From bcfdc9df046c159e973149f2a82acc99d07d7b83 Mon Sep 17 00:00:00 2001
From: Tom Lane <tgl@sss.pgh.pa.us>
Date: Mon, 22 Feb 1999 06:16:57 +0000
Subject: [PATCH] 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 | 35 ++----
src/backend/storage/ipc/ipci.c | 12 +-
src/backend/storage/ipc/shmem.c | 32 ++---
src/backend/storage/lmgr/lock.c | 38 ++----
src/backend/storage/smgr/mm.c | 31 ++---
src/backend/utils/hash/dynahash.c | 170 ++++++++++++++++----------
src/include/utils/hsearch.h | 33 ++---
7 files changed, 178 insertions(+), 173 deletions(-)
diff --git a/src/backend/storage/buffer/buf_init.c b/src/backend/storage/buffer/buf_init.c
index 4dd1b0cefe3..13a55f32768 100644
--- a/src/backend/storage/buffer/buf_init.c
+++ b/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;
}
diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c
index 49259807b40..2f5d1d3a193 100644
--- a/src/backend/storage/ipc/ipci.c
+++ b/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)
{
diff --git a/src/backend/storage/ipc/shmem.c b/src/backend/storage/ipc/shmem.c
index afc32068980..3653e844ab6 100644
--- a/src/backend/storage/ipc/shmem.c
+++ b/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);
}
/*
diff --git a/src/backend/storage/lmgr/lock.c b/src/backend/storage/lmgr/lock.c
index 58bd8d04491..ed9a5b31313 100644
--- a/src/backend/storage/lmgr/lock.c
+++ b/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;
}
diff --git a/src/backend/storage/smgr/mm.c b/src/backend/storage/smgr/mm.c
index 4ade36df00a..603ee590638 100644
--- a/src/backend/storage/smgr/mm.c
+++ b/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
diff --git a/src/backend/utils/hash/dynahash.c b/src/backend/utils/hash/dynahash.c
index f41b322189e..dcf4e4cce55 100644
--- a/src/backend/utils/hash/dynahash.c
+++ b/src/backend/utils/hash/dynahash.c
@@ -7,7 +7,7 @@
*
*
* IDENTIFICATION
- * $Header: /cvsroot/pgsql/src/backend/utils/hash/dynahash.c,v 1.18 1999/02/13 23:19:55 momjian Exp $
+ * $Header: /cvsroot/pgsql/src/backend/utils/hash/dynahash.c,v 1.19 1999/02/22 06:16:47 tgl Exp $
*
*-------------------------------------------------------------------------
*/
@@ -174,7 +174,9 @@ hash_create(int nelem, HASHCTL *info, int flags)
if (flags & HASH_SHARED_MEM)
{
- /* ctl structure is preallocated for shared memory tables */
+ /* ctl structure is preallocated for shared memory tables.
+ * Note that HASH_DIRSIZE had better be set as well.
+ */
hashp->hctl = (HHDR *) info->hctl;
hashp->segbase = (char *) info->segbase;
@@ -190,6 +192,7 @@ hash_create(int nelem, HASHCTL *info, int flags)
{
/* setup hash table defaults */
+ hashp->hctl = NULL;
hashp->alloc = (dhalloc_ptr) MEM_ALLOC;
hashp->dir = NULL;
hashp->segbase = NULL;
@@ -210,11 +213,6 @@ hash_create(int nelem, HASHCTL *info, int flags)
hctl->accesses = hctl->collisions = 0;
#endif
- if (flags & HASH_BUCKET)
- {
- hctl->bsize = info->bsize;
- hctl->bshift = my_log2(info->bsize);
- }
if (flags & HASH_SEGMENT)
{
hctl->ssize = info->ssize;
@@ -224,13 +222,12 @@ hash_create(int nelem, HASHCTL *info, int flags)
hctl->ffactor = info->ffactor;
/*
- * SHM hash tables have fixed maximum size (allocate a maximal sized
- * directory).
+ * SHM hash tables have fixed directory size passed by the caller.
