Replace mutex implementation, and fix a few bugs.

[btree] / threadskv10e.c

// btree version threadskv10e futex version
//      with reworked bt_deletekey code,
//      phase-fair re-entrant reader writer lock,
//      librarian page split code,
//      duplicate key management
//      bi-directional cursors
//      traditional buffer pool manager
//      ACID batched key-value updates
//      redo log for failure recovery
//      LSM B-trees for write optimization
//      and larger sized leaf pages than non-leaf

// 24 OCT 2014

// author: karl malbrain, malbrain@cal.berkeley.edu

/*
This work, including the source code, documentation
and related data, is placed into the public domain.

The orginal author is Karl Malbrain.

THIS SOFTWARE IS PROVIDED AS-IS WITHOUT WARRANTY
OF ANY KIND, NOT EVEN THE IMPLIED WARRANTY OF
MERCHANTABILITY. THE AUTHOR OF THIS SOFTWARE,
ASSUMES _NO_ RESPONSIBILITY FOR ANY CONSEQUENCE
RESULTING FROM THE USE, MODIFICATION, OR
REDISTRIBUTION OF THIS SOFTWARE.
*/

// Please see the project home page for documentation
// code.google.com/p/high-concurrency-btree

#define _FILE_OFFSET_BITS 64
#define _LARGEFILE64_SOURCE

#ifdef linux
#define _GNU_SOURCE
#include <xmmintrin.h>
#include <linux/futex.h>
#define SYS_futex 202
#endif

#ifdef unix
#include <unistd.h>
#include <stdio.h>
#include <stdlib.h>
#include <fcntl.h>
#include <sys/time.h>
#include <sys/mman.h>
#include <errno.h>
#include <pthread.h>
#include <limits.h>
#else
#define WIN32_LEAN_AND_MEAN
#include <windows.h>
#include <stdio.h>
#include <stdlib.h>
#include <time.h>
#include <fcntl.h>
#include <process.h>
#include <intrin.h>
#endif

#include <memory.h>
#include <string.h>
#include <stddef.h>

typedef unsigned long long      uid;
typedef unsigned long long      logseqno;

#ifndef unix
typedef unsigned long long      off64_t;
typedef unsigned short          ushort;
typedef unsigned int            uint;
#endif

#define BT_ro 0x6f72    // ro
#define BT_rw 0x7772    // rw

#define BT_maxbits              26                                      // maximum page size in bits
#define BT_minbits              9                                       // minimum page size in bits
#define BT_minpage              (1 << BT_minbits)       // minimum page size
#define BT_maxpage              (1 << BT_maxbits)       // maximum page size

//  BTree page number constants
#define ALLOC_page              0       // allocation page
#define ROOT_page               1       // root of the btree
#define LEAF_page               2       // first page of leaves

//      Number of levels to create in a new BTree

#define MIN_lvl                 2

/*
There are six lock types for each node in four independent sets: 
1. (set 1) AccessIntent: Sharable. Going to Read the node. Incompatible with NodeDelete. 
2. (set 1) NodeDelete: Exclusive. About to release the node. Incompatible with AccessIntent. 
3. (set 2) ReadLock: Sharable. Read the node. Incompatible with WriteLock. 
4. (set 2) WriteLock: Exclusive. Modify the node. Incompatible with ReadLock and other WriteLocks. 
5. (set 3) ParentModification: Exclusive. Change the node's parent keys. Incompatible with another ParentModification. 
6. (set 4) LinkModification: Exclusive. Update of a node's left link is underway. Incompatible with another LinkModification. 
*/

typedef enum{
        BtLockAccess = 1,
        BtLockDelete = 2,
        BtLockRead   = 4,
        BtLockWrite  = 8,
        BtLockParent = 16,
        BtLockLink   = 32
} BtLock;

typedef struct {
  union {
        struct {
          volatile unsigned char xcl[1];
          volatile unsigned char filler;
          volatile ushort waiters[1];
        } bits[1];
        uint value[1];
  };
} MutexLatch;

//      definition for reader/writer reentrant lock implementation

typedef struct {
  MutexLatch xcl[1];
  MutexLatch wrt[1];
  uint readers;
  ushort dup;           // re-entrant locks
  ushort tid;           // owner thread-no
  uint line;            // owner line #
} RWLock;

//  hash table entries

typedef struct {
        MutexLatch latch[1];
        uint entry;             // Latch table entry at head of chain
} BtHashEntry;

//      latch manager table structure

typedef struct {
        uid page_no;                    // latch set page number
        MutexLatch modify[1];   // modify entry lite latch
        RWLock readwr[1];               // read/write page lock
        RWLock access[1];               // Access Intent/Page delete
        RWLock parent[1];               // Posting of fence key in parent
        RWLock link[1];                 // left link update in progress
        uint split;                             // right split page atomic insert
        uint next;                              // next entry in hash table chain
        uint prev;                              // prev entry in hash table chain
        volatile ushort pin;    // number of accessing threads
        unsigned char dirty;    // page in cache is dirty
        unsigned char leaf;             // page in cache is a leaf
} BtLatchSet;

//      Define the length of the page record numbers

#define BtId 6

//      Page key slot definition.

//      Keys are marked dead, but remain on the page until
//      it cleanup is called. The fence key (highest key) for
//      a leaf page is always present, even after cleanup.

//      Slot types

//      In addition to the Unique keys that occupy slots
//      there are Librarian and Duplicate key
//      slots occupying the key slot array.

//      The Librarian slots are dead keys that
//      serve as filler, available to add new Unique
//      or Dup slots that are inserted into the B-tree.

//      The Duplicate slots have had their key bytes extended
//      by 6 bytes to contain a binary duplicate key uniqueifier.

typedef enum {
        Unique,
        Update,
        Librarian,
        Duplicate,
        Delete
} BtSlotType;

typedef struct {
        uint off:BT_maxbits;    // page offset for key start
        uint type:3;                    // type of slot
        uint dead:1;                    // set for deleted slot
} BtSlot;

//      The key structure occupies space at the upper end of
//      each page.  It's a length byte followed by the key
//      bytes.

typedef struct {
        unsigned char len;              // this can be changed to a ushort or uint
        unsigned char key[0];
} BtKey;

//      the value structure also occupies space at the upper
//      end of the page. Each key is immediately followed by a value.

typedef struct {
        unsigned char len;              // this can be changed to a ushort or uint
        unsigned char value[0];
} BtVal;

#define BT_maxkey       255             // maximum number of bytes in a key
#define BT_keyarray (BT_maxkey + sizeof(BtKey))

//      The first part of an index page.
//      It is immediately followed
//      by the BtSlot array of keys.

typedef struct BtPage_ {
        uint cnt;                                       // count of keys in page
        uint act;                                       // count of active keys
        uint min;                                       // next key offset
        uint garbage;                           // page garbage in bytes
        unsigned char lvl;                      // level of page
        unsigned char free;                     // page is on free chain
        unsigned char kill;                     // page is being deleted
        unsigned char nopromote;        // page is being constructed
        unsigned char filler1[6];       // padding to multiple of 8 bytes
        unsigned char right[BtId];      // page number to right
        unsigned char left[BtId];       // page number to left
        unsigned char filler2[2];       // padding to multiple of 8 bytes
        logseqno lsn;                           // log sequence number applied
        uid page_no;                            // this page number
} *BtPage;

//  The loadpage interface object

typedef struct {
        BtPage page;            // current page pointer
        BtLatchSet *latch;      // current page latch set
} BtPageSet;

//      structure for latch manager on ALLOC_page

typedef struct {
        struct BtPage_ alloc[1];                // next page_no in right ptr
        unsigned char freechain[BtId];  // head of free page_nos chain
        unsigned char leafchain[BtId];  // head of leaf page_nos chain
        unsigned long long leafpages;   // number of active leaf pages
        uint redopages;                                 // number of redo pages in file
        unsigned char leaf_xtra;                // leaf page size in xtra bits
        unsigned char page_bits;                // base page size in bits
} BtPageZero;

//      The object structure for Btree access

typedef struct {
        uint page_size;                         // base page size       
        uint page_bits;                         // base page size in bits       
        uint leaf_xtra;                         // leaf xtra bits       
#ifdef unix
        int idx;
#else
        HANDLE idx;
#endif
        BtPageZero *pagezero;           // mapped allocation page
        BtHashEntry *hashtable;         // the buffer pool hash table entries
        BtHashEntry *leaftable;         // the buffer pool hash table entries
        BtLatchSet *latchsets;          // mapped latch set from buffer pool
        BtLatchSet *leafsets;           // mapped latch set from buffer pool
        unsigned char *pagepool;        // mapped to the buffer pool pages
        unsigned char *leafpool;        // mapped to the leaf pool pages
        unsigned char *redobuff;        // mapped recovery buffer pointer
        logseqno lsn, flushlsn;         // current & first lsn flushed
        MutexLatch redo[1];                     // redo area lite latch
        MutexLatch lock[1];                     // allocation area lite latch
        MutexLatch maps[1];                     // mapping segments lite latch
        ushort thread_no[1];            // highest thread number issued
        ushort err_thread;                      // error thread number
        uint nlatchpage;                        // size of buffer pool & latchsets
        uint latchtotal;                        // number of page latch entries
        uint latchhash;                         // number of latch hash table slots
        uint latchvictim;                       // next latch entry to examine
        uint nleafpage;                         // size of leaf pool & leafsets
        uint leaftotal;                         // number of leaf latch entries
        uint leafhash;                          // number of leaf hash table slots
        uint leafvictim;                        // next leaf entry to examine
        uint leafpromote;                       // next leaf entry to promote
        uint redopage;                          // page number of redo buffer
        uint redolast;                          // last msync size of recovery buff
        uint redoend;                           // eof/end element in recovery buff
        int err;                                        // last error
        int line;                                       // last error line no
        int found;                                      // number of keys found by delete
        int reads, writes;                      // number of reads and writes
#ifndef unix
        HANDLE halloc;                          // allocation handle
        HANDLE hpool;                           // buffer pool handle
#endif
        volatile uint segments;         // number of memory mapped segments
        unsigned char *pages[64000];// memory mapped segments of b-tree
} BtMgr;

typedef struct {
        BtMgr *mgr;                                     // buffer manager for entire process
        BtMgr *main;                            // buffer manager for main btree
        BtPage cursor;                          // cached page frame for start/next
        ushort thread_no;                       // thread number
        unsigned char key[BT_keyarray]; // last found complete key
} BtDb;

//      atomic txn structures

typedef struct {
        logseqno reqlsn;        // redo log seq no required
        uint entry:31;          // latch table entry number
        uint reuse:1;           // reused previous page
} AtomicTxn;

//      Catastrophic errors

typedef enum {
        BTERR_ok = 0,
        BTERR_struct,
        BTERR_ovflw,
        BTERR_lock,
        BTERR_map,
        BTERR_read,
        BTERR_wrt,
        BTERR_atomic,
        BTERR_recovery
} BTERR;

#define CLOCK_bit 0x8000

// recovery manager entry types

typedef enum {
        BTRM_eof = 0,   // rest of buffer is emtpy
        BTRM_add,               // add a unique key-value to btree
        BTRM_dup,               // add a duplicate key-value to btree
        BTRM_del,               // delete a key-value from btree
        BTRM_upd,               // update a key with a new value
        BTRM_new,               // allocate a new empty page
        BTRM_old                // reuse an old empty page
} BTRM;

// recovery manager entry
//      structure followed by BtKey & BtVal

typedef struct {
        logseqno reqlsn;        // log sequence number required
        logseqno lsn;           // log sequence number for entry
        uint len;                       // length of entry
        unsigned char type;     // type of entry
        unsigned char lvl;      // level of btree entry pertains to
} BtLogHdr;

// B-Tree functions

extern void bt_close (BtDb *bt);
extern BtDb *bt_open (BtMgr *mgr, BtMgr *main);
extern BTERR bt_writepage (BtMgr *mgr, BtPage page, uid page_no, uint leaf);
extern BTERR bt_readpage (BtMgr *mgr, BtPage page, uid page_no, uint leaf);
extern void bt_lockpage(BtLock mode, BtLatchSet *latch, ushort thread_no, uint line);
extern void bt_unlockpage(BtLock mode, BtLatchSet *latch, ushort thread_no, uint line);
extern BTERR bt_insertkey (BtMgr *mgr, unsigned char *key, uint len, uint lvl, void *value, uint vallen, BtSlotType type, ushort thread_no);
extern BTERR  bt_deletekey (BtMgr *mgr, unsigned char *key, uint len, uint lvl, ushort thread_no);

extern int bt_findkey (BtDb *db, unsigned char *key, uint keylen, unsigned char *value, uint valmax);

extern uint bt_startkey (BtDb *db, unsigned char *key, uint len);
extern uint bt_nextkey   (BtDb *bt, uint slot);
extern uint bt_prevkey (BtDb *db, uint slot);
extern uint bt_lastkey (BtDb *db);

//      manager functions
extern BtMgr *bt_mgr (char *name, uint bits, uint leaf_xtra, uint poolsize, uint leafpool, uint redopages);
extern void bt_mgrclose (BtMgr *mgr);
extern logseqno bt_newredo (BtMgr *mgr, BTRM type, int lvl, BtKey *key, BtVal *val, ushort thread_no);
extern logseqno bt_txnredo (BtMgr *mgr, BtPage page, ushort thread_no);

//      atomic transaction functions
BTERR bt_atomicexec(BtMgr *mgr, BtPage source, logseqno lsn, int lsm, ushort thread_no);
BTERR bt_txnpromote (BtDb *bt);

//  The page is allocated from low and hi ends.
//  The key slots are allocated from the bottom,
//      while the text and value of the key
//  are allocated from the top.  When the two
//  areas meet, the page is split into two.

//  A key consists of a length byte, two bytes of
//  index number (0 - 65535), and up to 253 bytes
//  of key value.

//  Associated with each key is a value byte string
//      containing any value desired.

//  The b-tree root is always located at page 1.
//      The first leaf page of level zero is always
//      located on page 2.

//      The b-tree pages are linked with next
//      pointers to facilitate enumerators,
//      and provide for concurrency.

//      When to root page fills, it is split in two and
//      the tree height is raised by a new root at page
//      one with two keys.

//      Deleted keys are marked with a dead bit until
//      page cleanup. The fence key for a leaf node is
//      always present

//  To achieve maximum concurrency one page is locked at a time
//  as the tree is traversed to find leaf key in question. The right
//  page numbers are used in cases where the page is being split,
//      or consolidated.

//  Page 0 is dedicated to lock for new page extensions,
//      and chains empty pages together for reuse. It also
//      contains the latch manager hash table.

//      The ParentModification lock on a node is obtained to serialize posting
//      or changing the fence key for a node.

//      Empty pages are chained together through the ALLOC page and reused.

//      Access macros to address slot and key values from the page
//      Page slots use 1 based indexing.

