summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* [svn-r5440] Purpose:Quincey Koziol2002-05-2014-87/+54
| | | | | | | | | | | New feature Description: Add 'closing' parameter to H5FDflush and VFL "flush" functions, per http://hdf.ncsa.uiuc.edu/RFC/VFLFlush/VFLFlush.html Platforms tested: IRIX64 6.5 (modi4)
* [svn-r5439] Purpose:Quincey Koziol2002-05-204-28/+29
| | | | | | | | | | | Code cleanup Description: Change variable casts to use H5_ASSIGN_OVERFLOW, or check for overflow before the cast with H5_CHECK_OVERFLOW. Platforms tested: FreeBSD 4.5 (sleipnir)
* [svn-r5437] Pedro Vicente Nunes2002-05-187-22/+33
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Purpose: eliminate warnings on win32 Description: some conversion warnings were being issued Solution: added some type casts Platforms tested: w2000, linux changed the HDmemset to include a cast in win32; this was just to eliminate a compiler warning. probably the macro can also be used in unix #ifdef WIN32 #define HDmemset(X,C,Z) memset((void*)(X),C,Z) #else /* WIN32 */ #define HDmemset(X,C,Z) memset(X,C,Z) #endif /* WIN32 */ the list of previous warnings was D:\disk_w\hdf5\src\H5FDstdio.c(659) : warning C4244: 'initializing' : conversion from 'unsigned __int64 ' to 'unsigned int ', possible loss of data D:\disk_w\hdf5\src\H5Fcontig.c(435) : warning C4244: '=' : conversion from 'unsigned __int64 ' to 'unsigned int ', possible loss of data D:\disk_w\hdf5\src\H5Fcontig.c(497) : warning C4244: '=' : conversion from 'unsigned __int64 ' to 'unsigned int ', possible loss of data D:\disk_w\hdf5\src\H5Fcontig.c(915) : warning C4244: '=' : conversion from 'unsigned __int64 ' to 'unsigned int ', possible loss of data D:\disk_w\hdf5\src\H5Fcontig.c(982) : warning C4244: '=' : conversion from 'unsigned __int64 ' to 'unsigned int ', possible loss of data D:\disk_w\hdf5\src\H5Shyper.c(912) : warning C4244: '=' : conversion from 'unsigned __int64 ' to 'unsigned int ', possible loss of data D:\disk_w\hdf5\src\H5Shyper.c(995) : warning C4244: '=' : conversion from 'unsigned __int64 ' to 'unsigned int ', possible loss of data D:\disk_w\hdf5\src\H5Shyper.c(1936) : warning C4244: '=' : conversion from 'unsigned __int64 ' to 'unsigned int ', possible loss of data D:\disk_w\hdf5\src\H5Shyper.c(2019) : warning C4244: '=' : conversion from 'unsigned __int64 ' to 'unsigned int ', possible loss of data D:\disk_w\hdf5\src\H5Shyper.c(2862) : warning C4244: '=' : conversion from 'unsigned __int64 ' to 'unsigned int ', possible loss of data D:\disk_w\hdf5\src\H5Shyper.c(2864) : warning C4244: '=' : conversion from 'unsigned __int64 ' to 'unsigned int ', possible loss of data D:\disk_w\hdf5\src\H5Shyper.c(2948) : warning C4244: '=' : conversion from 'unsigned __int64 ' to 'unsigned int ', possible loss of data D:\disk_w\hdf5\src\H5Shyper.c(3690) : warning C4244: '=' : conversion from 'unsigned __int64 ' to 'unsigned int ', possible loss of data D:\disk_w\hdf5\src\H5Shyper.c(3692) : warning C4244: '=' : conversion from 'unsigned __int64 ' to 'unsigned int ', possible loss of data D:\disk_w\hdf5\src\H5Shyper.c(3776) : warning C4244: '=' : conversion from 'unsigned __int64 ' to 'unsigned int ', possible loss of data D:\disk_w\hdf5\src\H5Shyper.c(5167) : warning C4244: '+=' : conversion from '__int64 ' to 'unsigned int ', possible loss of data D:\disk_w\hdf5\src\H5Tvlen.c(371) : warning C4244: '=' : conversion from 'unsigned __int64 ' to 'unsigned int ', possible loss of data
* [svn-r5436] Pedro Vicente Nunes2002-05-184-4/+0
| | | | | | | | | Purpose: code clean up Description: removed some debugging comments regading the code warrior port Platforms tested: w2000
* [svn-r5435] Purpose:Quincey Koziol2002-05-171-0/+4
| | | | | | | | | | Document Code improvement below: Description: Propagated the "fill time" property into the parallel chunk allocation routine, allowing it to avoid writing fill values to each new chunk allocated. This improves the performance of chunked datasets in parallel I/O to be on par with contiguous datasets again (on modi4).
