/ Check-in [1d01c02a]
Login

Many hyperlinks are disabled.
Use anonymous login to enable hyperlinks.

Overview
Comment:Eliminate some unnecessary IO during a commit in full auto-vacuum mode. (CVS 5426)
Downloads: Tarball | ZIP archive | SQL archive
Timelines: family | ancestors | descendants | both | trunk
Files: files | file ages | folders
SHA1:1d01c02ab04f9cc0eb6a2a46a24d3ea63a617d5b
User & Date: danielk1977 2008-07-16 18:17:56
Context
2008-07-16
18:20
More improvements to the 64-bit integer conversion testing in func.test. (CVS 5427) check-in: 632bb3c7 user: drh tags: trunk
18:17
Eliminate some unnecessary IO during a commit in full auto-vacuum mode. (CVS 5426) check-in: 1d01c02a user: danielk1977 tags: trunk
18:04
Fix a test in func.test so that it works on machines that lack long double. (CVS 5425) check-in: 8686959d user: drh tags: trunk
Changes
Hide Diffs Unified Diffs Ignore Whitespace Patch

Changes to src/btree.c.

5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
....
2108
2109
2110
2111
2112
2113
2114
2115

2116
2117
2118
2119
2120
2121
2122
....
2123
2124
2125
2126
2127
2128
2129
2130
2131
2132
2133
2134
2135
2136
2137
....
2273
2274
2275
2276
2277
2278
2279
2280
2281
2282
2283
2284
2285
2286
2287
....
6026
6027
6028
6029
6030
6031
6032
6033
6034
6035
6036
6037
6038
6039
6040
....
6234
6235
6236
6237
6238
6239
6240
6241
6242
6243
6244
6245
6246
6247
6248
** a legal notice, here is a blessing:
**
**    May you do good and not evil.
**    May you find forgiveness for yourself and forgive others.
**    May you share freely, never taking more than you give.
**
*************************************************************************
** $Id: btree.c,v 1.482 2008/07/12 14:52:20 drh Exp $
**
** This file implements a external (disk-based) database using BTrees.
** See the header comment on "btreeInt.h" for additional information.
** Including a description of file format and an overview of operation.
*/
#include "btreeInt.h"

................................................................................
** database. The pDbPage reference remains valid.
*/
static int relocatePage(
  BtShared *pBt,           /* Btree */
  MemPage *pDbPage,        /* Open page to move */
  u8 eType,                /* Pointer map 'type' entry for pDbPage */
  Pgno iPtrPage,           /* Pointer map 'page-no' entry for pDbPage */
  Pgno iFreePage           /* The location to move pDbPage to */

){
  MemPage *pPtrPage;   /* The page that contains a pointer to pDbPage */
  Pgno iDbPage = pDbPage->pgno;
  Pager *pPager = pBt->pPager;
  int rc;

  assert( eType==PTRMAP_OVERFLOW2 || eType==PTRMAP_OVERFLOW1 || 
................................................................................
      eType==PTRMAP_BTREE || eType==PTRMAP_ROOTPAGE );
  assert( sqlite3_mutex_held(pBt->mutex) );
  assert( pDbPage->pBt==pBt );

  /* Move page iDbPage from its current location to page number iFreePage */
  TRACE(("AUTOVACUUM: Moving %d to free page %d (ptr page %d type %d)\n", 
      iDbPage, iFreePage, iPtrPage, eType));
  rc = sqlite3PagerMovepage(pPager, pDbPage->pDbPage, iFreePage);
  if( rc!=SQLITE_OK ){
    return rc;
  }
  pDbPage->pgno = iFreePage;

