Many hyperlinks are disabled.
Use anonymous login
to enable hyperlinks.
100 ancestors of e2c9f71a451e4404
|
2014-04-23
| ||
| 12:57 | Merge all recent trunk changes into the threads branch. (check-in: e2c9f71a user: drh tags: threads) | |
|
2014-04-21
| ||
| 13:36 | Comment tweaks on the test case for the [b75a9ca6b0] bug fix. (check-in: 65d2544a user: drh tags: trunk) | |
| 13:21 | Avoid discarding an ORDER BY clause in the case where an identical GROUP BY clauses uses an index to group, but not sort, the rows. Fix for [b75a9ca6b0]. (check-in: de9a490f user: dan tags: trunk) | |
|
2014-04-18
| ||
| 22:20 | Clean up the proper-subset cost adjustment logic to make it more compact and easier to read and so that full branch test coverage is more easily obtained. (check-in: 9a5d38c7 user: drh tags: trunk) | |
| 13:57 | Add to speedtest1.c the --threads option for setting the SQLITE_CONFIG_WORKER_THREADS configuration. (check-in: 5fce40c4 user: drh tags: threads) | |
| 13:40 | Fix harmless compiler warnings. (check-in: f8f72ecb user: drh tags: threads) | |
| 12:38 | Merge recent trunk changes into the threads branch. (check-in: 8729aa3e user: drh tags: threads) | |
| 00:49 | Add the SQLITE_RUNTIME_BYTEORDER compile-time option to force SQLite to check the processor byte-order at run-time. Add additional compile-time byte order checks for ARM, PPC, and SPARC. (check-in: 2c536387 user: drh tags: trunk) | |
| 00:06 | Add the SQLITE_TESTCTRL_BYTEORDER test control to interrogate SQLite's notion of the processor byte order and whether the byte order is known at compile-time or determined at run-time. (check-in: 9c696196 user: drh tags: trunk) | |
|
2014-04-17
| ||
| 08:57 | Fix build problems in vdbesort.c. Add further comments and changes to make things easier to understand. (check-in: 12b190db user: dan tags: threads) | |
|
2014-04-16
| ||
| 19:04 | Clarify the purpose of the nField argument passed to sqlite3VdbeSorterInit(). (check-in: c0c8cff1 user: dan tags: threads) | |
| 17:41 | Change the name of vdbeIncrInit2 to vdbePmaReaderIncrInit. Add a header comment to the same function. (check-in: 6622d876 user: dan tags: threads) | |
| 16:43 | Rework the way trees of MergeEngine objects are built in vdbesort.c to make it easier to follow. Fix memory leaks that could follow an OOM or IO error. Add various comments to explain functions in vdbesort.c. (check-in: 69026ec7 user: dan tags: threads) | |
|
2014-04-15
| ||
| 20:52 | Fix some problems to do with OOM conditions in vdbesort.c. Some problems remain. (check-in: 2f94f9ce user: dan tags: threads) | |
| 19:52 | Fix further code and documentation issues in vdbesort.c. (check-in: d03f5b86 user: dan tags: threads) | |
|
2014-04-14
| ||
| 19:48 | Make sure the undocumented sqlite_rename_parent() function does not crash even if it is invoked with NULL arguments. Ticket [264b970c4379fd4b94] (check-in: c6fa8d0d user: drh tags: trunk) | |
| 19:23 | Allow the sorter to begin returning data to the VDBE as soon as it is available, instead of waiting until all keys have been sorted. (check-in: cb0ab20c user: dan tags: threads) | |
| 18:41 | Improve performance in single-threaded mode by having the final merge pass keys directly to the VDBE, instead of going via a final PMA. (Closed-Leaf check-in: 02610cd9 user: dan tags: threads-experimental) | |
| 08:45 | Minor fixes so that builds with SQLITE_MAX_WORKER_THREADS=0 work. (check-in: e400bbbf user: dan tags: threads-experimental) | |
