Changes to make WAL more robust against SHM locking failures and OOM errors.

FossilOrigin-Name: ebf4041383c3cdddb5861960359abd209d4a1028

diff --git a/manifest b/manifest
index 7791d33a9605cc90a5db1fb6c0577e59ca2ff7e3..748756246d177cf93a473e96797a9c5d81d7bc7b 100644 (file)
--- a/manifest
+++ b/manifest
@@ -1,8 +1,8 @@
 -----BEGIN PGP SIGNED MESSAGE-----
 Hash: SHA1
 
-C In\sthe\srowhash.test,\smake\ssure\sglobal\svariables\sare\scleared\sprior\sto\suse.
-D 2010-06-02T14:43:30
+C Changes\sto\smake\sWAL\smore\srobust\sagainst\sSHM\slocking\sfailures\sand\sOOM\serrors.
+D 2010-06-02T14:45:51
 F Makefile.arm-wince-mingw32ce-gcc fcd5e9cd67fe88836360bb4f9ef4cb7f8e2fb5a0
 F Makefile.in a5cad1f8f3e021356bfcc6c77dc16f6f1952bbc3
 F Makefile.linux-gcc d53183f4aa6a9192d249731c90dbdffbd2c68654
@@ -155,7 +155,7 @@ F src/os.c 1516984144e26734f97748f891f1a04f9e294c2e
 F src/os.h 6f604986f0ef0ca288c2330b16051ff70b431e8c
 F src/os_common.h a8f95b81eca8a1ab8593d23e94f8a35f35d4078f
 F src/os_os2.c 665876d5eec7585226b0a1cf5e18098de2b2da19
-F src/os_unix.c 31511480da88947d39c9e12560a11d4495e19832
+F src/os_unix.c 7ae9257f65b32f20d777e880700a84e7dac8051e
 F src/os_win.c f815403c51a2adad30244374c801dd7fd2734567
 F src/pager.c acbef227bf158776449907c275c5d9332e4e52f9
 F src/pager.h 76466c3a5af56943537f68b1f16567101a0cd1d0
@@ -171,7 +171,7 @@ F src/resolve.c ac5f1a713cd1ae77f08b83cc69581e11bf5ae6f9
 F src/rowset.c 69afa95a97c524ba6faf3805e717b5b7ae85a697
 F src/select.c c03d8a0565febcde8c6a12c5d77d065fddae889b
 F src/shell.c fd4ccdb37c3b68de0623eb938a649e0990710714
-F src/sqlite.h.in 4f72e39f1edeccaf2a5913483bce2f704492b892
+F src/sqlite.h.in 60f4fcfe18bcb53ed0002989837b68aeeba79a5e
 F src/sqlite3ext.h 69dfb8116af51b84a029cddb3b35062354270c89
 F src/sqliteInt.h 4e26920ede6da1b6f154251fb4f89f65f78979e0
 F src/sqliteLimit.h 196e2f83c3b444c4548fc1874f52f84fdbda40f3
@@ -227,7 +227,7 @@ F src/vdbeblob.c 5327132a42a91e8b7acfb60b9d2c3b1c5c863e0e
 F src/vdbemem.c 2a82f455f6ca6f78b59fb312f96054c04ae0ead1
 F src/vdbetrace.c 864cef96919323482ebd9986f2132435115e9cc2
 F src/vtab.c a0f8a40274e4261696ef57aa806de2776ab72cda
-F src/wal.c 2bd4ef3bfeb2481e9edd3d1858ff8db0bc3a650d
+F src/wal.c e36af9e75768d1dc790392089b10cf272abccefa
 F src/wal.h 1c1c9feb629b7f4afcbe0b47f80f47c5551d3a02
 F src/walker.c 3112bb3afe1d85dc52317cb1d752055e9a781f8f
 F src/where.c 75fee9e255b62f773fcadd1d1f25b6f63ac7a356
@@ -818,14 +818,14 @@ F tool/speedtest2.tcl ee2149167303ba8e95af97873c575c3e0fab58ff
 F tool/speedtest8.c 2902c46588c40b55661e471d7a86e4dd71a18224
 F tool/speedtest8inst1.c 293327bc76823f473684d589a8160bde1f52c14e
 F tool/vdbe-compress.tcl d70ea6d8a19e3571d7ab8c9b75cba86d1173ff0f
-P c24ac3563f46fc6bcf7e45ee171afd3ae7d2faa8
-R 00a157c07ad0e7a71db4a2e73d42f8e4
+P 28efe0a40405567f3eaf334b6603282d5adac792
+R 9d892b2c2c5007a01a863a8338bf83dd
 U drh
-Z e46aac16978246262ba376cf49913464
+Z 8d64b4080a97b8f66f1bf92b76c05ca5
 -----BEGIN PGP SIGNATURE-----
 Version: GnuPG v1.4.6 (GNU/Linux)
 