* [svn-r5434] Purpose:Quincey Koziol2002-05-172-52/+78
| | | | | | | | | | | | | Code improvement. Description: Propagated the "fill time" property into the parallel chunk allocation routine, allowing it to avoid writing fill values to each new chunk allocated. This improves the performance of chunked datasets in parallel I/O to be on par with contiguous datasets again (on modi4). Platforms tested: IRIX64 6.5 (modi4)
* [svn-r5433] Purpose:Quincey Koziol2002-05-171-1/+1
| | | | | | | Code cleanup Description: Cleaned up the function header comment for H5Pset_fill_time.
* [svn-r5430] Purpose:Quincey Koziol2002-05-171-4/+8
| | | | | | | | | | | | | | | | | | Document Bug fix/Code improvement below: Description: Currently, the chunk data allocation routine invoked to allocate space for the entire dataset is inefficient. It writes out each chunk in the dataset, whether it is already allocated or not. Additionally, this happens not only when it is created, but also anytime it is opened for writing, or the dataset is extended. Worse, there's too much parallel I/O syncronization, which slows things down even more. Solution: Only attempt to write out chunks that don't already exist. Additionally, share the I/O writing between all the nodes, instead of writing everything with process 0. Then, only block with MPI_Barrier if chunks were actually created.
* [svn-r5429] Purpose:Quincey Koziol2002-05-175-130/+264
| | | | | | | | | | | | | | | | | | | | | Bug fix/Code improvement. Description: Currently, the chunk data allocation routine invoked to allocate space for the entire dataset is inefficient. It writes out each chunk in the dataset, whether it is already allocated or not. Additionally, this happens not only when it is created, but also anytime it is opened for writing, or the dataset is extended. Worse, there's too much parallel I/O syncronization, which slows things down even more. Solution: Only attempt to write out chunks that don't already exist. Additionally, share the I/O writing between all the nodes, instead of writing everything with process 0. Then, only block with MPI_Barrier if chunks were actually created. Platforms tested: IRIX64 6.5 (modi4)
* [svn-r5428] Binh-Minh Ribler2002-05-164-2/+81
| | | | | | | | | | | | | | | Purpose: New Feature - per library change Description: Added the new member function getMemberIndex to classes EnumType and CompType to match the new C API H5Tget_member_index. Given the name of a member of an enumeration or compound datatype, this new function queries the index of the member. Platforms: SunOS 5.7 (arabica) Linux 6.2 (eirene)
* [svn-r5427] Purpose:Binh-Minh Ribler2002-05-162-3/+48
| | | | | | | | | | | | | | | | | | New Feature - per library change Description: Added new member function DataSet::fillMemBuf for the new C API H5Dfill. Quincey's description of H5Dfill is as followed: Added new H5Dfill() routine to fill the elements in a selection for a memory buffer with a fill value. This is a user API wrapper around some internal routines which were needed for the fill-value modifications from Raymond as well as Pedro's code for reducing the size of a chunked dataset. Platforms: SunOS 5.7 (arabica) Linux 6.2 (eirene)
* [svn-r5426] Purpose:Albert Cheng2002-05-163-29/+79
| | | | | | | Folded the added feature of setting MPI Info object from v1.4 to this branch. Platforms tested: modi4
* [svn-r5422] Description:Albert Cheng2002-05-142-2/+2
| | | | | | | Corrected typos in the FUNC_ENTER macros of H5FD_sec2_flush and H5FD_log_flush. Platforms tested: modi4 (just tested it compiled fine.)