*/
if (flags & HASH_DIRSIZE)
{
- hctl->max_dsize = my_log2(info->max_size);
- hctl->dsize = my_log2(info->dsize);
+ hctl->max_dsize = info->max_dsize;
+ hctl->dsize = info->dsize;
}
/*
@@ -254,8 +251,8 @@ hash_create(int nelem, HASHCTL *info, int flags)
}
/*
- Allocate and initialize an HTAB structure
- */
+ * Set default HHDR parameters.
+ */
static int
hdefault(HTAB *hashp)
{
@@ -264,8 +261,6 @@ hdefault(HTAB *hashp)
MemSet(hashp->hctl, 0, sizeof(HHDR));
hctl = hashp->hctl;
- hctl->bsize = DEF_BUCKET_SIZE;
- hctl->bshift = DEF_BUCKET_SHIFT;
hctl->ssize = DEF_SEGSIZE;
hctl->sshift = DEF_SEGSIZE_SHIFT;
hctl->dsize = DEF_DIRSIZE;
@@ -295,38 +290,41 @@ init_htab(HTAB *hashp, int nelem)
SEG_OFFSET *segp;
int nbuckets;
int nsegs;
- int l2;
HHDR *hctl;
hctl = hashp->hctl;
/*
- * Divide number of elements by the fill factor and determine a
+ * Divide number of elements by the fill factor to determine a
* desired number of buckets. Allocate space for the next greater
* power of two number of buckets
*/
nelem = (nelem - 1) / hctl->ffactor + 1;
- l2 = my_log2(nelem);
- nbuckets = 1 << l2;
+ nbuckets = 1 << my_log2(nelem);
hctl->max_bucket = hctl->low_mask = nbuckets - 1;
hctl->high_mask = (nbuckets << 1) - 1;
+ /*
+ * Figure number of directory segments needed, round up to a power of 2
+ */
nsegs = (nbuckets - 1) / hctl->ssize + 1;
nsegs = 1 << my_log2(nsegs);
- if (nsegs > hctl->dsize)
- hctl->dsize = nsegs;
-
- /* Use two low order bits of points ???? */
-
/*
- * if ( !(hctl->mem = bit_alloc ( nbuckets )) ) return(-1); if (
- * !(hctl->mod = bit_alloc ( nbuckets )) ) return(-1);
+ * Make sure directory is big enough.
+ * If pre-allocated directory is too small, choke (caller screwed up).
*/
+ if (nsegs > hctl->dsize)
+ {
+ if (!(hashp->dir))
+ hctl->dsize = nsegs;
+ else
+ return -1;
+ }
- /* allocate a directory */
+ /* Allocate a directory */
if (!(hashp->dir))
{
hashp->dir = (SEG_OFFSET *) hashp->alloc(hctl->dsize * sizeof(SEG_OFFSET));
@@ -346,11 +344,9 @@ init_htab(HTAB *hashp, int nelem)
}
#if HASH_DEBUG
- fprintf(stderr, "%s\n%s%x\n%s%d\n%s%d\n%s%d\n%s%d\n%s%d\n%s%d\n%s%d\n%s%x\n%s%x\n%s%d\n%s%d\n",
+ fprintf(stderr, "%s\n%s%x\n%s%d\n%s%d\n%s%d\n%s%d\n%s%d\n%s%x\n%s%x\n%s%d\n%s%d\n",
"init_htab:",
"TABLE POINTER ", hashp,
- "BUCKET SIZE ", hctl->bsize,
- "BUCKET SHIFT ", hctl->bshift,
"DIRECTORY SIZE ", hctl->dsize,
"SEGMENT SIZE ", hctl->ssize,
"SEGMENT SHIFT ", hctl->sshift,
@@ -364,8 +360,51 @@ init_htab(HTAB *hashp, int nelem)
return 0;
}
+/*
+ * Estimate the space needed for a hashtable containing the given number
+ * of entries of given size.
+ * NOTE: this is used to estimate the footprint of hashtables in shared
+ * memory; therefore it does not count HTAB which is in local memory.
