summaryrefslogtreecommitdiffstats
path: root/perform
diff options
context:
space:
mode:
authorAlbert Cheng <acheng@hdfgroup.org>2002-06-01 23:01:48 (GMT)
committerAlbert Cheng <acheng@hdfgroup.org>2002-06-01 23:01:48 (GMT)
commit3ccff1357fe7708a56528ac64a79998f7c2a79eb (patch)
treeee818d4f07be52ecd45aea6693bb573c4d67c5ac /perform
parent44a72a6777520034f87f712697ce05de5929f288 (diff)
downloadhdf5-3ccff1357fe7708a56528ac64a79998f7c2a79eb.zip
hdf5-3ccff1357fe7708a56528ac64a79998f7c2a79eb.tar.gz
hdf5-3ccff1357fe7708a56528ac64a79998f7c2a79eb.tar.bz2
[svn-r5503] Purpose:
Features Description: Added a few macros to print rank and size of MPI_COMM_WORLD in a unifed style. Platforms tested: modi4, eirene, burrwhite
Diffstat (limited to 'perform')
-rw-r--r--perform/pio_engine.c483
-rw-r--r--perform/pio_perf.h7
2 files changed, 370 insertions, 120 deletions
diff --git a/perform/pio_engine.c b/perform/pio_engine.c
index 753b873..928129c 100644
--- a/perform/pio_engine.c
+++ b/perform/pio_engine.c
@@ -34,7 +34,7 @@
/* Macro definitions */
/* sizes of various items. these sizes won't change during program execution */
-#define ELMT_SIZE ((int)sizeof(int)) /* we're doing ints */
+#define ELMT_SIZE (sizeof(int)) /* we're doing ints */
#define GOTOERROR(errcode) { ret_code = errcode; goto done; }
#define GOTODONE { goto done; }
@@ -216,13 +216,6 @@ do_pio(parameters param)
}
}
- if (blk_size < 0 ) {
- HDfprintf(stderr,
- "interleaved I/O block size must be >= 0 (%Hd)\n",
- (long_long)blk_size);
- GOTOERROR(FAIL);
- }
-
/* Should only need blk_size <= buf_size. */
/* More restrictive condition for easier implementation for now. */
if (blk_size > 0 && (buf_size % blk_size)){
@@ -455,23 +448,35 @@ do_write(results *res, file_descr *fd, parameters *parms, long ndsets,
MPI_Offset mpi_offset;
MPI_Status mpi_status;
long ndset;
- off_t nelmts_written;
- size_t nelmts_towrite;
+ off_t nelmts_xfer;
+ size_t nelmts_toxfer;
char dname[64];
off_t dset_offset; /*dataset offset in a file */
off_t file_offset; /*file offset of the next transfer */
off_t dset_size; /*one dataset size in bytes */
- size_t nelmts_in_buf;
+ size_t nelmts_in_buf; /*how many element the buffer holds */
+ size_t nelmts_in_blk; /*how many element a block holds */
off_t elmts_begin; /*first elmt this process transfer */
off_t elmts_count; /*number of elmts this process transfer */
hid_t dcpl = -1; /* Dataset creation property list */
/* HDF5 variables */
- herr_t hrc; /*HDF5 return code */
- hsize_t h5dims[1]; /*dataset dim sizes */
- hid_t h5dset_space_id = -1; /*dataset space ID */
- hid_t h5mem_space_id = -1; /*memory dataspace ID */
- hid_t h5ds_id = -1; /*dataset handle */
+ herr_t hrc; /*HDF5 return code */
+ hsize_t h5dims[1]; /*dataset dim sizes */
+ hid_t h5dset_space_id = -1; /*dataset space ID */
+ hid_t h5mem_space_id = -1; /*memory dataspace ID */
+ hid_t h5ds_id = -1; /*dataset handle */
+ hsize_t h5mem_block[1]; /*memory space selection */
+ hsize_t h5mem_stride[1];
+ hsize_t h5mem_count[1];
+ hssize_t h5mem_start[1];
+#if 0
+ /* for future implementation */
+ hsize_t h5dset_block[1]; /*dset space selection */
+ hsize_t h5dset_stride[1];
+ hsize_t h5dset_count[1];
+ hssize_t h5dset_start[1];
+#endif
/* calculate dataset parameters. data type is always native C int */
dset_size = nelmts * ELMT_SIZE;
@@ -562,33 +567,103 @@ do_write(results *res, file_descr *fd, parameters *parms, long ndsets,
break;
}
- /* Calculate the first element and how many elements this process
- * transfer. First calculate the beginning element of this process
- * and the next process. Count of elements is the difference between
- * these two beginnings. This way, it avoids any rounding errors.
