void
__kmp_pop_sync( int gtid, enum cons_type ct, ident_t const * ident )
{
int tos;
struct cons_header *p = __kmp_threads[ gtid ]->th.th_cons;
tos = p->stack_top;
KE_TRACE( 10, ("__kmp_pop_sync (%d %d)\n", gtid, __kmp_get_gtid() ) );
if ( tos == 0 || p->s_top == 0 ) {
__kmp_error_construct( kmp_i18n_msg_CnsDetectedEnd, ct, ident );
};
if ( tos != p->s_top || p->stack_data[ tos ].type != ct ) {
__kmp_check_null_func();
__kmp_error_construct2(
kmp_i18n_msg_CnsExpectedEnd,
ct, ident,
& p->stack_data[ tos ]
);
};
if ( gtid < 0 ) {
__kmp_check_null_func();
};
KE_TRACE( 100, ( POP_MSG( p ) ) );
p->s_top = p->stack_data[ tos ].prev;
p->stack_data[ tos ].type = ct_none;
p->stack_data[ tos ].ident = NULL;
p->stack_top = tos - 1;
KE_DUMP( 1000, dump_cons_stack( gtid, p ) );
}
enum cons_type
__kmp_pop_workshare( int gtid, enum cons_type ct, ident_t const * ident )
{
int tos;
struct cons_header *p = __kmp_threads[ gtid ]->th.th_cons;
tos = p->stack_top;
KE_TRACE( 10, ("__kmp_pop_workshare (%d %d)\n", gtid, __kmp_get_gtid() ) );
if ( tos == 0 || p->w_top == 0 ) {
__kmp_error_construct( kmp_i18n_msg_CnsDetectedEnd, ct, ident );
}
if ( tos != p->w_top ||
( p->stack_data[ tos ].type != ct &&
/* below are two exceptions to the rule that construct types must match */
! ( p->stack_data[ tos ].type == ct_pdo_ordered && ct == ct_pdo ) &&
! ( p->stack_data[ tos ].type == ct_task_ordered && ct == ct_task )
)
) {
__kmp_check_null_func();
__kmp_error_construct2(
kmp_i18n_msg_CnsExpectedEnd,
ct, ident,
& p->stack_data[ tos ]
);
}
KE_TRACE( 100, ( POP_MSG( p ) ) );
p->w_top = p->stack_data[ tos ].prev;
p->stack_data[ tos ].type = ct_none;
p->stack_data[ tos ].ident = NULL;
p->stack_top = tos - 1;
KE_DUMP( 1000, dump_cons_stack( gtid, p ) );
return p->stack_data[ p->w_top ].type;
}
void
__kmp_check_barrier( int gtid, enum cons_type ct, ident_t const * ident )
{
struct cons_header *p = __kmp_threads[ gtid ]->th.th_cons;
KE_TRACE( 10, ("__kmp_check_barrier (loc: %p, gtid: %d %d)\n", ident, gtid, __kmp_get_gtid() ) );
if ( ident != 0 ) {
__kmp_check_null_func();
}
if ( p->w_top > p->p_top ) {
/* we are already in a WORKSHARING construct for this PARALLEL region */
__kmp_error_construct2(
kmp_i18n_msg_CnsInvalidNesting,
ct, ident,
& p->stack_data[ p->w_top ]
);
}
if (p->s_top > p->p_top) {
/* we are already in a SYNC construct for this PARALLEL region */
__kmp_error_construct2(
kmp_i18n_msg_CnsInvalidNesting,
ct, ident,
& p->stack_data[ p->s_top ]
);
}
}
void
__kmp_push_workshare( int gtid, enum cons_type ct, ident_t const * ident )
{
int tos;
struct cons_header *p = __kmp_threads[ gtid ]->th.th_cons;
KE_TRACE( 10, ("__kmp_push_workshare (%d %d)\n", gtid, __kmp_get_gtid() ) );
__kmp_check_workshare( gtid, ct, ident );
KE_TRACE( 100, ( PUSH_MSG( ct, ident ) ) );
tos = ++p->stack_top;
p->stack_data[ tos ].type = ct;
p->stack_data[ tos ].prev = p->w_top;
p->stack_data[ tos ].ident = ident;
p->stack_data[ tos ].name = NULL;
p->w_top = tos;
KE_DUMP( 1000, dump_cons_stack( gtid, p ) );
}
void
__kmp_push_sync( int gtid, enum cons_type ct, ident_t const * ident, kmp_user_lock_p lck )
{
int tos;
struct cons_header *p = __kmp_threads[ gtid ]->th.th_cons;
KMP_ASSERT( gtid == __kmp_get_gtid() );
KE_TRACE( 10, ("__kmp_push_sync (gtid=%d)\n", gtid ) );
__kmp_check_sync( gtid, ct, ident, lck );
KE_TRACE( 100, ( PUSH_MSG( ct, ident ) ) );
tos = ++ p->stack_top;
p->stack_data[ tos ].type = ct;
p->stack_data[ tos ].prev = p->s_top;
p->stack_data[ tos ].ident = ident;
p->stack_data[ tos ].name = lck;
p->s_top = tos;
KE_DUMP( 1000, dump_cons_stack( gtid, p ) );
}
void
__kmp_push_parallel( int gtid, ident_t const * ident )
{
int tos;
struct cons_header *p = __kmp_threads[ gtid ]->th.th_cons;
KMP_DEBUG_ASSERT( __kmp_threads[ gtid ]-> th.th_cons );
KE_TRACE( 10, ("__kmp_push_parallel (%d %d)\n", gtid, __kmp_get_gtid() ) );
KE_TRACE( 100, ( PUSH_MSG( ct_parallel, ident ) ) );
if ( p->stack_top >= p->stack_size ) {
__kmp_expand_cons_stack( gtid, p );
}; // if
tos = ++p->stack_top;
p->stack_data[ tos ].type = ct_parallel;
p->stack_data[ tos ].prev = p->p_top;
p->stack_data[ tos ].ident = ident;
p->stack_data[ tos ].name = NULL;
p->p_top = tos;
KE_DUMP( 1000, dump_cons_stack( gtid, p ) );
}
struct cons_header *
__kmp_allocate_cons_stack( int gtid )
{
struct cons_header *p;
/* TODO for monitor perhaps? */
if ( gtid < 0 ) {
__kmp_check_null_func();
}; // if
KE_TRACE( 10, ("allocate cons_stack (%d)\n", gtid ) );
p = (struct cons_header *) __kmp_allocate( sizeof( struct cons_header ) );
p->p_top = p->w_top = p->s_top = 0;
p->stack_data = (struct cons_data *) __kmp_allocate( sizeof( struct cons_data ) * (MIN_STACK+1) );
p->stack_size = MIN_STACK;
p->stack_top = 0;
p->stack_data[ 0 ].type = ct_none;
p->stack_data[ 0 ].prev = 0;
p->stack_data[ 0 ].ident = NULL;
return p;
}
void
__kmp_check_workshare( int gtid, enum cons_type ct, ident_t const * ident )
{
struct cons_header *p = __kmp_threads[ gtid ]->th.th_cons;
KMP_DEBUG_ASSERT( __kmp_threads[ gtid ]-> th.th_cons );
KE_TRACE( 10, ("__kmp_check_workshare (%d %d)\n", gtid, __kmp_get_gtid() ) );
if ( p->stack_top >= p->stack_size ) {
__kmp_expand_cons_stack( gtid, p );
}; // if
if ( p->w_top > p->p_top &&
!(IS_CONS_TYPE_TASKQ(p->stack_data[ p->w_top ].type) && IS_CONS_TYPE_TASKQ(ct))) {
// We are already in a WORKSHARE construct for this PARALLEL region.