+ * NB: assumes that all hash structure parameters have default values!
+ */
+long
+hash_estimate_size(long num_entries, long keysize, long datasize)
+{
+ long size = 0;
+ long nBuckets,
+ nSegments,
+ nRecordAllocs,
+ recordSize;
+
+ /* estimate number of buckets wanted */
+ nBuckets = 1L << my_log2((num_entries - 1) / DEF_FFACTOR + 1);
+ /* # of segments needed for nBuckets */
+ nSegments = (nBuckets - 1) / DEF_SEGSIZE + 1;
+
+ /* fixed control info */
+ size += MAXALIGN(sizeof(HHDR)); /* but not HTAB, per above */
+ /* directory, assumed to be of size DEF_DIRSIZE */
+ size += MAXALIGN(DEF_DIRSIZE * sizeof(SEG_OFFSET));
+ /* segments */
+ size += nSegments * MAXALIGN(DEF_SEGSIZE * sizeof(BUCKET_INDEX));
+ /* records --- allocated in groups of BUCKET_ALLOC_INCR */
+ recordSize = sizeof(BUCKET_INDEX) + keysize + datasize;
+ recordSize = MAXALIGN(recordSize);
+ nRecordAllocs = (num_entries - 1) / BUCKET_ALLOC_INCR + 1;
+ size += nRecordAllocs * BUCKET_ALLOC_INCR * recordSize;
+
+ return size;
+}
+
+
/********************** DESTROY ROUTINES ************************/
+/*
+ * XXX this sure looks thoroughly broken to me --- tgl 2/99.
+ * It's freeing every entry individually --- but they weren't
+ * allocated individually, see bucket_alloc!! Why doesn't it crash?
+ */
+
void
hash_destroy(HTAB *hashp)
{
@@ -602,7 +641,7 @@ hash_search(HTAB *hashp,
/* link into chain */
*prevIndexPtr = currIndex;
- /* copy key and data */
+ /* copy key into record */
destAddr = (char *) &(curr->key);
memmove(destAddr, keyPtr, hctl->keysize);
curr->next = INVALID_INDEX;
@@ -617,13 +656,10 @@ hash_search(HTAB *hashp,
*/
if (++hctl->nkeys / (hctl->max_bucket + 1) > hctl->ffactor)
{
-
- /*
- * fprintf(stderr,"expanding on '%s'\n",keyPtr);
- * hash_stats("expanded table",hashp);
+ /* NOTE: failure to expand table is not a fatal error,
+ * it just means we have to run at higher fill factor than we wanted.
*/
- if (!expand_table(hashp))
- return NULL;
+ expand_table(hashp);
}
return &(curr->key);
}
@@ -724,23 +760,25 @@ expand_table(HTAB *hashp)
#endif
hctl = hashp->hctl;
- new_bucket = ++hctl->max_bucket;
- old_bucket = (hctl->max_bucket & hctl->low_mask);
+ new_bucket = hctl->max_bucket + 1;
new_segnum = new_bucket >> hctl->sshift;
new_segndx = MOD(new_bucket, hctl->ssize);
if (new_segnum >= hctl->nsegs)
{
-
- /* Allocate new segment if necessary */
+ /* Allocate new segment if necessary -- could fail if dir full */
if (new_segnum >= hctl->dsize)
- dir_realloc(hashp);
+ if (! dir_realloc(hashp))
+ return 0;
if (!(hashp->dir[new_segnum] = seg_alloc(hashp)))
return 0;
hctl->nsegs++;
}
+ /* OK, we got a new bucket */
+ hctl->max_bucket++;
+ old_bucket = (hctl->max_bucket & hctl->low_mask);
if (new_bucket > hctl->high_mask)
{
@@ -789,31 +827,32 @@ static int
dir_realloc(HTAB *hashp)
{
char *p;
- char **p_ptr;
+ char *old_p;
+ long new_dsize;
long old_dirsize;
long new_dirsize;
-
