Many hyperlinks are disabled.
Use anonymous login
to enable hyperlinks.
Overview
| Comment: | Rename the PCache1.mxPinned field to n90pct (since it is 90% of nMax) in order to distinguish it from PGroup.mxPinned. Fix the computation of n90pct so that it does not overflow adversely on excessively large cache sizes. |
|---|---|
| Downloads: | Tarball | ZIP archive | SQL archive |
| Timelines: | family | ancestors | descendants | both | trunk |
| Files: | files | file ages | folders |
| SHA1: |
c85202baac49b22a4be36b9733d44f01 |
| User & Date: | drh 2011-01-26 00:07:03 |
Context
|
2011-01-26
| ||
| 06:13 | Changes to exclusive2.test to make it more deterministic. check-in: 84b0c2bc user: dan tags: trunk | |
| 00:07 | Rename the PCache1.mxPinned field to n90pct (since it is 90% of nMax) in order to distinguish it from PGroup.mxPinned. Fix the computation of n90pct so that it does not overflow adversely on excessively large cache sizes. check-in: c85202ba user: drh tags: trunk | |
|
2011-01-25
| ||
| 18:30 | Fix new compiler warnings in pcache1.c that were introduced by the recent performance enhancement patches of [e14649301138b684]. check-in: c17703ec user: drh tags: trunk | |
Changes
Changes to src/pcache1.c.
70 70 ** The PGroup mutex must be held when accessing nMax. 71 71 */ 72 72 PGroup *pGroup; /* PGroup this cache belongs to */ 73 73 int szPage; /* Size of allocated pages in bytes */ 74 74 int bPurgeable; /* True if cache is purgeable */ 75 75 unsigned int nMin; /* Minimum number of pages reserved */ 76 76 unsigned int nMax; /* Configured "cache_size" value */ 77 - unsigned int mxPinned; /* nMax*9/10 */ 77 + unsigned int n90pct; /* nMax*9/10 */ 78 78 79 79 /* Hash table of all pages. The following variables may only be accessed 80 80 ** when the accessor is holding the PGroup mutex. 81 81 */ 82 82 unsigned int nRecyclable; /* Number of pages in the LRU list */ 83 83 unsigned int nPage; /* Total number of pages in apHash */ 84 84 unsigned int nHash; /* Number of slots in apHash[] */ ................................................................................ 595 595 PCache1 *pCache = (PCache1 *)p; 596 596 if( pCache->bPurgeable ){ 597 597 PGroup *pGroup = pCache->pGroup; 598 598 pcache1EnterMutex(pGroup); 599 599 pGroup->nMaxPage += (nMax - pCache->nMax); 600 600 pGroup->mxPinned = pGroup->nMaxPage + 10 - pGroup->nMinPage; 601 601 pCache->nMax = nMax; 602 - pCache->mxPinned = nMax*9/10; 602 + pCache->n90pct = pCache->nMax*9/10; 603 603 pcache1EnforceMaxPage(pGroup); 604 604 pcache1LeaveMutex(pGroup); 605 605 } 606 606 } 607 607 608 608 /* 609 609 ** Implementation of the sqlite3_pcache.xPagecount method. ................................................................................ 705 705 #ifdef SQLITE_MUTEX_OMIT 706 706 pGroup = pCache->pGroup; 707 707 #endif 708 708 709 709 710 710 /* Step 3: Abort if createFlag is 1 but the cache is nearly full */ 711 711 nPinned = pCache->nPage - pCache->nRecyclable; 712 + assert( nPinned>=0 ); 712 713 assert( pGroup->mxPinned == pGroup->nMaxPage + 10 - pGroup->nMinPage ); 713 - assert( pCache->mxPinned == pCache->nMax*9/10 ); 714 + assert( pCache->n90pct == pCache->nMax*9/10 ); 714 715 if( createFlag==1 && ( 715 716 nPinned>=pGroup->mxPinned 716 - || nPinned>=(int)pCache->mxPinned 717 + || nPinned>=(int)pCache->n90pct 717 718 || pcache1UnderMemoryPressure(pCache) 718 719 )){ 719 720 goto fetch_out; 720 721 } 721 722 722 723 if( pCache->nPage>=pCache->nHash && pcache1ResizeHash(pCache) ){ 723 724 goto fetch_out;