summaryrefslogtreecommitdiffstats
path: root/src/H5FDmulti.c
Commit message (Collapse)AuthorAgeFilesLines
* [svn-r7411] Purpose:Quincey Koziol2003-08-261-2/+1
| | | | | | | | | | | Code cleanup Description: De-linted more modules Platforms tested: FreeBSD 4.8 (sleipnir) w/ & w/o stream enabled too minor to require h5committest
* [svn-r7393] Purpose:Quincey Koziol2003-08-231-89/+92
| | | | | | | | | | | Code cleanup Description: De-linted more code Platforms tested: FreeBSD 4.8 (sleipnir) w/parallel too minor to require h5committest
* [svn-r7299] Purpose:Quincey Koziol2003-08-081-34/+34
| | | | | | | | | | Code refactoring Description: Chase new error API usage by adding error class to calls to H5Epush(). Platforms tested: h5committested
* [svn-r7265] *** empty log message ***Raymond Lu2003-07-261-26/+26
|
* [svn-r6501] Purpose:Bill Wendling2003-03-191-4/+17
| | | | | | | | | | | | | | New Feature Description: Added support for the lock and unlock function calls for file drivers. Only FPHDF5 uses this feature. All of these drivers don't define lock or unlock methods. Platforms tested: Linux, Modi4, Sol Misc. update:
* [svn-r6387] Purpose:Quincey Koziol2003-02-101-13/+9
| | | | | | | | | | | | | | | | | | | | | | | | | | | Bug Fix Description: Metadata cache in parallel I/O can cause hangs in applications which perform independent I/O on chunked datasets, because the metadata cache can attempt to flush out dirty metadata from only a single process, instead of collectively from all processes. Solution: Pass a dataset transfer property list down from every API function which could possibly trigger metadata I/O. Then, split the metadata cache into two sets of entries to allow dirty metadata to be set aside when a hash table collision occurs during independent I/O. Platforms tested: Tested h5committest {arabica (fortran), eirene (fortran, C++) modi4 (parallel, fortran)} FreeBSD 4.7 (sleipnir) serial & parallel Misc. update: Updated release_docs/RELEASE
* [svn-r6252] Purpose:Quincey Koziol2003-01-091-6/+18
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | 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-r5951] Raymond Lu2002-09-301-0/+35
| | | | | | | | | | | Purpose: New API functions Description: Added API functions to return pointer to low-level file handle (H5Fget_vfd_handle and H5FDget_vfd_handle) and related property list setting functions(H5Pset_family_offset and H5Pset_multi_type). Platforms tested: Linux 2.2(eirene), Solaris 2.7(arabica), IRIX64 6.5(modi4)
* [svn-r5536] Purpose:Quincey Koziol2002-06-051-0/+1
| | | | | | | | | | | | | | | | | New feature. Description: Added a "small data" block allocation mechanism to the library, similar to the mechanism used for allocating metadata currently. See the RFC for more details: http://hdf.ncsa.uiuc.edu/RFC/SmallData/SmallData.html This reduces the number of I/O operations which hit the disk for my test program from 19 to 15 (i.e. from 393 to 15, overall). Platforms tested: Solaris 2.7 (arabica) w/FORTRAN and FreeBSD 4.5 (sleipnir) w/C++
* [svn-r5444] Purpose:Quincey Koziol2002-05-201-24/+24
| | | | | | | | | | Code cleanup Description: Clean up warnings on IRIX64 6.5 (modi4) Platforms tested: IRIX64 6.5 (modi4)
* [svn-r5440] Purpose:Quincey Koziol2002-05-201-4/+4
| | | | | | | | | | | New feature Description: Add 'closing' parameter to H5FDflush and VFL "flush" functions, per http://hdf.ncsa.uiuc.edu/RFC/VFLFlush/VFLFlush.html Platforms tested: IRIX64 6.5 (modi4)
* [svn-r5403] Purpose:Quincey Koziol2002-05-131-3/+3
| | | | | | | | | | Back out changes. Description: Back out changes to VFL 'flush' API function, pending review. Platforms tested: IRIX64 6.5 (modi4)
* [svn-r5393] Purpose:Quincey Koziol2002-05-101-3/+3
| | | | | | | | | | | | | | | | | | | | | | New Feature Description: The VFL flush function is called immediately before a file is closed. This can cause duplicate syncronization actions to occur, if the VFL close function also performs them. Solution: Added 'closing' parameter to VFL 'flush' operation. This allows the VFL flush function to bypass operations that will be duplicated within the VFL close function. Additionally, use the 'closing' parameter to bypass calls to MPI_File_sync() when set. Since MPI_File_close() also syncronizes the file, this avoids the terrible performance hit taken when calling MPI_File_sync() as the file is closing. Platforms tested: IRIX64 6.5 (modi4)
* [svn-r5165] Purpose:Albert Cheng2002-04-101-1/+13
| | | | | | | | | | | Feature Description: Added comments that this is intended as an example of file driver. Added a macro detection at the end to flag if non-public definitions are used. (only check for H5private.h which is mostly likely used. Not bullet proof at all.) Platforms tested: eirene.
