X-Git-Url: http://mj.ucw.cz/gitweb/?a=blobdiff_plain;f=lib%2Fsorter%2Farray.c;h=dd7659e70f04a2ccc18da447f6a9bdb269070eab;hb=a5ff98a53789157a6c96e58b2385bb898d688a22;hp=a6c217e8b2dca296f86edc5a79cef8630f71508c;hpb=f1b56d292bbe248bf29c078d27986cdfbb1b3d39;p=libucw.git

diff --git a/lib/sorter/array.c b/lib/sorter/array.c
index a6c217e8..dd7659e7 100644
--- a/lib/sorter/array.c
+++ b/lib/sorter/array.c
@@ -13,17 +13,22 @@
 #include "lib/sorter/common.h"
 
 #include <string.h>
+#include <alloca.h>
 
-#define ASORT_MIN_RADIX 5000		// FIXME: var?
 #define ASORT_MIN_SHIFT 2
 
 static void
 asort_radix(struct asort_context *ctx, void *array, void *buffer, uns num_elts, uns hash_bits, uns swapped_output)
 {
+  // swap_output == 0 if result should be returned in `array', otherwise in `buffer'
   uns buckets = (1 << ctx->radix_bits);
   uns shift = (hash_bits > ctx->radix_bits) ? (hash_bits - ctx->radix_bits) : 0;
   uns cnt[buckets];
 
+#if 0
+  static int reported[64];
+  if (!reported[hash_bits]++)
+#endif
   DBG(">>> n=%d h=%d s=%d sw=%d", num_elts, hash_bits, shift, swapped_output);
 
   bzero(cnt, sizeof(cnt));
@@ -43,7 +48,7 @@ asort_radix(struct asort_context *ctx, void *array, void *buffer, uns num_elts,
   for (uns i=0; i<buckets; i++)
     {
       uns n = cnt[i] - pos;
-      if (n < ASORT_MIN_RADIX || shift < ASORT_MIN_SHIFT)
+      if (n * ctx->elt_size < sorter_radix_threshold || shift < ASORT_MIN_SHIFT)
 	{
 	  ctx->quicksort(buffer, n);
 	  if (!swapped_output)
@@ -57,24 +62,387 @@ asort_radix(struct asort_context *ctx, void *array, void *buffer, uns num_elts,
     }
 }
 
