This is an automated email from the git hooks/post-receive script. It was generated because a ref change was pushed to the repository containing the project "MPICH primary repository". The branch, master has been updated via 04a1907aaed46486857f995e4b8332cfdcd865e7 (commit) via c03c3a79f19e772c1e2bd5fe32b6bc8117e59ca4 (commit) via 26fb5462d44b835e04ca192341bf9ff490def266 (commit) via c68494ad0be062336b6108ad8f1be4f70132a7de (commit) via 4b9c244961103d1c192143b7a819c062c0145382 (commit) from 5dac839337d807f45f46ffd8c1d3905badb2f910 (commit) Those revisions listed above that are new to this repository have not appeared on any other notification email; so we list those revisions in full, below. - Log ----------------------------------------------------------------- http://git.mpich.org/mpich.git/commitdiff/04a1907aaed46486857f995e4b8332cfdc... commit 04a1907aaed46486857f995e4b8332cfdcd865e7 Author: Rob Latham <[email protected]> Date: Tue Aug 5 12:51:58 2014 -0500 Promote 'len' to something bigger than int It's now possible to have a very large file domain (one exceeding 2 GiB). We need to now relax some assertions and promote this type. Signed-off-by: Ken Raffenetti <[email protected]> diff --git a/src/mpi/romio/adio/ad_gpfs/ad_gpfs_aggrs.c b/src/mpi/romio/adio/ad_gpfs/ad_gpfs_aggrs.c index c83d91f..517db04 100644 --- a/src/mpi/romio/adio/ad_gpfs/ad_gpfs_aggrs.c +++ b/src/mpi/romio/adio/ad_gpfs/ad_gpfs_aggrs.c @@ -536,8 +536,8 @@ void ADIOI_GPFS_Calc_my_req(ADIO_File fd, ADIO_Offset *offset_list, ADIO_Offset if (count_my_req_per_proc[i]) { my_req[i].offsets = (ADIO_Offset *) ADIOI_Malloc(count_my_req_per_proc[i] * sizeof(ADIO_Offset)); - my_req[i].lens = (int *) - ADIOI_Malloc(count_my_req_per_proc[i] * sizeof(int)); + my_req[i].lens = + ADIOI_Malloc(count_my_req_per_proc[i] * sizeof(ADIO_Offset)); count_my_req_procs++; } my_req[i].count = 0; /* will be incremented where needed @@ -574,8 +574,7 @@ void ADIOI_GPFS_Calc_my_req(ADIO_File fd, ADIO_Offset *offset_list, ADIO_Offset * and the associated count. */ my_req[proc].offsets[l] = off; - ADIOI_Assert(fd_len == (int) fd_len); - my_req[proc].lens[l] = (int) fd_len; + my_req[proc].lens[l] = fd_len; my_req[proc].count++; while (rem_len > 0) { @@ -595,8 +594,7 @@ void ADIOI_GPFS_Calc_my_req(ADIO_File fd, ADIO_Offset *offset_list, ADIO_Offset rem_len -= fd_len; my_req[proc].offsets[l] = off; - ADIOI_Assert(fd_len == (int) fd_len); - my_req[proc].lens[l] = (int) fd_len; + my_req[proc].lens[l] = fd_len; my_req[proc].count++; } } @@ -718,8 +716,8 @@ void ADIOI_GPFS_Calc_others_req(ADIO_File fd, int count_my_req_procs, others_req[i].offsets = (ADIO_Offset *) ADIOI_Malloc(count_others_req_per_proc[i]*sizeof(ADIO_Offset)); - others_req[i].lens = (int *) - ADIOI_Malloc(count_others_req_per_proc[i]*sizeof(int)); + others_req[i].lens = + ADIOI_Malloc(count_others_req_per_proc[i]*sizeof(ADIO_Offset)); if ( (MPIR_Upint)others_req[i].offsets < (MPIR_Upint)recvBufForOffsets ) recvBufForOffsets = others_req[i].offsets; @@ -813,7 +811,7 @@ void ADIOI_GPFS_Calc_others_req(ADIO_File fd, int count_my_req_procs, sdispls[i] = (int) ( ( (MPIR_Upint)my_req[i].lens - (MPIR_Upint)sendBufForLens ) / - (MPIR_Upint) sizeof(int) ); + (MPIR_Upint) sizeof(ADIO_Offset) ); /* Receive these offsets from process i. */ rcounts[i] = count_others_req_per_proc[i]; @@ -823,14 +821,14 @@ void ADIOI_GPFS_Calc_others_req(ADIO_File fd, int count_my_req_procs, rdispls[i] = (int) ( ( (MPIR_Upint)others_req[i].lens - (MPIR_Upint)recvBufForLens ) / - (MPIR_Upint) sizeof(int) ); + (MPIR_Upint) sizeof(ADIO_Offset) ); } /* Exchange the lengths */ MPI_Alltoallv(sendBufForLens, - scounts, sdispls, MPI_INT, + scounts, sdispls, ADIO_OFFSET, recvBufForLens, - rcounts, rdispls, MPI_INT, + rcounts, rdispls, ADIO_OFFSET, fd->comm); /* Clean up */ diff --git a/src/mpi/romio/adio/ad_gpfs/ad_gpfs_rdcoll.c b/src/mpi/romio/adio/ad_gpfs/ad_gpfs_rdcoll.c index 2172702..92b6336 100644 --- a/src/mpi/romio/adio/ad_gpfs/ad_gpfs_rdcoll.c +++ b/src/mpi/romio/adio/ad_gpfs/ad_gpfs_rdcoll.c @@ -829,8 +829,8 @@ static void ADIOI_R_Exchange_data(ADIO_File fd, void *buf, ADIOI_Flatlist_node tmp = others_req[i].lens[k]; others_req[i].lens[k] = partial_send[i]; } - MPI_Type_hindexed(count[i], - &(others_req[i].lens[start_pos[i]]), + ADIOI_Type_create_hindexed_x(count[i], + &(others_req[i].lens[start_pos[i]]), &(others_req[i].mem_ptrs[start_pos[i]]), MPI_BYTE, &send_type); /* absolute displacement; use MPI_BOTTOM in send */ diff --git a/src/mpi/romio/adio/ad_gpfs/ad_gpfs_wrcoll.c b/src/mpi/romio/adio/ad_gpfs/ad_gpfs_wrcoll.c index ba0ded5..d049340 100644 --- a/src/mpi/romio/adio/ad_gpfs/ad_gpfs_wrcoll.c +++ b/src/mpi/romio/adio/ad_gpfs/ad_gpfs_wrcoll.c @@ -931,8 +931,8 @@ static void ADIOI_W_Exchange_data(ADIO_File fd, const void *buf, char *write_buf tmp_len[i] = others_req[i].lens[k]; others_req[i].lens[k] = partial_recv[i]; } - MPI_Type_hindexed(count[i], - &(others_req[i].lens[start_pos[i]]), + ADIOI_Type_create_hindexed_x(count[i], + &(others_req[i].lens[start_pos[i]]), &(others_req[i].mem_ptrs[start_pos[i]]), MPI_BYTE, recv_types+j); /* absolute displacements; use MPI_BOTTOM in recv */ @@ -1286,7 +1286,7 @@ static void ADIOI_Heap_merge(ADIOI_Access *others_req, int *count, { typedef struct { ADIO_Offset *off_list; - int *len_list; + ADIO_Offset *len_list; int nelem; } heap_struct; diff --git a/src/mpi/romio/adio/ad_lustre/ad_lustre_aggregate.c b/src/mpi/romio/adio/ad_lustre/ad_lustre_aggregate.c index a43f565..cd55282 100644 --- a/src/mpi/romio/adio/ad_lustre/ad_lustre_aggregate.c +++ b/src/mpi/romio/adio/ad_lustre/ad_lustre_aggregate.c @@ -215,8 +215,8 @@ void ADIOI_LUSTRE_Calc_my_req(ADIO_File fd, ADIO_Offset *offset_list, my_req[i].offsets = (ADIO_Offset *) ADIOI_Malloc(count_my_req_per_proc[i] * sizeof(ADIO_Offset)); - my_req[i].lens = (int *) ADIOI_Malloc(count_my_req_per_proc[i] * - sizeof(int)); + my_req[i].lens = ADIOI_Malloc(count_my_req_per_proc[i] * + sizeof(ADIO_Offset)); count_my_req_procs++; } my_req[i].count = 0; /* will be incremented where needed later */ diff --git a/src/mpi/romio/adio/ad_lustre/ad_lustre_wrcoll.c b/src/mpi/romio/adio/ad_lustre/ad_lustre_wrcoll.c index 58d75c9..3299ad5 100644 --- a/src/mpi/romio/adio/ad_lustre/ad_lustre_wrcoll.c +++ b/src/mpi/romio/adio/ad_lustre/ad_lustre_wrcoll.c @@ -656,7 +656,7 @@ static void ADIOI_LUSTRE_W_Exchange_data(ADIO_File fd, const void *buf, j = 0; for (i = 0; i < nprocs; i++) { if (recv_size[i]) { - MPI_Type_hindexed(count[i], + ADIOI_Type_create_hindexed_x(count[i], &(others_req[i].lens[start_pos[i]]), &(others_req[i].mem_ptrs[start_pos[i]]), MPI_BYTE, recv_types + j); diff --git a/src/mpi/romio/adio/common/ad_aggregate.c b/src/mpi/romio/adio/common/ad_aggregate.c index 4749256..55b5544 100644 --- a/src/mpi/romio/adio/common/ad_aggregate.c +++ b/src/mpi/romio/adio/common/ad_aggregate.c @@ -335,8 +335,8 @@ void ADIOI_Calc_my_req(ADIO_File fd, ADIO_Offset *offset_list, ADIO_Offset *len_ if (count_my_req_per_proc[i]) { my_req[i].offsets = (ADIO_Offset *) ADIOI_Malloc(count_my_req_per_proc[i] * sizeof(ADIO_Offset)); - my_req[i].lens = (int *) - ADIOI_Malloc(count_my_req_per_proc[i] * sizeof(int)); + my_req[i].lens = + ADIOI_Malloc(count_my_req_per_proc[i] * sizeof(ADIO_Offset)); count_my_req_procs++; } my_req[i].count = 0; /* will be incremented where needed @@ -373,8 +373,7 @@ void ADIOI_Calc_my_req(ADIO_File fd, ADIO_Offset *offset_list, ADIO_Offset *len_ * and the associated count. */ my_req[proc].offsets[l] = off; - ADIOI_Assert(fd_len == (int) fd_len); - my_req[proc].lens[l] = (int) fd_len; + my_req[proc].lens[l] = fd_len; my_req[proc].count++; while (rem_len != 0) { @@ -394,8 +393,7 @@ void ADIOI_Calc_my_req(ADIO_File fd, ADIO_Offset *offset_list, ADIO_Offset *len_ rem_len -= fd_len; my_req[proc].offsets[l] = off; - ADIOI_Assert(fd_len == (int) fd_len); - my_req[proc].lens[l] = (int) fd_len; + my_req[proc].lens[l] = fd_len; my_req[proc].count++; } } @@ -463,8 +461,8 @@ void ADIOI_Calc_others_req(ADIO_File fd, int count_my_req_procs, others_req[i].count = count_others_req_per_proc[i]; others_req[i].offsets = (ADIO_Offset *) ADIOI_Malloc(count_others_req_per_proc[i]*sizeof(ADIO_Offset)); - others_req[i].lens = (int *) - ADIOI_Malloc(count_others_req_per_proc[i]*sizeof(int)); + others_req[i].lens = + ADIOI_Malloc(count_others_req_per_proc[i]*sizeof(ADIO_Offset)); others_req[i].mem_ptrs = (MPI_Aint *) ADIOI_Malloc(count_others_req_per_proc[i]*sizeof(MPI_Aint)); count_others_req_procs++; @@ -485,7 +483,7 @@ void ADIOI_Calc_others_req(ADIO_File fd, int count_my_req_procs, ADIO_OFFSET, i, i+myrank, fd->comm, &requests[j]); j++; MPI_Irecv(others_req[i].lens, others_req[i].count, - MPI_INT, i, i+myrank+1, fd->comm, &requests[j]); + ADIO_OFFSET, i, i+myrank+1, fd->comm, &requests[j]); j++; } } @@ -496,7 +494,7 @@ void ADIOI_Calc_others_req(ADIO_File fd, int count_my_req_procs, ADIO_OFFSET, i, i+myrank, fd->comm, &requests[j]); j++; MPI_Isend(my_req[i].lens, my_req[i].count, - MPI_INT, i, i+myrank+1, fd->comm, &requests[j]); + ADIO_OFFSET, i, i+myrank+1, fd->comm, &requests[j]); j++; } } diff --git a/src/mpi/romio/adio/common/ad_read_coll.c b/src/mpi/romio/adio/common/ad_read_coll.c index 35090a4..c5202a3 100644 --- a/src/mpi/romio/adio/common/ad_read_coll.c +++ b/src/mpi/romio/adio/common/ad_read_coll.c @@ -853,8 +853,8 @@ static void ADIOI_R_Exchange_data(ADIO_File fd, void *buf, ADIOI_Flatlist_node tmp = others_req[i].lens[k]; others_req[i].lens[k] = partial_send[i]; } - MPI_Type_hindexed(count[i], - &(others_req[i].lens[start_pos[i]]), + ADIOI_Type_create_hindexed_x(count[i], + &(others_req[i].lens[start_pos[i]]), &(others_req[i].mem_ptrs[start_pos[i]]), MPI_BYTE, &send_type); /* absolute displacement; use MPI_BOTTOM in send */ diff --git a/src/mpi/romio/adio/common/ad_write_coll.c b/src/mpi/romio/adio/common/ad_write_coll.c index bf92a75..d585f9e 100644 --- a/src/mpi/romio/adio/common/ad_write_coll.c +++ b/src/mpi/romio/adio/common/ad_write_coll.c @@ -611,8 +611,8 @@ static void ADIOI_W_Exchange_data(ADIO_File fd, void *buf, char *write_buf, tmp_len[i] = others_req[i].lens[k]; others_req[i].lens[k] = partial_recv[i]; } - MPI_Type_hindexed(count[i], - &(others_req[i].lens[start_pos[i]]), + ADIOI_Type_create_hindexed_x(count[i], + &(others_req[i].lens[start_pos[i]]), &(others_req[i].mem_ptrs[start_pos[i]]), MPI_BYTE, recv_types+j); /* absolute displacements; use MPI_BOTTOM in recv */ @@ -974,7 +974,7 @@ void ADIOI_Heap_merge(ADIOI_Access *others_req, int *count, { typedef struct { ADIO_Offset *off_list; - int *len_list; + ADIO_Offset *len_list; int nelem; } heap_struct; diff --git a/src/mpi/romio/adio/include/adioi.h b/src/mpi/romio/adio/include/adioi.h index 26aad93..65e2bcd 100644 --- a/src/mpi/romio/adio/include/adioi.h +++ b/src/mpi/romio/adio/include/adioi.h @@ -307,9 +307,7 @@ struct ADIOI_Fns_struct { as array of structures indexed by process number. */ typedef struct { ADIO_Offset *offsets; /* array of offsets */ - int *lens; /* array of lengths */ - /* consider aints or offsets for lens? Seems to be used as in-memory - buffer lengths, so it should be < 2G and ok as an int */ + ADIO_Offset *lens; /* array of lengths */ MPI_Aint *mem_ptrs; /* array of pointers. used in the read/write phase to indicate where the data is stored in memory */ @@ -320,7 +318,7 @@ typedef struct { file realms among other things */ typedef struct { ADIO_Offset *offsets; /* array of offsets */ - int *lens; /* array of lengths */ + ADIO_Offset *lens; /* array of lengths */ int count; /* size of above arrays */ } ADIOI_Offlen; http://git.mpich.org/mpich.git/commitdiff/c03c3a79f19e772c1e2bd5fe32b6bc8117... commit c03c3a79f19e772c1e2bd5fe32b6bc8117e59ca4 Author: Rob Latham <[email protected]> Date: Thu Aug 7 21:45:43 2014 -0500 create a big-count HINDEXED type standard-HINDEXED takes an integer array for its "array of lengths" parameter. ROMIO can generate array_of_lengths that are too big for an integer. Jump through hoops. Signed-off-by: Ken Raffenetti <[email protected]> diff --git a/src/mpi/romio/adio/common/Makefile.mk b/src/mpi/romio/adio/common/Makefile.mk index 043b198..6cfd118 100644 --- a/src/mpi/romio/adio/common/Makefile.mk +++ b/src/mpi/romio/adio/common/Makefile.mk @@ -68,5 +68,6 @@ romio_other_sources += \ adio/common/system_hints.c \ adio/common/hint_fns.c \ adio/common/ad_threaded_io.c \ - adio/common/p2p_aggregation.c + adio/common/p2p_aggregation.c \ + adio/common/utils.c diff --git a/src/mpi/romio/adio/common/utils.c b/src/mpi/romio/adio/common/utils.c new file mode 100644 index 0000000..9814220 --- /dev/null +++ b/src/mpi/romio/adio/common/utils.c @@ -0,0 +1,83 @@ +/* -*- Mode: C; c-basic-offset:4 ; indent-tabs-mode:nil ; -*- */ +/* + * + * Copyright (C) 2014 UChicgo/Argonne, LLC. + * See COPYRIGHT notice in top-level directory. + */ + +#include <adio.h> +#include <limits.h> + + +/* utility function for creating large contiguous types: algorithim from BigMPI + * https://github.com/jeffhammond/BigMPI */ + +static int type_create_contiguous_x(MPI_Count count, + MPI_Datatype oldtype, MPI_Datatype *newtype) +{ + /* to make 'count' fit MPI-3 type processing routines (which take integer + * counts), we construct a type consisting of N INT_MAX chunks followed by + * a remainder. e.g for a count of 4000000000 bytes you would end up with + * one 2147483647-byte chunk followed immediately by a 1852516353-byte + * chunk */ + MPI_Datatype chunks, remainder; + MPI_Aint lb, extent, disps[2]; + int blocklens[2]; + MPI_Datatype types[2]; + + MPI_Count c = count/INT_MAX; + MPI_Count r = count%INT_MAX; + + MPI_Type_vector(c, INT_MAX, INT_MAX, oldtype, &chunks); + MPI_Type_contiguous(r, oldtype, &remainder); + + MPI_Type_get_extent(oldtype, &lb, &extent); + + blocklens[0] = 1; blocklens[1] = 1; + disps[0] = 0; disps[1] = c*extent*INT_MAX; + types[0] = chunks; types[1] = remainder; + + MPI_Type_create_struct(2, blocklens, disps, types, newtype); + + MPI_Type_free(&chunks); + MPI_Type_free(&remainder); + + return MPI_SUCCESS; +} +/* like MPI_Type_create_hindexed, except array_of_lengths can be a larger datatype. + * + * Hindexed provides 'count' pairs of (displacement, length), but what if + * length is longer than an integer? We will create 'count' types, using + * contig if length is small enough, or something more complex if not */ + +int ADIOI_Type_create_hindexed_x(int count, + const MPI_Count array_of_blocklengths[], + const MPI_Aint array_of_displacements[], + MPI_Datatype oldtype, + MPI_Datatype *newtype) +{ + int i, ret; + MPI_Datatype *types; + int *blocklens; + + types = ADIOI_Malloc(count*sizeof(MPI_Datatype)); + blocklens = ADIOI_Malloc(count*sizeof(int)); + + for(i=0; i<count; i++) { + blocklens[i] = 1; + type_create_contiguous_x(array_of_blocklengths[i], oldtype, &(types[i])); + } + + ret = MPI_Type_create_struct(count, blocklens, array_of_displacements, + types, newtype); + for (i=0; i< count; i++) + MPI_Type_free(&(types[i])); + ADIOI_Free(types); + ADIOI_Free(blocklens); + + return ret; +} + +/* + * vim: ts=8 sts=4 sw=4 noexpandtab + */ diff --git a/src/mpi/romio/adio/include/adioi.h b/src/mpi/romio/adio/include/adioi.h index 944e099..26aad93 100644 --- a/src/mpi/romio/adio/include/adioi.h +++ b/src/mpi/romio/adio/include/adioi.h @@ -590,6 +590,11 @@ int MPIR_Status_set_bytes(MPI_Status *status, MPI_Datatype datatype, MPI_Count n int ADIOI_Uses_generic_read(ADIO_File fd); int ADIOI_Uses_generic_write(ADIO_File fd); int ADIOI_Err_create_code(const char *myname, const char *filename, int my_errno); +int ADIOI_Type_create_hindexed_x(int count, + const MPI_Count array_of_blocklengths[], + const MPI_Aint array_of_displacements[], + MPI_Datatype oldtype, + MPI_Datatype *newtype); int ADIOI_FAKE_IODone(ADIO_Request *request, ADIO_Status *status, http://git.mpich.org/mpich.git/commitdiff/26fb5462d44b835e04ca192341bf9ff490... commit 26fb5462d44b835e04ca192341bf9ff490def266 Author: Rob Latham <[email protected]> Date: Wed Aug 6 11:23:04 2014 -0500 portability: only BGQ uses the MPIX routines mpix.h only exists for the pami device. BGPLATFORM is our closest proxy for "need this header for some PAMI-implemented routines" Signed-off-by: Ken Raffenetti <[email protected]> diff --git a/src/mpi/romio/adio/ad_gpfs/ad_gpfs_wrcoll.c b/src/mpi/romio/adio/ad_gpfs/ad_gpfs_wrcoll.c index 17299f2..ba0ded5 100644 --- a/src/mpi/romio/adio/ad_gpfs/ad_gpfs_wrcoll.c +++ b/src/mpi/romio/adio/ad_gpfs/ad_gpfs_wrcoll.c @@ -17,7 +17,9 @@ #include "ad_gpfs.h" #include "ad_gpfs_aggrs.h" +#ifdef BGQPLATFORM #include <mpix.h> +#endif #ifdef AGGREGATION_PROFILE #include "mpe.h" http://git.mpich.org/mpich.git/commitdiff/c68494ad0be062336b6108ad8f1be4f701... commit c68494ad0be062336b6108ad8f1be4f70132a7de Author: Rob Latham <[email protected]> Date: Wed Aug 6 14:58:01 2014 -0500 clean up bigtype test a bit - free memory in test to appease valgrind - more helpful failure message Signed-off-by: Ken Raffenetti <[email protected]> diff --git a/test/mpi/io/bigtype.c b/test/mpi/io/bigtype.c index 571a0a3..d500714 100644 --- a/test/mpi/io/bigtype.c +++ b/test/mpi/io/bigtype.c @@ -119,11 +119,14 @@ int main(int argc, char** argv) for (k=0 ; k<NUM_X*NUM_Y ; k++) { if(buf_read[k] != buf_write[k]) { - fprintf(stderr, "Verfiy Failed at index %zu\n", k); + fprintf(stderr, "Verfiy Failed index %zu: expected %d found %d\n", + k, buf_write[k], buf_read[k]); assert(0); } } + free(buf_write); + free(buf_read); MPI_File_close(&fh); MPI_Type_free (&mem_type); http://git.mpich.org/mpich.git/commitdiff/4b9c244961103d1c192143b7a819c062c0... commit 4b9c244961103d1c192143b7a819c062c0145382 Author: Rob Latham <[email protected]> Date: Fri Aug 8 11:13:07 2014 -0500 shameful but necessary wholesale reformatting DOS-newlines snuck into this file when I wasn't paying attention. Signed-off-by: Ken Raffenetti <[email protected]> diff --git a/src/mpi/romio/adio/ad_gpfs/ad_gpfs_aggrs.c b/src/mpi/romio/adio/ad_gpfs/ad_gpfs_aggrs.c index 350a336..c83d91f 100644 --- a/src/mpi/romio/adio/ad_gpfs/ad_gpfs_aggrs.c +++ b/src/mpi/romio/adio/ad_gpfs/ad_gpfs_aggrs.c @@ -1,848 +1,848 @@ -/* ---------------------------------------------------------------- */ -/* (C)Copyright IBM Corp. 2007, 2008 */ -/* ---------------------------------------------------------------- */ -/** - * \file ad_gpfs_aggrs.c - * \brief The externally used function from this file is is declared in ad_gpfs_aggrs.h - */ - -/* -*- Mode: C; c-basic-offset:4 ; -*- */ -/* - * Copyright (C) 1997-2001 University of Chicago. - * See COPYRIGHT notice in top-level directory. - */ - - -#include "adio.h" -#include "adio_cb_config_list.h" -#include "ad_gpfs.h" -#include "ad_gpfs_aggrs.h" - -#ifdef AGGREGATION_PROFILE -#include "mpe.h" -#endif - - -#ifdef USE_DBG_LOGGING - #define AGG_DEBUG 1 -#endif - -#ifndef TRACE_ERR -# define TRACE_ERR(format...) -#endif - -/* Comments copied from common: - * This file contains four functions: - * - * ADIOI_Calc_aggregator() - * ADIOI_Calc_file_domains() - * ADIOI_Calc_my_req() - * ADIOI_Calc_others_req() - * - * The last three of these were originally in ad_read_coll.c, but they are - * also shared with ad_write_coll.c. I felt that they were better kept with - * the rest of the shared aggregation code. - */ - -/* Discussion of values available from above: - * - * ADIO_Offset st_offsets[0..nprocs-1] - * ADIO_Offset end_offsets[0..nprocs-1] - * These contain a list of start and end offsets for each process in - * the communicator. For example, an access at loc 10, size 10 would - * have a start offset of 10 and end offset of 19. - * int nprocs - * number of processors in the collective I/O communicator - * ADIO_Offset min_st_offset - * ADIO_Offset fd_start[0..nprocs_for_coll-1] - * starting location of "file domain"; region that a given process will - * perform aggregation for (i.e. actually do I/O) - * ADIO_Offset fd_end[0..nprocs_for_coll-1] - * start + size - 1 roughly, but it can be less, or 0, in the case of - * uneven distributions - */ - -/* Description from common/ad_aggregate.c. (Does it completely apply to bg?) - * ADIOI_Calc_aggregator() - * - * The intention here is to implement a function which provides basically - * the same functionality as in Rajeev's original version of - * ADIOI_Calc_my_req(). He used a ceiling division approach to assign the - * file domains, and we use the same approach here when calculating the - * location of an offset/len in a specific file domain. Further we assume - * this same distribution when calculating the rank_index, which is later - * used to map to a specific process rank in charge of the file domain. - * - * A better (i.e. more general) approach would be to use the list of file - * domains only. This would be slower in the case where the - * original ceiling division was used, but it would allow for arbitrary - * distributions of regions to aggregators. We'd need to know the - * nprocs_for_coll in that case though, which we don't have now. - * - * Note a significant difference between this function and Rajeev's old code: - * this code doesn't necessarily return a rank in the range - * 0..nprocs_for_coll; instead you get something in 0..nprocs. This is a - * result of the rank mapping; any set of ranks in the communicator could be - * used now. - * - * Returns an integer representing a rank in the collective I/O communicator. - * - * The "len" parameter is also modified to indicate the amount of data - * actually available in this file domain. - */ -/* - * This is more general aggregator search function which does not base on the assumption - * that each aggregator hosts the file domain with the same size - */ -int ADIOI_GPFS_Calc_aggregator(ADIO_File fd, - ADIO_Offset off, - ADIO_Offset min_off, - ADIO_Offset *len, - ADIO_Offset fd_size, - ADIO_Offset *fd_start, - ADIO_Offset *fd_end) -{ - int rank_index, rank; - ADIO_Offset avail_bytes; - TRACE_ERR("Entering ADIOI_GPFS_Calc_aggregator\n"); - - ADIOI_Assert ( (off <= fd_end[fd->hints->cb_nodes-1] && off >= min_off && fd_start[0] >= min_off ) ); - - /* binary search --> rank_index is returned */ - int ub = fd->hints->cb_nodes; - int lb = 0; - /* get an index into our array of aggregators */ - /* Common code for striping - bg doesn't use it but it's - here to make diff'ing easier. - rank_index = (int) ((off - min_off + fd_size)/ fd_size - 1); - - if (fd->hints->striping_unit > 0) { - * wkliao: implementation for file domain alignment - fd_start[] and fd_end[] have been aligned with file lock - boundaries when returned from ADIOI_Calc_file_domains() so cannot - just use simple arithmatic as above * - rank_index = 0; - while (off > fd_end[rank_index]) rank_index++; - } - bg does it's own striping below - */ - rank_index = fd->hints->cb_nodes / 2; - while ( off < fd_start[rank_index] || off > fd_end[rank_index] ) { - if ( off > fd_end [rank_index] ) { - lb = rank_index; - rank_index = (rank_index + ub) / 2; - } - else - if ( off < fd_start[rank_index] ) { - ub = rank_index; - rank_index = (rank_index + lb) / 2; - } - } - /* we index into fd_end with rank_index, and fd_end was allocated to be no - * bigger than fd->hins->cb_nodes. If we ever violate that, we're - * overrunning arrays. Obviously, we should never ever hit this abort */ - if (rank_index >= fd->hints->cb_nodes || rank_index < 0) { - FPRINTF(stderr, "Error in ADIOI_Calc_aggregator(): rank_index(%d) >= fd->hints->cb_nodes (%d) fd_size=%lld off=%lld\n", - rank_index,fd->hints->cb_nodes,fd_size,off); - MPI_Abort(MPI_COMM_WORLD, 1); - } - /* DBG_FPRINTF ("ADIOI_GPFS_Calc_aggregator: rank_index = %d\n", - rank_index ); */ - - /* - * remember here that even in Rajeev's original code it was the case that - * different aggregators could end up with different amounts of data to - * aggregate. here we use fd_end[] to make sure that we know how much - * data this aggregator is working with. - * - * the +1 is to take into account the end vs. length issue. - */ - avail_bytes = fd_end[rank_index] + 1 - off; - if (avail_bytes < *len && avail_bytes > 0) { - /* this file domain only has part of the requested contig. region */ - - *len = avail_bytes; - } - - /* map our index to a rank */ - /* NOTE: FOR NOW WE DON'T HAVE A MAPPING...JUST DO 0..NPROCS_FOR_COLL */ - rank = fd->hints->ranklist[rank_index]; - TRACE_ERR("Leaving ADIOI_GPFS_Calc_aggregator\n"); - - return rank; -} - -/* - * Compute a dynamic access range based file domain partition among I/O aggregators, - * which align to the GPFS block size - * Divide the I/O workload among "nprocs_for_coll" processes. This is - * done by (logically) dividing the file into file domains (FDs); each - * process may directly access only its own file domain. - * Additional effort is to make sure that each I/O aggregator get - * a file domain that aligns to the GPFS block size. So, there will - * not be any false sharing of GPFS file blocks among multiple I/O nodes. - * - * The common version of this now accepts a min_fd_size and striping_unit. - * It doesn't seem necessary here (using GPFS block sizes) but keep it in mind - * (e.g. we could pass striping unit instead of using fs_ptr->blksize). - */ -void ADIOI_GPFS_Calc_file_domains(ADIO_File fd, - ADIO_Offset *st_offsets, - ADIO_Offset *end_offsets, - int nprocs, - int nprocs_for_coll, - ADIO_Offset *min_st_offset_ptr, - ADIO_Offset **fd_start_ptr, - ADIO_Offset **fd_end_ptr, - ADIO_Offset *fd_size_ptr, - void *fs_ptr) -{ - ADIO_Offset min_st_offset, max_end_offset, *fd_start, *fd_end, *fd_size; - int i, aggr; - TRACE_ERR("Entering ADIOI_GPFS_Calc_file_domains\n"); - blksize_t blksize; - -#ifdef AGGREGATION_PROFILE - MPE_Log_event (5004, 0, NULL); -#endif - -# if AGG_DEBUG - static char myname[] = "ADIOI_GPFS_Calc_file_domains"; - DBG_FPRINTF(stderr, "%s(%d): %d aggregator(s)\n", - myname,__LINE__,nprocs_for_coll); -# endif - if (fd->blksize <= 0) - /* default to 1M if blksize unset */ - fd->blksize = 1048576; - blksize = fd->blksize; - -# if AGG_DEBUG - DBG_FPRINTF(stderr,"%s(%d): Blocksize=%ld\n",myname,__LINE__,blksize); -# endif -/* find min of start offsets and max of end offsets of all processes */ - min_st_offset = st_offsets [0]; - max_end_offset = end_offsets[0]; - for (i=1; i<nprocs; i++) { - min_st_offset = ADIOI_MIN(min_st_offset, st_offsets[i]); - max_end_offset = ADIOI_MAX(max_end_offset, end_offsets[i]); - } - - /* DBG_FPRINTF(stderr, "_calc_file_domains, min_st_offset, max_ - = %qd, %qd\n", min_st_offset, max_end_offset );*/ - - /* determine the "file domain (FD)" of each process, i.e., the portion of - the file that will be "owned" by each process */ - - ADIO_Offset gpfs_ub = (max_end_offset +blksize-1) / blksize * blksize - 1; - ADIO_Offset gpfs_lb = min_st_offset / blksize * blksize; - ADIO_Offset gpfs_ub_rdoff = (max_end_offset +blksize-1) / blksize * blksize - 1 - max_end_offset; - ADIO_Offset gpfs_lb_rdoff = min_st_offset - min_st_offset / blksize * blksize; - ADIO_Offset fd_gpfs_range = gpfs_ub - gpfs_lb + 1; - - int naggs = nprocs_for_coll; - - /* Tweak the file domains so that no fd is smaller than a threshold. We - * have to strike a balance between efficency and parallelism: somewhere - * between 10k processes sending 32-byte requests and one process sending a - * 320k request is a (system-dependent) sweet spot - - This is from the common code - the new min_fd_size parm that we didn't implement. - (And common code uses a different declaration of fd_size so beware) - - if (fd_size < min_fd_size) - fd_size = min_fd_size; - */ - fd_size = (ADIO_Offset *) ADIOI_Malloc(nprocs_for_coll * sizeof(ADIO_Offset)); - *fd_start_ptr = (ADIO_Offset *) ADIOI_Malloc(nprocs_for_coll * sizeof(ADIO_Offset)); - *fd_end_ptr = (ADIO_Offset *) ADIOI_Malloc(nprocs_for_coll * sizeof(ADIO_Offset)); - fd_start = *fd_start_ptr; - fd_end = *fd_end_ptr; - - /* each process will have a file domain of some number of gpfs blocks, but - * the division of blocks is not likely to be even. Some file domains will - * be "large" and others "small" - * - * Example: consider 17 blocks distributed over 3 aggregators. - * nb_cn_small = 17/3 = 5 - * naggs_large = 17 - 3*(17/3) = 17 - 15 = 2 - * naggs_small = 3 - 2 = 1 - * - * and you end up with file domains of {5-blocks, 6-blocks, 6-blocks} - * - * what about (relatively) small files? say, a file of 1000 blocks - * distributed over 2064 aggregators: - * nb_cn_small = 1000/2064 = 0 - * naggs_large = 1000 - 2064*(1000/2064) = 1000 - * naggs_small = 2064 - 1000 = 1064 - * and you end up with domains of {0, 0, 0, ... 1, 1, 1 ...