#define slotptr(page, slot) (((BtSlot *)(page+1)) + (slot-1))
#define keyptr(page, slot) ((BtKey*)((unsigned char*)(page) + slotptr(page, slot)->off))
#define valptr(page, slot) ((BtVal*)(keyptr(page,slot)->key + keyptr(page,slot)->len))

void bt_putid(unsigned char *dest, uid id)
{
int i = BtId;

        while( i-- )
                dest[i] = (unsigned char)id, id >>= 8;
}

uid bt_getid(unsigned char *src)
{
uid id = 0;
int i;

        for( i = 0; i < BtId; i++ )
                id <<= 8, id |= *src++; 

        return id;
}

//      lite weight spin lock Latch Manager

int sys_futex(void *addr1, int op, int val1, struct timespec *timeout, void *addr2, int val3)
{
        return syscall(SYS_futex, addr1, op, val1, timeout, addr2, val3);
}

void bt_mutexlock(MutexLatch *latch)
{
uint idx, waited = 0;
MutexLatch prev[1];

 while( 1 ) {
  for( idx = 0; idx < 100; idx++ ) {
        *prev->value = __sync_fetch_and_or (latch->value, 1);
        if( !*prev->bits->xcl ) {
          if( waited )
                __sync_fetch_and_sub (latch->bits->waiters, 1);
          return;
        }
  }

  if( !waited ) {
        __sync_fetch_and_add (latch->bits->waiters, 1);
        *prev->bits->waiters += 1;
        waited++;
  }

  sys_futex (latch->value, FUTEX_WAIT_PRIVATE, *prev->value, NULL, NULL, 0);
 }
}

int bt_mutextry(MutexLatch *latch)
{
        return !__sync_lock_test_and_set (latch->bits->xcl, 1);
}

void bt_releasemutex(MutexLatch *latch)
{
        *latch->bits->xcl = 0;

        if( *latch->bits->waiters )
                sys_futex( latch->value, FUTEX_WAKE_PRIVATE, 1, NULL, NULL, 0 );
}

//      reader/writer lock implementation

void WriteLock (RWLock *lock, ushort tid, uint line)
{
        if( lock->tid == tid ) {
                lock->dup++;
                return;
        }
        bt_mutexlock (lock->xcl);
        bt_mutexlock (lock->wrt);
        bt_releasemutex (lock->xcl);

        lock->line = line;
        lock->tid = tid;
}

void WriteRelease (RWLock *lock)
{
        if( lock->dup ) {
                lock->dup--;
                return;
        }
        lock->tid = 0;
        bt_releasemutex (lock->wrt);
}

void ReadLock (RWLock *lock, ushort tid)
{
        bt_mutexlock (lock->xcl);

        if( !__sync_fetch_and_add (&lock->readers, 1) )
                bt_mutexlock (lock->wrt);

        bt_releasemutex (lock->xcl);
}

void ReadRelease (RWLock *lock)
{
        if( __sync_fetch_and_sub (&lock->readers, 1) == 1 )
                bt_releasemutex (lock->wrt);
}

//      recovery manager -- flush dirty pages

void bt_flushlsn (BtMgr *mgr, ushort thread_no)
{
uint cnt3 = 0, cnt2 = 0, cnt = 0;
uint entry, segment;
BtLatchSet *latch;
BtPage page;

  //    flush dirty pool pages to the btree

fprintf(stderr, "Start flushlsn  ");
  for( entry = 1; entry < mgr->latchtotal; entry++ ) {
        page = (BtPage)(((uid)entry << mgr->page_bits) + mgr->pagepool);
        latch = mgr->latchsets + entry;
        bt_mutexlock (latch->modify);
        bt_lockpage(BtLockRead, latch, thread_no, __LINE__);

        if( latch->dirty ) {
          bt_writepage(mgr, page, latch->page_no, 0);
          latch->dirty = 0, cnt++;
        }
if( latch->pin & ~CLOCK_bit )
cnt2++;
        bt_unlockpage(BtLockRead, latch, thread_no, __LINE__);
        bt_releasemutex (latch->modify);
  }
  for( entry = 1; entry < mgr->leaftotal; entry++ ) {
        page = (BtPage)(((uid)entry << mgr->page_bits << mgr->leaf_xtra) + mgr->leafpool);
        latch = mgr->leafsets + entry;
        bt_mutexlock (latch->modify);
        bt_lockpage(BtLockRead, latch, thread_no, __LINE__);

        if( latch->dirty ) {
          bt_writepage(mgr, page, latch->page_no, 1);
          latch->dirty = 0, cnt++;
        }
if( latch->pin & ~CLOCK_bit )
cnt2++;
        bt_unlockpage(BtLockRead, latch, thread_no, __LINE__);
        bt_releasemutex (latch->modify);
  }
fprintf(stderr, "End flushlsn %d pages  %d pinned\n", cnt, cnt2);
fprintf(stderr, "begin sync");
        for( segment = 0; segment < mgr->segments; segment++ )
          if( msync (mgr->pages[segment], (uid)65536 << mgr->page_bits, MS_SYNC) < 0 )
                fprintf(stderr, "msync error %d line %d\n", errno, __LINE__);
fprintf(stderr, " end sync\n");
}

//      recovery manager -- process current recovery buff on startup
//      this won't do much if previous session was properly closed.

BTERR bt_recoveryredo (BtMgr *mgr)
{
BtLogHdr *hdr, *eof;
uint offset = 0;
BtKey *key;
BtVal *val;

  hdr = (BtLogHdr *)mgr->redobuff;
  mgr->flushlsn = hdr->lsn;

  while( 1 ) {
        hdr = (BtLogHdr *)(mgr->redobuff + offset);
        switch( hdr->type ) {
        case BTRM_eof:
                mgr->lsn = hdr->lsn;
                return 0;
        case BTRM_add:          // add a unique key-value to btree
        
        case BTRM_dup:          // add a duplicate key-value to btree
        case BTRM_del:          // delete a key-value from btree
        case BTRM_upd:          // update a key with a new value
        case BTRM_new:          // allocate a new empty page
        case BTRM_old:          // reuse an old empty page
                return 0;
        }
  }
}

//      recovery manager -- append new entry to recovery log
//      flush dirty pages to disk when it overflows.

logseqno bt_newredo (BtMgr *mgr, BTRM type, int lvl, BtKey *key, BtVal *val, ushort thread_no)
{
uint size = mgr->page_size * mgr->pagezero->redopages - sizeof(BtLogHdr);
uint amt = sizeof(BtLogHdr);
BtLogHdr *hdr, *eof;
uint last, end;

        bt_mutexlock (mgr->redo);

        if( key )
          amt += key->len + val->len + sizeof(BtKey) + sizeof(BtVal);

        //      see if new entry fits in buffer
        //      flush and reset if it doesn't

        if( amt > size - mgr->redoend ) {
          mgr->flushlsn = mgr->lsn;
          if( msync (mgr->redobuff + (mgr->redolast & ~0xfff), mgr->redoend - (mgr->redolast & ~0xfff) + sizeof(BtLogHdr), MS_SYNC) < 0 )
                fprintf(stderr, "msync error %d line %d\n", errno, __LINE__);
          mgr->redolast = 0;
          mgr->redoend = 0;
          bt_flushlsn(mgr, thread_no);
        }

        //      fill in new entry & either eof or end block

        hdr = (BtLogHdr *)(mgr->redobuff + mgr->redoend);

        hdr->len = amt;
        hdr->type = type;
        hdr->lvl = lvl;
        hdr->lsn = ++mgr->lsn;

        mgr->redoend += amt;

        eof = (BtLogHdr *)(mgr->redobuff + mgr->redoend);
        memset (eof, 0, sizeof(BtLogHdr));

        //  fill in key and value

        if( key ) {
          memcpy ((unsigned char *)(hdr + 1), key, key->len + sizeof(BtKey));
          memcpy ((unsigned char *)(hdr + 1) + key->len + sizeof(BtKey), val, val->len + sizeof(BtVal));
        }

        eof = (BtLogHdr *)(mgr->redobuff + mgr->redoend);
        memset (eof, 0, sizeof(BtLogHdr));
        eof->lsn = mgr->lsn;

        last = mgr->redolast & ~0xfff;
        end = mgr->redoend;

        if( end - last + sizeof(BtLogHdr) >= 32768 )
          if( msync (mgr->redobuff + last, end - last + sizeof(BtLogHdr), MS_SYNC) < 0 )
                fprintf(stderr, "msync error %d line %d\n", errno, __LINE__);
          else
                mgr->redolast = end;

        bt_releasemutex(mgr->redo);
        return hdr->lsn;
}

//      recovery manager -- append transaction to recovery log
//      flush dirty pages to disk when it overflows.

logseqno bt_txnredo (BtMgr *mgr, BtPage source, ushort thread_no)
{
uint size = mgr->page_size * mgr->pagezero->redopages - sizeof(BtLogHdr);
uint amt = 0, src, type;
BtLogHdr *hdr, *eof;
uint last, end;
logseqno lsn;
BtKey *key;
BtVal *val;

        //      determine amount of redo recovery log space required

        for( src = 0; src++ < source->cnt; ) {
          key = keyptr(source,src);
          val = valptr(source,src);
          amt += key->len + val->len + sizeof(BtKey) + sizeof(BtVal);
          amt += sizeof(BtLogHdr);
        }

        bt_mutexlock (mgr->redo);

        //      see if new entry fits in buffer
        //      flush and reset if it doesn't

        if( amt > size - mgr->redoend ) {
          mgr->flushlsn = mgr->lsn;
          if( msync (mgr->redobuff + (mgr->redolast & ~0xfff), mgr->redoend - (mgr->redolast & ~0xfff) + sizeof(BtLogHdr), MS_SYNC) < 0 )
                fprintf(stderr, "msync error %d line %d\n", errno, __LINE__);
          mgr->redolast = 0;
          mgr->redoend = 0;
          bt_flushlsn (mgr, thread_no);
        }

        //      assign new lsn to transaction

        lsn = ++mgr->lsn;

        //      fill in new entries

        for( src = 0; src++ < source->cnt; ) {
          key = keyptr(source, src);
          val = valptr(source, src);

          switch( slotptr(source, src)->type ) {
          case Unique:
                type = BTRM_add;
                break;
          case Duplicate:
                type = BTRM_dup;
                break;
          case Delete:
                type = BTRM_del;
                break;
          }

          amt = key->len + val->len + sizeof(BtKey) + sizeof(BtVal);
          amt += sizeof(BtLogHdr);

          hdr = (BtLogHdr *)(mgr->redobuff + mgr->redoend);
          hdr->len = amt;
          hdr->type = type;
          hdr->lsn = lsn;
          hdr->lvl = 0;

          //  fill in key and value

          memcpy ((unsigned char *)(hdr + 1), key, key->len + sizeof(BtKey));
          memcpy ((unsigned char *)(hdr + 1) + key->len + sizeof(BtKey), val, val->len + sizeof(BtVal));

          mgr->redoend += amt;
        }

        eof = (BtLogHdr *)(mgr->redobuff + mgr->redoend);
        memset (eof, 0, sizeof(BtLogHdr));
        eof->lsn = lsn;

        last = mgr->redolast & ~0xfff;
        end = mgr->redoend;

        if( end - last + sizeof(BtLogHdr) >= 32768 )
          if( msync (mgr->redobuff + last, end - last + sizeof(BtLogHdr), MS_SYNC) < 0 )
                fprintf(stderr, "msync error %d line %d\n", errno, __LINE__);
          else
                mgr->redolast = end;

        bt_releasemutex(mgr->redo);
        return lsn;
}

//      sync a single btree page to disk

BTERR bt_syncpage (BtMgr *mgr, BtPage page, BtLatchSet *latch)
{
uint segment = latch->page_no >> 16;
uint page_size = mgr->page_size;
BtPage perm;

        if( bt_writepage (mgr, page, latch->page_no, latch->leaf) )
                return mgr->err;

        if( !page->lvl )
                page_size <<= mgr->leaf_xtra;

        perm = (BtPage)(mgr->pages[segment] + ((latch->page_no & 0xffff) << mgr->page_bits));

        if( msync (perm, page_size, MS_SYNC) < 0 )
          fprintf(stderr, "msync error %d line %d\n", errno, __LINE__);

        latch->dirty = 0;
        return 0;
}

//      read page into buffer pool from permanent location in Btree file

BTERR bt_readpage (BtMgr *mgr, BtPage page, uid page_no, uint leaf)
{
int flag = PROT_READ | PROT_WRITE;
uint page_size = mgr->page_size;
uint off = 0, segment, fragment;
unsigned char *perm;

  if( leaf )
        page_size <<= mgr->leaf_xtra;

  fragment = page_no & 0xffff;
  segment = page_no >> 16;
  mgr->reads++;

  while( off < page_size ) {
        if( fragment >> 16 )
                segment++, fragment = 0;

        if( segment < mgr->segments ) {
          perm = mgr->pages[segment] + ((uid)fragment << mgr->page_bits);

          memcpy ((unsigned char *)page + off, perm, mgr->page_size);
          off += mgr->page_size;
          fragment++;
          continue;
        }

        bt_mutexlock (mgr->maps);

        if( segment < mgr->segments ) {
                bt_releasemutex (mgr->maps);
                continue;
        }

        mgr->pages[mgr->segments] = mmap (0, (uid)65536 << mgr->page_bits, flag, MAP_SHARED, mgr->idx, (uid)mgr->segments << (mgr->page_bits + 16));
        mgr->segments++;

        bt_releasemutex (mgr->maps);
  }

  return 0;
}

//      write page to permanent location in Btree file

BTERR bt_writepage (BtMgr *mgr, BtPage page, uid page_no, uint leaf)
{
int flag = PROT_READ | PROT_WRITE;
uint page_size = mgr->page_size;
uint off = 0, segment, fragment;
unsigned char *perm;

  if( leaf )
        page_size <<= mgr->leaf_xtra;

  fragment = page_no & 0xffff;
  segment = page_no >> 16;
  mgr->writes++;

  while( off < page_size ) {
        if( fragment >> 16 )
                segment++, fragment = 0;

        if( segment < mgr->segments ) {
          perm = mgr->pages[segment] + ((uid)fragment << mgr->page_bits);
          memcpy (perm, (unsigned char *)page + off, mgr->page_size);
          off += mgr->page_size;
          fragment++;
          continue;
        }

        bt_mutexlock (mgr->maps);

        if( segment < mgr->segments ) {
                bt_releasemutex (mgr->maps);
                continue;
        }

        mgr->pages[mgr->segments] = mmap (0, (uid)65536 << mgr->page_bits, flag, MAP_SHARED, mgr->idx, (uid)mgr->segments << (mgr->page_bits + 16));
        mgr->segments++;
        bt_releasemutex (mgr->maps);
  }

  return 0;
}

//      set CLOCK bit in latch
//      decrement pin count

int value;

void bt_unpinlatch (BtLatchSet *latch, ushort thread_no, uint line)
{
        bt_mutexlock(latch->modify);
        latch->pin |= CLOCK_bit;
        latch->pin--;

        bt_releasemutex(latch->modify);
}

//  return the btree cached page address

BtPage bt_mappage (BtMgr *mgr, BtLatchSet *latch)
{
uid entry = latch - (latch->leaf ? mgr->leafsets : mgr->latchsets);
BtPage page;

  if( latch->leaf )
        page = (BtPage)((entry << mgr->page_bits << mgr->leaf_xtra) + mgr->leafpool);
  else
        page = (BtPage)((entry << mgr->page_bits) + mgr->pagepool);

  return page;
}

//  return next available leaf entry
//        and with latch entry locked

uint bt_availleaf (BtMgr *mgr)
{
BtLatchSet *latch;
uint entry;

  while( 1 ) {
#ifdef unix
        entry = __sync_fetch_and_add (&mgr->leafvictim, 1) + 1;
#else
        entry = _InterlockedIncrement (&mgr->leafvictim);
#endif
        entry %= mgr->leaftotal;

        if( !entry )
                continue;

        latch = mgr->leafsets + entry;

        if( !bt_mutextry(latch->modify) )
                continue;

        //  return this entry if it is not pinned

        if( !latch->pin )
                return entry;

        //      if the CLOCK bit is set
        //      reset it to zero.

