summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* [svn-r6269] Purpose:Quincey Koziol2003-01-1312-34/+74
| | | | | | | | | | | | | Code cleanup Description: Various code cleanups to allow the development branch to be compiled with a C++ compiler (i.e. CC=g++ ) Platforms tested: Tested h5committest {arabica (fortran), eirene (fortran, C++) modi4 (parallel, fortran)} FreeBSD 4.7 (sleipnir) C++
* [svn-r6268] Purpose:Quincey Koziol2003-01-111-4/+4
| | | | | | | | | | | Code improvement Description: Add extra pair of braces to API versions of FUNC_ENTER/FUNC_LEAVE macros, to make API <-> non-API mismatches obvious. Platforms tested: FreeBSD 4.7 (sleipnir)
* [svn-r6267] Snapshot version 1.5 release 45HDF Admin2003-01-1111-36/+42
|
* [svn-r6266] Purpose:Quincey Koziol2003-01-1084-1526/+1505
| | | | | | | | | | | | Code cleanup/new feature. Description: Split FUNC_LEAVE into API and non-API specific versions. This allows a solution to compiling this branch with C++, as well as reducing the size of the binaries produced. Platforms tested: FreeBSD 4.7 (sleipnir) w/serial, parallel (including MPE) & thread-safe
* [svn-r6264] Purpose:Quincey Koziol2003-01-101-0/+2
| | | | Update manifest with missing test files...
* [svn-r6263] Purpose:Quincey Koziol2003-01-101-2/+7
| | | | | | | | | | | | | | | Update test Description: Some of the performance improvements changed the errors that are reported for creating a duplicated dataset in a file, causing this test's hard-coded errors to check for to fail. Solution: Updated error stack reported for dataset creation failure. Platforms tested: FreeBSD 4.7 (sleipnir) w/threads
* [svn-r6255] Purpose:Quincey Koziol2003-01-0919-231/+1065
| | | | | | | | | | Code cleanup Description: Clean up a few more warnings and update dependencies. Platforms tested: Linux 2.2.18smp (eirene) serial & parallel
* [svn-r6252] Purpose:Quincey Koziol2003-01-09112-3143/+7705
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Lots of performance improvements & a couple new internal API interfaces. Description: Performance Improvements: - Cached file offset & length sizes in shared file struct, to avoid constantly looking them up in the FCPL. - Generic property improvements: - Added "revision" number to generic property classes to speed up comparisons. - Changed method of storing properties from using a hash-table to the TBBT routines in the library. - Share the propery names between classes and the lists derived from them. - Removed redundant 'def_value' buffer from each property. - Switching code to use a "copy on write" strategy for properties in each list, where the properties in each list are shared with the properties in the class, until a property's value is changed in a list. - Fixed error in layout code which was allocating too many buffers. - Redefined public macros of the form (H5open()/H5check, <variable>) internally to only be (<variable>), avoiding innumerable useless calls to H5open() and H5check_version(). - Reuse already zeroed buffers in H5F_contig_fill instead of constantly re-zeroing them. - Don't write fill values if writing entire dataset. - Use gettimeofday() system call instead of time() system when checking the modification time of a dataset. - Added reference counted string API and use it for tracking the names of objects opening in a file (for the ID->name code). - Removed redundant H5P_get() calls in B-tree routines. - Redefine H5T datatype macros internally to the library, to avoid calling H5check redundantly. - Keep dataspace information for dataset locally instead of reading from disk each time. Added new module to track open objects in a file, to allow this (which will be useful eventually for some FPH5 metadata caching issues). - Remove H5AC_find macro which was inlining metadata cache lookups, and call function instead. - Remove redundant memset() calls from H5G_namei() routine. - Remove redundant checking of object type when locating objects in metadata cache and rely on the address only. - Create default dataset object to use when default dataset creation property list is used to create datasets, bypassing querying for all the property list values. - Use default I/O vector size when performing raw data with the default dataset transfer property list, instead of querying for I/O vector size. - Remove H5P_DEFAULT internally to the library, replacing it with more specific default property list based on the type of