ports/openmpi
ports/openmpi
1
1
Форкнуть 0
Код Релизы Активность
openmpi/ompi/mca/common/ompio/common_ompio_file_open.c
Edgar Gabriel 2da601a350 common/ompio: fix an ordering problem during file_open 
the sharedfp component has to be selected and opened before
we set the default file view during file_open. Otherwise
there is a sperious error message from the sharefp_file_seek
operation that is called during the file_set_view.

Fixes Issue #5560

Signed-off-by: Edgar Gabriel <egabriel@central.uh.edu>
2018-08-20 10:23:32 -05:00

681 строка
21 KiB
C
Исходник Ответственный История

/*
* Copyright (c) 2004-2005 The Trustees of Indiana University and Indiana
* University Research and Technology
* Corporation. All rights reserved.
* Copyright (c) 2004-2017 The University of Tennessee and The University
* of Tennessee Research Foundation. All rights
* reserved.
* Copyright (c) 2004-2005 High Performance Computing Center Stuttgart,
* University of Stuttgart. All rights reserved.
* Copyright (c) 2004-2005 The Regents of the University of California.
* All rights reserved.
* Copyright (c) 2008-2018 University of Houston. All rights reserved.
* Copyright (c) 2015-2018 Research Organization for Information Science
* and Technology (RIST). All rights reserved.
* Copyright (c) 2016 Cisco Systems, Inc. All rights reserved.
* Copyright (c) 2017 IBM Corporation. All rights reserved.
* Copyright (c) 2018 DataDirect Networks. All rights reserved.
* $COPYRIGHT$
*
* Additional copyrights may follow
*
* $HEADER$
*/
#include "ompi_config.h"
#include "ompi/communicator/communicator.h"
#include "ompi/info/info.h"
#include "ompi/file/file.h"
#include "ompi/mca/io/base/base.h"
#include "ompi/mca/fs/fs.h"
#include "ompi/mca/fs/base/base.h"
#include "ompi/mca/fcoll/fcoll.h"
#include "ompi/mca/fcoll/base/base.h"
#include "ompi/mca/fbtl/fbtl.h"
#include "ompi/mca/fbtl/base/base.h"
#include "ompi/mca/sharedfp/sharedfp.h"
#include "ompi/mca/sharedfp/base/base.h"
#include <unistd.h>
#include <math.h>
#include "common_ompio.h"
#include "ompi/mca/topo/topo.h"
static mca_common_ompio_generate_current_file_view_fn_t generate_current_file_view_fn;
static mca_common_ompio_get_mca_parameter_value_fn_t get_mca_parameter_value_fn;
int mca_common_ompio_file_open (ompi_communicator_t *comm,
const char *filename,
int amode,
opal_info_t *info,
ompio_file_t *ompio_fh, bool use_sharedfp)
{
int ret = OMPI_SUCCESS;
int remote_arch;
ompio_fh->f_iov_type = MPI_DATATYPE_NULL;
ompio_fh->f_comm = MPI_COMM_NULL;
if ( ((amode&MPI_MODE_RDONLY)?1:0) + ((amode&MPI_MODE_RDWR)?1:0) +
((amode&MPI_MODE_WRONLY)?1:0) != 1 ) {
return MPI_ERR_AMODE;
}
if ((amode & MPI_MODE_RDONLY) &&
((amode & MPI_MODE_CREATE) || (amode & MPI_MODE_EXCL))) {
return MPI_ERR_AMODE;
}