__kmp_error_construct2( kmp_i18n_msg_CnsInvalidNesting, ct, ident, & p->stack_data[ p->w_top ] );
}; // if
if ( p->s_top > p->p_top ) {
// We are already in a SYNC construct for this PARALLEL region.
__kmp_error_construct2( kmp_i18n_msg_CnsInvalidNesting, ct, ident, & p->stack_data[ p->s_top ] );
}; // if
}
static void
__kmp_expand_cons_stack( int gtid, struct cons_header *p )
{
int i;
struct cons_data *d;
/* TODO for monitor perhaps? */
if (gtid < 0)
__kmp_check_null_func();
KE_TRACE( 10, ("expand cons_stack (%d %d)\n", gtid, __kmp_get_gtid() ) );
d = p->stack_data;
p->stack_size = (p->stack_size * 2) + 100;
/* TODO free the old data */
p->stack_data = (struct cons_data *) __kmp_allocate( sizeof( struct cons_data ) * (p->stack_size+1) );
for (i = p->stack_top; i >= 0; --i)
p->stack_data[i] = d[i];
/* NOTE: we do not free the old stack_data */
}
static void
__kmp_for_static_init(
ident_t *loc,
kmp_int32 global_tid,
kmp_int32 schedtype,
kmp_int32 *plastiter,
T *plower,
T *pupper,
typename traits_t< T >::signed_t *pstride,
typename traits_t< T >::signed_t incr,
typename traits_t< T >::signed_t chunk
) {
KMP_COUNT_BLOCK(OMP_FOR_static);
typedef typename traits_t< T >::unsigned_t UT;
typedef typename traits_t< T >::signed_t ST;
/* this all has to be changed back to TID and such.. */
register kmp_int32 gtid = global_tid;
register kmp_uint32 tid;
register kmp_uint32 nth;
register UT trip_count;
register kmp_team_t *team;
register kmp_info_t *th = __kmp_threads[ gtid ];
#if OMPT_SUPPORT && OMPT_TRACE
ompt_team_info_t *team_info = __ompt_get_teaminfo(0, NULL);
ompt_task_info_t *task_info = __ompt_get_taskinfo(0);
#endif
KMP_DEBUG_ASSERT( plastiter && plower && pupper && pstride );
KE_TRACE( 10, ("__kmpc_for_static_init called (%d)\n", global_tid));
#ifdef KMP_DEBUG
{
const char * buff;
// create format specifiers before the debug output
buff = __kmp_str_format(
"__kmpc_for_static_init: T#%%d sched=%%d liter=%%d iter=(%%%s," \
" %%%s, %%%s) incr=%%%s chunk=%%%s signed?<%s>\n",
traits_t< T >::spec, traits_t< T >::spec, traits_t< ST >::spec,
traits_t< ST >::spec, traits_t< ST >::spec, traits_t< T >::spec );
KD_TRACE(100, ( buff, global_tid, schedtype, *plastiter,
*plower, *pupper, *pstride, incr, chunk ) );
__kmp_str_free( &buff );
}
#endif
if ( __kmp_env_consistency_check ) {
__kmp_push_workshare( global_tid, ct_pdo, loc );
if ( incr == 0 ) {
__kmp_error_construct( kmp_i18n_msg_CnsLoopIncrZeroProhibited, ct_pdo, loc );
}
}
/* special handling for zero-trip loops */
if ( incr > 0 ? (*pupper < *plower) : (*plower < *pupper) ) {
if( plastiter != NULL )
*plastiter = FALSE;
/* leave pupper and plower set to entire iteration space */
*pstride = incr; /* value should never be used */
// *plower = *pupper - incr; // let compiler bypass the illegal loop (like for(i=1;i<10;i--)) THIS LINE CAUSED shape2F/h_tests_1.f TO HAVE A FAILURE ON A ZERO-TRIP LOOP (lower=1,\
upper=0,stride=1) - JPH June 23, 2009.
#ifdef KMP_DEBUG
{
const char * buff;
// create format specifiers before the debug output
buff = __kmp_str_format(
"__kmpc_for_static_init:(ZERO TRIP) liter=%%d lower=%%%s upper=%%%s stride = %%%s signed?<%s>, loc = %%s\n",
traits_t< T >::spec, traits_t< T >::spec, traits_t< ST >::spec, traits_t< T >::spec );
KD_TRACE(100, ( buff, *plastiter, *plower, *pupper, *pstride, loc->psource ) );
__kmp_str_free( &buff );
}
#endif
KE_TRACE( 10, ("__kmpc_for_static_init: T#%d return\n", global_tid ) );
#if OMPT_SUPPORT && OMPT_TRACE
if ((ompt_status == ompt_status_track_callback) &&
ompt_callbacks.ompt_callback(ompt_event_loop_begin)) {
ompt_callbacks.ompt_callback(ompt_event_loop_begin)(
team_info->parallel_id, task_info->task_id,
team_info->microtask);
}
#endif
return;
}
#if OMP_40_ENABLED
if ( schedtype > kmp_ord_upper ) {
// we are in DISTRIBUTE construct
schedtype += kmp_sch_static - kmp_distribute_static; // AC: convert to usual schedule type
tid = th->th.th_team->t.t_master_tid;
team = th->th.th_team->t.t_parent;
} else
#endif
{
tid = __kmp_tid_from_gtid( global_tid );
team = th->th.th_team;
}
/* determine if "for" loop is an active worksharing construct */
if ( team -> t.t_serialized ) {
/* serialized parallel, each thread executes whole iteration space */
if( plastiter != NULL )
*plastiter = TRUE;
/* leave pupper and plower set to entire iteration space */
*pstride = (incr > 0) ? (*pupper - *plower + 1) : (-(*plower - *pupper + 1));
#ifdef KMP_DEBUG
{
const char * buff;
// create format specifiers before the debug output
buff = __kmp_str_format(
"__kmpc_for_static_init: (serial) liter=%%d lower=%%%s upper=%%%s stride = %%%s\n",
traits_t< T >::spec, traits_t< T >::spec, traits_t< ST >::spec );
KD_TRACE(100, ( buff, *plastiter, *plower, *pupper, *pstride ) );
__kmp_str_free( &buff );
}
#endif
KE_TRACE( 10, ("__kmpc_for_static_init: T#%d return\n", global_tid ) );
#if OMPT_SUPPORT && OMPT_TRACE
if ((ompt_status == ompt_status_track_callback) &&
ompt_callbacks.ompt_callback(ompt_event_loop_begin)) {
ompt_callbacks.ompt_callback(ompt_event_loop_begin)(
team_info->parallel_id, task_info->task_id,