property list needed. - Remove redundant memset() calls in object header message (H5O*) routines. - Remove redunant memset() calls in data I/O routines. - Split free-list allocation routines into malloc() and calloc()- like routines, instead of one combined routine. - Remove lots of indirection in H5O*() routines. - Simplify metadata cache entry comparison routine (used when flushing entire cache out). - Only enable metadata cache statistics when H5AC_DEBUG is turned on, instead of always tracking them. - Simplify address comparison macro (H5F_addr_eq). - Remove redundant metadata cache entry protections during dataset creation by protecting the object header once and making all the modifications necessary for the dataset creation before unprotecting it. - Reduce # of "number of element in extent" computations performed by computing and storing the value during dataspace creation. - Simplify checking for group location's file information, when file has not been involving in file-mounting operations. - Use binary encoding for modification time, instead of ASCII. - Hoist H5HL_peek calls (to get information in a local heap) out of loops in many group routine. - Use static variable for iterators of selections, instead of dynamically allocation them each time. - Lookup & insert new entries in one step, avoiding traversing group's B-tree twice. - Fixed memory leak in H5Gget_objname_idx() routine (tangential to performance improvements, but fixed along the way). - Use free-list for reference counted strings. - Don't bother copying object names into cached group entries, since they are re-created when an object is opened. The benchmark I used to measure these results created several thousand small (2K) datasets in a file and wrote out the data for them. This is Elena's "regular.c" benchmark. These changes resulted in approximately ~4.3x speedup of the development branch when compared to the previous code in the development branch and ~1.4x speedup compared to the release branch. Additionally, these changes reduce the total memory used (code and data) by the development branch by ~800KB, bringing the development branch back into the same ballpark as the release branch. I'll send out a more detailed description of the benchmark results as a followup note. New internal API routines: Added "reference counted strings" API for tracking strings that get used by multiple owners without duplicating the strings. Added "ternary search tree" API for text->object mappings. Platforms tested: Tested h5committest {arabica (fortran), eirene (fortran, C++) modi4 (parallel, fortran)} Other platforms/configurations tested? FreeBSD 4.7 (sleipnir) serial & parallel Solaris 2.6 (baldric) serial
* [svn-r6236] Purpose: CopyrightBinh-Minh Ribler2003-01-0769-58/+1004
| | | | | | | | | Description: Added copyright notice to C++ API files, including *.h, *.cpp, and Makefile.in Platforms: Linux 6.2 (eirene)
* [svn-r6235] Purpose: CopyrightBinh-Minh Ribler2003-01-071-3/+13
| | | | | | | | Description: Added copyright notice to C++ API files, including *.h, *.cpp, and Makefile.in Platforms: Linux 6.2 (eirene)
* [svn-r6234] Updated with the new MPE feature.Albert Cheng2003-01-041-0/+2
|
* [svn-r6233] Purpose:Albert Cheng2003-01-031-0/+1
| | | | Updated with the newly added file, H5MPprivate.h.
* [svn-r6232] Purpose:Albert Cheng2003-01-035-5/+467
| | | | | | | | | | | | | | | | new feature Description: Added MPICH/MPE instrumentation support. All source code are bracketed by the macro H5_HAVE_MPE. Use "--enable-mpe" to configure it in. Currently only worked in Eirene because the MPE library is not installed in all machines yet. The added file, H5MPprivate.h, holds HDF5/MPE related defintions. Platforms tested: Tested h5committest {arabica (fortran), eirene (fortran, C++) modi4 (parallel, fortran)}? Yep. Other platforms/configurations tested? --enable-mpe feature tested in Eirene.
* [svn-r6231] Purpose:Albert Cheng2003-01-032-4/+151
| | | | | | | | | | | | | | | new feature Description: Added MPICH/MPE instrumentation support. All source code are bracketed by the macro H5_HAVE_MPE. Use "--enable-mpe" to configure it in. Currently only worked in Eirene because the MPE library is not installed in all machines yet. Platforms tested: Tested h5committest {arabica (fortran), eirene (fortran, C++) modi4 (parallel, fortran)}? Yep. Other platforms/configurations tested? --enable-mpe feature tested in Eirene.
