates messages that are used in more than one
header into the ISOHM heap.
Tested on:
Mac OS X/32 10.4.9 (amazon)
FreeBSD/32 6.2 (duty)
Linux/32 2.6 (chicago)
Linux/64 2.6 (chicago2)
Change H5[D|G|T]<foo>_expand() "temporary" API routines to
H5[D|G|T]<foo>2() "versioned" routines. Also added
H5[D|G|T](create|commit)_anon() routines to continue to allow "anonymous"
objects to be created in a file.
Tested on:
Mac OS X/32 10.4.9 (amazon)
FreeBSD/32 6.2 (duty)
FreeBSD/64 6.2 (liberty)
Linux/32 2.6 (chicago)
Linux/64 2.6 (chicago2)
Tested platform:
Kagiso only since it is only a comment block change. If it works in one
machine, it should work in all, I hope. Still need to check the parallel
build on copper.
More cleanups on the object header message handling code, to make it
easier to work with and move forward on the creation order coding.
Various other minor cleanups & warning fixes.
Tested on:
FreeBSD/32 6.2 (duty)
Add support & tests for using shared datatypes with shared & un-shared
attributes. Involves some fairly icky code to make the "copy on write"
paradigm for shared attributes work.
Tested on:
Linux/32 2.6 (chicago)
Linux/64 2.6 (chicago2)
Finish implementation of H5Literate() [still needs real tests]
Clean up datatype macro warnings a bit more.
Unify iterator callback macros and put up in public header file, with the
iterator directions.
Tested on:
Mac OS X/32 10.4.8 (amazon)
FreeBSD/32 4.11 (sleipnir)
Linux/32 2.4 (heping)
Linux/64 2.4 (mir)
AIX/32 5.? (copper)
This feature is still in progress; Shared Object Header Messages are not
complete as a feature and are not thoroughly tested. There are still
"TODO" comments in the code (comments with the word "JAMES" in them,
so as not to be confused with other TODO comments).
Hopefully this checkin will reduce the liklihood of conflicts as I finish
implementing this feature.
All current tests pass on juniper, copper (parallel), heping, kagiso, and mir.
Add "use the latest format" support for dataspace object header encode/
decode routines and clean up format a bit for the latest format (new to 1.8.x
releases)
Remove storing 'perm' parameter for array datatypes in memory and the file,
and add test to make certain that if any user applications are attempting to
store them, we get some reports back. (Should be unlikely, since the RefMan
says that the parameter is not implemented and is unsupported).
Carry those changes into the tests, etc.
Clean up a bunch more compiler warnings.
Tested on:
FreeBSD/32 4.11 (sleipnir) w/threadsafe
Linux/32 2.4 (heping) w/FORTRAN & C++
Linux/64 2.4 (mir) w/enable-1.6-compat
Code cleanup
Description:
Trim trailing whitespace in Makefile.am and C/C++ source files to make
diffing changes easier.
Platforms tested:
None necessary, whitespace only change
Description: Mac OS 10.4 on PowerPC chip has some errors to convert (unsigned)
long long to long double. When the bit sequences are 0x003ff..., 0x007fff...,
0x00ffff..., 0x01ffff..., 0x7fffff..., the converted values are twice as big
as they should be.
Solution: Detect the error in configure and disable the compiler conversion
and test case.
Platforms tested: h5committest, Mac OS 10.4, and fuss.
Description: Data conversion from long double to (unsigned) long long
returns some incorrect values on Mac OS 10.4 and SGI IRIX64 6.5. The
conversions start to go wrong when the long double is
20041683600089727.779961 (0x4351ccf385ebc8a0bfcc2a...). If adjusting
the values higher by assigning 0x...c8a0cf... or 0x...c8a0df..., the
converted values go wild.
Solution: Detect this error in configure.in and disable compiler
conversions from long double to (unsigned) long long for Mac OS 10.4
and IRIX64 6.5.
Platforms tested: h5committest, Mac OS 10.4, and IRIX64 6.5.
New feature
Description:
Check in baseline for compact group revisions, which radically revises the
source code for managing groups and object headers.
WARNING!!!! WARNING!!!! WARNING!!!! WARNING!!!! WARNING!!!! WARNING!!!!
WARNING!!!! WARNING!!!! WARNING!!!! WARNING!!!! WARNING!!!! WARNING!!!!
This initiates the "unstable" phase of the 1.7.x branch, leading up
to the 1.8.0 release. Please test this code, but do _NOT_ keep files created
with it - the format will change again before the release and you will not
be able to read your old files!!!
WARNING!!!! WARNING!!!! WARNING!!!! WARNING!!!! WARNING!!!! WARNING!!!!