} - * - * it might be a good idea instead of having all the zeros up front, to - * "mix" those zeros into the fd_size array. that way, no pset/bridge-set - * is left with zero work. In fact, even if the small file domains aren't - * zero, it's probably still a good idea to mix the "small" file domains - * across the fd_size array to keep the io nodes in balance */ - - - ADIO_Offset n_gpfs_blk = fd_gpfs_range / blksize; - ADIO_Offset nb_cn_small = n_gpfs_blk/naggs; - ADIO_Offset naggs_large = n_gpfs_blk - naggs * (n_gpfs_blk/naggs); - ADIO_Offset naggs_small = naggs - naggs_large; - -#ifdef BGQPLATFORM - if (gpfsmpio_balancecontig == 1) { - /* File domains blocks are assigned to aggregators in a breadth-first - * fashion relative to the ions - additionally, file domains on the - * aggregators sharing the same bridgeset and ion have contiguous - * offsets. */ - - // initialize everything to small - for (i=0; i<naggs; i++) - fd_size[i] = nb_cn_small * blksize; - - // go thru and distribute the large across the bridges - - /* bridelistoffset: agg rank list offsets using the bridgelist - each - * entry is created by adding up the indexes for the aggs from all - * previous bridges */ - int *bridgelistoffset = - (int *) ADIOI_Malloc(fd->hints->fs_hints.bg.numbridges*sizeof(int)); - /* tmpbridgelistnum: copy of the bridgelistnum whose entries can be - * decremented to keep track of bridge assignments during the actual - * large block assignments to the agg rank list*/ - int *tmpbridgelistnum = - (int *) ADIOI_Malloc(fd->hints->fs_hints.bg.numbridges*sizeof(int)); - - int j; - for (j=0;j<fd->hints->fs_hints.bg.numbridges;j++) { - int k, bridgerankoffset = 0; - for (k=0;k<j;k++) { - bridgerankoffset += fd->hints->fs_hints.bg.bridgelistnum[k]; - } - bridgelistoffset[j] = bridgerankoffset; - } - - for (j=0;j<fd->hints->fs_hints.bg.numbridges;j++) - tmpbridgelistnum[j] = fd->hints->fs_hints.bg.bridgelistnum[j]; - int bridgeiter = 0; - - /* distribute the large blocks across the aggs going breadth-first - * across the bridgelist - this distributes the fd sizes across the - * ions, so later in the file domain assignment when it iterates thru - * the ranklist the offsets will be contiguous within the bridge and - * ion as well */ - for (j=0;j<naggs_large;j++) { - int foundbridge = 0; - int numbridgelistpasses = 0; - while (!foundbridge) { - if (tmpbridgelistnum[bridgeiter] > 0) { - foundbridge = 1; - /* - printf("bridgeiter is %d tmpbridgelistnum[bridgeiter] is %d bridgelistoffset[bridgeiter] is %d\n",bridgeiter,tmpbridgelistnum[bridgeiter],bridgelistoffset[bridgeiter]); - printf("naggs is %d bridgeiter is %d bridgelistoffset[bridgeiter] is %d tmpbridgelistnum[bridgeiter] is %d\n",naggs, bridgeiter,bridgelistoffset[bridgeiter],tmpbridgelistnum[bridgeiter]); - printf("naggs is %d bridgeiter is %d setting fd_size[%d]\n",naggs, bridgeiter,bridgelistoffset[bridgeiter]+(fd->hints->bridgelistnum[bridgeiter]-tmpbridgelistnum[bridgeiter])); - */ - int currentbridgelistnum = - (fd->hints->fs_hints.bg.bridgelistnum[bridgeiter]- - tmpbridgelistnum[bridgeiter]); - int currentfdsizeindex = bridgelistoffset[bridgeiter] + - currentbridgelistnum; - fd_size[currentfdsizeindex] = (nb_cn_small+1) * blksize; - tmpbridgelistnum[bridgeiter]--; - } - if (bridgeiter == (fd->hints->fs_hints.bg.numbridges-1)) { - /* guard against infinite loop - should only ever make 1 pass - * thru bridgelist */ - ADIOI_Assert(numbridgelistpasses == 0); - numbridgelistpasses++; - bridgeiter = 0; - } - else - bridgeiter++; - } - } - ADIOI_Free(tmpbridgelistnum); - ADIOI_Free(bridgelistoffset); - - } else { - /* BG/L- and BG/P-style distribution of file domains: simple allocation of - * file domins to each aggregator */ - for (i=0; i<naggs; i++) { - if (i < naggs_small) { - fd_size[i] = nb_cn_small * blksize; - } else { - fd_size[i] = (nb_cn_small+1) * blksize; - } - } - } -#ifdef balancecontigtrace - int myrank; - MPI_Comm_rank(fd->comm,&myrank); - if (myrank == 0) { - fprintf(stderr,"naggs_small is %d nb_cn_small is %d\n",naggs_small,nb_cn_small); - for (i=0; i<naggs; i++) { - fprintf(stderr,"fd_size[%d] set to %d agg rank is %d\n",i,fd_size[i],fd->hints->ranklist[i]); - } - } -#endif - -#else // not BGQ platform - for (i=0; i<naggs; i++) { - if (i < naggs_small) { - fd_size[i] = nb_cn_small * blksize; - } else { - fd_size[i] = (nb_cn_small+1) * blksize; - } - } - -#endif - - -# if AGG_DEBUG - DBG_FPRINTF(stderr,"%s(%d): " - "gpfs_ub %llu, " - "gpfs_lb %llu, " - "gpfs_ub_rdoff %llu, " - "gpfs_lb_rdoff %llu, " - "fd_gpfs_range %llu, " - "n_gpfs_blk %llu, " - "nb_cn_small %llu, " - "naggs_large %llu, " - "naggs_small %llu, " - "\n", - myname,__LINE__, - gpfs_ub , - gpfs_lb , - gpfs_ub_rdoff, - gpfs_lb_rdoff, - fd_gpfs_range, - n_gpfs_blk , - nb_cn_small , - naggs_large , - naggs_small - ); -# endif - - fd_size[0] -= gpfs_lb_rdoff; - fd_size[naggs-1] -= gpfs_ub_rdoff; - - /* compute the file domain for each aggr */ - ADIO_Offset offset = min_st_offset; - for (aggr=0; aggr<naggs; aggr++) { - fd_start[aggr] = offset; - fd_end [aggr] = offset + fd_size[aggr] - 1; - offset += fd_size[aggr]; - } - - *fd_size_ptr = fd_size[0]; - *min_st_offset_ptr = min_st_offset; - -#ifdef AGGREGATION_PROFILE - MPE_Log_event (5005, 0, NULL); -#endif - ADIOI_Free (fd_size); - TRACE_ERR("Leaving ADIOI_GPFS_Calc_file_domains\n"); -} - -/* - * ADIOI_GPFS_Calc_my_req() overrides ADIOI_Calc_my_req for the default implementation - * is specific for static file domain partitioning. - * - * ADIOI_Calc_my_req() - calculate what portions of the access requests - * of this process are located in the file domains of various processes - * (including this one) - */ -void ADIOI_GPFS_Calc_my_req(ADIO_File fd, ADIO_Offset *offset_list, ADIO_Offset *len_list, - int contig_access_count, ADIO_Offset - min_st_offset, ADIO_Offset *fd_start, - ADIO_Offset *fd_end, ADIO_Offset fd_size, - int nprocs, - int *count_my_req_procs_ptr, - int **count_my_req_per_proc_ptr, - ADIOI_Access **my_req_ptr, - int **buf_idx_ptr) -/* Possibly reconsider if buf_idx's are ok as int's, or should they be aints/offsets? - They are used as memory buffer indices so it seems like the 2G limit is in effect */ -{ - int *count_my_req_per_proc, count_my_req_procs, *buf_idx; - int i, l, proc; - ADIO_Offset fd_len, rem_len, curr_idx, off; - ADIOI_Access *my_req; - TRACE_ERR("Entering ADIOI_GPFS_Calc_my_req\n"); - -#ifdef AGGREGATION_PROFILE - MPE_Log_event (5024, 0, NULL); -#endif - *count_my_req_per_proc_ptr = (int *) ADIOI_Calloc(nprocs,sizeof(int)); - count_my_req_per_proc = *count_my_req_per_proc_ptr; -/* count_my_req_per_proc[i] gives the no. of contig. requests of this - process in process i's file domain. calloc initializes to zero. - I'm allocating memory of size nprocs, so that I can do an - MPI_Alltoall later on.*/ - - buf_idx = (int *) ADIOI_Malloc(nprocs*sizeof(int)); -/* buf_idx is relevant only if buftype_is_contig. - buf_idx[i] gives the index into user_buf where data received - from proc. i should be placed. This allows receives to be done - without extra buffer. This can't be done if buftype is not contig. */ - - /* initialize buf_idx to -1 */ - for (i=0; i < nprocs; i++) buf_idx[i] = -1; - - /* one pass just to calculate how much space to allocate for my_req; - * contig_access_count was calculated way back in ADIOI_Calc_my_off_len() - */ - for (i=0; i < contig_access_count; i++) { - /* short circuit offset/len processing if len == 0 - * (zero-byte read/write */ - if (len_list[i] == 0) - continue; - off = offset_list[i]; - fd_len = len_list[i]; - /* note: we set fd_len to be the total size of the access. then - * ADIOI_Calc_aggregator() will modify the value to return the - * amount that was available from the file domain that holds the - * first part of the access. - */ - /* BES */ - proc = ADIOI_GPFS_Calc_aggregator(fd, off, min_st_offset, &fd_len, fd_size, - fd_start, fd_end); - count_my_req_per_proc[proc]++; - - /* figure out how much data is remaining in the access (i.e. wasn't - * part of the file domain that had the starting byte); we'll take - * care of this data (if there is any) in the while loop below. - */ - rem_len = len_list[i] - fd_len; - - while (rem_len > 0) { - off += fd_len; /* point to first remaining byte */ - fd_len = rem_len; /* save remaining size, pass to calc */ - proc = ADIOI_GPFS_Calc_aggregator(fd, off, min_st_offset, &fd_len, - fd_size, fd_start, fd_end); - - count_my_req_per_proc[proc]++; - rem_len -= fd_len; /* reduce remaining length by amount from fd */ - } - } - -/* now allocate space for my_req, offset, and len */ - - *my_req_ptr = (ADIOI_Access *) - ADIOI_Malloc(nprocs*sizeof(ADIOI_Access)); - my_req = *my_req_ptr; - - count_my_req_procs = 0; - for (i=0; i < nprocs; i++) { - if (count_my_req_per_proc[i]) { - my_req[i].offsets = (ADIO_Offset *) - ADIOI_Malloc(count_my_req_per_proc[i] * sizeof(ADIO_Offset)); - my_req[i].lens = (int *) - ADIOI_Malloc(count_my_req_per_proc[i] * sizeof(int)); - count_my_req_procs++; - } - my_req[i].count = 0; /* will be incremented where needed - later */ - } - -/* now fill in my_req */ - curr_idx = 0; - for (i=0; i<contig_access_count; i++) { - /* short circuit offset/len processing if len == 0 - * (zero-byte read/write */ - if (len_list[i] == 0) - continue; - off = offset_list[i]; - fd_len = len_list[i]; - proc = ADIOI_GPFS_Calc_aggregator(fd, off, min_st_offset, &fd_len, fd_size, - fd_start, fd_end); - - /* for each separate contiguous access from this process */ - if (buf_idx[proc] == -1) - { - ADIOI_Assert(curr_idx == (int) curr_idx); - buf_idx[proc] = (int) curr_idx; - } - - l = my_req[proc].count; - curr_idx += fd_len; - - rem_len = len_list[i] - fd_len; - - /* store the