aboutsummaryrefslogtreecommitdiff
path: root/src/os_unix.c
Commit message (Expand)AuthorAge
* Change the name ZERO_DAMAGE to the more descriptive POWERSAFE_OVERWRITE.drh2011-12-23
* Merge the latest trunk changes into the statvfs branch.drh2011-12-23
|\
| * Add interfaces sqlite3_uri_boolean() and sqlite3_uri_int64() which aredrh2011-12-23
* | Add the ability to enable or disable SQLITE_IOCAP_ZERO_DAMAGE using a URIdrh2011-12-21
* | Add SQLITE_IOCAP_ZERO_DAMAGE and enable it for both unix and windows. Usedrh2011-12-17
* | Fix a bad #endif with the previous check-in on this branch.drh2011-12-17
* | Add support for statvfs() in os_unix.c, for determining the sector size.drh2011-12-17
|/
* Merge the nx-devkit changes into trunk. This includes the newdrh2011-12-14
|\
| * Add a file-control that will discover the name of the bottom-level VFS anddrh2011-12-14
| * Backport a minimal set of changes needed to get 8+3 filenames and thedrh2011-12-12
* | Fix os_unix.c so that, unless 8.3 filenames are actually in use, journal and ...dan2011-12-12
* | Fix harmless compiler warnings.drh2011-12-11
|/
* Make the unix VFS tolerant of read() calls that return less than thedrh2011-11-07
* Use mkdir() and rmdir() rather than open() and unlink() to create the lockdrh2011-11-04
* Fix an issue with finding the access permissions of journal files whendrh2011-10-20
* Added the tool/warnings-clang.sh script. Changes so that there are nodrh2011-10-15
* Add assert() statements and eliminate needless variable assignments in orderdrh2011-10-14
* Fix a harmless compiler warning introduced into os_unix.c by one of the drh2011-10-13
* The date/time functions return NULL if the xCurrentTime ordrh2011-10-12
* Change the behavior of the readonly_shm=1 query parameter so that it neverdrh2011-10-11
* When finding the appropriate file permissions for journal files withdrh2011-10-05
* Formerly, we enabled fdatasync() on linux only. But now we learn thatdrh2011-08-31
* Always include the unixShm.id field, even when not debugging.drh2011-08-31
* Make sure SQLITE_FCNTL_SIZE_HINT a no-op if the chunk size is not greater tha...mistachkin2011-08-30
* Back out [05c9832e5f6eb] since it was causing a performance regression withdrh2011-08-29
|\
* | Fix several harmless compiler warnings and a documentation bug.drh2011-08-26
* | Fix the SQLITE_DISABLE_DIRSYNC compile time option.dan2011-08-25
* | If the application-defined openDirectory() function returns SQLITE_CANTOPEN,drh2011-08-23
* | Change to test instrumentation of os_unix.c: Regard IO errors that occur with...dan2011-08-23
* | When retrying a write() after an EINTR error on unix, be sure to alsodrh2011-08-19
* | Make the openDirectory routine in os_unix.c overrideable so that it candrh2011-08-10
* | In os_unix.c, do not open the directory containing the database file whendrh2011-08-08
* | Allow the unlink() system call to be overridden in os_unix.c.drh2011-08-08
* | Merge changes for the new sqlite3_file_control() that will cause the -wal and...mistachkin2011-07-28
|\ \
| * | Test cases added. Fix the query mode.drh2011-07-26
| * | Prototype change for a new sqlite3_file_control() that will cause thedrh2011-07-26
* | | Use osStat() instead of stat() consistently in os_unix.cdrh2011-07-28
|/ /
* / Enable the SQLITE_FCNTL_SIZE_HINT on unix even if SQLITE_FCNTL_CHUNK_SIZEdrh2011-07-25
|/
* In os_unix.c, check if the ESTALE macro is defined before using it.dan2011-07-15
* Fix some harmless compiler warnings that were occurring with THREADSAFE=0.drh2011-06-24
* Rework the autoconf script to better deal with utime() and dlopen().drh2011-06-20
* Make use of the sqlite3GetBoolean() interface for more robust processingdrh2011-06-01
* Simplify the wal-readonly branch so that it does not require changes todrh2011-06-01
* Merge the latest trunk changes into the wal-readonly branch.drh2011-05-31
|\
| * Add the ability to limit filenames to 8+3 using thedrh2011-05-17
* | Add experimental support for read-only connections to WAL databases.dan2011-05-10
|/
* Invoke the unix open() system call through a wrapper to avoid problems drh2011-04-25
* Change the prototype for the open() system call to agree with Posix. Thoughdrh2011-04-17
* Fix #ifs involving SQLITE_ENABLE_LOCKING_STYLE so that they check the value ofdrh2011-04-15
* Make sure the unix backend returns SQLITE_FULL if write() fails with andrh2011-04-15