* [svn-r5419] Purpose:Quincey Koziol2002-05-141-0/+11
| | | | | | | | | | | Bug fix. Description: Added barrier to flush routine to prevent race condition where file could be truncated. Platforms tested: IRIX64 6.5 (modi4)
* [svn-r5418] Purpose:Quincey Koziol2002-05-141-30/+26
| | | | | | | | | | | | | | | | Code cleanup/Performance enhancement Description: The code to extend the file size in H5FD_mpio_flush is getting run even when the file size doesn't change. Also, it's sort of sidestepping MPI-I/O when extending the file, instead of using MPI-I/O features to set the file's size. Solution: Only extend the file's size when the allocated size has increased. Also use MPI_File_set_size() to change the file's size. Platforms tested: IRIX64 6.5 (modi4)
* [svn-r5415] Purpose:Quincey Koziol2002-05-141-0/+2
| | | | Document Bug Fix
* [svn-r5414] Purpose:Quincey Koziol2002-05-141-0/+8
| | | | | | | | | | | | | | | Bug Fix Description: When H5Freopen is called, the file to reopen's 'intent' (read/write permissions, etc) is not being copied to the new file's 'intent' information. Solution: Copy it. :-) Platforms tested: IRIX64 6.5 (modi4)
* [svn-r5412] Purpose:Quincey Koziol2002-05-141-17/+50
| | | | | | | | | | | | | | | | Bug fix Description: Calling MPI_Get_count needs to be done with the same MPI type as was used for the transfer and we are always using MPI_BYTE, even when a different MPI type was used for the transfer. Solution: Only query MPI_Get_count with MPI_BYTE when we really used MPI_BYTE for the transfer. Wait for later to query MPI_Get_count with other MPI types. Platforms tested: IRIX64 6.5 (modi4)
* [svn-r5409] Purpose:Quincey Koziol2002-05-131-0/+2
| | | | Document Performance enhancement
* [svn-r5408] Purpose:Quincey Koziol2002-05-133-2/+48
| | | | | | | | | | | | | | | Performance enhancement Description: Doing an MPI_File_sync() just before a file is closed causing a large performance loss. Solution: Add flag to MPI file driver to avoid performance the MPI_File_sync() when the flag is set before a call to H5F_flush(). Platforms tested: IRIX64 6.5 (modi4)
* [svn-r5407] Purpose:Albert Cheng2002-05-133-26/+34
| | | | | | | | | | | | | | | | | | Bug fix Description: Was not able to handle data size (file size) larger than 32bits. Was using long, which is only 4 bytes big in SP, thus overflowing into negative when trying to address 2GB or larger. Solution: Changed those variables involved in file size/offset calculation to type off_t. (If a certain system/compiler has off_t defined as 4 bytes, it can't write to file size larger than 2GB anyway.) Note that the lseek of SP with -D_LARGE_FILE still fails for offset larger than 2GB (works for 2GB). That has to be fixed soon. Platforms tested: burrwhite (linux 2.4) and modi4 parallel.
* [svn-r5406] Purpose:Quincey Koziol2002-05-131-120/+121
| | | | | | | | | | Code cleanup Description: Restructure H5F_close code to use just one H5F_flush call. Platforms tested: IRIX64 6.5 (modi4)
* [svn-r5404] Purpose:Quincey Koziol2002-05-132-12/+0
| | | | | | | Back out change Description: Back out description of VFL 'flush' change.
