Caching for layouts and harfbuzz faces

This patch adds caching for both layouts and for HarfBuzz face objects.
The granularity of the cache for layouts is words, so it splits the
input string at word boundaries (using a heuristic). There are is also
some refactoring to reduce the amount of allocation and copying, and
movement towards properly supporting contexts.

The size of the caches is a fixed number of entries; thus, it is
possible to consume a large amount of memory by filling the cache with
lots of large strings. This should be refined towards a scheme that
bounds the total memory used by the cache.

This patch fixes bug 15237293 "Regression: Measure performance is
significantly slower with minikin".

Change-Id: Ie8176857e2d78656ce5479a7c04969819ef2718d

1 files changed, 10 insertions, 1 deletions

diff --git a/include/minikin/FontCollection.h b/include/minikin/FontCollection.h
index dc48f8e..c6c2ed0 100644
--- a/include/minikin/FontCollection.h
+++ b/include/minikin/FontCollection.h
@@ -54,9 +54,12 @@ public:
         int start;
         int end;
     };
+
     void itemize(const uint16_t *string, size_t string_length, FontStyle style,
             std::vector<Run>* result) const;
-    private:
+
+    uint32_t getId() const;
+private:
     static const int kLogCharsPerPage = 8;
     static const int kPageMask = (1 << kLogCharsPerPage) - 1;
 
@@ -70,6 +73,12 @@ public:
         size_t end;
     };
 
+    // static for allocating unique id's
+    static uint32_t sNextId;
+
+    // unique id for this font collection (suitable for cache key)
+    uint32_t mId;
+
     // Highest UTF-32 code point that can be mapped
     uint32_t mMaxChar;