Update
Description:
A lot of modifications for the FPHDF5 stuff:
H5AC.c
H5ACprivate.h - Removed AC_find (it's replaced with AC_protect
and AC_unprotect). Added flushing if it's an FPHDF5 driver and
we're doing an AC_set or AC_unprotect with the dirty flag set.
H5B.c - Split up the B_flush function into different functions
since the one function was doing serialization which is better
left as a separate entity.
H5D.c - Removed some FPHDF5 code that was incorrect
H5F.c - Split up the F_flush function so that it no longer
allocates file space. Created new functions (F_init_superblock,
F_read_superblock, and F_write_superblock) for greater modularity
and so that the FPHDF5 non-captain processes can read the
superblock after the captain process writes it.
H5FD.c - Error message correction.
H5FDfphdf5.c - Removed MPI barrier call that wasn't needed.
H5FPclient.c
H5FPserver.c - Modified so that if a process requests data that
isn't exactly aligned, we can return it if we have the block that
contains the requested address.
H5G.c
H5Gent.c
H5Gnode.c
H5HL.c
H5HLpkg.h
H5HLprivate.h
H5Oefl.c - Removed the H5HL_peek function since it was doing a
(now unsafe) holding of the information in the cache. Replaced
with protect and unprotect calls.
H5TB.c - Error fix. The TB_dless function wasn't working
properly.
H5Gstab.c - Format change.
H5err.txt
H5Edefin.h
H5Einit.h
H5Epubgen.h
H5Eterm.h - Added new error code.
Platforms tested:
Modi4 (paralle, Fortran)
Sol (Fortran)
Linux (C++, Fortran)
Copper (Parallel, Fortran)
Misc. update:
Code cleanup
Description:
Cleanup H5Pclose code and also switch a few internal users of H5Pclose
to use H5I_dec_ref instead.
Platforms tested:
FreeBSD 4.8 (sleipnir)
too small for h5committest
Code cleanup & commentary
Description:
Add comments to the recent FPHDF5 additions to this file to describe why
the changes shouldn't be in this file and attempt to find the proper home for
them.
Platforms tested:
FreeBSD 4.8 (sleipnir)
just comments mostly, no need to run h5committest
Update
Description:
Added an extra flag to the clear functions that triggers a "destroy"
of the object being cleared if necessary. This is a fix for the
FPHDF5 stuff which had an object sticking around after it was
cleared. (In FPHDF5, some processes are in charge of destroying the
object, but all processes might actually allocate the object.)
Platforms tested:
AIX (Copper: Fortran & C++)
Linux (Verbena: Fortran & C++)
IRIX (Modi4: Parallel & Fortran)
(Sol is down)
Misc. update:
Bug fix
Description:
Object references were not getting written out correctly to the file, with
recent changes to their memory structure.
Solution:
Convert the object references correctly.
Platforms tested:
h5committested
Code cleanup
Description:
Changed lines like:
H5E_clear(H5E_get_my_stack());
to:
H5E_clear(NULL);
which performs the same operation.
Platforms tested:
h5committested
Bug fix and refactored code
Description:
Return the file space we have allocated already when a dataset creation
fails.
Also track the change to the H5Fget_obj_<foo> APIs
Platforms tested:
FreeBSD 4.8 (sleipnir)
h5committest
Bug fix
Description:
The fill time in a dataset with no fill value information created with an
older version of the library was getting set to H5D_FILL_TIME_ALLOC instead of
the new default H5D_FILL_TIME_IFSET and was causing H5Dcreate() calls with that
dataset creation property list to fail now.
Solution:
Set the new default in the fill time initialization for missing fill value
information.
Platforms tested:
FreeBSD 4.8 (sleipnir)
h5committest
New feature/Bug fix
Description:
Add new fill time value - H5D_FILL_TIME_IFSET which writes the fill value
to a dataset if the user has defined one, otherwise not writing the fill value
to the dataset.
Platforms tested:
FreeBSD 4.8 (sleipnir) serial & parallel
h5committest
Bug fix
Description:
Chunked datasets with early space allocation and unlimited dimensions were
running into problems where the dataset's "layout" message was marked as
constant too early, preventing the dataset's dimensions from being extended.
Solution:
Change logic for marking the layout message constant to wait a bit longer.