team_info->microtask);
}
#endif
return;
}
nth = team->t.t_nproc;
if ( nth == 1 ) {
if( plastiter != NULL )
*plastiter = TRUE;
*pstride = (incr > 0) ? (*pupper - *plower + 1) : (-(*plower - *pupper + 1));
#ifdef KMP_DEBUG
{
const char * buff;
// create format specifiers before the debug output
buff = __kmp_str_format(
"__kmpc_for_static_init: (serial) liter=%%d lower=%%%s upper=%%%s stride = %%%s\n",
traits_t< T >::spec, traits_t< T >::spec, traits_t< ST >::spec );
KD_TRACE(100, ( buff, *plastiter, *plower, *pupper, *pstride ) );
__kmp_str_free( &buff );
}
#endif
KE_TRACE( 10, ("__kmpc_for_static_init: T#%d return\n", global_tid ) );
#if OMPT_SUPPORT && OMPT_TRACE
if ((ompt_status == ompt_status_track_callback) &&
ompt_callbacks.ompt_callback(ompt_event_loop_begin)) {
ompt_callbacks.ompt_callback(ompt_event_loop_begin)(
team_info->parallel_id, task_info->task_id,
team_info->microtask);
}
#endif
return;
}
/* compute trip count */
if ( incr == 1 ) {
trip_count = *pupper - *plower + 1;
} else if (incr == -1) {
trip_count = *plower - *pupper + 1;
} else {
if ( incr > 1 ) { // the check is needed for unsigned division when incr < 0
trip_count = (*pupper - *plower) / incr + 1;
} else {
trip_count = (*plower - *pupper) / ( -incr ) + 1;
}
}
if ( __kmp_env_consistency_check ) {
/* tripcount overflow? */
if ( trip_count == 0 && *pupper != *plower ) {
__kmp_error_construct( kmp_i18n_msg_CnsIterationRangeTooLarge, ct_pdo, loc );
}
}
/* compute remaining parameters */
switch ( schedtype ) {
case kmp_sch_static:
{
if ( trip_count < nth ) {
KMP_DEBUG_ASSERT(
__kmp_static == kmp_sch_static_greedy || \
__kmp_static == kmp_sch_static_balanced
); // Unknown static scheduling type.
if ( tid < trip_count ) {
*pupper = *plower = *plower + tid * incr;
} else {
*plower = *pupper + incr;
}
if( plastiter != NULL )
*plastiter = ( tid == trip_count - 1 );
} else {
if ( __kmp_static == kmp_sch_static_balanced ) {
register UT small_chunk = trip_count / nth;
register UT extras = trip_count % nth;
*plower += incr * ( tid * small_chunk + ( tid < extras ? tid : extras ) );
*pupper = *plower + small_chunk * incr - ( tid < extras ? 0 : incr );
if( plastiter != NULL )
*plastiter = ( tid == nth - 1 );
} else {
register T big_chunk_inc_count = ( trip_count/nth +
( ( trip_count % nth ) ? 1 : 0) ) * incr;
register T old_upper = *pupper;
KMP_DEBUG_ASSERT( __kmp_static == kmp_sch_static_greedy );
// Unknown static scheduling type.
*plower += tid * big_chunk_inc_count;
*pupper = *plower + big_chunk_inc_count - incr;
if ( incr > 0 ) {
if( *pupper < *plower )
*pupper = i_maxmin< T >::mx;
if( plastiter != NULL )
*plastiter = *plower <= old_upper && *pupper > old_upper - incr;
if ( *pupper > old_upper ) *pupper = old_upper; // tracker C73258
} else {
if( *pupper > *plower )
*pupper = i_maxmin< T >::mn;
if( plastiter != NULL )
*plastiter = *plower >= old_upper && *pupper < old_upper - incr;
if ( *pupper < old_upper ) *pupper = old_upper; // tracker C73258
}
}
}
break;
}
case kmp_sch_static_chunked:
{
register ST span;
if ( chunk < 1 ) {
chunk = 1;
}
span = chunk * incr;
*pstride = span * nth;
*plower = *plower + (span * tid);
*pupper = *plower + span - incr;
if( plastiter != NULL )
*plastiter = (tid == ((trip_count - 1)/( UT )chunk) % nth);
break;
}
default:
KMP_ASSERT2( 0, "__kmpc_for_static_init: unknown scheduling type" );
break;
}
#if USE_ITT_BUILD
// Report loop metadata
if ( KMP_MASTER_TID(tid) && __itt_metadata_add_ptr && __kmp_forkjoin_frames_mode == 3 &&
#if OMP_40_ENABLED
th->th.th_teams_microtask == NULL &&
#endif
team->t.t_active_level == 1 )
{
kmp_uint64 cur_chunk = chunk;
// Calculate chunk in case it was not specified; it is specified for kmp_sch_static_chunked
if ( schedtype == kmp_sch_static ) {
cur_chunk = trip_count / nth + ( ( trip_count % nth ) ? 1 : 0);
}
// 0 - "static" schedule
__kmp_itt_metadata_loop(loc, 0, trip_count, cur_chunk);
}
#endif
#ifdef KMP_DEBUG
{
const char * buff;
// create format specifiers before the debug output
buff = __kmp_str_format(
"__kmpc_for_static_init: liter=%%d lower=%%%s upper=%%%s stride = %%%s signed?<%s>\n",
traits_t< T >::spec, traits_t< T >::spec, traits_t< ST >::spec, traits_t< T >::spec );
KD_TRACE(100, ( buff, *plastiter, *plower, *pupper, *pstride ) );
__kmp_str_free( &buff );
}
#endif
KE_TRACE( 10, ("__kmpc_for_static_init: T#%d return\n", global_tid ) );
#if OMPT_SUPPORT && OMPT_TRACE
if ((ompt_status == ompt_status_track_callback) &&
ompt_callbacks.ompt_callback(ompt_event_loop_begin)) {
ompt_callbacks.ompt_callback(ompt_event_loop_begin)(
team_info->parallel_id, task_info->task_id, team_info->microtask);
}
#endif
return;
}
static void
__kmp_team_static_init(
ident_t *loc,
kmp_int32 gtid,
kmp_int32 *p_last,
T *p_lb,
T *p_ub,
typename traits_t< T >::signed_t *p_st,
typename traits_t< T >::signed_t incr,
typename traits_t< T >::signed_t chunk
) {
// The routine returns the first chunk distributed to the team and
// stride for next chunks calculation.
// Last iteration flag set for the team that will execute
// the last iteration of the loop.
