Add new OpenMP 4.5 affinity API

This change introduces the new OpenMP 4.5 affinity api surrounding
OpenMP Places. There are six new entry points:

Typically called in serial region:
 * omp_get_num_places - returns the number of places available to the execution
       environment in the place list.
 * omp_get_place_num_procs - returns the number of processors available to the
       execution environment in the specified place.
 * omp_get_place_proc_ids - returns the numerical identifiers of the processors
       available to the execution environment in the specified place.

Typically called inside parallel region:
 * omp_get_place_num - returns the place number of the place to which the
       encountering thread is bound.
 * omp_get_partition_num_places - returns the number of places in the place
       partition of the innermost implicit task.
 * omp_get_partition_place_nums - returns the list of place numbers
       corresponding to the places in the place-var ICV of the innermost
       implicit task.

Differential Revision: http://reviews.llvm.org/D17417

llvm-svn: 261915

8 files changed, 274 insertions, 0 deletions

diff --git a/openmp/runtime/src/dllexports b/openmp/runtime/src/dllexports
index f14da7079e1..a8a70c0a364 100644
--- a/openmp/runtime/src/dllexports
+++ b/openmp/runtime/src/dllexports
@@ -498,6 +498,12 @@ kmp_set_warnings_off                        780
     omp_init_lock_with_hint                 870
     omp_init_nest_lock_with_hint            871
     omp_get_max_task_priority               872
+    omp_get_num_places                      873
+    omp_get_place_num_procs                 874
+    omp_get_place_proc_ids                  875
+    omp_get_place_num                       876
+    omp_get_partition_num_places            877
+    omp_get_partition_place_nums            878
 %endif # OMP_41
 
 %ifndef stub
diff --git a/openmp/runtime/src/include/41/omp.h.var b/openmp/runtime/src/include/41/omp.h.var
index 4597d0eaa81..9b7ac71b1a9 100644
--- a/openmp/runtime/src/include/41/omp.h.var
+++ b/openmp/runtime/src/include/41/omp.h.var
@@ -153,6 +153,14 @@
 
     extern omp_proc_bind_t __KAI_KMPC_CONVENTION omp_get_proc_bind (void);
 
+    /* OpenMP 4.5 affinity API */
+    extern int  __KAI_KMPC_CONVENTION omp_get_num_places (void);
+    extern int  __KAI_KMPC_CONVENTION omp_get_place_num_procs (int);
+    extern void __KAI_KMPC_CONVENTION omp_get_place_proc_ids (int, int *);
+    extern int  __KAI_KMPC_CONVENTION omp_get_place_num (void);
+    extern int  __KAI_KMPC_CONVENTION omp_get_partition_num_places (void);
+    extern void __KAI_KMPC_CONVENTION omp_get_partition_place_nums (int *);
+
     extern void * __KAI_KMPC_CONVENTION  kmp_malloc  (size_t);
     extern void * __KAI_KMPC_CONVENTION  kmp_calloc  (size_t, size_t);
     extern void * __KAI_KMPC_CONVENTION  kmp_realloc (void *, size_t);
diff --git a/openmp/runtime/src/include/41/omp_lib.f.var b/openmp/runtime/src/include/41/omp_lib.f.var
index d3406e00601..fe705ad65af 100644
--- a/openmp/runtime/src/include/41/omp_lib.f.var
+++ b/openmp/runtime/src/include/41/omp_lib.f.var
@@ -180,6 +180,38 @@
             integer (kind=omp_proc_bind_kind) omp_get_proc_bind
           end function omp_get_proc_bind
 
+          function omp_get_num_places()
+            use omp_lib_kinds
+            integer (kind=omp_integer_kind) omp_get_num_places
+          end function omp_get_num_places
+
+          function omp_get_place_num_procs(place_num)
+            use omp_lib_kinds
+            integer (kind=omp_integer_kind) place_num
+            integer (kind=omp_integer_kind) omp_get_place_num_procs
+          end function omp_get_place_num_procs
+
+          subroutine omp_get_place_proc_ids(place_num, ids)
+            use omp_lib_kinds
+            integer (kind=omp_integer_kind) place_num
+            integer (kind=kmp_pointer_kind) ids
+          end subroutine omp_get_place_proc_ids
+
+          function omp_get_place_num()
+            use omp_lib_kinds
+            integer (kind=omp_integer_kind) omp_get_place_num
+          end function omp_get_place_num
+
+          function omp_get_partition_num_places()
+            use omp_lib_kinds
+            integer (kind=omp_integer_kind) omp_get_partition_num_places
+          end function omp_get_partition_num_places
+
+          subroutine omp_get_partition_place_nums(place_nums)
+            use omp_lib_kinds
+            integer (kind=kmp_pointer_kind) place_nums
+          end subroutine omp_get_partition_place_nums
+
           function omp_get_wtime()
             double precision omp_get_wtime
           end function omp_get_wtime
diff --git a/openmp/runtime/src/include/41/omp_lib.f90.var b/openmp/runtime/src/include/41/omp_lib.f90.var
index 18770e4c979..737a68e504b 100644
--- a/openmp/runtime/src/include/41/omp_lib.f90.var
+++ b/openmp/runtime/src/include/41/omp_lib.f90.var
@@ -183,6 +183,38 @@
             integer (kind=omp_proc_bind_kind) omp_get_proc_bind
           end function omp_get_proc_bind
 
