tag: added a pool for tag items

The new source tag_pool.c manages a pool of reference counted tag_item
objects.  This is used to merge tag items of the same type and value,
saving lots of memory.  Formerly, only the value itself was pooled,
wasting memory for all the pointers and tag_item structs.

The following results were measured with massif.  Started MPD on
amd64, typed "mpc", no song being played.  My music database contains
35k tagged songs.  The results are what massif reports as "peak".

 0.13.2:     total 14,131,392; useful 11,408,972; extra 2,722,420
 eric:       total 18,370,696; useful 15,648,182; extra 2,722,514
 mk f34f694: total 15,833,952; useful 13,111,470; extra 2,722,482
 mk now:     total 12,837,632; useful 10,626,383; extra 2,211,249

This patch set saves 20% memory, and does a good job in reducing heap
fragmentation.

1 files changed, 4 insertions, 5 deletions

diff --git a/src/tag.c b/src/tag.c
index 9192cb171..f99375110 100644
--- a/src/tag.c
+++ b/src/tag.c
@@ -17,6 +17,7 @@
  */
 
 #include "tag.h"
+#include "tag_pool.h"
 #include "myfprintf.h"
 #include "utils.h"
 #include "utf8.h"
@@ -248,7 +249,7 @@ static void deleteItem(struct tag *tag, int idx)
 	assert(idx < tag->numOfItems);
 	tag->numOfItems--;
 
-	free(tag->items[idx]);
+	tag_pool_put_item(tag->items[idx]);
 	/* free(tag->items[idx].value); */
 
 	if (tag->numOfItems - idx > 0) {
@@ -284,7 +285,7 @@ static void clearMpdTag(struct tag *tag)
 
 	for (i = 0; i < tag->numOfItems; i++) {
 		/* free(tag->items[i].value); */
-		free(tag->items[i]);
+		tag_pool_put_item(tag->items[i]);
 	}
 
 	if (tag->items)
@@ -373,9 +374,7 @@ static void appendToTagItems(struct tag *tag, enum tag_type type,
 			      tag->numOfItems * sizeof(*tag->items));
 
 	len = strlen(duplicated);
-	tag->items[i] = xmalloc(sizeof(*tag->items[i]) + len);
-	tag->items[i]->type = type;
-	memcpy(tag->items[i]->value, duplicated, len + 1);
+	tag->items[i] = tag_pool_get_item(type, duplicated, len);
 
 	free(duplicated);
 }