mpidpre.h

/*
 * Copyright (C) by Argonne National Laboratory
 *     See COPYRIGHT in top-level directory
 */

/* FIXME: This header should contain only the definitions exported to the
   mpiimpl.h level */

#ifndef MPIDPRE_H_INCLUDED
#define MPIDPRE_H_INCLUDED

/* Tell the compiler that we're going to declare struct MPIR_Request later */
struct MPIR_Request;

#if defined(HAVE_SYS_TYPES_H)
#include <sys/types.h>
#endif

#ifdef HAVE_HCOLL
#include "hcoll/api/hcoll_dte.h"
#endif

#define MPID_TAG_DEV_BITS 0

typedef struct {
#ifdef HAVE_HCOLL
    hcoll_datatype_t hcoll_datatype;
#endif
    int foo; /* Shut up the compiler */
} MPIDI_Devdt_t;
#define MPID_DEV_DATATYPE_DECL   MPIDI_Devdt_t   dev;

#include "mpid_thread.h"

#include "mpid_sched.h"

union MPIDI_CH3_Pkt;
struct MPIDI_VC;
struct MPIR_Request;

/* PktHandler function:
   vc  (INPUT) -- vc on which the packet was received
   pkt (INPUT) -- pointer to packet header (aligned access).
   data (INPUT) -- pointer to beginning of data
   buflen (I/O) -- IN: number of bytes received into receive buffer
                   OUT: number of bytes processed by the handler function
   req (OUTPUT) -- NULL, if the whole message has been processed by the handler
                   function, otherwise, pointer to the receive request for this
                   message.  The IOV will be set describing where the rest of the
                   message should be received.
   (This decl needs to come before mpidi_ch3_pre.h)
*/
typedef int MPIDI_CH3_PktHandler_Fcn(struct MPIDI_VC *vc, union MPIDI_CH3_Pkt *pkt, void *data,
				     intptr_t *buflen, struct MPIR_Request **req );

/* Include definitions from the channel which must exist before items in this 
   file (mpidpre.h) or the file it includes (mpiimpl.h) can be defined. */
#include "mpidi_ch3_pre.h"

/* FIXME: Who defines this name */
/* As of 8/1/06, no-one defined MSGS_UNORDERED.  We should consider 
   moving support for unordered messages to a different part of the code
   However, note that sequence numbers may be useful in other contexts, 
   including identifying messages when multithreaded (for better profiling
   tools) and handling cancellations (rather than relying on unique 
   request ids) 
*/
#if defined (MPIDI_CH3_MSGS_UNORDERED)
#define MPID_USE_SEQUENCE_NUMBERS
#endif

#if defined(MPID_USE_SEQUENCE_NUMBERS)
typedef unsigned long MPID_Seqnum_t;
#endif

#include "mpichconf.h"

#if CH3_RANK_BITS == 16 && !defined(HAVE_EXTENDED_CONTEXT_BITS)
typedef int16_t MPIDI_Rank_t;
typedef int16_t MPIDI_Context_id_t;
#else
typedef int32_t MPIDI_Rank_t;
typedef int32_t MPIDI_Context_id_t;
#endif /* CH3_RANK_BITS */

/* For the typical communication system for which the ch3 channel is
   appropriate, 16 bits is sufficient for the rank.  By also using 16
   bits for the context, we can reduce the size of the match
   information, which is beneficial for slower communication
   links. Further, this allows the total structure size to be 64 bits
   and the search operations can be optimized on 64-bit platforms. We
   use a union of the actual required structure with a uintptr_t, so
   in this optimized case, the "whole" field can be used for
   comparisons.

   Note that the MPICH code (in src/mpi) uses int for rank (and usually for 
   contextids, though some work is needed there).  

   Note:  We need to check for truncation of rank in MPID_Init - it should 
   confirm that the size of comm_world is less than 2^15, and in an communicator
   create (that may make use of dynamically created processes) that the
   size of the communicator is within range.

   If any part of the definition of this type is changed, those changes
   must be reflected in the debugger interface in src/mpi/debugger/dll_mpich.c
   and dbgstub.c
*/
typedef struct MPIDI_Message_match_parts {
    int32_t tag;
    MPIDI_Rank_t rank;
    MPIDI_Context_id_t context_id;
} MPIDI_Message_match_parts_t;
typedef union {
    MPIDI_Message_match_parts_t parts;
    uintptr_t whole;
} MPIDI_Message_match;

/* Provides MPIDI_CH3_Pkt_t.  Must come after MPIDI_Message_match definition. */
#include "mpidpkt.h"

