BLI_task: Add new 'BLI_task_parallel_range_finalize()'.

Together with the extended loop callback and userdata_chunk, this allows to perform
cumulative tasks (like aggregation) in a lockfree way using local userdata_chunk to store temp data,
and once all workers have finished, to merge those userdata_chunks in the finalize callback
(from calling thread, so no need to lock here either).

Note that this changes how userdata_chunk is handled (now fully from 'main' thread,
which means a given worker thread will always get the same userdata_chunk, without
being re-initialized anymore to init value at start of each iter chunk).

1 files changed, 13 insertions, 1 deletions

diff --git a/source/blender/blenlib/BLI_task.h b/source/blender/blenlib/BLI_task.h
index c511ec432ee..967e0be6d0a 100644
--- a/source/blender/blenlib/BLI_task.h
+++ b/source/blender/blenlib/BLI_task.h
@@ -119,11 +119,13 @@ size_t BLI_task_pool_tasks_done(TaskPool *pool);
 /* Parallel for routines */
 typedef void (*TaskParallelRangeFunc)(void *userdata, const int iter);
 typedef void (*TaskParallelRangeFuncEx)(void *userdata, void *userdata_chunk, const int iter, const int thread_id);
+typedef void (*TaskParallelRangeFuncFinalize)(void *userdata, void *userdata_chunk);
 void BLI_task_parallel_range_ex(
         int start, int stop,
         void *userdata,
         void *userdata_chunk,
-        const size_t userdata_chunk_size, TaskParallelRangeFuncEx func_ex,
+        const size_t userdata_chunk_size,
+        TaskParallelRangeFuncEx func_ex,
         const bool use_threading,
         const bool use_dynamic_scheduling);
 void BLI_task_parallel_range(
@@ -132,6 +134,16 @@ void BLI_task_parallel_range(
         TaskParallelRangeFunc func,
         const bool use_threading);
 
+void BLI_task_parallel_range_finalize(
+        int start, int stop,
+        void *userdata,
+        void *userdata_chunk,
+        const size_t userdata_chunk_size,
+        TaskParallelRangeFuncEx func_ex,
+        TaskParallelRangeFuncFinalize func_finalize,
+        const bool use_threading,
+        const bool use_dynamic_scheduling);
+
 typedef void (*TaskParallelListbaseFunc)(void *userdata,
                                          struct Link *iter,
                                          int index);