summaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAgeFilesLines
* Cut WiredTiger release 2.2.02.2.0Alex Gorrod2014-05-218-18/+96
|
* Fix s_docs script to work with -l again.Alex Gorrod2014-05-211-0/+1
|
* Fix a memory leak, thanks valgrind!Keith Bostic2014-05-201-4/+3
|
* Update.Keith Bostic2014-05-201-0/+2
|
* Add a -F option to thread, to create a separate file for every thread.Keith Bostic2014-05-205-38/+91
| | | | | | | | With this option, the command line combination "-FS -R 0 -W 10" will trigger the checkpoint problem where closing a file fails because there is an update for the file we can't write. Reference #844, #957.
* Fix a bug where rolling back the first transaction to update an LSM tree ↵Michael Cahill2014-05-201-4/+13
| | | | | | could leave the metadata inconsistent. refs #976
* In the fast-path for avoiding scans of the global transaction table, check ↵Michael Cahill2014-05-191-1/+4
| | | | | | that we haven't raced. refs #1016
* Merge pull request #1013 from wiredtiger/compression-can-failMichael Cahill2014-05-151-36/+95
|\ | | | | Handle raw compression giving up.
| * Allow the raw compression function to fail, in which case we write theKeith Bostic2014-05-141-36/+95
| | | | | | | | | | | | | | | | | | | | | | original data as a single page. Don't accumulate more data unless the raw compression function requests it (by returning EAGAIN), or if we're stuck because the raw compression function consumed all of the data and there's more data so we may need a key for the next block. Reference #1007, #1008.
* | Sort the subconfig lists.Keith Bostic2014-05-142-24/+24
| |
* | Pull out configuration ordering into its own section, for clarity.Keith Bostic2014-05-141-28/+37
| |
* | Fix initialization bug in async reconfig. #1002Susan LoVerso2014-05-142-1/+6
| |
* | Mention the WIREDTIGER_CONFIG environment variable.Keith Bostic2014-05-141-4/+7
| |
* | Removing trailing ^M from each line.Keith Bostic2014-05-141-77/+77
| |
* | Reset locked variable when releasing the lock. #976Susan LoVerso2014-05-141-0/+1
| |
* | Return ENOTSUP if someone tries to allocate an async op handleSusan LoVerso2014-05-142-1/+78
|/ | | | without async enabled. Add python test case. #1014
* Merge pull request #1006 from wiredtiger/async-javaAlex Gorrod2014-05-145-38/+714
|\ | | | | Async java
| * Fixes for Java async stability and performance.Don Anderson2014-05-132-64/+119
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Cache field ids and method ids for all async operations. Invalidate unpackers by setting fields, rather than calling a method to do it. Use javaClose to invalidate the swigCptr for a AsyncOp at the end of the callback. Any use of an AsyncOp after its lifetime has ended will give a NullPointerException. Declare that new_async_op throws WiredTigerException, which happens when there are no more slots available. Added a test with a million items that typically does no pauses when creating ops, but catches WiredTigerException and retries with a short sleep.
| * KNF fixes and comment typo.Don Anderson2014-05-132-13/+19
| | | | | | | | Name async_new_op async_flush using underscore (not camelcase) convention.
| * Support async operations in Java. refs #933.Don Anderson2014-05-115-30/+645
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | - Created a real async handler for java allowing callbacks into java code. Code to be registered must implement the new AsyncCallback interface. - Renamed classes/methods to Java-ish names. - Fixes to support 'this' removal for AsyncOp. - Java's cursor ops, like 'insert' set up the value unpackers after insert completes. For AsyncOp, we take a different approach, and invalidate the unpacker when the op completes, letting the value be fetched when needed. - Added new test AsyncTest that tests doing different numbers of async ops with different parameters, and also with different types of keys/values.
* | Add async to reconfig test. Fix code to retain defaults. #1002Susan LoVerso2014-05-132-5/+33
| |
* | Allow async to be reconfigured. #1002Susan LoVerso2014-05-137-34/+188
| |
* | Don't roll forward the metadata when opening a hot backup: the version ↵Michael Cahill2014-05-134-20/+34
| | | | | | | | | | | | | | | | | | | | restored from the backup is the right one to use. Also fix a long-standing performance issue in recovery: we always have a checkpoint LSN in the turtle file for the metadata: use that rather than scanning the whole log for metadata updates. Lastly, this change means hot backup no longer has to copy files in any particular order. refs #972, #976.
* | Reserve more memory for zlib to finish a raw buffer -- it turns out to need ↵Michael Cahill2014-05-131-1/+1
| | | | | | | | | | | | 24 bytes instead of 12. refs #1007
* | Coverity. Clean up unused variables.Susan LoVerso2014-05-121-19/+2
| |
* | Add backoff to sleep if no async work available. #1002Susan LoVerso2014-05-126-98/+128
| |
* | Be a little more defensive: a reasonable compression function might notKeith Bostic2014-05-121-11/+15
| | | | | | | | | | be checking the target allocation size. If compression succeeds, but it doesn't gain us at least an allocation unit, don't compress the data.
