summaryrefslogtreecommitdiffstats
path: root/src/H5Dchunk.c
Commit message (Collapse)AuthorAgeFilesLines
* Added an H5MM_memcpy call that checks for buffer overlap.Dana Robinson2019-03-161-21/+21
|
* Fix issue with direct chunk write not updating the "last chunk" indexNeil Fortner2019-03-061-3/+37
| | | | | cache. Fix issues involving datasets being "no allocated" when they contain cached raw data.
* Merge branch 'develop' into c_stdlib_posix_call_cleanupDana Robinson2019-02-191-70/+53
|\
| * Align develop with incoming hyperslab_updates branch changes.Quincey Koziol2019-02-131-48/+41
| |
| * Fix some collective metadata read issuesJordan Henderson2019-02-121-16/+6
| |
| * Delta reduction against hyperslab_updates branch.Quincey Koziol2019-01-111-6/+6
| |
* | C and POSIX call cleanupDana Robinson2019-01-111-7/+8
|/
* Removed a comment.Binh-Minh Ribler2019-01-031-1/+0
|
* Fixed CVE division-by-zero issuesBinh-Minh Ribler2018-12-231-3/+26
| | | | | | | | | | | | | | | | | Description: Fixed HDFFV-10577 and similar issues found in H5Dchunk.c. All the occurrences are in: H5D__create_chunk_map_single H5D__create_chunk_file_map_hyper H5D__chunk_allocate H5D__chunk_update_old_edge_chunks H5D__chunk_prune_by_extent H5D__chunk_copy_cb H5D__chunk_collective_fill Platforms tested: Linux/64 (jelly) Linux/64 (platypus) Darwin (osx1010test)
* Merge pull request #1335 in HDFFV/hdf5 from ↵Ray Lu2018-11-191-0/+3
|\ | | | | | | | | | | | | | | | | | | | | ~SONGYULU/hdf5_ray:bugfix/HDFFV-10571-cve-2018-17237-divided-by-zero to develop * commit 'c923cdad6e515c842f3795a5b6d754ad94021e09': HDFFV-10571: Minor format changes. HDFFV-10571: Minor change - reformatting the error check. HDFFV-10571: Minor change - adding the error check right after decoding of chunk dimension for safeguard. HDFFV-10571: Minor change - revised the comment to be clearer. HDFFV-10571 Divided by Zero vulnerability. Minor fix: I added an error check to make sure the chunk size is not zero.
| * HDFFV-10571: Minor format changes.Songyu Lu2018-11-151-1/+0
| |
| * HDFFV-10571: Minor change - adding the error check right after decoding of ↵Songyu Lu2018-11-151-4/+4
| | | | | | | | chunk dimension for safeguard.
| * HDFFV-10571: Minor change - revised the comment to be clearer.Songyu Lu2018-11-141-1/+1
| |
| * HDFFV-10571 Divided by Zero vulnerability. Minor fix: I added an error ↵Songyu Lu2018-11-141-0/+4
| | | | | | | | check to make sure the chunk size is not zero.
* | HDFFV-10601: I changed to a better way to calculate the number of chunks in ↵Songyu Lu2018-11-081-5/+3
| | | | | | | | a dataset.
* | HDFFV-10601 Issues with chunk cache hash value calcuation:Songyu Lu2018-11-021-12/+12
|/ | | | | | | | 1. H5D__chunk_hash_val: When the number of chunks in the fastest changing dimension is larger than the number of slots in the hash table, H5D__chunk_hash_val abandons the normal hash value calculation algorithm and simply uses the scaled dimension. This will cause chunks a selection that cuts across chunks in dimensions other than the fastest changing to all have the same hash value, and they will therefore always evict each other from the cache, having an obvious major performance impact. Eliminated the check for the number of slots in this function and always use the full algorithm. 2. H5D__chunk_init: When the scaled dimensions (number of chunks in each dimension) are calculated in H5D__chunk_init, a simple divide ("/") operator is used with the dataset size in elements and the chunk size in elements. While this is fine when the dataset size is an exact multiple of the chunk size, in other cases, since "/" rounds down, it results in a scaled dimension one less than it should (it ignores the partial edge chunk). This has trickle down effects on hash value calculation that can cause excess hash value collisions and therefore performance issues. Changed the calculation to (((dataset_size - 1) / chunk_size) + 1). Tested the build with Autotool and CMake.
* A trivial change to reformat a comment. The purpose is simply to try out Git.Songyu Lu2018-10-311-2/+2
|
* Remainder of vol_normalization changes (dataset, attribute, files, objects).Dana Robinson2018-09-241-37/+36
|
* Removed 'VOL' versions of FUNC_ENTER macros as well as empty functionsDana Robinson2018-09-181-4/+4
| | | | orginally intended to support the full SWMR feature.
