From 1ba06a8395d3223e7fde349fd5b733e2d8ac79ac Mon Sep 17 00:00:00 2001 From: Quincey Koziol Date: Fri, 12 Aug 2005 10:53:55 -0500 Subject: [svn-r11235] Purpose: Code cleanup Description: Refactor, simplify and cleanup Kent's recent checking for collective chunk I/O. There's a bug that I need to talk to Kent about and some more cleanups still, but this is reasonable for an interim point. Platforms tested: FreeBSD 4.11 (sleipnir) w/parallel Too minor for h5committest --- src/H5Dio.c | 841 ++++++++++++++++++++++++++-------------------------------- src/H5Dmpio.c | 28 +- src/H5Dpkg.h | 8 +- src/H5Smpio.c | 78 +++--- 4 files changed, 430 insertions(+), 525 deletions(-) diff --git a/src/H5Dio.c b/src/H5Dio.c index fa3f49e..3570cf6 100644 --- a/src/H5Dio.c +++ b/src/H5Dio.c @@ -78,7 +78,7 @@ static herr_t H5D_read(H5D_t *dataset, hid_t mem_type_id, static herr_t H5D_write(H5D_t *dataset, hid_t mem_type_id, const H5S_t *mem_space, const H5S_t *file_space, hid_t dset_xfer_plist, const void *buf); -static herr_t +static herr_t H5D_contig_read(H5D_io_info_t *io_info, hsize_t nelmts, const H5T_t *mem_type, const H5S_t *mem_space, const H5S_t *file_space, H5T_path_t *tpath, @@ -89,20 +89,20 @@ H5D_contig_write(H5D_io_info_t *io_info, hsize_t nelmts, const H5S_t *file_space, H5T_path_t *tpath, hid_t src_id, hid_t dst_id, const void *buf); static herr_t -H5D_chunk_read(H5D_io_info_t *io_info, hsize_t nelmts, +H5D_chunk_read(H5D_io_info_t *io_info, hsize_t nelmts, const H5T_t *mem_type, const H5S_t *mem_space, const H5S_t *file_space, H5T_path_t *tpath, hid_t src_id, hid_t dst_id, void *buf/*out*/); static herr_t -H5D_chunk_write(H5D_io_info_t *io_info, hsize_t nelmts, +H5D_chunk_write(H5D_io_info_t *io_info, hsize_t nelmts, const H5T_t *mem_type, const H5S_t *mem_space, const H5S_t *file_space, H5T_path_t *tpath, hid_t src_id, hid_t dst_id, const void *buf); #ifdef H5_HAVE_PARALLEL -static herr_t -H5D_io_assist_mpio(hid_t dxpl_id, H5D_dxpl_cache_t *dxpl_cache, +static herr_t +H5D_io_assist_mpio(hid_t dxpl_id, H5D_dxpl_cache_t *dxpl_cache, hbool_t *xfer_mode_changed); -static herr_t +static herr_t H5D_io_restore_mpio(hid_t dxpl_id); static htri_t H5D_get_collective_io_consensus(const H5F_t *file, @@ -113,15 +113,15 @@ static herr_t H5D_mpio_get_mini_chunk(const H5D_t *dset, const H5S_t *mem_space, const H5S_t *file_space, int *min_chunkf); -static herr_t +static herr_t H5D_obtain_duplicate_pid(hid_t dxpl_id, - hid_t* dp_id, + hid_t* dp_id, H5D_dxpl_cache_t **cache); #endif /* H5_HAVE_PARALLEL */ /* I/O info operations */ -static herr_t +static herr_t H5D_ioinfo_init(H5D_t *dset, const H5D_dxpl_cache_t *dxpl_cache, hid_t dxpl_id, hid_t dp_dxpl_id, H5D_dxpl_cache_t *dp_dxpl_cache, const H5S_t *mem_space, const H5S_t *file_space, @@ -371,7 +371,7 @@ H5D_get_dxpl_cache_real(hid_t dxpl_id, H5D_dxpl_cache_t *cache) /* Get the data transform property */ if(H5P_get(dx_plist, H5D_XFER_XFORM_NAME, &cache->data_xform_prop)<0) HGOTO_ERROR (H5E_PLIST, H5E_CANTGET, FAIL, "Can't retrieve data transform info") - + done: FUNC_LEAVE_NOAPI(ret_value) } /* H5D_get_dxpl_cache_real() */ @@ -427,7 +427,7 @@ done: * Purpose: Reads (part of) a DSET from the file into application * memory BUF. The part of the dataset to read is defined with * MEM_SPACE_ID and FILE_SPACE_ID. The data points are - * converted from their file type to the MEM_TYPE_ID specified. + * converted from their file type to the MEM_TYPE_ID specified. * Additional miscellaneous data transfer properties can be * passed to this function with the PLIST_ID argument. * @@ -447,12 +447,12 @@ done: * Return: Non-negative on success/Negative on failure * * Errors: - * ARGS BADTYPE Not a data space. - * ARGS BADTYPE Not a data type. - * ARGS BADTYPE Not a dataset. - * ARGS BADTYPE Not xfer parms. - * ARGS BADVALUE No output buffer. - * DATASET READERROR Can't read data. + * ARGS BADTYPE Not a data space. + * ARGS BADTYPE Not a data type. + * ARGS BADTYPE Not a dataset. + * ARGS BADTYPE Not xfer parms. + * ARGS BADVALUE No output buffer. + * DATASET READERROR Can't read data. * * Programmer: Robb Matzke * Thursday, December 4, 1997 @@ -639,7 +639,7 @@ done: * Changed the way to retrieve property for generic property list. * * Raymond Lu, 2002-2-26 - * For the new fill value design, data space can either be allocated + * For the new fill value design, data space can either be allocated * or not allocated at this stage. Fill value or data from space is * returned to outgoing buffer. * @@ -663,9 +663,8 @@ H5D_read(H5D_t *dataset, hid_t mem_type_id, const H5S_t *mem_space, H5D_dxpl_cache_t *dp_dxpl_cache=&_dp_dxpl_cache; /* Data transfer property cache */ hid_t dp_id; #ifdef H5_HAVE_PARALLEL - hbool_t xfer_mode_changed=FALSE; /* Whether the transfer mode was changed */ - H5FD_mpio_xfer_t xfer_mode; - + hbool_t xfer_mode_changed=FALSE; /* Whether the transfer mode was changed */ + H5FD_mpio_xfer_t xfer_mode; #ifdef H5_HAVE_INSTRUMENTED_LIBRARY int prop_value,new_value; htri_t check_prop; @@ -724,12 +723,12 @@ H5D_read(H5D_t *dataset, hid_t mem_type_id, const H5S_t *mem_space, /* Retrieve dataset properties */ /* */ - /* If space hasn't been allocated and not using external storage, + /* If space hasn't been allocated and not using external storage, * return fill value to buffer if fill time is upon allocation, or - * do nothing if fill time is never. If the dataset is compact and + * do nothing if fill time is never. If the dataset is compact and * fill time is NEVER, there is no way to tell whether part of data - * has been overwritten. So just proceed in reading. - */ + * has been overwritten. So just proceed in reading. + */ if(nelmts > 0 && dataset->shared->efl.nused==0 && ((dataset->shared->layout.type==H5D_CONTIGUOUS && !H5F_addr_defined(dataset->shared->layout.u.contig.addr)) || (dataset->shared->layout.type==H5D_CHUNKED && !H5F_addr_defined(dataset->shared->layout.u.chunk.addr)))) { @@ -785,12 +784,11 @@ H5D_read(H5D_t *dataset, hid_t mem_type_id, const H5S_t *mem_space, } /* end switch */ #ifdef H5_HAVE_PARALLEL - /* Obtain duplicate property list id. This is used to handle + /* Obtain duplicate property list id. This is used to handle collective chunk IO. */ - - if(H5D_obtain_duplicate_pid(dxpl_id,&dp_id,&dp_dxpl_cache)<0) + if(H5D_obtain_duplicate_pid(dxpl_id,&dp_id,&dp_dxpl_cache)<0) HGOTO_ERROR(H5E_PLIST, H5E_CANTGET, FAIL, "can't obtain duplicated property id") -#endif +#endif /* Set up I/O operation */ if(H5D_ioinfo_init(dataset,dxpl_cache,dxpl_id,dp_id,dp_dxpl_cache,mem_space,file_space,sconv_flags,&use_par_opt_io,&io_info)<0) HGOTO_ERROR (H5E_DATASET, H5E_UNSUPPORTED, FAIL, "unable to set up I/O operation") @@ -798,17 +796,17 @@ H5D_read(H5D_t *dataset, hid_t mem_type_id, const H5S_t *mem_space, #ifdef H5_HAVE_PARALLEL #ifdef H5_HAVE_INSTRUMENTED_LIBRARY /**** Test for collective chunk IO - notice the following code should be removed after + notice the following code should be removed after a more general collective chunk IO algorithm is applied. */ if(dataset->shared->layout.type == H5D_CHUNKED) { /*only check for chunking storage */ check_prop = H5Pexist(dxpl_id,H5D_XFER_COLL_CHUNK_NAME); - if(check_prop < 0) + if(check_prop < 0) HGOTO_ERROR(H5E_PLIST, H5E_UNSUPPORTED, FAIL, "unable to check property list"); if(check_prop > 0) { - if(H5Pget(dxpl_id,H5D_XFER_COLL_CHUNK_NAME,&prop_value)<0) - HGOTO_ERROR(H5E_PLIST, H5E_UNSUPPORTED, FAIL, "unable to get property value"); + if(H5Pget(dxpl_id,H5D_XFER_COLL_CHUNK_NAME,&prop_value)<0) + HGOTO_ERROR(H5E_PLIST, H5E_UNSUPPORTED, FAIL, "unable to get property value"); if(!use_par_opt_io) { new_value = 0; if(H5Pset(dxpl_id,H5D_XFER_COLL_CHUNK_NAME,&new_value)<0) @@ -873,7 +871,7 @@ done: * Added the code that when it detects it is not safe to process a * COLLECTIVE write request without hanging, it changes it to * INDEPENDENT calls. - * + * * Albert Cheng, 2000-11-27 * Changed to use the optimized MPIO transfer for Collective calls only. * @@ -881,8 +879,8 @@ done: * Changed the way to retrieve property for generic property list. * * Raymond Lu, 2002-2-26 - * For the new fill value design, space may not be allocated until - * this function