// The routine is called for dist_schedue(static,chunk) only.
typedef typename traits_t< T >::unsigned_t UT;
typedef typename traits_t< T >::signed_t ST;
kmp_uint32 team_id;
kmp_uint32 nteams;
UT trip_count;
T lower;
T upper;
ST span;
kmp_team_t *team;
kmp_info_t *th;
KMP_DEBUG_ASSERT( p_last && p_lb && p_ub && p_st );
KE_TRACE( 10, ("__kmp_team_static_init called (%d)\n", gtid));
#ifdef KMP_DEBUG
{
const char * buff;
// create format specifiers before the debug output
buff = __kmp_str_format( "__kmp_team_static_init enter: T#%%d liter=%%d "\
"iter=(%%%s, %%%s, %%%s) chunk %%%s; signed?<%s>\n",
traits_t< T >::spec, traits_t< T >::spec, traits_t< ST >::spec,
traits_t< ST >::spec, traits_t< T >::spec );
KD_TRACE(100, ( buff, gtid, *p_last, *p_lb, *p_ub, *p_st, chunk ) );
__kmp_str_free( &buff );
}
#endif
lower = *p_lb;
upper = *p_ub;
if( __kmp_env_consistency_check ) {
if( incr == 0 ) {
__kmp_error_construct( kmp_i18n_msg_CnsLoopIncrZeroProhibited, ct_pdo, loc );
}
if( incr > 0 ? (upper < lower) : (lower < upper) ) {
// The loop is illegal.
// Some zero-trip loops maintained by compiler, e.g.:
// for(i=10;i<0;++i) // lower >= upper - run-time check
// for(i=0;i>10;--i) // lower <= upper - run-time check
// for(i=0;i>10;++i) // incr > 0 - compile-time check
// for(i=10;i<0;--i) // incr < 0 - compile-time check
// Compiler does not check the following illegal loops:
// for(i=0;i<10;i+=incr) // where incr<0
// for(i=10;i>0;i-=incr) // where incr<0
__kmp_error_construct( kmp_i18n_msg_CnsLoopIncrIllegal, ct_pdo, loc );
}
}
th = __kmp_threads[gtid];
KMP_DEBUG_ASSERT(th->th.th_teams_microtask); // we are in the teams construct
team = th->th.th_team;
#if OMP_40_ENABLED
nteams = th->th.th_teams_size.nteams;
#endif
team_id = team->t.t_master_tid;
KMP_DEBUG_ASSERT(nteams == team->t.t_parent->t.t_nproc);
// compute trip count
if( incr == 1 ) {
trip_count = upper - lower + 1;
} else if(incr == -1) {
trip_count = lower - upper + 1;
} else {
trip_count = (ST)(upper - lower) / incr + 1; // cast to signed to cover incr<0 case
}
if( chunk < 1 )
chunk = 1;
span = chunk * incr;
*p_st = span * nteams;
*p_lb = lower + (span * team_id);
*p_ub = *p_lb + span - incr;
if ( p_last != NULL )
*p_last = (team_id == ((trip_count - 1)/(UT)chunk) % nteams);
// Correct upper bound if needed
if( incr > 0 ) {
if( *p_ub < *p_lb ) // overflow?
*p_ub = i_maxmin< T >::mx;
if( *p_ub > upper )
*p_ub = upper; // tracker C73258
} else { // incr < 0
if( *p_ub > *p_lb )
*p_ub = i_maxmin< T >::mn;
if( *p_ub < upper )
*p_ub = upper; // tracker C73258
}
#ifdef KMP_DEBUG
{
const char * buff;
// create format specifiers before the debug output
buff = __kmp_str_format( "__kmp_team_static_init exit: T#%%d team%%u liter=%%d "\
"iter=(%%%s, %%%s, %%%s) chunk %%%s\n",
traits_t< T >::spec, traits_t< T >::spec, traits_t< ST >::spec,
traits_t< ST >::spec );
KD_TRACE(100, ( buff, gtid, team_id, *p_last, *p_lb, *p_ub, *p_st, chunk ) );
__kmp_str_free( &buff );
}
#endif
}
static void
__kmp_dist_for_static_init(
ident_t *loc,
kmp_int32 gtid,
kmp_int32 schedule,
kmp_int32 *plastiter,
T *plower,
T *pupper,
T *pupperDist,
typename traits_t< T >::signed_t *pstride,
typename traits_t< T >::signed_t incr,
typename traits_t< T >::signed_t chunk
) {
KMP_COUNT_BLOCK(OMP_DISTR_FOR_static);
typedef typename traits_t< T >::unsigned_t UT;
typedef typename traits_t< T >::signed_t ST;
register kmp_uint32 tid;
register kmp_uint32 nth;
register kmp_uint32 team_id;
register kmp_uint32 nteams;
register UT trip_count;
register kmp_team_t *team;
kmp_info_t * th;
KMP_DEBUG_ASSERT( plastiter && plower && pupper && pupperDist && pstride );
KE_TRACE( 10, ("__kmpc_dist_for_static_init called (%d)\n", gtid));
#ifdef KMP_DEBUG
{
const char * buff;
// create format specifiers before the debug output
buff = __kmp_str_format(
"__kmpc_dist_for_static_init: T#%%d schedLoop=%%d liter=%%d "\
"iter=(%%%s, %%%s, %%%s) chunk=%%%s signed?<%s>\n",
traits_t< T >::spec, traits_t< T >::spec, traits_t< ST >::spec,
traits_t< ST >::spec, traits_t< T >::spec );
KD_TRACE(100, ( buff, gtid, schedule, *plastiter,
*plower, *pupper, incr, chunk ) );
__kmp_str_free( &buff );
}
#endif
if( __kmp_env_consistency_check ) {
__kmp_push_workshare( gtid, ct_pdo, loc );
if( incr == 0 ) {
__kmp_error_construct( kmp_i18n_msg_CnsLoopIncrZeroProhibited, ct_pdo, loc );
}
if( incr > 0 ? (*pupper < *plower) : (*plower < *pupper) ) {
// The loop is illegal.
// Some zero-trip loops maintained by compiler, e.g.:
// for(i=10;i<0;++i) // lower >= upper - run-time check
// for(i=0;i>10;--i) // lower <= upper - run-time check
// for(i=0;i>10;++i) // incr > 0 - compile-time check
// for(i=10;i<0;--i) // incr < 0 - compile-time check
// Compiler does not check the following illegal loops:
// for(i=0;i<10;i+=incr) // where incr<0
// for(i=10;i>0;i-=incr) // where incr<0
__kmp_error_construct( kmp_i18n_msg_CnsLoopIncrIllegal, ct_pdo, loc );
}
}
tid = __kmp_tid_from_gtid( gtid );
th = __kmp_threads[gtid];
KMP_DEBUG_ASSERT(th->th.th_teams_microtask); // we are in the teams construct
nth = th->th.th_team_nproc;
team = th->th.th_team;
#if OMP_40_ENABLED
nteams = th->th.th_teams_size.nteams;
#endif
team_id = team->t.t_master_tid;
KMP_DEBUG_ASSERT(nteams == team->t.t_parent->t.t_nproc);
// compute global trip count
if( incr == 1 ) {
trip_count = *pupper - *plower + 1;
} else if(incr == -1) {
trip_count = *plower - *pupper + 1;
} else {
trip_count = (ST)(*pupper - *plower) / incr + 1; // cast to signed to cover incr<0 case
}
*pstride = *pupper - *plower; // just in case (can be unused)
if( trip_count <= nteams ) {
KMP_DEBUG_ASSERT(
__kmp_static == kmp_sch_static_greedy || \
__kmp_static == kmp_sch_static_balanced
); // Unknown static scheduling type.
