Repair some pretty serious problems in dynahash.c and

shared memory space allocation.  It's a wonder we have not seen bug
reports traceable to this area ... it's quite clear that the routine
dir_realloc() has never worked correctly, for example.

src/backend/storage/buffer/buf_init.c

@@ -7,7 +7,7 @@
  *
  *
  * IDENTIFICATION
- *	  $Header: /cvsroot/pgsql/src/backend/storage/buffer/buf_init.c,v 1.21 1999/02/13 23:17:54 momjian Exp $
+ *	  $Header: /cvsroot/pgsql/src/backend/storage/buffer/buf_init.c,v 1.22 1999/02/22 06:16:50 tgl Exp $
  *
  *-------------------------------------------------------------------------
  */
@@ -33,7 +33,6 @@
 #include "storage/lmgr.h"
 #include "miscadmin.h"
 #include "utils/builtins.h"
-#include "utils/dynahash.h"
 #include "utils/hsearch.h"
 #include "utils/memutils.h"
 #include "executor/execdebug.h" /* for NDirectFileRead */
@@ -261,21 +260,11 @@ int
 BufferShmemSize()
 {
 	int			size = 0;
-	int			nbuckets;
-	int			nsegs;
-	int			tmp;
-
-	nbuckets = 1 << (int) my_log2((NBuffers - 1) / DEF_FFACTOR + 1);
-	nsegs = 1 << (int) my_log2((nbuckets - 1) / DEF_SEGSIZE + 1);
-
-	/* size of shmem index table */
-	size += MAXALIGN(my_log2(SHMEM_INDEX_SIZE) * sizeof(void *));		/* HTAB->dir */
-	size += MAXALIGN(sizeof(HHDR));		/* HTAB->hctl */
-	size += MAXALIGN(DEF_SEGSIZE * sizeof(SEGMENT));
-	size += BUCKET_ALLOC_INCR *
-		(MAXALIGN(sizeof(BUCKET_INDEX)) +
-		 MAXALIGN(SHMEM_INDEX_KEYSIZE) +
-		 MAXALIGN(SHMEM_INDEX_DATASIZE));
+
+	/* size of shmem index hash table */
+	size += hash_estimate_size(SHMEM_INDEX_SIZE,
+							   SHMEM_INDEX_KEYSIZE,
+							   SHMEM_INDEX_DATASIZE);
 
 	/* size of buffer descriptors */
 	size += MAXALIGN((NBuffers + 1) * sizeof(BufferDesc));
@@ -284,17 +273,13 @@ BufferShmemSize()
 	size += NBuffers * MAXALIGN(BLCKSZ);
 
 	/* size of buffer hash table */
-	size += MAXALIGN(my_log2(NBuffers) * sizeof(void *));		/* HTAB->dir */
-	size += MAXALIGN(sizeof(HHDR));		/* HTAB->hctl */
-	size += nsegs * MAXALIGN(DEF_SEGSIZE * sizeof(SEGMENT));
-	tmp = (int) ceil((double) NBuffers / BUCKET_ALLOC_INCR);
-	size += tmp * BUCKET_ALLOC_INCR *
-		(MAXALIGN(sizeof(BUCKET_INDEX)) +
-		 MAXALIGN(sizeof(BufferTag)) +
-		 MAXALIGN(sizeof(Buffer)));
+	size += hash_estimate_size(NBuffers,
+							   sizeof(BufferTag),
+							   sizeof(Buffer));
 