is called. Allocate and initialize space if it + * For the new fill value design, space may not be allocated until + * this function is called. Allocate and initialize space if it * hasn't been. * * QAK - 2002/04/02 @@ -907,8 +905,8 @@ H5D_write(H5D_t *dataset, hid_t mem_type_id, const H5S_t *mem_space, H5D_dxpl_cache_t _dp_dxpl_cache; /* Data transfer property cache buffer */ H5D_dxpl_cache_t *dp_dxpl_cache=&_dp_dxpl_cache; /* Data transfer property cache */ #ifdef H5_HAVE_PARALLEL - hbool_t xfer_mode_changed=FALSE; /* Whether the transfer mode was changed */ - H5FD_mpio_xfer_t xfer_mode; + hbool_t xfer_mode_changed=FALSE; /* Whether the transfer mode was changed */ + H5FD_mpio_xfer_t xfer_mode; int mpi_rank; #ifdef H5_HAVE_INSTRUMENTED_LIBRARY int prop_value,new_value; @@ -964,8 +962,8 @@ H5D_write(H5D_t *dataset, hid_t mem_type_id, const H5S_t *mem_space, if (!file_space) file_space = dataset->shared->space; - if (!mem_space) - mem_space = file_space; + if (!mem_space) + mem_space = file_space; if((snelmts = H5S_GET_SELECT_NPOINTS(mem_space))<0) HGOTO_ERROR (H5E_ARGS, H5E_BADVALUE, FAIL, "src dataspace has invalid selection") H5_ASSIGN_OVERFLOW(nelmts,snelmts,hssize_t,hsize_t); @@ -995,7 +993,7 @@ H5D_write(H5D_t *dataset, hid_t mem_type_id, const H5S_t *mem_space, /* */ /* Allocate data space and initialize it if it hasn't been. */ - if(nelmts > 0 && dataset->shared->efl.nused==0 && + if(nelmts > 0 && dataset->shared->efl.nused==0 && ((dataset->shared->layout.type==H5D_CONTIGUOUS && !H5F_addr_defined(dataset->shared->layout.u.contig.addr)) || (dataset->shared->layout.type==H5D_CHUNKED && !H5F_addr_defined(dataset->shared->layout.u.chunk.addr)))) { hssize_t file_nelmts; /* Number of elements in file dataset's dataspace */ @@ -1051,22 +1049,22 @@ H5D_write(H5D_t *dataset, hid_t mem_type_id, const H5S_t *mem_space, /* Set up I/O operation */ if(H5D_ioinfo_init(dataset,dxpl_cache,dxpl_id,dp_id,dp_dxpl_cache,mem_space,file_space,sconv_flags,&use_par_opt_io,&io_info)<0) HGOTO_ERROR (H5E_DATASET, H5E_UNSUPPORTED, FAIL, "unable to set up I/O operation") - + #ifdef H5_HAVE_PARALLEL #ifdef H5_HAVE_INSTRUMENTED_LIBRARY /**** Test for collective chunk IO - notice the following code should be removed after + notice the following code should be removed after a more general collective chunk IO algorithm is applied. */ if(dataset->shared->layout.type == H5D_CHUNKED) { /*only check for chunking storage */ - + check_prop = H5Pexist(dxpl_id,H5D_XFER_COLL_CHUNK_NAME); - if(check_prop < 0) + if(check_prop < 0) HGOTO_ERROR(H5E_PLIST, H5E_UNSUPPORTED, FAIL, "unable to check property list"); if(check_prop > 0) { - if(H5Pget(dxpl_id,H5D_XFER_COLL_CHUNK_NAME,&prop_value)<0) - HGOTO_ERROR(H5E_PLIST, H5E_UNSUPPORTED, FAIL, "unable to get property value"); + if(H5Pget(dxpl_id,H5D_XFER_COLL_CHUNK_NAME,&prop_value)<0) + HGOTO_ERROR(H5E_PLIST, H5E_UNSUPPORTED, FAIL, "unable to get property value"); if(!use_par_opt_io) { new_value = 0; if(H5Pset(dxpl_id,H5D_XFER_COLL_CHUNK_NAME,&new_value)<0) @@ -1137,7 +1135,7 @@ done: * *------------------------------------------------------------------------- */ -static herr_t +static herr_t H5D_contig_read(H5D_io_info_t *io_info, hsize_t nelmts, const H5T_t *mem_type, const H5S_t *mem_space, const H5S_t *file_space, H5T_path_t *tpath, @@ -1145,7 +1143,7 @@ H5D_contig_read(H5D_io_info_t *io_info, hsize_t nelmts, { H5D_t *dataset=io_info->dset; /* Local pointer to dataset info */ const H5D_dxpl_cache_t *dxpl_cache=io_info->dxpl_cache; /* Local pointer to dataset transfer info */ - herr_t status; /*function return status*/ + herr_t status; /*function return status*/ #ifdef H5S_DEBUG H5_timer_t timer; #endif @@ -1196,7 +1194,7 @@ H5D_contig_read(H5D_io_info_t *io_info, hsize_t nelmts, || dataset->shared->layout.type==H5D_COMPACT); H5_CHECK_OVERFLOW(nelmts,hsize_t,size_t); status = (io_info->ops.read)(io_info, - (size_t)nelmts, H5T_get_size(dataset->shared->type), + (size_t)nelmts, H5T_get_size(dataset->shared->type), file_space, mem_space, buf/*out*/); #ifdef H5S_DEBUG @@ -1237,7 +1235,7 @@ H5D_contig_read(H5D_io_info_t *io_info, hsize_t nelmts, target_size=(size_t)(nelmts*max_type_size); } /* end if */ request_nelmts = target_size / max_type_size; - + /* Sanity check elements in temporary buffer */ if (request_nelmts==0) HGOTO_ERROR (H5E_DATASET, H5E_CANTINIT, FAIL, "temporary buffer max size is too small") @@ -1296,10 +1294,10 @@ H5D_contig_read(H5D_io_info_t *io_info, hsize_t nelmts, #ifdef H5S_DEBUG H5_timer_begin(&timer); #endif - /* Sanity check that space is allocated, then read data from it */ + /* Sanity check that space is allocated, then read data from it */ assert(((dataset->shared->layout.type==H5D_CONTIGUOUS && H5F_addr_defined(dataset->shared->layout.u.contig.addr)) || (dataset->shared->layout.type==H5D_CHUNKED && H5F_addr_defined(dataset->shared->layout.u.chunk.addr))) - || dataset->shared->efl.nused>0 || + || dataset->shared->efl.nused>0 || dataset->shared->layout.type==H5D_COMPACT); n = H5D_select_fgath(io_info, file_space, &file_iter, smine_nelmts, @@ -1376,7 +1374,7 @@ done: H5FL_BLK_FREE(type_conv,tconv_buf); if (bkg_buf && NULL==dxpl_cache->bkgr_buf) H5FL_BLK_FREE(type_conv,bkg_buf); - + FUNC_LEAVE_NOAPI(ret_value) } /* end H5D_contig_read() */ @@ -1407,7 +1405,7 @@ H5D_contig_write(H5D_io_info_t *io_info, hsize_t nelmts, { H5D_t *dataset=io_info->dset; /* Local pointer to dataset info */ const H5D_dxpl_cache_t *dxpl_cache=io_info->dxpl_cache; /* Local pointer to dataset transfer info */ - herr_t status; /*function return status*/ + herr_t status; /*function return status*/ #ifdef H5S_DEBUG H5_timer_t timer; #endif @@ -1469,7 +1467,7 @@ H5D_contig_write(H5D_io_info_t *io_info, hsize_t nelmts, /* direct xfer accomplished successfully */ HGOTO_DONE(SUCCEED) } /* end if */ - + /* * This is the general case. */ @@ -1509,7 +1507,7 @@ H5D_contig_write(H5D_io_info_t *io_info, hsize_t nelmts, if (H5S_select_iter_init(&bkg_iter, file_space, dst_type_size)<0) HGOTO_ERROR (H5E_DATASET, H5E_CANTINIT, FAIL, "unable to initialize background selection information") bkg_iter_init=1; /*file selection iteration info has been initialized */ - + /* * Get a temporary buffer for type conversion unless the app has already * supplied one through the xfer properties. Instead of allocating a @@ -1518,7 +1516,7 @@ H5D_contig_write(H5D_io_info_t *io_info, hsize_t nelmts, * same size over and over. */ if(H5T_detect_class(dataset->shared->type, H5T_VLEN)) { - /* Old data is retrieved into background buffer for VL datatype. The + /* Old data is retrieved into background buffer for VL datatype. The * data is used later for freeing heap objects. */ need_bkg = H5T_BKG_YES; } else if (H5T_path_bkg(tpath)) { @@ -1573,7 +1571,7 @@ H5D_contig_write(H5D_io_info_t *io_info, hsize_t nelmts, #endif n = H5D_select_fgath(io_info, file_space, &bkg_iter, smine_nelmts, - bkg_buf/*out*/); + bkg_buf/*out*/); #ifdef H5S_DEBUG H5_timer_end(&(io_info->stats->stats[0].bkg_timer), &timer); @@ -1590,7 +1588,7 @@ H5D_contig_write(H5D_io_info_t *io_info, hsize_t nelmts, if (H5T_convert(tpath, src_id, dst_id, smine_nelmts, 0, 0, tconv_buf, bkg_buf, io_info->dxpl_id)<0) HGOTO_ERROR(H5E_DATASET, H5E_CANTINIT, FAIL, "data type conversion failed") - /* Do the data transform after the type conversion (since we're using dataset->shared->type). */ + /* Do the data transform after the type conversion (since we're using dataset->shared->type). */ if(!H5Z_xform_noop(dxpl_cache->data_xform_prop)) if( H5Z_xform_eval(dxpl_cache->data_xform_prop, tconv_buf, smine_nelmts, dataset->shared->type) < 0) HGOTO_ERROR(H5E_ARGS, H5E_BADVALUE, FAIL, "Error performing data transform") @@ -1665,7 +1663,7 @@ H5D_chunk_read(H5D_io_info_t *io_info, hsize_t nelmts, const H5D_dxpl_cache_t *dxpl_cache=io_info->dxpl_cache; /* Local pointer to dataset transfer info */ fm_map fm; /* File<->memory mapping */ H5SL_node_t *chunk_node; /* Current node in chunk skip list */ - herr_t status; /*function return status*/ + herr_t status; /*function return status*/ #ifdef H5S_DEBUG H5_timer_t timer; #endif @@ -1675,7 +1673,7 @@ H5D_chunk_read(H5D_io_info_t *io_info, hsize_t