summaryrefslogtreecommitdiffstats
path: root/src
Commit message (Collapse)AuthorAgeFilesLines
* [svn-r13745] Description:Quincey Koziol2007-05-1014-564/+370
| | | | | | | | | | | | 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-r13744] Description:Quincey Koziol2007-05-102-115/+68
| | | | | | | | | | | | | | | Initial changes to take advantage of readers/writer locking on metadata cache entries. Reduce the # of protect/unprotect pairs during B-tree iteration by holding reader lock on B-tree and symbol table nodes while iteration occurs. This also has the benefit of preventing an iteration callback from modifying the B-tree being iterated over (which would probably confuse the B-tree iteration code terribly) Tested on: Mac OS X/32 10.4.9 (amazon) Linux/32 2.6 (chicago) Linux/64 2.6 (chicago2)
* [svn-r13743] Description:Quincey Koziol2007-05-103-90/+201
| | | | | | | | | | | Don't delete chunks from the dataset's B-tree while we are iterating over the B-tree, when reducing the size of the dataset's dataspace with H5Dset_extent(). Tested on: Mac OS X/32 2.6 (amazon) Linux/32 2.6 (chicago) Linux/64 2.6 (chicago2)
* [svn-r13742] Description:Quincey Koziol2007-05-101-5/+28
| | | | | | | | | Avoid copying attribute in a few iteration situations. Tested on: Mac OS X/32 10.4.9 (amazon) Linux/32 2.6 (chicago) Linux/64 2.6 (chicago2)
* [svn-r13741] Description:Quincey Koziol2007-05-101-0/+3
| | | | | | | | | Minor formatting & warning reduction code cleanups Tested on: Mac OS X/32 10.4.9 (amazon) Linux/32 2.6 (chicago) Linux/64 2.6 (chicago2)
* [svn-r13733] Changed more parallel debugging info. to a more standard way by ↵MuQun Yang2007-05-041-33/+52
| | | | | | using H5DEBUG. Have tested at linux to make sure the change won't cause any compiling errors or testing errors.
* [svn-r13730] Remove debug information inside H5Dmpio.c when enable-debug is ↵MuQun Yang2007-05-031-48/+71
| | | | | | | | turned on. More code will be cleaned up tomorrow. Since only debugging information is removed, I tested this only at kagiso(linux mpich). Kent
* [svn-r13726] Add H5P_reset_external_file_test to the header file.Peter Cao2007-05-021-0/+1
|
* [svn-r13722] Description:Quincey Koziol2007-05-021-1/+1
| | | | | | | | Fix typo that causes non-Purified build to fail. Tested: FreeBSD/32 6.2 (duty) FreeBSD/64 6.2 (liberty)
* [svn-r13721] Description:Quincey Koziol2007-05-0113-178/+398
| | | | | | | | | | | | | Add "fail if unknown" and "mark if unknown" flags to object header messages. This gives the library a clean way to tell older libraries that a message should not be just ignored if it's unknown, but that other actions should occur. Tested on: Mac OS X/32 10.4.9 (amazon) FreeBSD/32 6.2 (duty) FreeBSD/64 6.2 (liberty) Linux/32 2.6 (chicago) Linux/64 2.6 (chicago2)
* [svn-r13717] Description:Quincey Koziol2007-05-012-4/+6
| | | | | | | | Clean up some formatting, comments and error messages. Tested on: Mac OS X/32 10.4.9 (amazon) Too minor to require h5committest
* [svn-r13714] Fix memory leak on objcopy.c testPeter Cao2007-04-301-0/+47
| | | | | H5Premove(dcpl, H5D_CRT_EXT_FILE_LIST_NAME) causes memory leak. Use new function H5P_reset_external_file_test(dcpl) instead.
* [svn-r13711] Description:Quincey Koziol2007-04-302-158/+239
| | | | | | | | Reduce size of fill value message in default setting case (since it can't be eliminated). Tested on: Mac OS X/32 10.4.9 (amazon)
* [svn-r13704] Change to H5G_get_refobj_name so that H5I_get_name returns the ↵Christian Chilan2007-04-271-1/+1
| | | | length of the name of a dereferenced object without counting the null terminator. Tested on smirom, copper, and cobalt.
* [svn-r13700] Updated the copyright generated by the program to the THG notice.Albert Cheng2007-04-241-2/+3
| | | | | Tested: kagiso (no committed test since it is trivial.)
* [svn-r13694] 1. changed the HDlseek to be lseek64 and HDfseek to be fseeko ↵Raymond Lu2007-04-231-2/+10
| | | | | | | | | if available, to support big files. 2. added a check in test/big.c to make sure the system supports big file. Tested on smirom because already tested the same changes for v1.6 on liberty, smirom, copper, and sol.