Platforms tested:
FreeBSD 4.8 (sleipnir) w/parallel
h5committest
Code cleanup
Description:
Limit the scope on more function prototypes/macros/typedefs.
Platforms tested:
FreeBSD 4.8 (sleipnir)
h5committest not necessary.
Bug fix.
Description:
When the same dataset is opened more than once and those references to the
object are left dangling, the library would not correctly shut down the dataset
API, potentially allowing the file to be truncated.
Solution:
"force" the dataset IDs to be released when the dataset API is being shut
down by the library.
Platforms tested:
FreeBSD 4.8 (sleipnir)
h5committested
Code cleanup.
Description:
Move many package or internal function prototypes and macro definitions
into tighter scope according to their current use.
Added more comments where appropriate.
Eliminate ancient, unused functions.
Added a couple "accessor" functions to get parts of data structures which
were moved out of scope.
Platforms tested:
h5committested
New feature/enhancement
Description:
Chunked datasets are handled poorly in several circumstances involving
certain selections and chunks that are too large for the chunk cache and/or
chunks with filters, causing the chunk to be read from disk multiple times.
Solution:
Rearrange raw data I/O infrastructure to handle chunked datasets in a much
more friendly way by creating a selection in memory and on disk for each chunk
in a chunked dataset and performing all of the I/O on that chunk at one time.
There are still some scalability (the current code attempts to
create a selection for all the chunks in the dataset, instead of just the
chunks that are accessed, requiring portions of the istore.c and fillval.c
tests to be commented out) and performance issues, but checking this in will
allow the changes to be tested by a much wider audience while I address the
remaining issues.
Platforms tested:
h5committested, FreeBSD 4.8 (sleipnir) serial & parallel, Linux 2.4 (eirene)
Code cleanup
Description:
Add & improve assertion that was removed in last commit, to check for
compact dataset storage.
Platforms tested:
FreeBSD 4.8 (sleipnir)
triple check not necessary.
New feature, sorta
Description:
Track changes to various internal APIs
Platforms tested:
FreeBSD 4.8 (sleipnir) w/C++
Linux 2.4 (burrwhite) w/FORTRAN
Solaris 2.7 (arabica) w/FORTRAN
IRIX64 6.5 (modi4) w/parallel & FORTRAN
(h5committest not run due to my ongoing difficulties with C++ on burrwhite).
New feature
Description:
Hooked up internal dataset creation call (H5D_create) to the new "can apply"
and "set local" callbacks for filters.
Solution:
Platforms tested:
FreeBSD 4.8 (sleipnir) w/szip
Linux 2.4 (sleipnir) w/szip
Solaris 2.7 (arabica) w/FORTRAN
IRIX64 6.5 (modi4) w/szip, FORTRAN & parallel
Misc. update:
Code cleanup
Description:
Clean up miscellaneous warnings which have crept into the code.
Fix "_POSIX_C_SOURCE not defined" warning on FreeBSD.
Adjust gcc compiler flags to be more concise for production mode.
Refactor the H5O code so that there is a stronger boundary between code
in the H5O package and code in the library which just calls H5O routines.
Platforms tested:
Tested h5committest {arabica (fortran), eirene (fortran, C++)
modi4 (parallel, fortran)}
FreeBSD 4.7 (sleipnir) serial & parallel and gcc 2.95.4 & gcc 3.2.2
Misc. update:
Update MANIFEST if you add or remove any file.
Bug Fix
Description:
Metadata cache in parallel I/O can cause hangs in applications which
perform independent I/O on chunked datasets, because the metadata cache
can attempt to flush out dirty metadata from only a single process, instead
of collectively from all processes.
Solution:
Pass a dataset transfer property list down from every API function which
could possibly trigger metadata I/O.
Then, split the metadata cache into two sets of entries to allow dirty
metadata to be set aside when a hash table collision occurs during
independent I/O.
Platforms tested:
Tested h5committest {arabica (fortran), eirene (fortran, C++)
modi4 (parallel, fortran)}
FreeBSD 4.7 (sleipnir) serial & parallel
Misc. update:
Updated release_docs/RELEASE
Bug fix
Description:
Currently, when the library encounters an object header message that isn't
know, it fails to open that object in the file.
Solution:
Allow the library to skip over the unknown object header message and
continue to process the remaining messages, in the hope that the skipped
message isn't important later. If it is important, it will be caught at
a higher level of the library.
