summaryrefslogtreecommitdiffstats
path: root/src/H5Cprivate.h
Commit message (Collapse)AuthorAgeFilesLines
* [svn-r17411] Description:Quincey Koziol2009-08-241-2/+0
| | | | | | | | | | | | | | | | | | | | | | | Make H5AC_flush just flush the cache and make H5AC_dest perform the proper parallel synchronization before destroying the cache. Also, further discriminate between 'closing' and 'non-closing' actions in H5F_flush. 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 (kagiso) w/PGI compilers, w/C++ & FORTRAN, w/threadsafe, in debug mode Linux/64-amd64 2.6 (smirom) 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, in production mode Linux/64-ia64 2.6 (cobalt) w/Intel compilers, w/C++ & FORTRAN, in production mode Linux/64-ia64 2.4 (tg-login3) w/parallel, w/FORTRAN, in production mode Linux/64-amd64 2.6 (abe) w/parallel, w/FORTRAN, in production mode Mac OS X/32 10.5.8 (amazon) in debug mode Mac OS X/32 10.5.8 (amazon) w/C++ & FORTRAN, w/threadsafe, in production mode
* [svn-r16982] Description:Quincey Koziol2009-05-261-2/+4
| | | | | | | | | | | | | | | | | | | | | | | | Add library private routines to allow metadata cache entries to be set as flush dependency children on extensible array entries. This will allow chunk proxies to be set as child flush dependencies for the extensible array, adding another necessary layer to the flush dependency chain for SWMR access. 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/C++ & FORTRAN, w/threadsafe, in debug mode Linux/64-amd64 2.6 (smirom) 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, in production mode Linux/64-ia64 2.6 (cobalt) w/Intel compilers, w/C++ & FORTRAN, in production mode Linux/64-ia64 2.4 (tg-login3) w/parallel, w/FORTRAN, in debug mode Linux/64-amd64 2.6 (abe) w/parallel, w/FORTRAN, in production mode Mac OS X/32 10.5.6 (amazon) in debug mode Mac OS X/32 10.5.6 (amazon) w/C++ & FORTRAN, w/threadsafe, in production mode
* [svn-r16938] Purpose:Mike McGreevy2009-05-121-1/+1
| | | | | | | | | | | | | | | | | | Modifying default cache configuration, and adding an #ifdef to allow for a separate default configuration when parallel is enabled. This is being modified in order to address an observed performance problem with the current default configuration. Description of Changes: - increasing maximum cache size from 16MB to 32MB - increasing maximum entry size from 10MB to 32MB - decreasing min_clean_fraction from 0.3 to 0.01 in serial case - increasing flash_multiple from 1.0 to 1.4 in serial case Tested: jam
* [svn-r16632] Description:Quincey Koziol2009-03-311-4/+2
| | | | | | | | | | | | | | | | | | | | | | Remove unnecessary 'udata' parameter from H5AC_set() and 'notify' callback (introduced with checkin that added the 'notify' callback) 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/C++ & FORTRAN, w/threadsafe, in debug mode Linux/64-amd64 2.6 (smirom) 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, in production mode Linux/64-ia64 2.6 (cobalt) w/Intel compilers, w/C++ & FORTRAN, in production mode Linux/64-ia64 2.4 (tg-login3) w/parallel, w/FORTRAN, in production mode Linux/64-amd64 2.6 (abe) w/parallel, w/FORTRAN, in production mode Mac OS X/32 10.5.6 (amazon) in debug mode Mac OS X/32 10.5.6 (amazon) w/C++ & FORTRAN, w/threadsafe, in production mode
* [svn-r16630] Description:Quincey Koziol2009-03-311-1/+19
| | | | | | | | | | | | | | | | | | | | | | | | | Add 'notify' callback to metadata cache clients, so that they can be aware when the cache has taken/will take certain actions. Notifications are urrently limited to when an entry has successfully entered the cache (via an insertion or a load from the file) and when an entry is about to be evicted from the cache. 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/C++ & FORTRAN, w/threadsafe, in debug mode Linux/64-amd64 2.6 (smirom) 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, in production mode Linux/64-ia64 2.6 (cobalt) w/Intel compilers, w/C++ & FORTRAN, in production mode Linux/64-ia64 2.4 (tg-login3) w/parallel, w/FORTRAN, in production mode Linux/64-amd64 2.6 (abe) w/parallel, w/FORTRAN, in production mode Mac OS X/32 10.5.6 (amazon) in debug mode Mac OS X/32 10.5.6 (amazon) w/C++ & FORTRAN, w/threadsafe, in production mode
* [svn-r16625] Description:Quincey Koziol2009-03-301-1/+3
| | | | | | | | | | | | | | | | | | | | | | Add mechanism for querying if an entry in the metadata cache is the parent or child in a flush dependency relationship. 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/C++ & FORTRAN, w/threadsafe, in debug mode Linux/64-amd64 2.6 (smirom) 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, in production mode Linux/64-ia64 2.6 (cobalt) w/Intel compilers, w/C++ & FORTRAN, in production mode Linux/64-ia64 2.4 (tg-login3) w/parallel, w/FORTRAN, in production mode Linux/64-amd64 2.6 (abe) w/parallel, w/FORTRAN, in production mode Mac OS X/32 10.5.6 (amazon) in debug mode Mac OS X/32 10.5.6 (amazon) w/C++ & FORTRAN, w/threadsafe, in production mode
* [svn-r16618] Description:Quincey Koziol2009-03-261-1/+34