- */
- elmts_begin = (long)(((double)nelmts)/pio_mpi_nprocs_g*pio_mpi_rank_g);
-
- if (pio_mpi_rank_g < (pio_mpi_nprocs_g - 1))
- elmts_count = (long)((((double)nelmts) / pio_mpi_nprocs_g * (pio_mpi_rank_g + 1))
- - (double)elmts_begin);
- else
- /* last process. Take whatever are left */
- elmts_count = nelmts - elmts_begin;
-
- nelmts_written = 0 ;
+ /* There are two kinds of transfer patterns, contiguous and interleaved.
+ * Let 0,1,2,...,n be data accessed by process 0,1,2,...,n
+ * where n is rank of the last process.
+ * In contiguous pattern, data are accessed as
+ * 000...111...222...nnn...
+ * In interleaved pattern, data are accessed as
+ * 012...n012...n...
+ * These are all in the scope of one dataset.
+ */
+ /* Calculate the total number of elements (elmts_count) to be
+ * transferred by this process. It may be different for different
+ * transfer pattern due to rounding to integral values.
+ */
+ if (blk_size==0){
+ /* Contiguous Pattern:
+ * Calculate the beginning element of this process and the next.
+ * elmts_count is the difference between these two beginnings.
+ * This way, it eliminates any rounding errors.
+ */
+ elmts_begin = (off_t)(((double)nelmts)/pio_mpi_nprocs_g*pio_mpi_rank_g);
+
+ /* Do not cast elmt_begin to other types, especially non-integral
+ * types, else it may introduce rounding discrepency. */
+ if (pio_mpi_rank_g < (pio_mpi_nprocs_g - 1))
+ elmts_count = (off_t)(((double)nelmts) / pio_mpi_nprocs_g * (pio_mpi_rank_g + 1))
+ - elmts_begin;
+ else
+ /* last process. Take whatever are left */
+ elmts_count = nelmts - elmts_begin;
+ }else{
+ /* Interleaved Pattern:
+ * Each process takes blk_size of elements, starting with the first
+ * process. So, the last process may have fewer or even none.
+ * Calculate the beginning element of this process and the next.
+ * The elmnts_begin here marks only the beginning of the first
+ * block accessed by this process.
+ */
+ /* Algorithm:
+ * First allocate equal blocks per process, i.e. one block each
+ * process for every block_size*nprocs.
+ * If there is remaining unallocated, give a block each to process
+ * starting at proc 0. The last process may get a partial block.