* [svn-r4696] Raymond Lu2001-12-111-0/+1
| | | | | | | | | | | | | | | | | Purpose: Modify H5Fclose behavior Description: The HDF5 actual file close behaves in several ways in terms of if there are still objects(dataset, group, datatype) opened in file. Solution: Added a new file access property, file close degree. It has four values, H5F_CLOSE_DEFAULT H5F_CLOSE_WEAK H5F_CLOSE_SEMI H5F_CLOSE_STRONG The way a file is closed is decided by these values. Platforms tested: IRIX64 6.5, SunOS 5.6, FreeBSD 4.4
* [svn-r4620] Purpose:Quincey Koziol2001-11-201-3/+0
| | | | | | | | | Code cleanup Description: Get rid of IDs from internal function calls and some small cleanups from the old-stype => generic property list conversion. Platforms tested: FreeBSD 4.4 (hawkwind)
* [svn-r4589] Purpose:Quincey Koziol2001-11-031-2/+4
| | | | | | | | Code cleanup Description: Clean up various compiler warnings from generic property updates. Platforms tested: FreeBSD 4.4 (hawkwind)
* [svn-r4584] Raymond Lu2001-11-021-4/+4
| | | | | | | | | Purpose: Switch from old property list to new generic property list. Description: Mainly changed H5Pcreat, H5Pclose, H5Pcopy. Platforms tested: IRIX64 6.5, FreeBSD, SunOS 5.7.
* [svn-r4572] Raymond Lu2001-10-251-4/+13
| | | | | | | Purpose: Followup file access property list changes. Platforms tested: IRIX64, SunOS 5.7, FreeBSD.
* [svn-r4473] Purpose:Quincey Koziol2001-09-251-25/+25
| | | | | | | | | | | | | | | | | | | | | Code cleanup for better compatibility with C++ compilers Description: C++ compilers are choking on our C code, for various reasons: we used our UNUSED macro incorrectly when referring to pointer types we used various C++ keywords as variables, etc. we incremented enum's with the ++ operator. Solution: Changed variables, etc.to avoid C++ keywords (new, class, typename, typeid, template) Fixed usage of UNUSED macro from this: char UNUSED *c to this: char * UNUSED c Switched the enums from x++ to x=x+1 Platforms tested: FreeBSD 4.4 (hawkwind)
* [svn-r4444] Purpose:Albert Cheng2001-09-171-2/+25
| | | | | | | | | | | | | | | | | Feature Description: The H5Pset_fapl_split() used to assume both meta and raw files have the same prefix name. This prohibited the option to specify the two files reside on different file systems. E.g., having meta-file as /user/home/data1.meta and raw-file as /pfs/data1.raw. Solution: Added feature that if the raw or meta extension string contains a "%s", it will be substituted by the filename given for H5Fopen or H5Fcreate. This is same as the multi-file syntax. If no %s is found, one is inserted at the beginning. This is the previous behavior. Platforms tested: Modi4 (parallel, serial) and eirene (parallel).
* [svn-r4312] Purpose:Quincey Koziol2001-08-061-4/+0
| | | | | | | | | Feature shift Description: Take out the v1.2.x compatibility stubs and put in the hooks for v1.4.x compatibility when needed. Platforms tested: FreeBSD 4.3 (hawkwind)
* [svn-r4181] Purpose:Quincey Koziol2001-07-101-4/+4
| | | | | | | | | Bug Fix, Code Cleanup, Code Optimization, etc. Description: Fold in the hyperslab speedups, clean up compile warnings and change a few things from using 'unsigned' or 'hsize_t' to use 'size_t' instead. Platforms tested: FreeBSD 4.3 (hawkwind), Solaris 2.7 (arabica), Irix64 6.5 (modi4)
* [svn-r3781] Purpose:Bill Wendling2001-04-051-3/+4
| | | | | | | | | | | | | | | | | | Update Description: Changed #include <hdf_file.h> construct to #include "hdf_file.h" so that the GNU compiler can more easily pick up the dependencies which it places in the .depend and Dependencies files. Also regenerated the Dependencies to go along with this. Platforms tested: Linux
* [svn-r3252] Purpose:Quincey Koziol2001-01-091-73/+79
| | | | | | | | | | | | | Code cleanup. Description: Fixed _lots_ (I mean _tons_) of warnings spit out by the gcc with the extra warnings. Including a few show-stoppers for compression on IRIX machines. Solution: Changed lots of variables' types to more sensible and consistent types, more range-checking, more variable typecasts, etc. Platforms tested: FreeBSD 4.2 (hawkwind), IRIX64-64 (modi4)
* [svn-r3220] ./hdf5/src/H5FDmulti.cRobb Matzke2000-12-291-2/+3
| | | | | | | 2000-12-29 08:55:42 Robb Matzke <matzke@llnl.gov> * H5FD_multi_query: The `flags' argument is advertised to be output only. Therefore I added code to zero its value before the feature bits are assigned.