 #ifdef BMTRACE
 	size += (BMT_LIMIT * sizeof(bmtrace)) + sizeof(long);
 #endif
+
 	return size;
 }

src/backend/storage/ipc/ipci.c

@@ -7,7 +7,7 @@
  *
  *
  * IDENTIFICATION
- *	  $Header: /cvsroot/pgsql/src/backend/storage/ipc/ipci.c,v 1.21 1999/02/21 01:41:44 tgl Exp $
+ *	  $Header: /cvsroot/pgsql/src/backend/storage/ipc/ipci.c,v 1.22 1999/02/22 06:16:49 tgl Exp $
  *
  *-------------------------------------------------------------------------
  */
@@ -72,11 +72,19 @@ CreateSharedMemoryAndSemaphores(IPCKey key, int maxBackends)
 	 * ----------------
 	 */
 	CreateSpinlocks(IPCKeyGetSpinLockSemaphoreKey(key));
-	size = BufferShmemSize() + LockShmemSize(maxBackends);
 
+	/*
+	 * Size of the primary shared-memory block is estimated via
+	 * moderately-accurate estimates for the big hogs, plus 100K for
+	 * the stuff that's too small to bother with estimating.
+	 * Then we add 10% for a safety margin.
+	 */
+	size = BufferShmemSize() + LockShmemSize(maxBackends);
 #ifdef STABLE_MEMORY_STORAGE
 	size += MMShmemSize();
 #endif
+	size += 100000;
+	size += size / 10;
 
 	if (DebugLvl > 1)
 	{

src/backend/storage/ipc/shmem.c

@@ -7,7 +7,7 @@
  *
  *
  * IDENTIFICATION
- *	  $Header: /cvsroot/pgsql/src/backend/storage/ipc/shmem.c,v 1.36 1999/02/13 23:18:13 momjian Exp $
+ *	  $Header: /cvsroot/pgsql/src/backend/storage/ipc/shmem.c,v 1.37 1999/02/22 06:16:48 tgl Exp $
  *
  *-------------------------------------------------------------------------
  */
@@ -65,7 +65,6 @@
 #include "storage/shmem.h"
 #include "storage/spin.h"
 #include "storage/proc.h"
-#include "utils/dynahash.h"
 #include "utils/hsearch.h"
 #include "utils/memutils.h"
 #include "access/xact.h"
@@ -215,7 +214,7 @@ InitShmem(unsigned int key, unsigned int size)
 	/* create OR attach to the shared memory shmem index */
 	info.keysize = SHMEM_INDEX_KEYSIZE;
 	info.datasize = SHMEM_INDEX_DATASIZE;
-	hash_flags = (HASH_ELEM);
+	hash_flags = HASH_ELEM;
 
 	/* This will acquire the shmem index lock, but not release it. */
 	ShmemIndex = ShmemInitHash("ShmemIndex",
@@ -340,8 +339,8 @@ ShmemIsValid(unsigned long addr)
  */
 HTAB *
 ShmemInitHash(char *name,		/* table string name for shmem index */
-			  long init_size,	/* initial size */
-			  long max_size,	/* max size of the table */
+			  long init_size,	/* initial table size */
+			  long max_size,	/* max size of the table (NOT USED) */
 			  HASHCTL *infoP,	/* info about key and bucket size */
 			  int hash_flags)	/* info about infoP */
 {
@@ -349,17 +348,20 @@ ShmemInitHash(char *name,		/* table string name for shmem index */
 	long	   *location;
 
 	/*
-	 * shared memory hash tables have a fixed max size so that the control
-	 * structures don't try to grow.  The segbase is for calculating
-	 * pointer values.	The shared memory allocator must be specified.
+	 * Hash tables allocated in shared memory have a fixed directory;
+	 * it can't grow or other backends wouldn't be able to find it.
+	 * The segbase is for calculating pointer values.
+	 * The shared memory allocator must be specified too.
 	 */
+	infoP->dsize = infoP->max_dsize = DEF_DIRSIZE;
 	infoP->segbase = (long *) ShmemBase;
 	infoP->alloc = ShmemAlloc;
-	infoP->max_size = max_size;
-	hash_flags |= HASH_SHARED_MEM;
+	hash_flags |= HASH_SHARED_MEM | HASH_DIRSIZE;
 
