SQLite

Check-in [5669ac4a40]
Login

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

Overview
Comment:Avoid reading frames that have already been checkpointed from the wal file.
Downloads: Tarball | ZIP archive
Timelines: family | ancestors | descendants | both | wal-read-change
Files: files | file ages | folders
SHA1: 5669ac4a40429abc3f44540fc9d2f3b79b404bdf
User & Date: dan 2015-08-12 19:42:08.239
Context
2015-08-13
20:23
When searching the wal file for a frame, do not search that part that was already checkpointed when the snapshot being read was at the head of the wal file. (check-in: 90760e7232 user: dan tags: wal-read-change)
2015-08-12
19:42
Avoid reading frames that have already been checkpointed from the wal file. (check-in: 5669ac4a40 user: dan tags: wal-read-change)
15:36
Minor optimization for fts5 API xInst(). (check-in: efb7c9c5d0 user: dan tags: trunk)
Changes
Unified Diff Ignore Whitespace Patch
Changes to src/wal.c.
2368
2369
2370
2371
2372
2373
2374


2375
2376
2377
2378
2379
2380
2381
2382
2383
2384
2385



2386
2387
2388
2389
2390
2391
2392
  Wal *pWal,                      /* WAL handle */
  Pgno pgno,                      /* Database page number to read data for */
  u32 *piRead                     /* OUT: Frame number (or zero) */
){
  u32 iRead = 0;                  /* If !=0, WAL frame to return data from */
  u32 iLast = pWal->hdr.mxFrame;  /* Last page in WAL for this reader */
  int iHash;                      /* Used to loop through N hash tables */



  /* This routine is only be called from within a read transaction. */
  assert( pWal->readLock>=0 || pWal->lockError );

  /* If the "last page" field of the wal-index header snapshot is 0, then
  ** no data will be read from the wal under any circumstances. Return early
  ** in this case as an optimization.  Likewise, if pWal->readLock==0, 
  ** then the WAL is ignored by the reader so return early, as if the 
  ** WAL were empty.
  */
  if( iLast==0 || pWal->readLock==0 ){



    *piRead = 0;
    return SQLITE_OK;
  }

  /* Search the hash table or tables for an entry matching page number
  ** pgno. Each iteration of the following for() loop searches one
  ** hash table (each hash table indexes up to HASHTABLE_NPAGE frames).







>
>










|
>
>
>







2368
2369
2370
2371
2372
2373
2374
2375
2376
2377
2378
2379
2380
2381
2382
2383
2384
2385
2386
2387
2388
2389
2390
2391
2392
2393
2394
2395
2396
2397
  Wal *pWal,                      /* WAL handle */
  Pgno pgno,                      /* Database page number to read data for */
  u32 *piRead                     /* OUT: Frame number (or zero) */
){
  u32 iRead = 0;                  /* If !=0, WAL frame to return data from */
  u32 iLast = pWal->hdr.mxFrame;  /* Last page in WAL for this reader */
  int iHash;                      /* Used to loop through N hash tables */
  u32 iFirst;
  int iMinHash;

  /* This routine is only be called from within a read transaction. */
  assert( pWal->readLock>=0 || pWal->lockError );

  /* If the "last page" field of the wal-index header snapshot is 0, then
  ** no data will be read from the wal under any circumstances. Return early
  ** in this case as an optimization.  Likewise, if pWal->readLock==0, 
  ** then the WAL is ignored by the reader so return early, as if the 
  ** WAL were empty.
  */
  if( iLast==0 
   || pWal->readLock==0 
   || iLast==(iFirst = walCkptInfo(pWal)->nBackfill)
  ){
    *piRead = 0;
    return SQLITE_OK;
  }

  /* Search the hash table or tables for an entry matching page number
  ** pgno. Each iteration of the following for() loop searches one
  ** hash table (each hash table indexes up to HASHTABLE_NPAGE frames).
2408
2409
2410
2411
2412
2413
2414