* [svn-r3010] Purpose:Quincey Koziol2000-11-281-5/+5
| | | | | | | | | | | | | | Bug fix Description: When the v1.2 compatibility code was turned on, internal functions in the library were getting confused. Solution: Separated guts of H5Pget_driver call into an API function (the definition of which depends on the compatibility switch) and an internal function which always behaves like the v1.3/4 function. Replaced API function calls in the library code with the internal function. Platforms tested: FreeBSD 4.2 (hawkwind)
* [svn-r3005] Purpose:Quincey Koziol2000-11-271-2/+2
| | | | | | | | Backward compatibility code Description: Add in code to allow the library to emulate the v1.2 API and behavior. Platforms tested: FreeBSD 4.2 (hawkwind)
* [svn-r2722] Purpose:Quincey Koziol2000-10-241-3/+6
| | | | | | | | | | Feature symmetry Description: A while ago I needed to get the 'type' of data being accessed during writes to the VFL driver, so I put in code to get the information down there. Albert asked for the same information during reads, so I've added that in. Tested: FreeBSD 4.1.1 (hawkwind)
* [svn-r2600] Purpose:Quincey Koziol2000-09-261-1/+31
| | | | | | | | | | | Implemented new feature Description: Added data sieve buffering code to raw I/O data path. This is enabled for all the VFL drivers except the mpio & core drivers. Also added two new API functions to control the sieve buffer size: H5Pset_sieve_buf_size() and H5Pget_sieve_buf_size(). Platforms tested: Solaris 2.6 (i.e. baldric)
* [svn-r2498] Added new VFL 'query' code and added new 'type of data' ↵Quincey Koziol2000-08-311-3/+4
| | | | | | parameter to write call.
* [svn-r2032] Purpose:Albert Cheng2000-03-141-1/+1
| | | | | | | | | | | | | | | | | | Bug Fix. Description: Tests failed if drivers multi or split is used (split is implemented as a special case of multi). When a file created via those two drivers is opened again, the file sizes retrieved are incorrect. These seemed to happen to non-little-endian machines only (sun, sgi failed, but linux succeed.) Solution: The error was because the superblock encoding code was keeping track in nseen the pairs of HADDR needed to be converted but did not tell the H5Tconvert 2*nseen to convert. Thus only the first half of the addresses were converted, the rest were left alone. Corrected the numbers of addresses to convert. Platforms tested: IRIX64, Solaris 2.7, Linux.
* [svn-r1916] Changed the #define TRUE to be consistent with the one in ↵Albert Cheng1999-12-201-1/+1
| | | | | | H5private.h which is also used by most Unix systems.
* [svn-r1836] Mainly fixed a bug in VL datatype comparisons which was causing ↵Quincey Koziol1999-11-171-27/+27
| | | | | | | non-equal VL types to compare as equal. Added some asserts to make certain nothing slips through again. Also cleaned up a few warnings from the SGI compiler.
* [svn-r1788] Converted all the VFL drivers except for the stdio and multi ↵Quincey Koziol1999-10-231-52/+175
| | | | | | drivers back into the "internal" HDF5 coding standard.
* [svn-r1695] Fixed various compiler warnings..Quincey Koziol1999-09-301-2/+2
|
* [svn-r1689] Mainly adding support for "native" variable-length strings (C ↵Quincey Koziol1999-09-291-3/+7
| | | | | | | | only currently), but I fixed lots of misc. compiler warnings in other code and also tracked down the memory overwrite bug that was causing the development branch to core dump on most machines.
