| Commit message (Collapse) | Author | Age | Files | Lines |
| |
|
|
|
|
|
|
| |
Updates Windows thread-safe code in H5TS.c to use _beginthread instead of CreateThread.
Tested on 64-bit Windows 7 with Visual Studio 2010 using CMake. Both 32- and 64-bit builds were tested.
|
|
|
|
| |
standard 2.8.6
|
|
|
|
|
|
|
| |
Back out r21782 while I figure out what the problem is with the change.
Tested on:
Daily tests... :-/
|
|
|
|
|
|
|
|
|
|
|
|
| |
Description:
When shrinking a chunked dataset, the library fills in the unused parts of
chunks that have been shrunk. The fill value buffer allocated for this purpose
had a maximum size of 1 MB, but the fill was performed in a single operation.
Therefore, if the amount of unused space in a chunk after being shrunk was
greater than 1 MB, the library would read off the end of the fill value buffer.
Changed the maximum fill buffer size to be equal to the chunk size.
Tested: durandal; jam, koala, heiwa (h5committest)
|
|
|
|
|
|
|
|
|
|
| |
Rearrange checks for reasons why we break collective I/O back to independent
I/O into "global" and "local" sections. We should try to minimize the checks
in the "local" section...
Tested on:
Mac OS X/32 10.7.2 (amazon) w/parallel
(too minor to require h5committest)
|
| |
|
| |
|
| |
|
|
|
|
|
|
| |
between ASCII and UTF8. I corrected it by adding a condition check in H5T_conv_s_s and H5T_conv_vlen to report an error under this situation.
Tested on jam, koala, linew.
|
| |
|
| |
|
|
|
|
| |
No test needed.
|
|
|
|
|
|
|
|
| |
the size of compound data type through H5Tset_size immedia
tely after the type was created. I fixed it in this commit.
Tested on jam, linew, and koala.
|
|
|
|
|
|
| |
OPTION command for solution folder and no packaging.
Tested: local linux
|
|
|
|
|
|
| |
needs some polish - the solution folder assignment should go closer to the target declaration and not all projects are grouped (parallel, c++, fortran, hl, and a few others).
Tested on Windows.
|
|
|
|
| |
Trivial change - tested on 64-bit Windows 7 with Visual Studio 2010.
|
|
|
|
| |
Trivial change - tested on jam and local Windows 7 (VS2010).
|
|
|
|
| |
Trivial change - tested on 64-bit Windows 7 with Visual Studio 2010.
|
|
|
|
|
|
| |
with the input parameter. Quiets warnings from Visual Studio 2008.
Trivial change but tested on 32-bit linux (jam), 64-bit linux (koala), 64-bit Solaris/Sun CC (linew), 64-bit Windows 7 (kituo) just to be safe.
|
|
|
|
|
|
| |
modification time format.
Trivial change - just made sure it compiled on Windows.
|
|
|
|
| |
Tested on Windows (minor change).
|
|
|
|
|
|
|
|
| |
Description:
An old patch was mistakenly committed in r21556. Replaced this fix with the
latest.
Tested: jam, koala, heiwa (h5committest)
|
|
|
|
|
|
|
|
|
|
| |
Recalculate the size of destination attribute message when the source and
destination versions are different during an object copy operation.
(Jira: HDFF-7718)
Tested on:
Mac OS X/32 10.7.2 (amazon) w/debug
(h5committest upcoming)
|
|
|
|
|
|
|
|
|
|
|
| |
Description:
When H5Ocopy failed due to attempting to expand an invalid reference in an
attribute (or at any point after the destination object header was inserted
into the cache), it failed to clean up correctly. Modified
H5O_copy_header_real to keep track of whether the object header was inserted
and take appropriate action on failure.
Tested: fedora (too minor for full commit test)
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Description:
Added new H5SL_TYPE_GENERIC skip list type, which uses void *'s as keys and a
client-supplied callback for key comparison. This was added to support the
upcoming "merge named datatype" feature for H5Ocopy, but may be used in other
places as well. Also added testing.
Also fixed a potential bug with the H5SL_TYPE_OBJ implementation, and added
testing for that.
Tested: jam, koala, heiwa (h5committest), durandal
|
|
|
|
|
|
| |
H5Omtime.c
Tested on local linux (minor change)
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
| |
Correct error in loading local heap prefix & data block from the file.