nelmts, size_t target_size; /*desired buffer size */ size_t request_nelmts; /*requested strip mine */ hsize_t smine_start; /*strip mine start loc */ - size_t n, smine_nelmts; /*elements per strip */ + size_t n, smine_nelmts; /*elements per strip */ H5S_sel_iter_t mem_iter; /*memory selection iteration info*/ hbool_t mem_iter_init=0; /*memory selection iteration info has been initialized */ H5S_sel_iter_t bkg_iter; /*background iteration info*/ @@ -1690,12 +1688,12 @@ H5D_chunk_read(H5D_io_info_t *io_info, hsize_t nelmts, #ifdef H5_HAVE_PARALLEL int count_chunk,mpi_rank, mpi_code,min_num_chunk,is_regular,all_regular; - hid_t temp_id; + hid_t temp_id; MPI_Comm comm; #endif FUNC_ENTER_NOAPI_NOINIT(H5D_chunk_read) - + /* Map elements between file and memory for each chunk*/ if(H5D_create_chunk_map(dataset, mem_type, file_space, mem_space, &fm)<0) HGOTO_ERROR(H5E_DATASET, H5E_CANTINIT, FAIL, "can't build chunk mapping") @@ -1719,13 +1717,15 @@ H5D_chunk_read(H5D_io_info_t *io_info, hsize_t nelmts, /* Get first node in chunk skip list */ chunk_node=H5SL_first(fm.fsel); + #ifdef H5_HAVE_PARALLEL if(io_info->dxpl_cache->xfer_mode == H5FD_MPIO_COLLECTIVE) { - if(H5D_mpio_get_mini_chunk(dataset,mem_space,file_space,&min_num_chunk)<0) + if(H5D_mpio_get_mini_chunk(dataset,mem_space,file_space,&min_num_chunk)<0) HGOTO_ERROR(H5E_DATASET, H5E_CANTGET, FAIL, "can't get minimum number of chunk") - } + } count_chunk = 0; #endif + /* Iterate through chunks to be operated on */ while(chunk_node) { H5D_chunk_info_t *chunk_info; /* chunk information */ @@ -1738,152 +1738,110 @@ H5D_chunk_read(H5D_io_info_t *io_info, hsize_t nelmts, store.chunk.index = chunk_info->index; #ifdef H5_HAVE_PARALLEL - - count_chunk++; - if(io_info->dxpl_cache->xfer_mode == H5FD_MPIO_COLLECTIVE) { + count_chunk++; + if(io_info->dxpl_cache->xfer_mode == H5FD_MPIO_COLLECTIVE) { /* If the number of chunk is greater than minimum number of chunk, Do independent read */ - if(count_chunk <= min_num_chunk) { + if(count_chunk <= min_num_chunk) { #ifndef H5_MPI_COMPLEX_DERIVED_DATATYPE_WORKS - if(H5S_SELECT_IS_REGULAR(chunk_info->fspace) == TRUE && - H5S_SELECT_IS_REGULAR(chunk_info->mspace) == TRUE) - is_regular = 1; - else is_regular = 0; - - /* Getting MPI communicator and rank */ - if((comm = H5F_mpi_get_comm(dataset->ent.file))==MPI_COMM_NULL) - HGOTO_ERROR(H5E_DATASPACE, H5E_CANTGET, FAIL, "can't retrieve MPI communicator") - if((mpi_rank = H5F_mpi_get_rank(dataset->ent.file))<0) - HGOTO_ERROR(H5E_DATASPACE, H5E_CANTGET, FAIL, "can't retrieve MPI rank") - - if (MPI_SUCCESS != (mpi_code= MPI_Reduce(&all_regular,&is_regular,1,MPI_INT,MPI_MIN,0,comm))) - HMPI_GOTO_ERROR(FAIL, "MPI_Reduce failed", mpi_code) - if (MPI_SUCCESS != (mpi_code= MPI_Bcast(&all_regular,1,MPI_INT,0,comm))) - HMPI_GOTO_ERROR(FAIL, "MPI_Bcast failed", mpi_code) -#endif - } - - if(count_chunk > min_num_chunk) { - temp_id = io_info->dxpl_id; - io_info->dxpl_id = io_info->dp_dxpl_id; - status = (io_info->ops_sca.read)(io_info, - chunk_info->chunk_points, H5T_get_size(dataset->shared->type), - chunk_info->fspace, chunk_info->mspace, - buf); - /* Check return value from optimized read */ - if (status<0) - HGOTO_ERROR(H5E_DATASET, H5E_READERROR, FAIL, "optimized read failed") + if(H5S_SELECT_IS_REGULAR(chunk_info->fspace) == TRUE && + H5S_SELECT_IS_REGULAR(chunk_info->mspace) == TRUE) + is_regular = 1; + else + is_regular = 0; + + /* Getting MPI communicator and rank */ + if((comm = H5F_mpi_get_comm(dataset->ent.file))==MPI_COMM_NULL) + HGOTO_ERROR(H5E_DATASPACE, H5E_CANTGET, FAIL, "can't retrieve MPI communicator") + if((mpi_rank = H5F_mpi_get_rank(dataset->ent.file))<0) + HGOTO_ERROR(H5E_DATASPACE, H5E_CANTGET, FAIL, "can't retrieve MPI rank") + + if (MPI_SUCCESS != (mpi_code= MPI_Reduce(&all_regular,&is_regular,1,MPI_INT,MPI_MIN,0,comm))) + HMPI_GOTO_ERROR(FAIL, "MPI_Reduce failed", mpi_code) + if (MPI_SUCCESS != (mpi_code= MPI_Bcast(&all_regular,1,MPI_INT,0,comm))) + HMPI_GOTO_ERROR(FAIL, "MPI_Bcast failed", mpi_code) +#endif /* H5_MPI_COMPLEX_DERIVED_DATATYPE_WORKS */ + } + + if(count_chunk > min_num_chunk) { + temp_id = io_info->dxpl_id; + io_info->dxpl_id = io_info->dp_dxpl_id; + status = (io_info->ops_sca.read)(io_info, + chunk_info->chunk_points, H5T_get_size(dataset->shared->type), + chunk_info->fspace, chunk_info->mspace, buf); io_info->dxpl_id = temp_id; - } - - - else if((H5S_SELECT_IS_REGULAR(chunk_info->fspace) == FALSE)|| - (H5S_SELECT_IS_REGULAR(chunk_info->mspace) == FALSE)){ - + } + else if((H5S_SELECT_IS_REGULAR(chunk_info->fspace) == FALSE) || + (H5S_SELECT_IS_REGULAR(chunk_info->mspace) == FALSE)) { #ifndef H5_MPI_COMPLEX_DERIVED_DATATYPE_WORKS - /* Perform the independent read operation */ - temp_id = io_info->dxpl_id; - io_info->dxpl_id = io_info->dp_dxpl_id; - status = (io_info->ops_sca.read)(io_info, - chunk_info->chunk_points, H5T_get_size(dataset->shared->type), - chunk_info->fspace, chunk_info->mspace, - buf); - /* Check return value from optimized read */ - if (status<0) - HGOTO_ERROR(H5E_DATASET, H5E_READERROR, FAIL, "optimized read failed") - io_info->dxpl_id = temp_id; -#else - - /* Perform the actual collective read operation */ - status = (io_info->ops.read)(io_info, - chunk_info->chunk_points, H5T_get_size(dataset->shared->type), - chunk_info->fspace, chunk_info->mspace, - buf); - /* Check return value from optimized read */ - if (status<0) - HGOTO_ERROR(H5E_DATASET, H5E_READERROR, FAIL, "optimized read failed") -#endif - } - - else { - /* For regular selection, + /* Perform the independent read operation */ + temp_id = io_info->dxpl_id; + io_info->dxpl_id = io_info->dp_dxpl_id; + status = (io_info->ops_sca.read)(io_info, + chunk_info->chunk_points, H5T_get_size(dataset->shared->type), + chunk_info->fspace, chunk_info->mspace, buf); + io_info->dxpl_id = temp_id; +#else /* H5_MPI_COMPLEX_DERIVED_DATATYPE_WORKS */ + /* Perform the actual collective read operation */ + status = (io_info->ops.read)(io_info, + chunk_info->chunk_points, H5T_get_size(dataset->shared->type), + chunk_info->fspace, chunk_info->mspace, buf); +#endif /* H5_MPI_COMPLEX_DERIVED_DATATYPE_WORKS */ + } + else { + /* For regular selection, if MPI_COMPLEX_DERIVED_DATATYPE is not defined, unless spaces for all processors are regular, independent read operation should be performed.*/ #ifndef H5_MPI_COMPLEX_DERIVED_DATATYPE_WORKS - if(!all_regular) { - - /* Perform the independent read operation */ - temp_id = io_info->dxpl_id; - io_info->dxpl_id = io_info->dp_dxpl_id; - status = (io_info->ops_sca.read)(io_info, - chunk_info->chunk_points, H5T_get_size(dataset->shared->type), - chunk_info->fspace, chunk_info->mspace, - buf); - /* Check return value from optimized read */ - if (status<0) - HGOTO_ERROR(H5E_DATASET, H5E_READERROR, FAIL, "optimized read failed") - io_info->dxpl_id = temp_id; - } - - else { - /* For regular collective read in parallel*/ - /* Perform the read operation */ - status = (io_info->ops.read)(io_info, - chunk_info->chunk_points, - H5T_get_size(dataset->shared->type), - chunk_info->fspace, chunk_info->mspace, - buf); - /* Check return value from optimized read */ - if (status<0) - HGOTO_ERROR(H5E_DATASET, H5E_READERROR, FAIL, "optimized read failed") - } -#else - - /* For regular collective read in parallel*/ - /* Perform the read operation */ - status = (io_info->ops.read)(io_info, - chunk_info->chunk_points, - H5T_get_size(dataset->shared->type), - chunk_info->fspace, chunk_info->mspace, - buf); - /* Check return value from optimized read */ - if (status<0) - HGOTO_ERROR(H5E_DATASET, H5E_READERROR, FAIL, "optimized read failed") -#endif + if(!all_regular) { + /* Perform the independent read operation */ + temp_id = io_info->dxpl_id; + io_info->dxpl_id = io_info->dp_dxpl_id; + status = (io_info->ops_sca.read)(io_info, + chunk_info->chunk_points, H5T_get_size(dataset->shared->type), + chunk_info->fspace, chunk_info->mspace, buf); + io_info->dxpl_id = temp_id; + } + else { + /* For regular collective read in parallel*/ + /* Perform the read operation */ + status = (io_info->ops.read)(io_info, + chunk_info->chunk_points, H5T_get_size(dataset->shared->type), + chunk_info->fspace, chunk_info->mspace, buf); } - - } - else { - /* For regular independent read in parallel*/ - /* Perform the read operation */ - status = (io_info->ops.read)(io_info, - chunk_info->chunk_points, - H5T_get_size(dataset->shared->type), - chunk_info->fspace, chunk_info->mspace, - buf); - /* Check return value from optimized read */ - if (status<0) - HGOTO_ERROR(H5E_DATASET, H5E_READERROR, FAIL, "optimized read failed") - } - -#else - /* Perform the actual read operation for sequential*/ - status = (io_info->ops.read)(io_info, - chunk_info->chunk_points, - H5T_get_size(dataset->shared->type), - chunk_info->fspace, chunk_info->mspace, - buf); - /* Check return value from optimized read */ - if (status<0) - HGOTO_ERROR(H5E_DATASET, H5E_READERROR, FAIL, "optimized read failed") -#endif +#else /* H5_MPI_COMPLEX_DERIVED_DATATYPE_WORKS */ + /* For regular collective read in parallel*/ + /* Perform the read operation */ + status = (io_info->ops.read)(io_info, + chunk_info->chunk_points, H5T_get_size(dataset->shared->type), + chunk_info->fspace, chunk_info->mspace, buf); +#endif /* H5_MPI_COMPLEX_DERIVED_DATATYPE_WORKS */ + } + } + else { + /* For regular independent read in parallel*/ + /* Perform the read operation */ + status = (io_info->ops.read)(io_info, + chunk_info->chunk_points, H5T_get_size(dataset->shared->type), + chunk_info->fspace, chunk_info->mspace, buf); + } +#else /* H5_HAVE_PARALLEL */ + /* Perform the actual read operation for sequential*/ + status = (io_info->ops.read)(io_info, + chunk_info->chunk_points, H5T_get_size(dataset->shared->type), + chunk_info->fspace, chunk_info->mspace, buf); +#endif /* H5_HAVE_PARALLEL */ + + /* Check return value from optimized read */ + if (status<0) + HGOTO_ERROR(H5E_DATASET, H5E_READERROR, FAIL, "optimized read failed") /* Get the next chunk node in the skip list */ chunk_node=H5SL_next(chunk_node); } /* end while */ - + #ifdef H5S_DEBUG H5_timer_end(&(io_info->stats->stats[1].read_timer), &timer); io_info->stats->stats[1].read_nbytes += nelmts * H5T_get_size(dataset->shared->type); @@ -1918,7 +1876,7 @@ H5D_chunk_read(H5D_io_info_t *io_info, hsize_t nelmts, target_size=(size_t)(nelmts*max_type_size); } /* end if */ request_nelmts = target_size / max_type_size; - + /* Sanity check elements in temporary buffer */ if (request_nelmts==0) HGOTO_ERROR (H5E_DATASET, H5E_CANTINIT, FAIL, "temporary buffer max size is too small") @@ -1951,7 +1909,7 @@ H5D_chunk_read(H5D_io_info_t *io_info, hsize_t nelmts, if((bkg_buf=H5FL_BLK_CALLOC(type_conv,(request_nelmts*dst_type_size)))==NULL) HGOTO_ERROR (H5E_RESOURCE, H5E_NOSPACE, FAIL, "memory allocation failed for background conversion") } /* end if */ - + /* Loop over all the chunks, performing I/O on each */ /* Get first node in chunk skip list */ @@ -1974,16 +1932,16 @@ H5D_chunk_read(H5D_io_info_t *io_info, hsize_t nelmts, if (H5S_select_iter_init(&bkg_iter, chunk_info->mspace, dst_type_size)<0) HGOTO_ERROR (H5E_DATASET, H5E_CANTINIT, FAIL, "unable to initialize background selection information") bkg_iter_init=1; /*file selection iteration info has been initialized */ - + /* Pass in chunk's coordinates in a union*/ store.chunk.offset = chunk_info->coords; store.chunk.index = chunk_info->index; - + for (smine_start=0; smine_startchunk_points; smine_start+=smine_nelmts) { /* Go figure out how many elements to read from the file */ assert(H5S_SELECT_ITER_NELMTS(&file_iter)==(chunk_info->chunk_points-smine_start)); smine_nelmts = (size_t)MIN(request_nelmts, (chunk_info->chunk_points-smine_start)); - + /* * Gather the data from disk into the data type conversion * buffer. Also gather data from application to background buffer @@ -1992,11 +1950,11 @@ H5D_chunk_read(H5D_io_info_t *io_info, hsize_t nelmts, #ifdef H5S_DEBUG H5_timer_begin(&timer); #endif - /* Sanity check that space is allocated, then read data from it */ + /* Sanity check that space is allocated, then read data from it */ assert(((dataset->shared->layout.type==H5D_CONTIGUOUS && H5F_addr_defined(dataset->shared->layout.u.contig.addr)) || (dataset->shared->layout.type==H5D_CHUNKED && H5F_addr_defined(dataset->shared->layout.u.chunk.addr))) || dataset->shared->efl.nused>0 || dataset->shared->layout.type==H5D_COMPACT); - n = H5D_select_fgath(io_info, + n = H5D_select_fgath(io_info, chunk_info->fspace, &file_iter, smine_nelmts, tconv_buf/*out*/); @@ -2022,11 +1980,11 @@ H5D_chunk_read(H5D_io_info_t *io_info, hsize_t nelmts, if (n!=smine_nelmts) HGOTO_ERROR (H5E_IO, H5E_READERROR, FAIL, "mem gather failed") } /* end if */ - + /* * Perform data type conversion. */ - if (H5T_convert(tpath, src_id, dst_id, smine_nelmts, 0, 0, + if (H5T_convert(tpath, src_id, dst_id, smine_nelmts, 0, 0, tconv_buf, bkg_buf, io_info->dxpl_id)<0) HGOTO_ERROR(H5E_DATASET, H5E_CANTINIT, FAIL, "data type conversion failed") @@ -2072,7 +2030,7 @@ H5D_chunk_read(H5D_io_info_t *io_info, hsize_t nelmts, /* Get the next chunk node in the skip list */ chunk_node=H5SL_next(chunk_node); } /* end while */ - + done: /* Release selection iterators, if necessary */ if(file_iter_init) { @@ -2092,11 +2050,11 @@ done: H5FL_BLK_FREE(type_conv,tconv_buf); if (bkg_buf && NULL==dxpl_cache->bkgr_buf) H5FL_BLK_FREE(type_conv,bkg_buf); - + /* Release chunk mapping information */ if(H5D_destroy_chunk_map(&fm) < 0) HDONE_ERROR(H5E_DATASET, H5E_CANTRELEASE, FAIL, "can't release chunk mapping") - + FUNC_LEAVE_NOAPI(ret_value) } /* H5D_chunk_read() */ @@ -2131,7 +2089,7 @@ H5D_chunk_write(H5D_io_info_t *io_info, hsize_t nelmts, const H5D_dxpl_cache_t *dxpl_cache=io_info->dxpl_cache; /* Local pointer to dataset transfer info */ fm_map fm; /* File<->memory mapping */ H5SL_node_t *chunk_node; /* Current node in chunk skip list */ - herr_t status; /*function return status*/ + herr_t status; /*function return status*/ #ifdef H5S_DEBUG H5_timer_t timer; #endif @@ -2141,7 +2099,7 @@ H5D_chunk_write(H5D_io_info_t *io_info, hsize_t nelmts, size_t target_size; /*desired buffer size */ size_t request_nelmts; /*requested strip mine */ hsize_t smine_start; /*strip mine start loc */ - size_t n, smine_nelmts; /*elements per strip */ + size_t n, smine_nelmts; /*elements per strip */ H5S_sel_iter_t mem_iter; /*memory selection iteration info*/ hbool_t mem_iter_init=0; /*memory selection iteration info has been initialized */ H5S_sel_iter_t bkg_iter; /*background iteration info*/ @@ -2158,15 +2116,15 @@ H5D_chunk_write(H5D_io_info_t *io_info, hsize_t nelmts, hid_t temp_id; int count_chunk,mpi_rank,mpi_code,min_num_chunk,is_regular,all_regular = 0; MPI_Comm comm; - + #endif FUNC_ENTER_NOAPI_NOINIT(H5D_chunk_write) - + /* Map elements between file and memory for each chunk*/ if(H5D_create_chunk_map(dataset, mem_type, file_space, mem_space, &fm)<0) HGOTO_ERROR(H5E_DATASET, H5E_CANTINIT, FAIL, "can't build chunk mapping") - + /* Set dataset storage for I/O info */ io_info->store=&store; @@ -2181,11 +2139,12 @@ H5D_chunk_write(H5D_io_info_t *io_info, hsize_t nelmts, #ifdef H5_HAVE_PARALLEL if(io_info->dxpl_cache->xfer_mode == H5FD_MPIO_COLLECTIVE) { - if(H5D_mpio_get_mini_chunk(dataset,mem_space,file_space,&min_num_chunk)<0) + if(H5D_mpio_get_mini_chunk(dataset,mem_space,file_space,&min_num_chunk)<0) HGOTO_ERROR(H5E_DATASET, H5E_CANTGET, FAIL, "can't get minimum number of chunk") - } + } count_chunk = 0; #endif + /* Get first node in chunk skip list */ chunk_node=H5SL_first(fm.fsel); @@ -2201,140 +2160,109 @@ H5D_chunk_write(H5D_io_info_t *io_info, hsize_t nelmts, store.chunk.index = chunk_info->index; #ifdef H5_HAVE_PARALLEL - - count_chunk++; - if(io_info->dxpl_cache->xfer_mode == H5FD_MPIO_COLLECTIVE) { + count_chunk++; + if(io_info->dxpl_cache->xfer_mode == H5FD_MPIO_COLLECTIVE) { /* If the number of chunk is greater than minimum number of chunk, Do independent write */ - if(count_chunk <= min_num_chunk) { + if(count_chunk <= min_num_chunk) { #ifndef H5_MPI_COMPLEX_DERIVED_DATATYPE_WORKS - if(H5S_SELECT_IS_REGULAR(chunk_info->fspace) == TRUE && - H5S_SELECT_IS_REGULAR(chunk_info->mspace) == TRUE) - is_regular = 1; - else is_regular = 0; - /* Getting MPI communicator and rank */ - if((comm = H5F_mpi_get_comm(dataset->ent.file))==MPI_COMM_NULL) - HGOTO_ERROR(H5E_DATASPACE, H5E_CANTGET, FAIL, "can't