-iD8DBQFMBm4VoxKgR168RlERAhb3AJ9bqFhxwgZ6DohhkaiSxdI+Dne3iQCfS41x
-jKYaDVEfVJDuwV1vKuW32gg=
-=jOCy
+iD8DBQFMBm6ioxKgR168RlERAqErAJ9ySZWhR4zIr4TMIPtUXND7LUScBACeK1dS
+Xn9N4779qMX9jCkGApiXX1s=
+=nr7w
 -----END PGP SIGNATURE-----

diff --git a/manifest.uuid b/manifest.uuid
index 8b9b8236f9b37ea879fe2bf2e50d8d25cba029e7..2f899a051369e6edb9fff6d7b69baf9e6e19ae77 100644 (file)
--- a/manifest.uuid
+++ b/manifest.uuid
@@ -1 +1 @@
-28efe0a40405567f3eaf334b6603282d5adac792
\ No newline at end of file
+ebf4041383c3cdddb5861960359abd209d4a1028
\ No newline at end of file

diff --git a/src/os_unix.c b/src/os_unix.c
index 9d4a780a616f65781afd1d72d49d9ce7503e6841..95f20bad9fe02dbe08ae82d962b07252eb73e1f1 100644 (file)
--- a/src/os_unix.c
+++ b/src/os_unix.c
@@ -3362,7 +3362,7 @@ static int unixShmOpen(
     rc = SQLITE_OK;
     if( unixShmSystemLock(pShmNode, F_WRLCK, UNIX_SHM_DMS, 1)==SQLITE_OK ){
       if( ftruncate(pShmNode->h, 0) ){
-        rc = SQLITE_IOERR;
+        rc = SQLITE_IOERR_SHMOPEN;
       }
     }
     if( rc==SQLITE_OK ){
@@ -3469,7 +3469,7 @@ static int unixShmSize(
       if( reqSize<=(int)sStat.st_size ) break;
     }else{
       *pNewSize = 0;
-      rc = SQLITE_IOERR;
+      rc = SQLITE_IOERR_SHMSIZE;
       break;
     }
     rc = ftruncate(pShmNode->h, reqSize);