* Merge pull request #1134 in HDFFV/hdf5 from ~BMRIBLER/hdf5_bmr_cpp4:develop ↵Binh-Minh Ribler2018-07-161-0/+2
|\ | | | | | | | | | | | | | | | | | | to develop Fixed HDFFV-10481 and HDFFV-10477 * commit '83ca39ba9a2d6852dd7754b533f7839e9d2a0107': Combined macro lines as Dana commented Platforms tested: Linux/64 (jelly) (very minor) Fixed division-by-zero issues Description: Fixed HDFFV-10481 and HDFFV-10477, division by 0. Fixed another occurrence beside what were reported. Also, changed a local variable to avoid an unnecessary cast. Platforms tested: Linux/64 (jelly) Linux/32 (jam) Darwin (osx1010test)
| * Combined macro lines as Dana commentedBinh-Minh Ribler2018-07-161-2/+1
| | | | | | | | | | Platforms tested: Linux/64 (jelly) (very minor)
| * Fixed division-by-zero issuesBinh-Minh Ribler2018-07-161-0/+3
| | | | | | | | | | | | | | | | | | | | | | Description: Fixed HDFFV-10481 and HDFFV-10477, division by 0. Fixed another occurrence beside what were reported. Also, changed a local variable to avoid an unnecessary cast. Platforms tested: Linux/64 (jelly) Linux/32 (jam) Darwin (osx1010test)
* | Merge pull request #1127 in HDFFV/hdf5 from ~JHENDERSON/hdf5:develop to developJordan Henderson2018-07-161-2/+35
|\ \ | |/ |/| | | | | | | | | | | | | | | | | | | | | | | | | | | * commit 'a8d6f100cdddbfc42d0c4abfeb4ceb8788b1b087': Add note about single chunk caching and serial library Add check for actually using the MPI file driver when caching one chunk Fix error message mentioning wrong MPI function used Fix for HDFFV-10509 Revise H5D__mpio_array_gatherv() to not allocate memory needlessly Add test to continually grow and shrink chunks Changes to test with checksum filter as well as deflate filter Eliminate warning about signed to unsigned conversion Remove unused local variable Fix bug in parallel reads of compressed data Add data verification to parallel filtered compound write tests Add seven of fourteen parallel filtered data partial read tests
| * Add note about single chunk caching and serial libraryJordan Henderson2018-07-131-1/+7
| |
| * Add check for actually using the MPI file driver when caching one chunkJordan Henderson2018-07-131-1/+15
| | | | | | Better explain rationale behind chunk caching issue fix
| * Fix for HDFFV-10509Jordan Henderson2018-07-041-2/+15
| |
* | Fixed HDFFV-10404Binh-Minh Ribler2018-07-131-4/+4
| | | | | | | | | | | | | | | | | | Description: Applied the typo fixes from user's report. The previous pull request couldn't be merged because it was too old, and it was too complicated for me to resolve conflicts. Platform tested: Linux/64 (jelly) - very minor
* | Revert "Merge pull request #1116 in HDFFV/hdf5 from ↵hdftest2018-06-241-2/+2
| | | | | | | | | | | | | | ~HDFTEST/hdf5_hft:hdf5_1_10 to develop" This reverts commit e9f476dad47ce593f13dacb77b1cc664d1f24e7b, reversing changes made to 2ff00b1b937ebe36ac6ddf590c16a4c27fc0b053.
* | Merge branch 'develop' into hdf5_1_10.syncM. Scot Breitenfeld2018-05-081-317/+194
|\ \ | |/
| * Merge branch 'develop' of https://bitbucket.hdfgroup.org/scm/hdffv/hdf5 into ↵Quincey Koziol2018-04-291-343/+547
| |\ | | | | | | | | | merge_hyperslab_updates
| | * Merge remote-tracking branch 'origin/develop' into merge_func_enter_volQuincey Koziol2018-03-211-1/+1
| | |\
| | * | Add API context interface and use it throughout the library.Quincey Koziol2018-03-151-306/+187
| | | |
| * | | Merge branch 'develop' of https://bitbucket.hdfgroup.org/scm/hdffv/hdf5 into ↵Quincey Koziol2017-04-111-5/+5
| |\ \ \ | | | | | | | | | | | | | | | merge_hyperslab_updates
| * | | | Further minor cleanupsQuincey Koziol2017-03-281-2/+2
| | | | |
| * | | | Checkpoint normalization against incoming hyperslab / selection / dataspaceQuincey Koziol2017-03-281-11/+7
| | | | | | | | | | | | | | | | | | | | improvements.
* | | | | Merge pull request #983 in HDFFV/hdf5 from develop to hdf5_1_10Larry Knox2018-03-201-1/+1
|\ \ \ \ \ | | |_|_|/ | |/| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | * commit '6a1627f2967099039e11b9e86cf347f6482bdf3c': Fix for HDFFV-10425 test failure with H5DOwrite_chunk and latest format Fix date. Correct the date in RELEASE.txt for H5DOread_chunk. Add release info for H5DOread_chunk. Updated RELEASE.txt Description: Wrappers for H5Tencode and H5Tdecode No code change. Fixed HDFFV-10151 Description: Added a document to the HDF5 C++ API Reference Manual to show the mapping from a C API to C++ wrappers. Fix memory leak simply by capturing allocation in var Change max value Chnage values so test will fail when new latest is added Exception changed Exception type changed Correct constant var names Java constants for new lib verbounds values
| * | | | Fix for HDFFV-10425 test failure with H5DOwrite_chunk and latest formatVailin Choi2018-03-201-1/+1
| | |_|/ | |/| |
* | | | Fix for daily test failures.Vailin Choi2017-12-071-2/+2
|/ / /
* | | Fixed an uninitialized filter callback struct in H5Dchunk.c andDana Robinson2017-11-211-34/+42
| | | | | | | | | | | | | | | | | | unified the naming and initialization of said struct throughout the library. This was causing a crash on VS2015 in debug mode when the debug heap complained. Fixes HDFFV-10330.