* [svn-r13692] Snapshot version 1.8 release 0 (beta1post0)HDF Tester2007-04-231-2/+2
|
* [svn-r13690] Description:Quincey Koziol2007-04-1911-53/+27
| | | | | | | | | | | | | 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-r13685] Description:Quincey Koziol2007-04-181-2/+2
| | | | Update the version string to "1.8.0-beta1post0"
* [svn-r13683] Description:hdf5-1_8_0-beta1Quincey Koziol2007-04-181-2/+2
| | | | | Change the release string for the beta1 release, so the tag in subversion is correct.
* [svn-r13682] Description:Quincey Koziol2007-04-184-19/+25
| | | | Run bin/reconfigure to update infrastructure files/information.
* [svn-r13660] Minor cleanups and bug fixes.Elena Pourmal2007-04-132-1/+3
| | | | | | | | | H5Zpublic.h file didn't include H5public.h; done H5Spublic.h had ghost function H5Sset_space; deleted New source files were added to make.com for VMS. h5gget_comment_c had a small bug in calculating buffer length; fixed Platforms tested: pahra and kagiso; too minor for h5committest
* [svn-r13650] Description:Quincey Koziol2007-04-122-6/+2
| | | | | | | | Clean up problems from error handling API changes in parallel and threadsafe builds. Tested on: FreeBSD/64 6.2 (liberty) w/parallel & threadsafe
* [svn-r13648] Description:Quincey Koziol2007-04-1212-1575/+1783
| | | | | | | | | | Rename new error handling API routines from H5E<foo>_stack() to H5E<foo>2(). Tested on: Mac OS X/32 10.4.9 (amazon) FreeBSD/32 6.2 (duty) FreeBSD/64 6.2 (liberty)
* [svn-r13636] Description:Quincey Koziol2007-04-1127-856/+1711
| | | | | | | | | | | | | | Change H5[D|G|T]<foo>_expand() "temporary" API routines to H5[D|G|T]<foo>2() "versioned" routines. Also added H5[D|G|T](create|commit)_anon() routines to continue to allow "anonymous" objects to be created in a file. Tested on: Mac OS X/32 10.4.9 (amazon) FreeBSD/32 6.2 (duty) FreeBSD/64 6.2 (liberty) Linux/32 2.6 (chicago) Linux/64 2.6 (chicago2)
* [svn-r13635] Description:Quincey Koziol2007-04-116-14/+17
| | | | | | | Clean up code and reduce compiler warnings... Tested on: Mac OS X/32 10.4.9 (amazon)
* [svn-r13618] Modified metadata cache to support multiple read only protects of John Mainzer2007-04-094-219/+601
| | | | | | | | | | 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-r13616] This check-in updates the bin/reconfigure file to point to the ↵Mike McGreevy2007-04-091-87/+81
| | | | | | | | most recent versions of the autotools. Updated autotool versions are: autoconf 2.61, automake 1.10.0, and libtool 1.5.22. Tested on kagiso.
* [svn-r13594] Change a comment line. No test needed.Raymond Lu2007-04-051-1/+1
|
* [svn-r13591] Problem:Albert Cheng2007-04-051-1/+2
| | | | | | | | | | | 4294967296LL is not supported by older C like Windows VC6. Solution: /* use <= 2**32 -1 to stay within 4 bytes integer range */ HDassert(chunk0_size <= 4294967295UL); Platforms tested: kelgia, copper and osage.
* [svn-r13590] Take out the unused header file errno.h.Raymond Lu2007-04-041-1/+0
|
* [svn-r13589] Pedro Vicente Nunes2007-04-041-0/+2
| | | | | | h5repack code cleaning (required reconfigure) tested: linux (32, 64, parallel), solaris
* [svn-r13588] A support of files bigger than 2GB for STDIO driver. Configure ↵Raymond Lu2007-04-044-23/+112
| | | | | | | | will check if fseeko is available. Using it instead of fseek can support big files because the offset is of type off_t not long int. Also added the test for STDIO in big.c.
