summaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAgeFilesLines
...
| * | | SERVER-21553 Review feedback.Michael Cahill2015-12-021-2/+3
| | | |
| * | | SERVER-21553 Remove no-longer-used variables.Michael Cahill2015-12-021-4/+0
| | | |
| * | | SERVER-21553 Finish rename to __wt_ref_addr_free.Michael Cahill2015-12-022-3/+3
| | | |
| * | | SERVER-21553 Free blocks during reverse splits.Michael Cahill2015-12-024-48/+57
|/ / /
* | | Merge pull request #2350 from wiredtiger/WT-2251-ref-addr-leakAlex Gorrod2015-12-016-33/+35
|\ \ \
| * | | WT-2251 Clear the original ref->addr before an in-memory split to avoid ↵Michael Cahill2015-12-011-0/+11
| | | | | | | | | | | | | | | | use-after-free.
| * | | Merge branch 'develop' into WT-2251-ref-addr-leakMichael Cahill2015-12-013-23/+28
| |\ \ \ | |/ / / |/| | |
* | | | Merge pull request #2349 from wiredtiger/SERVER-21691Alex Gorrod2015-12-013-23/+28
|\ \ \ \ | | | | | | | | | | SERVER-21691 Avoid insert stalls
| * | | | SERVER-21691 Don't retry eviction of a page if transaction state hasn't changed.Michael Cahill2015-12-012-15/+17
| | | | | | | | | | | | | | | | | | | | This check was moved out of __wt_page_can_evict, which meant it only applied to LRU eviction. Move it back so that we don't repeatedly try forced eviction that has no chance of succeeding.
| * | | | SERVER-21691 Avoid blocking in-memory splits.Michael Cahill2015-12-011-8/+11
|/ / / / | | | | | | | | | | | | Only trylock the page's reconciliation lock (so in-memory splits aren't blocked by slow splits of siblings), don't give up as soon as a checkpoint starts.
| * | | WT-2251 Cleanup: no longer need to clear ref->addr after freeing.Michael Cahill2015-12-012-4/+0
| | | |
| * | | WT-2251 Free addresses when we discard deleted page references.Michael Cahill2015-12-016-29/+24
|/ / / | | | | | | | | | There was a lot of repeated code to check for offpage and free ref->addr, create an inlined function.
* | | Merge pull request #2346 from wiredtiger/WT-2249-eviction-stuckMichael Cahill2015-12-011-14/+17
|\ \ \ | |_|/ |/| | WT-2249 Keep eviction stuck until cache usage is under 100%.
| * | WT-2249 Review fix.Michael Cahill2015-12-011-2/+0
| | |
| * | WT-2249 Keep eviction stuck until cache usage is under 100%.Michael Cahill2015-11-301-12/+17
| | | | | | | | | | | | | | | | | | | | | Fix a real bug: STUCK is a cache flag, not a cache->state flag. Don't try to do eviction when dumping the cache, and only kill processes if no pages have been evicted since the STUCK flag was first set.
* | | Merge pull request #2347 from wiredtiger/WT-2250Keith Bostic2015-11-301-1/+1
|\ \ \ | | | | | | | | WT-2250 Minor fix. Use SET instead of INCRV for stat.
| * | | WT-2250 Minor fix. Use SET instead of INCRV for stat.Susan LoVerso2015-11-301-1/+1
|/ / /
* | | Don't complain if the long-running-transaction config is turned off forKeith Bostic2015-11-301-1/+1
|/ / | | | | | | fixed-length column stores.
* | Merge pull request #2340 from wiredtiger/WT-2244Michael Cahill2015-11-3010-113/+51
|\ \ | | | | | | WT-2244 - Trigger in-memory splits sooner.
| * \ Merge branch 'develop' into WT-2244Michael Cahill2015-11-304-5/+8
| |\ \ | |/ / |/| |
* | | Merge pull request #2344 from wiredtiger/wt-2248-session-closeMichael Cahill2015-11-301-0/+3
|\ \ \ | | | | | | | | WT-2248: WT_SESSION.close is updating WT_CONNECTION_IMPL.default_session
| * | | WT-2248: The code in WT_SESSION.close does work using the WT_CONNECTION_IMPLKeith Bostic2015-11-281-0/+3
|/ / / | | | | | | | | | | | | default session, but that means the API_END_RET_NOTFOUND_MAP session cleanup happens in the context of the default session, which isn't correct.
* | | Update a comment.Keith Bostic2015-11-271-2/+2
| | |
* | | Fix a few comments around cursor in-use count support.Keith Bostic2015-11-272-3/+3
| | |
| * | Rename __wt_page_can_split to be __wt_leaf_page_can_split, we splitKeith Bostic2015-11-273-5/+5
| | | | | | | | | | | | | | | internal pages a lot more these days. (It's really row-store leaf page, but that's a bug we'll eventualy fix.)
| * | Merge branch 'develop' into WT-2244Keith Bostic2015-11-271-2/+1
| |\ \ | |/ / |/| |
* | | __wt_page_modify_clear calls __wt_page_is_modified internally.Keith Bostic2015-11-271-2/+1
| | |
| * | Fix a comment, we no longer check a transaction value.Keith Bostic2015-11-271-3/+1
| | |
| * | whitespaceMichael Cahill2015-11-271-1/+1
| | |
| * | WT-2244 Don't wait for transactions after splits.Michael Cahill2015-11-277-55/+8
| | | | | | | | | | | | | | | | | | | | | Now that we split before pages become completely full, and we can block splits outright with a flag, don't use transaction visibilty to block subsequent splits. This had the effect of limiting each page to a single split during long-running transactions including checkpoints.
