Signed-off-by: Huiwei Lu <huiweilu@mcs.anl.gov>

This reverts commit 274a5a70.

See #2128

Signed-off-by: Huiwei Lu <huiweilu@mcs.anl.gov>

Fixes #2128

Signed-off-by: Huiwei Lu <huiweilu@mcs.anl.gov>

When the memory overhead for memory usage tracing is bigger than
the threshold value, a warning message is generated to indicate the
amount of the memory used in the tracing. Environment variables
to control the threshold value were added.

See #2068.

Signed-off-by: Pavan Balaji <balaji@anl.gov>

We were creating duplicating information in the operation structure
and in the packet structure when the message is actually issued.
Since most of the information is the same anyway, this patch just
embeds a packet structure into the operation structure.

Signed-off-by: Xin Zhao <xinzhao3@illinois.edu>

Paul Coffman authored Jul 15, 2014 and Rob Latham committed Jul 16, 2014

use ADIOI_Assert to guard againt infinite loop;  simpler easier-to-read
code for fd_size assignment

Signed-off-by: Rob Latham <robl@mcs.anl.gov>

Paul Coffman authored Jul 11, 2014 and Rob Latham committed Jul 16, 2014

P2PCONTIG memory leak fixes - MPI_Request cleanups via MPI_Wait for MPI_Isends

Signed-off-by: Rob Latham <robl@mcs.anl.gov>

Paul Coffman authored Jul 11, 2014 and Rob Latham committed Jul 16, 2014

The BGQPLATFORM bridgelist and bridgelistnum hints need to be freed in
ADIOI_Close because of the way the "create on one, reopen on all" logic
and the hint processing logic interact.

Signed-off-by: Rob Latham <robl@mcs.anl.gov>

Paul Coffman authored Jul 07, 2014 and Rob Latham committed Jul 16, 2014

After a call to ADIOI_P2PContig(Write/Read)Aggregation, code exits early
from ADIOI_GPFS_(Write/Read)StridedColl and was not freeing some data
structures

Signed-off-by: Rob Latham <robl@mcs.anl.gov>

(ibm) D197049

Signed-off-by: Michael Blocksome <blocksom@us.ibm.com>

This reverts commit 1c5c5945

.

Reopens #1788  (datatype performance tests failing)

but better to have poor performance than incorrect performance

Closes #2115 (RMA fails with derived type containing struct of struct)
Closes #2126 (Data Integrity issue in MPI_Gather ...)

Conflicts:
	src/mpid/common/datatype/dataloop/dataloop_optimize.c

but only because a subsequent commit removed bits of this optimization.
This commit fully removes this optimization, but we leave behind test
cases to help us make sure we get it right next time.  We also leave
behind some additional debugging support routines.

Signed-off-by: Junchao Zhang <jczhang@mcs.anl.gov>

Incorrect preprocessor directives caused when porting Python code to C.

Signed-off-by: Wesley Bland <wbland@anl.gov>

Test this case so we don't have another regression. Thanks for test case
contributed by Lisandro Dalcin (dalcinl@gmail.com).

See #2129

Signed-off-by: Ken Raffenetti <raffenet@mcs.anl.gov>

Wesley Bland authored Jul 10, 2014 and Kenneth Raffenetti committed Jul 15, 2014

The refcounting for isends needs to be reset to 0 sometimes if an error occurs
before passing the request back to the user. This was mistakenly cleaned up in
[1e171ff6

] and now needs to be set back (along with a better comment
explaining the problem.

Thanks for the bug report to Lisandro Dalcin (dalcinl@gmail.com).

Fixes #2129

Signed-off-by: Ken Raffenetti <raffenet@mcs.anl.gov>

Pavan Balaji authored Jul 14, 2014 and Kenneth Raffenetti committed Jul 14, 2014

Signed-off-by: Ken Raffenetti <raffenet@mcs.anl.gov>

Signed-off-by: Sangmin Seo <sseo@anl.gov>

https://svn.mpi-forum.org/trac/mpi-forum-web/ticket/405

 says :
"The use of the datatype MPI_CHAR in the MPI tool information interface implies a
null-terminated character array, i.e. a string in the C language. If a variable has
type MPI_CHAR, the value of the count parameter returned by MPI_T_CVAR_HANDLE_ALLOC
and MPI_T_PVAR_HANDLE_ALLOC will include the null-terminating character."

Unlike cvars of type MPI_INT etc, MPI_CHAR cvars need explicit memory allocation for their storage.
Note: In MPI standard, MPI_CHAR is not in types of pvar.

Fixes #2021

Signed-off-by: Sangmin Seo <sseo@anl.gov>

Junchao Zhang authored Jul 14, 2014 and Sangmin Seo committed Jul 14, 2014

Make sure user buf wont overflow cvar which is a string

Signed-off-by: Sangmin Seo <sseo@anl.gov>

Junchao Zhang authored Jul 14, 2014 and Sangmin Seo committed Jul 14, 2014

The old value (4096) was arbitrarily chosen. To save memory and to make cvar
tests cover the MPI_CHAR case easily, we set it to a smaller value, which is
still big enough to hold reasonable cvar strings.