| 07:30 | Improve use of multiple threads in sqlite3VdbeSorterRewind(). (check-in: e1bdc4b8 user: dan tags: threads-experimental) | |
|
2014-04-13
| ||
| 19:28 | Remove the unused second argument from sqlite3ExprCachePop(). Add an ALWAYS() on an always-true conditional in sqlite3VdbeResolveLabel(). (check-in: ab23abf3 user: drh tags: trunk) | |
| 19:15 | Make sure column cache elements are cleared correctly when jumping over code for key generation in a partial index. Fix for ticket [2ea3e9fe6379fc3f6]. (check-in: 3122b836 user: drh tags: trunk) | |
|
2014-04-12
| ||
| 19:34 | Fix many issues with new code. (check-in: 62c406a0 user: dan tags: threads-experimental) | |
|
2014-04-11
| ||
| 19:43 | Avoid having the sorter merge too many PMAs at a time when incrementally merging data following a SorterRewind(). (check-in: 98bf0307 user: dan tags: threads-experimental) | |
|
2014-04-10
| ||
| 02:24 | Modify the OR-clause handling so that it can safely deal with OR-clause with 17 or more terms. Fix for ticket [10fb063b1179be53ea0b53bb]. (check-in: a67b5312 user: drh tags: trunk) | |
|
2014-04-09
| ||
| 20:04 | Experimental multi-threaded sorting changes to allow the sorter to begin returning items to the VDBE before all data is sorted. (check-in: f9d5e09a user: dan tags: threads-experimental) | |
|
2014-04-04
| ||
| 22:44 | Fix harmless compiler warnings. (check-in: e54dded2 user: drh tags: threads) | |
| 21:40 | Fix typo in a Windows threading support routine. (check-in: 5e3dfa27 user: mistachkin tags: threads) | |
| 18:49 | Performance improvements when reading large blobs, especially if SQLITE_DIRECT_OVERFLOW_READ is defined. (check-in: 2312eb6a user: drh tags: trunk) | |
| 18:37 | Merge the latest trunk changes into the threads branch. (check-in: 39ac79cf user: drh tags: threads) | |
| 18:20 | Merge changes to the query planner that strive to ensure that any index usage that is a proper subset of some other index usage always has a slightly higher cost. (check-in: 683dd379 user: drh tags: trunk) | |
| 14:12 | Ensure the "PRAGMA journal_mode=WAL" works coming from any other journal_mode with ATTACH-ed databases. (check-in: e54330b4 user: drh tags: trunk) | |
| 07:52 | Add test file sort3.test, which should have been part of commit [9d3351b8d7]. (check-in: dceed2c8 user: dan tags: threads) | |
| 02:13 | Change vdbeSorterExtendFile() so that it makes a best effort to create the PMA file of the desired size, but does not return an error if unable. (check-in: 217814bc user: drh tags: threads) | |
|
2014-04-03
| ||
| 16:42 | Merge all recent changes from trunk. (check-in: a0910079 user: drh tags: threads) | |
| 16:29 | Use OP_Copy instead of OP_SCopy when moving results out of a subquery, to prevent the subquery results from changing out from under the outer query. Fix for ticket [98825a79ce1456]. Problem introduced by check-in [1e64dd782a126f48d78]. (check-in: d5513dfa user: drh tags: trunk) | |
| 16:25 | Fix an integer overflow problem in the sorter. (check-in: 9d3351b8 user: dan tags: threads) | |
| 16:16 | Use OP_Copy instead of OP_SCopy when moving results out of a subquery, to prevent the subquery results from changing out from under the outer query. Fix for ticket [98825a79ce1456]. Problem introduced by check-in [1e64dd782a126f48d78]. (check-in: ec6a0624 user: drh tags: branch-3.8.4) | |
| 14:31 | Fix a typo in the "Synopsis:" comment for the OP_VFilter opcode. (check-in: 48ecdd4a user: drh tags: trunk) | |
| 14:29 | Fix minor errors causing compilation to fail with SQLITE_MAX_WORKER_THREADS set to a value greater than zero. (check-in: 0561272a user: dan tags: threads) | |