+#ifdef CONFIG_UCW_THREADS
+
+#include "lib/threads.h"
+#include "lib/workqueue.h"
+#include "lib/eltpool.h"
+
+static uns asort_threads_use_count;
+static uns asort_threads_ready;
+static struct worker_pool asort_thread_pool;
+
+void
+asort_start_threads(uns run)
+{
+  ucwlib_lock();
+  asort_threads_use_count++;
+  if (run && !asort_threads_ready)
+    {
+      SORT_XTRACE(2, "Initializing thread pool (%d threads)", sorter_threads);
+      asort_thread_pool.num_threads = sorter_threads;
+      worker_pool_init(&asort_thread_pool);
+      asort_threads_ready = 1;
+    }
+  ucwlib_unlock();
+}
+
+void
+asort_stop_threads(void)
+{
+  ucwlib_lock();
+  if (!--asort_threads_use_count && asort_threads_ready)
+    {
+      SORT_XTRACE(2, "Shutting down thread pool");
+      worker_pool_cleanup(&asort_thread_pool);
+      asort_threads_ready = 0;
+    }
+  ucwlib_unlock();
+}
+
+struct qs_work {
+  struct work w;
+  struct asort_context *ctx;
+  void *array;
+  uns num_elts;
+  int left, right;
+#define LR_UNDEF -100
+};
+
+static void
+qs_handle_work(struct worker_thread *thr UNUSED, struct work *ww)
+{
+  struct qs_work *w = (struct qs_work *) ww;
+
+  DBG("Thread %d: got %d elts", thr->id, w->num_elts);
+  if (w->num_elts * w->ctx->elt_size < sorter_thread_threshold)
+    {
+      w->ctx->quicksort(w->array, w->num_elts);
+      w->left = w->right = LR_UNDEF;
+    }
+  else
+    w->ctx->quicksplit(w->array, w->num_elts, &w->left, &w->right);
+  DBG("Thread %d: returning l=%d r=%d", thr->id, w->left, w->right);
+}
+
+static struct qs_work *
+qs_alloc_work(struct asort_context *ctx)
+{
+  struct qs_work *w = ep_alloc(ctx->eltpool);
+  w->w.priority = 0;
+  w->w.go = qs_handle_work;
+  w->ctx = ctx;
+  return w;
+}
+
+static void
+threaded_quicksort(struct asort_context *ctx)
+{
+  struct work_queue q;
+  struct qs_work *v, *w;
+
+  asort_start_threads(1);
+  work_queue_init(&asort_thread_pool, &q);
+  ctx->eltpool = ep_new(sizeof(struct qs_work), 1000);
+
+  w = qs_alloc_work(ctx);
+  w->array = ctx->array;
+  w->num_elts = ctx->num_elts;
+  work_submit(&q, &w->w);
+
+  while (v = (struct qs_work *) work_wait(&q))
+    {
+      if (v->left != LR_UNDEF)
+	{
+	  if (v->right > 0)
+	    {
+	      w = qs_alloc_work(ctx);
+	      w->array = v->array;
+	      w->num_elts = v->right + 1;
+	      w->w.priority = v->w.priority + 1;
+	      work_submit(&q, &w->w);
+	    }
+	  if (v->left < (int)v->num_elts - 1)
+	    {
+	      w = qs_alloc_work(ctx);
+	      w->array = v->array + v->left * ctx->elt_size;
+	      w->num_elts = v->num_elts - v->left;
+	      w->w.priority = v->w.priority + 1;
+	      work_submit(&q, &w->w);
+	    }
+	}
+      ep_free(ctx->eltpool, v);
+    }
+
+  ep_delete(ctx->eltpool);
+  work_queue_cleanup(&q);
+  asort_stop_threads();
+}
+
+struct rs_work {
+  struct work w;
+  struct asort_context *ctx;
+  void *array, *buffer;		// Like asort_radix().
+  uns num_elts;
+  uns shift;
+  uns swap_output;
+  uns cnt[0];
+};
+
+static void
+rs_count(struct worker_thread *thr UNUSED, struct work *ww)
+{
+  struct rs_work *w = (struct rs_work *) ww;
+
+  DBG("Thread %d: Counting %d items, shift=%d", thr->id, w->num_elts, w->shift);
+  w->ctx->radix_count(w->array, w->num_elts, w->cnt, w->shift);
+  DBG("Thread %d: Counting done", thr->id);
+}
+
+static void
+rs_split(struct worker_thread *thr UNUSED, struct work *ww)
+{
+  struct rs_work *w = (struct rs_work *) ww;
+
+  DBG("Thread %d: Splitting %d items, shift=%d", thr->id, w->num_elts, w->shift);
+  w->ctx->radix_split(w->array, w->buffer, w->num_elts, w->cnt, w->shift);
+  DBG("Thread %d: Splitting done", thr->id);
+}
+
+static void
+rs_finish(struct worker_thread *thr UNUSED, struct work *ww)
+{
+  struct rs_work *w = (struct rs_work *) ww;
+
+  if (thr)
+    DBG("Thread %d: Finishing %d items, shift=%d", thr->id, w->num_elts, w->shift);
+  if (w->shift < ASORT_MIN_SHIFT || w->num_elts * w->ctx->elt_size < sorter_radix_threshold)
+    {
+      w->ctx->quicksort(w->array, w->num_elts);
+      if (w->swap_output)
+	memcpy(w->buffer, w->array, w->num_elts * w->ctx->elt_size);
+    }
+  else
+    asort_radix(w->ctx, w->array, w->buffer, w->num_elts, w->shift, w->swap_output);
+  if (thr)
+    DBG("Thread %d: Finishing done", thr->id);
+}
+
+static void
+rs_wait_small(struct asort_context *ctx)
+{
+  struct rs_work *w;
+
+  while (w = (struct rs_work *) work_wait(ctx->rs_work_queue))
+    {
+      DBG("Reaping small chunk of %d items", w->num_elts);
+      ep_free(ctx->eltpool, w);
+    }
+}
+
+static void
+rs_radix(struct asort_context *ctx, void *array, void *buffer, uns num_elts, uns hash_bits, uns swapped_output)
+{
+  uns buckets = (1 << ctx->radix_bits);
