summaryrefslogtreecommitdiffstats
path: root/src/H5Fprivate.h
Commit message (Collapse)AuthorAgeFilesLines
* [svn-r21136] Description:Quincey Koziol2011-07-211-35/+72
| | | | | | | | | | Bring r21135 from trunk to 1.8 branch: More code cleanups to reduce coupling with the H5F package. Tested on: FreeBSD/32 8.2 (loyalty) w/gcc4.6, w/C++ & FORTRAN, in debug mode (h5committested on trunk)
* [svn-r21134] Description:Quincey Koziol2011-07-201-0/+12
| | | | | | | | | | | | | | | | | | Bring r21133 from trunk to 1.8 branch: More code cleanups to reduce coupling between packages that use the H5F internal routines, but really aren't part of the H5F "package". Tested on: FreeBSD/32 8.2 (loyalty) w/gcc4.6, w/C++ & FORTRAN, in debug mode FreeBSD/64 8.2 (freedom) w/gcc4.6, w/C++ & FORTRAN, in debug mode Linux/32 2.6 (jam) w/PGI compilers, w/default API=1.8.x, w/C++ & FORTRAN, w/threadsafe, in debug mode Linux/64-amd64 2.6 (koala) w/Intel compilers, w/default API=1.6.x, w/C++ & FORTRAN, in production mode Solaris/32 2.10 (linew) w/deprecated symbols disabled, w/C++ & FORTRAN, w/szip filter, w/threadsafe, in production mode Linux/PPC 2.6 (heiwa) w/C++ & FORTRAN, w/threadsafe, in debug mode
* [svn-r21130] Description:Quincey Koziol2011-07-201-8/+19
| | | | | | | | | | | | | | | | Bring r21126 from trunk to 1.8 branch: Clean up some compiler warnings, but mainly an interim checkin on the path of reducing code coupling in the library by removing most of the places which were accessing H5F package definitions. Tested on: Mac OS X/32 10.6.8 (amazon) w/debug FreeBSD/32 8.2 (loyalty) w/debug FreeBSD/64 8.2 (freedom) w/debug Linux/32 2.6 (jam) w/debug Linux/64 2.6 (koala) w/debug Linux/64 2.6 (heiwa) w/debug
* [svn-r20114] Merge r20112 (open_file_pool) into 1_8 branch.Neil Fortner2011-02-171-0/+6
| | | | Tested: jam (open_file_pool branch tested with h5committest)
* [svn-r18032] Description:Quincey Koziol2009-12-171-7/+5
| | | | | | | | | | | | | | | | | Bring r18031 from trunk to 1.8 branch: Bring r18030 from merge_metadata_journaling branch to trunk: Bring "brush clearing" changes (whitespace & style issues, mostly) from metadata_journaling branch to the "merging" branch, to converge the trunk and the metadata_journaling branch. Also, some other minor cleanups along the way. Tested on: FreeBSD/32 6.3 (duty) in debug mode FreeBSD/64 6.3 (liberty) w/C++ & FORTRAN, in debug mode (h5committested on trunk)
* [svn-r17972] Description:Quincey Koziol2009-12-071-2/+1
| | | | | | | | | | | | Bring r17971 from trunk to 1.8 branch: Allow the core VFD to properly support opening backing store files through symbolic links and have the external links in the file be treated in the same way as for the sec2 driver. Tested on: FreeBSD/32 6.3 (duty) in debug mode (h5committested on trunk)
* [svn-r17944] Description:Quincey Koziol2009-12-011-2/+6
| | | | | | | | | | | Bring r17943 from trunk to 1.8 branch: Finish refactoring v2 B-trees so that they can have client callback context provided to the encode/decode callbacks. Tested on: FreeBSD/64 6.3 (liberty) (h5committested on trunk)
* [svn-r17941] Description:Quincey Koziol2009-12-011-3/+10
| | | | | | | | | | | | Bring r17939 from trunk to 1.8 branch: Handle external links from symlinked files by adding another check to look for "child" files for links from the actual location of the "parent" file, instead of from the location of the symlink. Tested on: FreeBSD/32 6.3 (duty) (h5committested on trunk)
* [svn-r17886] Description:Quincey Koziol2009-11-131-2/+2
| | | | | | | | | | | | | | | | | Bring r17885 from trunk to 1.8 branch: Convert 'huge' fractal heap object management code to use refactored v2 B-tree routines. Also, switch internal representation of 'sizeof_addr' and 'sizeof_size' in files to be uint8_t's instead of size_t's, since they are stored in single bytes in the file format. Various other minor compiler warning cleanups and error call tweaks. Tested on: FreeBSD/32 6.3 (duty) in debug mode (h5committested on trunk
* [svn-r17404] Description:Quincey Koziol2009-08-211-9/+1
