GitLab

iwriteatallf is failing on Linux machines of Jenkins test.
Mark iwriteatallf as xfail for now.

Ticket #2201

No reviewer

Move MPIX_Aint_add and MPIX_Aint_diff to MPI_ functions.

See MPI forum ticket #349:
  http://svn.mpi-forum.org/trac/mpi-forum-web/ticket/349

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

This commits moves the related internal MPIX_ names to MPI_ functions,
according to the new MPI 3.1 standard.
Functions affected:
  MPI_File_iread_all
  MPI_File_iwrite_all
  MPI_File_iread_at_all
  MPI_File_iwrite_at_all

See MPI forum ticket #273:
  http://svn.mpi-forum.org/trac/mpi-forum-web/ticket/273

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

comm_idup_overlap fails on mpich-mxm with "clang,async,ib"
configuration. It was not fully tested on mxm because mxm was added
after comm_idup patch. The reason of failing is still unknown. All other
platforms and configurations are fine.

Ticket #2238

No reviewer

These tests are still failing after a while and FT will not be an
announced feature for 3.2. Mark the tests as xfail for now.

No reviewer

The error code will be introduced in MPI-3.1. The code is used in
MPI_T_pvar_read, which has motivated for adding the error code.

See also https://svn.mpi-forum.org/trac/mpi-forum-web/ticket/400

Signed-off-by: Antonio J. Pena <apenya@mcs.anl.gov>

See https://svn.mpi-forum.org/trac/mpi-forum-web/ticket/377

Signed-off-by: Antonio J. Pena <apenya@mcs.anl.gov>

See https://svn.mpi-forum.org/trac/mpi-forum-web/ticket/377

Signed-off-by: Antonio J. Pena <apenya@mcs.anl.gov>

The new routines are defined in upcoming MPI-3.1.
See https://svn.mpi-forum.org/trac/mpi-forum-web/ticket/377

Signed-off-by: Antonio J. Pena <apenya@mcs.anl.gov>

They are defined in upcoming MPI-3.1.
See https://svn.mpi-forum.org/trac/mpi-forum-web/ticket/377

Signed-off-by: Antonio J. Pena <apenya@mcs.anl.gov>

artem.v.yalozo authored Feb 11, 2015 and Pavan Balaji committed Feb 27, 2015

Change-Id: Id7cb0c5122eb04eea7f2c5ea541b97a48ca71df3
Signed-off-by: Pavan Balaji <balaji@anl.gov>

Make it clearer that the meappend_large is passed a usable set of match_bits.
No further conversion is needed.
Signed-off-by: Antonio Pena Monferrer <apenya@mcs.anl.gov>

Moves the message length information from the payload to the message
header to save on overhead.
Signed-off-by: Antonio Pena Monferrer <apenya@mcs.anl.gov>

Use the match bits to encode a process's rank in MPI_COMM_WORLD, to be
retrieved by the receiving process. This is the same method used by the
2-sided Portals code, and allows the header to be used for other purposes.
Signed-off-by: Antonio Pena Monferrer <apenya@mcs.anl.gov>

Since we use the same portal for puts and gets in the RMA code, we
differentiate them with the otherwise unused match_bits. This commit
removes the sequenced tag numbers, instead relying on a simple 2 tag
system and the natural matching order of the portal.
Signed-off-by: Antonio Pena Monferrer <apenya@mcs.anl.gov>

When the async progress thread blocked the progress engine and yielded
control, if a thread started waiting inside a wait routine, e.g.,
ADIOI_GEN_iwc_wait_fn, of NBC I/O implementation, a deadlock happened.
The thread waiting continuously called MPI_Test to make progress, but
the progress engine did not make progress because it was blocked due to
the async progress thread.  The async progress thread tried to acquire
the lock, but the waiting thread did not release the lock because it
did not finish the wait routine.  Thus, it was a deadlock. This patch
fixes this deadlock problem by forcing the waiting thread to yield if
the progress engine has been blocked by another thread.

Fixes #2202
Signed-off-by: Rob Latham <robl@mcs.anl.gov>

Wesley Bland authored Feb 25, 2015 and Rob Latham committed Feb 26, 2015

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

Rob Latham authored Feb 16, 2015 and Kenneth Raffenetti committed Feb 25, 2015

lustre, or specifically the header files lustre brings in (quota.h) show
a problem with caddr_t not being defined if XOPEN_SOURCE is set.  We
provided this define so we could make use of pread/pwrite, but instead
we will use our replacement pread/pwrite if one does not exist.