* [svn-r6229] Purpose:Albert Cheng2003-01-031-1/+2
| | | | | | | | | | | Minor fix Description: By mistake, the MPI_Error handling macros were inserted outside the overall macro bracket. Moved them back inside the bracket. Platforms tested: Tested h5committest {arabica (fortran), eirene (fortran, C++) modi4 (parallel, fortran)}? Yep.
* [svn-r6228] Description:Albert Cheng2003-01-031-3/+0
| | | | | | | | When I removed the option (--enable-parallel=mpich), which did not work, I failed to run autoheader to update this file. I am checking in the updated version now. Platforms tested: No need to test since it just removed a #undefine and comments.
* [svn-r6225] Purpose:Albert Cheng2003-01-021-3/+7
| | | | | | | | | | Bug fix Description: SGI shell treats $0 as function name if used in a function. Set the name here to avoid that ambiguity and better style too. Platforms tested: Just in modi4 and burrwhite since it is a shell script and the change is machine independent.
* [svn-r6223] Purpose:Binh-Minh Ribler2002-12-222-8/+8
| | | | | | | | | | | | | | | | | | | | | | Bug fix - informed by Kent Description: Some identifiers were flagged as unresolved symbols when building with c++ on Windows. The reason is the name of these identifiers were changed by the c++ compiler, also called name mangling. Solution: Moved #ifdef __cplusplus extern "C" { #endif to include the offending identifiers so the c++ compiler will take them as is. Platforms: SunOS 5.7 (arabica) Linux 6.2 (eirene) Windows 2000
* [svn-r6221] Snapshot version 1.5 release 44HDF Admin2002-12-2110-35/+37
|
* [svn-r6220] Purpose:Bill Wendling2002-12-194-156/+299
| | | | | | | | | | Update Description: A few generic changes to the FPHDF5 code. Some error messages reworked a bit. Cleaning up in case of failure improved in some cases. Added another field to the synchronization messages... Platforms tested: Linux
* [svn-r6219] Purpose:Bill Wendling2002-12-191-1/+0
| | | | | | | | | Update Description: Removed the EFL field from the FPHDF5 structure as the property list already has this info. Platforms tested: Linux
* [svn-r6218] Purpose:Bill Wendling2002-12-191-47/+17
| | | | | | | | | | Bug Fix Description: Was encoding a value as DTYPE instead of MTIME, which would cause a couple of errors. Also, got rid of the EFL field as it's encoded in the PLIST already... Platforms tested: Linux...Modest test, FPHDF5 not fully implemented just yet.
* [svn-r6217] Elena Pourmal2002-12-191-1/+89
| | | | | | | | | | | | | Purpose: Bug fix Description: man pages for the following F90 functions were missing: h5tget_super_f h5tarray_create_f h5tget_array_ndims_f h5tget_array_dims_f Solution: Added missing man pages Platforms tested: IE
* [svn-r6215] Added Files:Albert Cheng2002-12-184-87/+56
| | | | | | | | | | | | | | | fortran/config/powerpc-ibm-aix5.x To cover AIX 5.x. Removed Files: fortran/config/powerpc-ibm-aix4.2.1.0 fortran/config/powerpc-ibm-aix4.3.2.0 These files are outdated and have been replaced by powerpc-ibm-aix4.x. Modified Files: MANIFEST Update it. Platforms tested: Only on LLNL blue. LLNL SP are the only machines that use these files and that I have access.
* [svn-r6213] Raymond Lu2002-12-161-29/+54
| | | | | | | | | | | Purpose: bug fix Description: some arrays were too big, running out of memory limit for some machines. Solution: change to dynamic memory allocation. Platforms tested: arabica, sleipnir
* [svn-r6210] Frank Baker2002-12-161-18/+129
| | | | | | | | | | Purpose: Adding H5Gmove2 and H5Glink2. Description: Added H5Gmove2 and H5Glink2. Renamed H5Gmove parameters for consistency. Platforms tested: IE 5
* [svn-r6209] Frank Baker2002-12-161-6/+55
| | | | | | | | | | | Purpose: Adding new H5Fget_obj_* functions. (Development branch only) Description: Added H5Fget_obj_count. Stubbed in H5Fget_obj_ids, but the entry is not filled in. Reordered entries to group similar functions. Platforms tested: IE 5
* [svn-r6208] Raymond Lu2002-12-161-0/+2
| | | | | Misc. update: Update an error assertion improvement for nil VL string.