Signed-off-by: Sangmin Seo <sseo@anl.gov>

Paul Coffman authored Jul 09, 2014 and Rob Latham committed Jul 14, 2014

missed some comment and algorithm changes for ADIOI_PE_gen_agg_ranklist
during branch migration

Signed-off-by: Rob Latham <robl@mcs.anl.gov>

Paul Coffman authored Jul 08, 2014 and Rob Latham committed Jul 14, 2014

Signed-off-by: Rob Latham <robl@mcs.anl.gov>

Paul Coffman authored May 11, 2014 and Rob Latham committed Jul 14, 2014

Signed-off-by: Rob Latham <robl@mcs.anl.gov>

Paul Coffman authored Apr 04, 2014 and Rob Latham committed Jul 14, 2014

The MP_IOAGENT_CNT is another environment variable to control I/O
aggregation selection. Teach ROMIO about this to ease PE->ROMIO transistion.

Signed-off-by: Rob Latham <robl@mcs.anl.gov>

Paul Coffman authored Mar 31, 2014 and Rob Latham committed Jul 14, 2014

Signed-off-by: Rob Latham <robl@mcs.anl.gov>

Paul Coffman authored Mar 30, 2014 and Rob Latham committed Jul 14, 2014

We extend gpfs filesystem to PE.  Right now, the main accomodation is
for ROMIO aggregation selection using MP_IOTASKLIST to mimic current PE
functionality

Signed-off-by: Rob Latham <robl@mcs.anl.gov>

On FreeBSD and Darwin, the read and write system calls think 2147483648 (2^31:
larger than a signed 32 bit integer can hold) is an invalid amount of data.
Maybe, despite taking a size_t parameter, something internally treats the count
as an int.  Fine, we'll take measures to never write more than 2^31-1 bytes.

Signed-off-by: Ken Raffenetti <raffenet@mcs.anl.gov>

Mohamad Chaarawi authored Jul 11, 2014 and Rob Latham committed Jul 11, 2014

Commit [e527fa1d

] missed an update to an assertion in the read path: with
bufcount promoted to MPI_Count, making sure it does not overflow an
'unsigned' is incorrect.

Signed-off-by: Rob Latham <robl@mcs.anl.gov>

Testcase for "RMA fails with derived type containing struct of struct"

See #2115

declare "handle_error" with a const string to make compiler happy

this optimization in ROMIO's noncontigous type handling detects if a
user-defined type will fit into a single write request (e.g. a call to
MPI_Type_create_subarray that was actually the entire array).  Code was
passing in a count of bytes, which can overflow an int. since we know
the type is contiguous, simply pass a count of that type (not bytes) to
write_contig

Signed-off-by: Mohamad Chaarawi <chaarawi@hdfgroup.org>

Xin Zhao authored Jul 13, 2014 and Pavan Balaji committed Jul 13, 2014

See "Notes for memory barriers in RMA synchronizations" in
src/mpid/ch3/src/ch3u_rma_sync.c.

Signed-off-by: Pavan Balaji <balaji@anl.gov>

Xin Zhao authored Jul 13, 2014 and Pavan Balaji committed Jul 13, 2014

Signed-off-by: Pavan Balaji <balaji@anl.gov>

The f2c/c2f handle conversion functions are only defined in C in the standard.
We provide non-standard extensions in Fortran. All functions, except MPI_File_f2c/c2f,
are trival to implement. For MPI_File_f2c/c2f, we bind the Fortran functions to C.
We do not provide PMPI version for them since it is not required by the standard.

No review since F08 binding is experimental now.

We were creating the rank list for the window start group twice
earlier, once for synchronization and once for the actual issuing of
the operations.  This patch combines them into a single creation of
the array.

Signed-off-by: Xin Zhao <xinzhao3@illinois.edu>

We were using a stack variable earlier, so when FLUSH from a different
thread tried to send that data out it couldn't access that address.

Signed-off-by: Xin Zhao <xinzhao3@illinois.edu>

MPI_Win_shared_query() currently returns the base address of the passed in rank.
All tasks other than task 0 assume that the first set of data belongs to task 0,
second set belongs to task 1 and so on. The assumption caused the data integrity
problem found by win_shared.c.

To fix the problem, instead of returning the base address of a rank,
MPI_Win_shared_query() should return the base address of shared segment
allocated for the window.

(ibm) D198663

Signed-off-by: Michael Blocksome <blocksom@us.ibm.com>

So that f08 tests can be tested separately.

No review since F08 binding is experimental now.

When the debugging option about memory usage tracing is turned on,
additional memory overhead for tracing may be significant especially
when a large number of small memory allocations are requested, as
reported in #2068. However, since the tracing data is necessary to
keep track of memory usage, it cannot be removed. Here, we provide
a warning message when the memory overhead for tracing is over the
threshold value.

See #2068

Signed-off-by: Xin Zhao <xinzhao3@illinois.edu>