* [svn-r5403] Purpose:Quincey Koziol2002-05-1313-59/+44
| | | | | | | | | | Back out changes. Description: Back out changes to VFL 'flush' API function, pending review. Platforms tested: IRIX64 6.5 (modi4)
* [svn-r5401] Elena Pourmal2002-05-131-15/+0
| | | | | | | | | Purpose: Code clean-up Description: Removed unused HAVE_XXXX macros from the H5f90i.h file. Platforms tested: HP-UX 11.00 (kelgia) and IRIX64-6.5 (modi4)
* [svn-r5399] Purpose:Quincey Koziol2002-05-111-1/+1
| | | | | | | | | | Bug Fix Description: Missed adding the 'closing' parameter to one 'flush' function. Platforms tested: FreeBSD 4.5 (sleipnir)
* [svn-r5397] Purpose:Bill Wendling2002-05-101-3/+3
| | | | | | | | | Bug Fix Description: Put quotes around some variables so that the shell script doesn't complain to us about "too many arguements" for a test. Platforms tested: Linux
* [svn-r5396] Purpose:Bill Wendling2002-05-101-3/+3
| | | | | | | | | | Bug Fix Description: Placed quotes around strings which could be empty so that the shell script doesn't complain to us about "too many arguments for test" problems. Platforms tested: Linux
* [svn-r5395] Purpose:Bill Wendling2002-05-101-0/+9
| | | | | | | Update Description: Added documentation on how you can install in a different directory than the one you specified during configuration.
* [svn-r5394] Purpose:Quincey Koziol2002-05-102-1/+13
| | | | Document new VFL flush parameter.
* [svn-r5393] Purpose:Quincey Koziol2002-05-1013-46/+60
| | | | | | | | | | | | | | | | | | | | | | New Feature Description: The VFL flush function is called immediately before a file is closed. This can cause duplicate syncronization actions to occur, if the VFL close function also performs them. Solution: Added 'closing' parameter to VFL 'flush' operation. This allows the VFL flush function to bypass operations that will be duplicated within the VFL close function. Additionally, use the 'closing' parameter to bypass calls to MPI_File_sync() when set. Since MPI_File_close() also syncronizes the file, this avoids the terrible performance hit taken when calling MPI_File_sync() as the file is closing. Platforms tested: IRIX64 6.5 (modi4)
* [svn-r5390] Purpose:Quincey Koziol2002-05-107-99/+11
| | | | | | | | | | | | | | | | | | | | | | Code cleanup Description: The parallel I/O file driver is optimized to only write metadata with one process (and broadcast the results to the other processes). This is currently enabled by a separate call to H5FD_mpio_tas_allsame() before each metadata write to the file. This can easily lead to problems where the prelude function call is omitted before the actual write code or, in a threaded environment, lead to race condititions where the value set is reset before being used. Solution: Since we only want to write metadata from one process, key off of the 'type' parameter (which has information about whether the data being written it metadata or raw data) to H5FD_mpio_write() as the method for determining whether to only write from one process or not. Platforms tested: IRIX64 6.5 (modi4)
* [svn-r5389] Purpose:Quincey Koziol2002-05-101-0/+2
| | | | Update release notes about rotating metadata writes.
* [svn-r5386] Purpose:Quincey Koziol2002-05-101-4/+2
| | | | | | | | | | | | | | | | Bug Fix Description: Currently, only process 0 is writing attribute data to a file. This is incorrect, because the raw data for attributes is cached in memory until the object header is written and other processes are not able to read the correct attribute information. Solution: Have all processes participate in writing the attribute data. Platforms tested: IRIX64 6.5 (modi4)
* [svn-r5385] Purpose:Quincey Koziol2002-05-101-4/+21
| | | | | | | | | | | | | | | New Feature Description: Currently, only process 0 writes metadata to disk, leading to a potential performance bottleneck as the other processors wait for it to catch up. Solution: Rotate the metadata responsibilities among all processes, speading out the work. Platforms tested: IRIX64 6.5 (modi4)
* [svn-r5383] Purpose:Bill Wendling2002-05-091-0/+12
| | | | | | Update Description: Updated how to compile HDF5 with Intel compilers (ecc or icc).