* [svn-r6207] Raymond Lu2002-12-161-0/+100
| | | | | | | | | Purpose: Add more test Description: More tests on nil and zero-sized VL strings. Platforms tested: arabica, eirene
* [svn-r6206] Raymond Lu2002-12-131-2/+2
| | | | | | | | | Purpose: Improve error handling Description: make error handling better for VL string null pointer. Platform tested: arabica
* [svn-r6205] MuQun Yang2002-12-131-0/+0
| | | | | | | | | | | | | Purpose: update windows project file Description: Solution: Platforms tested: windows 2000(not affect other platforms) Misc. update: Update MANIFEST if you add or remove any file. Update release_docs/RELEASE for bug fixes, new features, etc. Update applicable document files too.
* [svn-r6204] MuQun Yang2002-12-131-32/+62
| | | | | | | | | | | | | | | | | Purpose: bug fix Description: Need to use HDopen, HDread for windows support. off_t in windows is 32-bit; however, windows can support 64-bit file. So we have to change off_t to int64 in order that >4G file can be supported. Solution: See above Platforms tested: linux2.2.18smp, windows 2000 Misc. update: Update MANIFEST if you add or remove any file. Update release_docs/RELEASE for bug fixes, new features, etc. Update applicable document files too.
* [svn-r6203] MuQun Yang2002-12-131-5/+6
| | | | | | | | | | | | | | | | Purpose: bug fixed for windows Description: windows fopen functions by default opens a text mode file. It will fail for HDF5 file. So to use HDopen HDread HDclose for test. Solution: Change from fopen, fread, fclose to HDopen, HDread, HDclose. Platforms tested: linux2.2.18smp, windows 2000 Misc. update: Update MANIFEST if you add or remove any file. Update release_docs/RELEASE for bug fixes, new features, etc. Update applicable document files too.
* [svn-r6202] MuQun Yang2002-12-133-4/+12
| | | | | | | | | | | | | | | | Purpose: Fix bugs for windows platform. Description: 1. Update HDlseek for windows 2. Update H5Otime.c for normal time(non-daylight time) 3. Add H5_DLL in front of two functions so that windows DLL can work. Solution: Platforms tested: windows 2000, eirene(since most updates have WIN32 around it, not need to test other platforms) Misc. update: Update MANIFEST if you add or remove any file. Update release_docs/RELEASE for bug fixes, new features, etc. Update applicable document files too.
* [svn-r6201] Raymond Lu2002-12-121-6/+6
| | | | | | | | | | | Purpose: Bug fix(kinda) Description: Windows handle "fopen, fseek, fread, fclose" C functions slightly different. Solution: change to HDopen, HDlseek, HDread, HDclose to cover both Unix and Windows. Platforms tested: arabica
* [svn-r6199] Raymond Lu2002-12-121-2/+3
| | | | | | | Purpose: error-detection improvement Description: verify string parameter in function H5T_vlen_str_mem_getlen isn't nil.