| | | | | | | | | | | | | | | | | | | | | | Bring r17402 & r17403 from trunk to 1.8 branch: r17402: Switch from using a 'flags' parameter to the flush routine(s), to just using a single 'closing' boolean parameter, since that's all we're doing with the flags anyway and this makes things more obvious. r17403: Eliminate 'closing' flag to dataset flush routine, since all cached dataset information has already been flushed when a file is closed and the datasets themselves will be closed (and the dataset caches destroyed). Skip calling more flush routines when the file is closing, in preparation for eventual separation of the 'flush' concept from the 'shutdown the cache' concept. Tested on: FreeBSD/32 6.3 (duty) in debug mode (h5committest performed on trunk)
* [svn-r17401] Description:Quincey Koziol2009-08-211-2/+1
| | | | | | | | | | | | | | | Bring r17400 from trunk to 1.8 branch: Refactor H5F_flush into two routines, one which handles the recursion/ iteration necessary to handle mounted files; and one which just handles the flush operation for a single file. Also, remove H5F_FLUSH_INVALIDATE and switch all code to depend on H5F_FLUSH_CLOSING. Tested on: FreeBSD/32 6.3 (duty) in debug mode (h5committested on trunk)
* [svn-r17374] Description:Quincey Koziol2009-08-181-3/+6
| | | | | | | | | | | Bring r17365 from trunk to 1.8 branch: Final merge of changes from sblock_mdc branch back to trunk. The superblock is now managed by the metadata cache. Tested on: FreeBSD/32 6.3 (duty) in debug mode (h5committest performed on trunk)
* [svn-r17314] Description:Quincey Koziol2009-08-101-0/+1
| | | | | | | | | | | | | | | | | | | | | | Bring r17313 from trunk to 1.8 branch: Bring back various minor code cleanups from the file free space branch 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-r17139] Description:Quincey Koziol2009-07-011-0/+2
| | | | | | | | | | | | Bring r17138 from trunk into 1.8 branch: Disable use of temporary file space allocation when using a parallel VFD, until we've made changes to broadcast the new address of the metadata when it is relocated down into 'normal' file space. Tested on: Mac OS X/32 10.5.8 (amazon) (h5committested on trunk)
* [svn-r17083] Description:Quincey Koziol2009-06-181-0/+3
| | | | | | | | | | | | | | | | | | | | | | | Bring r17010,17017,17021-17022,17029-17031,17035 back from trunk: "temporary" file allocation space changes. 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 10.5.7 (amazon) in debug mode Mac OS X/32 10.5.7 (amazon) w/C++ & FORTRAN, w/threadsafe, in production mode
* [svn-r16368] Description:Quincey Koziol2009-01-271-1/+4
| | | | | | | | | | | | Bring r16367 back from trunk: Refactor internal address encode/decode routines slightly, to allow for more flexible use. Tested on: Mac OS X/32 (amazon) FreeBSD/32 (duty) (too minor to require h5committest)
* [svn-r16282] Purpose: Add functions to allow more flexible traversal of ↵Neil Fortner2009-01-081-0/+3
| | | | | | | | | | | external links. Description: Added H5Pset/get_elink_cb to allow the user to specify a callback function to be called whenever an external link is traversed. Added H5Pset/get_elink_acc_flags to allow the user to specify the file access flags to use to open the target file of an external link. All these properties are set on a LAPL. Tested: jam (trunk version tested with h5committest)
* [svn-r16183] Description:Quincey Koziol2008-12-111-1/+1
| | | | | | | | | | Bring r16182 back from trunk: Rename internal routines, variables, macros, typedefs, etc. for chunked dataset storage from "istore" to some variant of "chunk" or "btree". Tested on: FreeBSD/32 6.3 (duty) in debug mode
* [svn-r16087] Purpose: Port H5Pset_chunk_cache, H5Pget_chunk_cache, and ↵Neil Fortner2008-11-171-4/+4
| | | | | | | | H5Dget_access_plist functions to 1.8 branch Tested: kagiso, jam (trunk version tested with h5committest)
* [svn-r15801] Description:Quincey Koziol2008-10-071-40/+40
| | | | | | | | | | | | | | | | | | | | | | | Bring r15800 back from trunk: - File free space branch changes through r15794 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-r15797] Purpose: Close bug #1322Neil Fortner2008-10-071-2/+2
| | | | | | | | | Description: Fixes a possible datatype id leak that could occur during compound datatype conversion, or more precisely, when unregistering those conversions. Datatype ids normally registered by the library are no longer visible to the application via H5Fget_obj_ids and H5Fget_obj_count. Tested: kagiso, linew, smirom (h5committest)
* [svn-r15707] I changed the return values of H5Fget_obj_ids and ↵Raymond Lu2008-09-261-2/+2
| | | | | | | | H5Fget_obj_count to ssize_t and modified C++ and Fortran API functions. This is for bug #1245. Tested on kagiso - I've tested the same change using h5committest.