        latch->pin &= ~CLOCK_bit;
        bt_releasemutex(latch->modify);
  }
}

//  return next available latch entry
//        and with latch entry locked

uint bt_availnext (BtMgr *mgr)
{
BtLatchSet *latch;
uint entry;

  while( 1 ) {
#ifdef unix
        entry = __sync_fetch_and_add (&mgr->latchvictim, 1) + 1;
#else
        entry = _InterlockedIncrement (&mgr->latchvictim);
#endif
        entry %= mgr->latchtotal;

        if( !entry )
                continue;

        latch = mgr->latchsets + entry;

        if( !bt_mutextry(latch->modify) )
                continue;

        //  return this entry if it is not pinned

        if( !latch->pin )
                return entry;

        //      if the CLOCK bit is set
        //      reset it to zero.

        latch->pin &= ~CLOCK_bit;
        bt_releasemutex(latch->modify);
  }
}

//      pin leaf in leaf buffer pool
//      return with latchset pinned

BtLatchSet *bt_pinleaf (BtMgr *mgr, uid page_no, ushort thread_no)
{
uint hashidx = page_no % mgr->leafhash;
BtLatchSet *latch;
uint entry, idx;
BtPage page;

  //  try to find our entry

  bt_mutexlock(mgr->leaftable[hashidx].latch);

  if( entry = mgr->leaftable[hashidx].entry )
   do {
        latch = mgr->leafsets + entry;

        if( page_no == latch->page_no )
                break;
   } while( entry = latch->next );

  //  found our entry: increment pin

  if( entry ) {
        latch = mgr->leafsets + entry;
        bt_mutexlock(latch->modify);
        latch->pin |= CLOCK_bit;
        latch->pin++;

        bt_releasemutex(latch->modify);
        bt_releasemutex(mgr->leaftable[hashidx].latch);
        return latch;
  }

  //  find and reuse unpinned entry

trynext:
  entry = bt_availleaf (mgr);
  latch = mgr->leafsets + entry;

  idx = latch->page_no % mgr->leafhash;

  //  if latch is on a different hash chain
  //    unlink from the old page_no chain

  if( latch->page_no )
   if( idx != hashidx ) {

        //  skip over this entry if latch not available

    if( !bt_mutextry (mgr->leaftable[idx].latch) ) {
          bt_releasemutex(latch->modify);
          goto trynext;
        }

        if( latch->prev )
          mgr->leafsets[latch->prev].next = latch->next;
        else
          mgr->leaftable[idx].entry = latch->next;

        if( latch->next )
          mgr->leafsets[latch->next].prev = latch->prev;

    bt_releasemutex (mgr->leaftable[idx].latch);
   }

  page = (BtPage)(((uid)entry << mgr->page_bits << mgr->leaf_xtra) + mgr->leafpool);

  //  update permanent page area in btree from buffer pool
  //    no read-lock is required since page is not pinned.

  if( latch->dirty )
        if( mgr->err = bt_writepage (mgr, page, latch->page_no, 1) )
          return mgr->line = __LINE__, mgr->err_thread = thread_no, NULL;
        else
          latch->dirty = 0;

  if( bt_readpage (mgr, page, page_no, 1) )
        return mgr->line = __LINE__, NULL;

  //  link page as head of hash table chain
  //  if this is a never before used entry,
  //  or it was previously on a different
  //  hash table chain. Otherwise, just
  //  leave it in its current hash table
  //  chain position.

  if( !latch->page_no || hashidx != idx ) {
        if( latch->next = mgr->leaftable[hashidx].entry )
          mgr->leafsets[latch->next].prev = entry;

        mgr->leaftable[hashidx].entry = entry;
    latch->prev = 0;
  }

  //  fill in latch structure

  latch->pin = CLOCK_bit | 1;
  latch->page_no = page_no;
  latch->leaf = 1;

  bt_releasemutex (latch->modify);
  bt_releasemutex (mgr->leaftable[hashidx].latch);
  return latch;
}

//      pin page in non-leaf buffer pool
//      return with latchset pinned

BtLatchSet *bt_pinlatch (BtMgr *mgr, uid page_no, ushort thread_no)
{
uint hashidx = page_no % mgr->latchhash;
BtLatchSet *latch;
uint entry, idx;
BtPage page;

  //  try to find our entry

  bt_mutexlock(mgr->hashtable[hashidx].latch);

  if( entry = mgr->hashtable[hashidx].entry ) do
  {
        latch = mgr->latchsets + entry;

        if( page_no == latch->page_no )
                break;
  } while( entry = latch->next );

  //  found our entry: increment pin

  if( entry ) {
        latch = mgr->latchsets + entry;
        bt_mutexlock(latch->modify);
        latch->pin |= CLOCK_bit;
        latch->pin++;
        bt_releasemutex(latch->modify);
        bt_releasemutex(mgr->hashtable[hashidx].latch);
        return latch;
  }

  //  find and reuse unpinned entry

trynext:
  entry = bt_availnext (mgr);
  latch = mgr->latchsets + entry;

  idx = latch->page_no % mgr->latchhash;

  //  if latch is on a different hash chain
  //    unlink from the old page_no chain

  if( latch->page_no )
   if( idx != hashidx ) {

        //  skip over this entry if latch not available

    if( !bt_mutextry (mgr->hashtable[idx].latch) ) {
          bt_releasemutex(latch->modify);
          goto trynext;
        }

        if( latch->prev )
          mgr->latchsets[latch->prev].next = latch->next;
        else
          mgr->hashtable[idx].entry = latch->next;

        if( latch->next )
          mgr->latchsets[latch->next].prev = latch->prev;

    bt_releasemutex (mgr->hashtable[idx].latch);
   }

  page = (BtPage)(((uid)entry << mgr->page_bits) + mgr->pagepool);

  //  update permanent page area in btree from buffer pool
  //    no read-lock is required since page is not pinned.

  if( latch->dirty )
        if( mgr->err = bt_writepage (mgr, page, latch->page_no, 0) )
          return mgr->line = __LINE__, mgr->err_thread = thread_no, NULL;
        else
          latch->dirty = 0;

  if( bt_readpage (mgr, page, page_no, 0) )
        return mgr->line = __LINE__, NULL;

  //  link page as head of hash table chain
  //  if this is a never before used entry,
  //  or it was previously on a different
  //  hash table chain. Otherwise, just
  //  leave it in its current hash table
  //  chain position.

  if( !latch->page_no || hashidx != idx ) {
        if( latch->next = mgr->hashtable[hashidx].entry )
          mgr->latchsets[latch->next].prev = entry;

        mgr->hashtable[hashidx].entry = entry;
    latch->prev = 0;
  }

  //  fill in latch structure

  latch->pin = CLOCK_bit | 1;
  latch->page_no = page_no;
  latch->leaf = 0;

  bt_releasemutex (latch->modify);
  bt_releasemutex (mgr->hashtable[hashidx].latch);
  return latch;
}
  
void bt_mgrclose (BtMgr *mgr)
{
BtLatchSet *latch;
BtLogHdr *eof;
uint num = 0;
BtPage page;
uint slot;

        //      flush previously written dirty pages
        //      and write recovery buffer to disk

        fdatasync (mgr->idx);

        if( mgr->redoend ) {
                eof = (BtLogHdr *)(mgr->redobuff + mgr->redoend);
                memset (eof, 0, sizeof(BtLogHdr));
        }

        //      write remaining dirty pool pages to the btree

        for( slot = 1; slot < mgr->latchtotal; slot++ ) {
                page = (BtPage)(((uid)slot << mgr->page_bits) + mgr->pagepool);
                latch = mgr->latchsets + slot;

                if( latch->dirty ) {
                        bt_writepage(mgr, page, latch->page_no, 0);
                        latch->dirty = 0, num++;
                }
        }

        //      write remaining dirty leaf pages to the btree

        for( slot = 1; slot < mgr->leaftotal; slot++ ) {
                page = (BtPage)(((uid)slot << mgr->page_bits << mgr->leaf_xtra) + mgr->leafpool);
                latch = mgr->leafsets + slot;

                if( latch->dirty ) {
                        bt_writepage(mgr, page, latch->page_no, 1);
                        latch->dirty = 0, num++;
                }
        }

        //      clear redo recovery buffer on disk.

        if( mgr->pagezero->redopages ) {
                eof = (BtLogHdr *)mgr->redobuff;
                memset (eof, 0, sizeof(BtLogHdr));
                eof->lsn = mgr->lsn;
                if( msync (mgr->redobuff, 4096, MS_SYNC) < 0 )
                  fprintf(stderr, "msync error %d line %d\n", errno, __LINE__);
        }

        fprintf(stderr, "%d buffer pool pages flushed\n", num);

#ifdef unix
        while( mgr->segments )
                munmap (mgr->pages[--mgr->segments], (uid)65536 << mgr->page_bits);

        munmap (mgr->pagepool, (uid)mgr->nlatchpage << mgr->page_bits);
        munmap (mgr->leafpool, (uid)mgr->nleafpage << mgr->page_bits);
        munmap (mgr->pagezero, mgr->page_size);
#else
        FlushViewOfFile(mgr->pagezero, 0);
        UnmapViewOfFile(mgr->pagezero);
        UnmapViewOfFile(mgr->pagepool);
        CloseHandle(mgr->halloc);
        CloseHandle(mgr->hpool);
#endif
#ifdef unix
        close (mgr->idx);
        free (mgr);
#else
        VirtualFree (mgr->redobuff, 0, MEM_RELEASE);
        FlushFileBuffers(mgr->idx);
        CloseHandle(mgr->idx);
        GlobalFree (mgr);
#endif
}

//      close and release memory

void bt_close (BtDb *bt)
{
#ifdef unix
        if( bt->cursor )
                free (bt->cursor);
#else
        if( bt->cursor)
                VirtualFree (bt->cursor, 0, MEM_RELEASE);
#endif
        free (bt);
}

//  open/create new btree buffer manager

//      call with file_name, BT_openmode, bits in page size (e.g. 16),
//              size of page pool (e.g. 262144) and leaf pool

BtMgr *bt_mgr (char *name, uint pagebits, uint leafxtra, uint nodemax, uint leafmax, uint redopages)
{
uint lvl, attr, last, slot, idx;
uint nlatchpage, latchhash;
uint nleafpage, leafhash;
unsigned char value[BtId];
int flag, initit = 0;
BtPageZero *pagezero;
BtLatchSet *latch;
off64_t size;
uint amt[1];
BtMgr* mgr;
BtKey* key;
BtVal *val;

        // determine sanity of page size and buffer pool

        if( leafxtra + pagebits > BT_maxbits )
                fprintf (stderr, "pagebits + leafxtra > maxbits\n"), exit(1);

        if( pagebits < BT_minbits )
                fprintf (stderr, "pagebits < minbits\n"), exit(1);

#ifdef unix
        mgr = calloc (1, sizeof(BtMgr));

        mgr->idx = open ((char*)name, O_RDWR | O_CREAT, 0666);

        if( mgr->idx == -1 ) {
                fprintf (stderr, "Unable to create/open btree file %s\n", name);
                return free(mgr), NULL;
        }
#else
        mgr = GlobalAlloc (GMEM_FIXED|GMEM_ZEROINIT, sizeof(BtMgr));
        attr = FILE_ATTRIBUTE_NORMAL;
        mgr->idx = CreateFile(name, GENERIC_READ| GENERIC_WRITE, FILE_SHARE_READ|FILE_SHARE_WRITE, NULL, OPEN_ALWAYS, attr, NULL);

        if( mgr->idx == INVALID_HANDLE_VALUE ) {
                fprintf (stderr, "Unable to create/open btree file %s\n", name);
                return GlobalFree(mgr), NULL;
        }
#endif

#ifdef unix
        pagezero = valloc (BT_maxpage);
        *amt = 0;

        // read minimum page size to get root info
        //      to support raw disk partition files
        //      check if page_bits == 0 on the disk.

        if( size = lseek (mgr->idx, 0L, 2) )
                if( pread(mgr->idx, pagezero, BT_minpage, 0) == BT_minpage )
                        if( pagezero->page_bits ) {
                                pagebits = pagezero->page_bits;
                                leafxtra = pagezero->leaf_xtra;
                        } else
                                initit = 1;
                else
                        return free(mgr), free(pagezero), NULL;
        else
                initit = 1;
#else
        pagezero = VirtualAlloc(NULL, BT_maxpage, MEM_COMMIT, PAGE_READWRITE);
        size = GetFileSize(mgr->idx, amt);

        if( size || *amt ) {
                if( !ReadFile(mgr->idx, (char *)pagezero, BT_minpage, amt, NULL) )
                        return bt_mgrclose (mgr), NULL;
                pagebits = pagezero->page_bits;
                leafxtra = pagezero->leaf_xtra;
        } else
                initit = 1;
#endif

        mgr->page_size = 1 << pagebits;
        mgr->page_bits = pagebits;
        mgr->leaf_xtra = leafxtra;

        //  calculate number of latch hash table entries

        mgr->nlatchpage = ((uid)nodemax/16 * sizeof(BtHashEntry) + mgr->page_size - 1) >> mgr->page_bits;

        mgr->nlatchpage += nodemax;             // size of the buffer pool in pages
        mgr->nlatchpage += (sizeof(BtLatchSet) * nodemax + mgr->page_size - 1) >> mgr->page_bits;
        mgr->latchtotal = nodemax;

        //  calculate number of leaf hash table entries

        mgr->nleafpage = ((uid)leafmax/16 * sizeof(BtHashEntry) + mgr->page_size - 1) >> mgr->page_bits;

        mgr->nleafpage += leafmax << leafxtra;          // size of the leaf pool in pages
        mgr->nleafpage += (sizeof(BtLatchSet) * leafmax + mgr->page_size - 1) >> mgr->page_bits;
        mgr->leaftotal = leafmax;

        mgr->redopage = LEAF_page + (1 << leafxtra);

        if( !initit )
                goto mgrlatch;

        // initialize an empty b-tree with latch page, root page, page of leaves
        // and page(s) of latches and page pool cache

        ftruncate (mgr->idx, (mgr->redopage + pagezero->redopages) << mgr->page_bits);
        memset (pagezero, 0, 1 << pagebits);
        pagezero->alloc->lvl = MIN_lvl - 1;
        pagezero->redopages = redopages;
        pagezero->page_bits = mgr->page_bits;
        pagezero->leaf_xtra = leafxtra;

        bt_putid(pagezero->alloc->right, mgr->redopage + pagezero->redopages);
        pagezero->leafpages = 1;

        //  initialize left-most LEAF page in
        //      alloc->left and count of active leaf pages.

        bt_putid (pagezero->alloc->left, LEAF_page);

        if( bt_writepage (mgr, pagezero->alloc, 0, 0) ) {
                fprintf (stderr, "Unable to create btree page zero\n");
                return bt_mgrclose (mgr), NULL;
        }

        memset (pagezero, 0, 1 << pagebits);

        for( lvl=MIN_lvl; lvl--; ) {
        BtSlot *node = slotptr(pagezero->alloc, 1);
                node->off = mgr->page_size;

                if( !lvl )
                        node->off <<= mgr->leaf_xtra;

                node->off -= 3 + (lvl ? BtId + sizeof(BtVal): sizeof(BtVal));
                key = keyptr(pagezero->alloc, 1);
                key->len = 2;           // create stopper key
                key->key[0] = 0xff;
                key->key[1] = 0xff;

                bt_putid(value, MIN_lvl - lvl + 1);
                val = valptr(pagezero->alloc, 1);
                val->len = lvl ? BtId : 0;
                memcpy (val->value, value, val->len);

                pagezero->alloc->min = node->off;
                pagezero->alloc->lvl = lvl;
                pagezero->alloc->cnt = 1;
                pagezero->alloc->act = 1;
                pagezero->alloc->page_no = MIN_lvl - lvl;

                if( bt_writepage (mgr, pagezero->alloc, MIN_lvl - lvl, !lvl) ) {
                        fprintf (stderr, "Unable to create btree page\n");
                        return bt_mgrclose (mgr), NULL;
                }
        }

mgrlatch:
#ifdef unix
        free (pagezero);
#else
        VirtualFree (pagezero, 0, MEM_RELEASE);
#endif