WARNING!!!! WARNING!!!! WARNING!!!! WARNING!!!! WARNING!!!! WARNING!!!!
Solution:
There's too many changes to really describe them all, but some of them
include:
- Stop abusing the H5G_entry_t structure and split it into two separate
structures for non-symbol table node use within the library: H5O_loc_t
for object locations in a file and H5G_name_t to store the path to
an opened object. H5G_entry_t is now only used for storing symbol
table entries on disk.
- Retire H5G_namei() in favor of a more general mechanism for traversing
group paths and issuing callbacks on objects located. This gets us out
of the business of hacking H5G_namei() for new features, generally.
- Revised H5O* routines to take a H5O_loc_t instead of H5G_entry_t
- Lots more...
Platforms tested:
h5committested and maybe another dozen configurations.... :-)
Code cleanup
Description:
Merge back changes from "compact group" work that improve the
infrastructure of the library and may impact others. In this round of
merging, that includes:
- Move datatype allocation into single internal routine, instead of
duplicated code that was spread out in a dozen or so places.
- Clean up guts of object header routines (H5O_*) to allow for some of
the fancieroperations that need to be performed on groups, along with
some general improvements.
- Added a new error code
- Some minor cleanups in other code....
Platforms tested:
FreeBSD 4.11 (sleipnir)
Linux 2.4
Mac OS X
Description: Changed the names of macros defined in H5Tpkg.h from
H5T_CONV_INTERIM_XXX_XXX to H5T_CONV_INTERNAL_XXX_XXX to be more
descriptive.
Platforms tested: fuss - simple change
Description: Changed the names of the interim macros for data conversion
checked in yesterday from the SOURCE_DESTINATION style (like FP_FP or
LDOUBLE_LLONG) to the H5T_CONV_INTERIM_SOURCE_DESTINATION style (like
H5T_CONV_INTERIM_FP_FP OR H5T_CONV_INTERIM_LDOUBLE_LLONG), to be more
descriptive. These macros are defined in H5Tpkg.h and used in H5Tconv.c
and H5T.c for interim purpose.
Platforms tested: fuss - simple change.
Description: Added 2 new configure options, --enable-exception and --enable-accuracy.
--enable-exception lets the library check whether user's exception handling
functions are present during compiler data conversions and use them if they are.
When it's disabled, this step is skipped to improve conversion speed. This
step isn't implemented yet for soft conversions because there would be little
gain in speed.
--enable-accuracy guarantees data accuracy during data conversions. It means
the library will choose compiler conversions only if the accurate data is
secured. Otherwise, the library will go for the library's own conversions. If
this option is disabled, the library uses compiler conversions in favor of
their speed as long as they work even if data can be incorrect.
Platforms tested: h5committest and fuss. Some systems may fail after this
checkin.
Code cleanup
Description:
Trim trailing whitespace, which is making 'diff'ing the two branches
difficult.
Solution:
Ran this script in each directory:
foreach f (*.[ch] *.cpp)
sed 's/[[:blank:]]*$//' $f > sed.out && mv sed.out $f
end
Platforms tested:
FreeBSD 4.11 (sleipnir)
Too minor to require h5committest
Description: Data conversion function H5T_conv_i32le_f64le() was unused.
Because we have new conversion functions for integers and floating numbers,
this old function is taken out.
Platforms tested: fuss - simple change.
Misc. update:
Optimization
Description:
Speed up I/O on enumerated datatypes (including those nested in compound
datatypes, arrays, etc.) if the destination datatype is a proper superset of
the source datatype.
Solution:
Detect the situation and treat as no-op datatype conversion.
Platforms tested:
FreeBSD 4.11 (sleipnir)
Too minor to require h5committest
Description: Somehow, the hardware conversions between "long double" and other native floating-point
types were left out.
Solution: Added the hardware conversion functions in H5Tconv.c and test cases in dtypes.c.
Platforms tested: h5committest and fuss.
Misc. update: updated MANIFEST to replace bin/reconfigure.sh with bin/reconfigure
Description: Added support of hardware conversion between "long double" and integers(mainly
in H5Tconv.c) and some test cases(mainly in test/dtypes.c).
Platforms tested: h5committest and fuss.
Misc. update: RELEASE.txt
Bug Fix/Code Cleanup/Doc Cleanup/Optimization/Branch Sync :-)
Description:
Generally speaking, this is the "signed->unsigned" change to selections.
However, in the process of merging code back, things got stickier and stickier
until I ended up doing a big "sync the two branches up" operation. So... I
brought back all the "infrastructure" fixes from the development branch to the
release branch (which I think were actually making some improvement in
performance) as well as fixed several bugs which had been fixed in one branch,
but not the other.
I've also tagged the repository before making this checkin with the label
"before_signed_unsigned_changes".