* | Document dst and dst_len for WT_COMPRESSOR::compress.Keith Bostic2014-05-121-0/+6
| |
* | Fix a comment.Keith Bostic2014-05-121-1/+1
| |
* | Don't loop on calls to deflate until Z_OK isn't returned, we're not addingKeith Bostic2014-05-121-3/+3
| | | | | | | | | | additional buffer memory so no progress can be made (in other words, a Z_OK return from deflate means compression wasn't successful).
* | Remove decrement of the destination length by one, when setting up forKeith Bostic2014-05-121-1/+1
| | | | | | | | | | | | compression: one version of this code reserved a byte to figure out whether there were multiple blocks to inflate, but that isn't necessary for zlib.
* | Don't ignore the return value from deflateEnd.Keith Bostic2014-05-121-1/+4
| |
* | Merge pull request #1004 from wiredtiger/stat-log-reconfigMichael Cahill2014-05-129-139/+167
|\ \ | |/ |/| Stat log reconfig
| * Update checkpoint server reconfigure to match stat log semantics.Alex Gorrod2014-05-091-61/+11
| |
| * Make statistics log server reconfigurable.Alex Gorrod2014-05-098-78/+156
| |
* | If zlib compression fails because there's not enough room in the destinationKeith Bostic2014-05-101-7/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | buffer, deflateEnd will return Z_BUF_ERROR: From the docs (http://www.zlib.net/manual.html): deflate() returns Z_OK if some progress has been made (more input processed or more output produced), Z_STREAM_END if all input has been consumed and all output has been produced (only when flush is set to Z_FINISH), Z_STREAM_ERROR if the stream state was inconsistent (for example if next_in or next_out was NULL), Z_BUF_ERROR if no progress is possible (for example avail_in or avail_out was zero). Note that Z_BUF_ERROR is not fatal, and deflate() can be called again with more input and more output space to continue compressing. Also, I don't think there's any reason to loop around deflate calls when Z_FINISH is specified, I think deflate will return Z_OK the first time it's called, then return Z_BUF_ERROR the second time it's called, because no progress is possible. From the docs: If the parameter flush is set to Z_FINISH, pending input is processed, pending output is flushed and deflate returns with Z_STREAM_END if there was enough output space; if deflate returns with Z_OK, this function must be called again with Z_FINISH and more output space (updated avail_out) but no more input data, until it returns with Z_STREAM_END or an error. After deflate has returned Z_STREAM_END, the only possible operations on the stream are deflateReset or deflateEnd. Reference #1005.
* | whitespaceKeith Bostic2014-05-101-8/+4
| |
* | Run the "standard" extensions through the style checks, mostly add aKeith Bostic2014-05-1011-180/+272
| | | | | | | | | | | | | | bunch of function header comments. Change the compression source files to all have roughly identical patterns.
* | Coverity #1212124 Resource leak.Keith Bostic2014-05-091-4/+9
| |
* | Coverity #1212126: Uninitialized scalar variable.Keith Bostic2014-05-091-4/+4
| |
* | Fix a leak of the scratch buffer array in the dummy session: don't allocate ↵Michael Cahill2014-05-091-17/+8
| | | | | | | | scratch buffers until we have a real session.
* | typoMichael Cahill2014-05-091-1/+1
| |
* | Speed up checkpoints by doing a better job of skipping pages that can't ↵Michael Cahill2014-05-098-92/+47
| | | | | | | | | | | | | | | | contain changes that need to be included. Before the checkpoint pass, we have written every dirty leaf page at read-committed isolation. So the only pages that checkpoint has to write are internal pages, or leaf pages that were dirtied after the "write leaves" phase but before the checkpoint transaction started. Now that we separate out the allocation of a transaction ID from setting up a snapshot, we can distinguish between update transactions and the transaction we're using for the checkpoint. refs #954, #963, #1001
* | Fix a leak of the buffer used to hold config files.Michael Cahill2014-05-091-5/+8
|/
* Use the default cache size in test/thread: 5MB causes odd side effects that ↵Michael Cahill2014-05-091-1/+1
| | | | make the test slow, particularly with valgrind.
* Merge branch 'row-skip-append-check' into developMichael Cahill2014-05-091-6/+6
|\
| * Change search-near to always set the "insert" flag, LSM depends on it.Keith Bostic2014-05-081-6/+6
| |
* | Merge pull request #1003 from wiredtiger/row-skip-append-checkMichael Cahill2014-05-096-33/+50
|\ \ | |/ | | Row skip append check
| * Don't do an append test regardless if searching the tree, but moreKeith Bostic2014-05-085-26/+36
| | | | | | | | | | | | | | importantly, don't clear the "appending" flag for a read operation. This makes no difference in terms of performance, but means we don't let read operations interfere with our understanding of an appending workload.
| * Comment cleanup.Keith Bostic2014-05-081-3/+2
| |
View Lorry Controller Status