George Bosilca 21b940887a Tricky stuff !!! If we post a receive for ZERO bytes and we match it ...

with something with a different size ... well we segfault. The reason was
that the logic in the PML OB1 call the convertor based on the length
of he data on the wire and not the length of the data that the receiver
expects.

In other words, this is only half a patch :) It fix the problem, but we
still have to make sure the unpack is not called at all when the receiver
expect ZERO bytes.

This commit was SVN r18474.

2008-05-21 23:31:34 +00:00

..

convertor_internal.h

Look like a big commit but in fact it address only one issue. The way we're working with

2006-10-17 20:20:58 +00:00

convertor.c

This commit represents a bunch of work on a Mercurial side branch. As

2008-05-13 20:00:55 +00:00

convertor.h

Tricky stuff !!! If we post a receive for ZERO bytes and we match it

2008-05-21 23:31:34 +00:00

copy_functions_heterogeneous.c

This commit represents a bunch of work on a Mercurial side branch. As

2008-05-13 20:00:55 +00:00

copy_functions.c

Look like a big commit but in fact it address only one issue. The way we're working with

2006-10-17 20:20:58 +00:00

datatype_checksum.h

Keep the += by now. The only checksum that we have require it.

2006-06-16 03:01:16 +00:00

datatype_internal.h

This commit represents a bunch of work on a Mercurial side branch. As

2008-05-13 20:00:55 +00:00

datatype_memcpy.h

Last step for the generation of 2 set of pack/unpack functions. One with

2006-03-17 08:04:59 +00:00

datatype_pack.c

This commit represents a bunch of work on a Mercurial side branch. As

2008-05-13 20:00:55 +00:00

datatype_pack.h

This commit represents a bunch of work on a Mercurial side branch. As

2008-05-13 20:00:55 +00:00

datatype_prototypes.h

These functions dont have to be exported. The outside world

2006-11-08 16:49:34 +00:00

datatype_unpack.c

This commit represents a bunch of work on a Mercurial side branch. As

2008-05-13 20:00:55 +00:00

datatype_unpack.h

This commit represents a bunch of work on a Mercurial side branch. As

2008-05-13 20:00:55 +00:00

datatype.h

Minimize the dependencies for the datatype header file (datatype.h)

2008-02-10 21:40:47 +00:00

dt_add.c

This commit represents a bunch of work on a Mercurial side branch. As

2008-05-13 20:00:55 +00:00

dt_args.c

Shift the architecture calculation from the ompi/datatype engine to the opal/util area. This allows us to compute the architecture earlier in the launch and communicate it outside of the modex.

2008-04-17 20:43:56 +00:00

dt_copy.c

This commit represents a bunch of work on a Mercurial side branch. As

2008-05-13 20:00:55 +00:00

dt_create_dup.c

Minimize the dependencies for the datatype header file (datatype.h)

2008-02-10 21:40:47 +00:00

dt_create_indexed.c

Minimize the dependencies for the datatype header file (datatype.h)

2008-02-10 21:40:47 +00:00

dt_create_struct.c

Minimize the dependencies for the datatype header file (datatype.h)

2008-02-10 21:40:47 +00:00

dt_create_vector.c

Minimize the dependencies for the datatype header file (datatype.h)

2008-02-10 21:40:47 +00:00

dt_create.c

Replace the ompi_pointer_array with opal_pointer_array. The next step

2007-12-21 06:02:00 +00:00

dt_destroy.c

Minimize the dependencies for the datatype header file (datatype.h)

2008-02-10 21:40:47 +00:00

dt_external32.c

Shift the architecture calculation from the ompi/datatype engine to the opal/util area. This allows us to compute the architecture earlier in the launch and communicate it outside of the modex.

2008-04-17 20:43:56 +00:00

dt_get_count.c

Initialize the length before using it.

2006-11-01 22:07:21 +00:00

dt_match_size.c

Replace the ompi_pointer_array with opal_pointer_array. The next step

2007-12-21 06:02:00 +00:00

dt_module.c

This commit represents a bunch of work on a Mercurial side branch. As

2008-05-13 20:00:55 +00:00

dt_optimize.c

Finally the convertor export a convenience function to allow a consistent

2007-03-30 22:02:45 +00:00

dt_sndrcv.c

Rewrite the ompi_ddt_sndrcv function.

2006-11-12 05:56:02 +00:00

fake_stack.c

This commit represents a bunch of work on a Mercurial side branch. As

2008-05-13 20:00:55 +00:00

Makefile.am

Shift the architecture calculation from the ompi/datatype engine to the opal/util area. This allows us to compute the architecture earlier in the launch and communicate it outside of the modex.

2008-04-17 20:43:56 +00:00

position.c

This commit represents a bunch of work on a Mercurial side branch. As

2008-05-13 20:00:55 +00:00