1
1
openmpi/ompi/mca/mtl/portals/mtl_portals.h

250 строки
9.3 KiB
C
Исходник Обычный вид История

/*
* Copyright (c) 2004-2006 The Trustees of Indiana University and Indiana
* University Research and Technology
* Corporation. All rights reserved.
* Copyright (c) 2004-2007 The University of Tennessee and The University
* of Tennessee Research Foundation. All rights
* reserved.
* Copyright (c) 2004-2005 High Performance Computing Center Stuttgart,
* University of Stuttgart. All rights reserved.
* Copyright (c) 2004-2005 The Regents of the University of California.
* All rights reserved.
* $COPYRIGHT$
*
* Additional copyrights may follow
*
* $HEADER$
*/
#ifndef MTL_PORTALS_H_HAS_BEEN_INCLUDED
#define MTL_PORTALS_H_HAS_BEEN_INCLUDED
#include "ompi_config.h"
#include "opal/class/opal_list.h"
#include "ompi/class/ompi_free_list.h"
#include "ompi/mca/mtl/mtl.h"
#include "ompi/mca/mtl/base/base.h"
- Split the datatype engine into two parts: an MPI specific part in OMPI and a language agnostic part in OPAL. The convertor is completely moved into OPAL. This offers several benefits as described in RFC http://www.open-mpi.org/community/lists/devel/2009/07/6387.php namely: - Fewer basic types (int* and float* types, boolean and wchar - Fixing naming scheme to ompi-nomenclature. - Usability outside of the ompi-layer. - Due to the fixed nature of simple opal types, their information is completely known at compile time and therefore constified - With fewer datatypes (22), the actual sizes of bit-field types may be reduced from 64 to 32 bits, allowing reorganizing the opal_datatype structure, eliminating holes and keeping data required in convertor (upon send/recv) in one cacheline... This has implications to the convertor-datastructure and other parts of the code. - Several performance tests have been run, the netpipe latency does not change with this patch on Linux/x86-64 on the smoky cluster. - Extensive tests have been done to verify correctness (no new regressions) using: 1. mpi_test_suite on linux/x86-64 using clean ompi-trunk and ompi-ddt: a. running both trunk and ompi-ddt resulted in no differences (except for MPI_SHORT_INT and MPI_TYPE_MIX_LB_UB do now run correctly). b. with --enable-memchecker and running under valgrind (one buglet when run with static found in test-suite, commited) 2. ibm testsuite on linux/x86-64 using clean ompi-trunk and ompi-ddt: all passed (except for the dynamic/ tests failed!! as trunk/MTT) 3. compilation and usage of HDF5 tests on Jaguar using PGI and PathScale compilers. 4. compilation and usage on Scicortex. - Please note, that for the heterogeneous case, (-m32 compiled binaries/ompi), neither ompi-trunk, nor ompi-ddt branch would successfully launch. This commit was SVN r21641.