| | | | | | | | | | | | | | | | | | | | | | Modify metadata cache flush dependency feature to allow it to work with entries that are pinned through the cache API calls. 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/C++ & FORTRAN, w/threadsafe, in debug mode Linux/64-amd64 2.6 (smirom) 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, in production mode Linux/64-ia64 2.6 (cobalt) w/Intel compilers, w/C++ & FORTRAN, in production mode Linux/64-ia64 2.4 (tg-login3) w/parallel, w/FORTRAN, in production mode Linux/64-amd64 2.6 (abe) w/parallel, w/FORTRAN, in production mode Mac OS X/32 10.5.6 (amazon) in debug mode Mac OS X/32 10.5.6 (amazon) w/C++ & FORTRAN, w/threadsafe, in production mode
* [svn-r16600] Description:Quincey Koziol2009-03-241-1/+1
| | | | | | | | | | | | | | | | | | Ensure that metadata cache entries with flush dependencies are written to the file in the correct order (for serial I/O, at least). 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/C++ & FORTRAN, w/threadsafe, in debug mode Linux/64-ia64 2.6 (cobalt) w/Intel compilers, w/C++ & FORTRAN, in production mode Linux/64-ia64 2.4 (tg-login3) w/parallel, w/FORTRAN, in production mode Linux/64-amd64 2.6 (abe) w/parallel, w/FORTRAN, in production mode Mac OS X/32 10.5.6 (amazon) in debug mode Mac OS X/32 10.5.6 (amazon) w/C++ & FORTRAN, w/threadsafe, in production mode
* [svn-r16587] Description:Quincey Koziol2009-03-171-1/+21
| | | | | | | | | | | | | | | | | | | | Add infrastructure & tests for "flush dependencies" in metadata cache, which allow relationships that specify which order to flush metadata entries in. 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/C++ & FORTRAN, w/threadsafe, in debug mode Solaris/32 2.10 (linew) w/deprecated symbols disabled, w/C++ & FORTRAN, w/szip filter, in production mode Linux/64-ia64 2.6 (cobalt) w/Intel compilers, w/C++ & FORTRAN, in production mode Linux/64-ia64 2.4 (tg-login3) w/parallel, w/FORTRAN, in production mode Linux/64-amd64 2.6 (abe) w/parallel, w/FORTRAN, in production mode Mac OS X/32 10.5.6 (amazon) in debug mode Mac OS X/32 10.5.6 (amazon) w/C++ & FORTRAN, w/threadsafe, in production mode
* [svn-r15800] Description:Quincey Koziol2008-10-071-1/+16
| | | | | | | | | | | | | | | | | | | | | | | Bring file free space branch changes through r15795 into trunk, which includes a fair bit of code cleanup & rearrangement along with a couple of bug fixes also. Tested on: Mac OS X/32 10.5.5 (amazon) in debug mode Mac OS X/32 10.5.5 (amazon) w/C++ & FORTRAN, w/threadsafe, in production mode FreeBSD/32 6.3 (duty) in debug mode FreeBSD/64 6.3 (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/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, in production mode Linux/64-ia64 2.6 (cobalt) w/Intel compilers, w/C++ & FORTRAN, in production mode Linux/64-ia64 2.4 (tg-login3) w/parallel, w/FORTRAN, in production mode Linux/64-amd64 2.6 (abe) w/parallel, w/FORTRAN, in production mode
* [svn-r15628] Description:Quincey Koziol2008-09-161-29/+29
| | | | | | | | | | | | | Remove trailing whitespace from C/C++ source files, with the following script: foreach f (*.[ch] *.cpp) sed 's/[[:blank:]]*$//' $f > sed.out && mv sed.out $f end Tested on: Mac OS X/32 10.5.5 (amazon) No need for h5committest, just whitespace changes...
* [svn-r14445] Added code to detect the situation in which a metadata cache ↵John Mainzer2008-01-201-2/+31
| | | | | | | | | | | | | | entry flush callback functions modifies the skip list or LRU out from under a partial or complete flush of the cache. This is a test for a situation that should not occur, but we test anyway. Also added code to skip longer tests in cache_api in express tests. Tested serial and parallel on phoenix (debian --x86-32), and commit test. Elena commit tested as well, and ran a manual test under MacOS X.
* [svn-r14442] Modified metadata cache to adapt more quickly to the insertion of aJohn Mainzer2008-01-181-24/+81
| | | | | | | | | | | | | | | | | large entry, or to a large increases in the size of an existing entry. This required some additions to the cache configuration structure, and thus will require changes in the metadata cache documentation. The basic idea is to monitor the size of entries as they are loaded, inserted, or increased in size. If the size of the entry (or increase) exceeds some user selected fraction of the size of the cache, increase the size of the cache. Note that this fix was designed quickly -- while it deals with the use case that exposed the problem, we may have to revisit the issue later. Tested serial and parallel on Phoenix, and h5committest.
* [svn-r14065] Second try at checking in the "cork the cache" modification.John Mainzer2007-08-101-2/+2
| | | | | | | | First attempt contained slightly dated versions of the files listed below. Must also note that the mods were tested with h5committest, and two other linux builds.
* [svn-r14064] Implemented function allowing the user to cork the metadata cache.John Mainzer2007-08-101-2/+8
| | | | Also implemented associated test code.