| 02:54 | Refactor local object and method names in vdbesort.c so that their names more closely reflect their actual use. (check-in: d284e30e user: drh tags: threads) | |
| 00:32 | In the command-line shell, run set writable_schema before running the ".clone" command. (check-in: 9d2ae634 user: drh tags: trunk) | |
|
2014-04-02
| ||
| 18:58 | Add a big introductory comment to vdbesort.c explaining its operation at a high level. Also adjust some symbolic names and fix other comment issues in that file. (check-in: eef60f1b user: drh tags: threads) | |
| 15:15 | Fix some problems with OOM handling in vdbesort.c. (check-in: 47e702bd user: dan tags: threads) | |
| 14:38 | Change the name of the SorterThread object to "SortSubtask" to avoid confusion with the SQLiteThread object. (check-in: 4ee2d910 user: drh tags: threads) | |
|
2014-04-01
| ||
| 18:41 | When sorting data for a CREATE INDEX statement in single-threaded mode, assume that keys are delivered to the sorter in primary key order. Also fix various comments that had fallen out of date. (check-in: 821d1ac4 user: dan tags: threads) | |
| 15:38 | Even if compile time option SQLITE_MAX_WORKER_THREADS is set to one or greater, set the default number of worker threads to zero. Distribute data more evenly between threads in sqlite3VdbeSorterWrite() to improve performance when sorting large amounts of data. Add new test file sort2.test. (check-in: 643c86a0 user: dan tags: threads) | |
| 10:19 | Fix a problem with OOM handling in the sorter code. (check-in: 59cd5229 user: dan tags: threads) | |
| 01:24 | Remove an unnecessary conditional, replacing it with an assert(). Improvements to comments. (Closed-Leaf check-in: 834bf1c3 user: drh tags: overflow-pgno-cache) | |
|
2014-03-31
| ||
| 23:57 | Fix a compiler warning when SQLITE_DIRECT_OVERFLOW_READ is defined. Minor performance enhancement and size reduction. (check-in: 96385dc4 user: drh tags: overflow-pgno-cache) | |
| 22:03 | Merge in the latest changes and fixes from trunk. (check-in: f6211540 user: drh tags: overflow-pgno-cache) | |
| 20:05 | Remove an unnecessary conditional. (Closed-Leaf check-in: 7473c4df user: drh tags: query-plan-experiments) | |
| 19:57 | Add the SQLITE_MAX_WORKER_THREADS compile time option. And the SQLITE_CONFIG_WORKER_THREADS sqlite3_config() switch. (check-in: 2774710d user: dan tags: threads) | |
| 19:49 | Also make sure an index that is a proper subset of some other index has a higher cost than that other index. Add test cases. (check-in: b7830d23 user: drh tags: query-plan-experiments) | |
| 18:24 | Make sure that an index that covers a proper superset of the WHERE clause terms of some other index has a lower cost than the other index. (check-in: ea8b0910 user: drh tags: query-plan-experiments) | |
| 13:42 | Avoid a (harmless) buffer overread that is possible on an OOM when MEMSYS5 is engaged. (check-in: b3296267 user: drh tags: trunk) | |
|
2014-03-29
| ||
| 21:16 | Experiments in picking better query plans, especially when the usage of one index is a subset of another. (check-in: 8f869ca7 user: drh tags: query-plan-experiments) | |
| 10:01 | Fix a broken assert() in vdbesort.c. (check-in: 18d1b402 user: dan tags: threads) | |
| 09:34 | Fix a problem in vdbesort.c causing spurious SQLITE_NOMEM errors when using memsys3 or memsys5. (check-in: a683c05f user: dan tags: threads) | |