 	/* look it up in the shmem index */
-	location = ShmemInitStruct(name, my_log2(max_size) + sizeof(HHDR), &found);
+	location = ShmemInitStruct(name,
+							   sizeof(HHDR) + DEF_DIRSIZE * sizeof(SEG_OFFSET),
+							   &found);
 
 	/*
 	 * shmem index is corrupted.	Let someone else give the error
@@ -375,13 +377,11 @@ ShmemInitHash(char *name,		/* table string name for shmem index */
 	if (found)
 		hash_flags |= HASH_ATTACH;
 
-	/* these structures were allocated or bound in ShmemInitStruct */
-	/* control information and parameters */
+	/* Now provide the header and directory pointers */
 	infoP->hctl = (long *) location;
-	/* directory for hash lookup */
-	infoP->dir = (long *) (location + sizeof(HHDR));
+	infoP->dir = (long *) (((char*) location) + sizeof(HHDR));
 
-	return hash_create(init_size, infoP, hash_flags);;
+	return hash_create(init_size, infoP, hash_flags);
 }
 
 /*

src/backend/storage/lmgr/lock.c

@@ -7,7 +7,7 @@
  *
  *
  * IDENTIFICATION
- *	  $Header: /cvsroot/pgsql/src/backend/storage/lmgr/lock.c,v 1.44 1999/02/21 03:49:22 scrappy Exp $
+ *	  $Header: /cvsroot/pgsql/src/backend/storage/lmgr/lock.c,v 1.45 1999/02/22 06:16:52 tgl Exp $
  *
  * NOTES
  *	  Outside modules can create a lock table and acquire/release
@@ -42,7 +42,6 @@
 #include "storage/spin.h"
 #include "storage/proc.h"
 #include "storage/lock.h"
-#include "utils/dynahash.h"
 #include "utils/hsearch.h"
 #include "utils/memutils.h"
 #include "utils/palloc.h"
@@ -1481,36 +1480,21 @@ int
 LockShmemSize(int maxBackends)
 {
 	int			size = 0;
-	int			nLockEnts = NLOCKENTS(maxBackends);
-	int			nLockBuckets,
-				nLockSegs;
-	int			nXidBuckets,
-				nXidSegs;
-
-	nLockBuckets = 1 << (int) my_log2((nLockEnts - 1) / DEF_FFACTOR + 1);
-	nLockSegs = 1 << (int) my_log2((nLockBuckets - 1) / DEF_SEGSIZE + 1);
-
-	nXidBuckets = 1 << (int) my_log2((NLOCKS_PER_XACT - 1) / DEF_FFACTOR + 1);
-	nXidSegs = 1 << (int) my_log2((nLockBuckets - 1) / DEF_SEGSIZE + 1);
 
+	size += MAXALIGN(sizeof(PROC_HDR)); /* ProcGlobal */
 	size += MAXALIGN(maxBackends * sizeof(PROC)); /* each MyProc */
 	size += MAXALIGN(maxBackends * sizeof(LOCKMETHODCTL));		/* each
 																 * lockMethodTable->ctl */
-	size += MAXALIGN(sizeof(PROC_HDR)); /* ProcGlobal */
 
-	size += MAXALIGN(my_log2(nLockEnts) * sizeof(void *));
-	size += MAXALIGN(sizeof(HHDR));
-	size += nLockSegs * MAXALIGN(DEF_SEGSIZE * sizeof(SEGMENT));
-	size += nLockEnts *			/* XXX not multiple of BUCKET_ALLOC_INCR? */
-		(MAXALIGN(sizeof(BUCKET_INDEX)) +
-		 MAXALIGN(sizeof(LOCK)));		/* contains hash key */
-
-	size += MAXALIGN(my_log2(maxBackends) * sizeof(void *));
-	size += MAXALIGN(sizeof(HHDR));
-	size += nXidSegs * MAXALIGN(DEF_SEGSIZE * sizeof(SEGMENT));
-	size += maxBackends *		/* XXX not multiple of BUCKET_ALLOC_INCR? */
-		(MAXALIGN(sizeof(BUCKET_INDEX)) +
-		 MAXALIGN(sizeof(XIDLookupEnt)));		/* contains hash key */
+	/* lockHash table */
+	size += hash_estimate_size(NLOCKENTS(maxBackends),
+							   sizeof(LOCKTAG),
+							   sizeof(LOCK));
+
+	/* xidHash table */
+	size += hash_estimate_size(maxBackends,
+							   XID_TAGSIZE,
+							   sizeof(XIDLookupEnt));
 