if ((amode & MPI_MODE_RDWR) && (amode & MPI_MODE_SEQUENTIAL)) {
return MPI_ERR_AMODE;
}
ompio_fh->f_rank = ompi_comm_rank (comm);
ompio_fh->f_size = ompi_comm_size (comm);
remote_arch = opal_local_arch;
ompio_fh->f_convertor = opal_convertor_create (remote_arch, 0);
if ( true == use_sharedfp ) {
ret = ompi_comm_dup (comm, &ompio_fh->f_comm);
if ( OMPI_SUCCESS != ret ) {
goto fn_fail;
}
}
else {
/* No need to duplicate the communicator if the file_open is called
from the sharedfp component, since the comm used as an input
is already a dup of the user level comm. */
ompio_fh->f_comm = comm;
}
ompio_fh->f_fstype = NONE;
ompio_fh->f_amode = amode;
ompio_fh->f_info = info;
/* set some function pointers required for fcoll, fbtls and sharedfp modules*/
ompio_fh->f_generate_current_file_view=generate_current_file_view_fn;
ompio_fh->f_get_mca_parameter_value=get_mca_parameter_value_fn;
ompio_fh->f_filename = filename;
mca_common_ompio_set_file_defaults (ompio_fh);
ompio_fh->f_split_coll_req = NULL;
ompio_fh->f_split_coll_in_use = false;
/*Initialize the print_queues queues here!*/
mca_common_ompio_initialize_print_queue(&ompio_fh->f_coll_write_time);
mca_common_ompio_initialize_print_queue(&ompio_fh->f_coll_read_time);
/* This fix is needed for data seiving to work with
two-phase collective I/O */
if ( OMPIO_MCA_GET(ompio_fh, overwrite_amode) && !(amode & MPI_MODE_SEQUENTIAL) ) {
if ((amode & MPI_MODE_WRONLY)){
amode -= MPI_MODE_WRONLY;
amode += MPI_MODE_RDWR;
}
}
/*--------------------------------------------------*/
if (OMPI_SUCCESS != (ret = mca_fs_base_file_select (ompio_fh,
NULL))) {
opal_output(1, "mca_fs_base_file_select() failed\n");
goto fn_fail;
}
if (OMPI_SUCCESS != (ret = mca_fbtl_base_file_select (ompio_fh,
NULL))) {
opal_output(1, "mca_fbtl_base_file_select() failed\n");
goto fn_fail;
}
ompio_fh->f_sharedfp_component = NULL; /*component*/
ompio_fh->f_sharedfp = NULL; /*module*/
ompio_fh->f_sharedfp_data = NULL; /*data*/
if ( true == use_sharedfp ) {
if (OMPI_SUCCESS != (ret = mca_sharedfp_base_file_select (ompio_fh, NULL))) {
opal_output ( ompi_io_base_framework.framework_output,
"mca_sharedfp_base_file_select() failed\n");
ompio_fh->f_sharedfp = NULL; /*module*/
/* Its ok to not have a shared file pointer module as long as the shared file
** pointer operations are not used. However, the first call to any file_read/write_shared
** function will return an error code.
*/
}
}
else {
ompio_fh->f_flags |= OMPIO_SHAREDFP_IS_SET;
}
ret = ompio_fh->f_fs->fs_file_open (comm,
filename,
amode,
info,
ompio_fh);
if ( OMPI_SUCCESS != ret ) {
#ifdef OMPIO_DEBUG
opal_output(1, "fs_file failed, error code %d\n", ret);
#endif
goto fn_fail;
}
if ( true == use_sharedfp ) {
/* open the file once more for the shared file pointer if required.