if (hashp->hctl->max_dsize != NO_MAX_DSIZE)
return 0;
/* Reallocate directory */
- old_dirsize = hashp->hctl->dsize * sizeof(SEGMENT *);
- new_dirsize = old_dirsize << 1;
+ new_dsize = hashp->hctl->dsize << 1;
+ old_dirsize = hashp->hctl->dsize * sizeof(SEG_OFFSET);
+ new_dirsize = new_dsize * sizeof(SEG_OFFSET);
- p_ptr = (char **) hashp->dir;
+ old_p = (char *) hashp->dir;
p = (char *) hashp->alloc((unsigned long) new_dirsize);
+
if (p != NULL)
{
- memmove(p, *p_ptr, old_dirsize);
- MemSet(*p_ptr + old_dirsize, 0, new_dirsize - old_dirsize);
- free((char *) *p_ptr);
- *p_ptr = p;
- hashp->hctl->dsize = new_dirsize;
+ memmove(p, old_p, old_dirsize);
+ MemSet(p + old_dirsize, 0, new_dirsize - old_dirsize);
+ MEM_FREE((char *) old_p);
+ hashp->dir = (SEG_OFFSET *) p;
+ hashp->hctl->dsize = new_dsize;
return 1;
}
return 0;
-
}
@@ -823,15 +862,14 @@ seg_alloc(HTAB *hashp)
SEGMENT segp;
SEG_OFFSET segOffset;
-
segp = (SEGMENT) hashp->alloc((unsigned long)
- sizeof(SEGMENT) * hashp->hctl->ssize);
+ sizeof(BUCKET_INDEX) * hashp->hctl->ssize);
if (!segp)
return 0;
MemSet((char *) segp, 0,
- (long) sizeof(SEGMENT) * hashp->hctl->ssize);
+ (long) sizeof(BUCKET_INDEX) * hashp->hctl->ssize);
segOffset = MAKE_HASHOFFSET(hashp, segp);
return segOffset;
@@ -849,10 +887,11 @@ bucket_alloc(HTAB *hashp)
BUCKET_INDEX tmpIndex,
lastIndex;
+ /* Each bucket has a BUCKET_INDEX header plus user data. */
bucketSize = sizeof(BUCKET_INDEX) + hashp->hctl->keysize + hashp->hctl->datasize;
/* make sure its aligned correctly */
- bucketSize += sizeof(long *) - (bucketSize % sizeof(long *));
+ bucketSize = MAXALIGN(bucketSize);
/*
* tmpIndex is the shmem offset into the first bucket of the array.
@@ -869,8 +908,10 @@ bucket_alloc(HTAB *hashp)
lastIndex = hashp->hctl->freeBucketIndex;
hashp->hctl->freeBucketIndex = tmpIndex;
- /* initialize each bucket to point to the one behind it */
- for (i = 0; i < (BUCKET_ALLOC_INCR - 1); i++)
+ /* initialize each bucket to point to the one behind it.
+ * NOTE: loop sets last bucket incorrectly; we fix below.
+ */
+ for (i = 0; i < BUCKET_ALLOC_INCR; i++)
{
tmpBucket = GET_BUCKET(hashp, tmpIndex);
tmpIndex += bucketSize;
@@ -879,20 +920,21 @@ bucket_alloc(HTAB *hashp)
/*
* the last bucket points to the old freelist head (which is probably
- * invalid or we wouldnt be here)
+ * invalid or we wouldn't be here)
*/
tmpBucket->next = lastIndex;
return 1;
}
-/* calculate the log base 2 of num */
+/* calculate ceil(log base 2) of num */
int
my_log2(long num)
{
- int i = 1;
- int limit;
+ int i;
+ long limit;
- for (i = 0, limit = 1; limit < num; limit = 2 * limit, i++);
+ for (i = 0, limit = 1; limit < num; i++, limit <<= 1)
+ ;
return i;
}
diff --git a/src/include/utils/hsearch.h b/src/include/utils/hsearch.h
index 50b8b3095c6..4e5676eb887 100644
--- a/src/include/utils/hsearch.h
+++ b/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
--
GitLab