| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
|
|
|
|
|
| |
a callback closed
a different ID in the same type. Added a new skiplist routine,
H5SL_try_free_safe, which iterates over items, freeing some of them, and which
intercepts and defers attempts to remove from the list outside of the main
iteration. Changed H5I_clear_type to use this function.
Tested: jam, koala, ostrich (h5committest); ummon
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
- Updated the preprocessor logic that defines the H5_ATTR_* macros. The Solaris
Studio compiler understands __attribute__ but not the specific attributes
that we use. The H5_ATTR_* macros are now defined as blank with that
compiler which significantly cleans up the warnings, making debugging
easier.
- Replaced a few lingering __attribute__ macros with H5_ATTR_* markup.
Tested on: h5committest
Solaris w/ Solaris Studio (emu)
|
| |
|
| |
|
| |
|
| |
|
| |
|
|
|
|
|
|
|
|
| |
this by modifying
metadata cache clients to clear unused memory in the buffer after serializing.
Tested: ummon; jam, koala, ostrich (h5committest)
|
| |
|
|
|
|
|
|
|
|
|
| |
Separate H5AC layer from using package-scoped pieces of the H5C layer,
moving from including H5Cpkg.h to H5Cprivate.h.
Tested on:
MacOSX/64 10.10.3 (amazon) w/serial, parallel & production
Linux/32 2.6.x (jam) w/serial & parallel
|
|
|
|
|
|
|
|
|
| |
Correct error with braces around macro (in production builds), also clean
up a few warnings.
Tested on:
MacOSX/64 10.10.3 (amazon) w/serial, parallel & production
Linux/32 2.6 (jam) w/serial & parallel
|
|
|
|
|
|
|
|
|
|
| |
Clean up parameters to H5C__flush_single_entry: accept cache entry instead
of address, and make new flag for deleting entry from skiplist on destroy
instead of a separate parameter to the routine.
Tested on:
MacOSX/64 10.10.3 (amazon) w/serial & parallel
Linux/32 2.6.x (jam) w/serial & parallel
|
|
|
|
|
|
|
|
| |
Split parallel metadata cache code into separate source code modules.
Tested on:
MacOSX/64 10.10.3 (amazon) w/serial & parallel
(too minor for h5committest)
|
|
|
|
|
|
|
|
|
|
| |
Bring 'fail if unknown always' object header message flag from the
avoid_truncate branch, strengthen error checking for unknown messages, and try
to catch object header code configurations at compile time (instead of runtime).
Tested on:
MacOSX/64 10.10.3 (amazon) w/serial & parallel
(h5committest forthcoming)
|
|
|
|
| |
Remove some commented lines.
|
|
|
|
| |
possibly all) daily testing. When the behavior is acceptable, the changes should be merged into the regular release script as a --revision option.
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Changes that allow the autotools to correctly deal with inline
functions.
- Inline markup schemes are now checked with AC_COMPILE_IFELSE
tests instead of AC_C_INLINE.
- The inline designator is now H5_INLINE instead of H5_inline.
This is now set in H5private.h instead of H5pubconf.h.
Tested on: h5committest
|
| |
|
| |
|
| |
|
| |
|
| |
|
| |
|
|
|
|
|
|
|
|
|
| |
Merge v3 metadata cache changes to trunk. Yay! :-)
Tested on:
MacOSX/64 10.10.3 (amazon) w/serial & parallel
Linux/32 2.6.x (jam) w/serial & parallel
(daily tested on branch)
|
|
|
|
| |
dataset. They were updated to match the output of the file. (HDFFV-637)
|
| |
|
|
|
|
|
|
|
| |
- Moving truncate to H5F_flush() where it was done a long time ago.
- Add a second H5AC_flush() call after the truncate is done to make sure we write the correct EOA.
tested with h5committest
|
|
|
|
|
|
|
|
|
| |
Clean up compiler warnings before merging in v3 metadata cache changes
from branch.
Tested on:
MacOSX/64 10.10.3 (amazon) w/serial & parallel
Linux/32 2.6.x (jam) w/serial & parallel
|
| |
|
| |
|
|
|
|
|
|
|
|
|
| |
Large code cleanups and optimizations, in preparation for the upcoming
v3 metadata cache merge.
Tested on:
MacOSX/64 10.10.3 (amazon) w/serial & parallel
Linux/32 2.6.x (jam) w/serial & parallel
|
|
|
|
|
|
| |
selection as duplicate points are not checked to determine
that it is really an "entire chunk". The bug appears from running the mdset.c test. This is a short-term fix for the problem.
|
| |
|
|
|
|
|
|
| |
is_finalized need to be checked before making any MPI calls.
tested with h5committest.
|
| |
|
| |
|
| |
|
|
|
|
|
|
|
| |
- Check for mpi is_initialized and is_finalized before making MPI calls in init/terminate of HDF5.
- Add a test to make sure we can make serial HDF5 calls from rank 0 after MPI_Finalize is called.
tested with h5committest
|
| |
|
| |
|
|
|
|
|
|
|
|
|
|
| |
Normalize trunk against the metadata_cache_merge branch, in preparation
for merging the branch into the trunk.
Tested on:
MacOSX/64 10.10.3 (amazon) w/serial & parallel
Linux/64 2.6.x (koala) w/serial
Linux/32 2.6.x (jam) w/serial & parallel
|
| |
|
|
|
|
|
|
|
|
| |
* H5_ATTR_FORMAT(X,Y,Z) __attribute__((format(X, Y, Z)))
- Rename UNUSED attribute characterstic to H5_ATTR_UNUSED.
- Rename NORETURN attribute characterstic to H5_ATTR_NORETURN
tested with h5committest.
|
|
|
|
|
|
| |
Remove H5T_INIT_INTERFACE from H5Tconv
Add macro in H5Tconv to check overflow when using H5T_CONV_uS
|
| |
|
| |
|
| |
|
|
|
|
| |
Tested: none -- comments only.
|
|
|
|
| |
usage from public routines.
|