retrieve MPI communicator") - if((mpi_rank = H5F_mpi_get_rank(dataset->ent.file))<0) - HGOTO_ERROR(H5E_DATASPACE, H5E_CANTGET, FAIL, "can't retrieve MPI rank") - if (MPI_SUCCESS != (mpi_code= MPI_Reduce(&all_regular,&is_regular,1,MPI_INT,MPI_MIN,0,comm))) - HMPI_GOTO_ERROR(FAIL, "MPI_Reduce failed", mpi_code) - if (MPI_SUCCESS != (mpi_code= MPI_Bcast(&all_regular,1,MPI_INT,0,comm))) - HMPI_GOTO_ERROR(FAIL, "MPI_Bcast failed", mpi_code) -#endif - } - if(count_chunk > min_num_chunk) { - temp_id = io_info->dxpl_id; - io_info->dxpl_id = io_info->dp_dxpl_id; - fflush(stdout); - status = (io_info->ops_sca.write)(io_info, - chunk_info->chunk_points, H5T_get_size(dataset->shared->type), - chunk_info->fspace, chunk_info->mspace, - buf); - /* Check return value from optimized write */ - if (status<0) - HGOTO_ERROR(H5E_DATASET, H5E_WRITEERROR, FAIL, "optimized write failed") + if(H5S_SELECT_IS_REGULAR(chunk_info->fspace) == TRUE && + H5S_SELECT_IS_REGULAR(chunk_info->mspace) == TRUE) + is_regular = 1; + else + is_regular = 0; + + /* Getting MPI communicator and rank */ + if((comm = H5F_mpi_get_comm(dataset->ent.file))==MPI_COMM_NULL) + HGOTO_ERROR(H5E_DATASPACE, H5E_CANTGET, FAIL, "can't retrieve MPI communicator") + if((mpi_rank = H5F_mpi_get_rank(dataset->ent.file))<0) + HGOTO_ERROR(H5E_DATASPACE, H5E_CANTGET, FAIL, "can't retrieve MPI rank") + + if (MPI_SUCCESS != (mpi_code= MPI_Reduce(&all_regular,&is_regular,1,MPI_INT,MPI_MIN,0,comm))) + HMPI_GOTO_ERROR(FAIL, "MPI_Reduce failed", mpi_code) + if (MPI_SUCCESS != (mpi_code= MPI_Bcast(&all_regular,1,MPI_INT,0,comm))) + HMPI_GOTO_ERROR(FAIL, "MPI_Bcast failed", mpi_code) +#endif /* H5_MPI_COMPLEX_DERIVED_DATATYPE_WORKS */ + } + + if(count_chunk > min_num_chunk) { + temp_id = io_info->dxpl_id; + io_info->dxpl_id = io_info->dp_dxpl_id; + status = (io_info->ops_sca.write)(io_info, + chunk_info->chunk_points, H5T_get_size(dataset->shared->type), + chunk_info->fspace, chunk_info->mspace, buf); + + /* Check return value from optimized write */ + if (status<0) + HGOTO_ERROR(H5E_DATASET, H5E_WRITEERROR, FAIL, "optimized write failed") io_info->dxpl_id = temp_id; - } - - - else if((H5S_SELECT_IS_REGULAR(chunk_info->fspace) == FALSE)|| - (H5S_SELECT_IS_REGULAR(chunk_info->mspace) == FALSE)){ + } + else if((H5S_SELECT_IS_REGULAR(chunk_info->fspace) == FALSE) || + (H5S_SELECT_IS_REGULAR(chunk_info->mspace) == FALSE)) { #ifndef H5_MPI_COMPLEX_DERIVED_DATATYPE_WORKS - /* Perform the independent write operation */ - - temp_id = io_info->dxpl_id; - io_info->dxpl_id = io_info->dp_dxpl_id; - status = (io_info->ops_sca.write)(io_info, - chunk_info->chunk_points, H5T_get_size(dataset->shared->type), - chunk_info->fspace, chunk_info->mspace, - buf); - - /* Check return value from optimized write */ - if (status<0) - HGOTO_ERROR(H5E_DATASET, H5E_WRITEERROR, FAIL, "optimized write failed") - io_info->dxpl_id = temp_id; -#else - - /* Perform the actual collective write operation */ - status = (io_info->ops.write)(io_info, - chunk_info->chunk_points, H5T_get_size(dataset->shared->type), - chunk_info->fspace, chunk_info->mspace, - buf); - /* Check return value from optimized write */ - if (status<0) - HGOTO_ERROR(H5E_DATASET, H5E_WRITEERROR, FAIL, "optimized write failed") -#endif - } - - else { + /* Perform the independent write operation */ + temp_id = io_info->dxpl_id; + io_info->dxpl_id = io_info->dp_dxpl_id; + status = (io_info->ops_sca.write)(io_info, + chunk_info->chunk_points, H5T_get_size(dataset->shared->type), + chunk_info->fspace, chunk_info->mspace, buf); + io_info->dxpl_id = temp_id; +#else /* H5_MPI_COMPLEX_DERIVED_DATATYPE_WORKS */ + /* Perform the actual collective write operation */ + status = (io_info->ops.write)(io_info, + chunk_info->chunk_points, H5T_get_size(dataset->shared->type), + chunk_info->fspace, chunk_info->mspace, buf); +#endif /* H5_MPI_COMPLEX_DERIVED_DATATYPE_WORKS */ + } + else { #ifndef H5_MPI_COMPLEX_DERIVED_DATATYPE_WORKS - if(!all_regular) { - - /* Perform the independent write operation */ - temp_id = io_info->dxpl_id; - io_info->dxpl_id = io_info->dp_dxpl_id; - status = (io_info->ops_sca.write)(io_info, - chunk_info->chunk_points, H5T_get_size(dataset->shared->type), - chunk_info->fspace, chunk_info->mspace, - buf); - /* Check return value from optimized read */ - if (status<0) - HGOTO_ERROR(H5E_DATASET, H5E_READERROR, FAIL, "optimized read failed") - io_info->dxpl_id = temp_id; - } - else { - /* For regular selection, perform the collective write operation */ - status = (io_info->ops.write)(io_info, + if(!all_regular) { + /* Perform the independent write operation */ + temp_id = io_info->dxpl_id; + io_info->dxpl_id = io_info->dp_dxpl_id; + status = (io_info->ops_sca.write)(io_info, + chunk_info->chunk_points, H5T_get_size(dataset->shared->type), + chunk_info->fspace, chunk_info->mspace, buf); + io_info->dxpl_id = temp_id; + } + else { + /* For regular selection, perform the collective write operation */ + status = (io_info->ops.write)(io_info, + chunk_info->chunk_points, H5T_get_size(dataset->shared->type), + chunk_info->fspace, chunk_info->mspace, buf); + } +#else /* H5_MPI_COMPLEX_DERIVED_DATATYPE_WORKS */ + status = (io_info->ops.write)(io_info, + chunk_info->chunk_points, H5T_get_size(dataset->shared->type), + chunk_info->fspace, chunk_info->mspace, buf); +#endif /* H5_MPI_COMPLEX_DERIVED_DATATYPE_WORKS */ + } + } + else { + /* For independent parallel write*/ + /* Perform the write operation */ + status = (io_info->ops.write)(io_info, chunk_info->chunk_points, H5T_get_size(dataset->shared->type), - chunk_info->fspace, chunk_info->mspace, - buf); - /* Check return value from optimized write */ - if (status<0) - HGOTO_ERROR(H5E_DATASET, H5E_WRITEERROR, FAIL, "optimized write failed") - } -#else + chunk_info->fspace, chunk_info->mspace, buf); + } - status = (io_info->ops.write)(io_info, - chunk_info->chunk_points, H5T_get_size(dataset->shared->type), - chunk_info->fspace, chunk_info->mspace, - buf); - /* Check return value from optimized write */ - if (status<0) - HGOTO_ERROR(H5E_DATASET, H5E_WRITEERROR, FAIL, "optimized write failed") - -#endif - } - } - else { - /* For independent parallel write*/ - /* Perform the write operation */ - status = (io_info->ops.write)(io_info, - chunk_info->chunk_points, H5T_get_size(dataset->shared->type), - chunk_info->fspace, chunk_info->mspace, - buf); - /* Check return value from optimized write */ - if (status<0) - HGOTO_ERROR(H5E_DATASET, H5E_WRITEERROR, FAIL, "optimized write failed") - } +#else /* H5_HAVE_PARALLEL */ + /* Perform the actual write operation for sequential*/ + status = (io_info->ops.write)(io_info, + chunk_info->chunk_points, H5T_get_size(dataset->shared->type), + chunk_info->fspace, chunk_info->mspace, buf); +#endif /* H5_HAVE_PARALLEL */ -#else - /* Perform the actual write operation for sequential*/ - status = (io_info->ops.write)(io_info, - chunk_info->chunk_points, H5T_get_size(dataset->shared->type), - chunk_info->fspace, chunk_info->mspace, - buf); - /* Check return value from optimized write */ - if (status<0) - HGOTO_ERROR(H5E_DATASET, H5E_WRITEERROR, FAIL, "optimized write failed") -#endif + /* Check return value from optimized write */ + if (status<0) + HGOTO_ERROR(H5E_DATASET, H5E_WRITEERROR, FAIL, "optimized write failed") /* Get the next chunk node in the skip list */ chunk_node=H5SL_next(chunk_node); } /* end while */ - + #ifdef H5S_DEBUG H5_timer_end(&(io_info->stats->stats[0].write_timer), &timer); io_info->stats->stats[0].write_nbytes += nelmts * H5T_get_size(mem_type); @@ -2344,13 +2272,13 @@ H5D_chunk_write(H5D_io_info_t *io_info, hsize_t nelmts, /* direct xfer accomplished successfully */ HGOTO_DONE(SUCCEED) } /* end if */ - + /* * This is the general case (type conversion, usually). */ if(nelmts==0) HGOTO_DONE(SUCCEED) - + /* Compute element sizes and other parameters */ src_type_size = H5T_get_size(mem_type); dst_type_size = H5T_get_size(dataset->shared->type); @@ -2382,7 +2310,7 @@ H5D_chunk_write(H5D_io_info_t *io_info, hsize_t nelmts, * same size over and over. */ if(H5T_detect_class(dataset->shared->type, H5T_VLEN)) { - /* Old