See #1973
Signed-off-by: Ken Raffenetti <raffenet@mcs.anl.gov>

Rob Latham authored Feb 17, 2015 and Kenneth Raffenetti committed Feb 25, 2015

We can test at configure-time if pread/pwrite is available.
XOPEN_SOURCE set low enough will not provide this feature, for example
(as is the case with --enable-strict).
Signed-off-by: Ken Raffenetti <raffenet@mcs.anl.gov>

The RMA tests including ACC require the same basic datatype elements on
sender/receiver sides. However, the mtest was using double-based ddt for
one side and char for the other side. This patch unifies the basic dt to
double.
Signed-off-by: Xin Zhao <xinzhao3@illinois.edu>

When an operation is completed in an MPIC function, it should no longer
have the error bits set since that information has now been captured in
the errflag. This prevents unnecessary assert failures.
Signed-off-by: Huiwei Lu <huiweilu@mcs.anl.gov>

See also b21dc461 and 67577443

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

This issue was added in commit [54362c00

].
Signed-off-by: Igor Ivanov <Igor.Ivanov@itseez.com>
Signed-off-by: Wesley Bland <wbland@anl.gov>

Requests that aren't for receive operations don't have anything in them,
so trying to process them only generates valgrind warnings with
potentially unsafe behavior.
Signed-off-by: Huiwei Lu <huiweilu@mcs.anl.gov>

The function to check whether an operation was an anysource receive was
checking all request kinds, even if they weren't receives. This limits
that check to only receives to avoid examining an uninitialized
variable.
Signed-off-by: Huiwei Lu <huiweilu@mcs.anl.gov>

This is an initial limited implementation for CUDA support. This is not
performance optimized and only for testing.

(ibm) D202477
Signed-off-by: Su Huang <suhuang@us.ibm.com>

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

The comments are no longer significant for
new RMA infrastructure.
Signed-off-by: Pavan Balaji <balaji@anl.gov>

Here req->dev.user_count is used when receiving FOP/CAS response
data on origin in PktHandler_FOPResp and PktHandler_CASResp. Since
the count always be 1, we did not set rma_op->result_count, and
we directly set req->dev.user_count to 1 in packet handlers.
Signed-off-by: Pavan Balaji <balaji@anl.gov>

When issuing RMA packets, we do not need to
store target_win_handle in the request on
origin side but only need to store source_win_handle.
Because when the response data is back, we
only needs to use source_win_handle on origin
size. This patch simplifies the code in this way.
Signed-off-by: Pavan Balaji <balaji@anl.gov>

For GET-like RMA packets and response packets (GACC,
GET, FOP, CAS, GACC_RESP, GET_RESP, FOP_RESP, CAS_RESP),
originally we carry source_win_handle in packet struct
in order to locate window handle on origin side in the
packet handler of response packets. However, this is
not necessary because source_win_handle can be stored
in the request on the origin side. This patch delete
source_win_handle from those packets to reduce the size
of packet union.
Signed-off-by: Pavan Balaji <balaji@anl.gov>

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

do_accumulate_op() does more comprehensive work on ACC
computation than OP function. For example, MPI_REPLACE
is not defined as predefined computation and therefore
not handled by OP function, but it is safely handled
in do_accumulate_op(). This patch replace OP function
with do_accumulate_op() on target side.
Signed-off-by: Pavan Balaji <balaji@anl.gov>

In this patch we replace "=" with memcpy function
when assigning structure content to another struct.
Using "=" in this case is not compatible for llvm
compiler.
Signed-off-by: Pavan Balaji <balaji@anl.gov>

In this patch, we replace one argument of function
finish_op_on_target, "packet(op) type", with "has_response_data".
Since finish_op_on_target does not care what specific
packet(op) type it is processing on, but only cares
about if the current op has response data (like GET/GACC),
changing the argument in this way can simplify the
code by avoiding acquiring packet(op) type everytime
before calling finish_op_on_target.
Signed-off-by: Pavan Balaji <balaji@anl.gov>

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

Originally we add "immed_data" and "immed_len" areas to RMA packets,
in order to piggyback small amount of data with packet header to
reduce number of packets (Note that "immed_len" is necessary when
the piggybacked data is not the entire data). However, those areas
potentially increase the packet union size and worsen the two-sided
communication. This patch fixes this issue.

In this patch, we remove "immed_data" and "immed_len" from normal
"MPIDI_CH3_Pkt_XXX_t" operation type (e.g. MPIDI_CH3_Pkt_put_t), and
we introduce new "MPIDI_CH3_Pkt_XXX_immed_t" packt type for each
operation (e.g. MPIDI_CH3_Pkt_put_immed_t).

"MPIDI_CH3_Pkt_XXX_immed_t" is used when (1) both origin and target
are basic datatypes, AND, (2) the data to be sent can be entirely fit
into the header. By doing this, "MPIDI_CH3_Pkt_XXX_immed_t" needs
"immed_data" area but can drop "immed_len" area. Also, since it only
works with basic target datatype, it can drop "dataloop_size" area
as well. All operations that do not satisfy (1) or (2) will use
normal "MPIDI_CH3_Pkt_XXX_t" type.

Originally we always piggyback FOP data into the packet header,
which makes the packet size too large. In this patch we split the
FOP operaton into IMMED packets and normal packets.

Because CAS only work with 2 basic datatype and non-complex
elements, the data amount is relatively small, we always piggyback
the data with packet header and only use "MPIDI_CH3_Pkt_XXX_immed_t"
packet type for CAS.
Signed-off-by: Pavan Balaji <balaji@anl.gov>

This patch just does code refactoring for RMA operation rountines
to make the code structure clearer. This patch does not change any
functionality.

After code refactoring, in each operation routine, for non-SHM operations
we do the work in the following order:

(1) allocate a new op struct;
(2) fill areas in op struct, except for packet struct in op struct;
(3) initialize packet struct in op struct, fill areas in packet struct;
(4) enqueue op to data structure on window.
Signed-off-by: Pavan Balaji <balaji@anl.gov>