| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
|
|
|
|
| |
Revert r25273, 25283 & 25439 (the hyperslab improvement changes). They
are buggy and it's taking me a long time to correct the problem. I'll check
in a revised form of the changes when I've got them straightened out.
Tested on:
Mac OSX 10.10.0 (amazon) w/gcc 4.9.x, C++, FORTRAN
Linux 2.6.x (jam) w/parallel
|
| |
|
| |
|
|
|
|
|
|
|
| |
but the test was not changed (still doing H5A_create(...) < 0).
Fixed the error by changing to compare against NULL.
Tested: ADA AIX machine where the old code was flagged as an error by the AIX compiler.
|
| |
|
| |
|
| |
|
|
|
|
|
|
|
|
| |
passes NULL to equally divide all the address space between all the members.
Before there was one chunk of the address space not being used.
tested h5committest, and multi vfd make check on jam.
|
| |
|
|
|
|
| |
Tested: platypus
|
|
|
|
|
|
|
|
|
|
| |
is passed
as NULL. The library is supposed in that case to equally divide the
address space among all members, but there was a bug causing an
overflow in the assignment.
tested with h5commitest
|
|
|
|
|
|
| |
have two opposing use cases, only the enum is common.
Tested: local linux and windows
|
| |
|
| |
|
|
|
|
| |
read/write to handle big counts.
|
|
|
|
|
|
|
|
| |
libraries. This will prevent 'dll' being appended to the tool name on windows. See HDFFV-8292.
Windows DLL linkage problem because of defines needed.
Tested: local linux
|
|
|
|
|
|
| |
libraries. This will prevent 'dll' being appended to the tool name on windows. See HDFFV-8292.
Tested: local linux
|
| |
|
| |
|
| |
|
|
|
|
| |
fix bugs where hid_t were assumed to be 32 bit ints or herr_t.
|
| |
|
|
|
|
| |
resize instead.
|
| |
|
|
|
|
|
|
| |
Make the fill value operation for chunked datasets in parallel collective/fast.
tested with h5committest
|
|
|
|
|
|
|
| |
H5Pset_core_write_tracking(). This ensures that the behavior
matches the description in the reference manual.
Tested on a local linux VM. This is a very minor change.
|
| |
|
|
|
|
| |
hid_t. Bug exposed in VOL branch.
|
|
|
|
| |
tested Jam. minor change.
|
|
|
|
|
|
|
|
|
| |
Merge 64-bit ID changes from branch to trunk. (Plus a few minor cleanups
that aren't on the branch)
Tested on:
Mac OSX/64 10.9.4 (amazon) w/C++ & FORTRAN
(h5committested on branch already for a week)
|
|
|
|
|
|
| |
will be needed if H5Pget_create_plist is called.
tested jam. minor change.
|
| |
|
|
|
|
| |
tested with h5committest.
|
|
|
|
|
|
|
|
| |
Initial tweaks to align with VOL branch.
Tested on:
Mac OSX/64 10.9.3 (amazon) w/gcc 4.9
(h5committest not required, too small)
|
|
|
|
|
|
|
|
| |
Minor tweaks to updated hyperslab code, before larger revision coming.
Tested on:
Mac OSX/64 10.9.3 (amazon) w/gcc 4.9 & parallel
(h5committest forthcoming)
|
| |
|
| |
|
| |
|
|
|
|
|
| |
MPI_File_set_size as they are supported by most MPI implementations.
- fix bug in t_mpi.c (HDFFV-8856)
|
| |
|
|
|
|
| |
allocating space for zero-length chunked dataset and asserting.
|
| |
|
| |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
uninitialized variable bug in passing.
Added the H5C_get_entry_ptr_from_addr() and H5C_verify_entry_type()
in H5C.c, along with their pass through functions in H5AC.c. Note
that H5C_get_entry_ptr_from_addr() is quite slimey, and should be
avoided if possible.
Updated fractal heap to use flush ordering to force children
in the hdr / iblock / dblock tree to be flushed prior to parents.
This is part of a port of the fractal heap to the V3 cache, although
I gather that it will also be useful for SWMR.
Added sanity checking code to the fractal heap to verify that direct
blocks are flushed before their parent indirect blocks, child indirect
block are flushed prior to their parent indirect blocks, and that the
root iblock or dblock is flushed prior to the header
Tested on:
Jam: Serial and parallel
Phoenix (Debian): serial, debug and production
amazon (MacOS): serial
|
| |
|
| |
|
|
|
|
|
|
|
|
|
|
| |
Correct error in new hyperslab selection test to use the right # of
dimensions for start/stride/count/block arrays. Also, switch H5Sselect_select
to H5Smodify_select.
Tested on:
MacOSX/64 10.9.3 (amazon) w/gcc 4.9.1, OpenMPI, C++, FORTRAN & threadsafe
(Too minor to require h5committest)
|
|
|
|
|
|
|
|
| |
when the thread-safe shared library is built on Windows.
Fixes HDFFV-8837.
Tested on 64-bit Windows 7 with VS2012.
|
|
|
|
|
|
|
|
|
|
|
| |
Bring in Chao/Neil/my changes to optimize hyperslab selection operations
further, along with 3 new public API routines: H5Scombine_hyperslab(),
H5Sselect_select() and H5Scombine_select(), along with many minor cleanups to
the code and fixing a few compiler warnings.
Tested on:
Mac OSX/64 10.9.3 w/gcc 4.9.x and parallel w/OpenMPI
(h5commttest forthcoming)
|