** Can be disabled by the user if no shared file pointer operations
** are used by his application.
*/
if ( NULL != ompio_fh->f_sharedfp ) {
ret = ompio_fh->f_sharedfp->sharedfp_file_open(comm,
filename,
amode,
info,
ompio_fh);
if ( OMPI_SUCCESS != ret ) {
goto fn_fail;
}
}
}
/* Set default file view */
mca_common_ompio_set_view(ompio_fh,
0,
&ompi_mpi_byte.dt,
&ompi_mpi_byte.dt,
"native",
info);
/* If file has been opened in the append mode, move the internal
file pointer of OMPIO to the very end of the file. */
if ( ompio_fh->f_amode & MPI_MODE_APPEND ) {
OMPI_MPI_OFFSET_TYPE current_size;
mca_sharedfp_base_module_t * shared_fp_base_module;
ompio_fh->f_fs->fs_file_get_size( ompio_fh,
&current_size);
mca_common_ompio_set_explicit_offset (ompio_fh, current_size);
if ( true == use_sharedfp ) {
if ( NULL != ompio_fh->f_sharedfp ) {
shared_fp_base_module = ompio_fh->f_sharedfp;
ret = shared_fp_base_module->sharedfp_seek(ompio_fh,current_size, MPI_SEEK_SET);
if ( MPI_SUCCESS != ret ) {
opal_output(1, "mca_common_ompio_file_open: Could not adjust position of "
"shared file pointer with MPI_MODE_APPEND\n");
ret = MPI_ERR_OTHER;
goto fn_fail;
}
}
}
}
return OMPI_SUCCESS;
fn_fail:
/* no need to free resources here, since the destructor
* is calling mca_io_ompio_file_close, which actually gets
*rid of all allocated memory items */
return ret;
}
int mca_common_ompio_file_close (ompio_file_t *ompio_fh)
{
int ret = OMPI_SUCCESS;
int delete_flag = 0;
char name[256];
ret = ompio_fh->f_comm->c_coll->coll_barrier ( ompio_fh->f_comm, ompio_fh->f_comm->c_coll->coll_barrier_module);
if ( OMPI_SUCCESS != ret ) {
/* Not sure what to do */
opal_output (1,"mca_common_ompio_file_close: error in Barrier \n");
return ret;
}
if(OMPIO_MCA_GET(ompio_fh, coll_timing_info)){
strcpy (name, "WRITE");
if (!mca_common_ompio_empty_print_queue(ompio_fh->f_coll_write_time)){
ret = mca_common_ompio_print_time_info(ompio_fh->f_coll_write_time,
name,
ompio_fh);
if (OMPI_SUCCESS != ret){
printf("Error in print_time_info ");
}
}
strcpy (name, "READ");
if (!mca_common_ompio_empty_print_queue(ompio_fh->f_coll_read_time)){
ret = mca_common_ompio_print_time_info(ompio_fh->f_coll_read_time,
name,
ompio_fh);
if (OMPI_SUCCESS != ret){
printf("Error in print_time_info ");
}
}
}
if ( ompio_fh->f_amode & MPI_MODE_DELETE_ON_CLOSE ) {
delete_flag = 1;
}
/*close the sharedfp file*/
if( NULL != ompio_fh->f_sharedfp ){
ret = ompio_fh->f_sharedfp->sharedfp_file_close(ompio_fh);
}
if ( NULL != ompio_fh->f_fs ) {
/* The pointer might not be set if file_close() is
** called from the file destructor in case of an error
** during file_open()
*/
ret = ompio_fh->f_fs->fs_file_close (ompio_fh);
}
if ( delete_flag ) {
ret = mca_common_ompio_file_delete ( ompio_fh->f_filename, &(MPI_INFO_NULL->super) );
}
if ( NULL != ompio_fh->f_fs ) {
mca_fs_base_file_unselect (ompio_fh);
}
if ( NULL != ompio_fh->f_fbtl ) {
mca_fbtl_base_file_unselect (ompio_fh);
}
if ( NULL != ompio_fh->f_fcoll ) {
mca_fcoll_base_file_unselect (ompio_fh);
}