diff --git a/src/sqlite.h.in b/src/sqlite.h.in
index e6a1659b94d4db1a6551c12c696b959991e48a08..952f389544ef3fc7705c00518f1ae7757f39b837 100644 (file)
--- a/src/sqlite.h.in
+++ b/src/sqlite.h.in
@@ -388,7 +388,7 @@ int sqlite3_exec(
 #define SQLITE_NOTFOUND    12   /* NOT USED. Table or record not found */
 #define SQLITE_FULL        13   /* Insertion failed because database is full */
 #define SQLITE_CANTOPEN    14   /* Unable to open the database file */
-#define SQLITE_PROTOCOL    15   /* NOT USED. Database lock protocol error */
+#define SQLITE_PROTOCOL    15   /* Database lock protocol error */
 #define SQLITE_EMPTY       16   /* Database is empty */
 #define SQLITE_SCHEMA      17   /* The database schema changed */
 #define SQLITE_TOOBIG      18   /* String or BLOB exceeds size limit */
@@ -444,6 +444,9 @@ int sqlite3_exec(
 #define SQLITE_IOERR_LOCK              (SQLITE_IOERR | (15<<8))
 #define SQLITE_IOERR_CLOSE             (SQLITE_IOERR | (16<<8))
 #define SQLITE_IOERR_DIR_CLOSE         (SQLITE_IOERR | (17<<8))
+#define SQLITE_IOERR_SHMOPEN           (SQLITE_IOERR | (18<<8))
+#define SQLITE_IOERR_SHMSIZE           (SQLITE_IOERR | (19<<8))
+#define SQLITE_IOERR_SHMLOCK           (SQLITE_IOERR | (20<<8))
 #define SQLITE_LOCKED_SHAREDCACHE      (SQLITE_LOCKED |  (1<<8))
 #define SQLITE_BUSY_RECOVERY           (SQLITE_BUSY   |  (1<<8))
 #define SQLITE_CANTOPEN_NOTEMPDIR      (SQLITE_CANTOPEN | (1<<8))

diff --git a/src/wal.c b/src/wal.c
index dc143b4b17626e5cb35273baf01bae49edc22905..397a592b5d8afe682c6a00febb89b25ca9bbab76 100644 (file)
--- a/src/wal.c
+++ b/src/wal.c
@@ -377,6 +377,9 @@ struct Wal {
   WalIndexHdr hdr;           /* Wal-index header for current transaction */
   char *zWalName;            /* Name of WAL file */
   u32 nCkpt;                 /* Checkpoint sequence counter in the wal-header */
+#ifdef SQLITE_DEBUG
+  u8 lockError;              /* True if a locking error has occurred */
+#endif
 };
 
 /*
@@ -629,6 +632,7 @@ static int walLockShared(Wal *pWal, int lockIdx){
                         SQLITE_SHM_LOCK | SQLITE_SHM_SHARED);
   WALTRACE(("WAL%p: acquire SHARED-%s %s\n", pWal,
             walLockName(lockIdx), rc ? "failed" : "ok"));
+  VVA_ONLY( pWal->lockError = (rc!=SQLITE_OK && rc!=SQLITE_BUSY); )
   return rc;
 }
 static void walUnlockShared(Wal *pWal, int lockIdx){
@@ -644,6 +648,7 @@ static int walLockExclusive(Wal *pWal, int lockIdx, int n){
                         SQLITE_SHM_LOCK | SQLITE_SHM_EXCLUSIVE);
   WALTRACE(("WAL%p: acquire EXCLUSIVE-%s cnt=%d %s\n", pWal,
             walLockName(lockIdx), n, rc ? "failed" : "ok"));
+  VVA_ONLY( pWal->lockError = (rc!=SQLITE_OK && rc!=SQLITE_BUSY); )
   return rc;
 }
 static void walUnlockExclusive(Wal *pWal, int lockIdx, int n){
@@ -1693,7 +1698,7 @@ static int walIndexReadHdr(Wal *pWal, int *pChanged){
 ** so it takes care to hold an exclusive lock on the corresponding
 ** WAL_READ_LOCK() while changing values.
 */
-static int walTryBeginRead(Wal *pWal, int *pChanged, int useWal){
+static int walTryBeginRead(Wal *pWal, int *pChanged, int useWal, int cnt){
   volatile WalIndexHdr *pHdr;     /* Header of the wal-index */
   volatile WalCkptInfo *pInfo;    /* Checkpoint information in wal-index */
   u32 mxReadMark;                 /* Largest aReadMark[] value */
@@ -1703,6 +1708,12 @@ static int walTryBeginRead(Wal *pWal, int *pChanged, int useWal){
 