// only masters of some teams get single iteration, other threads get nothing
if( team_id < trip_count && tid == 0 ) {
*pupper = *pupperDist = *plower = *plower + team_id * incr;
} else {
*pupperDist = *pupper;
*plower = *pupper + incr; // compiler should skip loop body
}
if( plastiter != NULL )
*plastiter = ( tid == 0 && team_id == trip_count - 1 );
} else {
// Get the team's chunk first (each team gets at most one chunk)
if( __kmp_static == kmp_sch_static_balanced ) {
register UT chunkD = trip_count / nteams;
register UT extras = trip_count % nteams;
*plower += incr * ( team_id * chunkD + ( team_id < extras ? team_id : extras ) );
*pupperDist = *plower + chunkD * incr - ( team_id < extras ? 0 : incr );
if( plastiter != NULL )
*plastiter = ( team_id == nteams - 1 );
} else {
register T chunk_inc_count =
( trip_count / nteams + ( ( trip_count % nteams ) ? 1 : 0) ) * incr;
register T upper = *pupper;
KMP_DEBUG_ASSERT( __kmp_static == kmp_sch_static_greedy );
// Unknown static scheduling type.
*plower += team_id * chunk_inc_count;
*pupperDist = *plower + chunk_inc_count - incr;
// Check/correct bounds if needed
if( incr > 0 ) {
if( *pupperDist < *plower )
*pupperDist = i_maxmin< T >::mx;
if( plastiter != NULL )
*plastiter = *plower <= upper && *pupperDist > upper - incr;
if( *pupperDist > upper )
*pupperDist = upper; // tracker C73258
if( *plower > *pupperDist ) {
*pupper = *pupperDist; // no iterations available for the team
goto end;
}
} else {
if( *pupperDist > *plower )
*pupperDist = i_maxmin< T >::mn;
if( plastiter != NULL )
*plastiter = *plower >= upper && *pupperDist < upper - incr;
if( *pupperDist < upper )
*pupperDist = upper; // tracker C73258
if( *plower < *pupperDist ) {
*pupper = *pupperDist; // no iterations available for the team
goto end;
}
}
}
// Get the parallel loop chunk now (for thread)
// compute trip count for team's chunk
if( incr == 1 ) {
trip_count = *pupperDist - *plower + 1;
} else if(incr == -1) {
trip_count = *plower - *pupperDist + 1;
} else {
trip_count = (ST)(*pupperDist - *plower) / incr + 1;
}
KMP_DEBUG_ASSERT( trip_count );
switch( schedule ) {
case kmp_sch_static:
{
if( trip_count <= nth ) {
KMP_DEBUG_ASSERT(
__kmp_static == kmp_sch_static_greedy || \
__kmp_static == kmp_sch_static_balanced
); // Unknown static scheduling type.
if( tid < trip_count )
*pupper = *plower = *plower + tid * incr;
else
*plower = *pupper + incr; // no iterations available
if( plastiter != NULL )
if( *plastiter != 0 && !( tid == trip_count - 1 ) )
*plastiter = 0;
} else {
if( __kmp_static == kmp_sch_static_balanced ) {
register UT chunkL = trip_count / nth;
register UT extras = trip_count % nth;
*plower += incr * (tid * chunkL + (tid < extras ? tid : extras));
*pupper = *plower + chunkL * incr - (tid < extras ? 0 : incr);
if( plastiter != NULL )
if( *plastiter != 0 && !( tid == nth - 1 ) )
*plastiter = 0;
} else {
register T chunk_inc_count =
( trip_count / nth + ( ( trip_count % nth ) ? 1 : 0) ) * incr;
register T upper = *pupperDist;
KMP_DEBUG_ASSERT( __kmp_static == kmp_sch_static_greedy );
// Unknown static scheduling type.
*plower += tid * chunk_inc_count;
*pupper = *plower + chunk_inc_count - incr;
if( incr > 0 ) {
if( *pupper < *plower )
*pupper = i_maxmin< T >::mx;
if( plastiter != NULL )
if( *plastiter != 0 && !(*plower <= upper && *pupper > upper - incr) )
*plastiter = 0;
if( *pupper > upper )
*pupper = upper;//tracker C73258
} else {
if( *pupper > *plower )
*pupper = i_maxmin< T >::mn;
if( plastiter != NULL )
if( *plastiter != 0 && !(*plower >= upper && *pupper < upper - incr) )
*plastiter = 0;
if( *pupper < upper )
*pupper = upper;//tracker C73258
}
}
}
break;
}
case kmp_sch_static_chunked:
{
register ST span;
if( chunk < 1 )
chunk = 1;
span = chunk * incr;
*pstride = span * nth;
*plower = *plower + (span * tid);
*pupper = *plower + span - incr;
if( plastiter != NULL )
if( *plastiter != 0 && !(tid == ((trip_count - 1) / ( UT )chunk) % nth) )
*plastiter = 0;
break;
}
default:
KMP_ASSERT2( 0, "__kmpc_dist_for_static_init: unknown loop scheduling type" );
break;
}
}
end:;
#ifdef KMP_DEBUG
{
const char * buff;
// create format specifiers before the debug output
buff = __kmp_str_format(
"__kmpc_dist_for_static_init: last=%%d lo=%%%s up=%%%s upDist=%%%s "\
"stride=%%%s signed?<%s>\n",
traits_t< T >::spec, traits_t< T >::spec, traits_t< T >::spec,
traits_t< ST >::spec, traits_t< T >::spec );
KD_TRACE(100, ( buff, *plastiter, *plower, *pupper, *pupperDist, *pstride ) );
__kmp_str_free( &buff );
}
#endif
KE_TRACE( 10, ("__kmpc_dist_for_static_init: T#%d return\n", gtid ) );
return;
}
__kmp_check_sync( int gtid, enum cons_type ct, ident_t const * ident, kmp_user_lock_p lck )
#endif
{
struct cons_header *p = __kmp_threads[ gtid ]->th.th_cons;
KE_TRACE( 10, ("__kmp_check_sync (gtid=%d)\n", __kmp_get_gtid() ) );
if (p->stack_top >= p->stack_size)
__kmp_expand_cons_stack( gtid, p );
if (ct == ct_ordered_in_parallel || ct == ct_ordered_in_pdo || ct == ct_ordered_in_taskq ) {
if (p->w_top <= p->p_top) {
/* we are not in a worksharing construct */
#ifdef BUILD_PARALLEL_ORDERED
/* do not report error messages for PARALLEL ORDERED */
KMP_ASSERT( ct == ct_ordered_in_parallel );