if ( NULL != ompio_fh->f_sharedfp) {
mca_sharedfp_base_file_unselect (ompio_fh);
}
if (NULL != ompio_fh->f_io_array) {
free (ompio_fh->f_io_array);
ompio_fh->f_io_array = NULL;
}
if (NULL != ompio_fh->f_init_aggr_list) {
free (ompio_fh->f_init_aggr_list);
ompio_fh->f_init_aggr_list = NULL;
}
if (NULL != ompio_fh->f_aggr_list) {
free (ompio_fh->f_aggr_list);
ompio_fh->f_aggr_list = NULL;
}
if (NULL != ompio_fh->f_init_procs_in_group) {
free (ompio_fh->f_init_procs_in_group);
ompio_fh->f_init_procs_in_group = NULL;
}
if (NULL != ompio_fh->f_procs_in_group) {
free (ompio_fh->f_procs_in_group);
ompio_fh->f_procs_in_group = NULL;
}
if (NULL != ompio_fh->f_decoded_iov) {
free (ompio_fh->f_decoded_iov);
ompio_fh->f_decoded_iov = NULL;
}
if (NULL != ompio_fh->f_convertor) {
free (ompio_fh->f_convertor);
ompio_fh->f_convertor = NULL;
}
if (NULL != ompio_fh->f_datarep) {
free (ompio_fh->f_datarep);
ompio_fh->f_datarep = NULL;
}
if ( NULL != ompio_fh->f_coll_write_time ) {
free ( ompio_fh->f_coll_write_time );
ompio_fh->f_coll_write_time = NULL;
}
if ( NULL != ompio_fh->f_coll_read_time ) {
free ( ompio_fh->f_coll_read_time );
ompio_fh->f_coll_read_time = NULL;
}
if (MPI_DATATYPE_NULL != ompio_fh->f_iov_type) {
ompi_datatype_destroy (&ompio_fh->f_iov_type);
ompio_fh->f_iov_type=MPI_DATATYPE_NULL;
}
if ( MPI_DATATYPE_NULL != ompio_fh->f_etype ) {
ompi_datatype_destroy (&ompio_fh->f_etype);
}
if ( MPI_DATATYPE_NULL != ompio_fh->f_filetype ){
ompi_datatype_destroy (&ompio_fh->f_filetype);
}
if ( MPI_DATATYPE_NULL != ompio_fh->f_orig_filetype ){
ompi_datatype_destroy (&ompio_fh->f_orig_filetype);
}
if (MPI_COMM_NULL != ompio_fh->f_comm && !(ompio_fh->f_flags & OMPIO_SHAREDFP_IS_SET) ) {
ompi_comm_free (&ompio_fh->f_comm);
}
return ret;
}
int mca_common_ompio_file_get_size (ompio_file_t *ompio_fh,
OMPI_MPI_OFFSET_TYPE *size)
{
int ret = OMPI_SUCCESS;
ret = ompio_fh->f_fs->fs_file_get_size (ompio_fh, size);
return ret;
}
int mca_common_ompio_file_get_position (ompio_file_t *fh,
OMPI_MPI_OFFSET_TYPE *offset)
{
OMPI_MPI_OFFSET_TYPE off;
/* No. of copies of the entire file view */
off = (fh->f_offset - fh->f_disp)/fh->f_view_extent;
/* No. of elements per view */
off *= (fh->f_view_size / fh->f_etype_size);
/* No of elements used in the current copy of the view */
off += fh->f_total_bytes / fh->f_etype_size;
*offset = off;
return OMPI_SUCCESS;
}
int mca_common_ompio_set_file_defaults (ompio_file_t *fh)
{
if (NULL != fh) {
char char_stripe[MPI_MAX_INFO_VAL];
ompi_datatype_t *types[2];
int blocklen[2] = {1, 1};
ptrdiff_t d[2], base;
int i, flag;
fh->f_io_array = NULL;
fh->f_perm = OMPIO_PERM_NULL;
fh->f_flags = 0;
fh->f_bytes_per_agg = OMPIO_MCA_GET(fh, bytes_per_agg);
opal_info_get (fh->f_info, "cb_buffer_size", MPI_MAX_INFO_VAL, char_stripe, &flag);
if ( flag ) {
/* Info object trumps mca parameter value */
sscanf ( char_stripe, "%d", &fh->f_bytes_per_agg );
OMPIO_MCA_PRINT_INFO(fh, "cb_buffer_size", char_stripe, "");
}
fh->f_atomicity = 0;
fh->f_fs_block_size = 4096;
fh->f_offset = 0;
fh->f_disp = 0;
