DM: Push GPU-parent child tasks to the front of the queue.

include/utils/SkThreadPool.h

 /*
  * Copyright 2012 Google Inc.
  *
  * Use of this source code is governed by a BSD-style license that can be
  * found in the LICENSE file.
  */
 
 #ifndef SkThreadPool_DEFINED
 #define SkThreadPool_DEFINED
 
@@ skipping 32 matching lines @@
     ~SkTThreadPool();
 
     /**
      * Queues up an SkRunnable to run when a thread is available, or synchronously if count is 0.
      * Does not take ownership. NULL is a safe no-op.  If T is not void, the runnable will be passed
      * a reference to a T on the thread's local stack.
      */
     void add(SkTRunnable<T>*);
 
     /**
+     * Same as add, but adds the runnable as the very next to run rather than enqueueing it.
+     */
+    void addNext(SkTRunnable<T>*);
+
+    /**
      * Block until all added SkRunnables have completed.  Once called, calling add() is undefined.
      */
     void wait();
 
  private:
     struct LinkedRunnable {
         SkTRunnable<T>* fRunnable;  // Unowned.
         SK_DECLARE_INTERNAL_LLIST_INTERFACE(LinkedRunnable);
     };
 
     enum State {
         kRunning_State,  // Normal case.  We've been constructed and no one has called wait().
         kWaiting_State,  // wait has been called, but there still might be work to do or being done.
         kHalting_State,  // There's no work to do and no thread is busy.  All threads can shut down.
     };
 
+    void addSomewhere(SkTRunnable<T>* r,
+                      void (SkTInternalLList<LinkedRunnable>::*)(LinkedRunnable*));
+
     SkTInternalLList<LinkedRunnable> fQueue;
     SkCondVar                        fReady;
     SkTDArray<SkThread*>             fThreads;
     State                            fState;
     int                              fBusyThreads;
 
     static void Loop(void*);  // Static because we pass in this.
 };
 
 template <typename T>
@@ skipping 25 matching lines @@
 };
 
 template <>
 struct ThreadLocal<void> {
     void run(SkTRunnable<void>* r) { r->run(); }
 };
 
 }  // namespace SkThreadPoolPrivate
 
 template <typename T>
-void SkTThreadPool<T>::add(SkTRunnable<T>* r) {
+void SkTThreadPool<T>::addSomewhere(SkTRunnable<T>* r,
+                                    void (SkTInternalLList<LinkedRunnable>::* f)(LinkedRunnable*)) {
     if (r == NULL) {
         return;
     }
 
     if (fThreads.isEmpty()) {
         SkThreadPoolPrivate::ThreadLocal<T> threadLocal;
         threadLocal.run(r);
         return;
     }
 
     LinkedRunnable* linkedRunnable = SkNEW(LinkedRunnable);
     linkedRunnable->fRunnable = r;
     fReady.lock();
     SkASSERT(fState != kHalting_State);  // Shouldn't be able to add work when we're halting.
-    fQueue.addToHead(linkedRunnable);
+    (fQueue.*f)(linkedRunnable);
     fReady.signal();
     fReady.unlock();
 }
 
+template <typename T>
+void SkTThreadPool<T>::add(SkTRunnable<T>* r) {
+    this->addSomewhere(r, &SkTInternalLList<LinkedRunnable>::addToTail);
+}
+
+template <typename T>
+void SkTThreadPool<T>::addNext(SkTRunnable<T>* r) {
+    this->addSomewhere(r, &SkTInternalLList<LinkedRunnable>::addToHead);
+}
+
 
 template <typename T>
 void SkTThreadPool<T>::wait() {
     fReady.lock();
     fState = kWaiting_State;
     fReady.broadcast();
     fReady.unlock();
 
     // Wait for all threads to stop.
     for (int i = 0; i < fThreads.count(); i++) {
@@ skipping 23 matching lines @@
             if (kHalting_State == pool->fState) {
                 pool->fReady.unlock();
                 return;
             }
             // wait yields the lock while waiting, but will have it again when awoken.
             pool->fReady.wait();
         }
         // We've got the lock back here, no matter if we ran wait or not.
 
         // The queue is not empty, so we have something to run.  Claim it.
-        LinkedRunnable* r = pool->fQueue.tail();
+        LinkedRunnable* r = pool->fQueue.head();
 
         pool->fQueue.remove(r);
 
         // Having claimed our SkRunnable, we now give up the lock while we run it.
         // Otherwise, we'd only ever do work on one thread at a time, which rather
         // defeats the point of this code.
         pool->fBusyThreads++;
         pool->fReady.unlock();
 
         // OK, now really do the work.
         threadLocal.run(r->fRunnable);
         SkDELETE(r);
 
         // Let everyone know we're not busy.
         pool->fReady.lock();
         pool->fBusyThreads--;
         pool->fReady.unlock();
     }
 
     SkASSERT(false); // Unreachable.  The only exit happens when pool->fState is kHalting_State.
 }
 
 typedef SkTThreadPool<void> SkThreadPool;
 
 #endif