+ */
+ off_t remain_nelmts, remain_begin; /* unallocated remaining*/
+
+ nelmts_in_blk = blk_size/ELMT_SIZE;
+ elmts_begin = (off_t)(nelmts_in_blk*pio_mpi_rank_g);
+
+ /* must use integer calculation next */
+ /* allocate equal blocks per process */
+ elmts_count = (nelmts / (off_t)(nelmts_in_blk*pio_mpi_nprocs_g)) *
+ (off_t)nelmts_in_blk;
+ remain_nelmts = nelmts % ((off_t)(nelmts_in_blk*pio_mpi_nprocs_g));
+
+ /* allocate any remaining */
+ remain_begin = (off_t)(nelmts_in_blk*pio_mpi_rank_g);
+ if (remain_nelmts > remain_begin){
+ /* it gets something */
+ if (remain_nelmts > (remain_begin+(off_t)nelmts_in_blk)){
+ /* one full block */
+ elmts_count += nelmts_in_blk;
+ }else{
+ /* only a partial block */
+ elmts_count += remain_nelmts - remain_begin;
+ }
+ }
+ }
+ /* debug */
+ if (pio_debug_level >= 4) {
+ HDprint_rank(output);
+ HDfprintf(output, "Debug(do_write): "
+ "nelmts_in_blk=%Hd, elmts_begin=%Hd, elmts_count=%Hd\n",
+ (long_long)nelmts_in_blk, (long_long)elmts_begin,
+ (long_long)elmts_count);
+ }
+
+
+ /* The task is to transfer elmts_count elements, starting at
+ * elmts_begin position, using transfer buffer of buf_size bytes.
+ * If blk_size > 0, select blk_size at a time, in round robin
+ * fashion, according to number of process. Otherwise, select
+ * all elmt_count in contiguous.
+ */
+ nelmts_xfer = 0 ;
/* Start "raw data" write timer */
set_time(res->timers, HDF5_RAW_WRITE_FIXED_DIMS, START);
- while (nelmts_written < elmts_count){
- nelmts_towrite = elmts_count - nelmts_written;
-
- if ((elmts_count - nelmts_written) >= nelmts_in_buf) {
- nelmts_towrite = nelmts_in_buf;
+ while (nelmts_xfer < elmts_count){
+ /* transfer one buffer of data each round */
+ /* Note: because size_t is unsigned, avoid expressions that */
+ /* can be negative. */
+ if ((nelmts_xfer + (off_t)nelmts_in_buf) <= elmts_count) {
+ nelmts_toxfer = nelmts_in_buf;
} else {
- /* last write of a partial buffer */
- nelmts_towrite = elmts_count - nelmts_written;
+ /* last transfer of a partial buffer */
+ nelmts_toxfer = elmts_count - nelmts_xfer;
}
#ifdef AKCDEBUG
@@ -597,8 +672,8 @@ do_write(results *res, file_descr *fd, parameters *parms, long ndsets,
int *intptr = (int *)buffer;
register int i;
- for (i = 0; i < nelmts_towrite; ++i)
- *intptr++ = nelmts_towrite + i;
+ for (i = 0; i < nelmts_toxfer; ++i)
+ *intptr++ = nelmts_toxfer + i;
}
#endif
@@ -606,43 +681,87 @@ do_write(results *res, file_descr *fd, parameters *parms, long ndsets,
/* Calculate offset of write within a dataset/file */
switch (parms->io_type) {
case POSIXIO:
- /* need to (off_t) the elmnts_begin expression because they */
- /* may be of smaller sized integer types */
- file_offset = dset_offset + (off_t)(elmts_begin + nelmts_written)*ELMT_SIZE;
-
- /* only care if seek returns error */
- rc = POSIXSEEK(fd->posixfd, file_offset) < 0 ? -1 : 0;
- VRFY((rc==0), "POSIXSEEK");
- /* check if all bytes are written */
- rc = ((nelmts_towrite*ELMT_SIZE) ==