        // mlock the first segment of 64K pages

        flag = PROT_READ | PROT_WRITE;
        mgr->pages[0] = mmap (0, (uid)65536 << mgr->page_bits, flag, MAP_SHARED, mgr->idx, 0);
        mgr->segments = 1;

        if( mgr->pages[0] == MAP_FAILED ) {
                fprintf (stderr, "Unable to mmap first btree segment, error = %d\n", errno);
                return bt_mgrclose (mgr), NULL;
        }

        mgr->pagezero = (BtPageZero *)mgr->pages[0];
        mlock (mgr->pagezero, mgr->page_size);

        mgr->redobuff = mgr->pages[0] + (mgr->redopage << mgr->page_bits);
        mlock (mgr->redobuff, mgr->pagezero->redopages << mgr->page_bits);

        //      allocate pool buffers

        mgr->pagepool = mmap (0, (uid)mgr->nlatchpage << mgr->page_bits, flag, MAP_ANONYMOUS | MAP_SHARED, -1, 0);
        if( mgr->pagepool == MAP_FAILED ) {
                fprintf (stderr, "Unable to mmap anonymous buffer pool pages, error = %d\n", errno);
                return bt_mgrclose (mgr), NULL;
        }

        mgr->leafpool = mmap (0, (uid)mgr->nleafpage << mgr->page_bits, flag, MAP_ANONYMOUS | MAP_SHARED, -1, 0);
        if( mgr->leafpool == MAP_FAILED ) {
                fprintf (stderr, "Unable to mmap anonymous leaf pool pages, error = %d\n", errno);
                return bt_mgrclose (mgr), NULL;
        }

        mgr->latchsets = (BtLatchSet *)(mgr->pagepool + ((uid)mgr->latchtotal << mgr->page_bits));
        mgr->hashtable = (BtHashEntry *)(mgr->latchsets + mgr->latchtotal);
        mgr->latchhash = (mgr->pagepool + ((uid)mgr->nlatchpage << mgr->page_bits) - (unsigned char *)mgr->hashtable) / sizeof(BtHashEntry);

        mgr->leafsets = (BtLatchSet *)(mgr->leafpool + ((uid)mgr->leaftotal << mgr->page_bits << mgr->leaf_xtra));
        mgr->leaftable = (BtHashEntry *)(mgr->leafsets + mgr->leaftotal);
        mgr->leafhash = (mgr->leafpool + ((uid)mgr->nleafpage << mgr->page_bits) - (unsigned char *)mgr->leaftable) / sizeof(BtHashEntry);

        return mgr;
}

//      open BTree access method
//      based on buffer manager

BtDb *bt_open (BtMgr *mgr, BtMgr *main)
{
BtDb *bt = malloc (sizeof(*bt));

        memset (bt, 0, sizeof(*bt));
        bt->main = main;
        bt->mgr = mgr;
#ifdef unix
        bt->cursor = valloc (mgr->page_size << mgr->leaf_xtra);
#else
        bt->cursor = VirtualAlloc(NULL, mgr->page_size << mgr->leaf_xtra, MEM_COMMIT, PAGE_READWRITE);
#endif
#ifdef unix
        bt->thread_no = __sync_fetch_and_add (mgr->thread_no, 1) + 1;
#else
        bt->thread_no = _InterlockedIncrement16(mgr->thread_no, 1);
#endif
        return bt;
}

//  compare two keys, return > 0, = 0, or < 0
//  =0: keys are same
//  -1: key2 > key1
//  +1: key2 < key1
//  as the comparison value

int keycmp (BtKey* key1, unsigned char *key2, uint len2)
{
uint len1 = key1->len;
int ans;

        if( ans = memcmp (key1->key, key2, len1 > len2 ? len2 : len1) )
                return ans;

        if( len1 > len2 )
                return 1;
        if( len1 < len2 )
                return -1;

        return 0;
}

// place write, read, or parent lock on requested page_no.

void bt_lockpage(BtLock mode, BtLatchSet *latch, ushort thread_no, uint line)
{
        switch( mode ) {
        case BtLockRead:
                ReadLock (latch->readwr, thread_no);
                break;
        case BtLockWrite:
//if(latch->leaf)
//fprintf(stderr, "WrtRqst %d by %d at %d\n", (uint)latch->page_no, thread_no, line); 
                WriteLock (latch->readwr, thread_no, line);
//if(latch->leaf)
//fprintf(stderr, "WrtLock %d by %d at %d\n", (uint)latch->page_no, thread_no, line); 
                break;
        case BtLockAccess:
                ReadLock (latch->access, thread_no);
                break;
        case BtLockDelete:
                WriteLock (latch->access, thread_no, line);
                break;
        case BtLockParent:
                WriteLock (latch->parent, thread_no, line);
                break;
        case BtLockLink:
                WriteLock (latch->link, thread_no, line);
                break;
        }
}

// remove write, read, or parent lock on requested page

void bt_unlockpage(BtLock mode, BtLatchSet *latch, ushort thread_no, uint line)
{
        switch( mode ) {
        case BtLockRead:
                ReadRelease (latch->readwr);
                break;
        case BtLockWrite:
//if(latch->leaf)
//fprintf(stderr, "Un Lock %d by %d at %d\n", (uint)latch->page_no, thread_no, line); 
                WriteRelease (latch->readwr);
                break;
        case BtLockAccess:
                ReadRelease (latch->access);
                break;
        case BtLockDelete:
                WriteRelease (latch->access);
                break;
        case BtLockParent:
                WriteRelease (latch->parent);
                break;
        case BtLockLink:
                WriteRelease (latch->link);
                break;
        }
}

//      allocate a new page
//      return with page latched, but unlocked.

int bt_newpage(BtMgr *mgr, BtPageSet *set, BtPage contents, ushort thread_no)
{
uint page_size = mgr->page_size, page_xtra = 0;
unsigned char *freechain;
uid page_no;

        if( contents->lvl )
                freechain = mgr->pagezero->freechain;
        else {
                freechain = mgr->pagezero->leafchain;
                mgr->pagezero->leafpages++;
                page_xtra = mgr->leaf_xtra;
                page_size <<= page_xtra;
        }

        //      lock allocation page

        bt_mutexlock(mgr->lock);

        // use empty chain first
        // else allocate new page

        if( page_no = bt_getid(freechain) ) {
                if( set->latch = contents->lvl ? bt_pinlatch (mgr, page_no, thread_no) : bt_pinleaf (mgr, page_no, thread_no) )
                        set->page = bt_mappage (mgr, set->latch);
                else
                        return mgr->line = __LINE__, mgr->err_thread = thread_no, mgr->err = BTERR_struct;

                bt_putid(freechain, bt_getid(set->page->right));

                //  the page is currently nopromote and this
                //  will keep bt_txnpromote out.

                //      contents will replace this bit
                //  and pin will keep bt_txnpromote out

                contents->page_no = page_no;
                contents->nopromote = 0;
                set->latch->dirty = 1;

                memcpy (set->page, contents, page_size);

//              if( msync (mgr->pagezero, mgr->page_size, MS_SYNC) < 0 )
//                fprintf(stderr, "msync error %d line %d\n", errno, __LINE__);

                bt_releasemutex(mgr->lock);
                return 0;
        }

        page_no = bt_getid(mgr->pagezero->alloc->right);
        bt_putid(mgr->pagezero->alloc->right, page_no+(1 << page_xtra));

        // unlock allocation latch and
        //      extend file into new page.

//      if( msync (mgr->pagezero, mgr->page_size, MS_SYNC) < 0 )
//        fprintf(stderr, "msync error %d line %d\n", errno, __LINE__);
        bt_releasemutex(mgr->lock);

        //      keep bt_txnpromote out of this page
        contents->nopromote = 1;
        contents->page_no = page_no;
        if( pwrite (mgr->idx, contents, page_size, page_no << mgr->page_bits) < page_size )
                fprintf(stderr, "Write %d error %d\n", (uint)page_no, errno);
        if( set->latch = contents->lvl ? bt_pinlatch (mgr, page_no, thread_no) : bt_pinleaf (mgr, page_no, thread_no) )
                set->page = bt_mappage (mgr, set->latch);
        else
                return mgr->err_thread = thread_no, mgr->err;

        // now pin will keep bt_txnpromote out

        set->page->nopromote = 0;
        set->latch->dirty = 1;
        return 0;
}

//  find slot in page for given key at a given level

int bt_findslot (BtPage page, unsigned char *key, uint len)
{
uint diff, higher = page->cnt, low = 1, slot;
uint good = 0;

        //        make stopper key an infinite fence value

        if( bt_getid (page->right) )
                higher++;
        else
                good++;

        //      low is the lowest candidate.
        //  loop ends when they meet

        //  higher is already
        //      tested as .ge. the passed key.

        while( diff = higher - low ) {
                slot = low + ( diff >> 1 );
                if( keycmp (keyptr(page, slot), key, len) < 0 )
                        low = slot + 1;
                else
                        higher = slot, good++;
        }

        //      return zero if key is on right link page

        return good ? higher : 0;
}

//  find and load page at given level for given key
//      leave page rd or wr locked as requested

int bt_loadpage (BtMgr *mgr, BtPageSet *set, unsigned char *key, uint len, uint lvl, BtLock lock, ushort thread_no)
{
uid page_no = ROOT_page, prevpage_no = 0;
uint drill = 0xff, slot;
BtLatchSet *prevlatch;
uint mode, prevmode;
BtPage prevpage;
BtVal *val;
BtKey *ptr;

  //  start at root of btree and drill down

  do {
        // determine lock mode of drill level
        mode = (drill == lvl) ? lock : BtLockRead; 

        if( set->latch = drill ? bt_pinlatch (mgr, page_no, thread_no) : bt_pinleaf (mgr, page_no, thread_no) )
          set->page = bt_mappage (mgr, set->latch);
        else
          return 0;

        // obtain access lock using lock chaining with Access mode

        if( page_no > ROOT_page )
          bt_lockpage(BtLockAccess, set->latch, thread_no, __LINE__);

        //      release & unpin parent or left sibling page

        if( prevpage_no ) {
          bt_unlockpage(prevmode, prevlatch, thread_no, __LINE__);
          bt_unpinlatch (prevlatch, thread_no, __LINE__);
          prevpage_no = 0;
        }

        // obtain mode lock using lock coupling through AccessLock

        bt_lockpage(mode, set->latch, thread_no, __LINE__);

        // grab our fence key

        ptr=keyptr(set->page,set->page->cnt);

        if( set->page->free )
                return mgr->err = BTERR_struct, mgr->err_thread = thread_no, mgr->line = __LINE__, 0;

        if( page_no > ROOT_page )
          bt_unlockpage(BtLockAccess, set->latch, thread_no, __LINE__);

        // re-read and re-lock root after determining actual level of root

        if( set->page->lvl != drill) {
                if( set->latch->page_no != ROOT_page )
                        return mgr->err = BTERR_struct, mgr->err_thread = thread_no, mgr->line = __LINE__, 0;
                        
                drill = set->page->lvl;

                if( lock != BtLockRead && drill == lvl ) {
                  bt_unlockpage(mode, set->latch, thread_no, __LINE__);
                  bt_unpinlatch (set->latch, thread_no, __LINE__);
                  continue;
                }
        }

        prevpage_no = set->latch->page_no;
        prevlatch = set->latch;
        prevpage = set->page;
        prevmode = mode;

        //  if requested key is beyond our fence,
        //      slide to the right

        if( keycmp (ptr, key, len) < 0 )
          if( page_no = bt_getid(set->page->right) )
                continue;

        //  if page is part of a delete operation,
        //      slide to the left;

        if( set->page->kill ) {
          bt_lockpage(BtLockLink, set->latch, thread_no, __LINE__);
          page_no = bt_getid(set->page->left);
          bt_unlockpage(BtLockLink, set->latch, thread_no, __LINE__);
          continue;
        }

        //  find key on page at this level
        //  and descend to requested level

        if( slot = bt_findslot (set->page, key, len) ) {
          if( drill == lvl )
                return slot;

          // find next non-dead slot -- the fence key if nothing else

          while( slotptr(set->page, slot)->dead )
                if( slot++ < set->page->cnt )
                  continue;
                else
                  return mgr->err = BTERR_struct, mgr->err_thread = thread_no, mgr->line = __LINE__, 0;

          val = valptr(set->page, slot);

          if( val->len == BtId )
                page_no = bt_getid(val->value);
          else
                return mgr->line = __LINE__, mgr->err_thread = thread_no, mgr->err = BTERR_struct, 0;

          drill--;
          continue;
         }

        //  slide right into next page

        page_no = bt_getid(set->page->right);

  } while( page_no );

  // return error on end of right chain

  mgr->line = __LINE__, mgr->err_thread = thread_no, mgr->err = BTERR_struct;
  return 0;     // return error
}

//      return page to free list
//      page must be delete & write locked
//      and have no keys pointing to it.

void bt_freepage (BtMgr *mgr, BtPageSet *set, ushort thread_no)
{
unsigned char *freechain;

        if( set->page->lvl )
                freechain = mgr->pagezero->freechain;
        else {
                freechain = mgr->pagezero->leafchain;
                mgr->pagezero->leafpages--;
        }

        //      lock allocation page

        bt_mutexlock (mgr->lock);

        //      store chain

        memcpy(set->page->right, freechain, BtId);
        bt_putid(freechain, set->latch->page_no);
        set->latch->dirty = 1;
        set->page->free = 1;

//      if( msync (mgr->pagezero, mgr->page_size, MS_SYNC) < 0 )
//        fprintf(stderr, "msync error %d line %d\n", errno, __LINE__);

        // unlock released page
        // and unlock allocation page

        bt_unlockpage (BtLockDelete, set->latch, thread_no, __LINE__);
        bt_unlockpage (BtLockWrite, set->latch, thread_no, __LINE__);
        bt_unpinlatch (set->latch, thread_no, __LINE__);
        bt_releasemutex (mgr->lock);
}

//      a fence key was deleted from a page
//      push new fence value upwards

BTERR bt_fixfence (BtMgr *mgr, BtPageSet *set, uint lvl, ushort thread_no)
{
unsigned char leftkey[BT_keyarray], rightkey[BT_keyarray];
unsigned char value[BtId];
BtKey* ptr;
uint idx;

        //      remove the old fence value

        ptr = keyptr(set->page, set->page->cnt);
        memcpy (rightkey, ptr, ptr->len + sizeof(BtKey));
        memset (slotptr(set->page, set->page->cnt--), 0, sizeof(BtSlot));
        set->latch->dirty = 1;

        //  cache new fence value

        ptr = keyptr(set->page, set->page->cnt);
        memcpy (leftkey, ptr, ptr->len + sizeof(BtKey));

        bt_lockpage (BtLockParent, set->latch, thread_no, __LINE__);
        bt_unlockpage (BtLockWrite, set->latch, thread_no, __LINE__);

        //      insert new (now smaller) fence key

        bt_putid (value, set->latch->page_no);
        ptr = (BtKey*)leftkey;

        if( bt_insertkey (mgr, ptr->key, ptr->len, lvl+1, value, BtId, Unique, thread_no) )
          return mgr->err_thread = thread_no, mgr->err;

        //      now delete old fence key

        ptr = (BtKey*)rightkey;

        if( bt_deletekey (mgr, ptr->key, ptr->len, lvl+1, thread_no) )
                return mgr->err_thread = thread_no, mgr->err;

        bt_unlockpage (BtLockParent, set->latch, thread_no, __LINE__);
        bt_unpinlatch(set->latch, thread_no, __LINE__);
        return 0;
}

//      root has a single child
//      collapse a level from the tree

BTERR bt_collapseroot (BtMgr *mgr, BtPageSet *root, ushort thread_no)
{
BtPageSet child[1];
uid page_no;
BtVal *val;
uint idx;

  // find the child entry and promote as new root contents

  do {
        for( idx = 0; idx++ < root->page->cnt; )
          if( !slotptr(root->page, idx)->dead )
                break;

        val = valptr(root->page, idx);

        if( val->len == BtId )
                page_no = bt_getid (valptr(root->page, idx)->value);
        else
                return mgr->line = __LINE__, mgr->err_thread = thread_no, mgr->err = BTERR_struct;

        if( child->latch = bt_pinlatch (mgr, page_no, thread_no) )
                child->page = bt_mappage (mgr, child->latch);
        else
                return mgr->err_thread = thread_no, mgr->err;

        bt_lockpage (BtLockDelete, child->latch, thread_no, __LINE__);
        bt_lockpage (BtLockWrite, child->latch, thread_no, __LINE__);

        memcpy (root->page, child->page, mgr->page_size);
        root->latch->dirty = 1;

        bt_freepage (mgr, child, thread_no);

  } while( root->page->lvl > 1 && root->page->act == 1 );

  bt_unlockpage (BtLockWrite, root->latch, thread_no, __LINE__);
  bt_unpinlatch (root->latch, thread_no, __LINE__);
  return 0;
}

//  delete a page and manage key
//  call with page writelocked

//      returns with page unpinned
//      from the page pool.

