Many hyperlinks are disabled.
Use anonymous login
to enable hyperlinks.
Overview
| Comment: | Check at the write() call to work around the msdos bug in OSX actually succeeds and throw an error if it does not. #ifdef out the work-around for all platforms other than OSX. Ticket #3633. (CVS 6237) |
|---|---|
| Downloads: | Tarball | ZIP archive | SQL archive |
| Timelines: | family | ancestors | descendants | both | trunk |
| Files: | files | file ages | folders |
| SHA1: |
b054b569172c53f4e185e4a64f41d087 |
| User & Date: | drh 2009-02-03 15:27:02 |
Context
|
2009-02-03
| ||
| 15:39 | Remove the text of the sqlite3VdbeMemSanity() routine, which was already commented out. (CVS 6238) check-in: a3c26077 user: drh tags: trunk | |
| 15:27 | Check at the write() call to work around the msdos bug in OSX actually succeeds and throw an error if it does not. #ifdef out the work-around for all platforms other than OSX. Ticket #3633. (CVS 6237) check-in: b054b569 user: drh tags: trunk | |
| 13:51 | Add a special rule to the amalgamation generator to deal with the sqlite3OsDlSym function. Ticket #3631. (CVS 6236) check-in: 876f874c user: drh tags: trunk | |
Changes
Changes to src/os_unix.c.
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
...
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
|
** * Definitions of sqlite3_io_methods objects for all locking ** methods plus "finder" functions for each locking method. ** * sqlite3_vfs method implementations. ** * Locking primitives for the proxy uber-locking-method. (MacOSX only) ** * Definitions of sqlite3_vfs objects for all locking methods ** plus implementations of sqlite3_os_init() and sqlite3_os_end(). ** ** $Id: os_unix.c,v 1.238 2009/01/16 23:47:42 drh Exp $ */ #include "sqliteInt.h" #if SQLITE_OS_UNIX /* This file is used on unix only */ /* ** There are various methods for file locking used for concurrency ** control: ................................................................................ pFile->lastErrno = errno; #ifdef EOVERFLOW if( pFile->lastErrno==EOVERFLOW ) return SQLITE_NOLFS; #endif return SQLITE_IOERR; } /* On OS X on an msdos filesystem, the inode number is reported ** incorrectly for zero-size files. See ticket #3260. To work ** around this problem (we consider it a bug in OS X, not SQLite) ** we always increase the file size to 1 by writing a single byte ** prior to accessing the inode number. The one byte written is ** an ASCII 'S' character which also happens to be the first byte ** in the header of every SQLite database. In this way, if there ** is a race condition such that another thread has already populated ** the first page of the database, no damage is done. */ if( statbuf.st_size==0 ){ write(fd, "S", 1); rc = fstat(fd, &statbuf); if( rc!=0 ){ pFile->lastErrno = errno; return SQLITE_IOERR; } } memset(&lockKey, 0, sizeof(lockKey)); lockKey.fid.dev = statbuf.st_dev; #if OS_VXWORKS lockKey.fid.pId = pFile->pId; #else lockKey.fid.ino = statbuf.st_ino; |
|
>
|
>
>
>
>
|
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
...
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
|
** * Definitions of sqlite3_io_methods objects for all locking ** methods plus "finder" functions for each locking method. ** * sqlite3_vfs method implementations. ** * Locking primitives for the proxy uber-locking-method. (MacOSX only) ** * Definitions of sqlite3_vfs objects for all locking methods ** plus implementations of sqlite3_os_init() and sqlite3_os_end(). ** ** $Id: os_unix.c,v 1.239 2009/02/03 15:27:02 drh Exp $ */ #include "sqliteInt.h" #if SQLITE_OS_UNIX /* This file is used on unix only */ /* ** There are various methods for file locking used for concurrency ** control: ................................................................................ pFile->lastErrno = errno; #ifdef EOVERFLOW if( pFile->lastErrno==EOVERFLOW ) return SQLITE_NOLFS; #endif return SQLITE_IOERR; } #ifdef __APPLE__ /* On OS X on an msdos filesystem, the inode number is reported ** incorrectly for zero-size files. See ticket #3260. To work ** around this problem (we consider it a bug in OS X, not SQLite) ** we always increase the file size to 1 by writing a single byte ** prior to accessing the inode number. The one byte written is ** an ASCII 'S' character which also happens to be the first byte ** in the header of every SQLite database. In this way, if there ** is a race condition such that another thread has already populated ** the first page of the database, no damage is done. */ if( statbuf.st_size==0 ){ rc = write(fd, "S", 1); if( rc!=1 ){ return SQLITE_IOERR; } rc = fstat(fd, &statbuf); if( rc!=0 ){ pFile->lastErrno = errno; return SQLITE_IOERR; } } #endif memset(&lockKey, 0, sizeof(lockKey)); lockKey.fid.dev = statbuf.st_dev; #if OS_VXWORKS lockKey.fid.pId = pFile->pId; #else lockKey.fid.ino = statbuf.st_ino; |