   assert( pWal->readLock<0 );     /* Not currently locked */
 
+  /* Take steps to avoid spinning forever if there is a protocol error. */
+  if( cnt>5 ){
+    if( cnt>100 ) return SQLITE_PROTOCOL;
+    sqlite3OsSleep(pWal->pVfs, 1);
+  }
+
   if( !useWal ){
     rc = walIndexReadHdr(pWal, pChanged);
     if( rc==SQLITE_BUSY ){
@@ -1820,9 +1831,10 @@ static int walTryBeginRead(Wal *pWal, int *pChanged, int useWal){
 */
 int sqlite3WalBeginReadTransaction(Wal *pWal, int *pChanged){
   int rc;                         /* Return code */
+  int cnt = 0;                    /* Number of TryBeginRead attempts */
 
   do{
-    rc = walTryBeginRead(pWal, pChanged, 0);
+    rc = walTryBeginRead(pWal, pChanged, 0, ++cnt);
   }while( rc==WAL_RETRY );
   walIndexUnmap(pWal);
   return rc;
@@ -1859,8 +1871,8 @@ int sqlite3WalRead(
   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 called from within a read transaction */
-  assert( pWal->readLock>=0 );
+  /* 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
@@ -1982,7 +1994,7 @@ int sqlite3WalRead(
 ** Set *pPgno to the size of the database file (or zero, if unknown).
 */
 void sqlite3WalDbsize(Wal *pWal, Pgno *pPgno){
-  assert( pWal->readLock>=0 );
+  assert( pWal->readLock>=0 || pWal->lockError );
   *pPgno = pWal->hdr.nPage;
 }
 
@@ -2133,6 +2145,8 @@ int sqlite3WalSavepointUndo(Wal *pWal, u32 *aWalData){
 */
 static int walRestartLog(Wal *pWal){
   int rc = SQLITE_OK;
+  int cnt;
+
   if( pWal->readLock==0 
    && SQLITE_OK==(rc = walIndexMap(pWal, walMappingSize(pWal->hdr.mxFrame)))
   ){
@@ -2163,9 +2177,10 @@ static int walRestartLog(Wal *pWal){
     }
     walUnlockShared(pWal, WAL_READ_LOCK(0));
     pWal->readLock = -1;
+    cnt = 0;
     do{
       int notUsed;
-      rc = walTryBeginRead(pWal, &notUsed, 1);
+      rc = walTryBeginRead(pWal, &notUsed, 1, ++cnt);
     }while( rc==WAL_RETRY );
 
     /* Unmap the wal-index before returning. Otherwise the VFS layer may
@@ -2418,22 +2433,25 @@ int sqlite3WalCallback(Wal *pWal){
 */
 int sqlite3WalExclusiveMode(Wal *pWal, int op){
   int rc;
-  assert( pWal->writeLock==0 && pWal->readLock>=0 );
+  assert( pWal->writeLock==0 );
+  /* pWal->readLock is usually set, but might be -1 if there was a prior OOM */
+  assert( pWal->readLock>=0 || pWal->lockError );
   if( op==0 ){
     if( pWal->exclusiveMode ){
       pWal->exclusiveMode = 0;
-      if( walLockShared(pWal, WAL_READ_LOCK(pWal->readLock))!=SQLITE_OK ){
+      if( pWal->readLock>=0 
+       && walLockShared(pWal, WAL_READ_LOCK(pWal->readLock))!=SQLITE_OK
+      ){
         pWal->exclusiveMode = 1;
       }
       rc = pWal->exclusiveMode==0;
     }else{
-      /* No changes.  Either already in locking_mode=NORMAL or else the 
-      ** acquisition of the read-lock failed.  The pager must continue to
-      ** hold the database exclusive lock. */
+      /* Already in locking_mode=NORMAL */
       rc = 0;
     }
   }else if( op>0 ){
     assert( pWal->exclusiveMode==0 );
+    assert( pWal->readLock>=0 );
     walUnlockShared(pWal, WAL_READ_LOCK(pWal->readLock));
     pWal->exclusiveMode = 1;
     rc = 1;