1
1
openmpi/opal/mca/btl/ugni/btl_ugni_frag.h

159 строки
5.5 KiB
C
Исходник Обычный вид История

/* -*- Mode: C; c-basic-offset:4 ; indent-tabs-mode:nil -*- */
/*
* Copyright (c) 2011-2013 Los Alamos National Security, LLC. All rights
* reserved.
* Copyright (c) 2011 UT-Battelle, LLC. All rights reserved.
* Copyright (c) 2013 The University of Tennessee and The University
* of Tennessee Research Foundation. All rights
* reserved.
* $COPYRIGHT$
*
* Additional copyrights may follow
*
* $HEADER$
*/
#if !defined(MCA_BTL_UGNI_FRAG_H)
#define MCA_BTL_UGNI_FRAG_H
#include "btl_ugni.h"
#include "btl_ugni_endpoint.h"
typedef struct mca_btl_ugni_segment_t {
mca_btl_base_segment_t base;
gni_mem_handle_t memory_handle;
uint8_t extra_bytes[3];
uint8_t extra_byte_count;
} mca_btl_ugni_segment_t;
typedef struct mca_btl_ugni_send_frag_hdr_t {
uint32_t lag;
} mca_btl_ugni_send_frag_hdr_t;
typedef struct mca_btl_ugni_send_ex_frag_hdr_t {
mca_btl_ugni_send_frag_hdr_t send;
uint8_t pml_header[128];
} mca_btl_ugni_send_ex_frag_hdr_t;
typedef struct mca_btl_ugni_rdma_frag_hdr_t {
void *ctx;
} mca_btl_ugni_rdma_frag_hdr_t;
typedef struct mca_btl_ugni_eager_frag_hdr_t {
mca_btl_ugni_send_frag_hdr_t send;
mca_btl_ugni_segment_t src_seg;
void *ctx;
} mca_btl_ugni_eager_frag_hdr_t;
typedef struct mca_btl_ugni_eager_ex_frag_hdr_t {
mca_btl_ugni_eager_frag_hdr_t eager;
uint8_t pml_header[128];
} mca_btl_ugni_eager_ex_frag_hdr_t;
typedef union mca_btl_ugni_frag_hdr_t {
mca_btl_ugni_send_frag_hdr_t send;
mca_btl_ugni_send_ex_frag_hdr_t send_ex;
mca_btl_ugni_rdma_frag_hdr_t rdma;
mca_btl_ugni_eager_frag_hdr_t eager;
mca_btl_ugni_eager_ex_frag_hdr_t eager_ex;
} mca_btl_ugni_frag_hdr_t;
enum {
MCA_BTL_UGNI_FRAG_BUFFERED = 1, /* frag data is buffered */
MCA_BTL_UGNI_FRAG_COMPLETE = 2, /* smsg complete for frag */
MCA_BTL_UGNI_FRAG_EAGER = 4, /* eager get frag */
MCA_BTL_UGNI_FRAG_IGNORE = 8, /* ignore local smsg completion */
MCA_BTL_UGNI_FRAG_SMSG_COMPLETE = 16 /* SMSG has completed for this message */
};
struct mca_btl_ugni_base_frag_t;
typedef void (*frag_cb_t) (struct mca_btl_ugni_base_frag_t *, int);
typedef struct mca_btl_ugni_base_frag_t {
mca_btl_base_descriptor_t base;
uint32_t msg_id;
uint16_t hdr_size;
uint16_t flags;
mca_btl_ugni_frag_hdr_t hdr;
mca_btl_ugni_segment_t segments[2];
George did the work and deserves all the credit for it. Ralph did the merge, and deserves whatever blame results from errors in it :-) WHAT: Open our low-level communication infrastructure by moving all necessary components (btl/rcache/allocator/mpool) down in OPAL All the components required for inter-process communications are currently deeply integrated in the OMPI layer. Several groups/institutions have express interest in having a more generic communication infrastructure, without all the OMPI layer dependencies. This communication layer should be made available at a different software level, available to all layers in the Open MPI software stack. As an example, our ORTE layer could replace the current OOB and instead use the BTL directly, gaining access to more reactive network interfaces than TCP. Similarly, external software libraries could take advantage of our highly optimized AM (active message) communication layer for their own purpose. UTK with support from Sandia, developped a version of Open MPI where the entire communication infrastucture has been moved down to OPAL (btl/rcache/allocator/mpool). Most of the moved components have been updated to match the new schema, with few exceptions (mainly BTLs where I have no way of compiling/testing them). Thus, the completion of this RFC is tied to being able to completing this move for all BTLs. For this we need help from the rest of the Open MPI community, especially those supporting some of the BTLs. A non-exhaustive list of BTLs that qualify here is: mx, portals4, scif, udapl, ugni, usnic. This commit was SVN r32317.
