summaryrefslogtreecommitdiffstats
path: root/src
Commit message (Collapse)AuthorAgeFilesLines
* Quiets most const warnings in the fractal heap code (#1188)Dana Robinson2021-11-133-9/+32
|
* Fix MPI Comm and Info leak in H5FD__mpio_delete (#1189)jhendersonHDF2021-11-121-0/+6
|
* Cleans up pointer and ID reset code in H5Pint.c (#1182)Dana Robinson2021-11-101-74/+112
|
* Quiets const warning in H5RS code (#1181)Dana Robinson2021-11-101-2/+10
|
* Fix H5DS warnings related to new H5DSwith_new_ref and H5VLobject_is_native ↵jhendersonHDF2021-11-102-11/+15
| | | | APIs (#1184)
* Fixes FUNC_ENTER warnings in VFDs due to recent init changes (#1178)Dana Robinson2021-11-106-15/+9
|
* Fixes const issues in the version 2 B-trees (#1172)Dana Robinson2021-11-092-24/+21
| | | | The operations that were changed are fundamentally not const since the shadow operation can modify the node structure when SWMR is in use.
* Re-enable collective metadata reads after disabling for chunk lookup (#1173)jhendersonHDF2021-11-091-7/+21
|
* Simplify function enter macros for performance benefits (#1024)David Young2021-11-0879-1988/+1056
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | * Take a stab at using constructors to initialize instead of function-entry macros. This is a work in progress. It's good enough to run `many_dsets`. * Committing clang-format changes * Add the `many_dsets` benchmark and some scripts I used on jelly for setting up the build/test environment and for recording/flame-graphing profiles. * Committing clang-format changes * Change my Makefile and environment script to work both on jelly and on mayll (and probably on Summit). * Disable clang-format "fix." * Replace the `if (!H5_TERM_GLOBAL)` test in each FUNC_ENTER_ macro with `if (true)`. * Fix bad grammar in a comment. * Instead of labeling the H5*__init_package routines constructors, fold each into an initialization routine, H5*_init(), and call each of the H5*_init() routines. Call most of the H5*_init() routines from H5_init_library() in an explicit order that I found out earlier by instrumenting each __init_package routine and running the library tests. Roll H5FD*__init_package routines into H5FD*_init() routines. This change ends just-in-time initialization of package dependencies by package initializers. Don't track in per-package variables (H5_PKG_INIT_VAR) whether each package has been initialized. Instead, track in a single library variable whether the whole library is initialized or not. Drive the initialization of packages by H5_init_library() with a table of initializer routines. Also drive the termination of packages by H5_term_library() with a table. Perform initialization as needed from FUNC_ENTER_API_INIT(err). This basically restores the old behavior of that macro. Delete a bunch of #definitions in H5private.h that have fallen out of use with these changes. * Committing clang-format changes * Undo the bad auto-formatting that appears to have occurred in spite of my disabling it. Bracket some code in /* clang-format off */ /* clang-format on */ to prevent a recurrence. * Remove a diagnostic abort(). * Fix a logic error: print a comma between every package terminator run, and don't print an initial comma. * Complete the changes I started in H5_term_library() that undo the bad auto-formatting. Stop tracking whether package "tops" were initialized in per-package variables H5*_top_package_initialize_s. H5_term_library() takes care of that for them. Remove H5R_top_term_package() and H5R_term_package(), they don't do anything. * Committing clang-format changes * NFCI. Simplify macro text: replace `if (true) {` with `{`. * Fix formatting and suppress clang-format on a longer range. * Quiet some unused label, unused variable complaints that cropped up after I simplified the FUNC_ENTER_ macros for the sake of performance. * Committing clang-format changes * Delete some programs and scripts that don't belong in the pull request. * Use the right function-entry macro. * Use a sensible format and disable auto-formatting. * Stop calling do-nothing initializer H5FS_init(). Delete it. * Document what changes to make if the default VFD changes. * While I am here, change an `await_prior` flag on the terminator table to `true` to match the previous, non-table-driven code that was here. Found the oversight making the following changes: NFCI: insert an empty line and copy over slightly-edited comments from the previous version, where those comments still correctly explained how library termination operated. * NFCI: lower a staircase. * Replace every occurrence of FUNC_ENTER_NOAPI_INIT(...) with H5_PUSH_FUNC since that is all that that macro does any more. Quiet a bunch of new warnings by changing FUNC_ENTER_NOAPI(...) to FUNC_ENTER_NOAPI_NOERR and removing disused `done:` labels. * NFCI: add curly