| Commit message (Expand) | Author | Age | Files | Lines |
* | Only trigger the eviction server to look for pages that would force eviction ... | Michael Cahill | 2014-12-22 | 3 | -4/+4 |
* | Use the eviction server to write pages with READGEN_OLDEST set. | Alex Gorrod | 2014-12-19 | 3 | -4/+4 |
* | Set split-gen when deepening the tree, add paranoia that the eviction server ... | Michael Cahill | 2014-12-18 | 1 | -1/+2 |
* | Don't try to set and clear session->split_gen in WT_INTL_FOREACH_BEGIN: there... | Michael Cahill | 2014-12-18 | 4 | -6/+16 |
* | Detect write-write conflicts before no-overwrite cursors decide to skip an up... | Michael Cahill | 2014-12-17 | 1 | -15/+59 |
* | We document block_compressor supports a value of "none", but didn't | Keith Bostic | 2014-12-16 | 1 | -1/+1 |
* | We documented that huffman_key and huffman_value took "none" as an | Keith Bostic | 2014-12-16 | 1 | -13/+21 |
* | Even if we don't track any overflow pages during our read of the file, | Keith Bostic | 2014-12-15 | 1 | -4/+2 |
* | Most WiredTiger calloc calls are to allocate a single structure, | Keith Bostic | 2014-12-15 | 5 | -18/+18 |
* | Clean up a few __wt_calloc() calls that can be __wt_calloc_def() instead. | Keith Bostic | 2014-12-15 | 1 | -2/+2 |
* | Merge branch 'develop' into overflow-relax | Keith Bostic | 2014-12-09 | 1 | -1/+1 |
|\ |
|
| * | Merge branch 'develop' into log-compress | Michael Cahill | 2014-12-09 | 3 | -51/+72 |
| |\ |
|
| * \ | Merge branch 'develop' into log-compress | Michael Cahill | 2014-12-08 | 1 | -1/+1 |
| |\ \ |
|
| * \ \ | Merge branch 'develop' into log-compress | Don Anderson | 2014-12-04 | 6 | -8441/+5 |
| |\ \ \ |
|
| * \ \ \ | Merge branch 'orig-log-compress' into log-compress | Don Anderson | 2014-12-02 | 1 | -1/+1 |
| |\ \ \ \ |
|
| | * \ \ \ | Merge branch 'develop' into log-compress | Don Anderson | 2014-12-02 | 9 | -266/+748 |
| | |\ \ \ \ |
|
| | * | | | | | Fixed check to see if compression helped. In some cases (e.g. snappy), | Don Anderson | 2014-12-02 | 1 | -1/+1 |
* | | | | | | | Separate the btree maximum key/value sizes from the underlying page | Keith Bostic | 2014-12-09 | 2 | -122/+65 |
| |_|_|_|_|/
|/| | | | | |
|
* | | | | | | Merge branch 'develop' into force-evict-many-deletes | Michael Cahill | 2014-12-08 | 1 | -1/+1 |
|\ \ \ \ \ \
| | |_|_|_|/
| |/| | | | |
|
| * | | | | | Merge branch 'develop' into none-config | Alex Gorrod | 2014-12-08 | 7 | -8437/+3 |
| |\ \ \ \ \
| | | |_|_|/
| | |/| | | |
|
| * | | | | | Check for configurations of "none" in a wrapper around the config string pars... | Michael Cahill | 2014-12-08 | 1 | -1/+1 |
* | | | | | | Revert changes to flow control that are no longer required. | Michael Cahill | 2014-12-08 | 2 | -57/+32 |
* | | | | | | Check whether to evict pages with many deletes where: | Michael Cahill | 2014-12-08 | 2 | -8/+16 |
* | | | | | | Don't clear the count of deleted items on every iteration, clear it | Keith Bostic | 2014-12-05 | 2 | -2/+5 |
* | | | | | | Maintain the count of deleted items in the cursor until it leaves the | Keith Bostic | 2014-12-05 | 2 | -71/+30 |