/*
 * THIS IS OBSOLETE AND UNUSED, BUT RETAINED FOR ITS DESCRIPTIONS OF THE
 * VARIOUS STATES.  Note that this is not entirely accurate, as the 
 * CA_COMPLETE state could depend on the packet type (e.g., for RMA 
 * operations).
 *
 * MPIDI_CA_t
 *
 * An enumeration of the actions to perform when the requested I/O operation 
 * has completed.
 *
 * MPIDI_CH3_CA_COMPLETE - The last operation for this request has completed.
 * The completion counter should be decremented.  If
 * it has reached zero, then the request should be released by calling 
 * MPIR_Request_free().
 *
 * MPIDI_CH3_CA_UNPACK_UEBUF_AND_COMPLETE - This is a special case of the 
 * MPIDI_CH3_CA_COMPLETE.  The data for an unexpected
 * eager messaage has been stored into a temporary buffer and needs to be 
 * copied/unpacked into the user buffer before the
 * completion counter can be decremented, etc.
 *
 * MPIDI_CH3_CA_UNPACK_SRBUF_AND_COMPLETE - This is a special case of the 
 * MPIDI_CH3_CA_COMPLETE.  The data from the completing
 * read has been stored into a temporary send/receive buffer and needs to be 
 * copied/unpacked into the user buffer before the
 * completion counter can be decremented, etc.
 *
 * MPIDI_CH3_CA_RELOAD_IOV - This request contains more segments of data than 
 * the IOV or buffer space allow.  Since the
 * previously request operation has completed, the IOV in the request should 
 * be reload at this time.
 *
 * MPIDI_CH3_CA_UNPACK_SRBUF_AND_RELOAD_IOV - This is a special case of the 
 * MPIDI_CH3_CA_RELOAD_IOV.  The data from the
 * completing read operation has been stored into a temporary send/receive 
 * buffer and needs to be copied/unpacked into the user
 * buffer before the IOV is reloaded.
 *
 * MPIDI_CH3_CA_END_CH3 - This not a real action, but rather a marker.  
 * All actions numerically less than MPID_CA_END are defined
 * by channel device.  Any actions numerically greater than MPIDI_CA_END are 
 * internal to the channel instance and must be handled
 * by the channel instance.
 */

#define MPID_Comm_commit_pre_hook(comm_) MPIDI_CH3I_Comm_commit_pre_hook(comm_)
#define MPID_Comm_commit_post_hook(comm_) MPIDI_CH3I_Comm_commit_post_hook(comm_)
#define MPID_Comm_free_hook(comm_) MPIDI_CH3I_Comm_destroy_hook(comm_)
#define MPID_Comm_set_hints(comm_, info_) MPIDI_CH3I_Comm_set_hints(comm_, info_)

#ifndef HAVE_MPIDI_VCRT
#define HAVE_MPIDI_VCRT
typedef struct MPIDI_VC * MPIDI_VCR;
#endif

typedef struct MPIDI_CH3I_comm
{
    int anysource_enabled;  /* TRUE iff this anysource recvs can be posted on this communicator */
    int last_ack_rank;      /* The rank of the last acknowledged failure */
    int waiting_for_revoke; /* The number of other processes from which we are
                             * waiting for a revoke message before we can release
                             * the context id */

    int is_disconnected;    /* set to TRUE if this communicator was
                             * disconnected as a part of
                             * MPI_COMM_DISCONNECT; FALSE otherwise. */

    struct MPIDI_VCRT *vcrt;          /* virtual connection reference table */
    struct MPIDI_VCRT *local_vcrt;    /* local virtual connection reference table */

    struct MPIR_Comm *next; /* next pointer for list of communicators */
    struct MPIR_Comm *prev; /* prev pointer for list of communicators */
    MPIDI_CH3I_CH_comm_t ch;
}
MPIDI_CH3I_comm_t;

#define MPID_DEV_COMM_DECL MPIDI_CH3I_comm_t dev;

#ifndef DEFINED_REQ
#define DEFINED_REQ
#if defined(MPID_USE_SEQUENCE_NUMBERS)
#   define MPIDI_REQUEST_SEQNUM	\
        MPID_Seqnum_t seqnum;
#else
#   define MPIDI_REQUEST_SEQNUM
#endif

/* Here we add RMA sync types to specify types
 * of synchronizations the origin is going to
 * perform to the target. */

/* There are four kinds of synchronizations: NONE,
 * FLUSH_LOCAL, FLUSH, UNLOCK.
 * (1) NONE means there is no special synchronization,
 * origin just issues as many operations as it can,
 * excluding the last operation which is a piggyback
 * candidate;
 * (2) FLUSH_LOCAL means origin wants to do a
 * FLUSH_LOCAL sync and issues out all pending
 * operations including the piggyback candidate;
 * (3) FLUSH means origin wants to do a FLUSH sync
 * and issues out all pending operations including
 * the last op piggybacked with a FLUSH flag to
 * detect remote completion;
 * (4) UNLOCK means origin issues all pending operations
 * including the last op piggybacked with an UNLOCK
 * flag to release the lock on target and detect remote
 * completion.
 * Note that FLUSH_LOCAL is a superset of NONE, FLUSH
 * is a superset of FLUSH_LOCAL, and UNLOCK is a superset
 * of FLUSH.
 */
/* We start with an arbitrarily chosen number (58), to help with
 * debugging when a sync type is not initialized or wrongly
 * initialized. */
enum MPIDI_RMA_sync_types {
    MPIDI_RMA_SYNC_NONE = 58,
    MPIDI_RMA_SYNC_FLUSH_LOCAL,
    MPIDI_RMA_SYNC_FLUSH,
    MPIDI_RMA_SYNC_UNLOCK
};

/* We start with an arbitrarily chosen number (63), to help with
 * debugging when a window state is not initialized or wrongly
 * initialized. */
enum MPIDI_RMA_states {
    /* window-wide states */
    MPIDI_RMA_NONE = 63,
    MPIDI_RMA_FENCE_ISSUED,           /* access / exposure */
    MPIDI_RMA_FENCE_GRANTED,          /* access / exposure */
    MPIDI_RMA_PSCW_ISSUED,            /* access */
    MPIDI_RMA_PSCW_GRANTED,           /* access */
    MPIDI_RMA_PSCW_EXPO,              /* exposure */
    MPIDI_RMA_PER_TARGET,             /* access */
    MPIDI_RMA_LOCK_ALL_CALLED,        /* access */
    MPIDI_RMA_LOCK_ALL_ISSUED,        /* access */
    MPIDI_RMA_LOCK_ALL_GRANTED,       /* access */