- POSIXWRITE(fd->posixfd, buffer, nelmts_towrite*ELMT_SIZE));
- VRFY((rc != 0), "POSIXWRITE");
+ if (blk_size==0){
+ /* Contiguous pattern */
+ /* need to (off_t) the elmnts_begin expression because they */
+ /* may be of smaller sized integer types */
+ file_offset = dset_offset + (off_t)(elmts_begin + nelmts_xfer)*ELMT_SIZE;
+
+ /* only care if seek returns error */
+ rc = POSIXSEEK(fd->posixfd, file_offset) < 0 ? -1 : 0;
+ VRFY((rc==0), "POSIXSEEK");
+ /* check if all bytes are transferred */
+ rc = ((ssize_t)(nelmts_toxfer*ELMT_SIZE) ==
+ POSIXWRITE(fd->posixfd, buffer, nelmts_toxfer*ELMT_SIZE));
+ VRFY((rc != 0), "POSIXWRITE");
+ }else{
+ /* interleaved access pattern */
+ char *buf_p=buffer;
+ size_t xferred=0;
+ size_t toxfer=0;
+
+ file_offset = dset_offset + (off_t)(elmts_begin + nelmts_xfer)*ELMT_SIZE;
+ if (pio_debug_level >= 4) {
+HDprint_rank(output);
+HDfprintf(output,
+"Debug(do_write): "
+"nelmts_toxfer=%Hd, nelmts_xfer=%Hd\n"
+,
+(long_long)nelmts_toxfer, (long_long)nelmts_xfer);
+}
+ while (xferred < nelmts_toxfer){
+ if ((nelmts_toxfer - xferred) >= nelmts_in_blk)
+ toxfer = nelmts_in_blk;
+ else
+ toxfer = nelmts_toxfer - xferred;
+ /* Skip offset over blocks of other processes */
+ file_offset = dset_offset +
+ (off_t)(elmts_begin + (nelmts_xfer+xferred)*pio_mpi_nprocs_g)*ELMT_SIZE;
+ if (pio_debug_level >= 4) {
+HDprint_rank(output);
+HDfprintf(output,
+"Debug(do_write): "
+"nelmts_toxfer=%Hd, nelmts_xfer=%Hd"
+", toxfer=%Hd, xferred=%Hd"
+", file_offset=%Hd"
+"\n",
+(long_long)nelmts_toxfer, (long_long)nelmts_xfer,
+(long_long)toxfer, (long_long)xferred,
+(long_long)file_offset);
+}
+ /* only care if seek returns error */
+ rc = POSIXSEEK(fd->posixfd, file_offset) < 0 ? -1 : 0;
+ VRFY((rc==0), "POSIXSEEK");
+ /* check if all bytes are written */
+ rc = ((ssize_t)(toxfer*ELMT_SIZE) ==
+ POSIXWRITE(fd->posixfd, buf_p, toxfer*ELMT_SIZE));
+ VRFY((rc != 0), "POSIXWRITE");
+ xferred += toxfer;
+ }
+ }
break;
case MPIO:
- mpi_offset = dset_offset + (elmts_begin + nelmts_written)*ELMT_SIZE;
+ mpi_offset = dset_offset + (elmts_begin + nelmts_xfer)*ELMT_SIZE;
mrc = MPI_File_write_at(fd->mpifd, mpi_offset, buffer,
- (int)(nelmts_towrite*ELMT_SIZE), MPI_CHAR,
+ (int)(nelmts_toxfer*ELMT_SIZE), MPI_CHAR,
&mpi_status);
VRFY((mrc==MPI_SUCCESS), "MPIO_WRITE");
break;
case PHDF5:
/*set up the dset space id to select the segment to process */
{
- hsize_t block[1], stride[1], count[1];
- hssize_t start[1];
-
- start[0] = elmts_begin + nelmts_written;
- stride[0] = block[0] = nelmts_towrite;
- count[0] = 1;
+ h5mem_start[0] = elmts_begin + nelmts_xfer;
+ h5mem_stride[0] = h5mem_block[0] = nelmts_toxfer;
+ h5mem_count[0] = 1;
hrc = H5Sselect_hyperslab(h5dset_space_id, H5S_SELECT_SET,
- start, stride, count, block);
+ h5mem_start, h5mem_stride, h5mem_count, h5mem_block);
VRFY((hrc >= 0), "H5Sset_hyperslab");
/*setup the memory space id too. Only start is different */
- start[0] = 0;
+ h5mem_start[0] = 0;