* [svn-r6196] Raymond Lu2002-12-111-7/+477
| | | | | | | | | Purpose: more tests. Description: a few more test cases. Platforms tested: modi4, eirene, arabica
* [svn-r6195] Raymond Lu2002-12-114-105/+249
| | | | | | | | | | Purpose: Cleanup H5Tget_native_type code Description: improve code cosmetics. Solution: Platforms tested: modi4, eirene, arabica
* [svn-r6193] Snapshot version 1.5 release 43HDF Admin2002-12-0710-35/+37
|
* [svn-r6191] ./hdf5-devel/bin/traceRobb Matzke2002-12-041-3/+0
| | | | | | | | | | | Purpose: Missed a file Description: Missed the `trace' changes. Solution: Committing again Platforms tested: None
* [svn-r6190] ./hdf5-devel/tools/testfiles/tvldtypes1.lsRobb Matzke2002-12-041-3/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Purpose: New Feature; Optimization; Clean-up (Merged from 1.4 branch) Description: There is no symbolic constant to pass to functions that take an optional object ID for when the caller wants to indicate no object ID. In the past the caller always passed a negative integer. GPFS performs poorly. The h5ls tool decides whether to list the file name in the output based on a compile-time choice, which isn't always optimal at run time. Solution: Added a symbolic constant H5I_INVALID_HID. Added code to tell the mmfsd of GPFS to forego byte range token prefetching. h5ls decides whether to print the file name at runtime based on the number of objects being listed. Platforms tested: SuSE Linux (arborea), gcc and mpich-1.2.4 SunOS (baldric), gcc
* [svn-r6189] ./hdf5-devel/tools/testfiles/tstr-1.lsRobb Matzke2002-12-041-5/+5
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Purpose: New Feature; Optimization; Clean-up (Merged from 1.4 branch) Description: There is no symbolic constant to pass to functions that take an optional object ID for when the caller wants to indicate no object ID. In the past the caller always passed a negative integer. GPFS performs poorly. The h5ls tool decides whether to list the file name in the output based on a compile-time choice, which isn't always optimal at run time. Solution: Added a symbolic constant H5I_INVALID_HID. Added code to tell the mmfsd of GPFS to forego byte range token prefetching. h5ls decides whether to print the file name at runtime based on the number of objects being listed. Platforms tested: SuSE Linux (arborea), gcc and mpich-1.2.4 SunOS (baldric), gcc
* [svn-r6188] ./hdf5-devel/tools/testfiles/tslink-1.lsRobb Matzke2002-12-041-2/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Purpose: New Feature; Optimization; Clean-up (Merged from 1.4 branch) Description: There is no symbolic constant to pass to functions that take an optional object ID for when the caller wants to indicate no object ID. In the past the caller always passed a negative integer. GPFS performs poorly. The h5ls tool decides whether to list the file name in the output based on a compile-time choice, which isn't always optimal at run time. Solution: Added a symbolic constant H5I_INVALID_HID. Added code to tell the mmfsd of GPFS to forego byte range token prefetching. h5ls decides whether to print the file name at runtime based on the number of objects being listed. Platforms tested: SuSE Linux (arborea), gcc and mpich-1.2.4 SunOS (baldric), gcc
* [svn-r6187] ./hdf5-devel/tools/testfiles/tsaf.lsRobb Matzke2002-12-041-38/+38
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Purpose: New Feature; Optimization; Clean-up (Merged from 1.4 branch) Description: There is no symbolic constant to pass to functions that take an optional object ID for when the caller wants to indicate no object ID. In the past the caller always passed a negative integer. GPFS performs poorly. The h5ls tool decides whether to list the file name in the output based on a compile-time choice, which isn't always optimal at run time. Solution: Added a symbolic constant H5I_INVALID_HID. Added code to tell the mmfsd of GPFS to forego byte range token prefetching. h5ls decides whether to print the file name at runtime based on the number of objects being listed. Platforms tested: SuSE Linux (arborea), gcc and mpich-1.2.4 SunOS (baldric), gcc
* [svn-r6186] ./hdf5-devel/tools/testfiles/tnestcomp-1.lsRobb Matzke2002-12-041-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Purpose: New Feature; Optimization; Clean-up (Merged from 1.4 branch) Description: There is no symbolic constant to pass to functions that take an optional object ID for when the caller wants to indicate no object ID. In the past the caller always passed a negative integer. GPFS performs poorly. The h5ls tool decides whether to list the file name in the output based on a compile-time choice, which isn't always optimal at run time. Solution: Added a symbolic constant H5I_INVALID_HID. Added code to tell the mmfsd of GPFS to forego byte range token prefetching. h5ls decides whether to print the file name at runtime based on the number of objects being listed. Platforms tested: SuSE Linux (arborea), gcc and mpich-1.2.4 SunOS (baldric), gcc