braces around a multiline statement. * Quiet a signed/unsigned comparison warning. * Add some documentation about library initialization and shutdown. * Make sure that the library is initialized, or else that initialization is already underway, before performing any VFD's initialization. * Committing clang-format changes * Committing clang-format changes * Reduce differences from `develop` branch. * Always initialize `tot_init`. * Committing clang-format changes * Fix typo: H5SL_init initializes skip lists, not VOL. * Remove H5_TERM_GLOBAL test in H5T_init. H5T_init was unusual in that it tested H5_TERM_GLOBAL and exited early if it was set. No other module initializers did that, and I cannot find any reason that should be necessary. Tests still pass when I remove it, so away it goes. * Use HD prefix. * Add function header comments. * Drop the intermediate variable, it's only used once. * Extract subroutine `H5FDperform_init(hid_t (*init)(void))` that initializes the library, if necessary, before calling its VFD-initializer argument. Use H5FDperform_init in the definition of the symbols H5FD_<vfd> (e.g., H5FD_SEC2), which may be evaluated before the library is initialized, like so: ``` ``` I implement H5FDperform_init in its own source file, H5FDperform.c, and exclude that file from trace processing because the `bin/trace` cannot deal with the function-pointer type. * Straggler from last: add new source file src/H5FDperform.c. * Committing clang-format changes * Add a missing file to the MANIFEST. * Switch to FUNC_ENTER_API_NOINIT in H5FDperform_init() and hbool_t in H5_term_library(). Co-authored-by: github-actions <41898282+github-actions[bot]@users.noreply.github.com>
* Fix the function cast error in H5Dchunk.c and activate (#1170)David Young2021-11-081-1/+14
| | | `-Werror=cast-function-type`. Again.
* Assume C99 fixed sized ints exist, use them (#470)Sean McBride2021-11-0812-245/+126
| | | | | | | | | | | | | * Committing clang-format changes * Assume C99 fixed sized ints exist, use them * Assume H5_SIZEOF_LONG_DOUBLE != 0, `long double` has existed since C89 Note, this is only assuming that `long double` exists, no assumptions about its size have been touched. Didn't remove any code that does things like test if `long double` and `double` have different sizes. * Committing clang-format changes Co-authored-by: github-actions <41898282+github-actions[bot]@users.noreply.github.com>
* OESS-168: Remove clang warnings. (#1124)H. Joe Lee2021-11-052-12/+22
| | | | | | | | | | | * OESS-168: Remove clang warnings. * OESS-168: Remove clang warnings. * Committing clang-format changes * OESS-168: Address @derobins review for FALLTHROUGH. Co-authored-by: github-actions <41898282+github-actions[bot]@users.noreply.github.com>
* New references for Dimension Scale APIs (#1139)jhendersonHDF2021-11-053-25/+68
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | * Enable usage of new-style references with dimension scale APIs * Add API to check if an object ID represents a native connector object * Modified code to use new function H5DSwith_new_ref to determine if new references should be used with Dimension Scales. The new function return TRUE if non-native connector is used or if H5_DIMENSION_SCALES_WITH_NEW_REF varible is define at configure time (--enable-dimension-scales-with-new-ref). Tested on jelly. ToDo: generate testing file on BE system and enable the test; add flag to CMake; test netCDF-4 with the new references. * Adding new test files generated on BE system (hedgehog) created by 32 and 64-bit library. test_ds chokes on test_ds_le_new_ref.h5 on BE system; test passes for test_ds_be_new_ref-32bit.h5 for the 32-bit library and fails for the 64-bit library, and vice versa. I am checking the files for further investigation; but current implementation of the new references is not portable between LE and BE systems, and 32 and 64-bit systems. * Minor fixes for testing issues * Update test_ds.c Enabled broken test; tests pass now. * Update RELEASE.txt Documented new option to use new references with the HDF5 dimension scales APIs (H5DS*). * Update MANIFEST for new 32-bit new-style references test file for H5DS APIs * Update 'dimension scales w/ new-style refs' feature based on review Co-authored-by: Elena <epourmal@hdfgroup.org>
* OESS-168: Remove clang warnings. (#1105)H. Joe Lee2021-11-041-33/+33
| | | | | | | | | | | | | * OESS-168: Remove clang warnings. * Committing clang-format changes * OESS-168: Address @soumagne review. * Committing clang-format changes * OESS-168: Address @soumagne review. Co-authored-by: github-actions <41898282+github-actions[bot]@users.noreply.github.com>
* OESS-168: Remove clang warnings. (#1127)H. Joe Lee2021-11-021-6/+6
| | | | | * OESS-168: Remove clang warnings. * OESS-168: Address @lrknox clang-format review.