2009-07-13 04:56:31 +00:00
#include "opal/datatype/opal_convertor.h"
#include "ompi/mca/common/portals/common_portals.h"
- Replace combinations of #if defined (c_plusplus) defined (__cplusplus) followed by extern "C" { and the closing counterpart by BEGIN_C_DECLS and END_C_DECLS. Notable exceptions are: - opal/include/opal_config_bottom.h: This is our generated code, that itself defines BEGIN_C_DECL and END_C_DECL - ompi/mpi/cxx/mpicxx.h: Here we do not include opal_config_bottom.h: - Belongs to external code: opal/mca/backtrace/darwin/MoreBacktrace/MoreDebugging/MoreBacktrace.c opal/mca/backtrace/darwin/MoreBacktrace/MoreDebugging/MoreBacktrace.h - opal/include/opal/prefetch.h: Has C++ specific macros that are protected: - Had #if ... } #endif _and_ END_C_DECLS (aka end up with 2x END_C_DECLS) ompi/mca/btl/openib/btl_openib.h - opal/event/event.h has #ifdef __cplusplus as BEGIN_C_DECLS... - opal/win32/ompi_process.h: had extern "C"\n {... opal/win32/ompi_process.h: dito - ompi/mca/btl/pcie/btl_pcie_lex.l: needed to add *_C_DECLS ompi/mpi/f90/test/align_c.c: dito - ompi/debuggers/msgq_interface.h: used #ifdef __cplusplus - ompi/mpi/f90/xml/common-C.xsl: Amend Tested on linux using --with-openib and --with-mx The following do not contain either opal_config.h, orte_config.h or ompi_config.h (but possibly other header files, that include one of the above): ompi/mca/bml/r2/bml_r2_ft.h ompi/mca/btl/gm/btl_gm_endpoint.h ompi/mca/btl/gm/btl_gm_proc.h ompi/mca/btl/mx/btl_mx_endpoint.h ompi/mca/btl/ofud/btl_ofud_endpoint.h ompi/mca/btl/ofud/btl_ofud_frag.h ompi/mca/btl/ofud/btl_ofud_proc.h ompi/mca/btl/openib/btl_openib_mca.h ompi/mca/btl/portals/btl_portals_endpoint.h ompi/mca/btl/portals/btl_portals_frag.h ompi/mca/btl/sctp/btl_sctp_endpoint.h ompi/mca/btl/sctp/btl_sctp_proc.h ompi/mca/btl/tcp/btl_tcp_endpoint.h ompi/mca/btl/tcp/btl_tcp_ft.h ompi/mca/btl/tcp/btl_tcp_proc.h ompi/mca/btl/template/btl_template_endpoint.h ompi/mca/btl/template/btl_template_proc.h ompi/mca/btl/udapl/btl_udapl_eager_rdma.h ompi/mca/btl/udapl/btl_udapl_endpoint.h ompi/mca/btl/udapl/btl_udapl_mca.h ompi/mca/btl/udapl/btl_udapl_proc.h ompi/mca/mtl/mx/mtl_mx_endpoint.h ompi/mca/mtl/mx/mtl_mx.h ompi/mca/mtl/psm/mtl_psm_endpoint.h ompi/mca/mtl/psm/mtl_psm.h ompi/mca/pml/cm/pml_cm_component.h ompi/mca/pml/csum/pml_csum_comm.h ompi/mca/pml/dr/pml_dr_comm.h ompi/mca/pml/dr/pml_dr_component.h ompi/mca/pml/dr/pml_dr_endpoint.h ompi/mca/pml/dr/pml_dr_recvfrag.h ompi/mca/pml/example/pml_example.h ompi/mca/pml/ob1/pml_ob1_comm.h ompi/mca/pml/ob1/pml_ob1_component.h ompi/mca/pml/ob1/pml_ob1_endpoint.h ompi/mca/pml/ob1/pml_ob1_rdmafrag.h ompi/mca/pml/ob1/pml_ob1_recvfrag.h ompi/mca/pml/v/pml_v_output.h opal/include/opal/prefetch.h opal/mca/timer/aix/timer_aix.h opal/util/qsort.h test/support/components.h This commit was SVN r21855. The following SVN revision numbers were found above: r2 --> open-mpi/ompi@58fdc188553052bc2e893ba28fb28fddbe78435a
2009-08-20 11:42:18 +00:00
BEGIN_C_DECLS
struct mca_mtl_portals_module_t {
mca_mtl_base_module_t base;
ptl_handle_ni_t ptl_ni_h;
size_t eager_limit;
ptl_handle_eq_t ptl_eq_h;
ptl_handle_eq_t ptl_unex_eq_h;
/* long unex msgs - insert posted recvs before this */
ptl_handle_me_t ptl_unex_long_me_h;
/* send catchall - insert short unex buffers before this */
ptl_handle_me_t ptl_send_catchall_me_h;
/* catchall for ack portal */
ptl_handle_me_t ptl_ack_catchall_me_h;
/* catchall for read portal */
ptl_handle_me_t ptl_read_catchall_me_h;
/* for zero-length sends and acks */
ptl_handle_md_t ptl_zero_md_h;
ompi_free_list_t event_fl;
int ptl_recv_short_mds_num;
int ptl_recv_short_mds_size;
opal_list_t ptl_recv_short_blocks;