fh->f_position_in_file_view = 0;
fh->f_index_in_file_view = 0;
fh->f_total_bytes = 0;
fh->f_init_procs_per_group = -1;
fh->f_init_procs_in_group = NULL;
fh->f_procs_per_group = -1;
fh->f_procs_in_group = NULL;
fh->f_init_num_aggrs = -1;
fh->f_init_aggr_list = NULL;
fh->f_num_aggrs = -1;
fh->f_aggr_list = NULL;
/* Default file View */
fh->f_iov_type = MPI_DATATYPE_NULL;
fh->f_stripe_size = 0;
/*Decoded iovec of the file-view*/
fh->f_decoded_iov = NULL;
fh->f_etype = MPI_DATATYPE_NULL;
fh->f_filetype = MPI_DATATYPE_NULL;
fh->f_orig_filetype = MPI_DATATYPE_NULL;
fh->f_datarep = NULL;
/*Create a derived datatype for the created iovec */
types[0] = &ompi_mpi_long.dt;
types[1] = &ompi_mpi_long.dt;
d[0] = (ptrdiff_t) fh->f_decoded_iov;
d[1] = (ptrdiff_t) &fh->f_decoded_iov[0].iov_len;
base = d[0];
for (i=0 ; i<2 ; i++) {
d[i] -= base;
}
ompi_datatype_create_struct (2,
blocklen,
d,
types,
&fh->f_iov_type);
ompi_datatype_commit (&fh->f_iov_type);
return OMPI_SUCCESS;
}
else {
return OMPI_ERROR;
}
}
int mca_common_ompio_file_delete (const char *filename,
struct opal_info_t *info)
{
int ret = OMPI_SUCCESS;
ompio_file_t *fh = NULL;
/* No locking required for file_delete according to my understanding.
One thread will succeed, the other ones silently ignore the
error that the file is already deleted.
*/
/* Create an incomplete file handle, it will basically only
contain the filename. It is needed to select the correct
component in the fs framework and call the file_remove
function corresponding to the file type.
*/
ret = mca_common_ompio_create_incomplete_file_handle(filename, &fh);
if (OMPI_SUCCESS != ret) {
return ret;
}
ret = mca_fs_base_file_select (fh, NULL);
if (OMPI_SUCCESS != ret) {
opal_output(1, "error in mca_common_ompio_file_delete: "
"mca_fs_base_file_select() failed\n");
free(fh);
return ret;
}
ret = fh->f_fs->fs_file_delete ( (char *)filename, NULL);
free(fh);
if (OMPI_SUCCESS != ret) {
return ret;
}
return OMPI_SUCCESS;
}
int mca_common_ompio_create_incomplete_file_handle (const char *filename,
ompio_file_t **fh)
{
ompio_file_t *file;
if (NULL == filename) {
opal_output(1, "error in mca_common_ompio_create_incomplete_file_handle"
", filename is NULL.\n");
return OMPI_ERROR;
}
file = calloc(1, sizeof(ompio_file_t));
if (NULL == file) {
opal_output(1, "Out of memory.\n");
return OMPI_ERR_OUT_OF_RESOURCE;
}
/* Do not use communicator */
file->f_comm = MPI_COMM_NULL;
file->f_rank = OMPIO_ROOT;
/* TODO:
- Maybe copy the info for the info layer
- Maybe do the same as a file open: first create an ompi_file_t,
then allocate f_io_selected_data,
then use the ompio_file_t stored in this data structure
*/
/* We don't need to create a copy of the filename,
this file handle is only temporary. */
file->f_filename = filename;
*fh = file;
return OMPI_SUCCESS;
}
int mca_common_ompio_decode_datatype (struct ompio_file_t *fh,
ompi_datatype_t *datatype,
int count,
const void *buf,
size_t *max_data,
struct iovec **iov,
uint32_t *iovec_count)
{
opal_convertor_t convertor;
size_t remaining_length = 0;
