ports/openmpi
ports/openmpi
1
1
Форкнуть 0
Код Релизы Активность
7 297 коммитов 3 Ветки 0 Теги 130 MiB
b990870b55
Граф коммитов

191 Коммитов

Автор SHA1 Сообщение Дата
George Bosilca
b990870b55 Keep only the fastest version of the homogeneous unpack. Reorder the code 
to improve the execution path. Some minor clean-ups.

This commit was SVN r9773.
 2006-04-28 16:19:54 +00:00
George Bosilca
e23c3008a6 Keep only the best [the fastest] version of the unpack function. 
This commit was SVN r9772.
 2006-04-28 16:18:29 +00:00
George Bosilca
5133c987e3 Move the code around to shorten the execution path for the most common data-types. 
This commit was SVN r9771.
 2006-04-28 16:17:28 +00:00
George Bosilca
0fd4e2d906 Optimize the execution path. 
This commit was SVN r9763.
 2006-04-27 22:25:53 +00:00
George Bosilca
faba013106 Move tests. Most of the checking are now disabled, only the performance one 
are still active.

This commit was SVN r9762.
 2006-04-27 22:25:02 +00:00
George Bosilca
ced5d564cf Make sure we set the COMPLETED flag on the convertor in all the cases. 
This commit was SVN r9761.
 2006-04-27 20:53:28 +00:00
George Bosilca
80a53c7aff More tests, some comments, some new data-types. First step for the performance framework. 
This commit was SVN r9760.
 2006-04-27 20:52:43 +00:00
George Bosilca
0fcae472aa Detect when the convertor is done with the conversion in a more 
generic way.

This commit was SVN r9756.
 2006-04-27 17:25:56 +00:00
George Bosilca
cabf703dad No need for the assert after a loop. 
This commit was SVN r9755.
 2006-04-27 17:25:21 +00:00
George Bosilca
aa11164fdc Update the test to match the latest version of the convertor. 
This commit was SVN r9754.
 2006-04-27 17:24:53 +00:00
George Bosilca
29fe545451 Remove all references to the ompi_mpi_long_long variable. The MPI_LONG_LONG is now 
a synonym to MPI_LONG_LONG_INT.

This commit was SVN r9703.
 2006-04-24 22:15:42 +00:00
George Bosilca
40bcb2fb36 Still some long long left over. Should be clean by now. 
This commit was SVN r9701.
 2006-04-24 21:40:16 +00:00
George Bosilca
a297a7ae67 MPI standard state that MPI_LONG_LONG and MPI_LONG_LONG_INT are synonyms. Thanks to 
Martin audet for finding out this one.

This commit was SVN r9699.
 2006-04-24 21:24:10 +00:00
Brian Barrett
453f9b870f * Rename two files that had conflicting basenames with files in mpi/c/. This 
can cause issues on some platforms (since they end up in the same .a file)

This commit was SVN r9685.
 2006-04-22 20:03:45 +00:00
George Bosilca
87269ba0aa Add 3 functions for Galen. The allow to import, export and reset the 
pending bytes from the convertor. These bytes are used to handle partial
conversions between peers.

This commit was SVN r9531.
 2006-04-05 03:04:50 +00:00
George Bosilca
af7930a498 Don't forget to add the new .h file to the distribution list. 
This commit was SVN r9437.
 2006-03-28 07:16:39 +00:00
George Bosilca
c93680074f Enable the master convertor and add a function to clean them all up. Without talking 
about the comments :)

This commit was SVN r9435.
 2006-03-28 05:11:48 +00:00
George Bosilca
deb1b237a7 First cut of a unique architecture base master convertor. It will hold the sizes 
of the predefined types as seen by the remote architecture, as well as some
pointers to the conversion functions. Still in progress, but it should just fine
for what we have today in Open MPI.

This commit was SVN r9433.
 2006-03-28 03:16:53 +00:00
Galen Shipman
af02867322 Use ones compiliment so we don't kill previously set flags.. 
This commit was SVN r9428.
 2006-03-27 16:12:22 +00:00
George Bosilca
52e5d90a53 ompi_dt_swap_bytes is the way to go. 
This commit was SVN r9424.
 2006-03-25 07:12:07 +00:00
Brian Barrett
b6ddded1fa * Fix bug in determining size of C++ bool when creating the predefined 
datatype MPI::BOOL.
* Add first round of support for heterogeneous platforms.  This includes
  endianness and size difference for C++ bool and Fortran LOGICAL.  It
  does not include differences in sizes for any other datatype or for
  different representations of floating point numbers.

This commit was SVN r9422.
 2006-03-25 02:53:41 +00:00
George Bosilca
b22836ea93 Reset the completed flag when we roll-back the convertor. 
This commit was SVN r9419.
 2006-03-24 20:33:53 +00:00
Tim Woodall
b09c2dc1cd correction of checksum for pending/buffered data 
This commit was SVN r9367.
 2006-03-22 16:47:12 +00:00
Brian Barrett
b74d3112ec * add missing header file to headers list so that it ends up in the dist 
tarball

This commit was SVN r9336.
 2006-03-20 04:37:33 +00:00
Brian Barrett
01603d3d9f * reorder copy_functions array to match current predefined datatype ordering 
and add missing datatypes.