BTERR bt_deletepage (BtMgr *mgr, BtPageSet *set, ushort thread_no, uint lvl)
{
unsigned char lowerfence[BT_keyarray];
uint page_size = mgr->page_size;
BtPageSet right[1], temp[1];
unsigned char value[BtId];
uid page_no, right2;
BtKey *ptr;

        if( !lvl )
                page_size <<= mgr->leaf_xtra;

        //  cache copy of original fence key
        //      that is being deleted.

        ptr = keyptr(set->page, set->page->cnt);
        memcpy (lowerfence, ptr, ptr->len + sizeof(BtKey));

        //      obtain locks on right page

        page_no = bt_getid(set->page->right);

        if( right->latch = lvl ? bt_pinlatch (mgr, page_no, thread_no) : bt_pinleaf (mgr, page_no, thread_no) )
                right->page = bt_mappage (mgr, right->latch);
        else
                return 0;

        bt_lockpage (BtLockWrite, right->latch, thread_no, __LINE__);

        if( right->page->kill )
                return mgr->line = __LINE__, mgr->err = BTERR_struct;

        // pull contents of right peer into our empty page
        //      preserving our left page number.

        bt_lockpage (BtLockLink, set->latch, thread_no, __LINE__);
        page_no = bt_getid(set->page->left);
        memcpy (set->page, right->page, page_size);
        bt_putid (set->page->left, page_no);
        bt_unlockpage (BtLockLink, set->latch, thread_no, __LINE__);

        set->page->page_no = set->latch->page_no;
        set->latch->dirty = 1;

        if( right2 = bt_getid (right->page->right) ) {
          if( temp->latch = lvl ? bt_pinlatch (mgr, right2, thread_no) : bt_pinleaf (mgr, right2, thread_no) )
                temp->page = bt_mappage (mgr, temp->latch);
          else
                return 0;

      bt_lockpage(BtLockLink, temp->latch, thread_no, __LINE__);
          bt_putid (temp->page->left, set->latch->page_no);
          bt_unlockpage(BtLockLink, temp->latch, thread_no, __LINE__);
          bt_unpinlatch (temp->latch, thread_no, __LINE__);
        } else {        // page is rightmost
          bt_mutexlock (mgr->lock);
          bt_putid (mgr->pagezero->alloc->left, set->latch->page_no);
          bt_releasemutex(mgr->lock);
        }

        // mark right page deleted and release lock

        right->latch->dirty = 1;
        right->page->kill = 1;

        // redirect higher key directly to our new node contents

        ptr = keyptr(right->page, right->page->cnt);
        bt_putid (value, set->latch->page_no);

        if( bt_insertkey (mgr, ptr->key, ptr->len, lvl+1, value, BtId, Update, thread_no) )
          return mgr->err;

        bt_unlockpage (BtLockWrite, right->latch, thread_no, __LINE__);

        //      delete orignal fence key in parent
        //      and unlock our page.

        ptr = (BtKey *)lowerfence;

        if( bt_deletekey (mgr, ptr->key, ptr->len, lvl+1, thread_no) )
          return mgr->err;

        bt_unlockpage (BtLockWrite, set->latch, thread_no, __LINE__);
        bt_unpinlatch (set->latch, thread_no, __LINE__);

        //      obtain delete and write locks to right node

        bt_lockpage (BtLockDelete, right->latch, thread_no, __LINE__);
        bt_lockpage (BtLockWrite, right->latch, thread_no, __LINE__);
        bt_freepage (mgr, right, thread_no);
        return 0;
}

//  find and delete key on page by marking delete flag bit
//  if page becomes empty, delete it from the btree

BTERR bt_deletekey (BtMgr *mgr, unsigned char *key, uint len, uint lvl, ushort thread_no)
{
uint slot, idx, found, fence;
BtPageSet set[1];
BtSlot *node;
BtKey *ptr;
BtVal *val;

        if( slot = bt_loadpage (mgr, set, key, len, lvl, BtLockWrite, thread_no) ) {
                node = slotptr(set->page, slot);
                ptr = keyptr(set->page, slot);
        } else
                return mgr->err_thread = thread_no, mgr->err;

        // if librarian slot, advance to real slot

        if( node->type == Librarian ) {
                ptr = keyptr(set->page, ++slot);
                node = slotptr(set->page, slot);
        }

        fence = slot == set->page->cnt;

        // delete the key, ignore request if already dead

        if( found = !keycmp (ptr, key, len) )
          if( found = node->dead == 0 ) {
                val = valptr(set->page,slot);
                set->page->garbage += ptr->len + val->len + sizeof(BtKey) + sizeof(BtVal);
                set->page->act--;

                //  mark node type as delete

                node->type = Delete;
                node->dead = 1;

                // collapse empty slots beneath the fence
                // on interiour nodes

                if( lvl )
                 while( idx = set->page->cnt - 1 )
                  if( slotptr(set->page, idx)->dead ) {
                        *slotptr(set->page, idx) = *slotptr(set->page, idx + 1);
                        memset (slotptr(set->page, set->page->cnt--), 0, sizeof(BtSlot));
                  } else
                        break;
          }

        if( !found )
                return 0;

        //      did we delete a fence key in an upper level?

        if( lvl && set->page->act && fence )
          if( bt_fixfence (mgr, set, lvl, thread_no) )
                return mgr->err_thread = thread_no, mgr->err;
          else
                return 0;

        //      do we need to collapse root?

        if( set->latch->page_no == ROOT_page && set->page->act == 1 )
          if( bt_collapseroot (mgr, set, thread_no) )
                return mgr->err_thread = thread_no, mgr->err;
          else
                return 0;

        //      delete empty page

        if( !set->page->act )
          return bt_deletepage (mgr, set, thread_no, set->page->lvl);

        set->latch->dirty = 1;
        bt_unlockpage(BtLockWrite, set->latch, thread_no, __LINE__);
        bt_unpinlatch (set->latch, thread_no, __LINE__);
        return 0;
}

//      advance to next slot

uint bt_findnext (BtDb *bt, BtPageSet *set, uint slot)
{
BtLatchSet *prevlatch;
uid page_no;

        if( slot < set->page->cnt )
                return slot + 1;

        prevlatch = set->latch;

        if( page_no = bt_getid(set->page->right) )
          if( set->latch = bt_pinleaf (bt->mgr, page_no, bt->thread_no) )
                set->page = bt_mappage (bt->mgr, set->latch);
          else
                return 0;
        else
          return bt->mgr->err = BTERR_struct, bt->mgr->err_thread = bt->thread_no, bt->mgr->line = __LINE__, 0;

        // obtain access lock using lock chaining with Access mode

        bt_lockpage(BtLockAccess, set->latch, bt->thread_no, __LINE__);

        bt_unlockpage(BtLockRead, prevlatch, bt->thread_no, __LINE__);
        bt_unpinlatch (prevlatch, bt->thread_no, __LINE__);

        bt_lockpage(BtLockRead, set->latch, bt->thread_no, __LINE__);
        bt_unlockpage(BtLockAccess, set->latch, bt->thread_no, __LINE__);
        return 1;
}

//      find unique key == given key, or first duplicate key in
//      leaf level and return number of value bytes
//      or (-1) if not found.

int bt_findkey (BtDb *bt, unsigned char *key, uint keylen, unsigned char *value, uint valmax)
{
BtPageSet set[1];
uint len, slot;
int ret = -1;
BtKey *ptr;
BtVal *val;

  if( slot = bt_loadpage (bt->mgr, set, key, keylen, 0, BtLockRead, bt->thread_no) )
   do {
        ptr = keyptr(set->page, slot);

        //      skip librarian slot place holder

        if( slotptr(set->page, slot)->type == Librarian )
                ptr = keyptr(set->page, ++slot);

        //      return actual key found

        memcpy (bt->key, ptr, ptr->len + sizeof(BtKey));
        len = ptr->len;

        if( slotptr(set->page, slot)->type == Duplicate )
                len -= BtId;

        //      not there if we reach the stopper key

        if( slot == set->page->cnt )
          if( !bt_getid (set->page->right) )
                break;

        // if key exists, return >= 0 value bytes copied
        //      otherwise return (-1)

        if( slotptr(set->page, slot)->dead )
                continue;

        if( keylen == len )
          if( !memcmp (ptr->key, key, len) ) {
                val = valptr (set->page,slot);
                if( valmax > val->len )
                        valmax = val->len;
                memcpy (value, val->value, valmax);
                ret = valmax;
          }

        break;

   } while( slot = bt_findnext (bt, set, slot) );

  bt_unlockpage (BtLockRead, set->latch, bt->thread_no, __LINE__);
  bt_unpinlatch (set->latch, bt->thread_no, __LINE__);
  return ret;
}

//      check page for space available,
//      clean if necessary and return
//      0 - page needs splitting
//      >0  new slot value

uint bt_cleanpage(BtMgr *mgr, BtPageSet *set, uint keylen, uint slot, uint vallen)
{
uint page_size = mgr->page_size;
BtPage page = set->page, frame;
uint cnt = 0, idx = 0;
uint max = page->cnt;
uint newslot = max;
BtKey *key;
BtVal *val;

        if( !set->page->lvl )
                page_size <<= mgr->leaf_xtra;

        if( page->min >= (max+2) * sizeof(BtSlot) + sizeof(*page) + keylen + sizeof(BtKey) + vallen + sizeof(BtVal))
                return slot;

        //      skip cleanup and proceed to split
        //      if there's not enough garbage
        //      to bother with.

        if( page->garbage < page_size / 5 )
                return 0;

        frame = malloc (page_size);
        memcpy (frame, page, page_size);

        // skip page info and set rest of page to zero

        memset (page+1, 0, page_size - sizeof(*page));
        set->latch->dirty = 1;

        page->min = page_size;
        page->garbage = 0;
        page->act = 0;

        // clean up page first by
        // removing deleted keys

        while( cnt++ < max ) {
                if( cnt == slot )
                        newslot = idx + 2;

                if( cnt < max || frame->lvl )
                  if( slotptr(frame,cnt)->dead )
                        continue;

                // copy the value across

                val = valptr(frame, cnt);
                page->min -= val->len + sizeof(BtVal);
                memcpy ((unsigned char *)page + page->min, val, val->len + sizeof(BtVal));

                // copy the key across

                key = keyptr(frame, cnt);
                page->min -= key->len + sizeof(BtKey);
                memcpy ((unsigned char *)page + page->min, key, key->len + sizeof(BtKey));

                // make a librarian slot

                slotptr(page, ++idx)->off = page->min;
                slotptr(page, idx)->type = Librarian;
                slotptr(page, idx)->dead = 1;

                // set up the slot

                slotptr(page, ++idx)->off = page->min;
                slotptr(page, idx)->type = slotptr(frame, cnt)->type;

                if( !(slotptr(page, idx)->dead = slotptr(frame, cnt)->dead) )
                        page->act++;
        }

        page->cnt = idx;
        free (frame);

        //      see if page has enough space now, or does it need splitting?

        if( page->min >= (idx+2) * sizeof(BtSlot) + sizeof(*page) + keylen + sizeof(BtKey) + vallen + sizeof(BtVal) )
                return newslot;

        return 0;
}

// split the root and raise the height of the btree

BTERR bt_splitroot(BtMgr *mgr, BtPageSet *root, BtLatchSet *right, ushort thread_no)
{  
unsigned char leftkey[BT_keyarray];
uint nxt = mgr->page_size;
unsigned char value[BtId];
BtPageSet left[1];
uid left_page_no;
BtPage frame;
BtKey *ptr;
BtVal *val;

        frame = malloc (mgr->page_size);
        memcpy (frame, root->page, mgr->page_size);

        //      save left page fence key for new root

        ptr = keyptr(root->page, root->page->cnt);
        memcpy (leftkey, ptr, ptr->len + sizeof(BtKey));

        //  Obtain an empty page to use, and copy the current
        //  root contents into it, e.g. lower keys

        if( bt_newpage(mgr, left, frame, thread_no) )
                return mgr->err_thread = thread_no, mgr->err;

        left_page_no = left->latch->page_no;
        bt_unpinlatch (left->latch, thread_no, __LINE__);
        free (frame);

        // preserve the page info at the bottom
        // of higher keys and set rest to zero

        memset(root->page+1, 0, mgr->page_size - sizeof(*root->page));

        // insert stopper key at top of newroot page
        // and increase the root height

        nxt -= BtId + sizeof(BtVal);
        bt_putid (value, right->page_no);
        val = (BtVal *)((unsigned char *)root->page + nxt);
        memcpy (val->value, value, BtId);
        val->len = BtId;

        nxt -= 2 + sizeof(BtKey);
        slotptr(root->page, 2)->off = nxt;
        ptr = (BtKey *)((unsigned char *)root->page + nxt);
        ptr->len = 2;
        ptr->key[0] = 0xff;
        ptr->key[1] = 0xff;

        // insert lower keys page fence key on newroot page as first key

        nxt -= BtId + sizeof(BtVal);
        bt_putid (value, left_page_no);
        val = (BtVal *)((unsigned char *)root->page + nxt);
        memcpy (val->value, value, BtId);
        val->len = BtId;

        ptr = (BtKey *)leftkey;
        nxt -= ptr->len + sizeof(BtKey);
        slotptr(root->page, 1)->off = nxt;
        memcpy ((unsigned char *)root->page + nxt, leftkey, ptr->len + sizeof(BtKey));
        
        bt_putid(root->page->right, 0);
        root->page->min = nxt;          // reset lowest used offset and key count
        root->page->cnt = 2;
        root->page->act = 2;
        root->page->lvl++;

        mgr->pagezero->alloc->lvl = root->page->lvl;