* [svn-r13690] Description:Quincey Koziol2007-04-191-15/+0
| | | | | | | | | | | | | Fully enable readers/writer locking in metadata cache and correct errors in cache clients which were modifying a cache entry while only holding a read lock on it. Tested on: Mac OS X/32 10.4.9 (amazon) Linux/32 2.6 (chicago) Linux/64 2.6 (chicago2) FreeBSD/32 6.2 (duty) FreeBSD/64 6.2 (liberty)
* [svn-r13618] Modified metadata cache to support multiple read only protects of John Mainzer2007-04-091-1/+42
| | | | | | | | | | cache entries. Added test code to test the new feature. Also some minor cleanum h5committested -- passed on copper and sol. Failed on osage with a configuration error that appears unrelated to my changes. Serial test (debug mode) passes on Phoenix (Linux x86 2.6 kernel), so I went ahead with the checkin.
* [svn-r13253] Updated all C and C++ style source code files with the THG ↵Albert Cheng2007-02-071-2/+3
| | | | | | | | | copyright notice. 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.
* [svn-r12774] Modified flush code in the metadata cache to allow it toJohn Mainzer2006-10-181-2/+16
| | | | | | | | | | | | | | | | | handle flush callbacks which can dirty other entries, and resize and/or rename the target entry. This feature is needed by the fractal heap code. Also added associated test code. H5Commit tested. Test failed on heping, but the error appears to be a syntax error in an un-related file. Tests on copper & sol passed, along with tests on phoenix.
* [svn-r12595] Modified H5C_insert_entry() to accept the H5C__PIN_ENTRY_FLAG entryJohn Mainzer2006-08-171-12/+35
| | | | | | | | | | | flag, and pin an entry as it is inserted. The objective is to avoid some function call overhead in fheap. Also added matching test code in test/cache.c & test/cache_common.c (also testpar/t_cache.c checked in separately by accident) h5commit tested
* [svn-r12462] Committed a variety of metadata cache related changes:John Mainzer2006-07-131-0/+18
| | | | | | | | | | | | | | | | | | | | | | 1) Added trace file support to the metadata cache. This allows capture of all metadata cache calls in trace files for purposes of optimization and debuging. 2) Added an expunge entry function. This allows an entry to be deleteded from the cache without writing it to disk even if it is dirty. 3) Added a function call to resize pinned entries. 4) Added code to deal with entries that are dirty on load. This is needed in support of a bug fix which can alter object headers on load to repair files. 5) Added progress reporting code to the "MDC API smoke check" test in cache_api.c. To enable the progress reporting, set report_progress to TRUE in mdc_api_call_smoke_check(). Tested with h5committest, and a parallel test on phoenix (dual athelon linux box).
* [svn-r12440] Purpose:Quincey Koziol2006-06-271-19/+19
| | | | | | | | | | | 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
* [svn-r12374] Purpose:John Mainzer2006-05-241-0/+19
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | 1) Check in potential fix to unreproduceable bug in t_cache observed on Cobalt 2) Check in code supporting added pinned entry features in the metadata cache. Description: 1) Elena encountered a bug in t_cache when running on Cobalt. On investigation I found a potential race condition in t_cache that could explain her observation. 2) Quincey requested additions to the metadata cache allowing renaming of pinned entryies, and marking of pinned or protected entries as dirty. Solution: 1) Modified t_cache.c to prevent the potential race condition. Elena was unable to reproduce the bug on Cobalt, so we don't know if my fix actually addressed the problem. 2) Added code supporting the requested functions. The changes were relatively trivial, but required substantial new test code. Platforms tested: h5committest, serial test on heping, parallel test on phoenix. Misc. update:
* [svn-r12311] Purpose:John Mainzer2006-04-281-5/+54
| | | | | | | | | | | | | | | | | | | | | | | | | Add pinned entry capability to cache. Description: For frequently accessed cache entries, the protect/unprotect overhead is sometimes a bottleneck. Solution: Allow entries to be pinned in the cache. Pinned entries can't be evicted, but can be flushed or modified. Platforms tested: h5committested -- minus one small typo in test/cache.c whose fix was tested on copper and heping only. Misc. update:
* [svn-r11897] Purpose:John Mainzer2006-01-271-9/+16
| | | | | | | | | | | | | | | | | | | | | | | | | | | | 1) Add parallel test for metadata cache 2) Split serial test for metadata cache into two parts 3) Fix bug in which cache was flushed needlessly when the cache wasn't full. 4) Performance improvements 5) Update API for parallel cache coherency bug fix. Description: See above. Solution: See above. Platforms tested: h5committest Misc. update:
* [svn-r11470] Purpose:John Mainzer2005-09-271-9/+62
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Repair synchronization bug in the metadata cache in PHDF5 Also repair numerous other bugs that surfaced in testing the bug fix. Description: While operations modifying metadata must be collective, we allow independant reads. This allows metadata caches on different processes to adjust to different sizes, and to place the entries on their dirty lists in different orders. Since only process 0 actually writes metadata to disk (all other processes thought they did, but the writes were discarded on the theory that they had to be collective), this made it possible for another process to modify metadata, flush it, and then read it back in in its original form (pre-modification) form. The possibilities for file corruption should be obvious. Solution: Make the policy that only process 0 can write to file explicit, and visible to the metadata caches. Thus only process 0 may flush dirty entries -- all other caches must retain dirty entries until they are informed by process 0 that the entries are clean. Synchronization is handled by counting the bytes of dirty cache entries created, and then synching up between the caches whenever the sum exceeds an (eventually user specified) limit. Dirty metadata creation is consistent across all processes because all operations modifying metadata must be collective. This change uncovered may bugs which are repaired in this checkin. It also required modification of H5HL and H5O to allocate file space on insertion rather than on flush from cache. Platforms tested: H5committest, heping(parallel & serial) Misc. update:
* [svn-r11245] Purpose:Quincey Koziol2005-08-131-102/+102
| | | | | | | | | | | | | | | | | | | | 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
* [svn-r11014] Purpose:Quincey Koziol2005-07-021-5/+7
| | | | | | | | | | | | Code cleanup Description: Refactor metadata cache to merge "dirtied" flag in with other flags for H5AC_unprotect and H5C_unprotect. Platforms tested: FreeBSD 4.11 (sleipnir) h5committest
* [svn-r10978] Purpose:John Mainzer2005-06-241-0/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | Interim checkin of code changes moving management of the is_dirty flag into the cache code. Description: Prior to this checkin, management of the is_dirty flag was handled above the level of the metadata cache. This can no longer be allowed, as it introduces a race condition in the proposed fix for a cache coherency bug in PHDF5. Solution: Move management fo the is_dirty flag to the cache code proper. Entries are now marked as dirty via a flag on the unprotect call. Platforms tested: h5committested Misc. update:
* [svn-r10688] Purpose:John Mainzer2005-04-281-15/+19
| | | | | | | | | | | | | | | | | | | | | | | | | | Add API calls allowing user control of the metadata cache. Description: Prior to this update, the metadata cache was not configurable from outside the library. Solution: Add API calls allowing the user to configure the metadata cache either at file open time, or for any open file. Also added calls permitting the user to monitor cache size and hit rate. These latter facilities are needed for "manual" cache size control Platforms tested: h5committested Misc. update:
* [svn-r9850] Purpose:John Mainzer2005-01-201-6/+33
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | 1) Provide facilities in cache to allow us to avoid a potential cache consistency bug in the parallel case. 2) Clean up a off by one sanity checking bug. 3) Turn off execution of long running tests in debug mode. Description: 1) In the parallel case, all writes to metadata must be collective, but reads may not be. In pricipal, this allows us to different contents in different caches. This isn't a problem as long as the correct data is always on disk, but unless we can force certain writes immediately, that need not be the case. 2) & 3) should need no further explanation. Solution: 1) Add code allowing us to mark cache entries, and then force these entries to be flushed at a later time. Note that to actually avoid the bug, we will have to modify existing code to use these new features. 2) & 3) should need no further explanation. Platforms tested: heping (serial debug and production) committest (copper, sol, and heping). test failed on heping in the c++ portion of the build, but at Quincey's siggestion, I am proceeding with the checkin. Misc. update:
* [svn-r9727] Purpose:Quincey Koziol2004-12-291-5/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | 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:
* [svn-r9687] Purpose:John Mainzer2004-12-181-12/+343
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Modify the cache code (H5C) to support automatic cache resizing to adapt to the work load at run time. Description: Different applications require different sized caches to maintain an acceptable hit rate. This set of changes attempts to provide the ability to adjust to circumstances automatically. Solution: Added highly configurable code to allow the user to either set a fixed cache size, or allow the cache to grow and shrink according to conditions. If enabled, cache size increases are triggered when the hit rate drops below a user specified threshold in a user specified interval. Cache size reductions (if enabled) are triggered when either the hit rate exceeds some user specified threshold over a user specified interval, when the cache contains "enough" entries that haven't been accessed for a user specified interval, or some mix of the above. See the header comments on the H5C_auto_size_ctl_t structure in H5Cprivate.h for further details. At present, the cache resize configuration options are not accessible via the user API. Must add this. Platforms tested: h5committested, heping (serial), and copper (parallel) Misc. update:
* [svn-r9596] Purpose:Quincey Koziol2004-11-291-13/+7
| | | | | | | | | | | | | Code optimization Description: Retarget metadata cache code from using TBBT routines to using new skiplist routines, for a reasonable speedup (when dealing with dirty metadata) Platforms tested: FreeBSD 4.10 (sleipnir) w/parallel Solaris 2.7 (arabica) Too minor to require h5committest
* [svn-r9022] Purpose: Optimization of the cache code in H5C.John Mainzer2004-08-051-13/+59
| | | | | | | | | | | | | | | | | | | | | | | Description: Cache was running too slowly. Solution: Added a hash table for indexing. Retained the tree, but only for dirty entries. As we need to flush dirty entries in increasing address order, this is sufficient. Updated statistics collection code for the above. Converted a number of local functions into macros to avoid the function call overhead. Added code to disable the clean and dirty LRU lists in serial mode. Updated test code to account for the above changes. Platforms tested: h5committested + serial, parallel, and fp on Eirene. Misc. update:
* [svn-r8905] Purpose:MuQun Yang2004-07-201-1/+2