  /* If pDbPage was a btree-page, then it may have child pages and/or cells
  ** that point to overflow pages. The pointer map entries for all these
................................................................................
        }
        releasePage(pFreePg);
      }while( nFin!=0 && iFreePg>nFin );
      assert( iFreePg<iLastPg );
      
      rc = sqlite3PagerWrite(pLastPg->pDbPage);
      if( rc==SQLITE_OK ){
        rc = relocatePage(pBt, pLastPg, eType, iPtrPage, iFreePg);
      }
      releasePage(pLastPg);
      if( rc!=SQLITE_OK ){
        return rc;
      }
    }
  }
................................................................................
      assert( eType!=PTRMAP_ROOTPAGE );
      assert( eType!=PTRMAP_FREEPAGE );
      rc = sqlite3PagerWrite(pRoot->pDbPage);
      if( rc!=SQLITE_OK ){
        releasePage(pRoot);
        return rc;
      }
      rc = relocatePage(pBt, pRoot, eType, iPtrPage, pgnoMove);
      releasePage(pRoot);

      /* Obtain the page at pgnoRoot */
      if( rc!=SQLITE_OK ){
        return rc;
      }
      rc = sqlite3BtreeGetPage(pBt, pgnoRoot, &pRoot, 0);
................................................................................
        */
        MemPage *pMove;
        releasePage(pPage);
        rc = sqlite3BtreeGetPage(pBt, maxRootPgno, &pMove, 0);
        if( rc!=SQLITE_OK ){
          return rc;
        }
        rc = relocatePage(pBt, pMove, PTRMAP_ROOTPAGE, 0, iTable);
        releasePage(pMove);
        if( rc!=SQLITE_OK ){
          return rc;
        }
        rc = sqlite3BtreeGetPage(pBt, maxRootPgno, &pMove, 0);
        if( rc!=SQLITE_OK ){
          return rc;







|







 







|
>







 







|







 







|







 







|







 







|







5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
....
2108
2109
2110
2111
2112
2113
2114
2115
2116
2117
2118
2119
2120
2121
2122
2123
....
2124
2125
2126
2127
2128
2129
2130
2131
2132
2133
2134
2135
2136
2137
2138
....
2274
2275
2276
2277
2278
2279
2280
2281
2282
2283
2284
2285
2286
2287
2288
....
6027
6028
6029
6030
6031
6032
6033
6034
6035
6036
6037
6038
6039
6040
6041
....
6235
6236
6237
6238
6239
6240
6241
6242
6243
6244
6245
6246
6247
6248
6249
** a legal notice, here is a blessing:
**
**    May you do good and not evil.
**    May you find forgiveness for yourself and forgive others.
**    May you share freely, never taking more than you give.
**
*************************************************************************
** $Id: btree.c,v 1.483 2008/07/16 18:17:56 danielk1977 Exp $
**
** This file implements a external (disk-based) database using BTrees.
** See the header comment on "btreeInt.h" for additional information.
** Including a description of file format and an overview of operation.
*/
#include "btreeInt.h"

................................................................................
** database. The pDbPage reference remains valid.
*/
static int relocatePage(
  BtShared *pBt,           /* Btree */
  MemPage *pDbPage,        /* Open page to move */
  u8 eType,                /* Pointer map 'type' entry for pDbPage */
  Pgno iPtrPage,           /* Pointer map 'page-no' entry for pDbPage */
  Pgno iFreePage,          /* The location to move pDbPage to */
  int isCommit
){
  MemPage *pPtrPage;   /* The page that contains a pointer to pDbPage */
  Pgno iDbPage = pDbPage->pgno;
  Pager *pPager = pBt->pPager;
  int rc;

  assert( eType==PTRMAP_OVERFLOW2 || eType==PTRMAP_OVERFLOW1 || 
................................................................................
      eType==PTRMAP_BTREE || eType==PTRMAP_ROOTPAGE );
  assert( sqlite3_mutex_held(pBt->mutex) );
  assert( pDbPage->pBt==pBt );