    /* target-specific states */
    MPIDI_RMA_LOCK_CALLED,            /* access */
    MPIDI_RMA_LOCK_ISSUED,            /* access */
    MPIDI_RMA_LOCK_GRANTED,           /* access */
};

/* We start with an arbitrarily chosen number (19), to help with
 * debugging when a lock state is not initialized or wrongly
 * initialized. */
enum MPIDI_CH3_Lock_states {
    MPIDI_CH3_WIN_LOCK_NONE = 19,
    MPIDI_CH3_WIN_LOCK_CALLED,
    MPIDI_CH3_WIN_LOCK_REQUESTED,
    MPIDI_CH3_WIN_LOCK_GRANTED,
    MPIDI_CH3_WIN_LOCK_FLUSH
};

enum MPIDI_Win_info_arv_vals_accumulate_ordering {
    MPIDI_ACC_ORDER_RAR = 1,
    MPIDI_ACC_ORDER_RAW = 2,
    MPIDI_ACC_ORDER_WAR = 4,
    MPIDI_ACC_ORDER_WAW = 8
};

/* We start with an arbitrarily chosen number (11), to help with
 * debugging when an window info is not initialized or wrongly
 * initialized. */
enum MPIDI_Win_info_arg_vals_accumulate_ops {
    MPIDI_ACC_OPS_SAME_OP = 11,
    MPIDI_ACC_OPS_SAME_OP_NO_OP
};

struct MPIDI_Win_info_args {
    int no_locks;               /* valid flavor = all */
    int accumulate_ordering;
    int accumulate_ops;
    int same_size;
    int same_disp_unit;
    int alloc_shared_noncontig;
    int alloc_shm;
    int accumulate_granularity;
};

struct MPIDI_RMA_op;            /* forward decl from mpidrma.h */

typedef struct MPIDI_Win_basic_info {
    void *base_addr;
    MPI_Aint size;
    int disp_unit;
    MPI_Win win_handle;
} MPIDI_Win_basic_info_t;

#define MPIDI_DEV_WIN_DECL                                               \
    volatile int at_completion_counter;  /* completion counter for operations \
                                 targeting this window */                \
    void **shm_base_addrs; /* shared memory windows -- array of base     \
                              addresses of the windows of all processes  \
                              in this process's address space */         \
    MPIDI_Win_basic_info_t *basic_info_table;                            \
    volatile int current_lock_type;   /* current lock type on this window (as target)   \
                              * (none, shared, exclusive) */             \
    volatile int shared_lock_ref_cnt;                                    \
    struct MPIDI_RMA_Target_lock_entry volatile *target_lock_queue_head;  /* list of unsatisfied locks */  \
    struct MPIDI_Win_info_args info_args;                                \
    int shm_allocated; /* flag: TRUE iff this window has a shared memory \
                          region associated with it */                   \
    struct MPIDI_RMA_Op *op_pool_start; /* start pointer used for freeing */\
    struct MPIDI_RMA_Op *op_pool_head;  /* pool of operations */              \
    struct MPIDI_RMA_Target *target_pool_start; /* start pointer used for freeing */\
    struct MPIDI_RMA_Target *target_pool_head; /* pool of targets */          \
    struct MPIDI_RMA_Slot *slots;                                        \
    int num_slots;                                                       \
    struct {                                                             \
        enum MPIDI_RMA_states access_state;                              \
        enum MPIDI_RMA_states exposure_state;                            \
    } states;                                                            \
    int num_targets_with_pending_net_ops; /* keep track of number of     \
                                             targets that has non-empty  \
                                             net pending op list. */     \
    int *start_ranks_in_win_grp;                                         \
    int start_grp_size;                                                  \
    int lock_all_assert;                                                 \
    int lock_epoch_count; /* number of lock access epoch on this process */ \
    int outstanding_locks; /* when issuing multiple lock requests in     \
                            MPI_WIN_LOCK_ALL, this counter keeps track   \
                            of number of locks not being granted yet. */ \
    struct MPIDI_RMA_Target_lock_entry *target_lock_entry_pool_start;   \
    struct MPIDI_RMA_Target_lock_entry *target_lock_entry_pool_head;    \
    int current_target_lock_data_bytes;                                 \
    int sync_request_cnt; /* This counter tracks number of              \
                             incomplete sync requests (used in          \
                             Win_fence and PSCW). */                    \
    int active; /* specify if this window is active or not */           \
    struct MPIR_Win *prev;                                              \
    struct MPIR_Win *next;                                              \
    int outstanding_acks; /* keep track of # of outstanding ACKs window \
                             wide. */                                   \

extern struct MPIR_Win *MPIDI_RMA_Win_active_list_head, *MPIDI_RMA_Win_inactive_list_head;

extern int MPIDI_CH3I_RMA_Active_req_cnt;
extern int MPIDI_CH3I_RMA_Progress_hook_id;