hrc = H5Sselect_hyperslab(h5mem_space_id, H5S_SELECT_SET,
- start, stride, count, block);
+ h5mem_start, h5mem_stride, h5mem_count, h5mem_block);
VRFY((hrc >= 0), "H5Sset_hyperslab");
}
@@ -653,7 +772,7 @@ do_write(results *res, file_descr *fd, parameters *parms, long ndsets,
break;
}
- nelmts_written += nelmts_towrite;
+ nelmts_xfer += nelmts_toxfer;
}
/* Stop "raw data" write timer */
@@ -716,22 +835,34 @@ do_read(results *res, file_descr *fd, parameters *parms, long ndsets,
MPI_Offset mpi_offset;
MPI_Status mpi_status;
long ndset;
- size_t nelmts_toread;
- off_t nelmts_read;
+ off_t nelmts_xfer;
+ size_t nelmts_toxfer;
char dname[64];
off_t dset_offset; /*dataset offset in a file */
off_t file_offset; /*file offset of the next transfer */
off_t dset_size; /*one dataset size in bytes */
- size_t nelmts_in_buf;
+ size_t nelmts_in_buf; /*how many element the buffer holds */
+ size_t nelmts_in_blk; /*how many element a block holds */
off_t elmts_begin; /*first elmt this process transfer */
off_t elmts_count; /*number of elmts this process transfer */
/* HDF5 variables */
- herr_t hrc; /*HDF5 return code */
- hsize_t h5dims[1]; /*dataset dim sizes */
+ herr_t hrc; /*HDF5 return code */
+ hsize_t h5dims[1]; /*dataset dim sizes */
hid_t h5dset_space_id = -1; /*dataset space ID */
hid_t h5mem_space_id = -1; /*memory dataspace ID */
- hid_t h5ds_id = -1; /*dataset handle */
+ hid_t h5ds_id = -1; /*dataset handle */
+ hsize_t h5mem_block[1]; /*memory space selection */
+ hsize_t h5mem_stride[1];
+ hsize_t h5mem_count[1];
+ hssize_t h5mem_start[1];
+#if 0
+ /* for future implementation */
+ hsize_t h5dset_block[1]; /*dset space selection */
+ hsize_t h5dset_stride[1];
+ hsize_t h5dset_count[1];
+ hssize_t h5dset_start[1];
+#endif
/* calculate dataset parameters. data type is always native C int */
dset_size = nelmts * ELMT_SIZE;
@@ -784,57 +915,174 @@ do_read(results *res, file_descr *fd, parameters *parms, long ndsets,
break;
}
- /*
- * Calculate the first element and how many elements this process
- * transfer. First calculate the beginning element of this process
- * and the next process. Count of elements is the difference between
- * these two beginnings. This way, it avoids any rounding errors.
- */
- elmts_begin = (long)(((double)nelmts)/pio_mpi_nprocs_g*pio_mpi_rank_g);
-
- if (pio_mpi_rank_g < (pio_mpi_nprocs_g - 1))
- elmts_count = (long)((((double)nelmts) / pio_mpi_nprocs_g * (pio_mpi_rank_g + 1)) -
- (double)elmts_begin);
- else
- /* last process. Take whatever are left */
- elmts_count = nelmts - elmts_begin;
-
- nelmts_read = 0 ;
+ /* There are two kinds of transfer patterns, contiguous and interleaved.
+ * Let 0,1,2,...,n be data accessed by process 0,1,2,...,n
+ * where n is rank of the last process.
+ * In contiguous pattern, data are accessed as
+ * 000...111...222...nnn...
+ * In interleaved pattern, data are accessed as
+ * 012...n012...n...
+ * These are all in the scope of one dataset.