opal_list_t unexpected_messages;
int ptl_expected_queue_size;
int ptl_unexpected_queue_size;
/* for send-side copy blocks */
ptl_md_t ptl_short_md;
ptl_handle_md_t ptl_short_md_h;
int ptl_num_copy_blocks;
ptl_size_t ptl_copy_block_len;
int *ptl_copy_block_free_list;
int ptl_copy_block_first_free;
/* empty event queue for PtlMEMDPost() */
ptl_handle_eq_t ptl_empty_eq_h;
/* turn off aggressive polling of the unex msg event queue */
bool ptl_aggressive_polling;
/* use rendezvous for long messages */
bool ptl_use_rendezvous;
/* output channel for debugging */
int portals_output;
};
typedef struct mca_mtl_portals_module_t mca_mtl_portals_module_t;
extern mca_mtl_portals_module_t ompi_mtl_portals;
OMPI_DECLSPEC mca_mtl_base_component_2_0_0_t mca_mtl_portals_component;
struct ompi_mtl_portals_event_t {
struct ompi_free_list_item_t super;
ptl_event_t ev;
bool is_complete;
};
typedef struct ompi_mtl_portals_event_t ompi_mtl_portals_event_t;
OBJ_CLASS_DECLARATION(ompi_mtl_portals_event_t);
/* match/ignore bit manipulation
*
* 0123 4567 01234567 01234567 01234567 01234567 01234567 01234567 01234567
* | | |
* ^ | context id | source | message tag
* | | | |
* +---- protocol
*/
#define PTL_PROTOCOL_MASK 0xF000000000000000ULL
#define PTL_CONTEXT_MASK 0x0FFF000000000000ULL
#define PTL_SOURCE_MASK 0x0000FFFF00000000ULL
#define PTL_TAG_MASK 0x00000000FFFFFFFFULL
#define PTL_PROTOCOL_IGNR PTL_PROTOCOL_MASK
#define PTL_CONTEXT_IGNR PTL_CONTEXT_MASK
#define PTL_SOURCE_IGNR PTL_SOURCE_MASK
#define PTL_TAG_IGNR 0x000000007FFFFFFFULL
#define PTL_SHORT_MSG 0x1000000000000000ULL
#define PTL_LONG_MSG 0x2000000000000000ULL
#define PTL_READY_MSG 0x4000000000000000ULL
/* send posting */
#define PTL_SET_SEND_BITS(match_bits, contextid, source, tag, type) \
{ \
match_bits = contextid; \
match_bits = (match_bits << 16); \
match_bits |= source; \
match_bits = (match_bits << 32); \
match_bits |= (PTL_TAG_MASK & tag) | type; \
}
/* receive posting */
#define PTL_SET_RECV_BITS(match_bits, ignore_bits, contextid, source, tag) \
{ \
match_bits = 0; \
ignore_bits = PTL_PROTOCOL_IGNR; \
\
match_bits = contextid; \
match_bits = (match_bits << 16); \
\
if (MPI_ANY_SOURCE == source) { \
match_bits = (match_bits << 32); \
ignore_bits |= PTL_SOURCE_IGNR; \
} else { \
match_bits |= source; \
match_bits = (match_bits << 32); \
} \
\
if (MPI_ANY_TAG == tag) { \
ignore_bits |= PTL_TAG_IGNR; \
} else { \
match_bits |= (PTL_TAG_MASK & tag); \
} \
}
#define PTL_IS_SHORT_MSG(match_bits) \
(0 != (PTL_SHORT_MSG & match_bits))
#define PTL_IS_LONG_MSG(match_bits) \
(0 != (PTL_LONG_MSG & match_bits))
#define PTL_IS_READY_MSG(match_bits) \
(0 != (PTL_READY_MSG & match_bits))
#define PTL_IS_SYNC_MSG(event) \
(0 != event.hdr_data)
#define PTL_GET_TAG(match_bits) ((int)(match_bits & PTL_TAG_MASK))
#define PTL_GET_SOURCE(match_bits) ((int)((match_bits & PTL_SOURCE_MASK) >> 32))
/* MTL interface functions */
extern int ompi_mtl_portals_finalize(struct mca_mtl_base_module_t *mtl);
extern int ompi_mtl_portals_add_procs(struct mca_mtl_base_module_t* mtl,
size_t nprocs,
struct ompi_proc_t** procs,
struct mca_mtl_base_endpoint_t **mtl_peer_data);
extern int ompi_mtl_portals_del_procs(struct mca_mtl_base_module_t* mtl,
size_t nprocs,
struct ompi_proc_t** procs,
struct mca_mtl_base_endpoint_t **mtl_peer_data);
extern int ompi_mtl_portals_send(struct mca_mtl_base_module_t* mtl,
struct ompi_communicator_t* comm,
int dest,
int tag,