| | | | | | | | | | | | | | | | bug fix for windows testing Description: A function called H5C_stats_reset doesn't have H5_DLL in front of it, it cause windows DLL test failed. Solution: Add H5_DLL in front of it. Platforms tested: windows xp, sol 2.7, linux 2.4, aix 5.1 Misc. update:
* [svn-r8800] Purpose:Quincey Koziol2004-07-031-5/+5
| | | | | | | | | | | | | Code cleanup Description: Fix problems when compiling with C++ compiler. Also clean up some warnings with gcc 3.4.x Platforms tested: FreeBSD 4.10 (sleipnir) Too minor to require h5committest
* [svn-r8791] Purpose: Rewrote metadata cache (H5AC.c, etc.) to improve ↵John Mainzer2004-07-021-0/+374
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | performance. Description: Replaced the old metadata cache with a cache with a modified LRU replacement policy. This should improve the hit rate. Solution: Since we want to flush cache entries in increasing address order, I used the threaded binary B-tree code to store the cache entries. There is a fair bit of overhead here, so we may want to consider other options. While the code is designed to allow the support of other replacement algorithms, at present, only a modified version of LRU is supported. The modified LRU algorithm requires that a user selectable portion of the cache entries be clean. The clean entries are evicted first when writes are not permitted. If the pool of clean entries is used up, the cache grows beyond its user specified maximum size. The cache can also exceed its maximum size if the combined size of the protected (or locked) entries exceeds the maximum size of the cache. Platforms tested: eirene (serial, parallel, fp), h5committested Misc. update:
* [svn-r297] Switched templates to "property lists" and changed API prefix ↵Quincey Koziol1998-02-251-29/+0
| | | | from H5C to H5P
* [svn-r290] Changes since 19980206Robb Matzke1998-02-191-0/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | ---------------------- ./configure.in ./src/H5Fprivate.h ./src/H5Fsec2.c We now detect and use lseek64() on systems that have it (e.g., Irix64) and are able to generate >2GB files on Irix XFS file systems (and anything else that supports large files). This change also removed some warning messages from the Irix `-64' compiler. > $ ls -l istore.h5 > -rw-r--r-- 1 matzke meshtv 8605436856 Feb 17 14:03 istore.h5 ./configure.in ./src/H5Fprivate.h ./src/H5Fstdio.h We now detect and use fseek64() on systems that have it (e.g., Irix64) and are able to generate >2GB files on Irix XFS file systems (and anything else that supports large files). This change also removed some warning messages from the Iris `-64' compiler. ./src/H5E.c ./src/H5Epublic.h Added the H5E_OVERFLOW error to signal file address overflow. ./src/H5Fpublic.h ./examples/h5_chunk_read.c ./examples/h5_compound.c ./examples/h5_extend_write.c ./examples/h5_group.c ./examples/h5_read.c ./examples/h5_write.c ./html/Datasets.html ./html/Files.html ./html/H5.api.html ./html/H5.intro.html ./html/H5.sample_code.html ./html/ph5example.c ./html/review1.html ./test/cmpd_dset.c ./test/dsets.c ./test/extend.c ./test/tfile.c ./test/th5p.c ./test/theap.c ./test/tohdr.c ./test/tstab.c ./testpar/phdf5.c Renamed file access constants to follow the naming scheme. Also changed the base names a little to be more accurate as to what they do. The old names H5ACC_WRITE and H5ACC_OVERWRITE will temporarily work. H5ACC_DEFAULT --> H5F_ACC_RDONLY for H5Fopen() H5ACC_DEFAULT --> H5F_ACC_EXCL for H5Fcreate() H5ACC_WRITE --> H5F_ACC_RDWR H5ACC_OVERWRITE --> H5F_ACC_TRUNC Albert or Kim: The H5ACC_INDEPENDENT and H5ACC_COLLECTIVE macros in H5Fpublic.h should be an enum typedef and have names more like H5F_MPIO_INDEPENDENT and H5F_MPIO_COLLECTIVE. Also change the access_mode argument of H5Cset_mpio(). H5Fcreate() and H5Fopen() are more strict now about which flags are acceptable for the operation. ./src/H5Fprivate.h ./src/H5F.c ./src/H5C.c Changed the file access template to make it more general. A union contains a struct for each type of low-level driver and the default template is initialized at run-time. ./src/H5Fpublic.h ./src/H5F.c Added H5Fget_access_template() and cleaned up H5Fget_create_template(). ./src/H5C.c The H5Cset_mpi() no longer trashes the file access template when an error is detected. We check for errors and *then* update the file access template. Added H5C_close() so Albert and Kim have a place to release the MPI communicator and info from the file access property list. Kim or Albert: I notice in H5Cset_mpi() you copy the MPI communicator. Do you need to do something similar in H5C_copy()? ./src/H5F.c Added more error checking for the file creation and access property lists because it used to be possible to make the library dump core by swapping the creation and access property list ID numbers of H5Fcreate(). ./test/istore.c ./test/tfile.c ./test/th5p.c ./test/theap.c ./test/tohdr.c ./test/tstab.c ./testpar/testphdf5.c One must pass H5C_DEFAULT as the file creation or access property list in order to get the default property list. It is no longer possible to pass zero or any other arbitrary bad object ID. ./src/H5Fcore.c ./src/H5Ffamly.c ./src/H5Flow.c ./src/H5Fmpio.c ./src/H5Fsec2.c ./src/H5Fsplit.c ./src/H5Fstdio.c ./src/H5MF.c The file access property list is passed to all H5F_low_...() functions and to the drivers. ./src/H5Fcore.c The block size can be set at run time on a per-file basis instead of at compile time across all files. The "5000 items in a group test" now takes 1.6 seconds. ./src/H5private.h Removed inclusion of mpi.h and mpio.h since they're included from H5public.h. ./src/H5Cpublic.h ./src/H5C.c Added H5Cset_stdio(), H5Cset_sec2(), H5Cset_core(), H5Cset_split(), and H5Cset_family() in addition to the H5Cset_mpio() that Kim and Albert already wrote. We still need the H5Cget_driver() and an H5Cget...() counterpart for each of those functions. The split and family drivers still need a little work but I'm checking this in anyway.