  /* Move page iDbPage from its current location to page number iFreePage */
  TRACE(("AUTOVACUUM: Moving %d to free page %d (ptr page %d type %d)\n", 
      iDbPage, iFreePage, iPtrPage, eType));
  rc = sqlite3PagerMovepage(pPager, pDbPage->pDbPage, iFreePage, isCommit);
  if( rc!=SQLITE_OK ){
    return rc;
  }
  pDbPage->pgno = iFreePage;

  /* If pDbPage was a btree-page, then it may have child pages and/or cells
  ** that point to overflow pages. The pointer map entries for all these
................................................................................
        }
        releasePage(pFreePg);
      }while( nFin!=0 && iFreePg>nFin );
      assert( iFreePg<iLastPg );
      
      rc = sqlite3PagerWrite(pLastPg->pDbPage);
      if( rc==SQLITE_OK ){
        rc = relocatePage(pBt, pLastPg, eType, iPtrPage, iFreePg, nFin!=0);
      }
      releasePage(pLastPg);
      if( rc!=SQLITE_OK ){
        return rc;
      }
    }
  }
................................................................................
      assert( eType!=PTRMAP_ROOTPAGE );
      assert( eType!=PTRMAP_FREEPAGE );
      rc = sqlite3PagerWrite(pRoot->pDbPage);
      if( rc!=SQLITE_OK ){
        releasePage(pRoot);
        return rc;
      }
      rc = relocatePage(pBt, pRoot, eType, iPtrPage, pgnoMove, 0);
      releasePage(pRoot);

      /* Obtain the page at pgnoRoot */
      if( rc!=SQLITE_OK ){
        return rc;
      }
      rc = sqlite3BtreeGetPage(pBt, pgnoRoot, &pRoot, 0);
................................................................................
        */
        MemPage *pMove;
        releasePage(pPage);
        rc = sqlite3BtreeGetPage(pBt, maxRootPgno, &pMove, 0);
        if( rc!=SQLITE_OK ){
          return rc;
        }
        rc = relocatePage(pBt, pMove, PTRMAP_ROOTPAGE, 0, iTable, 0);
        releasePage(pMove);
        if( rc!=SQLITE_OK ){
          return rc;
        }
        rc = sqlite3BtreeGetPage(pBt, maxRootPgno, &pMove, 0);
        if( rc!=SQLITE_OK ){
          return rc;

Changes to src/pager.c.

14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
....
5180
5181
5182
5183
5184
5185
5186





5187
5188
5189
5190
5191
5192
5193
5194
5195
5196
5197
5198
5199
5200








5201
5202
5203
5204
5205
5206
5207
5208
....
5241
5242
5243
5244
5245
5246
5247
5248
5249

5250
5251
5252
5253
5254
5255
5256
** The pager is used to access a database disk file.  It implements
** atomic commit and rollback through the use of a journal file that
** is separate from the database file.  The pager also implements file
** locking to prevent two processes from writing the same database
** file simultaneously, or one process from reading the database while
** another is writing.
**
** @(#) $Id: pager.c,v 1.465 2008/07/11 03:34:10 drh Exp $
*/
#ifndef SQLITE_OMIT_DISKIO
#include "sqliteInt.h"
#include <assert.h>
#include <string.h>

/*
................................................................................
** meta-data associated with pPg (i.e. data stored in the nExtra bytes
** allocated along with the page) is the responsibility of the caller.
**
** A transaction must be active when this routine is called. It used to be
** required that a statement transaction was not active, but this restriction
** has been removed (CREATE INDEX needs to move a page when a statement
** transaction is active).