#else
__kmp_error_construct( kmp_i18n_msg_CnsBoundToWorksharing, ct, ident );
#endif /* BUILD_PARALLEL_ORDERED */
} else {
/* inside a WORKSHARING construct for this PARALLEL region */
if (!IS_CONS_TYPE_ORDERED(p->stack_data[ p->w_top ].type)) {
if (p->stack_data[ p->w_top ].type == ct_taskq) {
__kmp_error_construct2(
kmp_i18n_msg_CnsNotInTaskConstruct,
ct, ident,
& p->stack_data[ p->w_top ]
);
} else {
__kmp_error_construct2(
kmp_i18n_msg_CnsNoOrderedClause,
ct, ident,
& p->stack_data[ p->w_top ]
);
}
}
}
if (p->s_top > p->p_top && p->s_top > p->w_top) {
/* inside a sync construct which is inside a worksharing construct */
int index = p->s_top;
enum cons_type stack_type;
stack_type = p->stack_data[ index ].type;
if (stack_type == ct_critical ||
( ( stack_type == ct_ordered_in_parallel ||
stack_type == ct_ordered_in_pdo ||
stack_type == ct_ordered_in_taskq ) && /* C doesn't allow named ordered; ordered in ordered gets error */
p->stack_data[ index ].ident != NULL &&
(p->stack_data[ index ].ident->flags & KMP_IDENT_KMPC ))) {
/* we are in ORDERED which is inside an ORDERED or CRITICAL construct */
__kmp_error_construct2(
kmp_i18n_msg_CnsInvalidNesting,
ct, ident,
& p->stack_data[ index ]
);
}
}
} else if ( ct == ct_critical ) {
#if KMP_USE_DYNAMIC_LOCK
if ( lck != NULL && __kmp_get_user_lock_owner( lck, seq ) == gtid ) { /* this same thread already has lock for this critical section */
#else
if ( lck != NULL && __kmp_get_user_lock_owner( lck ) == gtid ) { /* this same thread already has lock for this critical section */
#endif
int index = p->s_top;
struct cons_data cons = { NULL, ct_critical, 0, NULL };
/* walk up construct stack and try to find critical with matching name */
while ( index != 0 && p->stack_data[ index ].name != lck ) {
index = p->stack_data[ index ].prev;
}
if ( index != 0 ) {
/* found match on the stack (may not always because of interleaved critical for Fortran) */
cons = p->stack_data[ index ];
}
/* we are in CRITICAL which is inside a CRITICAL construct of the same name */
__kmp_error_construct2( kmp_i18n_msg_CnsNestingSameName, ct, ident, & cons );
}
} else if ( ct == ct_master || ct == ct_reduce ) {
if (p->w_top > p->p_top) {
/* inside a WORKSHARING construct for this PARALLEL region */
__kmp_error_construct2(
kmp_i18n_msg_CnsInvalidNesting,
ct, ident,
& p->stack_data[ p->w_top ]
);
}
if (ct == ct_reduce && p->s_top > p->p_top) {
/* inside a another SYNC construct for this PARALLEL region */
__kmp_error_construct2(
kmp_i18n_msg_CnsInvalidNesting,
ct, ident,
& p->stack_data[ p->s_top ]
);
}; // if
}; // if
}
void
#if KMP_USE_DYNAMIC_LOCK
__kmp_push_sync( int gtid, enum cons_type ct, ident_t const * ident, kmp_user_lock_p lck, kmp_uint32 seq )
#else
__kmp_push_sync( int gtid, enum cons_type ct, ident_t const * ident, kmp_user_lock_p lck )
#endif
{
int tos;
struct cons_header *p = __kmp_threads[ gtid ]->th.th_cons;
KMP_ASSERT( gtid == __kmp_get_gtid() );
KE_TRACE( 10, ("__kmp_push_sync (gtid=%d)\n", gtid ) );
#if KMP_USE_DYNAMIC_LOCK
__kmp_check_sync( gtid, ct, ident, lck, seq );
#else
__kmp_check_sync( gtid, ct, ident, lck );
#endif
KE_TRACE( 100, ( PUSH_MSG( ct, ident ) ) );
tos = ++ p->stack_top;
p->stack_data[ tos ].type = ct;
p->stack_data[ tos ].prev = p->s_top;
p->stack_data[ tos ].ident = ident;
p->stack_data[ tos ].name = lck;
p->s_top = tos;
KE_DUMP( 1000, dump_cons_stack( gtid, p ) );
}
static void
__kmp_for_static_init(
ident_t *loc,
kmp_int32 global_tid,
kmp_int32 schedtype,
kmp_int32 *plastiter,
T *plower,
T *pupper,
typename traits_t< T >::signed_t *pstride,
typename traits_t< T >::signed_t incr,
typename traits_t< T >::signed_t chunk
) {
typedef typename traits_t< T >::unsigned_t UT;
typedef typename traits_t< T >::signed_t ST;
/* this all has to be changed back to TID and such.. */
register kmp_int32 gtid = global_tid;
register kmp_uint32 tid;
register kmp_uint32 nth;
register UT trip_count;
register kmp_team_t *team;
KE_TRACE( 10, ("__kmpc_for_static_init called (%d)\n", global_tid));
#ifdef KMP_DEBUG
{
const char * buff;
// create format specifiers before the debug output
buff = __kmp_str_format(
"__kmpc_for_static_init: T#%%d sched=%%d liter=%%d iter=(%%%s," \
" %%%s, %%%s) incr=%%%s chunk=%%%s signed?<%s>\n",
traits_t< T >::spec, traits_t< T >::spec, traits_t< ST >::spec,
traits_t< ST >::spec, traits_t< ST >::spec, traits_t< T >::spec );
KD_TRACE(100, ( buff, global_tid, schedtype, *plastiter,
*plower, *pupper, *pstride, incr, chunk ) );
__kmp_str_free( &buff );
}
#endif
if ( __kmp_env_consistency_check ) {
__kmp_push_workshare( global_tid, ct_pdo, loc );
if ( incr == 0 ) {
__kmp_error_construct( kmp_i18n_msg_CnsLoopIncrZeroProhibited, ct_pdo, loc );
}
}
/* special handling for zero-trip loops */
if ( incr > 0 ? (*pupper < *plower) : (*plower < *pupper) ) {
*plastiter = FALSE;
/* leave pupper and plower set to entire iteration space */
*pstride = incr; /* value should never be used */
// *plower = *pupper - incr; // let compiler bypass the illegal loop (like for(i=1;i<10;i--)) THIS LINE CAUSED shape2F/h_tests_1.f TO HAVE A FAILURE ON A ZERO-TRIP LOOP (lower=1,\
upper=0,stride=1) - JPH June 23, 2009.