#ifdef MPIDI_CH3_WIN_DECL
#define MPID_DEV_WIN_DECL \
MPIDI_DEV_WIN_DECL \
MPIDI_CH3_WIN_DECL
#else
#define MPID_DEV_WIN_DECL \
MPIDI_DEV_WIN_DECL
#endif


typedef struct MPIDI_Request {
    MPIDI_Message_match match;
    MPIDI_Message_match mask;

    /* user_buf, user_count, and datatype needed to process 
       rendezvous messages. */
    void        *user_buf;
    MPI_Aint   user_count;
    MPI_Datatype datatype;
    int drop_data;

    /* msg_offset, and msgsize are used when processing
       non-contiguous datatypes */
    intptr_t msg_offset;
    intptr_t msgsize;
    intptr_t orig_msg_offset;

    /* Pointer to datatype for reference counting purposes */
    struct MPIR_Datatype* datatype_ptr;

    /* iov and iov_count define the data to be transferred/received.  
       iov_offset points to the current head element in the IOV */
    struct iovec iov[MPL_IOV_LIMIT];
    int iov_count;
    size_t iov_offset;

    /* OnDataAvail is the action to take when data is now available.
       For example, when an operation described by an iov has 
       completed.  This replaces the MPIDI_CA_t (completion action)
       field used through MPICH 1.0.4. */
    int (*OnDataAvail)( struct MPIDI_VC *, struct MPIR_Request *, int * );
    /* OnFinal is used in the following case:
       OnDataAvail is set to a function, and that function has processed
       all of the data.  At that point, the OnDataAvail function can
       reset OnDataAvail to OnFinal.  This is normally used when processing
       non-contiguous data, where there is one more action to take (such
       as a get-response) when processing of the non-contiguous data 
       completes. This value need not be initialized unless OnDataAvail
       is set to a non-null value (and then only in certain cases) */
    int (*OnFinal)( struct MPIDI_VC *, struct MPIR_Request *, int * );

    /* tmpbuf and tmpbuf_sz describe temporary storage used for things like 
       unexpected eager messages and packing/unpacking
       buffers.  tmpuf_off is the current offset into the temporary buffer. */
    void          *tmpbuf;
    intptr_t tmpbuf_off;
    intptr_t tmpbuf_sz;

    intptr_t recv_data_sz;
    MPI_Request    sender_req_id;

    unsigned int   state;
    int            cancel_pending;

    /* This field seems to be used for unexpected messages.  Unexpected messages
     * need to go through two steps: matching and receiving the data.  These
     * steps could happen in either order though, so this field is initialized
     * to 2.  It is decremented when the request is matched and also when all of
     * the data is available.  Once it reaches 0 it should be safe to copy from
     * the temporary buffer (if there is one) to the user buffer.  This field is
     * related to, but not quite the same thing as the completion counter (cc). */
    /* MT access should be controlled by the MSGQUEUE CS when the req is still
     * unexpected, exclusive access otherwise */
    int            recv_pending_count;

    /* The next several fields are used to hold state for ongoing RMA operations */
    MPI_Op op;
    /* For accumulate, since data is first read into a tmp_buf */
    void *real_user_buf;
    /* For derived datatypes at target. */
    void *flattened_type;
    /* req. handle needed to implement derived datatype gets.
     * It also used for remembering user request of request-based RMA operations. */
    MPI_Request request_handle;
    MPI_Win     target_win_handle;
    MPI_Win     source_win_handle;
    int pkt_flags; /* pkt_flags that were included in the original RMA packet header */
    struct MPIDI_RMA_Target_lock_entry *target_lock_queue_entry;
    MPI_Request resp_request_handle; /* Handle for get_accumulate response */

    void *ext_hdr_ptr; /* Pointer to extended packet header.
                        * It is allocated in RMA issuing/pkt_handler functions,
                        * and freed when release request. */
    MPI_Aint ext_hdr_sz;

    struct MPIDI_RMA_Target *rma_target_ptr;

    MPIDI_REQUEST_SEQNUM

    /* Occasionally, when a message cannot be sent, we need to cache the
       data that is required.  The fields above (such as userbuf and tmpbuf)
       are used for the message data.  However, we also need space for the
       message packet. This field provide a generic location for that.
       Question: do we want to make this a link instead of reserving 
       a fixed spot in the request? */
    MPIDI_CH3_Pkt_t pending_pkt;

    /* Notes about request_completed_cb:
     *
     *   1. The callback function is triggered when this requests
     *      completion count reaches 0.
     *
     *   2. The callback function should be nonblocking.
     *
     *   3. The callback function should not poke the progress engine,
     *      or call any function that pokes the progress engine.
     *
     *   4. The callback function can complete other requests, thus
     *      calling those requests' callback functions.  However, the
     *      recursion depth of request completion function is limited.
     *      If we ever need deeper recurisve calls, we need to change
     *      to an iterative design instead of a recursive design for
     *      request completion.
     *
     *   5. In multithreaded programs, since the callback function is
     *      nonblocking and never calls the progress engine, it would
     *      never yield the lock to other threads.  So the recursion
     *      should be multithreading-safe.
     */
    int (*request_completed_cb)(struct MPIR_Request *);

    /* partner send request when a receive request is created by the
     * sender (only used for self send) */
    struct MPIR_Request * partner_request;
} MPIDI_Request;
#define MPIR_REQUEST_DECL MPIDI_Request dev;