Platforms tested:
FreeBSD 4.10 (sleipnir) w/parallel & fphdf5
FreeBSD 4.10 (sleipnir) w/threadsafe
FreeBSD 4.10 (sleipnir) w/backward compatibility
Solaris 2.7 (arabica) w/"purify options"
Solaris 2.8 (sol) w/FORTRAN & C++
AIX 5.x (copper) w/parallel & FORTRAN
IRIX64 6.5 (modi4) w/FORTRAN
Linux 2.4 (heping) w/FORTRAN & C++
Misc. update:
Code cleanup
Description:
Clear up some inconsistencies, inefficiencies and possible errors between
the release and development branch.
Platforms tested:
FreeBSD 4.10 (sleipnir) w/parallel
Solaris 2.7 (arabica)
Linux 2.4 (heping) w/C++ & FORTRAN
Description: "char" was considered as always "signed char" in data type conversion. However, ISO C leaves
the definition of "char" to individual implementation. i.e. for IBM AIX C compiler, it's treated as "unsigned
char".
Solution: Changed all "char" to "signed char". Don't even do "char" anymore because its definition is up
to each vendor.
Platforms tested: h5committest
Bug fix
Description:
Close a couple of memory leaks
Platforms tested:
FreeBSD 4.10 (sleipnir)
Solaris 2.7 (arabica) w/purify
Linux 2.4 (verbena)
too minor for h5committest
Purpose:
Feature
Description:
Datatypes and groups now use H5FO "file object" code that was previously
only used by datasets. These objects will hold a file open if the file
is closed but they have not yet been closed. If these objects are unlinked
then relinked, they will not be destroyed. If they are opened twice (even
by two different names), both IDs will "see" changes made to the object
using the other ID.
When an object is opened using two different names (e.g., if a dataset was
opened under one name, then mounted and opened under its new name), calling
H5Iget_name() on a given hid_t will return the name used to open that hid_t,
not the current name of the object (this is a feature, and a change from the
previous behavior of datasets).
Solution:
Used H5FO code that was already in place for datasets. Broke H5D_t's, H5T_t's,
and H5G_t's into a "shared" struct and a private struct. The shared structs
(H5D_shared_t, etc.) hold the object's information and are used by all IDs
that point to a given object in the file. The private structs are pointed
to by the hid_t and contain the object's group entry information (including its
name) and a pointer to the shared struct for that object.
This changed the naming of structs throughout the library (e.g., datatype->size
is now datatype->shared->size). I added an updated H5Tinit.c to windows.zip.
Platforms tested:
Visual Studio 7, sleipnir, arabica, verbena
Misc. update:
Code cleanup & minor optimization
Description:
Re-work the way interface initialization routines are specified in the
library to avoid the overhead of checking for them in routines where there is
no interface initialization routine. This cleans up warnings with gcc 3.4,
reduces the library binary size a bit (about 2-3%) and should speedup the
library's execution slightly.
Platforms tested:
FreeBSD 4.10 (sleipnir) w/gcc34
h5committest
Code optimization
Description:
Use 'size_t' instead of 'hsize_t' to track the number of elements in
memory buffers, especially for type conversion.
Platforms tested:
Solaris 2.7 (arabica)
FreeBSD 4.10 (sleipnir) w/parallel
Too minor to require h5committest
Code optimization
Description:
Avoid a memory copy by directly reading from the variable-length sequence
buffer when there is no type conversion on the way to disk.
Platforms tested:
Solaris 2.7 (arabica)
FreeBSD 4.10 (sleipnir) w/parallel
Too minor to require h5committest
Code optimization
Description:
Eliminate many redundant lookups to check for no-op type conversion by
remembering that a type conversion path is the no-op path.
Also, don't allow non-no-op conversions which happen to be no-ops on a
particular machine (such as int<->long conversions on machines where int and
long are the same size and format, etc.) to replace the default no-op
conversion.
Platforms tested:
Solaris 2.7 (arabica)
FreeBSD 4.9 (sleipnir) w/parallel
Bug fix (sorta)
Description:
Add hack to allow the MS Visual Studio 6 compiler to build the library.
It cannot cast unsigned long long values to float or double values. So, add
another configuration macro to disable this conversion in the library. Just
the "hardware" conversion is disabled, so the library will still correctly
convert unsigned long long to float and double values, it will just happen
more slowly with the "software" conversion routine.
Platforms tested:
FreeBSD 4.9 (sleipnir) with "Windows" setting faked
inappropriate for h5committest
Description: The algorithm of H5T_bit_neg wasn't general enough.
Solution: Changed it to handle arbitory starting position and size
in a bit sequence.
Platforms tested: h5committest.