Sometimes the local heap's prefix could be loaded before the data block (e.g.
using H5Oget_info), but then when the data block was loaded later, the free
list information would get lost, causing the heap's size to grow larger than
necessary. This is Jira bug #HDFFV-7767
Tested on:
Mac OS X/32 10.7.2 (amazon) w/debug
(h5committest coming up)
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Description:
H5Ocopy could get confused when copying a named datatype containing an
attribute which used that named datatype as its datatype. This happened
because H5Ocopy would recurse into the attribute's datatype before the object
the attribute was in was fully copied (i.e. before the "post-copy" pass).
Modified H5Ocopy to avoid recursing before the post-copy step in this case.
Required many changes, including to how non-committed shared messages are
copied.
Tested: jam, koala, heiwa (h5committest); durandal
|
|
|
|
| |
is copied correctly deep in the H5P code. Tested on local linux (trivial change).
|
|
|
|
|
|
|
|
| |
Update tracing information with new types introduced in "actual I/O mode"
query routines.
Tested on:
bin/reconfigure on jam
|
|
|
|
|
|
| |
- Added comment about requiring _O_BINARY being set on Windows to avoid CR-LF issues.
Tested on Windows (minor changes).
|
|
|
|
|
|
|
|
| |
Tested on Linux, FreeBSD, Solaris, MacOSX to ensure they are found correctly and everything builds. No code currently relys on clock_gettime().
- Brought Windows and VMS H5pubconf.h files into sync with the Linux file.
Tested on Windows. Ray will test on VMS when this is pushed to the 1.8.x branch.
|
|
|
|
| |
Just checked to make sure it still compiled on Windows.
|
|
|
|
| |
pending a careful evaluation of enum conversion behavior.
|
| |
|
|
|
|
|
|
|
|
| |
sets the SEC2 driver, though it will still report H5FD_WINDOWS.
- Set svn:eol-style and svn:keyword properties on H5FDwindows.c/h.
Tested on 64- and 32-bit Windows 7 (doesn't affect non-Windows platforms).
|
|
|
|
|
|
|
|
|
|
| |
- Created typedefs and #defines to make POSIX I/O type-safe on Windows and true POSIX platforms.
- Added pre-checks and removed post-checks for POSIX I/O sizes. When the number of bytes to be sent exceeds the maximum return value, the operation is split into smaller subsets.
- General code tidying and cleanup.
Tested on: jam (64-bit linux), jam + threadsafe, jam + parallel, Windows 7 (64- and 32-bit), freedom (64-bt BSD).
|
| |
|
|
|
|
|
|
| |
instead of copying the pointer.
Tested on jam (simple change).
|
|
|
|
|
|
|
|
| |
correctly reflect how they respond to an input NULL string.
- H5MM_xstrdup() now reports memory allocation errors via the HDF5 error stack (was previously an assert).
Tested on local Windows via ctest (trivial change).
|
|
|
|
|
|
| |
H5O_move_msgs_forward() when writing attributes. h5committested.
Tests are checked into the performance suite.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Check in "actual I/O mode" feature to trunk. Will merge back to 1.8 branch
after it bakes over the weekend.
Tested on:
FreeBSD/32 8.2 (loyalty) w/gcc4.6, w/C++ & FORTRAN, in debug mode
FreeBSD/64 8.2 (freedom) w/gcc4.6, w/C++ & FORTRAN, in debug mode
Linux/32 2.6 (jam) w/PGI compilers, w/default API=1.8.x,
w/C++ & FORTRAN, w/threadsafe, in debug mode
Linux/64-amd64 2.6 (koala) w/Intel compilers, w/default API=1.6.x,
w/C++ & FORTRAN, in production mode
Solaris/32 2.10 (linew) w/deprecated symbols disabled, w/C++ & FORTRAN,
w/szip filter, w/threadsafe, in production mode
Linux/PPC 2.6 (heiwa) w/C++ & FORTRAN, w/threadsafe, in debug mode
Linux/64-ia64 2.6 (ember) w/Intel compilers, w/C++ & FORTRAN,
in production mode
Linux/64-amd64 2.6 (abe) w/parallel, w/FORTRAN, in debug mode
Mac OS X/32 10.6.8 (amazon) in debug mode
Mac OS X/32 10.6.8 (amazon) w/C++ & FORTRAN, w/threadsafe,
in production mode
|
| |
|
| |
|
|
|
|
|
|
| |
defines H5_HAVE_WIN32_API and H5_HAVE_VISUAL_STUDIO defines to use. These can be properly set during configurration.
Tested: windows and local linux - reviewed internally
|
| |
|
| |
|