| 06:27 | Add the optimization to avoid some unnecessary calls to sqlite3VdbeRecordUnpack() added to the trunk by [707ea170b3]. (check-in: fc4d04e6 user: dan tags: threads) | |
|
2014-03-28
| ||
| 19:18 | Merge latest changes from orderby-planning branch. (check-in: 4c7fb542 user: dan tags: threads) | |
| 18:35 | Merge the latest changes from trunk. (check-in: 3047a25f user: drh tags: orderby-planning) | |
| 14:41 | Disable the wal64k.test script for non-unix systems since it depends on unix-only features. (check-in: 27deb6e4 user: drh tags: trunk) | |
| 12:56 | Fix a harmless compiler warning. (check-in: a4e47150 user: drh tags: trunk) | |
| 03:12 | Enhance the sqlite3VdbeRecordCompare() routines so that if they encounter database corruption, they will set the UnpackedRecord.isCorrupt field and return 0. The sqlite3BtreeMovetoUnpacked() routine detects this and returns SQLITE_CORRUPT, causing the corruption to be reported back to the top-level. (check-in: 7fa85eaa user: drh tags: trunk) | |
|
2014-03-27
| ||
| 19:25 | Instead of allocating a single large buffer at the beginning of each sort operation, start with a small buffer and extend it using realloc() as required. (check-in: 81987c8c user: dan tags: orderby-planning) | |
| 18:36 | Minor cleanup of the code in the query planner that computes the costs estimates for the various plans. There are no changes to the costs at this time. But the code is slightly more readable now and that might facilitate future enhancements. (check-in: 9b4d7226 user: drh tags: trunk) | |
| 17:23 | Use xFetch() to access temporary files in vdbesort.c. Use a single large allocation instead of many small allocations when accumulating records in vdbesort.c. This is an interim commit - it allocates a buffer the size of the page-cache every time data is sorted. (check-in: f4ac1bf2 user: dan tags: orderby-planning) | |
| 14:05 | Enhance the logest.c utility with new operators: "dup", "inv", "log", and "nlogn". Provide help on an invalid input. (check-in: b4bd2a06 user: drh tags: trunk) | |
| 00:09 | Merge the latest trunk changes and the fix for the crash on a corrupt database. (check-in: 0b35346c user: drh tags: orderby-planning) | |
|
2014-03-26
| ||
| 19:45 | Merge from trunk the fix for the crash on a corrupt database. (check-in: 8cb2b02b user: drh tags: threads) | |
| 15:14 | Add an extra test case for the potential buffer overread patched by [28ddecff04]. (check-in: f585f5d7 user: dan tags: trunk) | |
| 15:05 | Add a test case to verify that the previous change avoids excess buffer overread in sqlite3VdbeRecordCompare(). (check-in: 2b28e8d5 user: drh tags: trunk) | |
| 14:51 | Earlier detection of corruption in sqlite3VdbeRecordCompare() in order to avoid a potential buffer overread. (check-in: 28ddecff user: drh tags: trunk) | |
| 12:02 | Add an assert() and a comment to clarify the operation of the vdbeRecordCompareInt() routine. (check-in: 851abdb8 user: drh tags: trunk) | |
|
2014-03-25
| ||
| 20:28 | Adjust the repeat counts on several tests in speedtest1 so that the relative time spent on the test is roughly the same regardless of --size. (check-in: 7922809e user: drh tags: trunk) | |
| 18:29 | Merge enhancements and fixes from trunk. (check-in: e005f2d6 user: drh tags: orderby-planning) | |
| 17:45 | Add another performance test case to speedtest1. This case is another ORDER BY test but this time without LIMIT. (check-in: 9ab7ffd5 user: drh tags: trunk) | |
| 17:07 | Fix a problem in the code added by [707ea170b3] causing vdbesort.c to sort unstably. (check-in: d3e640af user: dan tags: orderby-planning) | |