* [svn-r1585] Changes since 19990820Robb Matzke1999-08-241-190/+471
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | ---------------------- ./src/H5D.c Added additional elements to a variable initializer in H5Dvlen_get_buf_size() to shut up a warning message. Also added the API tracing call. ./src/H5F.c Added file opening optimizations. If the driver doesn't support the ability to determine when two file handles refer to the same file (like MPIO and GASS) then H5F_open() makes fewer calls to the driver's open callback. Also, if the tentative file access flags are the same as the original flags then H5F_open() makes fewer calls to the file device. ./src/H5FD.c ./src/H5FDprivate.h ./src/H5FDpublic.h Added H5FD_get_class() so the library can get information about what file driver callbacks are defined. This will be useful when more optimization functions are added to the VFL, such as for MPIO derived datatype I/O. ./src/H5FDcore.c ./src/H5FDfamily.c ./src/H5FDmpio.c ./src/H5FDmulti.c ./src/H5FDsec2.c The driver symbols (like H5FD_CORE, etc) are actually function calls. The functions were fixed to return correct values even after calling H5close(). ./src/H5FDmulti.c ./src/H5FDmulti.h Added support for opening a file when parts are missing (only if the caller explicitly allows that in the file access property list). Moved some common code sequences into macros or functions. Added better support for reopening files. All the application has to know is that the file is a multi file and the base name from which all the member names are created. More debugging output when the file is opened with the H5F_ACC_DEBUG flag. Fixed various bugs. ./src/H5Fistore.c Chunked raw data was accidently allocated as meta data instead of raw data. ./src/H5I.c The H5Iget_type() function fails when invoked with an old object ID (an ID which has been closed down). ./test/h5test.c Added an extra argument when setting the multi file access property lists so the test fails if it can't open one of the sub-files. ./tools/h5ls.c Improved the algorithm for deciding what file driver to use. It basically tries all of the predefined drivers and is now able to open family, split, and multi files without looking for special characters in the file name. Added `-e' and `--errors' switches which cause errors from libhdf5 to be reported on stderr in addition to the simple error message displayed by h5ls.
* [svn-r1572] Changes since 19990810Robb Matzke1999-08-171-0/+1393
---------------------- ./MANIFEST ./src/H5FDmulti.c [NEW] ./src/H5FDmulti.h [NEW] ./src/Makefile.in ./src/hdf5.h The split driver was reimplemented as a more general "multi" driver which is capable of splitting data into multiple files like the family driver except the partioning is done by memory usage type instead of address. The H5Pset_fapl_split() function just calls H5Pset_fapl_multi() with arguments which prepare to split the address space into two files: meta and raw data. This is the first version. I plan to allow the open() call to relax a bit which would allow one to open an hdf5 file when only the meta-data file is present. This would allow a very large file to be split and stored on tape and the relatively small meta file to be mirrored on disk to allow limited browsing of the file (any request for raw data would fail). ./src/H5private.h ./src/H5F.c ./src/H5FD.c ./src/H5FDprivate.h ./src/H5FDpublic.h ./src/H5FDcore.c ./src/H5FDfamily.c ./src/H5FDmpio.c ./src/H5FDsec2.c Added the ability for a file driver to store information in the superblock which would be needed if the file were opened again later for reading. The format is driver-defined which allows users to extend it however they like. ./doc/html/H5.format.html Added information about the new driver information block of the superblock. This is where file drivers store information they need in order to reopen the file later. ./src/H5F.c ./src/H5Fprivate.h ./src/H5FD.c ./src/H5FDprivate.h ./src/H5FDpublic.h ./src/H5FDcore.c ./src/H5FDfamily.c ./src/H5FDmpio.c ./src/H5FDsec2.c ./src/H5Fistore.c ./src/H5R.c The file access properties and the file access property list were decoupled, which allows the property list to more cleanly contain properties for various levels of the file and which allows the property list to be modified more cleanly when opening files. ./src/H5.c ./src/H5FDpublic.h Removed H5FD_MEM_META and H5FD_MEM_GROUP since they're never used. ./src/H5D.c Changed the way we detect the MPIO driver in all these special cases. ./src/H5F.c ./src/H5Rpublic.h ./test/tfile.c The default file sizeof(offset) was changed to be a function of haddr_t instead of hsize_t. THE H5RPUBLIC.H DEFINITIONS WILL HAVE PROBLEMS IF THE USER CREATES A FILE WITH NON-DEFAULT OFFSET AND SIZE SIZES! ./src/H5F.c Fixed an uninitialized memory access bug in file closing related to the VFL. ./src/H5T.c ./src/H5Tpublic.h Added an H5T_NATIVE_HADDR predefined datatype which corresponds to the `haddr_t' type. ./test/Makefile.in Reformatted long lines. ./test/big.c ./test/cmpd_dset.c ./test/dsets.c ./test/dtypes.c ./test/extend.c ./test/external.c Removed the H5F_ACC_DEBUG flag from file creation/open calls. ./test/big.c Plugged a memory leak. ./test/h5test.c Added support for the `multi' driver. Removed #warning about not having the stdio driver. Plans are to not implement it since the sec2 driver serves the same purpose and testing didn't show any difference in execution times between the two.
generated by cgit v0.12 at 2024-11-02 22:16:51 (GMT)