        // release and unpin root pages

        bt_unlockpage(BtLockWrite, root->latch, thread_no, __LINE__);
        bt_unpinlatch (root->latch, thread_no, __LINE__);

        bt_unpinlatch (right, thread_no, __LINE__);
        return 0;
}

//  split already locked full node
//      leave it locked.
//      return pool entry for new right
//      page, pinned & unlocked

uint bt_splitpage (BtMgr *mgr, BtPageSet *set, ushort thread_no, uint linkleft)
{
uint page_size = mgr->page_size;
BtPageSet right[1], temp[1];
uint cnt = 0, idx = 0, max;
uint lvl = set->page->lvl;
BtKey *key, *ptr;
BtVal *val, *src;
BtPage frame;
uid right2;
uint entry;
uint prev;

        if( !set->page->lvl )
                page_size <<= mgr->leaf_xtra;

        //  split higher half of keys to frame

        frame = malloc (page_size);
        memset (frame, 0, page_size);
        frame->min = page_size;
        max = set->page->cnt;
        cnt = max / 2;
        idx = 0;

        while( cnt++ < max ) {
                if( cnt < max || set->page->lvl )
                  if( slotptr(set->page, cnt)->dead )
                        continue;

                src = valptr(set->page, cnt);
                frame->min -= src->len + sizeof(BtVal);
                memcpy ((unsigned char *)frame + frame->min, src, src->len + sizeof(BtVal));

                key = keyptr(set->page, cnt);
                frame->min -= key->len + sizeof(BtKey);
                ptr = (BtKey*)((unsigned char *)frame + frame->min);
                memcpy (ptr, key, key->len + sizeof(BtKey));

                //      add librarian slot

                slotptr(frame, ++idx)->off = frame->min;
                slotptr(frame, idx)->type = Librarian;
                slotptr(frame, idx)->dead = 1;

                //  add actual slot

                slotptr(frame, ++idx)->off = frame->min;
                slotptr(frame, idx)->type = slotptr(set->page, cnt)->type;

                if( !(slotptr(frame, idx)->dead = slotptr(set->page, cnt)->dead) )
                        frame->act++;
        }

        frame->cnt = idx;
        frame->lvl = lvl;

        // link right node

        if( set->latch->page_no > ROOT_page ) {
                right2 = bt_getid (set->page->right);
                bt_putid (frame->right, right2);
                bt_putid (frame->left, set->latch->page_no);
        }

        //      get new free page and write higher keys to it.

        if( bt_newpage(mgr, right, frame, thread_no) )
                return 0;

        //      link far right's left pointer to new page

        if( linkleft && set->latch->page_no > ROOT_page )
         if( right2 ) {
          if( temp->latch = lvl ? bt_pinlatch (mgr, right2, thread_no) : bt_pinleaf (mgr, right2, thread_no) )
                temp->page = bt_mappage (mgr, temp->latch);
          else
                return 0;

      bt_lockpage(BtLockLink, temp->latch, thread_no, __LINE__);
          bt_putid (temp->page->left, right->latch->page_no);
          bt_unlockpage(BtLockLink, temp->latch, thread_no, __LINE__);
          bt_unpinlatch (temp->latch, thread_no, __LINE__);
         } else {       // page is rightmost
          bt_mutexlock (mgr->lock);
          bt_putid (mgr->pagezero->alloc->left, right->latch->page_no);
          bt_releasemutex(mgr->lock);
         }

        // process lower keys

        memcpy (frame, set->page, page_size);
        memset (set->page+1, 0, page_size - sizeof(*set->page));
        set->latch->dirty = 1;

        set->page->min = page_size;
        set->page->garbage = 0;
        set->page->act = 0;
        max /= 2;
        cnt = 0;
        idx = 0;

        //  assemble page of smaller keys

        while( cnt++ < max ) {
                if( slotptr(frame, cnt)->dead )
                        continue;
                val = valptr(frame, cnt);
                set->page->min -= val->len + sizeof(BtVal);
                memcpy ((unsigned char *)set->page + set->page->min, val, val->len + sizeof(BtVal));

                key = keyptr(frame, cnt);
                set->page->min -= key->len + sizeof(BtKey);
                memcpy ((unsigned char *)set->page + set->page->min, key, key->len + sizeof(BtKey));

                //      add librarian slot

                slotptr(set->page, ++idx)->off = set->page->min;
                slotptr(set->page, idx)->type = Librarian;
                slotptr(set->page, idx)->dead = 1;

                //      add actual slot

                slotptr(set->page, ++idx)->off = set->page->min;
                slotptr(set->page, idx)->type = slotptr(frame, cnt)->type;
                set->page->act++;
        }

        bt_putid(set->page->right, right->latch->page_no);
        set->page->cnt = idx;
        free(frame);

        entry = right->latch - (set->page->lvl ? mgr->latchsets : mgr->leafsets);
        return entry;
}

//      fix keys for newly split page
//      call with both pages pinned & locked
//  return unlocked and unpinned

BTERR bt_splitkeys (BtMgr *mgr, BtPageSet *set, BtLatchSet *right, ushort thread_no)
{
unsigned char leftkey[BT_keyarray], rightkey[BT_keyarray];
unsigned char value[BtId];
uint lvl = set->page->lvl;
BtPage page;
BtKey *ptr;

        // if current page is the root page, split it

        if( set->latch->page_no == ROOT_page )
                return bt_splitroot (mgr, set, right, thread_no);

        ptr = keyptr(set->page, set->page->cnt);
        memcpy (leftkey, ptr, ptr->len + sizeof(BtKey));

        page = bt_mappage (mgr, right);

        ptr = keyptr(page, page->cnt);
        memcpy (rightkey, ptr, ptr->len + sizeof(BtKey));

        // insert new fences in their parent pages

        bt_lockpage (BtLockParent, right, thread_no, __LINE__);

        bt_lockpage (BtLockParent, set->latch, thread_no, __LINE__);
        bt_unlockpage (BtLockWrite, set->latch, thread_no, __LINE__);

        // insert new fence for reformulated left block of smaller keys

        bt_putid (value, set->latch->page_no);
        ptr = (BtKey *)leftkey;

        if( bt_insertkey (mgr, ptr->key, ptr->len, lvl+1, value, BtId, Unique, thread_no) )
                return mgr->err_thread = thread_no, mgr->err;

        // switch fence for right block of larger keys to new right page

        bt_putid (value, right->page_no);
        ptr = (BtKey *)rightkey;

        if( bt_insertkey (mgr, ptr->key, ptr->len, lvl+1, value, BtId, Unique, thread_no) )
                return mgr->err_thread = thread_no, mgr->err;

        bt_unlockpage (BtLockParent, set->latch, thread_no, __LINE__);
        bt_unpinlatch (set->latch, thread_no, __LINE__);

        bt_unlockpage (BtLockParent, right, thread_no, __LINE__);
        bt_unpinlatch (right, thread_no, __LINE__);
        return 0;
}

//      install new key and value onto page
//      page must already be checked for
//      adequate space

BTERR bt_insertslot (BtMgr *mgr, BtPageSet *set, uint slot, unsigned char *key,uint keylen, unsigned char *value, uint vallen, uint type)
{
uint idx, librarian;
BtSlot *node;
BtKey *ptr;
BtVal *val;
int rate;

        //      if previous slot is a librarian slot, use it

        if( slot > 1 )
          if( slotptr(set->page, slot-1)->type == Librarian )
                slot--;

        // copy value onto page

        set->page->min -= vallen + sizeof(BtVal);
        val = (BtVal*)((unsigned char *)set->page + set->page->min);
        memcpy (val->value, value, vallen);
        val->len = vallen;

        // copy key onto page

        set->page->min -= keylen + sizeof(BtKey);
        ptr = (BtKey*)((unsigned char *)set->page + set->page->min);
        memcpy (ptr->key, key, keylen);
        ptr->len = keylen;
        
        //      find first empty slot

        for( idx = slot; idx < set->page->cnt; idx++ )
          if( slotptr(set->page, idx)->dead )
                break;

        // now insert key into array before slot,
        //      adding as many librarian slots as
        //      makes sense.

        if( idx == set->page->cnt ) {
        int avail = 4 * set->page->min / 5 - sizeof(*set->page) - ++set->page->cnt * sizeof(BtSlot);

                librarian = ++idx - slot;
                avail /= sizeof(BtSlot);

                if( avail < 0 )
                        avail = 0;

                if( librarian > avail )
                        librarian = avail;

                if( librarian ) {
                        rate = (idx - slot) / librarian;
                        set->page->cnt += librarian;
                        idx += librarian;
                } else
                        rate = 0;
        } else
                librarian = 0, rate = 0;

        while( idx > slot ) {
                //  transfer slot
                *slotptr(set->page, idx) = *slotptr(set->page, idx-librarian-1);
                idx--;

                //      add librarian slot per rate

                if( librarian )
                 if( (idx - slot + 1)/2 <= librarian * rate ) {
                        node = slotptr(set->page, idx--);
                        node->off = node[1].off;
                        node->type = Librarian;
                        node->dead = 1;
                        librarian--;
                  }
        }

        set->latch->dirty = 1;
        set->page->act++;

        //      fill in new slot

        node = slotptr(set->page, slot);
        node->off = set->page->min;
        node->type = type;
        node->dead = 0;
        return 0;
}

//  Insert new key into the btree at given level.
//      either add a new key or update/add an existing one

BTERR bt_insertkey (BtMgr *mgr, unsigned char *key, uint keylen, uint lvl, void *value, uint vallen, BtSlotType type, ushort thread_no)
{
uint slot, idx, len, entry;
BtPageSet set[1];
BtSlot *node;
BtKey *ptr;
BtVal *val;

  while( 1 ) { // find the page and slot for the current key
        if( slot = bt_loadpage (mgr, set, key, keylen, lvl, BtLockWrite, thread_no) ) {
                node = slotptr(set->page, slot);
                ptr = keyptr(set->page, slot);
        } else {
                if( !mgr->err )
                        mgr->line = __LINE__, mgr->err = BTERR_ovflw;
                return mgr->err_thread = thread_no, mgr->err;
        }

        // if librarian slot == found slot, advance to real slot

        if( node->type == Librarian ) {
//        if( !keycmp (ptr, key, keylen) ) {
                ptr = keyptr(set->page, ++slot);
                node = slotptr(set->page, slot);
          }

        //  if inserting a duplicate key or unique
        //      key that doesn't exist on the page,
        //      check for adequate space on the page
        //      and insert the new key before slot.

        switch( type ) {
        case Unique:
        case Duplicate:
          if( !keycmp (ptr, key, keylen) )
                break;

          if( slot = bt_cleanpage (mgr, set, keylen, slot, vallen) )
            if( bt_insertslot (mgr, set, slot, key, keylen, value, vallen, type) )
                  return mgr->err;
                else
                  goto insxit;

          if( entry = bt_splitpage (mgr, set, thread_no, 1) )
            if( !bt_splitkeys (mgr, set, mgr->latchsets + entry, thread_no) )
                  continue;

          return mgr->err_thread = thread_no, mgr->err;

        case Update:
          if( keycmp (ptr, key, keylen) )
                goto insxit;

          break;
        }

        // if key already exists, update value and return

        val = valptr(set->page, slot);

        if( val->len >= vallen ) {
          if( node->dead )
                set->page->act++;
          node->type = type;
          node->dead = 0;

          set->page->garbage += val->len - vallen;
          set->latch->dirty = 1;
          val->len = vallen;
          memcpy (val->value, value, vallen);
          goto insxit;
        }

        //  new update value doesn't fit in existing value area
        //      make sure page has room

        if( !node->dead )
          set->page->garbage += val->len + ptr->len + sizeof(BtKey) + sizeof(BtVal);
        else
          set->page->act++;

        node->type = type;
        node->dead = 0;

        if( slot = bt_cleanpage (mgr, set, keylen, slot, vallen) )
          break;

        if( entry = bt_splitpage (mgr, set, thread_no, 1) )
          if( !bt_splitkeys (mgr, set, mgr->latchsets + entry, thread_no) )
                continue;

        return mgr->err_thread = thread_no, mgr->err;
  }

  //  copy key and value onto page and update slot

  set->page->min -= vallen + sizeof(BtVal);
  val = (BtVal*)((unsigned char *)set->page + set->page->min);
  memcpy (val->value, value, vallen);
  val->len = vallen;

  set->latch->dirty = 1;
  set->page->min -= keylen + sizeof(BtKey);
  ptr = (BtKey*)((unsigned char *)set->page + set->page->min);
  memcpy (ptr->key, key, keylen);
  ptr->len = keylen;
        
  slotptr(set->page,slot)->off = set->page->min;

insxit:
  bt_unlockpage(BtLockWrite, set->latch, thread_no, __LINE__);
  bt_unpinlatch (set->latch, thread_no, __LINE__);
  return 0;
}

//      determine actual page where key is located
//  return slot number

uint bt_atomicpage (BtMgr *mgr, BtPage source, AtomicTxn *locks, uint src, BtPageSet *set)
{
BtKey *key = keyptr(source,src), *ptr;
unsigned char fence[BT_keyarray];
uint entry, slot;

        if( locks[src].reuse )
          entry = locks[src-1].entry;
        else
          entry = locks[src].entry;

        //      find where our key is located 
        //      on current page or pages split on
        //      same page txn operations.

        do {
                set->latch = mgr->leafsets + entry;
                set->page = bt_mappage (mgr, set->latch);

                if( slot = bt_findslot(set->page, key->key, key->len) ) {
                  if( slotptr(set->page, slot)->type == Librarian )
                        slot++;
                  if( locks[src].reuse )
                        locks[src].entry = entry;
                  return slot;
                }
        } while( entry = set->latch->split );

        ptr = keyptr (set->page, set->page->cnt);
        memcpy (fence, ptr, ptr->len + 1);

        mgr->line = __LINE__, mgr->err = BTERR_atomic;
        return 0;
}

BTERR bt_atomicinsert (BtMgr *mgr, BtPage source, AtomicTxn *locks, uint src, ushort thread_no, logseqno lsn)
{
BtKey *key = keyptr(source, src);
BtVal *val = valptr(source, src);
BtLatchSet *latch;
BtPageSet set[1];
uint entry, slot;

  while( slot = bt_atomicpage (mgr, source, locks, src, set) ) {
        if( slot = bt_cleanpage(mgr, set, key->len, slot, val->len) ) {
          if( bt_insertslot (mgr, set, slot, key->key, key->len, val->value, val->len, slotptr(source,src)->type) )
                return mgr->err_thread = thread_no, mgr->err;

          set->page->lsn = lsn;
          return 0;
        }

        //  split page

        if( entry = bt_splitpage (mgr, set, thread_no, 0) )
          latch = mgr->leafsets + entry;
        else
          return mgr->err_thread = thread_no, mgr->err;

        //      splice right page into split chain
        //      and WriteLock it

        bt_lockpage(BtLockWrite, latch, thread_no, __LINE__);
        latch->split = set->latch->split;
        set->latch->split = entry;
  }

  return mgr->line = __LINE__, mgr->err_thread = thread_no, mgr->err = BTERR_atomic;
}

//      perform delete from smaller btree
//  insert a delete slot if not found there

BTERR bt_atomicdelete (BtMgr *mgr, BtPage source, AtomicTxn *locks, uint src, ushort thread_no, logseqno lsn)
{
BtKey *key = keyptr(source, src);
BtPageSet set[1];
uint idx, slot;
BtSlot *node;
BtKey *ptr;
BtVal *val;

        if( slot = bt_atomicpage (mgr, source, locks, src, set) ) {
          node = slotptr(set->page, slot);
          ptr = keyptr(set->page, slot);
          val = valptr(set->page, slot);
        } else
          return mgr->line = __LINE__, mgr->err_thread = thread_no, mgr->err = BTERR_struct;

        //  if slot is not found, insert a delete slot

        if( keycmp (ptr, key->key, key->len) )
          if( bt_insertslot (mgr, set, slot, key->key, key->len, NULL, 0, Delete) )
                return mgr->err;

        //      if node is already dead,
        //      ignore the request.

        if( node->dead )
                return 0;

        set->page->garbage += ptr->len + val->len + sizeof(BtKey) + sizeof(BtVal);
        set->latch->dirty = 1;
        set->page->lsn = lsn;
        set->page->act--;

        node->dead = 0;
        __sync_fetch_and_add(&mgr->found, 1);
        return 0;
}

//      release master's splits from right to left

void bt_atomicrelease (BtMgr *mgr, uint entry, ushort thread_no)
{
BtLatchSet *latch = mgr->leafsets + entry;

        if( latch->split )
                bt_atomicrelease (mgr, latch->split, thread_no);

        latch->split = 0;
        bt_unlockpage(BtLockWrite, latch, thread_no, __LINE__);
        bt_unpinlatch(latch, thread_no, __LINE__);
}

int qsortcmp (BtSlot *slot1, BtSlot *slot2, BtPage page)
{
BtKey *key1 = (BtKey *)((char *)page + slot1->off);
BtKey *key2 = (BtKey *)((char *)page + slot2->off);

        return keycmp (key1, key2->key, key2->len);
}
//      atomic modification of a batch of keys.

