Improve aset.c's space management in contexts with small maxBlockSize.

The previous coding would allow requests up to half of maxBlockSize to be
treated as "chunks", but when that actually did happen, we'd waste nearly
half of the space in the malloc block containing the chunk, if no smaller
requests came along to fill it.  Avoid this scenario by limiting the
maximum size of a chunk to 1/8th maxBlockSize, so that we can waste no more
than 1/8th of the allocated space.  This will not change the behavior at
all for the default context size parameters (with large maxBlockSize),
but it will change the behavior when using ALLOCSET_SMALL_MAXSIZE.

In particular, there's no longer a need for spell.c to be overly concerned
about the request size parameters it uses, so remove a rather unhelpful
comment about that.

Merlin Moncure, per an idea of Tom Lane's

diff --git a/src/backend/tsearch/spell.c b/src/backend/tsearch/spell.c
index 8c0eaa7..be1663c 100644 (file)
--- a/src/backend/tsearch/spell.c
+++ b/src/backend/tsearch/spell.c
@@ -75,7 +75,7 @@ NIFinishBuild(IspellDict *Conf)
  * doesn't need that.  The cpalloc and cpalloc0 macros are just documentation
  * to indicate which allocations actually require zeroing.
  */
-#define COMPACT_ALLOC_CHUNK 8192       /* must be > aset.c's allocChunkLimit */
+#define COMPACT_ALLOC_CHUNK 8192       /* amount to get from palloc at once */
 #define COMPACT_MAX_REQ                1024    /* must be < COMPACT_ALLOC_CHUNK */
 
 static void *

diff --git a/src/backend/utils/mmgr/aset.c b/src/backend/utils/mmgr/aset.c
index e95dcb6..140b0c7 100644 (file)
--- a/src/backend/utils/mmgr/aset.c
+++ b/src/backend/utils/mmgr/aset.c
@@ -89,7 +89,9 @@
  *
  * With the current parameters, request sizes up to 8K are treated as chunks,
  * larger requests go into dedicated blocks.  Change ALLOCSET_NUM_FREELISTS
- * to adjust the boundary point.
+ * to adjust the boundary point.  (But in contexts with small maxBlockSize,
+ * we may set the allocChunkLimit to less than 8K, so as to avoid space
+ * wastage.)
  *--------------------
  */
 
@@ -97,6 +99,8 @@
 #define ALLOCSET_NUM_FREELISTS 11
 #define ALLOC_CHUNK_LIMIT      (1 << (ALLOCSET_NUM_FREELISTS-1+ALLOC_MINBITS))
 /* Size of largest chunk that we use a fixed size for */
+#define ALLOC_CHUNK_FRACTION   4
+/* We allow chunks to be at most 1/4 of maxBlockSize (less overhead) */
 
 /*--------------------
  * The first block allocated for an allocset has size initBlockSize.
@@ -380,15 +384,20 @@ AllocSetContextCreate(MemoryContext parent,
        /*
         * Compute the allocation chunk size limit for this context.  It can't be
         * more than ALLOC_CHUNK_LIMIT because of the fixed number of freelists.
-        * If maxBlockSize is small then requests exceeding the maxBlockSize
-        * should be treated as large chunks, too.      We have to have
-        * allocChunkLimit a power of two, because the requested and
-        * actually-allocated sizes of any chunk must be on the same side of the
-        * limit, else we get confused about whether the chunk is "big".
+        * If maxBlockSize is small then requests exceeding the maxBlockSize, or
+        * even a significant fraction of it, should be treated as large chunks
+        * too.  For the typical case of maxBlockSize a power of 2, the chunk size
+        * limit will be at most 1/8th maxBlockSize, so that given a stream of
+        * requests that are all the maximum chunk size we will waste at most
+        * 1/8th of the allocated space.
+        *
+        * We have to have allocChunkLimit a power of two, because the requested
+        * and actually-allocated sizes of any chunk must be on the same side of
+        * the limit, else we get confused about whether the chunk is "big".
         */
        context->allocChunkLimit = ALLOC_CHUNK_LIMIT;
-       while (context->allocChunkLimit >
-                  (Size) (maxBlockSize - ALLOC_BLOCKHDRSZ - ALLOC_CHUNKHDRSZ))
+       while ((Size) (context->allocChunkLimit + ALLOC_CHUNKHDRSZ) >
+                  (Size) ((maxBlockSize - ALLOC_BLOCKHDRSZ) / ALLOC_CHUNK_FRACTION))
                context->allocChunkLimit >>= 1;
 
        /*