GitLab

Added builtin op implementations for RMA-only ops.  MPI_REPLACE was not defined
in the op_table prior to this commit.  Also includes an alternate
implementation of the linked list test that uses fetch-and-op with these
operations to enable the use of shared locks.

Reviewer: goodell

Adds implementation of atomic fetch-and-op operation and several test cases.

Reviewer: buntinas

Reviewer: buntinas

Reviewer: buntinas

The source_win_handle is used to inform the target that the RMA access epoch
has ended.  This was removed from the CAS packet to save space and was replaced
with another packet type, MPIDI_CH3_PKT_CAS_UNLOCK, which conveys the same
information.

Reviewer: buntinas

This patch adds support for the new MPI-3 compare-and-swap routine.  The
implementation embeds the data into the packet header to reduce the number of
messages.

Reviewer: buntinas

Pulled bit manipulation arithmetic into macros for converting handles to
indices in MPIR_Op and MPIR_Op_check_dtype function pointer tables.

Reviewer: buntinas

Removed the old RMA virtual function infrastructure from CH3 -- this code was
all already dead.  Function overrides are already provided per-window in the
MPID_Win structure.  Overrides for non-window-specific (window creation)
operations will be added shortly.

Reviewer: buntinas

yet implemented below the ADI level.

[svn-r9963] Added code to complete RMA operations when large numbers of pending requests are created; allows codes such as Graph500 benchmark to complete

No reviewer.

No reviewer.

No reviewer.

[svn-r8069] Detect and report incorrect use of assert and/or rank in Win_lock and Win_unlock.  These detect an error in the current test suite that manifested itself as a storage leak.  Also corrected some man page information and instrumentation for RMA completion in non-fence modes

[svn-r7425] Fixed error in computing previous pointer location in RMA list and adding missing checks on error returns

[svn-r7416] Major improvement to RMA performance for long lists of operations, an immediate mode accumulate for single ints, store the MPID_Comm within the window, and added a basic performance instrumentation interface that was extensively used to improve the RMA performance (enabled with --enable-g=instr).  With these fixes, MPICH2 can run the one-sided version of the Graph500 benchmark at a respectable if not great rate

[svn-r7312] Correct use of MPI_ERR_NO_MEM (should be MPI_ERR_OTHER) for out-of-memory that does not pertain to MPI_Alloc_mem and add detail to error messages about out-of-memory

[svn-r7186] Fixes to allow job to continue with a node failure.  VCs associated with the failed process enter a failed state, and cannot be used for communication, but communication is still allowed with other processes.  Reviewed by goodell@.

Reviewed by buntinas.

No reviewer.

[svn-r7029] replaced NMPI_ functions with MPIR_ _impl for pack, unpack, pack_size, and group_translate_ranks.  reviewed by balaji@

[svn-r7012] replaced NMPI_[i]Send, [i]Recv, Wait[all] and Test with MPIR__impl, or directly called MPID_ functions.  Reviewed by jayesh@

When compiled for fine-grained threading, the completion counter serves
as a form of lockfree signalling.  As such, atomic access and memory
barriers must be used to ensure correctness.

In per-object mode, this code also contains valgrind client request annotations
to inform Helgrind/DRD/TSan about the lockfree signalling pattern.

No reviewer.

[svn-r6799] Replaced NMPI_Reduce_scatter_blocked with MPIR_Reduce_scatter_blocked_impl.  Reviewed by goodell@

This commit moves the existing tracing memory allocation and valgrind
code utilities to MPL.  This permits other code like gforker to use the
trmem code while allowing MPICH2 to remain thread-safe when using
tracing.  gforker is not updated to the MPL routines by this commit.

Reviewed by buntinas@.

The MPIDI_CH3U_Request_load_recv_iov function must fill the iov in the
rreq at offset 0 *and* set rreq->dev.iov_offset to 0 because callers of
this function as an OnDataAvail function don't know what action is
occurring in the handler and can't set the offset to zero themselves.

See r4099 and ticket #462 for some info on the previous test1_dt bug.

Reviewed by buntinas@.

[svn-r4411] Fixed nemesis to correctly set vc state (instead of setting all vcs to active.  Renamed MPIDI_Comm_get_vc to MPIDI_Comm_get_vc_set_active to alert caller to side-effect.  Added MPIDI_CHANGE_VC_STATE macro to set the vc state and call debugging macro.  Changed all places where vc state is changed to use this macro.  Reviewed by goodell@.

[svn-r4062] 1) Modifying the datatype code to find the number of contig blocks in an instance of an MPI derived datatype - Since finding the real number of contig blocks is not easy we find a reasonable upper bound instead. This also fixes the case where the number of contig blocks was uninitialized for contiguous datatypes. Refer ticket #428 for details (2) Adding a test case, test1_dt.c, to test the fix - Review @ rross, thakur

[svn-r3717] Added ATTRIBUTE((unused)) in several places; removed unused variables and code; added some fixmes

[svn-r3275] Updates for brief-global thread cs support (still requires updates to MPIR_Localcopy) and for new match structure defintion where used within debugging statements