proc, offset, and len information in an array - * of structures, my_req. Each structure contains the - * offsets and lengths located in that process's FD, - * and the associated count. - */ - my_req[proc].offsets[l] = off; - ADIOI_Assert(fd_len == (int) fd_len); - my_req[proc].lens[l] = (int) fd_len; - my_req[proc].count++; - - while (rem_len > 0) { - off += fd_len; - fd_len = rem_len; - proc = ADIOI_GPFS_Calc_aggregator(fd, off, min_st_offset, &fd_len, - fd_size, fd_start, fd_end); - - if (buf_idx[proc] == -1) - { - ADIOI_Assert(curr_idx == (int) curr_idx); - buf_idx[proc] = (int) curr_idx; - } - - l = my_req[proc].count; - curr_idx += fd_len; - rem_len -= fd_len; - - my_req[proc].offsets[l] = off; - ADIOI_Assert(fd_len == (int) fd_len); - my_req[proc].lens[l] = (int) fd_len; - my_req[proc].count++; - } - } - - - -#ifdef AGG_DEBUG - for (i=0; i<nprocs; i++) { - if (count_my_req_per_proc[i] > 0) { - DBG_FPRINTF(stderr, "data needed from %d (count = %d):\n", i, - my_req[i].count); - for (l=0; l < my_req[i].count; l++) { - DBG_FPRINTF(stderr, " off[%d] = %lld, len[%d] = %d\n", l, - my_req[i].offsets[l], l, my_req[i].lens[l]); - } - } - DBG_FPRINTF(stderr, "buf_idx[%d] = 0x%x\n", i, buf_idx[i]); - } -#endif - - *count_my_req_procs_ptr = count_my_req_procs; - *buf_idx_ptr = buf_idx; -#ifdef AGGREGATION_PROFILE - MPE_Log_event (5025, 0, NULL); -#endif - TRACE_ERR("Leaving ADIOI_GPFS_Calc_my_req\n"); -} - -/* - * ADIOI_Calc_others_req (copied to bg and switched to all to all for performance) - * - * param[in] count_my_req_procs Number of processes whose file domain my - * request touches. - * param[in] count_my_req_per_proc count_my_req_per_proc[i] gives the no. of - * contig. requests of this process in - * process i's file domain. - * param[in] my_req A structure defining my request - * param[in] nprocs Number of nodes in the block - * param[in] myrank Rank of this node - * param[out] count_others_req_proc_ptr Number of processes whose requests lie in - * my process's file domain (including my - * process itself) - * param[out] others_req_ptr Array of other process' requests that lie - * in my process's file domain - */ -void ADIOI_GPFS_Calc_others_req(ADIO_File fd, int count_my_req_procs, - int *count_my_req_per_proc, - ADIOI_Access *my_req, - int nprocs, int myrank, - int *count_others_req_procs_ptr, - ADIOI_Access **others_req_ptr) -{ - TRACE_ERR("Entering ADIOI_GPFS_Calc_others_req\n"); -/* determine what requests of other processes lie in this process's - file domain */ - -/* count_others_req_procs = number of processes whose requests lie in - this process's file domain (including this process itself) - count_others_req_per_proc[i] indicates how many separate contiguous - requests of proc. i lie in this process's file domain. */ - - int *count_others_req_per_proc, count_others_req_procs; - int i; - ADIOI_Access *others_req; - - /* Parameters for MPI_Alltoallv */ - int *scounts, *sdispls, *rcounts, *rdispls; - - /* Parameters for MPI_Alltoallv. These are the buffers, which - * are later computed to be the lowest address of all buffers - * to be sent/received for offsets and lengths. Initialize to - * the highest possible address which is the current minimum. - */ - void *sendBufForOffsets=(void*)0xFFFFFFFFFFFFFFFF, - *sendBufForLens =(void*)0xFFFFFFFFFFFFFFFF, - *recvBufForOffsets=(void*)0xFFFFFFFFFFFFFFFF, - *recvBufForLens =(void*)0xFFFFFFFFFFFFFFFF; - -/* first find out how much to send/recv and from/to whom */ -#ifdef AGGREGATION_PROFILE - MPE_Log_event (5026, 0, NULL); -#endif - /* Send 1 int to each process. count_my_req_per_proc[i] is the number of - * requests that my process will do to the file domain owned by process[i]. - * Receive 1 int from each process. count_others_req_per_proc[i] is the number of - * requests that process[i] will do to the file domain owned by my process. - */ - count_others_req_per_proc = (int *) ADIOI_Malloc(nprocs*sizeof(int)); -/* cora2a1=timebase(); */ -/*for(i=0;i<nprocs;i++) ?*/ - MPI_Alltoall(count_my_req_per_proc, 1, MPI_INT, - count_others_req_per_proc, 1, MPI_INT, fd->comm); - -/* total_cora2a+=timebase()-cora2a1; */ - - /* Allocate storage for an array of other nodes' accesses of our - * node's file domain. Also allocate storage for the alltoallv - * parameters. - */ - *others_req_ptr = (ADIOI_Access *) - ADIOI_Malloc(nprocs*sizeof(ADIOI_Access)); - others_req = *others_req_ptr; - - scounts = ADIOI_Malloc(nprocs*sizeof(int)); - sdispls = ADIOI_Malloc(nprocs*sizeof(int)); - rcounts = ADIOI_Malloc(nprocs*sizeof(int)); - rdispls = ADIOI_Malloc(nprocs*sizeof(int)); - - /* If process[i] has any requests in my file domain, - * initialize an ADIOI_Access structure that will describe each request - * from process[i]. The offsets, lengths, and buffer pointers still need - * to be obtained to complete the setting of this structure. - */ - count_others_req_procs = 0; - for (i=0; i<nprocs; i++) { - if (count_others_req_per_proc[i]) - { - others_req[i].count = count_others_req_per_proc[i]; - - others_req[i].offsets = (ADIO_Offset *) - ADIOI_Malloc(count_others_req_per_proc[i]*sizeof(ADIO_Offset)); - others_req[i].lens = (int *) - ADIOI_Malloc(count_others_req_per_proc[i]*sizeof(int)); - - if ( (MPIR_Upint)others_req[i].offsets < (MPIR_Upint)recvBufForOffsets ) - recvBufForOffsets = others_req[i].offsets; - if ( (MPIR_Upint)others_req[i].lens < (MPIR_Upint)recvBufForLens ) - recvBufForLens = others_req[i].lens; - - others_req[i].mem_ptrs = (MPI_Aint *) - ADIOI_Malloc(count_others_req_per_proc[i]*sizeof(MPI_Aint)); - - count_others_req_procs++; - } - else - { - others_req[i].count = 0; - others_req[i].offsets = NULL; - others_req[i].lens = NULL; - } - } - /* If no recv buffer was allocated in the loop above, make it NULL */ - if ( recvBufForOffsets == (void*)0xFFFFFFFFFFFFFFFF) recvBufForOffsets = NULL; - if ( recvBufForLens == (void*)0xFFFFFFFFFFFFFFFF) recvBufForLens = NULL; - - /* Now send the calculated offsets and lengths to respective processes */ - - /************************/ - /* Exchange the offsets */ - /************************/ - - /* Determine the lowest sendBufForOffsets/Lens */ - for (i=0; i<nprocs; i++) - { - if ( (my_req[i].count) && - ((MPIR_Upint)my_req[i].offsets <= (MPIR_Upint)sendBufForOffsets) ) - { - sendBufForOffsets = my_req[i].offsets; - } - - if ( (my_req[i].count) && - ((MPIR_Upint)my_req[i].lens <= (MPIR_Upint)sendBufForLens) ) - { - sendBufForLens = my_req[i].lens; - } - } - - /* If no send buffer was found in the loop above, make it NULL */ - if ( sendBufForOffsets == (void*)0xFFFFFFFFFFFFFFFF) sendBufForOffsets = NULL; - if ( sendBufForLens == (void*)0xFFFFFFFFFFFFFFFF) sendBufForLens = NULL; - - /* Calculate the displacements from the sendBufForOffsets/Lens */ - for (i=0; i<nprocs; i++) - { - /* Send these offsets to process i.*/ - scounts[i] = count_my_req_per_proc[i]; - if ( scounts[i] == 0 ) - sdispls[i] = 0; - else - sdispls[i] = (int) - ( ( (MPIR_Upint)my_req[i].offsets - - (MPIR_Upint)sendBufForOffsets ) / - (MPIR_Upint)sizeof(ADIO_Offset) ); - - /* Receive these offsets from process i.*/ - rcounts[i] = count_others_req_per_proc[i]; - if ( rcounts[i] == 0 ) - rdispls[i] = 0; - else - rdispls[i] = (int) - ( ( (MPIR_Upint)others_req[i].offsets - - (MPIR_Upint)recvBufForOffsets ) / - (MPIR_Upint)sizeof(ADIO_Offset) ); - } - - /* Exchange the offsets */ - MPI_Alltoallv(sendBufForOffsets, - scounts, sdispls, ADIO_OFFSET, - recvBufForOffsets, - rcounts, rdispls, ADIO_OFFSET, - fd->comm); - - /************************/ - /* Exchange the lengths */ - /************************/ - - for (i=0; i<nprocs; i++) - { - /* Send these lengths to process i.*/ - scounts[i] = count_my_req_per_proc[i]; - if ( scounts[i] == 0 ) - sdispls[i] = 0; - else - sdispls[i] = (int) - ( ( (MPIR_Upint)my_req[i].lens - - (MPIR_Upint)sendBufForLens ) / - (MPIR_Upint) sizeof(int) ); - - /* Receive these offsets from process i. */ - rcounts[i] = count_others_req_per_proc[i]; - if ( rcounts[i] == 0 ) - rdispls[i] = 0; - else - rdispls[i] = (int) - ( ( (MPIR_Upint)others_req[i].lens - - (MPIR_Upint)recvBufForLens ) / - (MPIR_Upint) sizeof(int) ); - } - - /* Exchange the lengths */ - MPI_Alltoallv(sendBufForLens, - scounts, sdispls, MPI_INT, - recvBufForLens, - rcounts, rdispls, MPI_INT, - fd->comm); - - /* Clean up */ - ADIOI_Free(count_others_req_per_proc); - ADIOI_Free (scounts); - ADIOI_Free (sdispls); - ADIOI_Free (rcounts); - ADIOI_Free (rdispls); - - *count_others_req_procs_ptr = count_others_req_procs; -#ifdef AGGREGATION_PROFILE - MPE_Log_event (5027, 0, NULL); -#endif - TRACE_ERR("Leaving ADIOI_GPFS_Calc_others_req\n"); -} +/* ---------------------------------------------------------------- */ +/* (C)Copyright IBM Corp. 2007, 2008 */ +/* ---------------------------------------------------------------- */ +/** + * \file ad_gpfs_aggrs.c + * \brief The externally used function from this file is is declared in ad_gpfs_aggrs.h + */ + +/* -*- Mode: C; c-basic-offset:4 ; -*- */ +/* + * Copyright (C) 1997-2001 University of Chicago. + * See COPYRIGHT notice in top-level directory. + */ + + +#include "adio.h" +#include "adio_cb_config_list.h" +#include "ad_gpfs.h" +#include "ad_gpfs_aggrs.h" + +#ifdef AGGREGATION_PROFILE +#include "mpe.h" +#endif + + +#ifdef USE_DBG_LOGGING + #define AGG_DEBUG 1 +#endif + +#ifndef TRACE_ERR +# define TRACE_ERR(format...) +#endif + +/* Comments copied from common: + * This file contains four functions: + * + * ADIOI_Calc_aggregator() + * ADIOI_Calc_file_domains() + * ADIOI_Calc_my_req() + * ADIOI_Calc_others_req() + * + * The