- Split the datatype engine into two parts: an MPI specific part in OMPI and a language agnostic part in OPAL. The convertor is completely moved into OPAL. This offers several benefits as described in RFC http://www.open-mpi.org/community/lists/devel/2009/07/6387.php namely: - Fewer basic types (int* and float* types, boolean and wchar - Fixing naming scheme to ompi-nomenclature. - Usability outside of the ompi-layer. - Due to the fixed nature of simple opal types, their information is completely known at compile time and therefore constified - With fewer datatypes (22), the actual sizes of bit-field types may be reduced from 64 to 32 bits, allowing reorganizing the opal_datatype structure, eliminating holes and keeping data required in convertor (upon send/recv) in one cacheline... This has implications to the convertor-datastructure and other parts of the code. - Several performance tests have been run, the netpipe latency does not change with this patch on Linux/x86-64 on the smoky cluster. - Extensive tests have been done to verify correctness (no new regressions) using: 1. mpi_test_suite on linux/x86-64 using clean ompi-trunk and ompi-ddt: a. running both trunk and ompi-ddt resulted in no differences (except for MPI_SHORT_INT and MPI_TYPE_MIX_LB_UB do now run correctly). b. with --enable-memchecker and running under valgrind (one buglet when run with static found in test-suite, commited) 2. ibm testsuite on linux/x86-64 using clean ompi-trunk and ompi-ddt: all passed (except for the dynamic/ tests failed!! as trunk/MTT) 3. compilation and usage of HDF5 tests on Jaguar using PGI and PathScale compilers. 4. compilation and usage on Scicortex. - Please note, that for the heterogeneous case, (-m32 compiled binaries/ompi), neither ompi-trunk, nor ompi-ddt branch would successfully launch. This commit was SVN r21641.
2009-07-13 04:56:31 +00:00
struct opal_convertor_t *convertor,
mca_pml_base_send_mode_t mode);
extern int ompi_mtl_portals_isend(struct mca_mtl_base_module_t* mtl,
struct ompi_communicator_t* comm,
int dest,
int tag,
- Split the datatype engine into two parts: an MPI specific part in OMPI and a language agnostic part in OPAL. The convertor is completely moved into OPAL. This offers several benefits as described in RFC http://www.open-mpi.org/community/lists/devel/2009/07/6387.php namely: - Fewer basic types (int* and float* types, boolean and wchar - Fixing naming scheme to ompi-nomenclature. - Usability outside of the ompi-layer. - Due to the fixed nature of simple opal types, their information is completely known at compile time and therefore constified - With fewer datatypes (22), the actual sizes of bit-field types may be reduced from 64 to 32 bits, allowing reorganizing the opal_datatype structure, eliminating holes and keeping data required in convertor (upon send/recv) in one cacheline... This has implications to the convertor-datastructure and other parts of the code. - Several performance tests have been run, the netpipe latency does not change with this patch on Linux/x86-64 on the smoky cluster. - Extensive tests have been done to verify correctness (no new regressions) using: 1. mpi_test_suite on linux/x86-64 using clean ompi-trunk and ompi-ddt: a. running both trunk and ompi-ddt resulted in no differences (except for MPI_SHORT_INT and MPI_TYPE_MIX_LB_UB do now run correctly). b. with --enable-memchecker and running under valgrind (one buglet when run with static found in test-suite, commited) 2. ibm testsuite on linux/x86-64 using clean ompi-trunk and ompi-ddt: all passed (except for the dynamic/ tests failed!! as trunk/MTT) 3. compilation and usage of HDF5 tests on Jaguar using PGI and PathScale compilers. 4. compilation and usage on Scicortex. - Please note, that for the heterogeneous case, (-m32 compiled binaries/ompi), neither ompi-trunk, nor ompi-ddt branch would successfully launch. This commit was SVN r21641.