* [svn-r13580] Description:Quincey Koziol2007-04-033-107/+154
| | | | | | | | | | | | Add version # and flags to external link format (as fields in a single byte), in order to accomodate future changes/expansions. 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-r13575] Problem:Elena Pourmal2007-04-031-1/+1
| | | | | | | | | | | dsets failed on kelgia with assertion failure in H5Ocache.c. Kelgia could not deal with the comparison of uint64_t variable with 4294967296 constant. Solution: By John's suggestions used suffix LL for the constant. Platforms tested: kelgia, kagiso, and smirom with icc compiler
* [svn-r13573] Pedro Vicente Nunes2007-04-021-0/+18
| | | | | | | | | | | | Bug fixes Reset external file list slots name_offset to a state when created (0) in H5P_dcrt_copy so that it conforms to an assertion in H5D_update_entry_info that assumes the name_offset is 0 at this point this fixes the problem of h5repack and external files, add a new test and files for an external file h5diff, check for an error return in H5D_get_storage_size tested linux 32, 64
* [svn-r13570] Description:Quincey Koziol2007-03-302-1/+38
| | | | | | | | Fix error when eliminating empty chunk from "middle" of list of chunks for an object header. Tested on: Mac OS X/32 10.4.9 (amazon)
* [svn-r13549] Description:Quincey Koziol2007-03-271-1/+2
| | | | | | | | | Check in changes from Elena and I to get pgcc compiler working again. Primarily (all?) changes to move from using 'hsize_t' as array index to using something else ('size_t') mostly. Tested on: Linux/32 2.4 kagiso w/pgcc
* [svn-r13544] Some minor adjustment from last round of checkin. Tested on kagisoRaymond Lu2007-03-261-4/+3
| | | | only because the changes are simple.
* [svn-r13542] Added a detection for zero-length VL data in H5T_conv_vlen to ↵Raymond Lu2007-03-261-1/+7
| | | | | | handle the case properly. This fix was discovered by the user while Bug #717 was being fixed. Modefied the test, too.
* [svn-r13534] Description:Quincey Koziol2007-03-211-101/+66
| | | | | | | | Minor code cleanups... Tested on: FreeBSD/32 6.2 (duty) Too minor to require h5committest
* [svn-r13531] When a dataset is created and no data is written, but fill ↵Raymond Lu2007-03-201-21/+77
| | | | | | | | | | | | value is defined, if there's VL type of data in the dataset, the address of the data is copied multiple times into the buffer, causing some trouble when the data is released. Instead, this checkin makes multiple copies of fill value first, then does conversion on each element so that each of them has a copy of the VL data. A test is added, too. Tested it on kagiso, sol, and copper.
* [svn-r13523] DescriptionAlbert Cheng2007-03-161-7/+8
| | | | | | | | | Switch the default setting for configure from --enable-production/--disable-debug to --disable-production/--enable-debug. Platform Tested: In kagiso with fortran and C++. No commit test because changes happened in configure and it is straight forward.
* [svn-r13520] Description:Quincey Koziol2007-03-162-2/+113
| | | | | | | | | Added H5Lexists() API routine, which determines if a link of a particular name exists in a group. Tested on: Linux/32 2.6 (chicago) Linux/64 2.6 (chicago2)
* [svn-r13518] Description:Quincey Koziol2007-03-161-1/+8
| | | | | | | | Add some sanity checking to improve the resilience of the object header decoding in the face of a bad reference, or a corrupted file. Tested on: Mac OS X/32 10.4.9 (amazon)
* [svn-r13517] To deal with the situation that compiler supports direct I/O ↵Raymond Lu2007-03-151-10/+22
| | | | | | but file system doesn't, added a condition check of writing aligned data. Skip the test for direct VFD if it's this situation.
* [svn-r13500] Description:Quincey Koziol2007-03-122-1/+4
| | | | | | | | | Roll changes to SOHM v2 B-tree records into format, to get the format for them stable while I'm working on debugging the actual support for the "unique, but sharable" feature. Tested on: FreeBSD/32 6.2 (duty)
* [svn-r13499] Description:Quincey Koziol2007-03-128-19/+74
| | | | | | | | Avoid storing the # of attributes in the "attribute info" message and regenerate it when the object is opened. Tested on: FreeBSD/32 6.2 (duty)
* [svn-r13498] Description:Quincey Koziol2007-03-1216-67/+164
| | | | | | | | | Eliminate storing # of links in "link info" message, regenerate it when the object is opened instead. Tested on: FreeBSD/32 6.2 (duty) Mac OS X/32 10.4.8 (amazon)
* [svn-r13497] Description:Quincey Koziol2007-03-119-107/+614
| | | | | | | | | | | | | | | | | | Move ref. count of # of links to an object out of the object header's prefix and make it a header message instead (since it's a "rare" occurence), eliminating some more space for each object in the file. Inserting this "ref. count" message exposed a flaw in the library's mechanism for locating a message to promote to another chunk and replace with a continuation message, which required some additional work to fix. It's still not completely robust, but it's working for more cases now and detects failures robustly. Reduced the minimum size of an object header chunk to just enough to contain a header message prefix and continuation message. Tested on: FreeBSD/32 6.2 (duty)
generated by cgit v0.12 at 2024-07-22 18:24:37 (GMT)