Platforms tested:
FreeBSD 4.7 (sleipnir)
Code cleanup/new feature.
Description:
Split FUNC_LEAVE into API and non-API specific versions. This allows a
solution to compiling this branch with C++, as well as reducing the size
of the binaries produced.
Platforms tested:
FreeBSD 4.7 (sleipnir) w/serial, parallel (including MPE) & thread-safe
Lots of performance improvements & a couple new internal API interfaces.
Description:
Performance Improvements:
- Cached file offset & length sizes in shared file struct, to avoid
constantly looking them up in the FCPL.
- Generic property improvements:
- Added "revision" number to generic property classes to speed
up comparisons.
- Changed method of storing properties from using a hash-table
to the TBBT routines in the library.
- Share the propery names between classes and the lists derived
from them.
- Removed redundant 'def_value' buffer from each property.
- Switching code to use a "copy on write" strategy for
properties in each list, where the properties in each list
are shared with the properties in the class, until a
property's value is changed in a list.
- Fixed error in layout code which was allocating too many buffers.
- Redefined public macros of the form (H5open()/H5check, <variable>)
internally to only be (<variable>), avoiding innumerable useless
calls to H5open() and H5check_version().
- Reuse already zeroed buffers in H5F_contig_fill instead of
constantly re-zeroing them.
- Don't write fill values if writing entire dataset.
- Use gettimeofday() system call instead of time() system when
checking the modification time of a dataset.
- Added reference counted string API and use it for tracking the
names of objects opening in a file (for the ID->name code).
- Removed redundant H5P_get() calls in B-tree routines.
- Redefine H5T datatype macros internally to the library, to avoid
calling H5check redundantly.
- Keep dataspace information for dataset locally instead of reading
from disk each time. Added new module to track open objects
in a file, to allow this (which will be useful eventually for
some FPH5 metadata caching issues).
- Remove H5AC_find macro which was inlining metadata cache lookups,
and call function instead.
- Remove redundant memset() calls from H5G_namei() routine.
- Remove redundant checking of object type when locating objects
in metadata cache and rely on the address only.
- Create default dataset object to use when default dataset creation
property list is used to create datasets, bypassing querying
for all the property list values.
- Use default I/O vector size when performing raw data with the
default dataset transfer property list, instead of querying for
I/O vector size.
- Remove H5P_DEFAULT internally to the library, replacing it with
more specific default property list based on the type of
property list needed.
- Remove redundant memset() calls in object header message (H5O*)
routines.
- Remove redunant memset() calls in data I/O routines.
- Split free-list allocation routines into malloc() and calloc()-
like routines, instead of one combined routine.
- Remove lots of indirection in H5O*() routines.
- Simplify metadata cache entry comparison routine (used when
flushing entire cache out).
- Only enable metadata cache statistics when H5AC_DEBUG is turned
on, instead of always tracking them.
- Simplify address comparison macro (H5F_addr_eq).
- Remove redundant metadata cache entry protections during dataset
creation by protecting the object header once and making all
the modifications necessary for the dataset creation before
unprotecting it.
- Reduce # of "number of element in extent" computations performed
by computing and storing the value during dataspace creation.
- Simplify checking for group location's file information, when file
has not been involving in file-mounting operations.
- Use binary encoding for modification time, instead of ASCII.
- Hoist H5HL_peek calls (to get information in a local heap)
out of loops in many group routine.
- Use static variable for iterators of selections, instead of
dynamically allocation them each time.
- Lookup & insert new entries in one step, avoiding traversing
group's B-tree twice.
- Fixed memory leak in H5Gget_objname_idx() routine (tangential to
performance improvements, but fixed along the way).
- Use free-list for reference counted strings.
- Don't bother copying object names into cached group entries,
since they are re-created when an object is opened.
The benchmark I used to measure these results created several thousand
small (2K) datasets in a file and wrote out the data for them. This is
Elena's "regular.c" benchmark.
These changes resulted in approximately ~4.3x speedup of the
development branch when compared to the previous code in the
development branch and ~1.4x speedup compared to the release
branch.
Additionally, these changes reduce the total memory used (code and
data) by the development branch by ~800KB, bringing the development
branch back into the same ballpark as the release branch.
I'll send out a more detailed description of the benchmark results
as a followup note.
New internal API routines:
Added "reference counted strings" API for tracking strings that get
used by multiple owners without duplicating the strings.