+          function omp_get_num_places() bind(c)
+            use omp_lib_kinds
+            integer (kind=omp_integer_kind) omp_get_num_places
+          end function omp_get_num_places
+
+          function omp_get_place_num_procs(place_num) bind(c)
+            use omp_lib_kinds
+            integer (kind=omp_integer_kind), value :: place_num
+            integer (kind=omp_integer_kind) omp_get_place_num_procs
+          end function omp_get_place_num_procs
+
+          subroutine omp_get_place_proc_ids(place_num, ids) bind(c)
+            use omp_lib_kinds
+            integer (kind=omp_integer_kind), value :: place_num
+            integer (kind=kmp_pointer_kind) ids
+          end subroutine omp_get_place_proc_ids
+
+          function omp_get_place_num() bind(c)
+            use omp_lib_kinds
+            integer (kind=omp_integer_kind) omp_get_place_num
+          end function omp_get_place_num
+
+          function omp_get_partition_num_places() bind(c)
+            use omp_lib_kinds
+            integer (kind=omp_integer_kind) omp_get_partition_num_places
+          end function omp_get_partition_num_places
+
+          subroutine omp_get_partition_place_nums(place_nums) bind(c)
+            use omp_lib_kinds
+            integer (kind=kmp_pointer_kind) place_nums
+          end subroutine omp_get_partition_place_nums
+
           function omp_get_wtime() bind(c)
             use omp_lib_kinds
             real (kind=kmp_double_kind) omp_get_wtime
diff --git a/openmp/runtime/src/include/41/omp_lib.h.var b/openmp/runtime/src/include/41/omp_lib.h.var
index fbdd8484f26..a91569dfa46 100644
--- a/openmp/runtime/src/include/41/omp_lib.h.var
+++ b/openmp/runtime/src/include/41/omp_lib.h.var
@@ -172,6 +172,38 @@
           integer (kind=omp_proc_bind_kind) omp_get_proc_bind
         end function omp_get_proc_bind
 
+        function omp_get_num_places() bind(c)
+          import
+          integer (kind=omp_integer_kind) omp_get_num_places
+        end function omp_get_num_places
+
+        function omp_get_place_num_procs(place_num) bind(c)
+          import
+          integer (kind=omp_integer_kind), value :: place_num
+          integer (kind=omp_integer_kind) omp_get_place_num_procs
+        end function omp_get_place_num_procs
+
+        subroutine omp_get_place_proc_ids(place_num, ids) bind(c)
+          import
+          integer (kind=omp_integer_kind), value :: place_num
+          integer (kind=kmp_pointer_kind) ids
+        end subroutine omp_get_place_proc_ids
+
+        function omp_get_place_num() bind(c)
+          import
+          integer (kind=omp_integer_kind) omp_get_place_num
+        end function omp_get_place_num
+
+        function omp_get_partition_num_places() bind(c)
+          import
+          integer (kind=omp_integer_kind) omp_get_partition_num_places
+        end function omp_get_partition_num_places
+
+        subroutine omp_get_partition_place_nums(place_nums) bind(c)
+          import
+          integer (kind=kmp_pointer_kind) place_nums
+        end subroutine omp_get_partition_place_nums
+
         function omp_get_wtime() bind(c)
           double precision omp_get_wtime
         end function omp_get_wtime
diff --git a/openmp/runtime/src/kmp_ftn_entry.h b/openmp/runtime/src/kmp_ftn_entry.h
index 2711c5d12fa..e654e7f2e9b 100644
--- a/openmp/runtime/src/kmp_ftn_entry.h
+++ b/openmp/runtime/src/kmp_ftn_entry.h
@@ -676,6 +676,139 @@ xexpand(FTN_GET_PROC_BIND)( void )
     #endif
 }
 