last three of these were originally in ad_read_coll.c, but they are + * also shared with ad_write_coll.c. I felt that they were better kept with + * the rest of the shared aggregation code. + */ + +/* Discussion of values available from above: + * + * ADIO_Offset st_offsets[0..nprocs-1] + * ADIO_Offset end_offsets[0..nprocs-1] + * These contain a list of start and end offsets for each process in + * the communicator. For example, an access at loc 10, size 10 would + * have a start offset of 10 and end offset of 19. + * int nprocs + * number of processors in the collective I/O communicator + * ADIO_Offset min_st_offset + * ADIO_Offset fd_start[0..nprocs_for_coll-1] + * starting location of "file domain"; region that a given process will + * perform aggregation for (i.e. actually do I/O) + * ADIO_Offset fd_end[0..nprocs_for_coll-1] + * start + size - 1 roughly, but it can be less, or 0, in the case of + * uneven distributions + */ + +/* Description from common/ad_aggregate.c. (Does it completely apply to bg?) + * ADIOI_Calc_aggregator() + * + * The intention here is to implement a function which provides basically + * the same functionality as in Rajeev's original version of + * ADIOI_Calc_my_req(). He used a ceiling division approach to assign the + * file domains, and we use the same approach here when calculating the + * location of an offset/len in a specific file domain. Further we assume + * this same distribution when calculating the rank_index, which is later + * used to map to a specific process rank in charge of the file domain. + * + * A better (i.e. more general) approach would be to use the list of file + * domains only. This would be slower in the case where the + * original ceiling division was used, but it would allow for arbitrary + * distributions of regions to aggregators. We'd need to know the + * nprocs_for_coll in that case though, which we don't have now. + * + * Note a significant difference between this function and Rajeev's old code: + * this code doesn't necessarily return a rank in the range + * 0..nprocs_for_coll; instead you get something in 0..nprocs. This is a + * result of the rank mapping; any set of ranks in the communicator could be + * used now. + * + * Returns an integer representing a rank in the collective I/O communicator. + * + * The "len" parameter is also modified to indicate the amount of data + * actually available in this file domain. + */ +/* + * This is more general aggregator search function which does not base on the assumption + * that each aggregator hosts the file domain with the same size + */ +int ADIOI_GPFS_Calc_aggregator(ADIO_File fd, + ADIO_Offset off, + ADIO_Offset min_off, + ADIO_Offset *len, + ADIO_Offset fd_size, + ADIO_Offset *fd_start, + ADIO_Offset *fd_end) +{ + int rank_index, rank; + ADIO_Offset avail_bytes; + TRACE_ERR("Entering ADIOI_GPFS_Calc_aggregator\n"); + + ADIOI_Assert ( (off <= fd_end[fd->hints->cb_nodes-1] && off >= min_off && fd_start[0] >= min_off ) ); + + /* binary search --> rank_index is returned */ + int ub = fd->hints->cb_nodes; + int lb = 0; + /* get an index into our array of aggregators */ + /* Common code for striping - bg doesn't use it but it's + here to make diff'ing easier. + rank_index = (int) ((off - min_off + fd_size)/ fd_size - 1); + + if (fd->hints->striping_unit > 0) { + * wkliao: implementation for file domain alignment + fd_start[] and fd_end[] have been aligned with file lock + boundaries when returned from ADIOI_Calc_file_domains() so cannot + just use simple arithmatic as above * + rank_index = 0; + while (off > fd_end[rank_index]) rank_index++; + } + bg does it's own striping below + */ + rank_index = fd->hints->cb_nodes / 2; + while ( off < fd_start[rank_index] || off > fd_end[rank_index] ) { + if ( off > fd_end [rank_index] ) { + lb = rank_index; + rank_index = (rank_index + ub) / 2; + } + else + if ( off < fd_start[rank_index] ) { + ub = rank_index; + rank_index = (rank_index + lb) / 2; + } + } + /* we index into fd_end with rank_index, and fd_end was allocated to be no + * bigger than fd->hins->cb_nodes. If we ever violate that, we're + * overrunning arrays. Obviously, we should never ever hit this abort */ + if (rank_index >= fd->hints->cb_nodes || rank_index < 0) { + FPRINTF(stderr, "Error in ADIOI_Calc_aggregator(): rank_index(%d) >= fd->hints->cb_nodes (%d) fd_size=%lld off=%lld\n", + rank_index,fd->hints->cb_nodes,fd_size,off); + MPI_Abort(MPI_COMM_WORLD, 1); + } + /* DBG_FPRINTF ("ADIOI_GPFS_Calc_aggregator: rank_index = %d\n", + rank_index ); */ + + /* + * remember here that even in Rajeev's original code it was the case that + * different aggregators could end up with different amounts of data to + * aggregate. here we use fd_end[] to make sure that we know how much + * data this aggregator is working with. + * + * the +1 is to take into account the end vs. length issue. + */ + avail_bytes = fd_end[rank_index] + 1 - off; + if (avail_bytes < *len && avail_bytes > 0) { + /* this file domain only has part of the requested contig. region */ + + *len = avail_bytes; + } + + /* map our index to a rank */ + /* NOTE: FOR NOW WE DON'T HAVE A MAPPING...JUST DO 0..NPROCS_FOR_COLL */ + rank = fd->hints->ranklist[rank_index]; + TRACE_ERR("Leaving ADIOI_GPFS_Calc_aggregator\n"); + + return rank; +} + +/* + * Compute a dynamic access range based file domain partition among I/O aggregators, + * which align to the GPFS block size + * Divide the I/O workload among "nprocs_for_coll" processes. This is + * done by (logically) dividing the file into file domains (FDs); each + * process may directly access only its own file domain. + * Additional effort is to make sure that each I/O aggregator get + * a file domain that aligns to the GPFS block size. So, there will + * not be any false sharing of GPFS file blocks among multiple I/O nodes. + * + * The common version of this now accepts a min_fd_size and striping_unit. + * It doesn't seem necessary here (using GPFS block sizes) but keep it in mind + * (e.g. we could pass striping unit instead of using fs_ptr->blksize). + */ +void ADIOI_GPFS_Calc_file_domains(ADIO_File fd, + ADIO_Offset *st_offsets, + ADIO_Offset *end_offsets, + int nprocs, + int nprocs_for_coll, + ADIO_Offset *min_st_offset_ptr, + ADIO_Offset **fd_start_ptr, + ADIO_Offset **fd_end_ptr, + ADIO_Offset *fd_size_ptr, + void *fs_ptr) +{ + ADIO_Offset min_st_offset, max_end_offset, *fd_start, *fd_end, *fd_size; + int i, aggr; + TRACE_ERR("Entering ADIOI_GPFS_Calc_file_domains\n"); + blksize_t blksize; + +#ifdef AGGREGATION_PROFILE + MPE_Log_event (5004, 0, NULL); +#endif + +# if AGG_DEBUG + static char myname[] = "ADIOI_GPFS_Calc_file_domains"; + DBG_FPRINTF(stderr, "%s(%d): %d aggregator(s)\n", + myname,__LINE__,nprocs_for_coll); +# endif + if (fd->blksize <= 0) + /* default to 1M if blksize unset */ + fd->blksize = 1048576; + blksize = fd->blksize; + +# if AGG_DEBUG + DBG_FPRINTF(stderr,"%s(%d): Blocksize=%ld\n",myname,__LINE__,blksize); +# endif +/* find min of start offsets and max of end offsets of all processes */ + min_st_offset = st_offsets [0]; + max_end_offset = end_offsets[0]; + for (i=1; i<nprocs; i++) { + min_st_offset = ADIOI_MIN(min_st_offset, st_offsets[i]); + max_end_offset = ADIOI_MAX(max_end_offset, end_offsets[i]); + } + + /* DBG_FPRINTF(stderr, "_calc_file_domains, min_st_offset, max_ + = %qd, %qd\n", min_st_offset, max_end_offset );*/ + + /* determine the "file domain (FD)" of each process, i.e., the portion of + the file that will be "owned" by each process */ + + ADIO_Offset gpfs_ub = (max_end_offset +blksize-1) / blksize * blksize - 1; + ADIO_Offset gpfs_lb = min_st_offset / blksize * blksize; + ADIO_Offset gpfs_ub_rdoff = (max_end_offset +blksize-1) / blksize * blksize - 1 - max_end_offset; + ADIO_Offset gpfs_lb_rdoff = min_st_offset - min_st_offset / blksize * blksize; + ADIO_Offset fd_gpfs_range = gpfs_ub - gpfs_lb + 1; + + int naggs = nprocs_for_coll; + + /* Tweak the file domains so that no fd is smaller than a threshold. We + * have to strike a balance between efficency and parallelism: somewhere + * between 10k processes sending 32-byte requests and one process sending a + * 320k request is a (system-dependent) sweet spot + + This is from the common code - the new min_fd_size parm that we didn't implement. + (And common code uses a different declaration of fd_size so beware) + + if (fd_size < min_fd_size) + fd_size = min_fd_size; + */ + fd_size = (ADIO_Offset *) ADIOI_Malloc(nprocs_for_coll * sizeof(ADIO_Offset)); + *fd_start_ptr = (ADIO_Offset *) ADIOI_Malloc(nprocs_for_coll * sizeof(ADIO_Offset)); + *fd_end_ptr = (ADIO_Offset *) ADIOI_Malloc(nprocs_for_coll * sizeof(ADIO_Offset)); + fd_start = *fd_start_ptr; + fd_end = *fd_end_ptr; + + /* each process will have a file domain of some number of gpfs blocks, but + * the division of blocks is not likely to be even. Some file domains will + * be "large" and others "small" + * + * Example: consider 17 blocks distributed over 3 aggregators. + * nb_cn_small = 17/3 = 5 + * naggs_large = 17 - 3*(17/3) = 17 - 15 = 2 + * naggs_small = 3 - 2 = 1 + * + * and you end up with file domains of {5-blocks, 6-blocks, 6-blocks} + * + * what about (relatively) small files? say, a file of 1000 blocks + * distributed over 2064 aggregators: + * nb_cn_small = 1000/2064 = 0 + * naggs_large = 1000 - 2064*(1000/2064) = 1000 + * naggs_small = 2064 - 1000 = 1064 + * and you end up with domains of {0, 0, 0, ... 1, 1, 1 ...