* [svn-r15677] Description:Quincey Koziol2008-09-221-0/+46
| | | | | | | | | | | | Bring r15676 back from trunk: Centralize all macros for declaring "magic numbers"/signatures for objects in the file into src/H5Fprivate.h, so it's easier to know what values have already been defined, etc. Tested on: Mac OS X/32 10.5.5 (amazon) Too minor to require h5committest
* [svn-r15486] Purpose: Allow library to shut down properly when objects have ↵Neil Fortner2008-08-191-2/+2
| | | | | | | | | | | | | | | | reference count > 1. Description: Added a new field 'app_count' to H5I_id_info_t struct, to track the reference count on an id due to the application. the old 'count' field tracks the total. Generally any id visible to the application gets placed in app_count. Added app_ref boolean parameter to H5I_inc_ref, H5I_dec_ref, H5I_register, H5I_clear_type, and a few other functions, to specify whether the operation(s) being performed on the id(s) are due to the application (TRUE) or not (FALSE). Test added for this case. Tested: kagiso, smirom, linew (h5committest)
* [svn-r15212] Description:Quincey Koziol2008-06-131-0/+2
| | | | | | | | | | | Bring back revision 15211 from trunk: Update the gcc flags for version 4.3 Clean up warnings Tested on: Mac OS X/32 10.5.3 (amazon)
* [svn-r14791] Add handling for external link:Vailin Choi2008-04-021-0/+3
| | | | | | | 1. src/H5F.c, src/H5system.c: formulate path for mainfile 2. src/H5Lexternal.c: add search for target file 3. configure.in: add H5_HAVE_WINDOW_PATH 4. test/links.c: add tests for external link
* [svn-r14284] Description:Quincey Koziol2007-11-241-0/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Add H5Lvisit_by_name() API routine to library. Eliminated all (five!) other group traversal routines and changed them all to use the new API routine. Cleaned up output of h5ls & h5stat: - Issue error when requesting recursive traversal of a file with the "group info" flag, but no group given - Print info about root group in all(?) appropriate situations - Don't print "verbose" information about root group until the root group is in the list of objects to display (mostly because h5ls & h5stat had a different twist on traversing the groups in a file that the other utilities) Tested on: FreeBSD/32 6.2 (duty) in debug mode FreeBSD/64 6.2 (liberty) w/C++ & FORTRAN, in debug mode Linux/32 2.6 (kagiso) w/PGI compilers, w/C++ & FORTRAN, w/threadsafe, in debug mode Linux/64-amd64 2.6 (smirom) w/default API=1.6.x, w/C++ & FORTRAN, in production mode Linux/64-ia64 2.6 (cobalt) w/Intel compilers, w/C++ & FORTRAN, in production mode Solaris/32 2.10 (linew) w/deprecated symbols disabled, w/C++ & FORTRAN, w/szip filter, in production mode Mac OS X/32 10.4.10 (amazon) in debug mode Linux/64-ia64 2.4 (tg-login3) w/parallel, w/FORTRAN, in production mode
* [svn-r13745] Description:Quincey Koziol2007-05-101-1/+1
| | | | | | | | | | | | Take further advantage of the readers/writer locking in the metadata cache and push the locking of the local heap up to the routines which call B-tree routines, which should drastically lower the number of metadata cache protect/ unprotect calls. Tested on: Mac OS X/32 10.4.9 (amazon) Linux/32 2.6 (chicago) Linux/64 2.6 (chicago2)
* [svn-r13477] Description:Quincey Koziol2007-03-081-0/+3
| | | | | | | | | | | | | | | | | | | | The main purpose of this checkin was to eliminate the space used for tracking creation time indices when there is no way they can be used (i.e. attributes can't be shared in the file and the user hasn't turned on attribute creation tracking), however there were some other minor changes which crept in: - Fix a cache locking deadlock when a shared attribute and one of its components end up in the same fractal heap direct block. (This is fixed the "slow" way for right now, until John has time to add support for readers/writer locking to the cache. - Optimize attribute copying when a copy will be kept during a v2 B-tree search. - When freeing a block on disk, attempt to merge it with the metadata and "small data" aggregators. Tested on: Mac OS X/32 10.4.8 (amazon) FreeBSD/32 6.2 (duty)