| 15:04 | Remove the sequence values from sorter records used by ORDER BY as well. (check-in: c3ae3697 user: dan tags: orderby-planning) | |
| 14:54 | Add an ORDER BY test case to speedtest1.c (check-in: 58812264 user: drh tags: trunk) | |
| 14:12 | Enable four sorting threads by default in the command-line shell. (check-in: 1cab8357 user: drh tags: threads) | |
| 13:17 | Merge all fixes and enhancements from trunk. (check-in: b415dfb6 user: drh tags: threads) | |
| 11:00 | Detect when a VdbeCursor is still pointing at a valid row but that row has moved, and invalidated the return from prior sqlite3BtreeDataFetch() or sqlite3BtreeKeyFetch() calls. (check-in: e6798871 user: drh tags: trunk) | |
|
2014-03-24
| ||
| 20:19 | Omit the sequence value from sorter records used by GROUP BY queries that cannot use an index. (check-in: 3f90abdd user: dan tags: orderby-planning) | |
| 18:08 | Avoid unnecessary moving of content between registers during an ORDER BY. (check-in: 4f472acc user: drh tags: orderby-planning) | |
| 16:30 | Remove unused variables Parse.nColCache and Parse.iColCache. (check-in: 4d7551ce user: dan tags: trunk) | |
| 15:00 | Fix arithmetic operators so that they do not change the affinity of their input operands. Ticket [a8a0d2996a]. (check-in: 221f8f94 user: drh tags: trunk) | |
| 11:23 | Avoid attempting to mmap memory from an offset that is not a multiple of the system page size on systems with page sizes larger than 32KB. (check-in: db7d62c8 user: dan tags: trunk) | |
| 09:34 | Remove an unnecessary temporary register allocation. (check-in: 5d506743 user: dan tags: orderby-planning) | |
| 02:20 | Remove a pointless OP_Once operation in ORDER BY clauses with LIMIT. (check-in: e6c59d23 user: drh tags: orderby-planning) | |
| 01:43 | Further enhancements to generated VDBE code for ORDER BY. (check-in: e7188fad user: drh tags: orderby-planning) | |
|
2014-03-23
| ||
| 18:47 | merge fixes from trunk (check-in: faf7f9ca user: drh tags: orderby-planning) | |
| 17:45 | Use only a single OP_MakeRecord instead of two when constructing entries to go onto a sorter. (check-in: d696cded user: drh tags: orderby-planning) | |
| 16:29 | Avoid a possible use of an uninitialized variable following an I/O or OOM error. (check-in: 641408a1 user: drh tags: trunk) | |
|
2014-03-22
| ||
| 02:19 | Clear the sort-order flags on the KeyInfo object associated with the OP_Compare/OP_Jump of the block-sort logic, in order to make the OP_Jump fully testable. (check-in: 9b2838f2 user: drh tags: trunk) | |
| 01:52 | Increase the version number to 3.8.5. (check-in: 9760879d user: drh tags: trunk) | |
| 00:27 | Fix the ORDER BY optimization logic so that it will do a block-sort on a partial DESC ORDER BY. This enhancement uncovered a memory leak in pushUntoSorter() which is also fixed. (check-in: c36f7461 user: drh tags: trunk) | |
|
2014-03-21
| ||
| 20:58 | Add the ability to use indices for the first few terms of an ORDER BY clause, then sort in batches to handle the later terms. (check-in: fa06a6fe user: drh tags: trunk) | |
| 19:56 | Change the names of SRT_DistTable and SRT_Table used by CTE to more meaningful SRT_DistFifo and SRT_Fifo, respectively. Simplify the IgnorableOrderby() macro in the process. (check-in: 45d8cc67 user: drh tags: trunk) | |
| 19:27 | Avoid leaking memory in an obscure case where the flattener adds an ORDER BY clause to the recursive part of a recursive query. (check-in: 1f413aca user: dan tags: trunk) | |