#if defined(MPIDI_CH3_REQUEST_DECL)
#define MPID_DEV_REQUEST_DECL			\
MPIR_REQUEST_DECL				\
MPIDI_CH3_REQUEST_DECL
#else
#define MPID_DEV_REQUEST_DECL			\
MPIR_REQUEST_DECL
#endif

#ifdef MPIDI_CH3_REQUEST_KIND_DECL
#define MPID_REQUEST_KIND_DECL MPIDI_CH3_REQUEST_KIND_DECL
#endif

#endif

/* FIXME: This ifndef test is a temp until mpidpre is cleaned of
   all items that do not belong (e.g., all items not needed by the
   top layers of MPICH) */
/* FIXME: The progress routines will be made into ch3-common definitions, not
   channel specific.  Channels that need more will need to piggy back or 
   otherwise override */
typedef struct {
#if defined(MPIDI_CH3_PROGRESS_STATE_DECL)
    MPIDI_CH3_PROGRESS_STATE_DECL
#else
    int foo;
#endif
} MPID_Progress_state;

typedef struct {
    int gpid[2];
} MPIDI_Gpid;

/* Tell initthread to prepare a private comm_world */
#define MPID_NEEDS_ICOMM_WORLD

int MPID_Init(int required, int *provided);

int MPID_InitCompleted( void );

int MPID_Allocate_vci(int *vci, bool is_shared);
int MPID_Deallocate_vci(int vci);

int MPID_Finalize(void);

int MPID_Abort( MPIR_Comm *comm, int mpi_errno, int exit_code, const char *error_msg );

int MPID_Open_port(MPIR_Info *, char *, int);
int MPID_Close_port(const char *);

int MPID_Comm_accept(const char *, MPIR_Info *, int, MPIR_Comm *, MPIR_Comm **);

int MPID_Comm_connect(const char *, MPIR_Info *, int, MPIR_Comm *, MPIR_Comm **);

int MPID_Comm_disconnect(MPIR_Comm *);

int MPID_Comm_spawn_multiple(int, char *[], char **[], const int [], MPIR_Info* [],
                             int, MPIR_Comm *, MPIR_Comm **, int []);

int MPID_Comm_failure_ack(MPIR_Comm *comm);

int MPID_Comm_failure_get_acked(MPIR_Comm *comm, MPIR_Group **failed_group_ptr);

int MPID_Comm_get_all_failed_procs(MPIR_Comm *comm_ptr, MPIR_Group **failed_group, int tag);

int MPID_Comm_revoke(MPIR_Comm *comm, int is_remote);

int MPID_Send( const void *buf, MPI_Aint count, MPI_Datatype datatype,
	       int dest, int tag, MPIR_Comm *comm, int attr,
	       MPIR_Request **request );

int MPID_Rsend( const void *buf, MPI_Aint count, MPI_Datatype datatype,
		int dest, int tag, MPIR_Comm *comm, int attr,
		MPIR_Request **request );

int MPID_Ssend( const void *buf, MPI_Aint count, MPI_Datatype datatype,
		int dest, int tag, MPIR_Comm *comm, int attr,
		MPIR_Request **request );

int MPID_Isend( const void *buf, MPI_Aint count, MPI_Datatype datatype,
		int dest, int tag, MPIR_Comm *comm, int attr,
		MPIR_Request **request );

int MPID_Irsend( const void *buf, MPI_Aint count, MPI_Datatype datatype,
		 int dest, int tag, MPIR_Comm *comm, int attr,
		 MPIR_Request **request );

int MPID_Issend( const void *buf, MPI_Aint count, MPI_Datatype datatype,
		 int dest, int tag, MPIR_Comm *comm, int attr,
		 MPIR_Request **request );

int MPID_Recv( void *buf, MPI_Aint count, MPI_Datatype datatype,
	       int source, int tag, MPIR_Comm *comm, int attr,
	       MPI_Status *status, MPIR_Request **request );

int MPID_Irecv( void *buf, MPI_Aint count, MPI_Datatype datatype,
		int source, int tag, MPIR_Comm *comm, int attr,
		MPIR_Request **request );

int MPID_Send_init( const void *buf, MPI_Aint count, MPI_Datatype datatype,
		    int dest, int tag, MPIR_Comm *comm, int attr,
		    MPIR_Request **request );

int MPID_Bsend_init(const void *, MPI_Aint, MPI_Datatype, int, int, MPIR_Comm *,
		   int, MPIR_Request **);
int MPID_Rsend_init( const void *buf, MPI_Aint count, MPI_Datatype datatype,
		     int dest, int tag, MPIR_Comm *comm, int attr,
		     MPIR_Request **request );
int MPID_Ssend_init( const void *buf, MPI_Aint count, MPI_Datatype datatype,
		     int dest, int tag, MPIR_Comm *comm, int attr,
		     MPIR_Request **request );

int MPID_Recv_init( void *buf, MPI_Aint count, MPI_Datatype datatype,
		    int source, int tag, MPIR_Comm *comm, int attr,
		    MPIR_Request **request );

int MPID_Startall(int count, MPIR_Request *requests[]);

int MPID_Bcast_init(void *buffer, MPI_Aint count, MPI_Datatype datatype, int root,
                    MPIR_Comm *comm_ptr, MPIR_Info* info_ptr, MPIR_Request **request);

int MPID_Allreduce_init(const void *sendbuf, void *recvbuf, MPI_Aint count, MPI_Datatype datatype,
                        MPI_Op op, MPIR_Comm * comm_ptr, MPIR_Info * info_ptr,
                        MPIR_Request ** request);