2014-07-26 00:47:28 +00:00
opal_common_ugni_post_desc_t post_desc;
mca_btl_base_endpoint_t *endpoint;
mca_btl_ugni_reg_t *registration;
ompi_free_list_t *my_list;
frag_cb_t cbfunc;
} mca_btl_ugni_base_frag_t;
typedef struct mca_btl_ugni_base_frag_t mca_btl_ugni_smsg_frag_t;
typedef struct mca_btl_ugni_base_frag_t mca_btl_ugni_rdma_frag_t;
typedef struct mca_btl_ugni_base_frag_t mca_btl_ugni_eager_frag_t;
#define MCA_BTL_UGNI_DESC_TO_FRAG(desc) \
((mca_btl_ugni_base_frag_t *)((uintptr_t) (desc) - offsetof (mca_btl_ugni_base_frag_t, post_desc)))
OBJ_CLASS_DECLARATION(mca_btl_ugni_smsg_frag_t);
OBJ_CLASS_DECLARATION(mca_btl_ugni_rdma_frag_t);
OBJ_CLASS_DECLARATION(mca_btl_ugni_eager_frag_t);
void mca_btl_ugni_frag_init (mca_btl_ugni_base_frag_t *frag, mca_btl_ugni_module_t *ugni_module);
static inline int mca_btl_ugni_frag_alloc (mca_btl_base_endpoint_t *ep,
ompi_free_list_t *list,
mca_btl_ugni_base_frag_t **frag)
{
ompi_free_list_item_t *item = NULL;
OMPI_FREE_LIST_GET_MT(list, item);
*frag = (mca_btl_ugni_base_frag_t *) item;
if (OPAL_LIKELY(NULL != item)) {
(*frag)->my_list = list;
(*frag)->endpoint = ep;
George did the work and deserves all the credit for it. Ralph did the merge, and deserves whatever blame results from errors in it :-) WHAT: Open our low-level communication infrastructure by moving all necessary components (btl/rcache/allocator/mpool) down in OPAL All the components required for inter-process communications are currently deeply integrated in the OMPI layer. Several groups/institutions have express interest in having a more generic communication infrastructure, without all the OMPI layer dependencies. This communication layer should be made available at a different software level, available to all layers in the Open MPI software stack. As an example, our ORTE layer could replace the current OOB and instead use the BTL directly, gaining access to more reactive network interfaces than TCP. Similarly, external software libraries could take advantage of our highly optimized AM (active message) communication layer for their own purpose. UTK with support from Sandia, developped a version of Open MPI where the entire communication infrastucture has been moved down to OPAL (btl/rcache/allocator/mpool). Most of the moved components have been updated to match the new schema, with few exceptions (mainly BTLs where I have no way of compiling/testing them). Thus, the completion of this RFC is tied to being able to completing this move for all BTLs. For this we need help from the rest of the Open MPI community, especially those supporting some of the BTLs. A non-exhaustive list of BTLs that qualify here is: mx, portals4, scif, udapl, ugni, usnic. This commit was SVN r32317.
2014-07-26 00:47:28 +00:00
return OPAL_SUCCESS;
}
George did the work and deserves all the credit for it. Ralph did the merge, and deserves whatever blame results from errors in it :-) WHAT: Open our low-level communication infrastructure by moving all necessary components (btl/rcache/allocator/mpool) down in OPAL All the components required for inter-process communications are currently deeply integrated in the OMPI layer. Several groups/institutions have express interest in having a more generic communication infrastructure, without all the OMPI layer dependencies. This communication layer should be made available at a different software level, available to all layers in the Open MPI software stack. As an example, our ORTE layer could replace the current OOB and instead use the BTL directly, gaining access to more reactive network interfaces than TCP. Similarly, external software libraries could take advantage of our highly optimized AM (active message) communication layer for their own purpose. UTK with support from Sandia, developped a version of Open MPI where the entire communication infrastucture has been moved down to OPAL (btl/rcache/allocator/mpool). Most of the moved components have been updated to match the new schema, with few exceptions (mainly BTLs where I have no way of compiling/testing them). Thus, the completion of this RFC is tied to being able to completing this move for all BTLs. For this we need help from the rest of the Open MPI community, especially those supporting some of the BTLs. A non-exhaustive list of BTLs that qualify here is: mx, portals4, scif, udapl, ugni, usnic. This commit was SVN r32317.