| * | WT-2244 Refine eviction during checkpoints.Michael Cahill2015-11-274-46/+23
| | | | | | | | | | | | | | | | | | In particular, take the checkpoint genertion into account when tracking the oldest ID in eviction, so that we retry eviction as soon as a checkpoint moves forward.
| * | Merge branch 'develop' into WT-2244Michael Cahill2015-11-279-42/+84
| |\ \ | |/ / |/| |
* | | Merge pull request #2341 from wiredtiger/SERVER-21553Michael Cahill2015-11-272-11/+14
|\ \ \ | | | | | | | | SERVER-21553 Enable fast-path truncate after splits.
| * | | SERVER-21553 Fix a line removed by mistake: set the address during splits.Michael Cahill2015-11-271-0/+1
| | | |
| * | | WT-2245 - Code Review changesDavid Hows2015-11-270-0/+0
| | | |
| * | | Merge branch 'develop' into SERVER-21553Michael Cahill2015-11-273-9/+23
| |\ \ \ | |/ / / |/| | |
* | | | Merge pull request #2339 from wiredtiger/WT-2243Alex Gorrod2015-11-273-9/+23
|\ \ \ \ | |_|_|/ |/| | | WT-2243 Don't keep transaction IDs pinned for reading from checkpoints.
| * | | WT-2243 Only clear btree cursor flags associated with position.Michael Cahill2015-11-262-5/+6
| | | | | | | | | | | | | | | | Previously, we were clearing the new "no transaction" flag every time a cursor was repositioned.
| * | | WT-2243 Don't keep transaction IDs pinned for reading from checkpoints.Michael Cahill2015-11-263-4/+17
| | | |
| | * | SERVER-21553 Enable fast-path truncate after splits.Michael Cahill2015-11-272-12/+14
| |/ / |/| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | A truncate operation attempts to mark leaf pages deleted without reading them into cache. One of the conditions that has to be met for that fast-path truncate of pages is that the leaf page not contain overflow items (or we would need to read it in order to delete the overflow items). The "no overflow" flag was not being preseved across internal page splits, so recent changes to splits were defeating fast-path truncation. Add tracking of the "no overflow" flag for in-memory page addresses so fast-path truncates work after internal pages are split.
* | | Fix some OS X clang warnings.Michael Cahill2015-11-272-2/+2
| | | | | | | | | | | | | | | ext/extractors/csv/csv_extractor.c:155:25: error: implicit conversion loses integer precision: 'long' to 'int' [-Werror,-Wshorten-64-to-32] src/cursor/cur_join.c:1022:16: error: implicit conversion loses integer precision: 'size_t' (aka 'unsigned long') to 'u_int' (aka 'unsigned int') [-Werror,-Wshorten-64-to-32]
* | | Merge pull request #2324 from wiredtiger/wt-2230-multi-split-error-pathAlex Gorrod2015-11-263-12/+24
|\ \ \ | | | | | | | | WT-2230: multi-split error path
| * | | WT-2230: incomplete change, __wt_free_ref can't reference ref->page,Keith Bostic2015-11-201-1/+1
| | | | | | | | | | | | | | | | use ref->home.
| * | | WT-2230: comment typo.Keith Bostic2015-11-201-2/+2
| | | |
| * | | WT-2230: The split_multi_inmem_fail function calls wt_free_ref withKeith Bostic2015-11-203-12/+24
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | ref->page as its page argument, which is used by wt_free_ref to check the page type and then free row-store keys instantiated in the WT_REF. That's an error, and we'd drop core because ref->page is freed by wt_free_ref before the page type is checked. Instead, pass in a page type to resolve questions about WT_REF.key, and use WT_REF.home to resolve questions about WT_REF.addr.
* | | | Merge pull request #2320 from wiredtiger/wt-2228-raw-compressionAlex Gorrod2015-11-261-8/+21
|\ \ \ \ | |_|/ / |/| | | WT-2228: avoid unnecessary raw-compression calls.
| * | | WT-2228: The WiredTiger reconciliation process calls raw compressionKeith Bostic2015-11-191-8/+21
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | (for example, if the engine is configured with zlib), even when the data chunk is smaller than the allocation size and compression is known to be a waste of time. When the allocation size is a significant percentage of the maximum block size, this can be half or more of the calls to raw compression.
| | | * WT-2244 - Trigger in-memory splits sooner.Michael Cahill2015-11-264-6/+16
| |_|/ |/| | | | | | | | Specifically, do an in-memory split when we hit 80% of memory_page_max.
* | | Merge pull request #2336 from wiredtiger/server-21619-dont-split-dead-treeAlex Gorrod2015-11-267-46/+46
|\ \ \ | | | | | | | | SERVER-21619 Don't do internal page splits after a tree is marked DEAD.
| * | | SERVER-21619 Revert an assertion change.Michael Cahill2015-11-261-2/+1
| | | |
View Lorry Controller Status