* | | | | | | Do the forced eviction call in the lower-level next/prev code. Now that we m... | Michael Cahill | 2014-12-05 | 2 | -70/+110 |
* | | | | | | Force eviction if we see many consecutive deletes when scanning through a page. | Michael Cahill | 2014-12-05 | 2 | -20/+30 |
* | | | | | | Move the READGEN_OLDEST logic into a function to make it less obscure. | Michael Cahill | 2014-12-05 | 3 | -4/+4 |
* | | | | | | Push the logic about whether to evict pages with too many deleted items down ... | Michael Cahill | 2014-12-05 | 2 | -39/+65 |
| |/ / / /
|/| | | | |
|
* | | | | | Merge branch 'develop' into log-recycle | Michael Cahill | 2014-12-04 | 6 | -8435/+1 |
|\ \ \ \ \ |
|
| * \ \ \ \ | Merge branch 'develop' into evict-rename | Michael Cahill | 2014-12-04 | 1 | -5/+4 |
| |\ \ \ \ \
| | |/ / / / |
|
| * | | | | | Move eviction and reconciliation files into their own subdirs. | Alex Gorrod | 2014-12-03 | 5 | -8444/+0 |
| * | | | | | Shuffle some eviction/btree file names. | Alex Gorrod | 2014-12-03 | 5 | -129/+138 |
| | |_|_|/
| |/| | | |
|
* | | | | | Merge branch 'develop' into log-recycle | Susan LoVerso | 2014-12-03 | 2 | -23/+21 |
|\ \ \ \ \
| | |/ / /
| |/| / /
| |_|/ /
|/| | | |
|
| * | | | Don't try in-memory splits when closing a tree. | Michael Cahill | 2014-12-03 | 1 | -1/+1 |
| * | | | Only do in-memory splits on large, dirty pages. | Michael Cahill | 2014-12-03 | 1 | -5/+4 |
| * | | | Don't attempt to split pages when checkpoints are active in a tree. | Michael Cahill | 2014-12-03 | 1 | -18/+17 |
| |/ / |
|
* | | | Make pre-alloc a boolean and convert connection log settings to flags. | Susan LoVerso | 2014-12-02 | 1 | -2/+2 |
|/ / |
|
* | | Use the original approach to finding the first key on a page: search the smal... | Michael Cahill | 2014-12-02 | 1 | -9/+9 |
* | | Remove stray debugging. | Michael Cahill | 2014-12-02 | 1 | -2/+0 |
* | | Instantiate a full key for the parent ref of a page during in-memory splits. ... | Michael Cahill | 2014-12-02 | 1 | -3/+35 |
* | | Inherit the original page's dirty transaction during a split so that checkpoi... | Michael Cahill | 2014-12-02 | 1 | -4/+3 |
* | | Don't force-evict pages that have just been through an in-memory split. | Michael Cahill | 2014-12-02 | 1 | -0/+16 |
* | | Merge branch 'develop' into split-right | Michael Cahill | 2014-12-02 | 1 | -0/+3 |
|\ \ |
|
| * | | Add a statistic to count how much eviction application threads are doing. | Michael Cahill | 2014-12-02 | 1 | -0/+3 |
* | | | Look for an in-memory split in the reconciliation path before checking | Alex Gorrod | 2014-12-02 | 1 | -13/+18 |
* | | | Add error-handling code to the insert split, if the split into the | Keith Bostic | 2014-12-01 | 1 | -43/+58 |
* | | | Reviewing the code, minor whitespace changes. | Keith Bostic | 2014-12-01 | 1 | -12/+7 |
* | | | Instead of checking for a single level-5 skip list entry, check for two | Keith Bostic | 2014-12-01 | 1 | -15/+17 |
* | | | whitespace | Keith Bostic | 2014-12-01 | 1 | -3/+2 |