* [svn-r214] Changes since 19980203Robb Matzke1998-02-031-0/+2
| | | | | | | | | | | ---------------------- ./src/H5C.c ./src/H5Cprivate.h ./src/H5D.c ./src/H5Dpublic.h Added H5Dget_create_parms(), trying to stay one step ahead of Elena ;-)
* [svn-r164] Changes since 19980122Robb Matzke1998-01-221-5/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | ---------------------- ./src/*.h Fixed indentation where indent(1) screwed up. This isn't by any means the final say, but it's better than it was. ./src/H5A.c ./src/H5Aprivate.h ./src/H5Apublic.h ./src/H5C.c ./src/H5D.c ./src/H5E.c ./src/H5F.c ./src/H5G.c ./src/H5M.c ./src/H5P.c ./src/H5T.c ./src/H5Tconv.c ./src/debug.c ./test/dtypes.c ./test/istore.c ./test/theap.c ./test/tohdr.c ./test/tstab.c Removed some atom functions from the API and made them library-scope. Also changed some names by removing the redundant `atom' from the name and by adding a `_' after the `H5A'.
* [svn-r157] Reformatted code with indent...Quincey Koziol1998-01-161-1/+1
|
* [svn-r142] ./src/H5C.cRobb Matzke1997-12-111-1/+0
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | ./src/H5Cprivate.h Changed H5Csetparm() to H5Cset_prop() and changed it to pass property values by value instead of reference. Changed H5Cgetparm() to H5Cget_prop(). Changed data types of file creation properties so the application can use `int' or `size_t' for most properties. Properties and their types are documented in H5Cget_prop(). Changed H5C_class() to H5Cget_class() to make the naming more uniform across packages. ./src/H5Cpublic.h Changed names of properties to make them more uniform across packages: H5F_USERBLOCK_SIZE became H5F_SIZEOF_USERBLOCK, H5F_OFFSET_SIZE became H5F_SIZEOF_ADDR, H5F_LENGTH_SIZE became H5F_SIZEOF_SIZE. ./src/H5D.c ./src/H5Dprivate.h Added support for chunked data storage. There still isn't any support for data type conversion or data space conversion, so chunked storage doesn't do much at this time. ./src/H5E.c ./src/H5Epublic.h Added the H5E_TEMPLATE major error number. I'm about the change the HGOTO_ERROR() and HRETURN_ERROR() macros to take an error string as an additional argument instead of specifying it as a comment. ./src/H5Fistore.c Added `const' to the `buf' argument of H5F_istore_write() although this temporarily causes a compiler warning. ./src/H5Fprivate.h Changed data types of certain file-related variables from types like `uint8' to `uintn' since the library data types should not depend on the file data types. ./src/H5P.c ./src/H5Pprivate.h ./src/H5Ppublic.h Moved H5Pget_lrank() to H5Pget_ndims(), and H5Pget_ldims() to H5Pget_dims() to make naming more uniform. Dimensions should always be of type `size_t'. Similarly for H5P_get_lrank() and H5P_get_ldims().
* [svn-r139] ./src/*.[ch]Robb Matzke1997-12-101-24/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Removed the interface initialization argument from FUNC_ENTER() and made it a locally-defined preprocessor symbol, INTERFACE_INIT. Changed `offset' to `address' and `length' to `size' in documentation so it's more consistent. `Offset' still appears occassionally when it refers to a byte offset within some other data structure. Moved interface termination function prototypes from public header files to .c files and made them static. ./src/H5.c ./src/H5public.h Added H5init() because it's possible that the predefined data types are not initialized. This happens only if the first call to the hdf5 library passes a predefined data type symbol as an argument. There should be some way to fix this... ./src/H5A.c ./src/H5Aprivate.h ./src/H5Apublic.h The free_func returns SUCCEED or FAIL, although the return value is ignored by H5A. This is so we can use the various H5*_close() functions to free things. H5Ainc_ref() and H5Adec_ref() are no longer public. Many of the other atom functions should also be made private, but I'll save that for later... Added additional template groups called H5_TEMPLATE_0 through H5_TEMPLATE_7 that are used by the various template subclasses. Increased the number of bits used for atom groups to prevent negative atoms. ./src/H5AC.c ./src/H5ACprivate.h Changed H5AC_new() to H5AC_create() to make names more consistent. ./src/H5B.c ./src/H5Bprivate.h Changed H5B_new() to H5B_create() to make names more consistent. ./src/H5C.c ./src/H5Cprivate.h ./src/H5Cpublic.h Now supports multiple subclasses of templates, although it's done with big switch statements. The default values for templates are defined in the source file to which that template belongs. This got rid of lots of needless preprocessor constants. Added H5Ccreate() to create a new template. Changed H5C_release() to H5Cclose() to make the naming