data is retrieved into background buffer for VL datatype. The + /* Old data is retrieved into background buffer for VL datatype. The * data is used later for freeing heap objects. */ need_bkg = H5T_BKG_YES; } else if (H5T_path_bkg(tpath)) { @@ -2429,16 +2357,16 @@ H5D_chunk_write(H5D_io_info_t *io_info, hsize_t nelmts, if (H5S_select_iter_init(&bkg_iter, chunk_info->fspace, dst_type_size)<0) HGOTO_ERROR (H5E_DATASET, H5E_CANTINIT, FAIL, "unable to initialize background selection information") bkg_iter_init=1; /*file selection iteration info has been initialized */ - + /*pass in chunk's coordinates in a union*/ store.chunk.offset = chunk_info->coords; store.chunk.index = chunk_info->index; - + for (smine_start=0; smine_startchunk_points; smine_start+=smine_nelmts) { /* Go figure out how many elements to read from the file */ assert(H5S_SELECT_ITER_NELMTS(&file_iter)==(chunk_info->chunk_points-smine_start)); smine_nelmts = (size_t)MIN(request_nelmts, (chunk_info->chunk_points-smine_start)); - + /* * Gather the data from disk into the data type conversion * buffer. Also gather data from application to background buffer @@ -2449,7 +2377,7 @@ H5D_chunk_write(H5D_io_info_t *io_info, hsize_t nelmts, #endif n = H5D_select_mgath(buf, chunk_info->mspace, &mem_iter, smine_nelmts, dxpl_cache, tconv_buf/*out*/); - + #ifdef H5S_DEBUG H5_timer_end(&(io_info->stats->stats[1].gath_timer), &timer); io_info->stats->stats[1].gath_nbytes += n * src_type_size; @@ -2464,7 +2392,7 @@ H5D_chunk_write(H5D_io_info_t *io_info, hsize_t nelmts, #endif n = H5D_select_fgath(io_info, chunk_info->fspace, &bkg_iter, smine_nelmts, - bkg_buf/*out*/); + bkg_buf/*out*/); #ifdef H5S_DEBUG H5_timer_end(&(io_info->stats->stats[0].bkg_timer), &timer); @@ -2478,7 +2406,7 @@ H5D_chunk_write(H5D_io_info_t *io_info, hsize_t nelmts, /* * Perform data type conversion. */ - if (H5T_convert(tpath, src_id, dst_id, smine_nelmts, 0, 0, + if (H5T_convert(tpath, src_id, dst_id, smine_nelmts, 0, 0, tconv_buf, bkg_buf, io_info->dxpl_id)<0) HGOTO_ERROR(H5E_DATASET, H5E_CANTINIT, FAIL, "data type conversion failed") @@ -2541,12 +2469,12 @@ done: if(H5S_SELECT_ITER_RELEASE(&bkg_iter)<0) HDONE_ERROR (H5E_DATASET, H5E_CANTFREE, FAIL, "Can't release selection iterator") } /* end if */ - + if (tconv_buf && NULL==dxpl_cache->tconv_buf) H5FL_BLK_FREE(type_conv,tconv_buf); if (bkg_buf && NULL==dxpl_cache->bkgr_buf) H5FL_BLK_FREE(type_conv,bkg_buf); - + /* Release chunk mapping information */ if(H5D_destroy_chunk_map(&fm) < 0) HDONE_ERROR(H5E_DATASET, H5E_CANTRELEASE, FAIL, "can't release chunk mapping") @@ -2573,14 +2501,14 @@ done: * *------------------------------------------------------------------------- */ -static herr_t -H5D_io_assist_mpio(hid_t dxpl_id, H5D_dxpl_cache_t *dxpl_cache, +static herr_t +H5D_io_assist_mpio(hid_t dxpl_id, H5D_dxpl_cache_t *dxpl_cache, hbool_t *xfer_mode_changed) -{ +{ herr_t ret_value = SUCCEED; /*return value */ FUNC_ENTER_NOAPI_NOINIT(H5D_io_assist_mpio) - + /* The following may not handle a collective call correctly * since it does not ensure all processes can handle the write * request according to the MPI collective specification. @@ -2612,7 +2540,7 @@ done: /*------------------------------------------------------------------------- * Function: H5D_io_restore_mpio * - * Purpose: Common logic for restoring MPI transfer mode + * Purpose: Common logic for restoring MPI transfer mode * * Return: Non-negative on success/Negative on failure * @@ -2623,15 +2551,15 @@ done: * *------------------------------------------------------------------------- */ -static herr_t +static herr_t H5D_io_restore_mpio(hid_t dxpl_id) -{ +{ H5P_genplist_t *dx_plist; /* Data transer property list */ H5FD_mpio_xfer_t xfer_mode; /*xfer_mode for this request */ herr_t ret_value = SUCCEED; /*return value */ FUNC_ENTER_NOAPI_NOINIT(H5D_io_restore_mpio) - + /* Get the dataset transfer property list */ if (NULL == (dx_plist = H5I_object(dxpl_id))) HGOTO_ERROR(H5E_ARGS, H5E_BADTYPE, FAIL, "not a dataset creation property list") @@ -2669,7 +2597,7 @@ done: *------------------------------------------------------------------------- */ static herr_t -H5D_create_chunk_map(const H5D_t *dataset, const H5T_t *mem_type, const H5S_t *file_space, +H5D_create_chunk_map(const H5D_t *dataset, const H5T_t *mem_type, const H5S_t *file_space, const H5S_t *mem_space, fm_map *fm) { H5S_t *tmp_mspace=NULL; /* Temporary memory dataspace */ @@ -2686,12 +2614,12 @@ H5D_create_chunk_map(const H5D_t *dataset, const H5T_t *mem_type, const H5S_t *f char bogus; /* "bogus" buffer to pass to selection iterator */ unsigned u; /* Local index variable */ herr_t ret_value = SUCCEED; /* Return value */ - + FUNC_ENTER_NOAPI_NOINIT(H5D_create_chunk_map) /* Get layout for dataset */ fm->layout = &(dataset->shared->layout); - + /* Check if the memory space is scalar & make equivalent memory space */ if((sm_ndims = H5S_GET_EXTENT_NDIMS(mem_space))<0) HGOTO_ERROR (H5E_DATASPACE, H5E_CANTGET, FAIL, "unable to get dimension number") @@ -2716,7 +2644,7 @@ H5D_create_chunk_map(const H5D_t *dataset, const H5T_t *mem_type, const H5S_t *f /* Set the number of dimensions for the memory dataspace */ H5_ASSIGN_OVERFLOW(fm->m_ndims,sm_ndims,int,unsigned); } /* end else */ - + /* Get dim number and dimensionality for each dataspace */ fm->f_ndims=f_ndims=dataset->shared->layout.u.chunk.ndims-1; @@ -2741,7 +2669,7 @@ H5D_create_chunk_map(const H5D_t *dataset, const H5T_t *mem_type, const H5S_t *f /* Round up to the next integer # of chunks, to accomodate partial chunks */ fm->chunks[u] = ((fm->f_dims[u]+dataset->shared->layout.u.chunk.dim[u])-1) / dataset->shared->layout.u.chunk.dim[u]; } /* end for */ - + /* Compute the "down" size of 'chunks' information */ if(H5V_array_down(f_ndims,fm->chunks,fm->down_chunks)<0) HGOTO_ERROR (H5E_INTERNAL, H5E_BADVALUE, FAIL, "can't compute 'down' sizes") @@ -2770,7 +2698,7 @@ H5D_create_chunk_map(const H5D_t *dataset, const H5T_t *mem_type, const H5S_t *f /* Create temporary datatypes for selection iteration */ if((f_tid = H5I_register(H5I_DATATYPE, H5T_copy(dataset->shared->type, H5T_COPY_ALL)))<0) HGOTO_ERROR (H5E_DATATYPE, H5E_CANTREGISTER, FAIL, "unable to register file datatype") - + /* Spaces aren't the same shape, iterate over the memory selection directly */ if(H5S_select_iterate(&bogus, f_tid, file_space, H5D_chunk_file_cb, fm)<0) HGOTO_ERROR (H5E_DATASET, H5E_CANTINIT, FAIL, "unable to create file chunk selections") @@ -2819,7 +2747,7 @@ H5D_create_chunk_map(const H5D_t *dataset, const H5T_t *mem_type, const H5S_t *f /* Make a copy of equivalent memory space */ if((tmp_mspace = H5S_copy(equiv_mspace,TRUE))==NULL) HGOTO_ERROR (H5E_DATASPACE, H5E_CANTCOPY, FAIL, "unable to copy memory space") - + /* De-select the mem space copy */ if(H5S_select_none(tmp_mspace)<0) HGOTO_ERROR (H5E_DATASPACE, H5E_CANTINIT, FAIL, "unable to de-select memory space") @@ -2832,7 +2760,7 @@ H5D_create_chunk_map(const H5D_t *dataset, const H5T_t *mem_type, const H5S_t *f if((f_tid = H5I_register(H5I_DATATYPE, H5T_copy(dataset->shared->type, H5T_COPY_ALL)))<0) HGOTO_ERROR (H5E_DATATYPE, H5E_CANTREGISTER, FAIL, "unable to register file datatype") } /* end if */ - + /* Create selection iterator for memory selection */ if((elmt_size=H5T_get_size(mem_type))==0) HGOTO_ERROR(H5E_DATATYPE, H5E_BADSIZE, FAIL, "datatype size invalid") @@ -2897,7 +2825,7 @@ done: if(H5S_hyper_denormalize_offset(file_space, old_offset)<0) HGOTO_ERROR (H5E_DATASET, H5E_BADSELECT, FAIL, "unable to normalize dataspace by offset") } /* end if */ - + FUNC_LEAVE_NOAPI(ret_value) } /* end H5D_create_chunk_map() */ @@ -2960,7 +2888,7 @@ static herr_t H5D_destroy_chunk_map(const fm_map *fm) { herr_t ret_value = SUCCEED; /*return value */ - + FUNC_ENTER_NOAPI_NOINIT(H5D_destroy_chunk_map) /* Free the chunk info skip list */ @@ -2996,7 +2924,7 @@ done: * *------------------------------------------------------------------------- */ -static herr_t +static herr_t H5D_create_chunk_file_map_hyper(const fm_map *fm) { hssize_t ssel_points; /* Number of elements in file selection */ @@ -3010,7 +2938,7 @@ H5D_create_chunk_file_map_hyper(const