} + * + * it might be a good idea instead of having all the zeros up front, to + * "mix" those zeros into the fd_size array. that way, no pset/bridge-set + * is left with zero work. In fact, even if the small file domains aren't + * zero, it's probably still a good idea to mix the "small" file domains + * across the fd_size array to keep the io nodes in balance */ + + + ADIO_Offset n_gpfs_blk = fd_gpfs_range / blksize; + ADIO_Offset nb_cn_small = n_gpfs_blk/naggs; + ADIO_Offset naggs_large = n_gpfs_blk - naggs * (n_gpfs_blk/naggs); + ADIO_Offset naggs_small = naggs - naggs_large; + +#ifdef BGQPLATFORM + if (gpfsmpio_balancecontig == 1) { + /* File domains blocks are assigned to aggregators in a breadth-first + * fashion relative to the ions - additionally, file domains on the + * aggregators sharing the same bridgeset and ion have contiguous + * offsets. */ + + // initialize everything to small + for (i=0; i<naggs; i++) + fd_size[i] = nb_cn_small * blksize; + + // go thru and distribute the large across the bridges + + /* bridelistoffset: agg rank list offsets using the bridgelist - each + * entry is created by adding up the indexes for the aggs from all + * previous bridges */ + int *bridgelistoffset = + (int *) ADIOI_Malloc(fd->hints->fs_hints.bg.numbridges*sizeof(int)); + /* tmpbridgelistnum: copy of the bridgelistnum whose entries can be + * decremented to keep track of bridge assignments during the actual + * large block assignments to the agg rank list*/ + int *tmpbridgelistnum = + (int *) ADIOI_Malloc(fd->hints->fs_hints.bg.numbridges*sizeof(int)); + + int j; + for (j=0;j<fd->hints->fs_hints.bg.numbridges;j++) { + int k, bridgerankoffset = 0; + for (k=0;k<j;k++) { + bridgerankoffset += fd->hints->fs_hints.bg.bridgelistnum[k]; + } + bridgelistoffset[j] = bridgerankoffset; + } + + for (j=0;j<fd->hints->fs_hints.bg.numbridges;j++) + tmpbridgelistnum[j] = fd->hints->fs_hints.bg.bridgelistnum[j]; + int bridgeiter = 0; + + /* distribute the large blocks across the aggs going breadth-first + * across the bridgelist - this distributes the fd sizes across the + * ions, so later in the file domain assignment when it iterates thru + * the ranklist the offsets will be contiguous within the bridge and + * ion as well */ + for (j=0;j<naggs_large;j++) { + int foundbridge = 0; + int numbridgelistpasses = 0; + while (!foundbridge) { + if (tmpbridgelistnum[bridgeiter] > 0) { + foundbridge = 1; + /* + printf("bridgeiter is %d tmpbridgelistnum[bridgeiter] is %d bridgelistoffset[bridgeiter] is %d\n",bridgeiter,tmpbridgelistnum[bridgeiter],bridgelistoffset[bridgeiter]); + printf("naggs is %d bridgeiter is %d bridgelistoffset[bridgeiter] is %d tmpbridgelistnum[bridgeiter] is %d\n",naggs, bridgeiter,bridgelistoffset[bridgeiter],tmpbridgelistnum[bridgeiter]); + printf("naggs is %d bridgeiter is %d setting fd_size[%d]\n",naggs, bridgeiter,bridgelistoffset[bridgeiter]+(fd->hints->bridgelistnum[bridgeiter]-tmpbridgelistnum[bridgeiter])); + */ + int currentbridgelistnum = + (fd->hints->fs_hints.bg.bridgelistnum[bridgeiter]- + tmpbridgelistnum[bridgeiter]); + int currentfdsizeindex = bridgelistoffset[bridgeiter] + + currentbridgelistnum; + fd_size[currentfdsizeindex] = (nb_cn_small+1) * blksize; + tmpbridgelistnum[bridgeiter]--; + } + if (bridgeiter == (fd->hints->fs_hints.bg.numbridges-1)) { + /* guard against infinite loop - should only ever make 1 pass + * thru bridgelist */ + ADIOI_Assert(numbridgelistpasses == 0); + numbridgelistpasses++; + bridgeiter = 0; + } + else + bridgeiter++; + } + } + ADIOI_Free(tmpbridgelistnum); + ADIOI_Free(bridgelistoffset); + + } else { + /* BG/L- and BG/P-style distribution of file domains: simple allocation of + * file domins to each aggregator */ + for (i=0; i<naggs; i++) { + if (i < naggs_small) { + fd_size[i] = nb_cn_small * blksize; + } else { + fd_size[i] = (nb_cn_small+1) * blksize; + } + } + } +#ifdef balancecontigtrace + int myrank; + MPI_Comm_rank(fd->comm,&myrank); + if (myrank == 0) { + fprintf(stderr,"naggs_small is %d nb_cn_small is %d\n",naggs_small,nb_cn_small); + for (i=0; i<naggs; i++) { + fprintf(stderr,"fd_size[%d] set to %d agg rank is %d\n",i,fd_size[i],fd->hints->ranklist[i]); + } + } +#endif + +#else // not BGQ platform + for (i=0; i<naggs; i++) { + if (i < naggs_small) { + fd_size[i] = nb_cn_small * blksize; + } else { + fd_size[i] = (nb_cn_small+1) * blksize; + } + } + +#endif + + +# if AGG_DEBUG + DBG_FPRINTF(stderr,"%s(%d): " + "gpfs_ub %llu, " + "gpfs_lb %llu, " + "gpfs_ub_rdoff %llu, " + "gpfs_lb_rdoff %llu, " + "fd_gpfs_range %llu, " + "n_gpfs_blk %llu, " + "nb_cn_small %llu, " + "naggs_large %llu, " + "naggs_small %llu, " + "\n", + myname,__LINE__, + gpfs_ub , + gpfs_lb , + gpfs_ub_rdoff, + gpfs_lb_rdoff, + fd_gpfs_range, + n_gpfs_blk , + nb_cn_small , + naggs_large , + naggs_small + ); +# endif + + fd_size[0] -= gpfs_lb_rdoff; + fd_size[naggs-1] -= gpfs_ub_rdoff; + + /* compute the file domain for each aggr */ + ADIO_Offset offset = min_st_offset; + for (aggr=0; aggr<naggs; aggr++) { + fd_start[aggr] = offset; + fd_end [aggr] = offset + fd_size[aggr] - 1; + offset += fd_size[aggr]; + } + + *fd_size_ptr = fd_size[0]; + *min_st_offset_ptr = min_st_offset; + +#ifdef AGGREGATION_PROFILE + MPE_Log_event (5005, 0, NULL); +#endif + ADIOI_Free (fd_size); + TRACE_ERR("Leaving ADIOI_GPFS_Calc_file_domains\n"); +} + +/* + * ADIOI_GPFS_Calc_my_req() overrides ADIOI_Calc_my_req for the default implementation + * is specific for static file domain partitioning. + * + * ADIOI_Calc_my_req() - calculate what portions of the access requests + * of this process are located in the file domains of various processes + * (including this one) + */ +void ADIOI_GPFS_Calc_my_req(ADIO_File fd, ADIO_Offset *offset_list, ADIO_Offset *len_list, + int contig_access_count, ADIO_Offset + min_st_offset, ADIO_Offset *fd_start, + ADIO_Offset *fd_end, ADIO_Offset fd_size, + int nprocs, + int *count_my_req_procs_ptr, + int **count_my_req_per_proc_ptr, + ADIOI_Access **my_req_ptr, + int **buf_idx_ptr) +/* Possibly reconsider if buf_idx's are ok as int's, or should they be aints/offsets? + They are used as memory buffer indices so it seems like the 2G limit is in effect */ +{ + int *count_my_req_per_proc, count_my_req_procs, *buf_idx; + int i, l, proc; + ADIO_Offset fd_len, rem_len, curr_idx, off; + ADIOI_Access *my_req; + TRACE_ERR("Entering ADIOI_GPFS_Calc_my_req\n"); + +#ifdef AGGREGATION_PROFILE + MPE_Log_event (5024, 0, NULL); +#endif + *count_my_req_per_proc_ptr = (int *) ADIOI_Calloc(nprocs,sizeof(int)); + count_my_req_per_proc = *count_my_req_per_proc_ptr; +/* count_my_req_per_proc[i] gives the no. of contig. requests of this + process in process i's file domain. calloc initializes to zero. + I'm allocating memory of size nprocs, so that I can do an + MPI_Alltoall later on.*/ + + buf_idx = (int *) ADIOI_Malloc(nprocs*sizeof(int)); +/* buf_idx is relevant only if buftype_is_contig. + buf_idx[i] gives the index into user_buf where data received + from proc. i should be placed. This allows receives to be done + without extra buffer. This can't be done if buftype is not contig. */ + + /* initialize buf_idx to -1 */ + for (i=0; i < nprocs; i++) buf_idx[i] = -1; + + /* one pass just to calculate how much space to allocate for my_req; + * contig_access_count was calculated way back in ADIOI_Calc_my_off_len() + */ + for (i=0; i < contig_access_count; i++) { + /* short circuit offset/len processing if len == 0 + * (zero-byte read/write */ + if (len_list[i] == 0) + continue; + off = offset_list[i]; + fd_len = len_list[i]; + /* note: we set fd_len to be the total size of the access. then + * ADIOI_Calc_aggregator() will modify the value to return the + * amount that was available from the file domain that holds the + * first part of the access. + */ + /* BES */ + proc = ADIOI_GPFS_Calc_aggregator(fd, off, min_st_offset, &fd_len, fd_size, + fd_start, fd_end); + count_my_req_per_proc[proc]++; + + /* figure out how much data is remaining in the access (i.e. wasn't + * part of the file domain that had the starting byte); we'll take + * care of this data (if there is any) in the while loop below. + */ + rem_len = len_list[i] - fd_len; + + while (rem_len > 0) { + off += fd_len; /* point to first remaining byte */ + fd_len = rem_len; /* save remaining size, pass to calc */ + proc = ADIOI_GPFS_Calc_aggregator(fd, off, min_st_offset, &fd_len, + fd_size, fd_start, fd_end); + + count_my_req_per_proc[proc]++; + rem_len -= fd_len; /* reduce remaining length by amount from fd */ + } + } + +/* now allocate space for my_req, offset, and len */ + + *my_req_ptr = (ADIOI_Access *) + ADIOI_Malloc(nprocs*sizeof(ADIOI_Access)); + my_req = *my_req_ptr; + + count_my_req_procs = 0; + for (i=0; i < nprocs; i++) { + if (count_my_req_per_proc[i]) { + my_req[i].offsets = (ADIO_Offset *) + ADIOI_Malloc(count_my_req_per_proc[i] * sizeof(ADIO_Offset)); + my_req[i].lens = (int *) + ADIOI_Malloc(count_my_req_per_proc[i] * sizeof(int)); + count_my_req_procs++; + } + my_req[i].count = 0; /* will be incremented where needed + later */ + } + +/* now fill in my_req */ + curr_idx = 0; + for (i=0; i<contig_access_count; i++) { + /* short circuit offset/len processing if len == 0 + * (zero-byte read/write */ + if (len_list[i] == 0) + continue; + off = offset_list[i]; + fd_len = len_list[i]; + proc = ADIOI_GPFS_Calc_aggregator(fd, off, min_st_offset, &fd_len, fd_size, + fd_start, fd_end); + + /* for each separate contiguous access from this process */ + if (buf_idx[proc] == -1) + { + ADIOI_Assert(curr_idx == (int) curr_idx); + buf_idx[proc] = (int) curr_idx; + } + + l = my_req[proc].count; + curr_idx += fd_len; + + rem_len = len_list[i] - fd_len; + + /* store the proc, offset, and len information in an array + * of structures, my_req. Each structure contains the + * offsets and lengths located in that process's FD, + * and the associated count. + */ + my_req[proc].offsets[l] = off; + ADIOI_Assert(fd_len == (int) fd_len); + my_req[proc].lens[l] = (int) fd_len; + my_req[proc].count++; + + while (rem_len > 0) { + off += fd_len; + fd_len = rem_len; + proc = ADIOI_GPFS_Calc_aggregator(fd, off, min_st_offset, &fd_len, + fd_size, fd_start, fd_end); + + if (buf_idx[proc] == -1) + { + ADIOI_Assert(curr_idx == (int) curr_idx); + buf_idx[proc] = (int) curr_idx; + } + + l = my_req[proc].count; + curr_idx += fd_len; + rem_len -= fd_len; + + my_req[proc].offsets[l] = off; + ADIOI_Assert(fd_len == (int) fd_len); + my_req[proc].lens[l] = (int) fd_len; + my_req[proc].count++; + } + } + + + +#ifdef AGG_DEBUG + for (i=0; i<nprocs; i++) { + if (count_my_req_per_proc[i] > 0) { + DBG_FPRINTF(stderr, "data needed from %d (count = %d):\n", i, + my_req[i].count); + for (l=0; l < my_req[i].count; l++) { + DBG_FPRINTF(stderr, " off[%d] = %lld, len[%d] = %d\n", l, + my_req[i].offsets[l], l, my_req[i].lens[l]); + } + } + DBG_FPRINTF(stderr, "buf_idx[%d] = 0x%x\n", i, buf_idx[i]); + } +#endif + + *count_my_req_procs_ptr = count_my_req_procs; + *buf_idx_ptr = buf_idx; +#ifdef AGGREGATION_PROFILE + MPE_Log_event (5025, 0, NULL); +#endif + TRACE_ERR("Leaving ADIOI_GPFS_Calc_my_req\n"); +} + +/* + * ADIOI_Calc_others_req (copied to bg and switched to all to all for performance) + * + * param[in] count_my_req_procs Number of processes whose file domain my + * request touches. + * param[in] count_my_req_per_proc count_my_req_per_proc[i] gives the no. of + * contig. requests of this process in + * process i's file domain. + * param[in] my_req A structure defining my request + * param[in] nprocs Number of nodes in the block + * param[in] myrank Rank of this node + * param[out] count_others_req_proc_ptr Number of processes whose requests lie in + * my process's file domain (including my + * process itself) + * param[out] others_req_ptr Array of other process' requests that lie + * in my process's file domain + */ +void ADIOI_GPFS_Calc_others_req(ADIO_File fd, int count_my_req_procs, + int *count_my_req_per_proc, + ADIOI_Access *my_req, + int nprocs, int myrank, + int *count_others_req_procs_ptr, + ADIOI_Access **others_req_ptr) +{ + TRACE_ERR("Entering ADIOI_GPFS_Calc_others_req\n"); +/* determine what requests of other processes lie in this process's + file domain */ + +/* count_others_req_procs = number of processes whose requests lie in + this process's file domain (including this process itself) + count_others_req_per_proc[i] indicates how many separate contiguous + requests of proc. i lie in this process's file domain. */ + + int *count_others_req_per_proc, count_others_req_procs; + int i; + ADIOI_Access *others_req; + + /* Parameters for MPI_Alltoallv */ + int *scounts, *sdispls, *rcounts, *rdispls; + + /* Parameters for MPI_Alltoallv. These are the buffers, which + * are later computed to be the lowest address of all buffers + * to be sent/received for offsets and lengths. Initialize to + * the highest possible address which is the current minimum. + */ + void *sendBufForOffsets=(void*)0xFFFFFFFFFFFFFFFF, + *sendBufForLens =(void*)0xFFFFFFFFFFFFFFFF, + *recvBufForOffsets=(void*)0xFFFFFFFFFFFFFFFF, + *recvBufForLens =(void*)0xFFFFFFFFFFFFFFFF; + +/* first find out how much to send/recv and from/to whom */ +#ifdef AGGREGATION_PROFILE + MPE_Log_event (5026, 0, NULL); +#endif + /* Send 1 int to each process. count_my_req_per_proc[i] is the number of + * requests that my process will do to the file domain owned by process[i]. + * Receive 1 int from each process. count_others_req_per_proc[i] is the number of + * requests that process[i] will do to the file domain owned by my process. + */ + count_others_req_per_proc = (int *) ADIOI_Malloc(nprocs*sizeof(int)); +/* cora2a1=timebase(); */ +/*for(i=0;i<nprocs;i++) ?*/ + MPI_Alltoall(count_my_req_per_proc, 1, MPI_INT, + count_others_req_per_proc, 1, MPI_INT, fd->comm); + +/* total_cora2a+=timebase()-cora2a1; */ + + /* Allocate storage for an array of other nodes' accesses of our + * node's file domain. Also allocate storage for the alltoallv + * parameters. + */ + *others_req_ptr = (ADIOI_Access *) + ADIOI_Malloc(nprocs*sizeof(ADIOI_Access)); + others_req = *others_req_ptr; + + scounts = ADIOI_Malloc(nprocs*sizeof(int)); + sdispls = ADIOI_Malloc(nprocs*sizeof(int)); + rcounts = ADIOI_Malloc(nprocs*sizeof(int)); + rdispls = ADIOI_Malloc(nprocs*sizeof(int)); + + /* If process[i] has any requests in my file domain, + * initialize an ADIOI_Access structure that will describe each request + * from process[i]. The offsets, lengths, and buffer pointers still need + * to be obtained to complete the setting of this structure. + */ + count_others_req_procs = 0; + for (i=0; i<nprocs; i++) { + if (count_others_req_per_proc[i]) + { + others_req[i].count = count_others_req_per_proc[i]; + + others_req[i].offsets = (ADIO_Offset *) + ADIOI_Malloc(count_others_req_per_proc[i]*sizeof(ADIO_Offset)); + others_req[i].lens = (int *) + ADIOI_Malloc(count_others_req_per_proc[i]*sizeof(int)); + + if ( (MPIR_Upint)others_req[i].offsets < (MPIR_Upint)recvBufForOffsets ) + recvBufForOffsets = others_req[i].offsets; + if ( (MPIR_Upint)others_req[i].lens < (MPIR_Upint)recvBufForLens ) + recvBufForLens = others_req[i].lens; + + others_req[i].mem_ptrs = (MPI_Aint *) + ADIOI_Malloc(count_others_req_per_proc[i]*sizeof(MPI_Aint)); + + count_others_req_procs++; + } + else + { + others_req[i].count = 0; + others_req[i].offsets = NULL; + others_req[i].lens = NULL; + } + } + /* If no recv buffer was allocated in the loop above, make it NULL */ + if ( recvBufForOffsets == (void*)0xFFFFFFFFFFFFFFFF) recvBufForOffsets = NULL; + if ( recvBufForLens == (void*)0xFFFFFFFFFFFFFFFF) recvBufForLens = NULL; + + /* Now send the calculated offsets and lengths to respective processes */ + + /************************/ + /* Exchange the offsets */ + /************************/ + + /* Determine the lowest sendBufForOffsets/Lens */ + for (i=0; i<nprocs; i++) + { + if ( (my_req[i].count) && + ((MPIR_Upint)my_req[i].offsets <= (MPIR_Upint)sendBufForOffsets) ) + { + sendBufForOffsets = my_req[i].offsets; + } + + if ( (my_req[i].count) && + ((MPIR_Upint)my_req[i].lens <= (MPIR_Upint)sendBufForLens) ) + { + sendBufForLens = my_req[i].lens; + } + } + + /* If no send buffer was found in the loop above, make it NULL */ + if ( sendBufForOffsets == (void*)0xFFFFFFFFFFFFFFFF) sendBufForOffsets = NULL; + if ( sendBufForLens == (void*)0xFFFFFFFFFFFFFFFF) sendBufForLens = NULL; + + /* Calculate the displacements from the sendBufForOffsets/Lens */ + for (i=0; i<nprocs; i++) + { + /* Send these offsets to process i.*/ + scounts[i] = count_my_req_per_proc[i]; + if ( scounts[i] == 0 ) + sdispls[i] = 0; + else + sdispls[i] = (int) + ( ( (MPIR_Upint)my_req[i].offsets - + (MPIR_Upint)sendBufForOffsets ) / + (MPIR_Upint)sizeof(ADIO_Offset) ); + + /* Receive these offsets from process i.*/ + rcounts[i] = count_others_req_per_proc[i]; + if ( rcounts[i] == 0 ) + rdispls[i] = 0; + else + rdispls[i] = (int) + ( ( (MPIR_Upint)others_req[i].offsets - + (MPIR_Upint)recvBufForOffsets ) / + (MPIR_Upint)sizeof(ADIO_Offset) ); + } + + /* Exchange the offsets */ + MPI_Alltoallv(sendBufForOffsets, + scounts, sdispls, ADIO_OFFSET, + recvBufForOffsets, + rcounts, rdispls, ADIO_OFFSET, + fd->comm); + + /************************/ + /* Exchange the lengths */ + /************************/ + + for (i=0; i<nprocs; i++) + { + /* Send these lengths to process i.*/ + scounts[i] = count_my_req_per_proc[i]; + if ( scounts[i] == 0 ) + sdispls[i] = 0; + else + sdispls[i] = (int) + ( ( (MPIR_Upint)my_req[i].lens - + (MPIR_Upint)sendBufForLens ) / + (MPIR_Upint) sizeof(int) ); + + /* Receive these offsets from process i. */ + rcounts[i] = count_others_req_per_proc[i]; + if ( rcounts[i] == 0 ) + rdispls[i] = 0; + else + rdispls[i] = (int) + ( ( (MPIR_Upint)others_req[i].lens - + (MPIR_Upint)recvBufForLens ) / + (MPIR_Upint) sizeof(int) ); + } + + /* Exchange the lengths */ + MPI_Alltoallv(sendBufForLens, + scounts, sdispls, MPI_INT, + recvBufForLens, + rcounts, rdispls, MPI_INT, + fd->comm); + + /* Clean up */ + ADIOI_Free(count_others_req_per_proc); + ADIOI_Free (scounts); + ADIOI_Free (sdispls); + ADIOI_Free (rcounts); + ADIOI_Free (rdispls); + + *count_others_req_procs_ptr = count_others_req_procs; +#ifdef AGGREGATION_PROFILE + MPE_Log_event (5027, 0, NULL); +#endif + TRACE_ERR("Leaving ADIOI_GPFS_Calc_others_req\n"); +} ----------------------------------------------------------------------- Summary of changes: src/mpi/romio/adio/ad_gpfs/ad_gpfs_aggrs.c | 1694 ++++++++++---------- src/mpi/romio/adio/ad_gpfs/ad_gpfs_rdcoll.c | 4 +- src/mpi/romio/adio/ad_gpfs/ad_gpfs_wrcoll.c | 8 +- src/mpi/romio/adio/ad_lustre/ad_lustre_aggregate.c | 4 +- src/mpi/romio/adio/ad_lustre/ad_lustre_wrcoll.c | 2 +- src/mpi/romio/adio/common/Makefile.mk | 3 +- src/mpi/romio/adio/common/ad_aggregate.c | 18 +- src/mpi/romio/adio/common/ad_read_coll.c | 4 +- src/mpi/romio/adio/common/ad_write_coll.c | 6 +- src/mpi/romio/adio/common/utils.c | 83 + src/mpi/romio/adio/include/adioi.h | 11 +- test/mpi/io/bigtype.c | 5 +- 12 files changed, 965 insertions(+), 877 deletions(-) create mode 100644 src/mpi/romio/adio/common/utils.c hooks/post-receive -- MPICH primary repository