* [svn-r5381] Purpose:Quincey Koziol2002-05-091-32/+19
| | | | | | | | | | | Small code cleanup/improvement Description: Query the MPI rank only once and store it in the file structure for each process. Platforms tested: IRIX64 6.5 (modi4)
* [svn-r5378] Frank Baker2002-05-072-6/+24
| | | | | | | | | | | | | | | | Purpose: Update h5dread/write_f parameters Description: fortran/h5d_FORTRAN.html h5dread/write_f -- in the 'reference' versions, which yesterday's revisions missed): Added new overloading to dims parameter description. Corrected prior dims type and noted that it is deprecated and will be removed in 1.6. fortran/h5a_FORTRAN.html Updated "Last modified" tag to reflect yesterday's revisions. Platforms tested: IE 5
* [svn-r5376] Description:Albert Cheng2002-05-071-0/+46
| | | | | | | Added the calculation and printing (when debug >=3) of the MPI_write/read statistics. Platforms tested: modi4
* [svn-r5375] Purpose:Albert Cheng2002-05-071-2/+2
| | | | | | Changed all API to use MPI_WTIME timer. Just to be consistent. Platforms tested: modi4
* [svn-r5374] Purpose:Bill Wendling2002-05-071-3/+3
| | | | | | | | | Name Change Description: Changed the macro's name from "MILLISECOND" to "MICROSECOND" since, as Quincey pointed out, that's what it is ;). Platforms tested: Linux
* [svn-r5373] Purpose:Albert Cheng2002-05-071-1/+1
| | | | | | | | | Bug fix Description: The system timer forgot to add previous cummulated time. Fixed. Platforms tested: modi4
* [svn-r5371] Frank Baker2002-05-072-4/+49
| | | | | | | | | | | | | | | Purpose: Bugfix -- Closes #702 Description: Commented out references to and descriptions of unimplemented H5S functions. Particularly in the UG document, Dataspaces.html. H5Sopen H5Slock H5Ssubspace H5Scommit H5Sselect_op H5Ssubspace_name H5Sis_subspace H5Sselect_order H5Ssubspace_location Added explanatory "Note to developers" in comments near top of Dataspaces.html. Platforms tested: IE 5
* [svn-r5369] Purpose:Quincey Koziol2002-05-071-4/+0
| | | | | | | | | | Code cleanup Description: Remove incorrect MPI_Barrier() call in data I/O loop. Platforms tested: IRIX64 6.5 (modi4)
* [svn-r5368] Purpose:Albert Cheng2002-05-074-5/+61
| | | | | | | | | | Bug fix, feature Description: Added code to time pure MPI_File_write_xxx and read_xxx routines. Moved the setting of the timer_g to when timer is created (new) and destroyed. Platforms tested: modi4pp
* [svn-r5367] Purpose:Bill Wendling2002-05-065-136/+141
| | | | | | | | | | | | | | | | | Modification Description: Put some of the variables into the global space so that random functions can access them. (The output and timer variables which point to the output file and timer resp.) Placed the debug level 4 prints into the timer module so that it's easier to automatically put timer start/stop printfs for new timers which are created. Added timers for MPI Time calls (called HDF5_MPI_WRITE and HDF5_MPI_READ). Platforms tested: Linux
* [svn-r5366] Frank Baker2002-05-062-18/+34
| | | | | | | | | | | Purpose: Update description of h5dread/write_f and h5aread/write_f dims parameter. Description: h5dread/write_f, h5aread/write_f Added new overloading to dims parameter description. Noted that prior dims type is deprecated and will be removed in 1.6. Platforms tested: IE 5
* [svn-r5364] Elena Pourmal2002-05-061-1/+12
| | | | | | | | Purpose: Maintenance Description: Added information about Parallel Fortran Support for HP-UX 11.00 SysV and write/read overloaded subroutines (bug #670)
* [svn-r5358] MuQun Yang2002-05-061-4/+4
| | | | | | | | | | | | Purpose: bug fix Description: In H5TB.c, previously use an unsigned variable to represent address difference that is maybe a negative value. This will cause a potential bug in the future when this section of code is run. Solution: use H5F_address_cmp macro to compare two addresses. Platforms tested: linux 2.2.18
* [svn-r5348] Purpose:Bill Wendling2002-05-033-7/+6
| | | | | | | | | | | Bug Fix Description: There was a duplicate declaration of print_indent() in the header and in the source files. Solution: Removed the public declaration and made the function static again. Platforms tested: Arabica