+ */
+ /* Calculate the total number of elements (elmts_count) to be
+ * transferred by this process. It may be different for different
+ * transfer pattern due to rounding to integral values.
+ */
+ if (blk_size==0){
+ /* Contiguous Pattern:
+ * Calculate the beginning element of this process and the next.
+ * elmts_count is the difference between these two beginnings.
+ * This way, it eliminates any rounding errors.
+ */
+ elmts_begin = (off_t)(((double)nelmts)/pio_mpi_nprocs_g*pio_mpi_rank_g);
+
+ /* Do not cast elmt_begin to other types, especially non-integral
+ * types, else it may introduce rounding discrepency. */
+ if (pio_mpi_rank_g < (pio_mpi_nprocs_g - 1))
+ elmts_count = (off_t)(((double)nelmts) / pio_mpi_nprocs_g * (pio_mpi_rank_g + 1))
+ - elmts_begin;
+ else
+ /* last process. Take whatever are left */
+ elmts_count = nelmts - elmts_begin;
+ }else{
+ /* Interleaved Pattern:
+ * Each process takes blk_size of elements, starting with the first
+ * process. So, the last process may have fewer or even none.
+ * Calculate the beginning element of this process and the next.
+ * The elmnts_begin here marks only the beginning of the first
+ * block accessed by this process.
+ */
+ /* Algorithm:
+ * First allocate equal blocks per process, i.e. one block each
+ * process for every block_size*nprocs.
+ * If there is remaining unallocated, give a block each to process
+ * starting at proc 0. The last process may get a partial block.
+ */
+ off_t remain_nelmts, remain_begin; /* unallocated remaining*/
+
+ nelmts_in_blk = blk_size/ELMT_SIZE;
+ elmts_begin = (off_t)(nelmts_in_blk*pio_mpi_rank_g);
+
+ /* must use integer calculation next */
+ /* allocate equal blocks per process */
+ elmts_count = (nelmts / (off_t)(nelmts_in_blk*pio_mpi_nprocs_g)) *
+ (off_t)nelmts_in_blk;
+ remain_nelmts = nelmts % ((off_t)(nelmts_in_blk*pio_mpi_nprocs_g));
+
+ /* allocate any remaining */
+ remain_begin = (off_t)(nelmts_in_blk*pio_mpi_rank_g);
+ if (remain_nelmts > remain_begin){
+ /* it gets something */
+ if (remain_nelmts > (remain_begin+(off_t)nelmts_in_blk)){
+ /* one full block */
+ elmts_count += nelmts_in_blk;
+ }else{
+ /* only a partial block */
+ elmts_count += remain_nelmts - remain_begin;
+ }
+ }
+ }
+ /* debug */
+ if (pio_debug_level >= 4) {
+ HDprint_rank(output);
+ HDfprintf(output, "Debug(do_read): "
+ "nelmts_in_blk=%Hd, elmts_begin=%Hd, elmts_count=%Hd\n",
+ (long_long)nelmts_in_blk, (long_long)elmts_begin,
+ (long_long)elmts_count);
+ }
+
+
+ /* The task is to transfer elmts_count elements, starting at
+ * elmts_begin position, using transfer buffer of buf_size bytes.
+ * If blk_size > 0, select blk_size at a time, in round robin
+ * fashion, according to number of process. Otherwise, select
+ * all elmt_count in contiguous.