 	return size;
 }

src/backend/storage/smgr/mm.c

@@ -10,7 +10,7 @@
  *
  *
  * IDENTIFICATION
- *	  $Header: /cvsroot/pgsql/src/backend/storage/smgr/Attic/mm.c,v 1.13 1999/02/13 23:18:36 momjian Exp $
+ *	  $Header: /cvsroot/pgsql/src/backend/storage/smgr/Attic/mm.c,v 1.14 1999/02/22 06:16:57 tgl Exp $
  *
  *-------------------------------------------------------------------------
  */
@@ -25,7 +25,6 @@
 #include "storage/shmem.h"
 #include "storage/spin.h"
 
-#include "utils/dynahash.h"
 #include "utils/hsearch.h"
 #include "utils/rel.h"
 #include "utils/memutils.h"
@@ -565,36 +564,20 @@ int
 MMShmemSize()
 {
 	int			size = 0;
-	int			nbuckets;
-	int			nsegs;
-	int			tmp;
 
 	/*
 	 * first compute space occupied by the (dbid,relid,blkno) hash table
 	 */
-
-	nbuckets = 1 << (int) my_log2((MMNBUFFERS - 1) / DEF_FFACTOR + 1);
-	nsegs = 1 << (int) my_log2((nbuckets - 1) / DEF_SEGSIZE + 1);
-
-	size += MAXALIGN(my_log2(MMNBUFFERS) * sizeof(void *));
-	size += MAXALIGN(sizeof(HHDR));
-	size += nsegs * MAXALIGN(DEF_SEGSIZE * sizeof(SEGMENT));
-	tmp = (int) ceil((double) MMNBUFFERS / BUCKET_ALLOC_INCR);
-	size += tmp * BUCKET_ALLOC_INCR *
-		(MAXALIGN(sizeof(BUCKET_INDEX)) +
-		 MAXALIGN(sizeof(MMHashEntry)));		/* contains hash key */
+	size += hash_estimate_size(MMNBUFFERS,
+				   0, /* MMHashEntry includes key */
+				   sizeof(MMHashEntry));
 
 	/*
 	 * now do the same for the rel hash table
 	 */
-
-	size += MAXALIGN(my_log2(MMNRELATIONS) * sizeof(void *));
-	size += MAXALIGN(sizeof(HHDR));
-	size += nsegs * MAXALIGN(DEF_SEGSIZE * sizeof(SEGMENT));
-	tmp = (int) ceil((double) MMNRELATIONS / BUCKET_ALLOC_INCR);
-	size += tmp * BUCKET_ALLOC_INCR *
-		(MAXALIGN(sizeof(BUCKET_INDEX)) +
-		 MAXALIGN(sizeof(MMRelHashEntry)));		/* contains hash key */
+	size += hash_estimate_size(MMNRELATIONS,
+				   0, /* MMRelHashEntry includes key */
+				   sizeof(MMRelHashEntry));
 
 	/*
 	 * finally, add in the memory block we use directly

src/include/utils/hsearch.h

@@ -6,7 +6,7 @@
  *
  * Copyright (c) 1994, Regents of the University of California
  *
- * $Id: hsearch.h,v 1.10 1999/02/13 23:22:22 momjian Exp $
+ * $Id: hsearch.h,v 1.11 1999/02/22 06:16:46 tgl Exp $
  *
  *-------------------------------------------------------------------------
  */
@@ -16,16 +16,23 @@
 