+#if OMP_41_ENABLED
+int FTN_STDCALL
+FTN_GET_NUM_PLACES( void )
+{
+    #if defined(KMP_STUB) || !KMP_AFFINITY_SUPPORTED
+        return 0;
+    #else
+        if ( ! TCR_4(__kmp_init_middle) ) {
+            __kmp_middle_initialize();
+        }
+        return __kmp_affinity_num_masks;
+    #endif
+}
+
+int FTN_STDCALL
+FTN_GET_PLACE_NUM_PROCS( int place_num )
+{
+    #if defined(KMP_STUB) || !KMP_AFFINITY_SUPPORTED
+        return 0;
+    #else
+        int i;
+        int retval = 0;
+        if ( ! TCR_4(__kmp_init_middle) ) {
+            __kmp_middle_initialize();
+        }
+        if ( place_num < 0 || place_num >= (int)__kmp_affinity_num_masks )
+            return 0;
+        kmp_affin_mask_t *mask = KMP_CPU_INDEX(__kmp_affinity_masks, place_num);
+        KMP_CPU_SET_ITERATE(i, mask) {
+            if ( !KMP_CPU_ISSET(i, mask) )
+                continue;
+            ++retval;
+        }
+        return retval;
+    #endif
+}
+
+void FTN_STDCALL
+FTN_GET_PLACE_PROC_IDS( int place_num, int *ids )
+{
+    #if defined(KMP_STUB) || !KMP_AFFINITY_SUPPORTED
+    // Nothing.
+    #else
+        int i,j;
+        if ( ! TCR_4(__kmp_init_middle) ) {
+            __kmp_middle_initialize();
+        }
+        if ( place_num < 0 || place_num >= (int)__kmp_affinity_num_masks )
+            return;
+        kmp_affin_mask_t *mask = KMP_CPU_INDEX(__kmp_affinity_masks, place_num);
+        j = 0;
+        KMP_CPU_SET_ITERATE(i, mask) {
+            if ( !KMP_CPU_ISSET(i, mask) )
+                continue;
+            ids[j++] = i;
+        }
+    #endif
+}
+
+int FTN_STDCALL
+FTN_GET_PLACE_NUM( void )
+{
+    #if defined(KMP_STUB) || !KMP_AFFINITY_SUPPORTED
+        return -1;
+    #else
+        int gtid;
+        kmp_info_t *thread;
+        if ( ! TCR_4(__kmp_init_middle) ) {
+            __kmp_middle_initialize();
+        }
+        gtid = __kmp_entry_gtid();
+        thread = __kmp_thread_from_gtid(gtid);
+        if ( thread->th.th_current_place < 0 )
+            return -1;
+        return thread->th.th_current_place;
+    #endif
+}
+
+int FTN_STDCALL
+FTN_GET_PARTITION_NUM_PLACES( void )
+{
+    #if defined(KMP_STUB) || !KMP_AFFINITY_SUPPORTED
+        return 0;
+    #else
+        int gtid, num_places, first_place, last_place;
+        kmp_info_t *thread;
+        if ( ! TCR_4(__kmp_init_middle) ) {
+            __kmp_middle_initialize();
+        }
+        gtid = __kmp_entry_gtid();
+        thread = __kmp_thread_from_gtid(gtid);
+        first_place = thread->th.th_first_place;
+        last_place = thread->th.th_last_place;
+        if ( first_place < 0 || last_place < 0 )
+            return 0;
+        if ( first_place <= last_place )
+            num_places = last_place - first_place + 1;
+        else
+            num_places = __kmp_affinity_num_masks - first_place + last_place + 1;
+        return num_places;
+    #endif
+}
+
+void FTN_STDCALL
+FTN_GET_PARTITION_PLACE_NUMS( int *place_nums ) {
+    #if defined(KMP_STUB) || !KMP_AFFINITY_SUPPORTED
+    // Nothing.
+    #else
+        int i, gtid, place_num, first_place, last_place, start, end;
+        kmp_info_t *thread;
+        if ( ! TCR_4(__kmp_init_middle) ) {
+            __kmp_middle_initialize();
+        }
+        gtid = __kmp_entry_gtid();
+        thread = __kmp_thread_from_gtid(gtid);
+        first_place = thread->th.th_first_place;
+        last_place = thread->th.th_last_place;
+        if ( first_place < 0 || last_place < 0 )
+            return;
+        if ( first_place <= last_place ) {
+            start = first_place;
+            end = last_place;
+        } else {
+            start = last_place;
+            end = first_place;
+        }
+        for (i = 0, place_num = start; place_num <= end; ++place_num, ++i) {
+            place_nums[i] = place_num;
+        }
+    #endif
+}
+#endif
+
 int FTN_STDCALL
 xexpand(FTN_GET_NUM_TEAMS)( void )
 {
diff --git a/openmp/runtime/src/kmp_ftn_os.h b/openmp/runtime/src/kmp_ftn_os.h
index d9157504cbc..04a64a7bd90 100644
--- a/openmp/runtime/src/kmp_ftn_os.h
+++ b/openmp/runtime/src/kmp_ftn_os.h
@@ -115,6 +115,12 @@
 