#ifdef KMP_DEBUG
{
const char * buff;
// create format specifiers before the debug output
buff = __kmp_str_format(
"__kmpc_for_static_init:(ZERO TRIP) liter=%%d lower=%%%s upper=%%%s stride = %%%s signed?<%s>, loc = %%s\n",
traits_t< T >::spec, traits_t< T >::spec, traits_t< ST >::spec, traits_t< T >::spec );
KD_TRACE(100, ( buff, *plastiter, *plower, *pupper, *pstride, loc->psource ) );
__kmp_str_free( &buff );
}
#endif
KE_TRACE( 10, ("__kmpc_for_static_init: T#%d return\n", global_tid ) );
return;
}
static void __kmp_for_static_init(ident_t *loc, kmp_int32 global_tid,
kmp_int32 schedtype, kmp_int32 *plastiter,
T *plower, T *pupper,
typename traits_t<T>::signed_t *pstride,
typename traits_t<T>::signed_t incr,
typename traits_t<T>::signed_t chunk
#if OMPT_SUPPORT && OMPT_OPTIONAL
,
void *codeptr
#endif
) {
KMP_COUNT_BLOCK(OMP_FOR_static);
KMP_TIME_PARTITIONED_BLOCK(FOR_static_scheduling);
typedef typename traits_t<T>::unsigned_t UT;
typedef typename traits_t<T>::signed_t ST;
/* this all has to be changed back to TID and such.. */
kmp_int32 gtid = global_tid;
kmp_uint32 tid;
kmp_uint32 nth;
UT trip_count;
kmp_team_t *team;
kmp_info_t *th = __kmp_threads[gtid];
#if OMPT_SUPPORT && OMPT_OPTIONAL
ompt_team_info_t *team_info = NULL;
ompt_task_info_t *task_info = NULL;
ompt_work_type_t ompt_work_type = ompt_work_loop;
static kmp_int8 warn = 0;
if (ompt_enabled.ompt_callback_work) {
// Only fully initialize variables needed by OMPT if OMPT is enabled.
team_info = __ompt_get_teaminfo(0, NULL);
task_info = __ompt_get_task_info_object(0);
// Determine workshare type
if (loc != NULL) {
if ((loc->flags & KMP_IDENT_WORK_LOOP) != 0) {
ompt_work_type = ompt_work_loop;
} else if ((loc->flags & KMP_IDENT_WORK_SECTIONS) != 0) {
ompt_work_type = ompt_work_sections;
} else if ((loc->flags & KMP_IDENT_WORK_DISTRIBUTE) != 0) {
ompt_work_type = ompt_work_distribute;
} else {
kmp_int8 bool_res =
KMP_COMPARE_AND_STORE_ACQ8(&warn, (kmp_int8)0, (kmp_int8)1);
if (bool_res)
KMP_WARNING(OmptOutdatedWorkshare);
}
KMP_DEBUG_ASSERT(ompt_work_type);
}
}
#endif
KMP_DEBUG_ASSERT(plastiter && plower && pupper && pstride);
KE_TRACE(10, ("__kmpc_for_static_init called (%d)\n", global_tid));
#ifdef KMP_DEBUG
{
char *buff;
// create format specifiers before the debug output
buff = __kmp_str_format(
"__kmpc_for_static_init: T#%%d sched=%%d liter=%%d iter=(%%%s,"
" %%%s, %%%s) incr=%%%s chunk=%%%s signed?<%s>\n",
traits_t<T>::spec, traits_t<T>::spec, traits_t<ST>::spec,
traits_t<ST>::spec, traits_t<ST>::spec, traits_t<T>::spec);
KD_TRACE(100, (buff, global_tid, schedtype, *plastiter, *plower, *pupper,
*pstride, incr, chunk));
__kmp_str_free(&buff);
}
#endif
if (__kmp_env_consistency_check) {
__kmp_push_workshare(global_tid, ct_pdo, loc);
if (incr == 0) {
__kmp_error_construct(kmp_i18n_msg_CnsLoopIncrZeroProhibited, ct_pdo,
loc);
}
}
/* special handling for zero-trip loops */
if (incr > 0 ? (*pupper < *plower) : (*plower < *pupper)) {
if (plastiter != NULL)
*plastiter = FALSE;
/* leave pupper and plower set to entire iteration space */
*pstride = incr; /* value should never be used */
// *plower = *pupper - incr;
// let compiler bypass the illegal loop (like for(i=1;i<10;i--))
// THE LINE COMMENTED ABOVE CAUSED shape2F/h_tests_1.f TO HAVE A FAILURE
// ON A ZERO-TRIP LOOP (lower=1, upper=0,stride=1) - JPH June 23, 2009.
#ifdef KMP_DEBUG
{
char *buff;
// create format specifiers before the debug output
buff = __kmp_str_format("__kmpc_for_static_init:(ZERO TRIP) liter=%%d "
"lower=%%%s upper=%%%s stride = %%%s "
"signed?<%s>, loc = %%s\n",
traits_t<T>::spec, traits_t<T>::spec,
traits_t<ST>::spec, traits_t<T>::spec);
KD_TRACE(100,
(buff, *plastiter, *plower, *pupper, *pstride, loc->psource));
__kmp_str_free(&buff);
}
#endif
KE_TRACE(10, ("__kmpc_for_static_init: T#%d return\n", global_tid));
#if OMPT_SUPPORT && OMPT_OPTIONAL
if (ompt_enabled.ompt_callback_work) {
ompt_callbacks.ompt_callback(ompt_callback_work)(
ompt_work_type, ompt_scope_begin, &(team_info->parallel_data),
&(task_info->task_data), 0, codeptr);
}
#endif
KMP_COUNT_VALUE(FOR_static_iterations, 0);
return;
}
#if OMP_40_ENABLED
// Although there are schedule enumerations above kmp_ord_upper which are not
// schedules for "distribute", the only ones which are useful are dynamic, so
// cannot be seen here, since this codepath is only executed for static
// schedules.