+ */
+ nelmts_xfer = 0 ;
/* Start "raw data" read timer */
set_time(res->timers, HDF5_RAW_READ_FIXED_DIMS, START);
- while (nelmts_read < elmts_count){
- nelmts_toread = elmts_count - nelmts_read;
-
- if ((elmts_count - nelmts_read) >= nelmts_in_buf)
- nelmts_toread = nelmts_in_buf;
- else
- /* last read of a partial buffer */
- nelmts_toread = elmts_count - nelmts_read;
+ while (nelmts_xfer < elmts_count){
+ /* transfer one buffer of data each round */
+ /* Note: because size_t is unsigned, avoid expressions that */
+ /* can be negative. */
+ if ((nelmts_xfer + (off_t)nelmts_in_buf) <= elmts_count) {
+ nelmts_toxfer = nelmts_in_buf;
+ } else {
+ /* last transfer of a partial buffer */
+ nelmts_toxfer = elmts_count - nelmts_xfer;
+ }
/* read */
/* Calculate offset of read within a dataset/file */
switch (parms->io_type){
case POSIXIO:
- /* need to (off_t) the elmnts_begin expression because they */
- /* may be of smaller sized integer types */
- file_offset = dset_offset + (off_t)(elmts_begin + nelmts_read)*ELMT_SIZE;
-
- /* only care if seek returns error */
- rc = POSIXSEEK(fd->posixfd, file_offset) < 0 ? -1 : 0;
- VRFY((rc==0), "POSIXSEEK");
- /* check if all bytes are read */
- rc = ((nelmts_toread*ELMT_SIZE) ==
- POSIXREAD(fd->posixfd, buffer, nelmts_toread*ELMT_SIZE));
- VRFY((rc != 0), "POSIXREAD");
+ if (blk_size==0){
+ /* Contiguous pattern */
+ /* need to (off_t) the elmnts_begin expression because they */
+ /* may be of smaller sized integer types */
+ file_offset = dset_offset + (off_t)(elmts_begin + nelmts_xfer)*ELMT_SIZE;
+
+ /* only care if seek returns error */
+ rc = POSIXSEEK(fd->posixfd, file_offset) < 0 ? -1 : 0;
+ VRFY((rc==0), "POSIXSEEK");
+ /* check if all bytes are transferred */
+ rc = ((ssize_t)(nelmts_toxfer*ELMT_SIZE) ==
+ POSIXREAD(fd->posixfd, buffer, nelmts_toxfer*ELMT_SIZE));
+ VRFY((rc != 0), "POSIXREAD");
+ }else{
+ /* interleaved access pattern */
+ char *buf_p=buffer;
+ size_t xferred=0;
+ size_t toxfer=0;
+
+ file_offset = dset_offset + (off_t)(elmts_begin + nelmts_xfer)*ELMT_SIZE;
+ if (pio_debug_level >= 4) {
+HDprint_rank(output);
+HDfprintf(output,
+"Debug(do_read): "
+"nelmts_toxfer=%Hd, nelmts_xfer=%Hd\n"
+,
+(long_long)nelmts_toxfer, (long_long)nelmts_xfer);
+}
+ while (xferred < nelmts_toxfer){
+ if ((nelmts_toxfer - xferred) >= nelmts_in_blk)
+ toxfer = nelmts_in_blk;
+ else
+ toxfer = nelmts_toxfer - xferred;
+ /* Skip offset over blocks of other processes */
+ file_offset = dset_offset +
+ (off_t)(elmts_begin + (nelmts_xfer+xferred)*pio_mpi_nprocs_g)*ELMT_SIZE;
+ if (pio_debug_level >= 4) {
+HDprint_rank(output);
+HDfprintf(output,
+"Debug(do_read):"
+"nelmts_toxfer=%Hd, nelmts_xfer=%Hd"
+", toxfer=%Hd, xferred=%Hd"
+", file_offset=%Hd"
+"\n",
+(long_long)nelmts_toxfer, (long_long)nelmts_xfer,
+(long_long)toxfer, (long_long)xferred,
+(long_long)file_offset);
+}