2009-07-13 04:56:31 +00:00
struct opal_convertor_t *convertor,
mca_pml_base_send_mode_t mode,
bool blocking,
mca_mtl_request_t *mtl_request);
extern int ompi_mtl_portals_irecv(struct mca_mtl_base_module_t* mtl,
struct ompi_communicator_t *comm,
int src,
int tag,
- Split the datatype engine into two parts: an MPI specific part in OMPI and a language agnostic part in OPAL. The convertor is completely moved into OPAL. This offers several benefits as described in RFC http://www.open-mpi.org/community/lists/devel/2009/07/6387.php namely: - Fewer basic types (int* and float* types, boolean and wchar - Fixing naming scheme to ompi-nomenclature. - Usability outside of the ompi-layer. - Due to the fixed nature of simple opal types, their information is completely known at compile time and therefore constified - With fewer datatypes (22), the actual sizes of bit-field types may be reduced from 64 to 32 bits, allowing reorganizing the opal_datatype structure, eliminating holes and keeping data required in convertor (upon send/recv) in one cacheline... This has implications to the convertor-datastructure and other parts of the code. - Several performance tests have been run, the netpipe latency does not change with this patch on Linux/x86-64 on the smoky cluster. - Extensive tests have been done to verify correctness (no new regressions) using: 1. mpi_test_suite on linux/x86-64 using clean ompi-trunk and ompi-ddt: a. running both trunk and ompi-ddt resulted in no differences (except for MPI_SHORT_INT and MPI_TYPE_MIX_LB_UB do now run correctly). b. with --enable-memchecker and running under valgrind (one buglet when run with static found in test-suite, commited) 2. ibm testsuite on linux/x86-64 using clean ompi-trunk and ompi-ddt: all passed (except for the dynamic/ tests failed!! as trunk/MTT) 3. compilation and usage of HDF5 tests on Jaguar using PGI and PathScale compilers. 4. compilation and usage on Scicortex. - Please note, that for the heterogeneous case, (-m32 compiled binaries/ompi), neither ompi-trunk, nor ompi-ddt branch would successfully launch. This commit was SVN r21641.
2009-07-13 04:56:31 +00:00
struct opal_convertor_t *convertor,
mca_mtl_request_t *mtl_request);
extern int ompi_mtl_portals_iprobe(struct mca_mtl_base_module_t* mtl,
struct ompi_communicator_t *comm,
int src,
int tag,
int *flag,
struct ompi_status_public_t *status);
extern int ompi_mtl_portals_imrecv(struct mca_mtl_base_module_t* mtl,
struct opal_convertor_t *convertor,
struct ompi_message_t **message,
struct mca_mtl_request_t *mtl_request);
extern int ompi_mtl_portals_improbe(struct mca_mtl_base_module_t *mtl,
struct ompi_communicator_t *comm,
int src,
int tag,
int *matched,
struct ompi_message_t **message,
struct ompi_status_public_t *status);
extern int ompi_mtl_portals_cancel(struct mca_mtl_base_module_t* mtl,
mca_mtl_request_t *mtl_request,
int flag);
extern int ompi_mtl_portals_add_comm(struct mca_mtl_base_module_t *mtl,
struct ompi_communicator_t *comm);
extern int ompi_mtl_portals_del_comm(struct mca_mtl_base_module_t *mtl,