if (schedtype > kmp_ord_upper) {
// we are in DISTRIBUTE construct
schedtype += kmp_sch_static -
kmp_distribute_static; // AC: convert to usual schedule type
tid = th->th.th_team->t.t_master_tid;
team = th->th.th_team->t.t_parent;
} else
#endif
{
tid = __kmp_tid_from_gtid(global_tid);
team = th->th.th_team;
}
/* determine if "for" loop is an active worksharing construct */
if (team->t.t_serialized) {
/* serialized parallel, each thread executes whole iteration space */
if (plastiter != NULL)
*plastiter = TRUE;
/* leave pupper and plower set to entire iteration space */
*pstride =
(incr > 0) ? (*pupper - *plower + 1) : (-(*plower - *pupper + 1));
#ifdef KMP_DEBUG
{
char *buff;
// create format specifiers before the debug output
buff = __kmp_str_format("__kmpc_for_static_init: (serial) liter=%%d "
"lower=%%%s upper=%%%s stride = %%%s\n",
traits_t<T>::spec, traits_t<T>::spec,
traits_t<ST>::spec);
KD_TRACE(100, (buff, *plastiter, *plower, *pupper, *pstride));
__kmp_str_free(&buff);
}
#endif
KE_TRACE(10, ("__kmpc_for_static_init: T#%d return\n", global_tid));
#if OMPT_SUPPORT && OMPT_OPTIONAL
if (ompt_enabled.ompt_callback_work) {
ompt_callbacks.ompt_callback(ompt_callback_work)(
ompt_work_type, ompt_scope_begin, &(team_info->parallel_data),
&(task_info->task_data), *pstride, codeptr);
}
#endif
return;
}
nth = team->t.t_nproc;
if (nth == 1) {
if (plastiter != NULL)
*plastiter = TRUE;
*pstride =
(incr > 0) ? (*pupper - *plower + 1) : (-(*plower - *pupper + 1));
#ifdef KMP_DEBUG
{
char *buff;
// create format specifiers before the debug output
buff = __kmp_str_format("__kmpc_for_static_init: (serial) liter=%%d "
"lower=%%%s upper=%%%s stride = %%%s\n",
traits_t<T>::spec, traits_t<T>::spec,
traits_t<ST>::spec);
KD_TRACE(100, (buff, *plastiter, *plower, *pupper, *pstride));
__kmp_str_free(&buff);
}
#endif
KE_TRACE(10, ("__kmpc_for_static_init: T#%d return\n", global_tid));
#if OMPT_SUPPORT && OMPT_OPTIONAL
if (ompt_enabled.ompt_callback_work) {
ompt_callbacks.ompt_callback(ompt_callback_work)(
ompt_work_type, ompt_scope_begin, &(team_info->parallel_data),
&(task_info->task_data), *pstride, codeptr);
}
#endif
return;
}
/* compute trip count */
if (incr == 1) {
trip_count = *pupper - *plower + 1;
} else if (incr == -1) {
trip_count = *plower - *pupper + 1;
} else if (incr > 0) {
// upper-lower can exceed the limit of signed type
trip_count = (UT)(*pupper - *plower) / incr + 1;
} else {
trip_count = (UT)(*plower - *pupper) / (-incr) + 1;
}
if (__kmp_env_consistency_check) {
/* tripcount overflow? */
if (trip_count == 0 && *pupper != *plower) {
__kmp_error_construct(kmp_i18n_msg_CnsIterationRangeTooLarge, ct_pdo,
loc);
}
}
KMP_COUNT_VALUE(FOR_static_iterations, trip_count);
/* compute remaining parameters */
switch (schedtype) {
case kmp_sch_static: {
if (trip_count < nth) {
KMP_DEBUG_ASSERT(
__kmp_static == kmp_sch_static_greedy ||
__kmp_static ==
kmp_sch_static_balanced); // Unknown static scheduling type.
if (tid < trip_count) {
*pupper = *plower = *plower + tid * incr;
} else {
*plower = *pupper + incr;
}
if (plastiter != NULL)
*plastiter = (tid == trip_count - 1);
} else {
if (__kmp_static == kmp_sch_static_balanced) {
UT small_chunk = trip_count / nth;
UT extras = trip_count % nth;
*plower += incr * (tid * small_chunk + (tid < extras ? tid : extras));
*pupper = *plower + small_chunk * incr - (tid < extras ? 0 : incr);
if (plastiter != NULL)
*plastiter = (tid == nth - 1);
} else {
T big_chunk_inc_count =
(trip_count / nth + ((trip_count % nth) ? 1 : 0)) * incr;
T old_upper = *pupper;
KMP_DEBUG_ASSERT(__kmp_static == kmp_sch_static_greedy);
// Unknown static scheduling type.
*plower += tid * big_chunk_inc_count;
*pupper = *plower + big_chunk_inc_count - incr;
if (incr > 0) {
if (*pupper < *plower)
*pupper = traits_t<T>::max_value;
if (plastiter != NULL)
*plastiter = *plower <= old_upper && *pupper > old_upper - incr;
if (*pupper > old_upper)
*pupper = old_upper; // tracker C73258
} else {
if (*pupper > *plower)
*pupper = traits_t<T>::min_value;
if (plastiter != NULL)
*plastiter = *plower >= old_upper && *pupper < old_upper - incr;
if (*pupper < old_upper)
*pupper = old_upper; // tracker C73258
}
}
}
*pstride = trip_count;
break;
}
case kmp_sch_static_chunked: {
ST span;
if (chunk < 1) {
chunk = 1;
}
span = chunk * incr;
*pstride = span * nth;
*plower = *plower + (span * tid);
*pupper = *plower + span - incr;
if (plastiter != NULL)
*plastiter = (tid == ((trip_count - 1) / (UT)chunk) % nth);
break;
}
#if OMP_45_ENABLED
case kmp_sch_static_balanced_chunked: {
T old_upper = *pupper;
// round up to make sure the chunk is enough to cover all iterations
UT span = (trip_count + nth - 1) / nth;
// perform chunk adjustment
chunk = (span + chunk - 1) & ~(chunk - 1);
span = chunk * incr;
*plower = *plower + (span * tid);
*pupper = *plower + span - incr;
if (incr > 0) {
if (*pupper > old_upper)
*pupper = old_upper;
} else if (*pupper < old_upper)
*pupper = old_upper;
if (plastiter != NULL)
*plastiter = (tid == ((trip_count - 1) / (UT)chunk));
break;
}
#endif
default:
KMP_ASSERT2(0, "__kmpc_for_static_init: unknown scheduling type");
break;
}
#if USE_ITT_BUILD
// Report loop metadata
if (KMP_MASTER_TID(tid) && __itt_metadata_add_ptr &&
__kmp_forkjoin_frames_mode == 3 &&
#if OMP_40_ENABLED
th->th.th_teams_microtask == NULL &&
#endif
team->t.t_active_level == 1) {
kmp_uint64 cur_chunk = chunk;
// Calculate chunk in case it was not specified; it is specified for
// kmp_sch_static_chunked
if (schedtype == kmp_sch_static) {
cur_chunk = trip_count / nth + ((trip_count % nth) ? 1 : 0);
}
// 0 - "static" schedule
__kmp_itt_metadata_loop(loc, 0, trip_count, cur_chunk);
}
#endif
#ifdef KMP_DEBUG
{
char *buff;
// create format specifiers before the debug output
buff = __kmp_str_format("__kmpc_for_static_init: liter=%%d lower=%%%s "
"upper=%%%s stride = %%%s signed?<%s>\n",
traits_t<T>::spec, traits_t<T>::spec,
traits_t<ST>::spec, traits_t<T>::spec);
KD_TRACE(100, (buff, *plastiter, *plower, *pupper, *pstride));
__kmp_str_free(&buff);
}
#endif
KE_TRACE(10, ("__kmpc_for_static_init: T#%d return\n", global_tid));
#if OMPT_SUPPORT && OMPT_OPTIONAL
if (ompt_enabled.ompt_callback_work) {
ompt_callbacks.ompt_callback(ompt_callback_work)(
ompt_work_type, ompt_scope_begin, &(team_info->parallel_data),
&(task_info->task_data), trip_count, codeptr);
}
#endif
return;
}