int MPID_Reduce_init(const void *sendbuf, void *recvbuf, MPI_Aint count, MPI_Datatype datatype,
                     MPI_Op op, int root, MPIR_Comm *comm_ptr, MPIR_Info* info_ptr,
                     MPIR_Request **request);

int MPID_Alltoall_init(const void *sendbuf, MPI_Aint sendcount, MPI_Datatype sendtype,
                       void *recvbuf, MPI_Aint recvcount, MPI_Datatype recvtype,
                       MPIR_Comm *comm_ptr, MPIR_Info* info_ptr, MPIR_Request** request);

int MPID_Alltoallv_init(const void *sendbuf, const MPI_Aint sendcounts[], const MPI_Aint sdispls[],
                        MPI_Datatype sendtype, void *recvbuf, const MPI_Aint recvcounts[],
                        const MPI_Aint rdispls[], MPI_Datatype recvtype, MPIR_Comm * comm_ptr,
                        MPIR_Info * info_ptr, MPIR_Request ** request);

int MPID_Alltoallw_init(const void *sendbuf, const MPI_Aint sendcounts[], const MPI_Aint sdispls[],
                        const MPI_Datatype sendtypes[], void *recvbuf, const MPI_Aint recvcounts[],
                        const MPI_Aint rdispls[], const MPI_Datatype recvtypes[],
                        MPIR_Comm * comm_ptr, MPIR_Info * info_ptr, MPIR_Request ** request);

int MPID_Allgather_init(const void *sendbuf, MPI_Aint sendcount, MPI_Datatype sendtype,
                        void *recvbuf, MPI_Aint recvcount, MPI_Datatype recvtype,
                        MPIR_Comm *comm_ptr, MPIR_Info* info_ptr, MPIR_Request** request);

int MPID_Allgatherv_init(const void *sendbuf, MPI_Aint sendcount, MPI_Datatype sendtype,
                         void *recvbuf, const MPI_Aint * recvcounts, const MPI_Aint * displs,
                         MPI_Datatype recvtype, MPIR_Comm *comm_ptr, MPIR_Info* info_ptr,
                         MPIR_Request** request);

int MPID_Reduce_scatter_block_init(const void *sendbuf, void *recvbuf, MPI_Aint recvcount,
                                   MPI_Datatype datatype, MPI_Op op, MPIR_Comm * comm,
                                   MPIR_Info * info, MPIR_Request ** request);

int MPID_Reduce_scatter_init(const void *sendbuf, void *recvbuf, const MPI_Aint recvcounts[],
                             MPI_Datatype datatype, MPI_Op op, MPIR_Comm * comm, MPIR_Info * info,
                             MPIR_Request ** request);

int MPID_Scan_init(const void *sendbuf, void *recvbuf, MPI_Aint count, MPI_Datatype datatype,
                   MPI_Op op, MPIR_Comm * comm, MPIR_Info * info, MPIR_Request ** request);

int MPID_Gather_init(const void *sendbuf, MPI_Aint sendcount, MPI_Datatype sendtype, void *recvbuf,
                     MPI_Aint recvcount, MPI_Datatype recvtype, int root, MPIR_Comm * comm,
                     MPIR_Info * info, MPIR_Request ** request);

int MPID_Gatherv_init(const void *sendbuf, MPI_Aint sendcount, MPI_Datatype sendtype, void *recvbuf,
                      const MPI_Aint recvcounts[], const MPI_Aint displs[], MPI_Datatype recvtype,
                      int root, MPIR_Comm * comm, MPIR_Info * info, MPIR_Request ** request);

int MPID_Scatter_init(const void *sendbuf, MPI_Aint sendcount, MPI_Datatype sendtype, void *recvbuf,
                      MPI_Aint recvcount, MPI_Datatype recvtype, int root, MPIR_Comm * comm,
                      MPIR_Info * info, MPIR_Request ** request);

int MPID_Scatterv_init(const void *sendbuf, const MPI_Aint sendcounts[], const MPI_Aint displs[],
                       MPI_Datatype sendtype, void *recvbuf, MPI_Aint recvcount,
                       MPI_Datatype recvtype, int root, MPIR_Comm * comm, MPIR_Info * info,
                       MPIR_Request ** request);

int MPID_Barrier_init(MPIR_Comm * comm, MPIR_Info * info, MPIR_Request ** request);

int MPID_Exscan_init(const void *sendbuf, void *recvbuf, MPI_Aint count, MPI_Datatype datatype,
                     MPI_Op op, MPIR_Comm * comm, MPIR_Info * info, MPIR_Request ** request);

int MPID_Neighbor_allgather_init(const void *sendbuf, MPI_Aint sendcount, MPI_Datatype sendtype,
                                 void *recvbuf, MPI_Aint recvcount, MPI_Datatype recvtype,
                                 MPIR_Comm * comm, MPIR_Info * info, MPIR_Request ** request);

int MPID_Neighbor_allgatherv_init(const void *sendbuf, MPI_Aint sendcount, MPI_Datatype sendtype,
                                  void *recvbuf, const MPI_Aint recvcounts[],
                                  const MPI_Aint displs[], MPI_Datatype recvtype, MPIR_Comm * comm,
                                  MPIR_Info * info, MPIR_Request ** request);