* | | Cleaned up misc warnings in src and test.Dana Robinson2017-11-101-1/+1
| | |
* | | Merge branch 'develop' of https://bitbucket.hdfgroup.org/scm/hdffv/hdf5 into ↵Jordan Henderson2017-08-111-1/+6
|\ \ \ | | | | | | | | | | | | develop
| * | | Fix for HDFFV-10217 infinite loop in H5VM_power2up().Vailin Choi2017-07-101-1/+6
| | | | | | | | | | | | | | | | | | | | | | | | The function H5VM_power2up() returns the next power of 2 for n. When n exceeds 2^63, it overflows and becomes 0 causing the infinite looping. The fix ensures that the function checks for n >= 2^63 and returns 0.
* | | | Merging in latest from upstream (HDFFV/hdf5:refs/heads/develop)Jordan Henderson2017-07-051-74/+387
|\ \ \ \ | |/ / / | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | * commit '6428a637752753cc8a9007c0c3c38b9ea9cc97fd': (165 commits) Skip HDFFV-10160 filenotclosed test for split/multi driver Temporary skip this test for split/multi driver. Further investigation to resolve the failure is needed. Please enter the commit message for your changes. Lines starting More minor whitespace cleanup Fixed a minor comment typo in H5R.c * Reverted removal of H5I_REFERENCE to avoid breaking binary compatibility. * Added a quick H5I_REFERENCE smoke check to test_misc19() in tmisc.c. Added missing free() call to test_misc19(). Purged the unused H5I_REFERENCE from the library. Added a test for VFD IDs to testhdf5. Revert "Partial removal of NDEBUG ifdefs from the library." Partial removal of NDEBUG ifdefs from the library. Removed compiler flag cleanup since Solaris chokes on it. Fix for HDFFV-10160 Modifications to fix the assertion/abort failure when the application does not close the file. Fixed a bug in autotools Fortran processing where a test for the presence of a C long double type always returned true. Updated the --enable-debug/production handling in configure.ac so that it more reliably indicates that the configure option can no longer be used. Previously, using --enable-debug="", etc. would not trigger an error (even though such usage is unlikely). Added awk processing to remove extraneous whitespace from H5_ flags in configure. Fixed a comment in configure.ac HDFFV-10219 another mode change HDFFV-10219 change subset to use BINARY mode for h5dump HDFFV-10219 default input binary to NATIVE Set default of var to NATIVE HDFFV-10219 - disable h5dump tests until h5import NATIVE is fixed ...
| * | | Fix for HDFFV-7853 H5Ocopy doesn't work with open identifiersVailin Choi2017-06-111-40/+107
| | | | | | | | | | | | | | | | Changes made so that raw data for dataset objects are copied from cached info when possible instead of flushing objects to file and read them back in again.
| * | | Revert "Merge pull request #567 in HDFFV/hdf5 from ↵Dana Robinson2017-06-091-12/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | ~DEROBINS/hdf5_der:develop to develop" This reverts commit 4242753848c44ab3b5d226e66225eac2f64db314, reversing changes made to b0e79fe6dd20ec7aa1b3e5f0f8b370639a4ef5bd.
| * | | * Made STATIC free lists EXTERN for H5O_layout_t.Dana Robinson2017-06-071-1/+1
| | | | | | | | | | | | | | | | * Made all non-global instances of H5O_layout_t in H5Pdcpl.c dynamic.
| * | | Updated H5O_layout_t dynamic work to use the H5FL interface.Dana Robinson2017-06-071-2/+5
| | | |
| * | | Made some H5O_layout_t stack allocations dynamic (quiets warnings).Dana Robinson2017-06-061-4/+9
| | | |
| * | | Incorporate additional code changes for the H5DOread_chunk patch from GE ↵Vailin Choi2017-05-151-21/+22
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Heathcare (HDFFV-9934) This is the similar set of changes that was checked in to 1.8 branch: Incorporate the code changes that were missing from the original patch: (1) Additional tests in hl/test/test_dset_opt.c (2) Fix in src/H5Dchunk.c for direct access when an entry is in chunk cache but not dirty Tested on platypus, ostrich, kituo, mayll, osx1010test, moohan, quail, emu.
| * | | Merged HDFFV-10187 direct chunk overwrite bugfix to develop from 1.10 branch.Dana Robinson2017-05-121-5/+11
| | | |
generated by cgit v0.12 at 2024-11-01 08:59:15 (GMT)