* Removes the "try free" behavior from the skip lists (#1126)Dana Robinson2021-10-222-470/+200
| | | | | | | | | | | | | | * Removes the "try free" behavior from the skip lists This was only used in the ID code when iterating and a callback could delete IDs. It is not used anywhere else in the library and is now pointless overhead. Also quiets the const warnings when returning stored elements. They only need to be const with respect to the skip list code, which should never modify them. The library can do whatever it wants with the elements it stored. * Formatted source
* OESS-168: Remove clang warnings. (#1117)H. Joe Lee2021-10-211-1/+1
|
* Fix map open to use correct VOL argument structure (#1116)jhendersonHDF2021-10-211-1/+3
| | | Fix map iterate by setting key memory type ID field
* OESS-168: Remove clang warnings. (#1076)H. Joe Lee2021-10-081-9/+9
|
* Fix bug with cross platform compatibility of references within vlens. (#1064)Neil Fortner2021-10-011-2/+2
| | | No testing yet.
* Fix Mirror, HDFS and ROS3 case statements for VFD value in H5trace (#1062)jhendersonHDF2021-09-301-6/+6
|
* Update compact dataset I/O routines to handle driver-level memory copy (#1054)jhendersonHDF2021-09-304-13/+145
| | | | | | | | | | * Update compact dataset I/O routines to handle driver-level memory copy * Combine compact I/O read/write struct into single I/O struct Rename CTL memory copy flag and H5Fquery routine to get file driver structure * Add RELEASE.txt entry for new "ctl" memory copy operation
* VFD plugins (#602)jhendersonHDF2021-09-2943-500/+2589
| | | | | | | | | | | | | | | | | * Implement support for loading of Virtual File Drivers as plugins Fix plugin caching for VOL connector and VFD plugins Fix plugin iteration to skip paths that can't be opened * Enable dynamic loading of VFDs with HDF5_DRIVER environment variable * Temporarily disable error reporting during H5F_open double file open * Default to using HDstat in h5_get_file_size for unknown VFDs * Use macros for some environment variables that HDF5 interprets * Update "null" and "ctl testing" VFDs
* Update version to 1.13.0-7 after snapshot release 1.13.0-6. (#1048)Larry Knox2021-09-291-2/+2
|
* Update Platforms Tested in RELEASE.txt. (#1045)hdf5-1_13_0-rc6Larry Knox2021-09-281-2/+2
| | | | | | * Update Platforms Tested in RELEASE.txt. Remove Tested Configuration Features Summary for updated replacement. * Set version to 1.13.0-6 for rc6 snapshot.