 /*
  * Constants
+ *
+ * A hash table has a top-level "directory", each of whose entries points
+ * to a "segment" of ssize bucket headers.  The maximum number of hash
+ * buckets is thus dsize * ssize (but dsize may be expansible).  Of course,
+ * the number of records in the table can be larger, but we don't want a
+ * whole lot of records per bucket or performance goes down.
+ *
+ * In a hash table allocated in shared memory, the directory cannot be
+ * expanded because it must stay at a fixed address.
  */
-#define DEF_BUCKET_SIZE		   256
-#define DEF_BUCKET_SHIFT	   8/* log2(BUCKET) */
 #define DEF_SEGSIZE			   256
-#define DEF_SEGSIZE_SHIFT			   8		/* log2(SEGSIZE)  */
+#define DEF_SEGSIZE_SHIFT	   8			/* log2(SEGSIZE)  */
 #define DEF_DIRSIZE			   256
-#define PRIME1				   37
+#define DEF_FFACTOR			   1			/* default fill factor */
+
+#define PRIME1				   37			/* for the hash function */
 #define PRIME2				   1048583
-#define DEF_FFACTOR			   1
-#define SPLTMAX				   8
 
 
 /*
@@ -46,10 +53,8 @@ typedef unsigned long SEG_OFFSET;
 
 typedef struct hashhdr
 {
-	long		bsize;			/* Bucket/Page Size */
-	long		bshift;			/* Bucket shift */
 	long		dsize;			/* Directory Size */
-	long		ssize;			/* Segment Size */
+	long		ssize;			/* Segment Size --- must be power of 2 */
 	long		sshift;			/* Segment shift */
 	long		max_bucket;		/* ID of Maximum bucket in use */
 	long		high_mask;		/* Mask to modulo into entire table */
@@ -59,8 +64,7 @@ typedef struct hashhdr
 	long		nsegs;			/* Number of allocated segments */
 	long		keysize;		/* hash key length in bytes */
 	long		datasize;		/* elem data length in bytes */
-	long		max_dsize;		/* 'dsize' limit if directory is fixed
-								 * size */
+	long		max_dsize;		/* 'dsize' limit if directory is fixed size */
 	BUCKET_INDEX freeBucketIndex;
 	/* index of first free bucket */
 #ifdef HASH_STATISTICS
@@ -83,14 +87,13 @@ typedef struct htab
 
 typedef struct hashctl
 {
-	long		bsize;			/* Bucket Size */
 	long		ssize;			/* Segment Size */
 	long		dsize;			/* Dirsize Size */
 	long		ffactor;		/* Fill factor */
 	long		(*hash) ();		/* Hash Function */
 	long		keysize;		/* hash key length in bytes */
 	long		datasize;		/* elem data length in bytes */
-	long		max_size;		/* limit to dsize if directory size is
+	long		max_dsize;		/* limit to dsize if directory size is
 								 * limited */
 	long	   *segbase;		/* base for calculating bucket + seg ptrs */
 	long	   *(*alloc) ();	/* memory allocation function */
@@ -100,7 +103,6 @@ typedef struct hashctl
 } HASHCTL;
 
 /* Flags to indicate action for hctl */
-#define HASH_BUCKET		0x001	/* Setting bucket size */
 #define HASH_SEGMENT	0x002	/* Setting segment size */
 #define HASH_DIRSIZE	0x004	/* Setting directory size */
 #define HASH_FFACTOR	0x008	/* Setting fill factor */
@@ -136,6 +138,7 @@ extern void hash_stats(char *where, HTAB *hashp);
 extern long *hash_search(HTAB *hashp, char *keyPtr, HASHACTION action,
 			bool *foundPtr);
 extern long *hash_seq(HTAB *hashp);
+extern long hash_estimate_size(long num_entries, long keysize, long datasize);
 
 /*
  * prototypes from functions in hashfn.c