more consistent. ./src/H5D.c ./src/H5Dprivate.h ./src/H5Dpublic.h Enhanced to use the new dataset interface, and uses the enhanced data type and data space interfaces, which haven't been completely implemented. The dataset interface doesn't handle non-contiguous storage, compression, or data type and space conversions yet. ./src/H5F.c ./src/H5Fprivate.h ./src/H5Fpublic.h Removed H5Fflush() since just calls H5F_flush(), which doesn't do what the user would probably think it does, namely, flush everything. It only flushes those things sitting in the H5AC cache and the boot block. Changed the `file_create_parms' field of H5F_low_t to just `create_parms' since the `file' part is obvious. ./src/H5Fistore.c Added some support for external files. Mostly just in the file format and not supported much by the library yet. I need to finish some dataset functions first. Changed H5F_istore_new() to H5F_istore_create() to make names more uniform across packages. ./src/H5Flow.c Flushing a file causes the file to be physically extended to the logical eof. This prevents H5F_open() from thinking a file has been truncated. Most of the time the file will already be that large, and when it isn't Unix will often just allocate the final block anyway. ./src/H5G.c ./src/H5Gent.c ./src/H5Gnode.c ./src/H5Gpkg.h ./src/H5Gprivate.h ./src/H5Gstab.c Removed H5G_basename() Removed (temporarily) data type information from symbol table entries and renamed H5G_CACHED_SDATA to H5G_CACHED_SDSPACE to reflect that it's a simple data space and has nothing to do with raw data. Changed H5G_node_new() to H5G_node_create() and H5G_stab_new() to H5G_stab_create() to make names more uniform across packages. Fixed an undefined address bug that happens when H5G_node_debug() program doesn't pass enough info to H5G_node_load(). ./src/H5H.c ./src/H5Hprivate.h Changed H5H_new() to H5H_create() to make the names more uniform across packages. ./src/H5M.c ./src/H5Mprivate.h ./src/H5Mpublic.h Nulled all the create functions. Most of the other callbacks are to public functions. Removed H5Mcreate(). Changed hobjtype_t to group_t since it has to be the same thing anyway. ./src/H5O.c ./src/H5Oprivate.h ./src/H5Osdim.c ./src/H5Osdtyp.c Changed H5O_SIM_DIM to H5O_SDSPACE (simple data space) since `simple data space' is its official name, not `simple dimensions'. Will eventually add H5O_CDSPACE for comples data spaces. Changed _sim_dim_ to _dspace_. Replaced H5O_SIM_DTYPE and the compound data type messages with a single H5O_DTYPE message. Changed _sim_dtype_ to _dtype_. Changed H5O_STD_STORE to H5O_CSTORE (contiguous storage) since contiguous storage is not necessarily standard. Changed _std_store_ to _cstore_ in H5Ocstore.c Added the H5O_EFL (external file list) message. Changed H5O_new() to H5O_create() to make names more uniform across packages. ./src/H5Oefl.c NEW External file list message for specifying which non-hdf5 files contain raw data for a dataset. ./src/H5P.c ./src/H5Pprivate.h ./src/H5Ppublic.h Renamed and moved data structures to make the names conform to our naming scheme. ./src/H5T.c ./src/H5Tprivate.h ./src/H5Tpublic.h ./src/H5Tpkg.h NEW Data structures redesigned to be more flexible. The interface was redesigned to make it more regular and to make some names more uniform across packages. ./src/H5detect.c Output was changed to produce a file that conforms to the hdf5 coding standard. ./src/Makefile.in Generates H5Tinit.c by running H5detect. ./src/debug.c Moved command argument processing.
* [svn-r129] Changes since 19970916Robb Matzke1997-10-201-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | ---------------------- ./config/depend.in Fixed backslashes in sed script because the H5Gnode.c dependency info was disappearing. You'll have to rerun config.status to rebuild the Makefiles unless you use gnu make. ./config/conclude.in Also removes emacs backup files, TAGS, and svf backup files. ./config/linux Grouped gcc flags and added provisions for debugging vs. production. ./html/H5.format.html Updated messages 0x0008, 0x0009, and 0x000A. ./html/storage.html Documentation describing storage schemes. ./src/Makefile.in ./test/Makefile.in New source files. ./src/H5A.c ./src/H5Apublic.h ./src/H5C.c Changed VOIDP to void* in a couple places. ./src/H5AC.c ./src/H5ACprivate.h ./src/H5B.c ./src/H5Bprivate.h ./src/H5G.c ./src/H5Gnode.c ./src/H5Gprivate.h ./src/H5H.c ./src/H5O.c Removed `const' from some variables because H5G_node_found() wanted to modify it's udata argument. Removing const there caused it to cascade to these other locations. ./src/H5AC.c ./src/H5ACprivate.h ./src/H5B.c ./src/H5Gnode.c ./src/H5Gstab.c ./src/H5H.c ./src/H5O.c Added an extra