Added "ternary search tree" API for text->object mappings.
Platforms tested:
Tested h5committest {arabica (fortran), eirene (fortran, C++)
modi4 (parallel, fortran)}
Other platforms/configurations tested?
FreeBSD 4.7 (sleipnir) serial & parallel
Solaris 2.6 (baldric) serial
Purpose:
New feature to H5Dget_offset
Description:
If user block is set, H5Dget_offset should be able to return the absolute
offset from the beginning of file.
Platforms tested:
eirene, arabica
Code Movement
Description:
Moved definition of the H5D_t data structure into the H5Dprivate.h
header file. With the FPHDF5 stuff, we need access to this data
structure.
Platforms tested:
Eirene...change small, so no need for all three tests.
Code cleanup
Description:
Cleaned up some error values that were incorrect, added some assertions
and refactored the integer & floating-point native conversions some.
Platforms tested:
FreeBSD 4.7 (sleipnir), too small for h5committest
Purpose:
New function.
Description:
H5Dget_offset returns the offset of a dataset's data relative to the
beginning of a file.
Platforms tested:
arabica(simple function, one test should be enough.)
Misc. update:
RELEASE.txt
New Feature. Support for new FPHDF5 Feature
Description:
More support for the FPHDF5 feature. This splits the H5D_create
function into separate parts which update the metadata cache. It was
necessary to split apart the H5O_create function to do a similar
thing since it would allocate real space on the hard disk.
I'm checking this up now so that I don't get too far away from what
the CVS repository has and so that others may look at the code and
retch^H^H^H^H^Hmarvel at it. If there are any comments, send them my
way.
There is one outstanding issue, though. I need to use non-API
functions in the FPHDF5 stuff. I'm using some APIs right now and
should migrate to using others. At the moment, I'm putting that on
the back burner until I'm more along in the implementation.
Platforms tested:
Arabica (C++)
Eirene (Parallel)
Modi4 (Parallel)
Another Oops
Description:
I accidentally checked in the H5D.c module when I shouldn't have.
Solution:
Rolled back to 1.281 version
Platforms tested:
Linux
Oops
Description:
The H5FPprivate.h file was being #included even if PARALLEL was
turned off.
Solution:
put the #include of the H5FPprivate.h file within the #ifdef block so
that it doesn't get included if H5_HAVE_FPHDF5 isn't defined.
Platforms tested:
Linux...
Code cleanup/Bug Fix
Description:
Re-do the ID->name code again, taking into account many more weird and
wonderful special cases stumbled across during the last set of test
writing.
Platforms tested:
FreeBSD 4.6 (sleipnir) w and w/o parallel
Linux 2.2.x (eirene) w/FORTRAN & C++
Solaris 2.7 (arabica) w/FORTRAN
IRIX64 6.5 (modi4) w/FORTRAN & parallel
Code cleanup
Description:
Clean up ID->name code:
- Reformat to better match library coding standard
- Changed several algorithms to be more efficient
- Integrated into library more smoothly
Platforms tested:
eirene w/FORTRAN & C++
arabica w/FORTRAN
modi4 w/FORTRAN & parallel
sleipnir
Purpose:
bug fix
Description:
memory leak regarding the ID to name buffer
Solution:
added a new function H5G_free_ent_name that is called on several places of the library
Platforms tested:
windows 2000
linux, with cpp
solaris, with fortran, cpp
irix64, with parallel, fortran
API name change
Description:
Change all "space time" references to "alloc time", including API functions
and macro definitions, etc.
Platforms tested:
FreeBSD 4.6 (sleipnir) w/C++
Solaris 2.7 (arabica) w/FORTRAN
IRIX64 6.5 (modi4) w/parallel & FORTRAN
Bug Fix
Description:
Some macros had "NULL" instead of "FAIL" as the return value.
Solution:
Change from NULL to FAIL.
Platforms tested:
Eirene (C++), Arabica (Fortran), Modi4 (pp and Fortran).
Purpose:
Added 'ID to name' support
Description:
There is a new API function H5Iget_name
Most of the changes are on H5G.c , regarding the symbol table entry struct H5G_entry_t
which has 2 new fields 'name' and 'old_name'
A new private function was introduced H5G_ent_copy, that does a deep copy
between 2 symbol table entries
The test file is getname.c
Platforms tested:
windows 2000, Linux, Solaris