* VFD ctl feature (#981)jhendersonHDF2021-09-2317-146/+526
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | * Added "ctl" callback to the VFD interface, and the associated H5FDctl() and H5FD_ctl() calls. Modified the MPIO VFD accordingly -- specifically: Added ctl() call with op-code support to expose rank, size, and communicator. Modified H5FD_mpi_get_rank(), H5FD_mpi_get_size(), and H5FD_mpi_get_comm() to use the new ctl() callback. In passing removed the const qualifier from the file parameter of these functions, as the file parameter of the ctl callback is not const. Deleted the old H5FD__mpio_mpi_rank(), H5FD__mpio_mpi_size(), and H5FD__mpio_communicator() calls from the MPIO VFD. Deleted H5FD_class_mpi_t from H5FDprivate.h, and modified the MPIO VFD accordingly. Note that all VFDs now use H5FD_class_t, with no special class for VFDs that that support MPI. Some minor touch ups to the Neil's selection I/O mods in passing. Tested serial and parallel, debug and production on charis and jelly. * Reserve a range of VFD "ctl" opcodes for library and experimental usage * Add "ctl" callbacks to passthrough VFDs * Add RELEASE.txt entry for "ctl" callback * Use H5FDopen with H5F_ACC_RDWR flag instead of H5F_ACC_TRUNC in vfd test * Remove handling of passthrough "ctl" flag from multi VFD * Move logic for testing H5FD_CTL__TEST_OPCODE into a testing VFD Revise description of "ctl" callback in RELEASE.txt Remove unused H5FD_CTL__NUM_OPCODES definition Fix some warnings in multi VFD Co-authored-by: mainzer <mainzer#hdfgroup.org>
* Modified gcc/clang warning suppression macros to account for some warnings ↵Sean McBride2021-09-1610-106/+134
| | | | | | | | | | | | | | | | | flags being supported by one compiler but not the other (#379) * Committing clang-format changes * Fixed GCC warning suppression pragmas to also work with clang H5_GCC_DIAG_ON remains gcc-only. Added a new H5_CLANG_DIAG_ON that's clang-only, but it's not used anywhere currently. Added a new H5_GCC_CLANG_DIAG_ON that works with both compilers, which afterall support mostly the same warnings. Changed almost all uses of H5_GCC_DIAG_ON to use H5_GCC_CLANG_DIAG_ON, with the exception of a couple, where they really were suppressing gcc-only warnings. * Committing clang-format changes Co-authored-by: github-actions <41898282+github-actions[bot]@users.noreply.github.com>
* Fix several warnings (#720)jhendersonHDF2021-09-1413-33/+47
|
* H5Ztrans: (feature) Improved H5Z_xform_noop() and H5Z_xform_create() … (#933)Jan-Willem Blokland2021-09-131-5/+13
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | * H5Ztrans: (feature) Improved H5Z_xform_noop() and H5Z_xform_create() function - Made a small improvement for H5Z_xform_noop() function. Now, the function will also return TRUE if the data transform function expression = "x". For this case, the HDF5 library behaves in a similar fashion as the case when no data transform function has been specified. - Improved the inline documentation of the function H5Z_xform_create() such it is more inline with the rest of the code. * Committing clang-format changes * H5Ztrans: (feature) Added 3 tests for improved H5Z_xform_noop() function - Added serial test with data transform expression = "x" to verify the improved H5Z_xform_noop() function behaves as expected. - Added 2 parallel tests with data transform expression = "x" in combination with a filter. Before, these tests will fail but with the improved H5Z_xform_noop() function they work and result in the expected behavior. - Small bug fix for one of parallel filter tests. * Committing clang-format changes * H5Ztrans: (feature) Added release note about detection of the simple data transform function "x". - Added a brief explanation about the implemented improvement of the detection of the simple data transform function "x" to the RELEASE.txt file. Co-authored-by: github-actions <41898282+github-actions[bot]@users.noreply.github.com>
* Fix various warnings and issues in doxygen doc generation (#994)Allen Byrne2021-09-103-43/+11
| | | | | * Fix various warnings and issues in doxygen doc generation * Add new file to list
* Update copyright headers with reproduction notice in 5 files (#993)Larry Knox2021-09-082-17/+14
| | | | | | | | | * Update copyright headers with reproduction notice in 5 files that did not match terms in COPYING file . * Committing clang-format changes Co-authored-by: github-actions <41898282+github-actions[bot]@users.noreply.github.com>
* Small doxygen fixes plus simple doc fixes (#989)Allen Byrne2021-09-042-10/+8
| | | | Change doxygen generated file location Clang format override for flags
* Synch CMake doxygen options with autotools configure (#983)Allen Byrne2021-09-021-1/+2
|