argument to H5AC_find_f() and H5AC_protect(). This arg gets passed to the load() method. Also added an extra argument to the H5AC_find() macro. ./src/H5B.c ./src/H5Bprivate.h ./src/H5Gnode.c Extra argument passed to the sizeof_rkey() method. ./src/H5Fprivate.c ./src/H5Fistore.c (new) Added indexed I/O operations. ./src/H5G.c ./src/H5Gnode.c ./src/H5Gprivate.h Beginning to add H5G_open/close and related bug fixes. ./src/H5Oprivate.h ./src/H5Oistore.c (new) Added the H5O_ISTORE messsage (0x0008) for indexed storage of objects. ./src/H5private.h Added extra braces around both sides of the FUNC_ENTER() and FUNC_LEAVE() macros so FUNC_ENTER() can appear before declarations or after executable statements the second case is used by H5G_namei() to initialize output arguments to sane values before FUNC_ENTER() might return failure. int f () { int decl1; printf ("This happens before FUNC_ENTER()\n"); FUNC_ENTER (...); int another_declaration; ./src/H5B.c ./src/H5Bprivate.h ./src/H5Gnode.c Extra arguments for key encoding and decoding. ./src/H5E.c ./src/H5Epublic.h ./src/H5Fistore.c ./src/H5Oistore.c ./src/H5Oprivate.h Indexed, chunked, sparse storage (not ready for general consumption yet). ./src/H5V.c (new) ./src/H5Vprivate.h (new) ./test/hyperslab.c (new) Vector, array, and hyperslab functions. ./src/H5B.c ./src/H5Bprivate.h ./src/H5Fistore.c ./src/H5Gnode.c ./src/H5V.c ./src/H5Vprivate.h ./test/hyperslab.c Added functionality for indexed storage. ./src/H5F.c Fixed problems with seek optimizing. Recommend we disable it until we can implement it in the file/address class since all of HDF5 must be aware of it. ./src/H5O.c Fixed comeent speling erorr :-) ./MANIFEST Added new files. ./config/conclude.in Added the word `Testing' to the test cases. So if a test program is called hyperslab then the make output will contain the line `Testing hyperslab'. ./config/linux The default file I/O library is Posix section 2 on my linux machine so I can do some I/O performance testing. ./src/H5C.c ./src/H5Cprivate.h ./src/H5Cpublic.h Added ability to set size of indexed-storage B-tree. ./src/H5D.c ./src/H5E.c ./src/H5Epublic.h ./src/H5F.c ./src/H5Fprivate.h ./src/H5G.c ./src/H5Gnode.c ./src/H5Gpkg.h ./src/H5Gprivate.h ./src/H5Gpublic.h ./src/H5Gshad.c ./src/H5Gstab.c ./test/stab.c Changed `directory' to `group' in numerous places. ./src/H5private.h The FILELIB constant can be set on the compile command-line. ./src/istore.c NEW Tests for indexed storage.
* [svn-r108] Changed all hatom_t types to hid_t. (Isn't 'sed' nifty.. :-)Quincey Koziol1997-09-241-5/+5
|
* [svn-r35] ./src/H5ACproto.hRobb Matzke1997-08-151-4/+31
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | ./src/H5Aproto.h ./src/H5Bproto.h ./src/H5Cproto.h ./src/H5Dproto.h ./src/H5Eproto.h ./src/H5Fproto.h ./src/H5Gproto.h ./src/H5Hproto.h ./src/H5MFproto.h ./src/H5MMproto.h ./src/H5Mproto.h ./src/H5Oproto.h ./src/H5Pproto.h ./src/H5Tproto.h ./src/H5proto.h These files were removed from the library and renamed by changing `proto' to `public'. ./src/H5ACpublic.h NEW ./src/H5Apublic.h NEW ./src/H5Bpublic.h NEW ./src/H5Cpublic.h NEW ./src/H5Dpublic.h NEW ./src/H5Epublic.h NEW ./src/H5Fpublic.h NEW ./src/H5Gpublic.h NEW ./src/H5Hpublic.h NEW ./src/H5MFpublic.h NEW ./src/H5MMpublic.h NEW ./src/H5Mpublic.h NEW ./src/H5Opublic.h NEW ./src/H5Ppublic.h NEW ./src/H5Tpublic.h NEW ./src/H5public.h NEW These files came from the old H5*proto.h files. ./src/Makefile ./src/Makefile.in NEW Removed. Now generated automatically from Makefile.in by running configure. ./src/h5oplat.h ./src/hdf5fort.h ./src/hdf5gen.h ./src/hdf5glob.h ./src/hdf5lims.h ./src/hdf5meta.h ./src/hdf5pabl.h ./src/hdf5plat.h ./src/hdf5port.h ./src/hdf5type.h Removed. The contents of these files has moved to other header files or source files depending on it's nature. ./src/H5.c ./src/H5A.c ./src/H5B.c ./src/H5C.c ./src/H5D.c ./src/H5E.c ./src/H5F.c ./src/H5G.c ./src/H5H.c ./src/H5M.c ./src/H5MF.c ./src/H5MM.c ./src/H5O.c ./src/H5Ocont.c ./src/H5Oname.c ./src/H5Onull.c ./src/H5Ostab.c ./src/H5P.c ./src/H5T.c Fixed include files. Moved some things from old headers into these files. ./src/H5ACprivate.h ./src/H5Aprivate.h ./src/H5Bprivate.h ./src/H5Cprivate.h ./src/H5Dprivate.h ./src/H5Eprivate.h ./src/H5Fprivate.h ./src/H5Gprivate.h ./src/H5Hprivate.h ./src/H5MFprivate.h ./src/H5MMprivate.h ./src/H5Mprivate.h ./src/H5Oprivate.h ./src/H5Pprivate.h ./src/H5Tprivate.h ./src/H5private.h ./src/debug.c Fixed include files. ./src/hdf5.h This is now the top-level *PUBLIC* include file. It should never appear in the library *.c files.
* [svn-r2] Oops...Quincey Koziol1997-07-301-0/+25
generated by cgit v0.12 at 2024-11-01 17:27:35 (GMT)