+  uns shift = (hash_bits > ctx->radix_bits) ? (hash_bits - ctx->radix_bits) : 0;
+  uns cnt[buckets];
+  uns blksize = num_elts / sorter_threads;
+  DBG(">>> n=%d h=%d s=%d blk=%d sw=%d", num_elts, hash_bits, shift, blksize, swapped_output);
+
+  // If there are any small chunks in progress, wait for them to finish
+  rs_wait_small(ctx);
+
+  // Start parallel counting
+  void *iptr = array;
+  for (uns i=0; i<sorter_threads; i++)
+    {
+      struct rs_work *w = ctx->rs_works[i];
+      w->w.priority = 0;
+      w->w.go = rs_count;
+      w->ctx = ctx;
+      w->array = iptr;
+      w->buffer = buffer;
+      w->num_elts = blksize;
+      if (i == sorter_threads-1)
+	w->num_elts += num_elts % sorter_threads;
+      w->shift = shift;
+      iptr += w->num_elts * ctx->elt_size;
+      bzero(w->cnt, sizeof(uns) * buckets);
+      work_submit(ctx->rs_work_queue, &w->w);
+    }
+
+  // Get bucket sizes from the counts
+  bzero(cnt, sizeof(cnt));
+  for (uns i=0; i<sorter_threads; i++)
+    {
+      struct rs_work *w = (struct rs_work *) work_wait(ctx->rs_work_queue);
+      ASSERT(w);
+      for (uns j=0; j<buckets; j++)
+	cnt[j] += w->cnt[j];
+    }
+
+  // Calculate bucket starts
+  uns pos = 0;
+  for (uns i=0; i<buckets; i++)
+    {
+      uns j = cnt[i];
+      cnt[i] = pos;
+      pos += j;
+    }
+  ASSERT(pos == num_elts);
+
+  // Start parallel splitting
+  for (uns i=0; i<sorter_threads; i++)
+    {
+      struct rs_work *w = ctx->rs_works[i];
+      w->w.go = rs_split;
+      for (uns j=0; j<buckets; j++)
+	{
+	  uns k = w->cnt[j];
+	  w->cnt[j] = cnt[j];
+	  cnt[j] += k;
+	}
+      work_submit(ctx->rs_work_queue, &w->w);
+    }
+  ASSERT(cnt[buckets-1] == num_elts);
+
+  // Wait for splits to finish
+  while (work_wait(ctx->rs_work_queue))
+    ;
+
+  // Recurse on buckets
+  pos = 0;
+  for (uns i=0; i<buckets; i++)
+    {
+      uns n = cnt[i] - pos;
+      if (!n)
+	continue;
+      if (n * ctx->elt_size < sorter_thread_threshold)
+	{
+	  struct rs_work *w = ep_alloc(ctx->eltpool);
+	  w->w.priority = 0;
+	  w->w.go = rs_finish;
+	  w->ctx = ctx;
+	  w->array = buffer;
+	  w->buffer = array;
+	  w->num_elts = n;
+	  w->shift = shift;
+	  w->swap_output = !swapped_output;
+	  if (n * ctx->elt_size < sorter_thread_chunk)
+	    {
+	      DBG("Sorting block %d+%d inline", pos, n);
+	      rs_finish(NULL, &w->w);
+	      ep_free(ctx->eltpool, w);
+	    }
+	  else
+	    {
+	      DBG("Scheduling block %d+%d", pos, n);
+	      work_submit(ctx->rs_work_queue, &w->w);
+	    }
+	}
+      else
+	rs_radix(ctx, buffer, array, n, shift, !swapped_output);
+      pos = cnt[i];
+      array += n * ctx->elt_size;
+      buffer += n * ctx->elt_size;
+    }
+}
+
+static void
+threaded_radixsort(struct asort_context *ctx, uns swap)
+{
+  struct work_queue q;
+
+  asort_start_threads(1);
+  work_queue_init(&asort_thread_pool, &q);
+
+  // Prepare work structures for counting and splitting.
+  // We use big_alloc(), because we want to avoid cacheline aliasing between threads.
+  ctx->rs_work_queue = &q;
+  ctx->rs_works = alloca(sizeof(struct rs_work *) * sorter_threads);
+  for (uns i=0; i<sorter_threads; i++)
+    ctx->rs_works[i] = big_alloc(sizeof(struct rs_work) + sizeof(uns) * (1 << ctx->radix_bits));
+
+  // Prepare a pool for all remaining small bits which will be sorted on background.
+  ctx->eltpool = ep_new(sizeof(struct rs_work), 1000);
+
+  // Do the big splitting
+  rs_radix(ctx, ctx->array, ctx->buffer, ctx->num_elts, ctx->hash_bits, swap);
+  for (uns i=0; i<sorter_threads; i++)
+    big_free(ctx->rs_works[i], sizeof(struct rs_work) + sizeof(uns) * (1 << ctx->radix_bits));
+
+  // Finish the small blocks
+  rs_wait_small(ctx);
+
+  ASSERT(!ctx->eltpool->num_allocated);
+  ep_delete(ctx->eltpool);
+  work_queue_cleanup(&q);
+  asort_stop_threads();
+}
+
+#else
+
+void asort_start_threads(uns run UNUSED) { }
+void asort_stop_threads(void) { }
+
+#endif
+
+static uns
+predict_swap(struct asort_context *ctx)
+{
+  uns bits = ctx->radix_bits;
+  uns elts = ctx->num_elts;
+  uns swap = 0;
+
+  while (elts * ctx->elt_size >= sorter_radix_threshold && bits >= ASORT_MIN_SHIFT)
+    {
+      DBG("Predicting pass: %d elts, %d bits", elts, bits);
+      swap = !swap;
+      elts >>= ctx->radix_bits;
+      bits = MAX(bits, ctx->radix_bits) - ctx->radix_bits;
+    }
+  return swap;
+}
+
 void
 asort_run(struct asort_context *ctx)
 {
   SORT_XTRACE(10, "Array-sorting %d items per %d bytes, hash_bits=%d", ctx->num_elts, ctx->elt_size, ctx->hash_bits);
+  uns allow_threads UNUSED = (sorter_threads > 1 &&
+			      ctx->num_elts * ctx->elt_size >= sorter_thread_threshold &&
+			      !(sorter_debug & SORT_DEBUG_ASORT_NO_THREADS));
 