int MPID_Neighbor_alltoall_init(const void *sendbuf, MPI_Aint sendcount, MPI_Datatype sendtype,
                                void *recvbuf, MPI_Aint recvcount, MPI_Datatype recvtype,
                                MPIR_Comm * comm, MPIR_Info * info, MPIR_Request ** request);


int MPID_Neighbor_alltoallv_init(const void *sendbuf, const MPI_Aint sendcounts[],
                                 const MPI_Aint sdispls[], MPI_Datatype sendtype, void *recvbuf,
                                 const MPI_Aint recvcounts[], const MPI_Aint rdispls[],
                                 MPI_Datatype recvtype, MPIR_Comm * comm, MPIR_Info * info,
                                 MPIR_Request ** request);

int MPID_Neighbor_alltoallw_init(const void *sendbuf, const MPI_Aint sendcounts[],
                                 const MPI_Aint sdispls[], const MPI_Datatype sendtypes[],
                                 void *recvbuf, const MPI_Aint recvcounts[],
                                 const MPI_Aint rdispls[], const MPI_Datatype recvtypes[],
                                 MPIR_Comm * comm, MPIR_Info * info, MPIR_Request ** request);

int MPID_Probe(int source, int tag, MPIR_Comm *comm, int attr, MPI_Status * status);
int MPID_Iprobe(int source, int tag, MPIR_Comm *comm, int attr, int *flag, MPI_Status *status);

int MPID_Mprobe(int source, int tag, MPIR_Comm *comm, int attr,
                MPIR_Request **message, MPI_Status *status);

int MPID_Improbe(int source, int tag, MPIR_Comm *comm, int attr,
                 int *flag, MPIR_Request **message, MPI_Status *status);

int MPID_Imrecv(void *buf, MPI_Aint count, MPI_Datatype datatype,
                MPIR_Request *message, MPIR_Request **rreqp);

int MPID_Mrecv(void *buf, MPI_Aint count, MPI_Datatype datatype,
               MPIR_Request *message, MPI_Status *status, MPIR_Request **rreq);

int MPID_Cancel_send(MPIR_Request *);
int MPID_Cancel_recv(MPIR_Request *);

int MPID_Psend_init(const void *buf, int partitions, MPI_Count count, MPI_Datatype datatype,
                    int dest, int tag, MPIR_Comm *comm, MPIR_Info *info,
                    MPIR_Request **request );
int MPID_Precv_init(void *buf, int partitions, MPI_Count count, MPI_Datatype datatype,
                    int source, int tag, MPIR_Comm *comm, MPIR_Info *info,
                    MPIR_Request **request );

int MPID_Pready_range(int partition_low, int partition_high, MPIR_Request *sreq);
int MPID_Pready_list(int length, const int array_of_partitions[], MPIR_Request *sreq);
int MPID_Parrived(MPIR_Request *rreq, int partition, int *flag);

MPI_Aint MPID_Aint_add(MPI_Aint base, MPI_Aint disp);

MPI_Aint MPID_Aint_diff(MPI_Aint addr1, MPI_Aint addr2);

int MPID_Win_create(void *, MPI_Aint, MPI_Aint, MPIR_Info *, MPIR_Comm *,
                    MPIR_Win **);
int MPID_Win_free(MPIR_Win **);

int MPID_Put(const void *, MPI_Aint, MPI_Datatype, int, MPI_Aint, MPI_Aint,
             MPI_Datatype, MPIR_Win *);
int MPID_Get(void *, MPI_Aint, MPI_Datatype, int, MPI_Aint, MPI_Aint,
             MPI_Datatype, MPIR_Win *);
int MPID_Accumulate(const void *, MPI_Aint, MPI_Datatype, int, MPI_Aint, MPI_Aint,
                    MPI_Datatype, MPI_Op, MPIR_Win *);

int MPID_Win_fence(int, MPIR_Win *);
int MPID_Win_post(MPIR_Group *group_ptr, int assert, MPIR_Win *win_ptr);
int MPID_Win_start(MPIR_Group *group_ptr, int assert, MPIR_Win *win_ptr);
int MPID_Win_test(MPIR_Win *win_ptr, int *flag);
int MPID_Win_wait(MPIR_Win *win_ptr);
int MPID_Win_complete(MPIR_Win *win_ptr);

int MPID_Win_lock(int lock_type, int dest, int assert, MPIR_Win *win_ptr);
int MPID_Win_unlock(int dest, MPIR_Win *win_ptr);

int MPID_Win_allocate(MPI_Aint size, MPI_Aint disp_unit, MPIR_Info *info,
                      MPIR_Comm *comm, void *baseptr, MPIR_Win **win);
int MPID_Win_allocate_shared(MPI_Aint size, MPI_Aint disp_unit, MPIR_Info *info_ptr, MPIR_Comm *comm_ptr,
                             void *base_ptr, MPIR_Win **win_ptr);
int MPID_Win_shared_query(MPIR_Win *win, int rank, MPI_Aint *size, int *disp_unit,
                          void *baseptr);
int MPID_Win_create_dynamic(MPIR_Info *info, MPIR_Comm *comm, MPIR_Win **win);
int MPID_Win_attach(MPIR_Win *win, void *base, MPI_Aint size);
int MPID_Win_detach(MPIR_Win *win, const void *base);
int MPID_Win_get_info(MPIR_Win *win, MPIR_Info **info_used);
int MPID_Win_set_info(MPIR_Win *win, MPIR_Info *info);