BTERR bt_atomictxn (BtDb *bt, BtPage source)
{
uint src, idx, slot, samepage, entry, que = 0;
BtKey *key, *ptr, *key2;
int result = 0;
BtSlot temp[1];
logseqno lsn;
int type;

  // stable sort the list of keys into order to
  //    prevent deadlocks between threads.
/*
  for( src = 1; src++ < source->cnt; ) {
        *temp = *slotptr(source,src);
        key = keyptr (source,src);

        for( idx = src; --idx; ) {
          key2 = keyptr (source,idx);
          if( keycmp (key, key2->key, key2->len) < 0 ) {
                *slotptr(source,idx+1) = *slotptr(source,idx);
                *slotptr(source,idx) = *temp;
          } else
                break;
        }
  }
*/
        qsort_r (slotptr(source,1), source->cnt, sizeof(BtSlot), (__compar_d_fn_t)qsortcmp, source);
  //  add entries to redo log

  if( bt->mgr->pagezero->redopages )
        lsn = bt_txnredo (bt->mgr, source, bt->thread_no);
  else
        lsn = 0;

  //  perform the individual actions in the transaction

  if( bt_atomicexec (bt->mgr, source, lsn, 0, bt->thread_no) )
        return bt->mgr->err;

  // if number of active pages
  // is greater than the buffer pool
  // promote page into larger btree

  if( bt->main )
   while( bt->mgr->pagezero->leafpages > bt->mgr->leaftotal - *bt->mgr->thread_no )
        if( bt_txnpromote (bt) )
          return bt->mgr->err;

  // return success

  return 0;
}

//      execute the source list of inserts/deletes

BTERR bt_atomicexec(BtMgr *mgr, BtPage source, logseqno lsn, int lsm, ushort thread_no)
{
uint src, idx, samepage, entry;
BtPageSet set[1], prev[1];
unsigned char value[BtId];
BtLatchSet *latch;
uid right_page_no;
AtomicTxn *locks;
BtKey *key, *ptr;
BtPage page;
BtVal *val;

  locks = calloc (source->cnt + 1, sizeof(AtomicTxn));

  // Load the leaf page for each key
  // group same page references with reuse bit

  for( src = 0; src++ < source->cnt; ) {
        key = keyptr(source, src);

        // first determine if this modification falls
        // on the same page as the previous modification
        //      note that the far right leaf page is a special case

        if( samepage = src > 1 )
          samepage = !bt_getid(set->page->right) || keycmp (ptr, key->key, key->len) >= 0;

        if( !samepage )
          if( bt_loadpage(mgr, set, key->key, key->len, 0, BtLockWrite, thread_no) )
                ptr = keyptr(set->page, set->page->cnt), set->latch->split = 0;
          else
                return mgr->err;

        entry = set->latch - mgr->leafsets;
        locks[src].reuse = samepage;
        locks[src].entry = entry;

        //      capture current lsn for master page

        locks[src].reqlsn = set->page->lsn;
  }

  // insert or delete each key
  // process any splits or merges
  // run through txn list backwards

  samepage = source->cnt + 1;

  for( src = source->cnt; src; src-- ) {
        if( locks[src].reuse )
          continue;

        //  perform the txn operations
        //      from smaller to larger on
        //  the same page

        for( idx = src; idx < samepage; idx++ )
         switch( slotptr(source,idx)->type ) {
         case Delete:
          if( bt_atomicdelete (mgr, source, locks, idx, thread_no, lsn) )
                return mgr->err;
          break;

        case Duplicate:
        case Unique:
          if( bt_atomicinsert (mgr, source, locks, idx, thread_no, lsn) )
                return mgr->err;
          break;

        default:
          bt_atomicpage (mgr, source, locks, idx, set);
          break;
        }

        //      after the same page operations have finished,
        //  process master page for splits or deletion.

        latch = prev->latch = mgr->leafsets + locks[src].entry;
        prev->page = bt_mappage (mgr, prev->latch);
        samepage = src;

        //  pick-up all splits from master page
        //      each one is already pinned & WriteLocked.

        while( entry = prev->latch->split ) {
          set->latch = mgr->leafsets + entry;
          set->page = bt_mappage (mgr, set->latch);

          // delete empty master page by undoing its split
          //  (this is potentially another empty page)
          //  note that there are no pointers to it yet

          if( !prev->page->act ) {
                memcpy (set->page->left, prev->page->left, BtId);
                memcpy (prev->page, set->page, mgr->page_size << mgr->leaf_xtra);
                bt_lockpage (BtLockDelete, set->latch, thread_no, __LINE__);
                prev->latch->split = set->latch->split;
                prev->latch->dirty = 1;
                bt_freepage (mgr, set, thread_no);
                continue;
          }

          // remove empty split page from the split chain
          // and return it to the free list. No other
          // thread has its page number yet.

          if( !set->page->act ) {
                memcpy (prev->page->right, set->page->right, BtId);
                prev->latch->split = set->latch->split;

                bt_lockpage (BtLockDelete, set->latch, thread_no, __LINE__);
                bt_freepage (mgr, set, thread_no);
                continue;
          }

          //  update prev's fence key

          ptr = keyptr(prev->page,prev->page->cnt);
          bt_putid (value, prev->latch->page_no);

          if( bt_insertkey (mgr, ptr->key, ptr->len, 1, value, BtId, Unique, thread_no) )
                return mgr->err;

          // splice in the left link into the split page

          bt_putid (set->page->left, prev->latch->page_no);

          if( lsm )
                bt_syncpage (mgr, prev->page, prev->latch);

          *prev = *set;
        }

        //  update left pointer in next right page from last split page
        //      (if all splits were reversed or none occurred, latch->split == 0)

        if( latch->split ) {
          //  fix left pointer in master's original (now split)
          //  far right sibling or set rightmost page in page zero

          if( right_page_no = bt_getid (prev->page->right) ) {
                if( set->latch = bt_pinleaf (mgr, right_page_no, thread_no) )
                  set->page = bt_mappage (mgr, set->latch);
                else
                  return mgr->err;

            bt_lockpage (BtLockLink, set->latch, thread_no, __LINE__);
            bt_putid (set->page->left, prev->latch->page_no);
                set->latch->dirty = 1;

            bt_unlockpage (BtLockLink, set->latch, thread_no, __LINE__);
                bt_unpinlatch (set->latch, thread_no, __LINE__);
          } else {      // prev is rightmost page
            bt_mutexlock (mgr->lock);
                bt_putid (mgr->pagezero->alloc->left, prev->latch->page_no);
            bt_releasemutex(mgr->lock);
          }

          //  switch the original fence key from the
          //  master page to the last split page.

          ptr = keyptr(prev->page,prev->page->cnt);
          bt_putid (value, prev->latch->page_no);

          if( bt_insertkey (mgr, ptr->key, ptr->len, 1, value, BtId, Update, thread_no) )
                return mgr->err;

          if( lsm )
                bt_syncpage (mgr, prev->page, prev->latch);

          //  unlock and unpin the split pages

          bt_atomicrelease (mgr, latch->split, thread_no);

          //  unlock and unpin the master page

          latch->split = 0;
          bt_unlockpage(BtLockWrite, latch, thread_no, __LINE__);
          bt_unpinlatch(latch, thread_no, __LINE__);
          continue;
        }

        //      since there are no splits remaining, we're
        //  finished if master page occupied

        if( prev->page->act ) {
          bt_unlockpage(BtLockWrite, prev->latch, thread_no, __LINE__);

          if( lsm )
                bt_syncpage (mgr, prev->page, prev->latch);

          bt_unpinlatch(prev->latch, thread_no, __LINE__);
          continue;
        }

        // any and all splits were reversed, and the
        // master page located in prev is empty, delete it

        if( bt_deletepage (mgr, prev, thread_no, 0) )
                return mgr->err;
  }

  free (locks);
  return 0;
}

//  pick & promote a page into the larger btree

BTERR bt_txnpromote (BtDb *bt)
{
uint entry, slot, idx;
BtPageSet set[1];
BtSlot *node;
BtKey *ptr;
BtVal *val;

  while( 1 ) {
#ifdef unix
        entry = __sync_fetch_and_add(&bt->mgr->leafpromote, 1);
#else
        entry = _InterlockedIncrement (&bt->mgr->leafpromote) - 1;
#endif
        entry %= bt->mgr->leaftotal;

        if( !entry )
                continue;

        set->latch = bt->mgr->leafsets + entry;

        //  skip this entry if it has never been used

        if( !set->latch->page_no )
          continue;

        if( !bt_mutextry(set->latch->modify) )
                continue;

        //  skip this entry if it is pinned

        if( set->latch->pin & ~CLOCK_bit ) {
          bt_releasemutex(set->latch->modify);
          continue;
        }

        set->page = bt_mappage (bt->mgr, set->latch);

        // entry has no right sibling

        if( !bt_getid (set->page->right) ) {
          bt_releasemutex(set->latch->modify);
          continue;
        }

        // entry interiour node or being killed or constructed

        if( set->page->lvl || set->page->nopromote || set->page->kill ) {
          bt_releasemutex(set->latch->modify);
          continue;
        }

        //  pin the page for our access
        //      and leave it locked for the
        //      duration of the promotion.

        set->latch->pin++;
        bt_lockpage (BtLockWrite, set->latch, bt->thread_no, __LINE__);
        bt_releasemutex(set->latch->modify);

        // transfer slots in our selected page to the main btree
if( !(entry % 100) )
fprintf(stderr, "Promote entry %d page %d, %d keys\n", entry, set->latch->page_no, set->page->act);

        if( bt_atomicexec (bt->main, set->page, 0, bt->mgr->pagezero->redopages ? 1 : 0, bt->thread_no) ) {
                fprintf (stderr, "Promote error = %d line = %d\n", bt->main->err, bt->main->line);
                return bt->main->err;
        }

        //  now delete the page

        if( bt_deletepage (bt->mgr, set, bt->thread_no, 0) )
                fprintf (stderr, "Promote: delete page err = %d, threadno = %d\n", bt->mgr->err, bt->mgr->err_thread);

        return bt->mgr->err;
  }
}

//      set cursor to highest slot on highest page

uint bt_lastkey (BtDb *bt)
{
uid page_no = bt_getid (bt->mgr->pagezero->alloc->left);
BtPageSet set[1];

        if( set->latch = bt_pinleaf (bt->mgr, page_no, bt->thread_no) )
                set->page = bt_mappage (bt->mgr, set->latch);
        else
                return 0;

    bt_lockpage(BtLockRead, set->latch, bt->thread_no, __LINE__);
        memcpy (bt->cursor, set->page, bt->mgr->page_size << bt->mgr->leaf_xtra);
    bt_unlockpage(BtLockRead, set->latch, bt->thread_no, __LINE__);
        bt_unpinlatch (set->latch, bt->thread_no, __LINE__);
        return bt->cursor->cnt;
}

//      return previous slot on cursor page

uint bt_prevkey (BtDb *bt, uint slot)
{
uid cursor_page = bt->cursor->page_no;
uid ourright, next, us = cursor_page;
BtPageSet set[1];

        if( --slot )
                return slot;

        ourright = bt_getid(bt->cursor->right);

goleft:
        if( !(next = bt_getid(bt->cursor->left)) )
                return 0;

findourself:
        cursor_page = next;

        if( set->latch = bt_pinleaf (bt->mgr, next, bt->thread_no) )
                set->page = bt_mappage (bt->mgr, set->latch);
        else
                return 0;

    bt_lockpage(BtLockRead, set->latch, bt->thread_no, __LINE__);
        memcpy (bt->cursor, set->page, bt->mgr->page_size << bt->mgr->leaf_xtra);
        bt_unlockpage(BtLockRead, set->latch, bt->thread_no, __LINE__);
        bt_unpinlatch (set->latch, bt->thread_no, __LINE__);
        
        next = bt_getid (bt->cursor->right);

        if( bt->cursor->kill )
                goto findourself;

        if( next != us )
          if( next == ourright )
                goto goleft;
          else
                goto findourself;

        return bt->cursor->cnt;
}

//  return next slot on cursor page
//  or slide cursor right into next page

uint bt_nextkey (BtDb *bt, uint slot)
{
BtPageSet set[1];
uid right;

  do {
        right = bt_getid(bt->cursor->right);

        while( slot++ < bt->cursor->cnt )
          if( slotptr(bt->cursor,slot)->dead )
                continue;
          else if( right || (slot < bt->cursor->cnt) ) // skip infinite stopper
                return slot;
          else
                break;

        if( !right )
                break;

        if( set->latch = bt_pinleaf (bt->mgr, right, bt->thread_no) )
                set->page = bt_mappage (bt->mgr, set->latch);
        else
                return 0;

    bt_lockpage(BtLockRead, set->latch, bt->thread_no, __LINE__);
        memcpy (bt->cursor, set->page, bt->mgr->page_size << bt->mgr->leaf_xtra);
        bt_unlockpage(BtLockRead, set->latch, bt->thread_no, __LINE__);
        bt_unpinlatch (set->latch, bt->thread_no, __LINE__);
        slot = 0;

  } while( 1 );

  return bt->mgr->err = 0;
}

//  cache page of keys into cursor and return starting slot for given key

uint bt_startkey (BtDb *bt, unsigned char *key, uint len)
{
BtPageSet set[1];
uint slot;

        // cache page for retrieval

        if( slot = bt_loadpage (bt->mgr, set, key, len, 0, BtLockRead, bt->thread_no) )
          memcpy (bt->cursor, set->page, bt->mgr->page_size << bt->mgr->leaf_xtra);
        else
          return 0;

        bt_unlockpage(BtLockRead, set->latch, bt->thread_no, __LINE__);
        bt_unpinlatch (set->latch, bt->thread_no, __LINE__);
        return slot;
}