2415
2416
2417
2418
2419
2420
2421
2422
2423
2424
2425
2426
2427
2428
2429
2430
2431
2432
2433
2434
2435
2436
2437
  **   (aPgno[iFrame]==pgno): 
  **     This condition filters out normal hash-table collisions.
  **
  **   (iFrame<=iLast): 
  **     This condition filters out entries that were added to the hash
  **     table after the current read-transaction had started.
  */

  for(iHash=walFramePage(iLast); iHash>=0 && iRead==0; iHash--){
    volatile ht_slot *aHash;      /* Pointer to hash table */
    volatile u32 *aPgno;          /* Pointer to array of page numbers */
    u32 iZero;                    /* Frame number corresponding to aPgno[0] */
    int iKey;                     /* Hash slot index */
    int nCollide;                 /* Number of hash collisions remaining */
    int rc;                       /* Error code */

    rc = walHashGet(pWal, iHash, &aHash, &aPgno, &iZero);
    if( rc!=SQLITE_OK ){
      return rc;
    }
    nCollide = HASHTABLE_NSLOT;
    for(iKey=walHash(pgno); aHash[iKey]; iKey=walNextHash(iKey)){
      u32 iFrame = aHash[iKey] + iZero;
      if( iFrame<=iLast && aPgno[aHash[iKey]]==pgno ){
        assert( iFrame>iRead || CORRUPT_DB );
        iRead = iFrame;
      }
      if( (nCollide--)==0 ){
        return SQLITE_CORRUPT_BKPT;
      }
    }







>
|














|







2413
2414
2415
2416
2417
2418
2419
2420
2421
2422
2423
2424
2425
2426
2427
2428
2429
2430
2431
2432
2433
2434
2435
2436
2437
2438
2439
2440
2441
2442
2443
  **   (aPgno[iFrame]==pgno): 
  **     This condition filters out normal hash-table collisions.
  **
  **   (iFrame<=iLast): 
  **     This condition filters out entries that were added to the hash
  **     table after the current read-transaction had started.
  */
  iMinHash = walFramePage(iFirst);
  for(iHash=walFramePage(iLast); iHash>=iMinHash && iRead==0; iHash--){
    volatile ht_slot *aHash;      /* Pointer to hash table */
    volatile u32 *aPgno;          /* Pointer to array of page numbers */
    u32 iZero;                    /* Frame number corresponding to aPgno[0] */
    int iKey;                     /* Hash slot index */
    int nCollide;                 /* Number of hash collisions remaining */
    int rc;                       /* Error code */

    rc = walHashGet(pWal, iHash, &aHash, &aPgno, &iZero);
    if( rc!=SQLITE_OK ){
      return rc;
    }
    nCollide = HASHTABLE_NSLOT;
    for(iKey=walHash(pgno); aHash[iKey]; iKey=walNextHash(iKey)){
      u32 iFrame = aHash[iKey] + iZero;
      if( iFrame<=iLast && iFrame>iFirst && aPgno[aHash[iKey]]==pgno ){
        assert( iFrame>iRead || CORRUPT_DB );
        iRead = iFrame;
      }
      if( (nCollide--)==0 ){
        return SQLITE_CORRUPT_BKPT;
      }
    }
Changes to test/wal6.test.
188
189
190
191
192
193
194










































195
196
197
} {}

db eval {SELECT test4('3.3.2')}

do_test 3.x {
  db2 close
} {}











































finish_test








>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>



188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
} {}

db eval {SELECT test4('3.3.2')}

do_test 3.x {
  db2 close
} {}

#-------------------------------------------------------------------------
# Check that if a wal file has been partially checkpointed, no frames are
# read from the checkpointed part.
#
reset_db
do_execsql_test 4.1 {
  PRAGMA page_size = 1024;
  PRAGMA journal_mode = wal;
  CREATE TABLE t1(a, b);
  CREATE TABLE t2(a, b);
  PRAGMA wal_checkpoint = truncate;
} {wal 0 0 0}

do_test 4.2 {
  execsql { INSERT INTO t1 VALUES(1, 2) }
  file size test.db-wal
} [wal_file_size 1 1024]

do_test 4.3 {
  sqlite3 db2 test.db
  execsql { 
    BEGIN;
    INSERT INTO t2 VALUES(3, 4);
  }
  execsql { PRAGMA wal_checkpoint = passive } db2
} {0 1 1}

do_test 4.3 {
  execsql { COMMIT }
  db2 close
  hexio_write test.db-wal 0 [string repeat 00 2000]
  sqlite3 db2 test.db
} {}

do_test 4.4.1 { 
  catchsql { SELECT * FROM t1 } db2 
} {0 {1 2}}
do_test 4.4.2 { 
  catchsql { SELECT * FROM t2 } db2 
} {1 {database disk image is malformed}}


finish_test