-  if (ctx->num_elts < ASORT_MIN_RADIX ||
+  if (ctx->num_elts * ctx->elt_size < sorter_radix_threshold ||
       ctx->hash_bits <= ASORT_MIN_SHIFT ||
       !ctx->radix_split ||
       (sorter_debug & SORT_DEBUG_ASORT_NO_RADIX))
     {
-      SORT_XTRACE(12, "Decided to use direct quicksort");
+#ifdef CONFIG_UCW_THREADS
+      if (allow_threads)
+	{
+	  SORT_XTRACE(12, "Decided to use parallel quicksort");
+	  threaded_quicksort(ctx);
+	  return;
+	}
+#endif
+      SORT_XTRACE(12, "Decided to use sequential quicksort");
       ctx->quicksort(ctx->array, ctx->num_elts);
     }
   else
     {
-      SORT_XTRACE(12, "Decided to use radix-sort");
-      // FIXME: select dest buffer
-      asort_radix(ctx, ctx->array, ctx->buffer, ctx->num_elts, ctx->hash_bits, 0);
+      uns swap = predict_swap(ctx);
+#ifdef CONFIG_UCW_THREADS
+      if (allow_threads)
+	{
+	  SORT_XTRACE(12, "Decided to use parallel radix-sort (swap=%d)", swap);
+	  threaded_radixsort(ctx, swap);
+	  return;
+	}
+#endif
+      SORT_XTRACE(12, "Decided to use sequential radix-sort (swap=%d)", swap);
+      asort_radix(ctx, ctx->array, ctx->buffer, ctx->num_elts, ctx->hash_bits, swap);
+      if (swap)
+	ctx->array = ctx->buffer;
     }
 
   SORT_XTRACE(11, "Array-sort finished");