This commit was SVN r9335.
 2006-03-20 01:16:30 +00:00
Tim Woodall
bd870519fd - modified convertor copy_and_prepare routines to accept an addition 
flag, new flags to be included when convertor is initialized
- modified pml/btl module defs and added stub functions for diagnostic
  output routines to dump state of queues / endpoints
- updates to data reliability pml

This commit was SVN r9329.
 2006-03-17 18:46:48 +00:00
George Bosilca
5819304916 Unfortunately, I cannot remove the log from SVN to erase all trace (for the sake 
of the future generations) of my own way to define function prototypes.
Shame on me ...

This commit was SVN r9327.
 2006-03-17 16:39:36 +00:00
George Bosilca
c2f001c0f2 Remove the dependency to datatype_memcpy.c. It's not yet ready for primetime. 
This commit was SVN r9326.
 2006-03-17 16:03:23 +00:00
George Bosilca
0fe57bb44c Allow the convertor to select between pack/unpack with or without 
checksum. Specify CONVERTOR_WITH_CHECKSUM when the convertor is
created and all the pack/unpack functions attached to this convertor
will be checksum aware.

This commit was SVN r9325.
 2006-03-17 08:27:17 +00:00
George Bosilca
6babf2f874 Rename files. 
This commit was SVN r9324.
 2006-03-17 08:10:37 +00:00
George Bosilca
9955eb2f2e Last step for the generation of 2 set of pack/unpack functions. One with 
checksum and the other without. Split the large files into smaller one, and
put similar functions together.

This commit was SVN r9323.
 2006-03-17 08:04:59 +00:00
George Bosilca
229f26dc55 First split of the datatype. More files and a cleaner distribution of functions 
in the corresponding files. There are few others changes to come ...

This commit was SVN r9319.
 2006-03-16 21:04:34 +00:00
George Bosilca
e79f35fe6c Copy the last fake DT_END_LOOP too when we duplicate the data. 
This commit was SVN r9309.
 2006-03-16 17:21:10 +00:00
Rainer Keller
2ce64c6fc7 - Add missing signed_char 
All hail the moog!

This commit was SVN r9299.
 2006-03-16 11:06:59 +00:00
George Bosilca
49c40f596b No data validation by default. 
This commit was SVN r9295.
 2006-03-16 01:35:49 +00:00
Tim Woodall
9e958f727f additional macros for csum 
This commit was SVN r9293.
 2006-03-16 00:19:55 +00:00
Brian Barrett
5c7751f9a2 * configure changes required to support the double compile of the ddt engine. 
still need to coordinate some Makefile.am changes with george tomorrow, but
  this is the stuff that will require re-autogening

This commit was SVN r9285.
 2006-03-15 03:35:30 +00:00
Graham Fagg
fcca36d280 Add missing ifdef 
This commit was SVN r9279.
 2006-03-14 20:48:50 +00:00
Tim Woodall
0b366e912c macros to compute checksum while moving data 
This commit was SVN r9278.
 2006-03-14 19:33:27 +00:00
George Bosilca
d5cee0cd75 Data validation should only be used with DR PML. 
This commit was SVN r9210.
 2006-03-06 20:01:26 +00:00
Tim Woodall
d350232c04 work in progress 
This commit was SVN r9209.
 2006-03-06 19:30:37 +00:00
Rainer Keller
5627fb5627 - Get rid of compiler warning: init type to DT_MAX_PREDEFINED 
and assert it is below when accessing ompi_ddt_basicDatatypes
 - datatype in ompi_ddt_pack_description_length is and will not be
   changed.

This commit was SVN r9193.
 2006-03-04 13:16:04 +00:00
Tim Woodall
274ee03df6 work in progress 
This commit was SVN r9192.
 2006-03-04 00:36:16 +00:00
George Bosilca
ca116e3cf3 Where we create a duplicate data-type for the get_content MPI function we have to attach the 
args from the old data-type. As this args are only read-only, I don't want to have several
copies around, so I add a reference count to them and simply attach them to the new
datatype. The args array now get freed only when the last reference dissapear.

This commit was SVN r9191.
 2006-03-03 20:51:36 +00:00
Rainer Keller
b8d1189091 - Use the actual _count parameter passed through the macro 
This commit was SVN r9189.
 2006-03-02 20:23:07 +00:00
George Bosilca
06def0b2d6 Not one .. but i. We're looking at the current position not at the first position 
in the buffer ...

This commit was SVN r9132.
 2006-02-23 15:59:17 +00:00
George Bosilca
0c1c650c91 Do not forget to commit the data-type once it is created from the 
packed description.

This commit was SVN r9125.
 2006-02-23 04:29:21 +00:00
George Bosilca
574c384ad4 This is the internal function so we should return the correct type. The 
external function is type_size and it's used only at the MPI level as MPI
only support integers for the size.

This commit was SVN r9124.
 2006-02-23 04:28:44 +00:00
George Bosilca
9dd522e350 Correct the MPI_COMBINER_DUP case when we generate the data description. Now, we have an 
optimized version here, where we pack only 2 integers instead of 6.

This commit was SVN r9102.
 2006-02-21 17:16:30 +00:00
George Bosilca
3ac541a722 Remove some debugging code. 
This commit was SVN r9101.
 2006-02-21 15:17:50 +00:00