*/
int sqlite3PagerMovepage(Pager *pPager, DbPage *pPg, Pgno pgno){
  PgHdr *pPgOld;  /* The page being overwritten. */
  int h;
  Pgno needSyncPgno = 0;

  pagerEnter(pPager);
  assert( pPg->nRef>0 );

  PAGERTRACE5("MOVE %d page %d (needSync=%d) moves to %d\n", 
      PAGERID(pPager), pPg->pgno, pPg->needSync, pgno);
  IOTRACE(("MOVE %p %d %d\n", pPager, pPg->pgno, pgno))

  pager_get_content(pPg);








  if( pPg->needSync ){
    needSyncPgno = pPg->pgno;
    assert( pPg->inJournal || (int)pgno>pPager->origDbSize );
    assert( pPg->dirty );
    assert( pPager->needSync );
  }

  /* Unlink pPg from its hash-chain */
................................................................................
  pPager->dirtyCache = 1;
  pPager->dbModified = 1;

  if( needSyncPgno ){
    /* If needSyncPgno is non-zero, then the journal file needs to be 
    ** sync()ed before any data is written to database file page needSyncPgno.
    ** Currently, no such page exists in the page-cache and the 
    ** Pager.pInJournal bit has been set. This needs to be remedied by loading
    ** the page into the pager-cache and setting the PgHdr.needSync flag.

    **
    ** If the attempt to load the page into the page-cache fails, (due
    ** to a malloc() or IO failure), clear the bit in the pInJournal[]
    ** array. Otherwise, if the page is loaded and written again in
    ** this transaction, it may be written to the database file before
    ** it is synced into the journal file. This way, it may end up in
    ** the journal file twice, but that is not a problem.







|







 







>
>
>
>
>

|












>
>
>
>
>
>
>
>
|







 







|
|
>







14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
....
5180
5181
5182
5183
5184
5185
5186
5187
5188
5189
5190
5191
5192
5193
5194
5195
5196
5197
5198
5199
5200
5201
5202
5203
5204
5205
5206
5207
5208
5209
5210
5211
5212
5213
5214
5215
5216
5217
5218
5219
5220
5221
....
5254
5255
5256
5257
5258
5259
5260
5261
5262
5263
5264
5265
5266
5267
5268
5269
5270
** The pager is used to access a database disk file.  It implements
** atomic commit and rollback through the use of a journal file that
** is separate from the database file.  The pager also implements file
** locking to prevent two processes from writing the same database
** file simultaneously, or one process from reading the database while
** another is writing.
**
** @(#) $Id: pager.c,v 1.466 2008/07/16 18:17:56 danielk1977 Exp $
*/
#ifndef SQLITE_OMIT_DISKIO
#include "sqliteInt.h"
#include <assert.h>
#include <string.h>

/*
................................................................................
** meta-data associated with pPg (i.e. data stored in the nExtra bytes
** allocated along with the page) is the responsibility of the caller.
**
** A transaction must be active when this routine is called. It used to be
** required that a statement transaction was not active, but this restriction
** has been removed (CREATE INDEX needs to move a page when a statement
** transaction is active).
**
** If the fourth argument, isCommit, is non-zero, then this page is being
** moved as part of a database reorganization just before the transaction 
** is being committed. In this case, it is guaranteed that the database page 
** pPg refers to will not be written to again within this transaction.
*/
int sqlite3PagerMovepage(Pager *pPager, DbPage *pPg, Pgno pgno, int isCommit){
  PgHdr *pPgOld;  /* The page being overwritten. */
  int h;
  Pgno needSyncPgno = 0;

  pagerEnter(pPager);
  assert( pPg->nRef>0 );

  PAGERTRACE5("MOVE %d page %d (needSync=%d) moves to %d\n", 
      PAGERID(pPager), pPg->pgno, pPg->needSync, pgno);
  IOTRACE(("MOVE %p %d %d\n", pPager, pPg->pgno, pgno))

  pager_get_content(pPg);

  /* If the journal needs to be sync()ed before page pPg->pgno can
  ** be written to, store pPg->pgno in local variable needSyncPgno.
  **
  ** If the isCommit flag is set, there is no need to remember that
  ** the journal needs to be sync()ed before database page pPg->pgno 
  ** can be written to. The caller has already promised not to write to it.
  */
  if( pPg->needSync && !isCommit ){
    needSyncPgno = pPg->pgno;
    assert( pPg->inJournal || (int)pgno>pPager->origDbSize );
    assert( pPg->dirty );
    assert( pPager->needSync );
  }