* [svn-r6185] ./hdf5-devel/tools/testfiles/tloop-1.lsRobb Matzke2002-12-041-4/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Purpose: New Feature; Optimization; Clean-up (Merged from 1.4 branch) Description: There is no symbolic constant to pass to functions that take an optional object ID for when the caller wants to indicate no object ID. In the past the caller always passed a negative integer. GPFS performs poorly. The h5ls tool decides whether to list the file name in the output based on a compile-time choice, which isn't always optimal at run time. Solution: Added a symbolic constant H5I_INVALID_HID. Added code to tell the mmfsd of GPFS to forego byte range token prefetching. h5ls decides whether to print the file name at runtime based on the number of objects being listed. Platforms tested: SuSE Linux (arborea), gcc and mpich-1.2.4 SunOS (baldric), gcc
* [svn-r6184] ./hdf5-devel/tools/testfiles/tgroup-1.lsRobb Matzke2002-12-041-14/+14
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Purpose: New Feature; Optimization; Clean-up (Merged from 1.4 branch) Description: There is no symbolic constant to pass to functions that take an optional object ID for when the caller wants to indicate no object ID. In the past the caller always passed a negative integer. GPFS performs poorly. The h5ls tool decides whether to list the file name in the output based on a compile-time choice, which isn't always optimal at run time. Solution: Added a symbolic constant H5I_INVALID_HID. Added code to tell the mmfsd of GPFS to forego byte range token prefetching. h5ls decides whether to print the file name at runtime based on the number of objects being listed. Platforms tested: SuSE Linux (arborea), gcc and mpich-1.2.4 SunOS (baldric), gcc
* [svn-r6183] ./hdf5-devel/tools/testfiles/tdset-1.lsRobb Matzke2002-12-041-2/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Purpose: New Feature; Optimization; Clean-up (Merged from 1.4 branch) Description: There is no symbolic constant to pass to functions that take an optional object ID for when the caller wants to indicate no object ID. In the past the caller always passed a negative integer. GPFS performs poorly. The h5ls tool decides whether to list the file name in the output based on a compile-time choice, which isn't always optimal at run time. Solution: Added a symbolic constant H5I_INVALID_HID. Added code to tell the mmfsd of GPFS to forego byte range token prefetching. h5ls decides whether to print the file name at runtime based on the number of objects being listed. Platforms tested: SuSE Linux (arborea), gcc and mpich-1.2.4 SunOS (baldric), gcc
* [svn-r6182] ./hdf5-devel/tools/testfiles/tcomp-1.lsRobb Matzke2002-12-041-10/+10
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Purpose: New Feature; Optimization; Clean-up (Merged from 1.4 branch) Description: There is no symbolic constant to pass to functions that take an optional object ID for when the caller wants to indicate no object ID. In the past the caller always passed a negative integer. GPFS performs poorly. The h5ls tool decides whether to list the file name in the output based on a compile-time choice, which isn't always optimal at run time. Solution: Added a symbolic constant H5I_INVALID_HID. Added code to tell the mmfsd of GPFS to forego byte range token prefetching. h5ls decides whether to print the file name at runtime based on the number of objects being listed. Platforms tested: SuSE Linux (arborea), gcc and mpich-1.2.4 SunOS (baldric), gcc
* [svn-r6181] ./hdf5-devel/tools/testfiles/tarray1.lsRobb Matzke2002-12-041-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Purpose: New Feature; Optimization; Clean-up (Merged from 1.4 branch) Description: There is no symbolic constant to pass to functions that take an optional object ID for when the caller wants to indicate no object ID. In the past the caller always passed a negative integer. GPFS performs poorly. The h5ls tool decides whether to list the file name in the output based on a compile-time choice, which isn't always optimal at run time. Solution: Added a symbolic constant H5I_INVALID_HID. Added code to tell the mmfsd of GPFS to forego byte range token prefetching. h5ls decides whether to print the file name at runtime based on the number of objects being listed. Platforms tested: SuSE Linux (arborea), gcc and mpich-1.2.4 SunOS (baldric), gcc
generated by cgit v0.12 at 2024-11-09 14:25:18 (GMT)