| Commit message (Collapse) | Author | Age | Files | Lines |
| |
|
| |
|
| |
|
|
|
|
|
|
|
| |
* Replace H5detect's build-time detection of C99 integer properties with a
table-driven routine, `H5T__init_native_int()`, that is run at library
initialization time.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* Do not conditionally compile code that uses a pointer dereference
and assignment to copy a potentially unaligned variable to aligned
automatic storage, or vice versa. Instead, always use naked `memcpy(3)`s.
Disassembling the generated code reveals that the `memcpy(3)`s optimize
(`-O3`) to a single `mov` instruction for x86_64, which is not strict
about alignment.
This change reduces the size of code and scripts by 143 lines, eases
our way to cross-compilation, and avoids invoking undefined behavior.
* Committing clang-format changes
* Per discussion, use HD and add comments.
Co-authored-by: github-actions <41898282+github-actions[bot]@users.noreply.github.com>
|
| |
|
|
|
|
|
|
| |
H5Pget_fapl_mpio() should return an MPI info object containing all the
MPI-IO hints used by the MPI library underneath, after the file is
opened. Some hints, such as cb_nodes (number of I/O aggregators), are
useful for HDF5 applications and I/O libraries built on top of HDF5.
|
| |
|
| |
|
| |
|
|
|
|
|
| |
* Removes the STATIC flavor of FUNC_ENTER macros
|
| |
|
| |
|
| |
|
|
|
|
|
|
|
|
|
| |
(version, subversion) (#1556)
* Fix MPI version check for MPI_Aint_diff usage
* Add H5_CHECK_MPI_VERSION macro
Adds new H5_CHECK_MPI_VERSION macro to check for MPI library version/subversion pair of at least the specified pair
|
| |
|
| |
|
| |
|
| |
|
|
|
|
| |
* Unnecessary extern C guards in cache headers
* Non-existent H5Xpublic.h includes hidden behind NOT_YET symbols
|
| |
|
| |
|
|
|
|
|
|
|
|
|
|
| |
From Vailin:
Fix the FSM bug when setting the FSM threshold to a non-default value.
Check for smaller or larger section size after merging and shrinking a
section, for this case is the section that is smaller than threshold
(see H5MF_xfree() in H5MF.c). It is possible for the section to be
smaller after merging/shrinking (see H5MF__sect_large_shrink() in
H5MFsection.c).
|
| |
|
| |
|
| |
|
|
|
|
|
|
| |
to MANIFEST (#1460)
|
| |
|
| |
|
| |
|
| |
|
|
|
| |
Co-authored-by: Larry Knox <lrknox@hdfgroup.org>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* Committing clang-format changes
* Spelling of preceed was corrected to proceed, but should have been
corrected to precede.
* Correct spelling correction of 'preceed' incorrectly to 'proceed'. It should be 'precede'.
* OESS-168: Remove clang warnings.
* OESS-168: Address @lrknox and @gnuoyd reviews.
* Eliminate clang warnings listed in PR #1310 without adding new ssize_t
variables.
* Committing clang-format changes
* Add H5_ATTR_UNUSED to wait_for_child call.
Remove unneeded casts in mirror_log calls.
* Keep ssize_t in mirror_server.c line 479.
Co-authored-by: github-actions <41898282+github-actions[bot]@users.noreply.github.com>
Co-authored-by: Hyo-Kyung Lee <hyoklee@hdfgroup.org>
|
| |
|
|
|
|
|
|
|
|
| |
* Fix issue with copying null new references. Fix assertion failure when
reference copying fails.
* Committing clang-format changes
Co-authored-by: github-actions <41898282+github-actions[bot]@users.noreply.github.com>
|
|
|
|
|
|
|
|
|
|
| |
* Implement big I/O support for independent reads. Add test for this.
* Committing clang-format changes
* Update big I/O code in H5FD__mpio_write() to match that in
H5FD__mpio_read().
Co-authored-by: github-actions <41898282+github-actions[bot]@users.noreply.github.com>
|
|
|
|
|
| |
(#1358)" (#1437)
This reverts commit 01092658a3095c31d7dc1ed1beebbd965095c244.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* Implement H5ESget_requests() to retrieve request pointers and
corresponding connector IDs from an event set. Add tests for this
function.
* Add "order" parameter to H5ESget_requests, to allow the user to specify
the order in which requests are returned.
* Fix bugs with H5ESinsert_request()
* Change H5ESget_requests() API to have separate input and output
parameters for the allocated array length(s) (in) and the number of events in
the event set (out).
* Fix issue with H5_now_usec() with 32 bit systems. Also improve event
set test.
* Committing clang-format changes
* Add comments to H5_now_usec() explaining rationale for uint64_t casts.
Co-authored-by: github-actions <41898282+github-actions[bot]@users.noreply.github.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
When resizing a pinned/protected cache entry, the metadata
cache code previously would wait until after resizing the
entry to attempt to log the newly-dirtied entry. This would
cause H5C_resize_entry to mark the entry as dirty and make
H5AC_resize_entry think that it doesn't need to add the
newly-dirtied entry to the dirty entries skiplist.
Thus, a subsequent H5AC__log_moved_entry would think it
needs to allocate a new entry for insertion into the dirty
entry skip list, since the entry doesn't exist on that list.
This causes an assertion failure, as the code to allocate a
new entry assumes that the entry is not dirty.
|
| |
|
| |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* Replaced many uses of sprintf with safer snprintf
Many very straightforward, but in a few cases added a length parameter to some private functions, because buffer length was otherwise unknowable.
* Removed unnecessary use of static on small buffers
This improves thread safety.
* Committing clang-format changes
Co-authored-by: github-actions <41898282+github-actions[bot]@users.noreply.github.com>
|
|
|
|
|
|
|
|
|
| |
* Remove const from the argv tools/tests main sig.
* also remove const from H5_get_option and parse_command_line.
* Committing clang-format changes
Co-authored-by: github-actions <41898282+github-actions[bot]@users.noreply.github.com>
|
|
|
|
|
|
|
| |
* Replaced several uses of sprintf with safer snprintf
* Committing clang-format changes
Co-authored-by: github-actions <41898282+github-actions[bot]@users.noreply.github.com>
|
| |
|
|
|
|
|
|
|
|
| |
Commit 61464d7e was a fix for a compiler error on OpenBSD where
int_fast8_t is in fact larger than 1 byte. However, the changes broke
H5detect.c generation of H5Tinit.c on Power8 big-endian, and probably on
all big-endian machines. Switching DETECT_BYTE to DETECT_I for 8 byte types doesn't work on our big-endian Power8 machine because switching the last parameter of DETECT_I_BYTE_CORE from "int" to the 8 byte type incorrectly sets dt->shared->u.atomic.order = H5T_ORDER_LE on a big-endian machine. A different fix to accomodate both big-endian and OpenBSD is needed. See commit 61464d7e for OpenBSD info.
Co-authored-by: github-actions <41898282+github-actions[bot]@users.noreply.github.com>
|
|
|
|
|
|
|
| |
* Fix several places where rank 0 can skip past collective MPI operations on failure
* Committing clang-format changes
Co-authored-by: github-actions <41898282+github-actions[bot]@users.noreply.github.com>
|
| |
|
|
|
|
|
|
| |
(#1355)
Co-authored-by: github-actions <41898282+github-actions[bot]@users.noreply.github.com>
|