| Age | Commit message (Expand) | Author |
|---|
| 2014-08-13 | Add flags to page release, respect WT_READ_NO_GEN to avoid forced eviction. | Michael Cahill |
| 2014-07-17 | Clean up __cursor_pos_clear: only clear the fields we need to clear and | Keith Bostic |
| 2014-07-17 | Rename __cursor_search_clear __cursor_pos_clear; it's the code that | Keith Bostic |
| 2014-07-07 | Encode the location/length of simple values in the WT_ROW pointer, along | Keith Bostic |
| 2014-07-02 | Merge pull request #1090 from wiredtiger/onpage-cell | Michael Cahill |
| 2014-07-01 | Minor code shuffling to simplify setup. | Keith Bostic |
| 2014-06-30 | Convert row-store fast on-page key processing from per-page to per key. | Keith Bostic |
| 2014-06-25 | Merge branch 'develop' into split-memory-management | Michael Cahill |
| 2014-06-20 | Free memory more aggressively after splits. Keep the list of allocated objec... | Michael Cahill |
| 2014-06-18 | Merge branch 'develop' into avoid-cell-unpack | Keith Bostic |
| 2014-06-14 | Fixes for WT_INSERT objects with invisible WT_UPDATE objects. For | Keith Bostic |
| 2014-05-30 | Experimental change to avoid unpacking cells in order to access | Keith Bostic |
| 2014-04-29 | Change __wt_buf_grrow_worker to copy non-local data into place, this | Keith Bostic |
| 2014-04-28 | Change the block manager to not touch the WT_ITEM.{mem,memsize} fields | Keith Bostic |
| 2014-03-29 | Experimental change: move the WT_PAGE.{parent,ref_hint} fields into the | Keith Bostic |
| 2014-01-09 | Avoid unpacking the key cell multiple times when returning key/value | Keith Bostic |
| 2014-01-07 | Update copyright notices for 2014. | Keith Bostic |
| 2013-12-10 | Avoid bouncing the session cursor count between zero and one when searching a... | Michael Cahill |
| 2013-11-15 | Move the cache full check to cursor enter, when there are no cursors positioned. | Michael Cahill |
| 2013-11-15 | Read-only workloads need a cache full check. | Michael Cahill |
| 2013-11-08 | Don't block the oldest reader in the system forever. Try harder to evict whe... | Michael Cahill |
| 2013-10-18 | There's magic code to handle overflow records that have been deleted but | Keith Bostic |
| 2013-10-05 | Comment: explain why we copy the prefix before growing the buffer. | Michael Cahill |
| 2013-10-05 | Fix a bug scanning through a memory-mapped file with overflow items (which is... | Michael Cahill |
| 2013-10-04 | minor cleanup, only assign to kb.{data,size} once. | Keith Bostic |
| 2013-10-05 | Always cache the last key returned for use in scans. Otherwise, we can do hu... | Michael Cahill |
| 2013-09-12 | Fix a bug where a read-uncommitted scan was keeping the oldest transaction ID | Michael Cahill |
| 2013-08-24 | The cursor's write-generation value is no longer used, it's only set and | Keith Bostic |
| 2013-07-30 | Wrap the btree cursor operations with code that saves/restores the cursor | Keith Bostic |
| 2013-07-20 | We do data copies to the return WT_ITEM inside of __wt_row_leaf_key_copy() | Keith Bostic |
| 2013-07-20 | Clarify a comment, no other change. | Keith Bostic |
| 2013-07-19 | Change "short" data items from 127B or less to be 63B or less, and use the | Keith Bostic |
| 2013-07-13 | __wt_row_key_copy is only called on row-store leaf pages -- make that | Keith Bostic |
| 2013-07-13 | Turn off Huffman encoding for keys on row-store internal pages. | Keith Bostic |
| 2013-07-13 | Remove support for prefix-compression on row-store internal pages. | Keith Bostic |
| 2013-05-15 | typo | Michael Cahill |
| 2013-05-15 | Fix ordering of application eviction so a session's read-committed transactio... | Michael Cahill |
| 2013-04-23 | Only evict a single page from __cursor_leave to avoid latency spikes. | Alex Gorrod |
| 2013-03-25 | Merge branch 'develop' into data-handles | Michael Cahill |
| 2013-03-25 | Merge branch 'develop' into data-handles | Michael Cahill |
| 2013-03-22 | Fix retry logic in autocommit transactions, check for errors from __wt_cache_... | Michael Cahill |
| 2013-03-22 | Move the cache full check for autocommit transactions out of the rollback pat... | Michael Cahill |
| 2013-03-02 | Attempt to avoid test/format getting stuck: check if the cache is full before | Michael Cahill |
| 2013-03-01 | Fix a fat-fingered change: we always need to release our snapshot when the la... | Michael Cahill |
| 2013-02-18 | Don't check if the cache is full when resetting a cursor if a transaction is ... | Michael Cahill |
| 2013-02-06 | Remove support for the no-cache configuration, revert back to maintaining | Keith Bostic |
| 2013-01-01 | Update copyright notice to 2013. | Keith Bostic |
| 2012-12-19 | Don't clear the cursor APP flags until the key or value is update. Otherwise... | Michael Cahill |
| 2012-12-19 | whitespace | Michael Cahill |
| 2012-12-19 | Feedback from Keith's review of the change to track whether cursors are point... | Michael Cahill |