  /* Unlink pPg from its hash-chain */
................................................................................
  pPager->dirtyCache = 1;
  pPager->dbModified = 1;

  if( needSyncPgno ){
    /* If needSyncPgno is non-zero, then the journal file needs to be 
    ** sync()ed before any data is written to database file page needSyncPgno.
    ** Currently, no such page exists in the page-cache and the 
    ** "is journaled" bitvec flag has been set. This needs to be remedied by
    ** loading the page into the pager-cache and setting the PgHdr.needSync 
    ** flag.
    **
    ** If the attempt to load the page into the page-cache fails, (due
    ** to a malloc() or IO failure), clear the bit in the pInJournal[]
    ** array. Otherwise, if the page is loaded and written again in
    ** this transaction, it may be written to the database file before
    ** it is synced into the journal file. This way, it may end up in
    ** the journal file twice, but that is not a problem.

Changes to src/pager.h.

9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
...
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
**    May you share freely, never taking more than you give.
**
*************************************************************************
** This header file defines the interface that the sqlite page cache
** subsystem.  The page cache subsystem reads and writes a file a page
** at a time and provides a journal for rollback.
**
** @(#) $Id: pager.h,v 1.76 2008/06/07 08:58:22 danielk1977 Exp $
*/

#ifndef _PAGER_H_
#define _PAGER_H_

/*
** If defined as non-zero, auto-vacuum is enabled by default. Otherwise
................................................................................
void sqlite3PagerSetSafetyLevel(Pager*,int,int);
const char *sqlite3PagerFilename(Pager*);
const sqlite3_vfs *sqlite3PagerVfs(Pager*);
sqlite3_file *sqlite3PagerFile(Pager*);
const char *sqlite3PagerDirname(Pager*);
const char *sqlite3PagerJournalname(Pager*);
int sqlite3PagerNosync(Pager*);
int sqlite3PagerMovepage(Pager*,DbPage*,Pgno);
void *sqlite3PagerGetData(DbPage *); 
void *sqlite3PagerGetExtra(DbPage *); 
int sqlite3PagerLockingMode(Pager *, int);
int sqlite3PagerJournalMode(Pager *, int);
i64 sqlite3PagerJournalSizeLimit(Pager *, i64);
void *sqlite3PagerTempSpace(Pager*);
int sqlite3PagerSync(Pager *pPager);







|







 







|







9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
...
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
**    May you share freely, never taking more than you give.
**
*************************************************************************
** This header file defines the interface that the sqlite page cache
** subsystem.  The page cache subsystem reads and writes a file a page
** at a time and provides a journal for rollback.
**
** @(#) $Id: pager.h,v 1.77 2008/07/16 18:17:56 danielk1977 Exp $
*/

#ifndef _PAGER_H_
#define _PAGER_H_

/*
** If defined as non-zero, auto-vacuum is enabled by default. Otherwise
................................................................................
void sqlite3PagerSetSafetyLevel(Pager*,int,int);
const char *sqlite3PagerFilename(Pager*);
const sqlite3_vfs *sqlite3PagerVfs(Pager*);
sqlite3_file *sqlite3PagerFile(Pager*);
const char *sqlite3PagerDirname(Pager*);
const char *sqlite3PagerJournalname(Pager*);
int sqlite3PagerNosync(Pager*);
int sqlite3PagerMovepage(Pager*,DbPage*,Pgno,int);
void *sqlite3PagerGetData(DbPage *); 
void *sqlite3PagerGetExtra(DbPage *); 
int sqlite3PagerLockingMode(Pager *, int);
int sqlite3PagerJournalMode(Pager *, int);
i64 sqlite3PagerJournalSizeLimit(Pager *, i64);
void *sqlite3PagerTempSpace(Pager*);
int sqlite3PagerSync(Pager *pPager);