uint32_t i;
uint32_t temp_count;
struct iovec *temp_iov=NULL;
size_t temp_data;
opal_convertor_clone (fh->f_convertor, &convertor, 0);
if (OMPI_SUCCESS != opal_convertor_prepare_for_send (&convertor,
&(datatype->super),
count,
buf)) {
opal_output (1, "Cannot attach the datatype to a convertor\n");
return OMPI_ERROR;
}
if ( 0 == datatype->super.size ) {
*max_data = 0;
*iovec_count = 0;
*iov = NULL;
return OMPI_SUCCESS;
}
remaining_length = count * datatype->super.size;
temp_count = OMPIO_IOVEC_INITIAL_SIZE;
temp_iov = (struct iovec*)malloc(temp_count * sizeof(struct iovec));
if (NULL == temp_iov) {
opal_output (1, "OUT OF MEMORY\n");
return OMPI_ERR_OUT_OF_RESOURCE;
}
while (0 == opal_convertor_raw(&convertor,
temp_iov,
&temp_count,
&temp_data)) {
#if 0
printf ("%d: New raw extraction (iovec_count = %d, max_data = %lu)\n",
fh->f_rank,temp_count, (unsigned long)temp_data);
for (i = 0; i < temp_count; i++) {
printf ("%d: \t{%p, %lu}\n",fh->f_rank,
temp_iov[i].iov_base,
(unsigned long)temp_iov[i].iov_len);
}
#endif
*iovec_count = *iovec_count + temp_count;
*max_data = *max_data + temp_data;
*iov = (struct iovec *) realloc (*iov, *iovec_count * sizeof(struct iovec));
if (NULL == *iov) {
opal_output(1, "OUT OF MEMORY\n");
free(temp_iov);
return OMPI_ERR_OUT_OF_RESOURCE;
}
for (i=0 ; i<temp_count ; i++) {
(*iov)[i+(*iovec_count-temp_count)].iov_base = temp_iov[i].iov_base;
(*iov)[i+(*iovec_count-temp_count)].iov_len = temp_iov[i].iov_len;
}
remaining_length -= temp_data;
temp_count = OMPIO_IOVEC_INITIAL_SIZE;
}
#if 0
printf ("%d: LAST raw extraction (iovec_count = %d, max_data = %d)\n",
fh->f_rank,temp_count, temp_data);
for (i = 0; i < temp_count; i++) {
printf ("%d: \t offset[%d]: %ld; length[%d]: %ld\n", fh->f_rank,i,temp_iov[i].iov_base, i,temp_iov[i].iov_len);
}
#endif
*iovec_count = *iovec_count + temp_count;
*max_data = *max_data + temp_data;
if ( temp_count > 0 ) {
*iov = (struct iovec *) realloc (*iov, *iovec_count * sizeof(struct iovec));
if (NULL == *iov) {
opal_output(1, "OUT OF MEMORY\n");
free(temp_iov);
return OMPI_ERR_OUT_OF_RESOURCE;
}
}
for (i=0 ; i<temp_count ; i++) {
(*iov)[i+(*iovec_count-temp_count)].iov_base = temp_iov[i].iov_base;
(*iov)[i+(*iovec_count-temp_count)].iov_len = temp_iov[i].iov_len;
}
remaining_length -= temp_data;
#if 0
if (0 == fh->f_rank) {
printf ("%d Entries: \n",*iovec_count);
for (i=0 ; i<*iovec_count ; i++) {
printf ("\t{%p, %d}\n",
(*iov)[i].iov_base,
(*iov)[i].iov_len);
}
}
#endif
if (remaining_length != 0) {
printf( "Not all raw description was been extracted (%lu bytes missing)\n",
(unsigned long) remaining_length );
}
free (temp_iov);
return OMPI_SUCCESS;
}
int mca_common_ompio_set_callbacks(mca_common_ompio_generate_current_file_view_fn_t generate_current_file_view,
mca_common_ompio_get_mca_parameter_value_fn_t get_mca_parameter_value)
{
generate_current_file_view_fn = generate_current_file_view;
get_mca_parameter_value_fn = get_mca_parameter_value;
return OMPI_SUCCESS;
}
Показать git blame Копировать постоянную ссылку