* [svn-r13465] Description:Quincey Koziol2007-03-061-1/+7
| | | | | | | | | | | | Fix bug in external links when "strong" file close degree is used. Make external link callbacks use standard error reporting mechanisms. Other minor code cleanups... Tested on: Mac OS X/32 10.4.8 (amazon) FreeBSD/32 6.2 (duty)
* [svn-r13427] Description:Quincey Koziol2007-02-271-14/+11
| | | | | | | Eliminate some unnecessary properties for unused aspects of superblock. Tested on: Mac OS X/32 10.4.8 (amazon)
* [svn-r13413] Description:Quincey Koziol2007-02-271-0/+20
| | | | | | | | | | | | | Further cleanups to superblock code: - Move most of the ISOHM-specific code into the H5SM module - Make only one checksum, for combined superblock & driver info block Minor other cleanups, etc. Tested on: Mac OS X/32 10.4.8 (amazon) FreeBSD/32 6.2 (duty)
* [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-r13015] Description:Quincey Koziol2006-12-041-0/+2
| | | | | | | | | | | Migrate more object header routines to use the H5O_msg_ prefix and put them into the src/H5Omessage.c code module. Tested on: Mac OS X/32 10.4.8 (amazon) FreeBSD/32 4.11 (sleipnir) Linux/32 2.4 (heping) AIX/32 5.? (copper)
* [svn-r12935] Added list-to-btree conversion, changed SOHM apis, and pushed ↵James Laird2006-11-171-4/+5
| | | | | | | | | | | SOHM table version and size information into the superblock to eliminate a read when loading it. This is a file format change, and hopefully the last one (knock on wood). Tested on kagiso and Windows (mostly just a SOHM change).
* [svn-r12902] Checkin of Shared Object Header Message work.James Laird2006-11-131-0/+6
| | | | | | | | | | | | This feature is still in progress; Shared Object Header Messages are not complete as a feature and are not thoroughly tested. There are still "TODO" comments in the code (comments with the word "JAMES" in them, so as not to be confused with other TODO comments). Hopefully this checkin will reduce the liklihood of conflicts as I finish implementing this feature. All current tests pass on juniper, copper (parallel), heping, kagiso, and mir.
* [svn-r12866] Description:Quincey Koziol2006-11-061-6/+2
| | | | | | | Clean up initialization of H5F & H5L APIs. Tested on: Linux/64 2.6 (chicago2)
* [svn-r12842] Description:Quincey Koziol2006-11-021-142/+39
| | | | | | | | | | | | | | | | Refactor generic property list initialization code to put property list specific routines in property list modules, instead of scattered to the four winds. Also, introduce property list class initialization objects, to make adding new property list classes in the library easier. Fix daily test failure by using H5Pget_elink_prefix() API routine instead of looking at the "raw" generic property list information. Tested on: Mac OS X/32 10.4.8 (amazon) FreeBSD/32 4.11 (sleipnir) w/threadsafe Linux/32 2.4 (heping) w/C++ & FORTRAN Linux/64 2.4 (mir) w/build-all & 1.6 compat
* [svn-r12739] Added Direct I/O driver to VFD. It's only supported by SGI ↵Raymond Lu2006-10-101-1/+1
| | | | | | | | | | Altix (cobalt). There's a configure option --enable-direct-vfd/--disable-direct-vfd to enable/disable Direct I/O support. The default is enabled. There's a small test in test/vfd.c. Another way to test it is to set environment variable HDF5_DRIVER to "direct" and run "make check" in the test/ directory. There'll be some further improvement in the following checkin including allowing user to provide memory boundary value, file block size, and copying buffer size.