* A batch of life-cycle examples for different modules (#654)Gerd Heber2021-09-0132-1383/+1166
| | | | | | | | | * Create a tag file for permalinks. * Added DOXYGEN_TAG_FILE. * Added Doxygen life-cycle examples for different modules. Co-authored-by: github-actions <41898282+github-actions[bot]@users.noreply.github.com>
* More various warnings (#958)Sean McBride2021-08-254-22/+22
| | | | | | | | | | | | | * Committing clang-format changes * Fixed various -Wdouble-promotion warnings * Fixed -Wshadow warning for `optopt` conflict On macOS at least, there is a global various named `optopt` in unistd.h. * Committing clang-format changes Co-authored-by: github-actions <41898282+github-actions[bot]@users.noreply.github.com>
* More clang tidy (#908)Sean McBride2021-08-236-42/+42
| | | | | | | | | | | | | | | | | | | * Pacify clang-analyzer-unix.cstring.NullArg * Apply some bugprone-suspicious-string-compare * Apply some readability-simplify-boolean-expr * Apply some readability-make-member-function-const * Apple some bugprone-macro-parentheses * Changed an f suffix to L for `long double` * Applied some readability-uppercase-literal-suffix automatically * Committing clang-format changes Co-authored-by: github-actions <41898282+github-actions[bot]@users.noreply.github.com>
* Develop windows compiler fixes (#912)Allen Byrne2021-08-191-0/+2
| | | | | | | | | | | | | | * Fix windows compiler issues Restore MS C++ Default exception handling behavior #885 Remove clang toolchain as it skips compiler checks * Add check and header gaurd for pwd.h * Revert change * Committing clang-format changes Co-authored-by: github-actions <41898282+github-actions[bot]@users.noreply.github.com>
* Correct number of args to Windows HDunsetenv macro (#939)Allen Byrne2021-08-191-1/+1
| | | | | * Correct number of args to Windows HDunsetenv macro * Correct format
* Fix formatting (#923)Allen Byrne2021-08-181-1/+1
|
* Fixes a bad memory read and unfreed memory in fsinfo code (#893)Dana Robinson2021-08-122-14/+22
| | | | | | | | | | | | | | * Fixes a bad memory read and unfreed memory in fsinfo code The segfaul from CVE-2020-10810 was fixed some time ago, but the illegal memory read and unfreed memory were not. This fix tracks some buffer sizes and errors out gracefully on errors, ensuring buffers are cleaned up and avoiding the H5FL infinite loop + abort on library close. * Committing clang-format changes Co-authored-by: github-actions <41898282+github-actions[bot]@users.noreply.github.com>
* Guard the SKIP_MPICXX macros against redefinition. (#892)David Wells2021-08-121-0/+4
|
* Unsigned comparison to 0 warning avoidance (#869)Larry Knox2021-08-021-3/+4
| | | | | | | * Added int variable for comparison with H5_VERS_RELEASE in H5.c to avoid warning that unsigned comparison < 0 is always false, which is known, but for later versions the comparison can possibly be true. * Better solution for avoiding warning that unsigned comparison < 0.
* Added int variable for comparison with H5_VERS_RELEASE in H5.c to avoid (#866)Larry Knox2021-07-301-1/+3
| | | | warning that unsigned comparison < 0 is always false, which is known, but for later versions the comparison can possibly be true.
* Updates a comment in H5mpi.c (#863)Dana Robinson2021-07-281-1/+1
| | | MPI_Type_struct() is an MPI-1 call we no longer use.
* Remove version-check for relnum in H5check. (#812)Allen Byrne2021-07-272-93/+154
| | | | | | | | | | | | | | | | | | | | | | | | | * Remove version-check for relnum in H5check. * Add in release exception code check * Fix typos * Fix more typos * Rework comments * format change * format whitespace * Library version must be less than or equal to headers * Need the NOT version of the compare * Enable release+1 check for PASS * Add release note * Update note
* Fixes a couple of warnings: (#849)Dana Robinson2021-07-221-2/+2
| | | | | | | - warn_unused_result warnings from h5jamgentest and j5stat_gentest that were triggered by write(2) calls where we didn't check the return value - A missing callback in the NULL VOL connector - A possibly uninitialized property list pointer
* Avoid popping API context when one wasn't pushed (#848)jhendersonHDF2021-07-221-3/+16
|
* Allow parallel filters feature for comm size of 1 (#840)jhendersonHDF2021-07-201-3/+11
|
* Remove checks for winsock2.h (#813)Dana Robinson2021-07-062-10/+9
| | | | This is only needed on Windows and always available, so there's no need to check for it.
* Fixes parallel issues from recent C99 changes (#809)Dana Robinson2021-06-3018-240/+248
| | | | | | | * Fixes parallel issues from recent C99 changes * Adds MPE FUNC --> __func__ changes missed in earlier PRs * Even more missed FUNC --> __func__ macros
generated by cgit v0.12 at 2024-11-07 05:51:42 (GMT)