+ /* only care if seek returns error */
+ rc = POSIXSEEK(fd->posixfd, file_offset) < 0 ? -1 : 0;
+ VRFY((rc==0), "POSIXSEEK");
+ /* check if all bytes are transferred */
+ rc = ((ssize_t)(toxfer*ELMT_SIZE) ==
+ POSIXREAD(fd->posixfd, buf_p, toxfer*ELMT_SIZE));
+ VRFY((rc != 0), "POSIXREAD");
+ xferred += toxfer;
+ }
+ }
break;
case MPIO:
- mpi_offset = dset_offset + (elmts_begin + nelmts_read)*ELMT_SIZE;
+ mpi_offset = dset_offset + (elmts_begin + nelmts_xfer)*ELMT_SIZE;
mrc = MPI_File_read_at(fd->mpifd, mpi_offset, buffer,
- (int)(nelmts_toread*ELMT_SIZE), MPI_CHAR,
+ (int)(nelmts_toxfer*ELMT_SIZE), MPI_CHAR,
&mpi_status);
VRFY((mrc==MPI_SUCCESS), "MPIO_read");
break;
@@ -842,20 +1090,17 @@ do_read(results *res, file_descr *fd, parameters *parms, long ndsets,
case PHDF5:
/*set up the dset space id to select the segment to process */
{
- hsize_t block[1], stride[1], count[1];
- hssize_t start[1];
-
- start[0] = elmts_begin + nelmts_read;
- stride[0] = block[0] = nelmts_toread;
- count[0] = 1;
+ h5mem_start[0] = elmts_begin + nelmts_xfer;
+ h5mem_stride[0] = h5mem_block[0] = nelmts_toxfer;
+ h5mem_count[0] = 1;
hrc = H5Sselect_hyperslab(h5dset_space_id, H5S_SELECT_SET,
- start, stride, count, block);
+ h5mem_start, h5mem_stride, h5mem_count, h5mem_block);
VRFY((hrc >= 0), "H5Sset_hyperslab");
/*setup the memory space id too. Only start is different */
- start[0] = 0;
+ h5mem_start[0] = 0;
hrc = H5Sselect_hyperslab(h5mem_space_id, H5S_SELECT_SET,
- start, stride, count, block);
+ h5mem_start, h5mem_stride, h5mem_count, h5mem_block);
VRFY((hrc >= 0), "H5Sset_hyperslab");
}
@@ -872,13 +1117,14 @@ do_read(results *res, file_descr *fd, parameters *parms, long ndsets,
int *intptr = (int *)buffer;
register int i;
- for (i = 0; i < nelmts_towrite; ++i)
+ for (i = 0; i < nelmts_toxfer; ++i)
/* TO BE IMPLEMENTED */
+ #error "NOT IMPLEMENTED YET"
;
}
#endif
- nelmts_read += nelmts_toread;
+ nelmts_xfer += nelmts_toxfer;
}
/* Stop "raw data" read timer */
@@ -1208,9 +1454,6 @@ stop_data_shipping(int handle)
#endif /* H5_HAVE_GPFS */
-#ifndef TIME_MPI
-#define TIME_MPI
-#endif
#ifdef TIME_MPI
/* instrument the MPI_File_wrirte_xxx and read_xxx calls to measure
* pure time spent in MPI_File code.
diff --git a/perform/pio_perf.h b/perform/pio_perf.h
index dd00ce9..89713c1 100644
--- a/perform/pio_perf.h
+++ b/perform/pio_perf.h
@@ -66,6 +66,13 @@ extern int pio_debug_level; /* The debug level:
* 4 - Even More Debugging (timer stuff)
*/
+#define HDprint_rank(f) /* print rank in MPI_COMM_WORLD */ \
+ HDfprintf(f, "%d: ", comm_world_rank_g);
+#define HDprint_size(f) /* print size of MPI_COMM_WORLD */ \
+ HDfprintf(f, "%d", comm_world_nprocs_g);
+#define HDprint_rank_size(f) /* print rank/size of MPI_COMM_WORLD */ \
+ HDfprintf(f, "%d/%d: ", comm_world_rank_g, comm_world_nprocs_g);
+
#ifdef __cplusplus
extern "C" {
#endif /* __cplusplus */