struct ompi_communicator_t *comm);
extern int ompi_mtl_portals_progress(void);
- Replace combinations of #if defined (c_plusplus) defined (__cplusplus) followed by extern "C" { and the closing counterpart by BEGIN_C_DECLS and END_C_DECLS. Notable exceptions are: - opal/include/opal_config_bottom.h: This is our generated code, that itself defines BEGIN_C_DECL and END_C_DECL - ompi/mpi/cxx/mpicxx.h: Here we do not include opal_config_bottom.h: - Belongs to external code: opal/mca/backtrace/darwin/MoreBacktrace/MoreDebugging/MoreBacktrace.c opal/mca/backtrace/darwin/MoreBacktrace/MoreDebugging/MoreBacktrace.h - opal/include/opal/prefetch.h: Has C++ specific macros that are protected: - Had #if ... } #endif _and_ END_C_DECLS (aka end up with 2x END_C_DECLS) ompi/mca/btl/openib/btl_openib.h - opal/event/event.h has #ifdef __cplusplus as BEGIN_C_DECLS... - opal/win32/ompi_process.h: had extern "C"\n {... opal/win32/ompi_process.h: dito - ompi/mca/btl/pcie/btl_pcie_lex.l: needed to add *_C_DECLS ompi/mpi/f90/test/align_c.c: dito - ompi/debuggers/msgq_interface.h: used #ifdef __cplusplus - ompi/mpi/f90/xml/common-C.xsl: Amend Tested on linux using --with-openib and --with-mx The following do not contain either opal_config.h, orte_config.h or ompi_config.h (but possibly other header files, that include one of the above): ompi/mca/bml/r2/bml_r2_ft.h ompi/mca/btl/gm/btl_gm_endpoint.h ompi/mca/btl/gm/btl_gm_proc.h ompi/mca/btl/mx/btl_mx_endpoint.h ompi/mca/btl/ofud/btl_ofud_endpoint.h ompi/mca/btl/ofud/btl_ofud_frag.h ompi/mca/btl/ofud/btl_ofud_proc.h ompi/mca/btl/openib/btl_openib_mca.h ompi/mca/btl/portals/btl_portals_endpoint.h ompi/mca/btl/portals/btl_portals_frag.h ompi/mca/btl/sctp/btl_sctp_endpoint.h ompi/mca/btl/sctp/btl_sctp_proc.h ompi/mca/btl/tcp/btl_tcp_endpoint.h ompi/mca/btl/tcp/btl_tcp_ft.h ompi/mca/btl/tcp/btl_tcp_proc.h ompi/mca/btl/template/btl_template_endpoint.h ompi/mca/btl/template/btl_template_proc.h ompi/mca/btl/udapl/btl_udapl_eager_rdma.h ompi/mca/btl/udapl/btl_udapl_endpoint.h ompi/mca/btl/udapl/btl_udapl_mca.h ompi/mca/btl/udapl/btl_udapl_proc.h ompi/mca/mtl/mx/mtl_mx_endpoint.h ompi/mca/mtl/mx/mtl_mx.h ompi/mca/mtl/psm/mtl_psm_endpoint.h ompi/mca/mtl/psm/mtl_psm.h ompi/mca/pml/cm/pml_cm_component.h ompi/mca/pml/csum/pml_csum_comm.h ompi/mca/pml/dr/pml_dr_comm.h ompi/mca/pml/dr/pml_dr_component.h ompi/mca/pml/dr/pml_dr_endpoint.h ompi/mca/pml/dr/pml_dr_recvfrag.h ompi/mca/pml/example/pml_example.h ompi/mca/pml/ob1/pml_ob1_comm.h ompi/mca/pml/ob1/pml_ob1_component.h ompi/mca/pml/ob1/pml_ob1_endpoint.h ompi/mca/pml/ob1/pml_ob1_rdmafrag.h ompi/mca/pml/ob1/pml_ob1_recvfrag.h ompi/mca/pml/v/pml_v_output.h opal/include/opal/prefetch.h opal/mca/timer/aix/timer_aix.h opal/util/qsort.h test/support/components.h This commit was SVN r21855. The following SVN revision numbers were found above: r2 --> open-mpi/ompi@58fdc188553052bc2e893ba28fb28fddbe78435a
2009-08-20 11:42:18 +00:00
END_C_DECLS
#endif /* MTL_PORTALS_H_HAS_BEEN_INCLUDED */