int MPID_Get_accumulate(const void *origin_addr, MPI_Aint origin_count,
                        MPI_Datatype origin_datatype, void *result_addr, MPI_Aint result_count,
                        MPI_Datatype result_datatype, int target_rank, MPI_Aint target_disp,
                        MPI_Aint target_count, MPI_Datatype target_datatype, MPI_Op op, MPIR_Win *win);
int MPID_Fetch_and_op(const void *origin_addr, void *result_addr,
                      MPI_Datatype datatype, int target_rank, MPI_Aint target_disp,
                      MPI_Op op, MPIR_Win *win);
int MPID_Compare_and_swap(const void *origin_addr, const void *compare_addr,
                          void *result_addr, MPI_Datatype datatype, int target_rank,
                          MPI_Aint target_disp, MPIR_Win *win);
int MPID_Rput(const void *origin_addr, MPI_Aint origin_count,
              MPI_Datatype origin_datatype, int target_rank, MPI_Aint target_disp,
              MPI_Aint target_count, MPI_Datatype target_datatype, MPIR_Win *win,
              MPIR_Request **request);
int MPID_Rget(void *origin_addr, MPI_Aint origin_count,
              MPI_Datatype origin_datatype, int target_rank, MPI_Aint target_disp,
              MPI_Aint target_count, MPI_Datatype target_datatype, MPIR_Win *win,
              MPIR_Request **request);
int MPID_Raccumulate(const void *origin_addr, MPI_Aint origin_count,
                     MPI_Datatype origin_datatype, int target_rank, MPI_Aint target_disp,
                     MPI_Aint target_count, MPI_Datatype target_datatype, MPI_Op op, MPIR_Win *win,
                     MPIR_Request **request);
int MPID_Rget_accumulate(const void *origin_addr, MPI_Aint origin_count,
                         MPI_Datatype origin_datatype, void *result_addr, MPI_Aint result_count,
                         MPI_Datatype result_datatype, int target_rank, MPI_Aint target_disp,
                         MPI_Aint target_count, MPI_Datatype target_datatype, MPI_Op op, MPIR_Win *win,
                         MPIR_Request **request);

int MPID_Win_lock_all(int assert, MPIR_Win *win);
int MPID_Win_unlock_all(MPIR_Win *win);
int MPID_Win_flush(int rank, MPIR_Win *win);
int MPID_Win_flush_all(MPIR_Win *win);
int MPID_Win_flush_local(int rank, MPIR_Win *win);
int MPID_Win_flush_local_all(MPIR_Win *win);
int MPID_Win_sync(MPIR_Win *win);

int MPID_Send_enqueue(const void *buf, MPI_Aint count, MPI_Datatype datatype,
                      int dest, int tag, MPIR_Comm * comm_ptr);
int MPID_Recv_enqueue(void *buf, MPI_Aint count, MPI_Datatype datatype,
                      int source, int tag, MPIR_Comm * comm_ptr, MPI_Status * status);
int MPID_Isend_enqueue(const void *buf, MPI_Aint count, MPI_Datatype datatype,
                       int dest, int tag, MPIR_Comm * comm_ptr, MPIR_Request ** req);
int MPID_Irecv_enqueue(void *buf, MPI_Aint count, MPI_Datatype datatype,
                       int source, int tag, MPIR_Comm * comm_ptr, MPIR_Request ** req);
int MPID_Wait_enqueue(MPIR_Request * req_ptr, MPI_Status * status);
int MPID_Waitall_enqueue(int count, MPI_Request * array_of_requests,
                         MPI_Status * array_of_statuses);

void MPID_Progress_start(MPID_Progress_state * state);
int MPID_Progress_wait(MPID_Progress_state * state);
void MPID_Progress_end(MPID_Progress_state * state);
int MPID_Progress_poke(void);

int MPID_Get_processor_name( char *name, int namelen, int *resultlen);
int MPID_Get_universe_size(int  * universe_size);

#define MPID_Request_create_from_comm(kind, comm) MPIR_Request_create(kind)
void MPID_Request_create_hook(MPIR_Request *);
void MPID_Request_free_hook(MPIR_Request *);
void MPID_Request_destroy_hook(MPIR_Request *);
#define MPID_Prequest_free_hook(req_) do {} while(0)
#define MPID_Part_request_free_hook(req_) do {} while(0)
int MPID_Request_complete(MPIR_Request *);

void *MPID_Alloc_mem(MPI_Aint size, MPIR_Info *info );
int MPID_Free_mem( void *ptr );

/* Prototypes and definitions for the node ID code.  This is used to support
   hierarchical collectives in a (mostly) device-independent way. */
int MPID_Get_node_id(MPIR_Comm *comm, int rank, int *id_p);
int MPID_Get_max_node_id(MPIR_Comm *comm, int *max_id_p);

int MPID_Type_commit_hook(MPIR_Datatype * type);
int MPID_Type_free_hook(MPIR_Datatype * type);
int MPID_Op_commit_hook(MPIR_Op * op);
int MPID_Op_free_hook(MPIR_Op * op);

MPL_STATIC_INLINE_PREFIX int MPID_Stream_create_hook(MPIR_Stream * stream)
{
    return MPI_SUCCESS;
}

MPL_STATIC_INLINE_PREFIX int MPID_Stream_free_hook(MPIR_Stream * stream)
{
    return MPI_SUCCESS;
}

#endif /* MPIDPRE_H_INCLUDED */