 #if OMP_41_ENABLED
     #define FTN_GET_MAX_TASK_PRIORITY            omp_get_max_task_priority
+    #define FTN_GET_NUM_PLACES                   omp_get_num_places
+    #define FTN_GET_PLACE_NUM_PROCS              omp_get_place_num_procs
+    #define FTN_GET_PLACE_PROC_IDS               omp_get_place_proc_ids
+    #define FTN_GET_PLACE_NUM                    omp_get_place_num
+    #define FTN_GET_PARTITION_NUM_PLACES         omp_get_partition_num_places
+    #define FTN_GET_PARTITION_PLACE_NUMS         omp_get_partition_place_nums
 #endif
 
 #endif /* KMP_FTN_PLAIN */
@@ -216,6 +222,12 @@
 
 #if OMP_41_ENABLED
     #define FTN_GET_MAX_TASK_PRIORITY            omp_get_max_task_priority_
+    #define FTN_GET_NUM_PLACES                   omp_get_num_places_
+    #define FTN_GET_PLACE_NUM_PROCS              omp_get_place_num_procs_
+    #define FTN_GET_PLACE_PROC_IDS               omp_get_place_proc_ids_
+    #define FTN_GET_PLACE_NUM                    omp_get_place_num_
+    #define FTN_GET_PARTITION_NUM_PLACES         omp_get_partition_num_places_
+    #define FTN_GET_PARTITION_PLACE_NUMS         omp_get_partition_place_nums_
 #endif
 
 #endif /* KMP_FTN_APPEND */
@@ -317,6 +329,12 @@
 
 #if OMP_41_ENABLED
     #define FTN_GET_MAX_TASK_PRIORITY            OMP_GET_MAX_TASK_PRIORITY
+    #define FTN_GET_NUM_PLACES                   OMP_GET_NUM_PLACES
+    #define FTN_GET_PLACE_NUM_PROCS              OMP_GET_PLACE_NUM_PROCS
+    #define FTN_GET_PLACE_PROC_IDS               OMP_GET_PLACE_PROC_IDS
+    #define FTN_GET_PLACE_NUM                    OMP_GET_PLACE_NUM
+    #define FTN_GET_PARTITION_NUM_PLACES         OMP_GET_PARTITION_NUM_PLACES
+    #define FTN_GET_PARTITION_PLACE_NUMS         OMP_GET_PARTITION_PLACE_NUMS
 #endif
 
 #endif /* KMP_FTN_UPPER */
@@ -418,6 +436,12 @@
 
 #if OMP_41_ENABLED
     #define FTN_GET_MAX_TASK_PRIORITY            OMP_GET_MAX_TASK_PRIORITY_
+    #define FTN_GET_NUM_PLACES                   OMP_GET_NUM_PLACES_
+    #define FTN_GET_PLACE_NUM_PROCS              OMP_GET_PLACE_NUM_PROCS_
+    #define FTN_GET_PLACE_PROC_IDS               OMP_GET_PLACE_PROC_IDS_
+    #define FTN_GET_PLACE_NUM                    OMP_GET_PLACE_NUM_
+    #define FTN_GET_PARTITION_NUM_PLACES         OMP_GET_PARTITION_NUM_PLACES_
+    #define FTN_GET_PARTITION_PLACE_NUMS         OMP_GET_PARTITION_PLACE_NUMS_
 #endif
 
 #endif /* KMP_FTN_UAPPEND */
diff --git a/openmp/runtime/src/kmp_runtime.c b/openmp/runtime/src/kmp_runtime.c
index 8801ed86b12..7a2fa7bac55 100644
--- a/openmp/runtime/src/kmp_runtime.c
+++ b/openmp/runtime/src/kmp_runtime.c
@@ -3834,6 +3834,13 @@ __kmp_register_root( int initial_thread )
     KMP_DEBUG_ASSERT( root->r.r_hot_team->t.t_bar[ bs_forkjoin_barrier ].b_arrived == KMP_INIT_BARRIER_STATE );
 
 #if KMP_AFFINITY_SUPPORTED
+# if OMP_40_ENABLED
+    root_thread->th.th_current_place = KMP_PLACE_UNDEFINED;
+    root_thread->th.th_new_place = KMP_PLACE_UNDEFINED;
+    root_thread->th.th_first_place = KMP_PLACE_UNDEFINED;
+    root_thread->th.th_last_place = KMP_PLACE_UNDEFINED;
+# endif
+
     if ( TCR_4(__kmp_init_middle) ) {
         __kmp_affinity_set_init_mask( gtid, TRUE );
     }