fm_map *fm) int curr_dim; /* Current dimension to increment */ unsigned u; /* Local index variable */ herr_t ret_value = SUCCEED; /* Return value */ - + FUNC_ENTER_NOAPI_NOINIT(H5D_create_chunk_file_map_hyper) /* Sanity check */ @@ -3060,7 +2988,7 @@ H5D_create_chunk_file_map_hyper(const fm_map *fm) (void)H5S_close(tmp_fchunk); HGOTO_ERROR(H5E_DATASPACE, H5E_CANTSELECT, FAIL, "can't create chunk selection") } /* end if */ - + /* Resize chunk's dataspace dimensions to size of chunk */ if(H5S_set_extent_real(tmp_fchunk,fm->chunk_dim)<0) { (void)H5S_close(tmp_fchunk); @@ -3173,7 +3101,7 @@ done: * *------------------------------------------------------------------------- */ -static herr_t +static herr_t H5D_create_chunk_mem_map_hyper(const fm_map *fm) { H5SL_node_t *curr_node; /* Current node in skip list */ @@ -3185,7 +3113,7 @@ H5D_create_chunk_mem_map_hyper(const fm_map *fm) hssize_t chunk_adjust[H5O_LAYOUT_NDIMS]; /* Adjustment to make to a particular chunk */ unsigned u; /* Local index variable */ herr_t ret_value = SUCCEED; /* Return value */ - + FUNC_ENTER_NOAPI_NOINIT(H5D_create_chunk_mem_map_hyper) /* Sanity check */ @@ -3292,7 +3220,7 @@ done: * *------------------------------------------------------------------------- */ -static herr_t +static herr_t H5D_chunk_file_cb(void UNUSED *elem, hid_t UNUSED type_id, unsigned ndims, const hsize_t *coords, void *_fm) { fm_map *fm = (fm_map*)_fm; /* File<->memory chunk mapping info */ @@ -3301,7 +3229,7 @@ H5D_chunk_file_cb(void UNUSED *elem, hid_t UNUSED type_id, unsigned ndims, const hsize_t chunk_index; /* Chunk index */ unsigned u; /* Local index variable */ herr_t ret_value = SUCCEED; /* Return value */ - + FUNC_ENTER_NOAPI_NOINIT(H5D_chunk_file_cb) /* Calculate the index of this chunk */ @@ -3408,7 +3336,7 @@ done: *------------------------------------------------------------------------- */ /* ARGSUSED */ -static herr_t +static herr_t H5D_chunk_mem_cb(void UNUSED *elem, hid_t UNUSED type_id, unsigned ndims, const hsize_t *coords, void *_fm) { fm_map *fm = (fm_map*)_fm; /* File<->memory chunk mapping info */ @@ -3416,7 +3344,7 @@ H5D_chunk_mem_cb(void UNUSED *elem, hid_t UNUSED type_id, unsigned ndims, const hsize_t coords_in_mem[H5O_LAYOUT_NDIMS]; /* Coordinates of element in memory */ hsize_t chunk_index; /* Chunk index */ herr_t ret_value = SUCCEED; /* Return value */ - + FUNC_ENTER_NOAPI_NOINIT(H5D_chunk_mem_cb) /* Calculate the index of this chunk */ @@ -3585,22 +3513,21 @@ done: * *------------------------------------------------------------------------- */ -static herr_t +static herr_t H5D_ioinfo_init(H5D_t *dset, const H5D_dxpl_cache_t *dxpl_cache, hid_t dxpl_id, hid_t #if !(defined H5_HAVE_PARALLEL || defined H5S_DEBUG) UNUSED #endif /* H5_HAVE_PARALLEL */ - dp_dxpl_id, H5D_dxpl_cache_t - + dp_dxpl_id, H5D_dxpl_cache_t #if !(defined H5_HAVE_PARALLEL || defined H5S_DEBUG) UNUSED #endif /* H5_HAVE_PARALLEL */ -*dp_dxpl_cache,const H5S_t + *dp_dxpl_cache, const H5S_t #if !(defined H5_HAVE_PARALLEL || defined H5S_DEBUG) UNUSED #endif /* H5_HAVE_PARALLEL */ - *mem_space, const H5S_t + *mem_space, const H5S_t #if !(defined H5_HAVE_PARALLEL || defined H5S_DEBUG) UNUSED #endif /* H5_HAVE_PARALLEL */ @@ -3608,7 +3535,7 @@ H5D_ioinfo_init(H5D_t *dset, const H5D_dxpl_cache_t *dxpl_cache, hid_t dxpl_id, #ifndef H5_HAVE_PARALLEL UNUSED #endif /* H5_HAVE_PARALLEL */ - flags, hbool_t + flags, hbool_t #ifndef H5_HAVE_PARALLEL UNUSED #endif /* H5_HAVE_PARALLEL */ @@ -3662,14 +3589,14 @@ H5D_ioinfo_init(H5D_t *dset, const H5D_dxpl_cache_t *dxpl_cache, hid_t dxpl_id, /* Check if we can use the optimized parallel I/O routines */ if(opt==TRUE) { /* Set the pointers to the MPI-specific routines */ - io_info->ops.read = H5D_mpio_select_read; + io_info->ops.read = H5D_mpio_select_read; io_info->ops.write = H5D_mpio_select_write; io_info->ops_sca.read = H5D_select_read; io_info->ops_sca.write = H5D_select_write; - *use_par_opt_io=TRUE; + /* Indicate that the I/O will use collective */ - } - /* end if */ + *use_par_opt_io=TRUE; + } /* end if */ else { /* Indicate that the I/O will _NOT_ be parallel, use independent IO */ *use_par_opt_io=FALSE; @@ -3694,135 +3621,121 @@ done: FUNC_LEAVE_NOAPI(ret_value) } /* end H5D_ioinfo_init() */ - #ifdef H5_HAVE_PARALLEL - /*------------------------------------------------------------------------- * Function: H5D_mpio_get_mini_chunk * * Purpose: Routine for obtaining minimum number of chunks to cover - hyperslab selection selected by all processors. - * + * hyperslab selection selected by all processors. * * Return: Non-negative on success/Negative on failure * * Programmer: - * * * Modifications: * *------------------------------------------------------------------------- */ - static herr_t H5D_mpio_get_mini_chunk(const H5D_t *dset, - const H5S_t *mem_space, + const H5S_t *mem_space, const H5S_t *file_space, - int *min_chunkf) { - - - hsize_t chunk_dim[H5O_LAYOUT_NDIMS]; /* Chunk dimensions */ - hsize_t startf[H5S_MAX_RANK], /* Selection start bounds */ - endf[H5S_MAX_RANK]; /* Selection end bounds */ - unsigned dim_rankf; /* Number of dimensions of file dataspace */ - int pcheck_hyper,check_hyper, /* Flags for checking if selection is in one chunk */ - tnum_chunkf, /* Number of chunks selection overlaps */ - max_chunkf, /* Maximum number of chunks selection overlaps */ - num_chunks_same; /* Flag indicating whether all processes have the same # of chunks to operate on */ - unsigned dim_chunks; /* Temporary number of chunks in a dimension */ - MPI_Comm comm; /* MPI communicator for file */ - int mpi_rank; /* Rank in MPI communicator */ - int mpi_code; /* MPI return code */ - unsigned u; /* Local index variable */ - herr_t ret_value; - - ret_value = SUCCEED; - FUNC_ENTER_NOAPI_NOINIT(H5D_mpio_get_mini_chunk); - /* Getting MPI communicator and rank */ - if((comm = H5F_mpi_get_comm(dset->ent.file))==MPI_COMM_NULL) - HGOTO_ERROR(H5E_DATASPACE, H5E_CANTGET, FAIL, "can't retrieve MPI communicator") - if((mpi_rank = H5F_mpi_get_rank(dset->ent.file))<0) - HGOTO_ERROR(H5E_DATASPACE, H5E_CANTGET, FAIL, "can't retrieve MPI rank") - - - dim_rankf = H5S_GET_EXTENT_NDIMS(file_space); - - if(H5S_SELECT_BOUNDS(file_space,startf,endf)==FAIL) - HGOTO_ERROR(H5E_DATASPACE, H5E_BADRANGE,FAIL, "invalid check for single selection blocks"); - - for(u=0; u < dset->shared->layout.u.chunk.ndims; u++) - chunk_dim[u] = dset->shared->layout.u.chunk.dim[u]; - - - /* Compute the number of chunks covered by the selection on this process */ - tnum_chunkf = 1; - for (u=0; uent.file))==MPI_COMM_NULL) + HGOTO_ERROR(H5E_DATASPACE, H5E_CANTGET, FAIL, "can't retrieve MPI communicator") + if((mpi_rank = H5F_mpi_get_rank(dset->ent.file))<0) + HGOTO_ERROR(H5E_DATASPACE, H5E_CANTGET, FAIL, "can't retrieve MPI rank") - done: - FUNC_LEAVE_NOAPI(ret_value); - -} + dim_rankf = H5S_GET_EXTENT_NDIMS(file_space); + + /* Get selection bounds */ + if(H5S_SELECT_BOUNDS(file_space,startf,endf)==FAIL) + HGOTO_ERROR(H5E_DATASPACE, H5E_BADRANGE,FAIL, "invalid check for single selection blocks"); + + /* Get chunk dimensions */ + for(u=0; u < dset->shared->layout.u.chunk.ndims; u++) + chunk_dim[u] = dset->shared->layout.u.chunk.dim[u]; + + /* Compute the number of chunks covered by the selection on this process */ + tnum_chunkf = 1; + for (u=0; ushared->layout.type == H5D_CHUNKED) - if(dset->shared->dcpl_cache.pline.nused>0) - HGOTO_DONE(FALSE); /* Perform the independent write operation */ - + if(dset->shared->layout.type == H5D_CHUNKED) + if(dset->shared->dcpl_cache.pline.nused>0) + HGOTO_DONE(FALSE); /* Perform the independent write operation */ done: FUNC_LEAVE_NOAPI(ret_value); - } /* H5D_mpio_opt_possible() */ @@ -219,7 +215,7 @@ H5D_mpio_spaces_xfer(H5D_io_info_t *io_info, size_t elmt_size, else { haddr_t chunk_addr; /* for collective chunk IO */ - assert(io_info->dset->shared->layout.type == H5D_CHUNKED); + assert(io_info->dset->shared->layout.type == H5D_CHUNKED); chunk_addr=H5D_istore_get_addr(io_info,NULL); addr = H5F_BASE_ADDR(io_info->dset->ent.file) + chunk_addr + mpi_file_offset; } @@ -288,8 +284,7 @@ H5D_mpio_select_read(H5D_io_info_t *io_info, FUNC_ENTER_NOAPI_NOFUNC(H5D_mpio_select_read); - - ret_value = H5D_mpio_spaces_xfer(io_info, elmt_size, file_space, + ret_value = H5D_mpio_spaces_xfer(io_info, elmt_size, file_space, mem_space, buf, 0/*read*/); FUNC_LEAVE_NOAPI(ret_value); @@ -321,11 +316,10 @@ H5D_mpio_select_write(H5D_io_info_t *io_info, FUNC_ENTER_NOAPI_NOFUNC(H5D_mpio_select_write); /*OKAY: CAST DISCARDS CONST QUALIFIER*/ - ret_value = H5D_mpio_spaces_xfer(io_info, elmt_size, file_space, + ret_value = H5D_mpio_spaces_xfer(io_info, elmt_size, file_space, mem_space, (void*)buf, 1/*write*/); FUNC_LEAVE_NOAPI(ret_value); } /* end H5D_mpio_spaces_write() */ - - #endif /* H5_HAVE_PARALLEL */ + diff --git a/src/H5Dpkg.h b/src/H5Dpkg.h index 7b2c9e3..3137f53 100644 --- a/src/H5Dpkg.h +++ b/src/H5Dpkg.h @@ -233,12 +233,12 @@ H5_DLL ssize_t H5D_contig_writevv(const H5D_io_info_t *io_info, /* Functions that operate on compact dataset storage */ H5_DLL ssize_t H5D_compact_readvv(const H5D_io_info_t *io_info, - size_t dset_max_nseq, size_t *dset_curr_seq, size_t dset_size_arr[], hsize_t dset_offset_arr[], - size_t mem_max_nseq, size_t *mem_curr_seq, size_t mem_size_arr[], hsize_t mem_offset_arr[], + size_t dset_max_nseq, size_t *dset_curr_seq, size_t dset_size_arr[], hsize_t dset_offset_arr[], + size_t mem_max_nseq, size_t *mem_curr_seq, size_t mem_size_arr[], hsize_t mem_offset_arr[], void *buf); H5_DLL ssize_t H5D_compact_writevv(const H5D_io_info_t *io_info, - size_t dset_max_nseq, size_t *dset_curr_seq, size_t dset_size_arr[], hsize_t dset_offset_arr[], - size_t mem_max_nseq, size_t *mem_curr_seq, size_t mem_size_arr[], hsize_t mem_offset_arr[], + size_t dset_max_nseq, size_t *dset_curr_seq, size_t dset_size_arr[], hsize_t dset_offset_arr[], + size_t mem_max_nseq, size_t *mem_curr_seq, size_t mem_size_arr[], hsize_t mem_offset_arr[], const void *buf); /* Functions that operate on indexed storage */ diff --git a/src/H5Smpio.c b/src/H5Smpio.c index a9b90ee..48f2358 100644 --- a/src/H5Smpio.c +++ b/src/H5Smpio.c @@ -66,7 +66,7 @@ H5S_mpio_span_hyper_type( const H5S_t *space, size_t elmt_size, hsize_t *extra_offset, hbool_t *is_derived_type ); -static herr_t H5S_obtain_datatype(const hsize_t size[], +static herr_t H5S_obtain_datatype(const hsize_t size[], H5S_hyper_span_t* span,MPI_Datatype *span_type, size_t elmt_size,int dimindex); @@ -314,10 +314,10 @@ H5S_mpio_hyper_type( const H5S_t *space, size_t elmt_size, goto empty; } } /* end else */ - + /********************************************************************** Compute array "offset[rank]" which gives the offsets for a multi- - dimensional array with dimensions "d[i].xtent" (i=0,1,...,rank-1). + dimensional array with dimensions "d[i].xtent" (i=0,1,...,rank-1). **********************************************************************/ offset[rank-1] = 1; max_xtent[rank-1] = d[rank-1].xtent; @@ -397,9 +397,9 @@ H5S_mpio_hyper_type( const H5S_t *space, size_t elmt_size, block_length[0] = 1; block_length[1] = 1; block_length[2] = 1; - + displacement[0] = 0; - + old_types[0] = MPI_LB; old_types[1] = outer_type; old_types[2] = MPI_UB; @@ -414,7 +414,7 @@ H5S_mpio_hyper_type( const H5S_t *space, size_t elmt_size, displacement, /* displacements */ old_types, /* old types */ &inner_type); /* new type */ - + MPI_Type_free (&outer_type); if (mpi_code!=MPI_SUCCESS) HMPI_GOTO_ERROR(FAIL, "couldn't resize MPI vector type", mpi_code); @@ -424,7 +424,7 @@ H5S_mpio_hyper_type( const H5S_t *space, size_t elmt_size, } } /* end for */ /*************************** -* End of loop, walking +* End of loop, walking * thru dimensions. ***************************/ @@ -469,7 +469,7 @@ done: /*------------------------------------------------------------------------- * Function: H5S_mpio_span_hyper_type * - * Purpose: Translate an HDF5 irregular hyperslab selection into an + * Purpose: Translate an HDF5 irregular hyperslab selection into an MPI type. * * Return: non-negative on success, negative on failure. @@ -482,9 +482,9 @@ done: * * Programmer: kyang * - */ + */ static herr_t -H5S_mpio_span_hyper_type( const H5S_t *space, +H5S_mpio_span_hyper_type( const H5S_t *space, size_t elmt_size, MPI_Datatype *new_type,/* out: */ size_t *count, @@ -499,26 +499,26 @@ H5S_mpio_span_hyper_type( const H5S_t *space, herr_t ret_value = SUCCEED; MPI_Aint extent,lb; - + FUNC_ENTER_NOAPI_NOINIT_NOFUNC(H5S_mpio_span_hyper_type); /* Check args */ assert (space); - + /* assert(sizeof(MPI_Aint) >= sizeof(elmt_size)); not sure the reason*/ rank = space->extent.rank; - /* size = HDcalloc((size_t)rank,sizeof(hsize_t)); */ + /* size = HDcalloc((size_t)rank,sizeof(hsize_t)); */ if (0==elmt_size) goto empty; size = space->extent.size; - if(size == 0) + if(size == 0) goto empty; - + odown = space->select.sel_info.hslab->span_lst; - if(odown == NULL) + if(odown == NULL) goto empty; ospan = odown->head; if(ospan == NULL) @@ -527,13 +527,13 @@ H5S_mpio_span_hyper_type( const H5S_t *space, /* obtain derived data type */ H5S_obtain_datatype(space->extent.size,ospan,&span_type,elmt_size,rank); MPI_Type_commit(&span_type); - + *new_type = span_type; /* fill in the remaining return values */ *count = 1; *extra_offset = 0; *is_derived_type = 1; - + HGOTO_DONE(SUCCEED); empty: @@ -551,8 +551,8 @@ done: /*------------------------------------------------------------------------- * Function: obtain datatype * - * Purpose: Obtain an MPI derived datatype based on span-tree - implementation + * Purpose: Obtain an MPI derived datatype based on span-tree + implementation * * Return: non-negative on success, negative on failure. * @@ -560,13 +560,13 @@ done: * * Programmer: kyang * - */ -static herr_t H5S_obtain_datatype(const hsize_t size[], + */ +static herr_t H5S_obtain_datatype(const hsize_t size[], H5S_hyper_span_t* span, MPI_Datatype *span_type, size_t elmt_size, - int dimindex) { - + int dimindex) +{ int innercount,outercount; MPI_Datatype bas_type; MPI_Datatype temp_type; @@ -579,7 +579,7 @@ static herr_t H5S_obtain_datatype(const hsize_t size[], H5S_hyper_span_info_t *down; H5S_hyper_span_t *tspan; int mpi_code; - herr_t ret_value; + herr_t ret_value = SUCCEED; #ifdef H5_HAVE_MPI2 MPI_Aint sizeaint,sizedtype; @@ -587,10 +587,10 @@ static herr_t H5S_obtain_datatype(const hsize_t size[], hsize_t total_lowd,total_lowd1; int i; int ret; - + FUNC_ENTER_NOAPI_NOINIT(H5S_obtain_datatype); - ret_value = SUCCEED; assert(span); + inner_type = NULL; down = NULL; tspan = NULL; @@ -625,8 +625,8 @@ static herr_t H5S_obtain_datatype(const hsize_t size[], inner_type = (MPI_Datatype *)HDcalloc((size_t)outercount,sizeof(int)); #endif - - + + tspan = span; outercount = 0; @@ -650,7 +650,7 @@ static herr_t H5S_obtain_datatype(const hsize_t size[], if(MPI_SUCCESS != (mpi_code = MPI_Type_hindexed(outercount,blocklen, disp,bas_type,span_type))) HMPI_GOTO_ERROR(FAIL, "MPI_Type_hindexed failed", mpi_code); - + } else {/* dimindex is the rank of the dimension */ @@ -664,9 +664,9 @@ static herr_t H5S_obtain_datatype(const hsize_t size[], for ( i = dimindex-1; i > 1; i--) total_lowd1 = total_lowd1 * size[i]; - + while(tspan){ - + /* Displacement should be in byte and should have dimension information */ /* First using MPI Type vector to build derived data type for this span only */ /* Need to calculate the disp in byte for this dimension. */ @@ -676,7 +676,7 @@ static herr_t H5S_obtain_datatype(const hsize_t size[], blocklen[outercount] = 1; /* generating inner derived datatype by using MPI_Type_hvector */ - if(FAIL == H5S_obtain_datatype(size,tspan->down->head,&temp_type,elmt_size,dimindex-1)) + if(FAIL == H5S_obtain_datatype(size,tspan->down->head,&temp_type,elmt_size,dimindex-1)) HGOTO_ERROR(H5E_DATASPACE, H5E_BADTYPE, FAIL,"couldn't obtain MPI derived data type"); if(MPI_SUCCESS != (mpi_code = MPI_Type_commit(&temp_type))) @@ -698,20 +698,18 @@ static herr_t H5S_obtain_datatype(const hsize_t size[], tspan = tspan->next; } - + /* building the whole vector datatype */ - - if(MPI_SUCCESS != (mpi_code = + if(MPI_SUCCESS != (mpi_code = MPI_Type_struct(outercount,blocklen,disp,inner_type,span_type))) HMPI_GOTO_ERROR(FAIL, "MPI_Type_struct failed", mpi_code); } - + if(inner_type != NULL){ if(down != NULL) { - for(i=0;i