2014-07-26 00:47:28 +00:00
return OPAL_ERR_OUT_OF_RESOURCE;
}
static inline int mca_btl_ugni_frag_return (mca_btl_ugni_base_frag_t *frag)
{
if (frag->registration) {
frag->endpoint->btl->super.btl_mpool->mpool_deregister(frag->endpoint->btl->super.btl_mpool,
(mca_mpool_base_registration_t *) frag->registration);
frag->registration = NULL;
}
frag->flags = 0;
OMPI_FREE_LIST_RETURN_MT(frag->my_list, (ompi_free_list_item_t *) frag);
George did the work and deserves all the credit for it. Ralph did the merge, and deserves whatever blame results from errors in it :-) WHAT: Open our low-level communication infrastructure by moving all necessary components (btl/rcache/allocator/mpool) down in OPAL All the components required for inter-process communications are currently deeply integrated in the OMPI layer. Several groups/institutions have express interest in having a more generic communication infrastructure, without all the OMPI layer dependencies. This communication layer should be made available at a different software level, available to all layers in the Open MPI software stack. As an example, our ORTE layer could replace the current OOB and instead use the BTL directly, gaining access to more reactive network interfaces than TCP. Similarly, external software libraries could take advantage of our highly optimized AM (active message) communication layer for their own purpose. UTK with support from Sandia, developped a version of Open MPI where the entire communication infrastucture has been moved down to OPAL (btl/rcache/allocator/mpool). Most of the moved components have been updated to match the new schema, with few exceptions (mainly BTLs where I have no way of compiling/testing them). Thus, the completion of this RFC is tied to being able to completing this move for all BTLs. For this we need help from the rest of the Open MPI community, especially those supporting some of the BTLs. A non-exhaustive list of BTLs that qualify here is: mx, portals4, scif, udapl, ugni, usnic. This commit was SVN r32317.
2014-07-26 00:47:28 +00:00
return OPAL_SUCCESS;
}
static inline void mca_btl_ugni_frag_complete (mca_btl_ugni_base_frag_t *frag, int rc) {
frag->flags |= MCA_BTL_UGNI_FRAG_COMPLETE;
BTL_VERBOSE(("frag complete. flags = %d", frag->base.des_flags));
/* call callback if specified */
if (frag->base.des_flags & MCA_BTL_DES_SEND_ALWAYS_CALLBACK) {
frag->base.des_cbfunc(&frag->endpoint->btl->super, frag->endpoint, &frag->base, rc);
}
if (frag->base.des_flags & MCA_BTL_DES_FLAGS_BTL_OWNERSHIP) {
mca_btl_ugni_frag_return (frag);
}
}
#define MCA_BTL_UGNI_FRAG_ALLOC_SMSG(ep, frag) \
mca_btl_ugni_frag_alloc((ep), &(ep)->btl->smsg_frags, &(frag))
#define MCA_BTL_UGNI_FRAG_ALLOC_RDMA(ep, frag) \
mca_btl_ugni_frag_alloc((ep), &(ep)->btl->rdma_frags, &(frag))
#define MCA_BTL_UGNI_FRAG_ALLOC_RDMA_INT(ep, frag) \
mca_btl_ugni_frag_alloc((ep), &(ep)->btl->rdma_int_frags, &(frag))
#define MCA_BTL_UGNI_FRAG_ALLOC_EAGER_SEND(ep, frag) \
mca_btl_ugni_frag_alloc((ep), &(ep)->btl->eager_frags_send, &(frag))
#define MCA_BTL_UGNI_FRAG_ALLOC_EAGER_RECV(ep, frag) \
mca_btl_ugni_frag_alloc((ep), &(ep)->btl->eager_frags_recv, &(frag))
#endif /* MCA_BTL_UGNI_FRAG_H */