#ifdef STANDALONE

#ifndef unix
double getCpuTime(int type)
{
FILETIME crtime[1];
FILETIME xittime[1];
FILETIME systime[1];
FILETIME usrtime[1];
SYSTEMTIME timeconv[1];
double ans = 0;

        memset (timeconv, 0, sizeof(SYSTEMTIME));

        switch( type ) {
        case 0:
                GetSystemTimeAsFileTime (xittime);
                FileTimeToSystemTime (xittime, timeconv);
                ans = (double)timeconv->wDayOfWeek * 3600 * 24;
                break;
        case 1:
                GetProcessTimes (GetCurrentProcess(), crtime, xittime, systime, usrtime);
                FileTimeToSystemTime (usrtime, timeconv);
                break;
        case 2:
                GetProcessTimes (GetCurrentProcess(), crtime, xittime, systime, usrtime);
                FileTimeToSystemTime (systime, timeconv);
                break;
        }

        ans += (double)timeconv->wHour * 3600;
        ans += (double)timeconv->wMinute * 60;
        ans += (double)timeconv->wSecond;
        ans += (double)timeconv->wMilliseconds / 1000;
        return ans;
}
#else
#include <time.h>
#include <sys/resource.h>

double getCpuTime(int type)
{
struct rusage used[1];
struct timeval tv[1];

        switch( type ) {
        case 0:
                gettimeofday(tv, NULL);
                return (double)tv->tv_sec + (double)tv->tv_usec / 1000000;

        case 1:
                getrusage(RUSAGE_SELF, used);
                return (double)used->ru_utime.tv_sec + (double)used->ru_utime.tv_usec / 1000000;

        case 2:
                getrusage(RUSAGE_SELF, used);
                return (double)used->ru_stime.tv_sec + (double)used->ru_stime.tv_usec / 1000000;
        }

        return 0;
}
#endif

void bt_poolaudit (BtMgr *mgr, char *type)
{
BtLatchSet *latch, test[1];
uint entry;

        memset (test, 0, sizeof(test));

        if( memcmp (test, mgr->latchsets, sizeof(test)) )
                fprintf(stderr, "%s latchset zero overwritten\n", type);

        if( memcmp (test, mgr->leafsets, sizeof(test)) )
                fprintf(stderr, "%s leafset zero overwritten\n", type);

        for( entry = 0; ++entry < mgr->latchtotal; ) {
                latch = mgr->latchsets + entry;

                if( *latch->modify->value )
                        fprintf(stderr, "%s latchset %d modifylocked for page %d\n", type, entry, latch->page_no);

                if( latch->pin & ~CLOCK_bit )
                        fprintf(stderr, "%s latchset %d pinned %d times for page %d\n", type, entry, latch->pin & ~CLOCK_bit, latch->page_no);
        }

        for( entry = 0; ++entry < mgr->leaftotal; ) {
                latch = mgr->leafsets + entry;

                if( *latch->modify->value )
                        fprintf(stderr, "%s leafset %d modifylocked for page %d\n", type, entry, latch->page_no);

                if( latch->pin & ~CLOCK_bit )
                        fprintf(stderr, "%s leafset %d pinned %d times for page %d\n", type, entry, latch->pin & ~CLOCK_bit, latch->page_no);
        }
}

typedef struct {
        char idx;
        char *type;
        char *infile;
        BtMgr *main;
        BtMgr *mgr;
        int num;
} ThreadArg;

//  standalone program to index file of keys
//  then list them onto std-out

#ifdef unix
void *index_file (void *arg)
#else
uint __stdcall index_file (void *arg)
#endif
{
int line = 0, found = 0, cnt = 0, idx;
uid next, page_no = LEAF_page;  // start on first page of leaves
int ch, len = 0, slot, type = 0;
unsigned char key[BT_maxkey];
unsigned char txn[65536];
ThreadArg *args = arg;
BtPage page, frame;
BtPageSet set[1];
uint nxt = 65536;
BtKey *ptr;
BtVal *val;
BtDb *bt;
FILE *in;

        bt = bt_open (args->mgr, args->main);
        page = (BtPage)txn;

        if( args->idx < strlen (args->type) )
                ch = args->type[args->idx];
        else
                ch = args->type[strlen(args->type) - 1];

        switch(ch | 0x20)
        {
        case 'd':
                type = Delete;

        case 'p':
                if( !type )
                        type = Unique;

                if( args->num )
                 if( type == Delete )
                  fprintf(stderr, "started TXN pennysort delete for %s\n", args->infile);
                 else
                  fprintf(stderr, "started TXN pennysort insert for %s\n", args->infile);
                else
                 if( type == Delete )
                  fprintf(stderr, "started pennysort delete for %s\n", args->infile);
                 else
                  fprintf(stderr, "started pennysort insert for %s\n", args->infile);

                if( in = fopen (args->infile, "rb") )
                  while( ch = getc(in), ch != EOF )
                        if( ch == '\n' )
                        {
                          line++;

                          if( !args->num ) {
                            if( bt_insertkey (bt->mgr, key, 10, 0, key + 10, len - 10, Unique, bt->thread_no) )
                                  fprintf(stderr, "Error %d Line: %d source: %d\n", bt->mgr->err, bt->mgr->line, line), exit(0);
                            len = 0;
                                continue;
                          }

                          nxt -= len - 10;
                          memcpy (txn + nxt, key + 10, len - 10);
                          nxt -= 1;
                          txn[nxt] = len - 10;
                          nxt -= 10;
                          memcpy (txn + nxt, key, 10);
                          nxt -= 1;
                          txn[nxt] = 10;
                          slotptr(page,++cnt)->off  = nxt;
                          slotptr(page,cnt)->type = type;
                          len = 0;

                          if( cnt < args->num )
                                continue;

                          page->cnt = cnt;
                          page->act = cnt;
                          page->min = nxt;

                          if( bt_atomictxn (bt, page) )
                                fprintf(stderr, "Error %d Line: %d by %d source: %d\n", bt->mgr->err, bt->mgr->line, bt->mgr->err_thread, line), exit(0);
                          nxt = sizeof(txn);
                          cnt = 0;

                        }
                        else if( len < BT_maxkey )
                                key[len++] = ch;
                fprintf(stderr, "finished %s for %d keys\n", args->infile, line);
                break;

        case 'w':
                fprintf(stderr, "started indexing for %s\n", args->infile);
                if( in = fopen (args->infile, "r") )
                  while( ch = getc(in), ch != EOF )
                        if( ch == '\n' )
                        {
                          line++;

                          if( bt_insertkey (bt->mgr, key, len, 0, NULL, 0, Unique, bt->thread_no) )
                                fprintf(stderr, "Error %d Line: %d source: %d\n", bt->mgr->err, bt->mgr->line, line), exit(0);
                          len = 0;
                        }
                        else if( len < BT_maxkey )
                                key[len++] = ch;
                fprintf(stderr, "finished %s for %d keys\n", args->infile, line);
                break;

        case 'f':
                fprintf(stderr, "started finding keys for %s\n", args->infile);
                if( in = fopen (args->infile, "rb") )
                  while( ch = getc(in), ch != EOF )
                        if( ch == '\n' )
                        {
                          line++;
                          if( bt_findkey (bt, key, len, NULL, 0) == 0 )
                                found++;
                          else if( bt->mgr->err )
                                fprintf(stderr, "Error %d Syserr %d Line: %d source: %d\n", bt->mgr->err, errno, bt->mgr->line, line), exit(0);
                          len = 0;
                        }
                        else if( len < BT_maxkey )
                                key[len++] = ch;
                fprintf(stderr, "finished %s for %d keys, found %d\n", args->infile, line, found);
                break;

        case 's':
                fprintf(stderr, "started scanning\n");
                
                do {
                        if( set->latch = bt_pinleaf (bt->mgr, page_no, bt->thread_no) )
                                set->page = bt_mappage (bt->mgr, set->latch);
                        else
                                fprintf(stderr, "unable to obtain latch"), exit(1);

                        bt_lockpage (BtLockRead, set->latch, bt->thread_no, __LINE__);
                        next = bt_getid (set->page->right);

                        for( slot = 0; slot++ < set->page->cnt; )
                         if( next || slot < set->page->cnt )
                          if( !slotptr(set->page, slot)->dead ) {
                                ptr = keyptr(set->page, slot);
                                len = ptr->len;

                            if( slotptr(set->page, slot)->type == Duplicate )
                                        len -= BtId;

                                fwrite (ptr->key, len, 1, stdout);
                                val = valptr(set->page, slot);
                                fwrite (val->value, val->len, 1, stdout);
                                fputc ('\n', stdout);
                                cnt++;
                           }

                        bt_unlockpage (BtLockRead, set->latch, bt->thread_no, __LINE__);
                        bt_unpinlatch (set->latch, bt->thread_no, __LINE__);
                } while( page_no = next );

                fprintf(stderr, " Total keys read %d\n", cnt);
                break;

        case 'r':
                fprintf(stderr, "started reverse scan\n");
                if( slot = bt_lastkey (bt) )
                   while( slot = bt_prevkey (bt, slot) ) {
                        if( slotptr(bt->cursor, slot)->dead )
                          continue;

                        ptr = keyptr(bt->cursor, slot);
                        len = ptr->len;

                        if( slotptr(bt->cursor, slot)->type == Duplicate )
                                len -= BtId;

                        fwrite (ptr->key, len, 1, stdout);
                        val = valptr(bt->cursor, slot);
                        fwrite (val->value, val->len, 1, stdout);
                        fputc ('\n', stdout);
                        cnt++;
                  }

                fprintf(stderr, " Total keys read %d\n", cnt);
                break;

        case 'c':
#ifdef unix
                posix_fadvise( bt->mgr->idx, 0, 0, POSIX_FADV_SEQUENTIAL);
#endif
                fprintf(stderr, "started counting\n");
                next = bt->mgr->redopage + bt->mgr->pagezero->redopages;

                while( page_no < bt_getid(bt->mgr->pagezero->alloc->right) ) {
                        if( bt_readpage (bt->mgr, bt->cursor, page_no, 1) )
                                break;

                        if( !bt->cursor->free && !bt->cursor->lvl )
                                cnt += bt->cursor->act;

                        bt->mgr->reads++;
                        page_no = next++;
                }
                
                cnt--;  // remove stopper key
                fprintf(stderr, " Total keys counted %d\n", cnt);
                break;
        }

        bt_close (bt);
#ifdef unix
        return NULL;
#else
        return 0;
#endif
}

typedef struct timeval timer;

int main (int argc, char **argv)
{
int idx, cnt, len, slot, err;
double start, stop;
#ifdef unix
pthread_t *threads;
#else
HANDLE *threads;
#endif
ThreadArg *args;
uint mainleafpool = 0;
uint mainleafxtra = 0;
uint redopages = 0;
uint poolsize = 0;
uint leafpool = 0;
uint leafxtra = 0;
uint mainpool = 0;
uint mainbits = 0;
int bits = 16;
float elapsed;
int num = 0;
char key[1];
BtMgr *main;
BtMgr *mgr;
BtKey *ptr;

        if( argc < 3 ) {
                fprintf (stderr, "Usage: %s idx_file main_file cmds [pagebits leafbits poolsize leafpool txnsize redopages mainbits mainleafbits mainpool mainleafpool src_file1 src_file2 ... ]\n", argv[0]);
                fprintf (stderr, "  where idx_file is the name of the cache btree file\n");
                fprintf (stderr, "  where main_file is the name of the main btree file\n");
                fprintf (stderr, "  cmds is a string of (c)ount/(r)ev scan/(w)rite/(s)can/(d)elete/(f)ind/(p)ennysort, with one character command for each input src_file. Commands with no input file need a placeholder.\n");
                fprintf (stderr, "  pagebits is the page size in bits for the cache btree\n");
                fprintf (stderr, "  leafbits is the number of xtra bits for a leaf page\n");
                fprintf (stderr, "  poolsize is the number of pages in buffer pool for the cache btree\n");
                fprintf (stderr, "  leafpool is the number of leaf pages in leaf pool for the cache btree\n");
                fprintf (stderr, "  txnsize = n to block transactions into n units, or zero for no transactions\n");
                fprintf (stderr, "  redopages = n to implement recovery buffer with n pages, or zero for no recovery buffer\n");
                fprintf (stderr, "  mainbits is the page size of the main btree in bits\n");
                fprintf (stderr, "  mainpool is the number of main pages in the main buffer pool\n");
                fprintf (stderr, "  mainleaf is the number of main pages in the main leaf pool\n");
                fprintf (stderr, "  src_file1 thru src_filen are files of keys separated by newline\n");
                exit(0);
        }

        start = getCpuTime(0);

        if( argc > 4 )
                bits = atoi(argv[4]);

        if( argc > 5 )
                leafxtra = atoi(argv[5]);

        if( argc > 6 )
                poolsize = atoi(argv[6]);

        if( !poolsize )
                fprintf (stderr, "no page pool\n"), exit(1);

        if( argc > 7 )
                leafpool = atoi(argv[7]);

        if( argc > 8 )
                num = atoi(argv[8]);

        if( argc > 9 )
                redopages = atoi(argv[9]);

        if( redopages > 65535 )
                fprintf (stderr, "Recovery buffer too large\n"), exit(1);

        if( argc > 10 )
                mainbits = atoi(argv[10]);

        if( argc > 11 )
                mainleafxtra = atoi(argv[11]);

        if( argc > 12 )
                mainpool = atoi(argv[12]);

        if( argc > 13 )
                mainleafpool = atoi(argv[13]);

        cnt = argc - 14;
#ifdef unix
        threads = malloc (cnt * sizeof(pthread_t));
#else
        threads = GlobalAlloc (GMEM_FIXED|GMEM_ZEROINIT, cnt * sizeof(HANDLE));
#endif
        args = malloc ((cnt + 1) * sizeof(ThreadArg));

        mgr = bt_mgr (argv[1], bits, leafxtra, poolsize, leafpool, redopages);

        if( !mgr ) {
                fprintf(stderr, "Index Open Error %s\n", argv[1]);
                exit (1);
        }

        if( mainbits ) {
                main = bt_mgr (argv[2], mainbits, mainleafxtra, mainpool, mainleafpool, 0);

                if( !main ) {
                        fprintf(stderr, "Index Open Error %s\n", argv[2]);
                        exit (1);
                }
        } else
                main = NULL;

        //      fire off threads

        if( cnt )
          for( idx = 0; idx < cnt; idx++ ) {
                args[idx].infile = argv[idx + 14];
                args[idx].type = argv[3];
                args[idx].main = main;
                args[idx].mgr = mgr;
                args[idx].num = num;
                args[idx].idx = idx;
#ifdef unix
                if( err = pthread_create (threads + idx, NULL, index_file, args + idx) )
                        fprintf(stderr, "Error creating thread %d\n", err);
#else
                threads[idx] = (HANDLE)_beginthreadex(NULL, 131072, index_file, args + idx, 0, NULL);
#endif
          }
        else {
                args[0].infile = argv[idx + 10];
                args[0].type = argv[3];
                args[0].main = main;
                args[0].mgr = mgr;
                args[0].num = num;
                args[0].idx = 0;
                index_file (args);
        }

        //      wait for termination

#ifdef unix
        for( idx = 0; idx < cnt; idx++ )
                pthread_join (threads[idx], NULL);
#else
        WaitForMultipleObjects (cnt, threads, TRUE, INFINITE);

        for( idx = 0; idx < cnt; idx++ )
                CloseHandle(threads[idx]);
#endif
        bt_poolaudit(mgr, "cache");

        if( main )
                bt_poolaudit(main, "main");

        fprintf(stderr, "cache %d reads %d writes %d found\n", mgr->reads, mgr->writes, mgr->found);
        if( main )
                fprintf(stderr, "main %d reads %d writes %d found\n", main->reads, main->writes, main->found);

        if( main )
                bt_mgrclose (main);
        bt_mgrclose (mgr);

        elapsed = getCpuTime(0) - start;
        fprintf(stderr, " real %dm%.3fs\n", (int)(elapsed/60), elapsed - (int)(elapsed/60)*60);
        elapsed = getCpuTime(1);
        fprintf(stderr, " user %dm%.3fs\n", (int)(elapsed/60), elapsed - (int)(elapsed/60)*60);
        elapsed = getCpuTime(2);
        fprintf(stderr, " sys  %dm%.3fs\n", (int)(elapsed/60), elapsed - (int)(elapsed/60)*60);
}

#endif  //STANDALONE