* [svn-r12708] Description:Quincey Koziol2006-10-031-0/+7
| | | | | | | | | | | | | | | | | | | Add the "use the latest version of the format" support to datatype messages. And a regression test to check that it's working. Also, found that we have been over-allocating space for compound datatype messages (when they have array datatypes or can use the latest version of the format) and trimmed the size back. Clean up datatype & dataspace encode/decode routines by having them allocate & release "fake" file structures, which gets them out of needing to "know" about the internals of file structures. Other minor whitespace/formatting cleanups, etc. Tested on: Linux/32 2.6 (chicago) Linux/64 2.6 (chicago2)
* [svn-r12656] Description:Quincey Koziol2006-09-111-0/+5
| | | | | | | | | | | | | | | Add "use the latest version of the file format" flag to the file access property list and internal file data structures. Fix bug where metadata block size was retrieved instead of the small data block size. Categorize property list routine prototypes in the public header file. Tested on: Mac OS/PPC 10.4 (amazon) Linux/32 2.6 (chicago) Linux/64 2.6 (chicago2)
* [svn-r12362] Purpose:Quincey Koziol2006-05-221-0/+4
| | | | | | | | | | | | | | | | | | | | | Code checkpoint Description: Update fractal heap code to allow objects to be deleted Fix bugs in fractal heap object insertion Improve free space manager code to allow sections to be merged and shrink the container Another try at making the Windows compilers happy... Platforms tested: FreeBSD 4.11 (sleipnir) Mac OS X.4/PPC (amazon) Linux 2.4 (chicago & heping) Solaris 2.9 (shanti) AIX 5.? (copper) w/parallel
* [svn-r12292] Purpose:Quincey Koziol2006-04-201-1/+0
| | | | | | | | | | | | | Code maintenance Description: Remove flexible parallel code Platforms tested: FreeBSD 4.11 (sleipnir) Linux 2.4 (heping) Solaris 2.9 (shanti) Linux 2.4/64 (mir)
* [svn-r12080] Purpose:Quincey Koziol2006-03-111-0/+30
| | | | | | | | | | | | New feature Description: Add "variable-length" 32-bit encode/decode macros, matching the other 64-bit ones, for later use in the fractal heap code. Platforms tested: FreeBSD 4.11 (sleipnir) Too minor for full h5committest
* [svn-r12004] Purpose:Quincey Koziol2006-03-041-42/+45
| | | | | | | | | | | | | | | | | New feature & code cleanup Description: Update fractal heap to be able to insert objects into a direct block hanging off the header. Extract "octal dump" code into separate routine so that both the local heaps & fractal heaps can use it. Other code cleanups & support to get this far. Platforms tested: FreeBSD 4.11 (sleipnir) Linux 2.4
* [svn-r11850] Purpose:James Laird2006-01-021-0/+4
| | | | | | | | | | | | | | Feature Description: Added character encoding and attribute creation property lists. Solution: Attributes' character encoding is set via the ACPL. The default is ASCII, with UTF-8 being the other option currently. Platforms tested: heping, mir, sleipnir, copper
* [svn-r11712] Purpose:Quincey Koziol2005-11-151-2/+5
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | New feature Description: Check in baseline for compact group revisions, which radically revises the source code for managing groups and object headers. WARNING!!!! WARNING!!!! WARNING!!!! WARNING!!!! WARNING!!!! WARNING!!!! WARNING!!!! WARNING!!!! WARNING!!!! WARNING!!!! WARNING!!!! WARNING!!!! This initiates the "unstable" phase of the 1.7.x branch, leading up to the 1.8.0 release. Please test this code, but do _NOT_ keep files created with it - the format will change again before the release and you will not be able to read your old files!!! WARNING!!!! WARNING!!!! WARNING!!!! WARNING!!!! WARNING!!!! WARNING!!!! WARNING!!!! WARNING!!!! WARNING!!!! WARNING!!!! WARNING!!!! WARNING!!!! Solution: There's too many changes to really describe them all, but some of them include: - Stop abusing the H5G_entry_t structure and split it into two separate structures for non-symbol table node use within the library: H5O_loc_t for object locations in a file and H5G_name_t to store the path to an opened object. H5G_entry_t is now only used for storing symbol table entries on disk. - Retire H5G_namei() in favor of a more general mechanism for traversing group paths and issuing callbacks on objects located. This gets us out of the business of hacking H5G_namei() for new features, generally. - Revised H5O* routines to take a H5O_loc_t instead of H5G_entry_t - Lots more... Platforms tested: h5committested and maybe another dozen configurations.... :-)
* [svn-r11470] Purpose:John Mainzer2005-09-271-0/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | 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-r11307] Purpose:Quincey Koziol2005-08-291-4/+4
| | | | | | | | | | | | Code cleanup Description: Clean up internals of group creation & iteration code. Platforms tested: FreeBSD 4.11 (sleipnir) Mac OS X (nile) Too minor to require h5committest
* [svn-r11245] Purpose:Quincey Koziol2005-08-131-11/+11
| | | | | | | | | | | | | | | | | | | | 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
generated by cgit v0.12 at 2024-11-09 20:36:16 (GMT)