| Rewrote / refactored page.h. Now, it uses a freelist to reclaim slotid's within a page, and is implemented as two seperate modules. The first handles lsn's, and (will soon) check the type of each page, and then delegate work to a particular implementation of the second module. |
|
More...
|
almost 20 years ago
|
| O_DIRECT is working now. |
|
More...
|
almost 20 years ago
|
| Moved some utility functions into their own file. |
|
More...
|
almost 20 years ago
|
| O_DIRECT (sort of...) |
|
More...
|
almost 20 years ago
|
| Refactored bufferMananger / pageCache, so that bufferMananger handles loading an unloading pages, and pageCache only implements the replacement policy. |
|
More...
|
almost 20 years ago
|
| Implemented (but didn't test) iterator over the records in a page. |
|
More...
|
almost 20 years ago
|
| Re-enabled non-blob tests. |
|
More...
|
almost 20 years ago
|
| Blobs are now re-entrant. |
|
More...
|
almost 20 years ago
|
| bufferMananger is (really!) re-entrant now, performed big refactoring to prevent deadlocks, check_transactional2 ran overnight without coring, w/ 25 threads, and 10000 allocate/write/reads per thread. (The whole test suite was run in a loop...) |
|
More...
|
almost 20 years ago
|
| Just committing some documentation updates before big refactoring. |
|
More...
|
almost 20 years ago
|
| bufferMananger is *really* reentrant now! (I think) However, the same thread may not load a page more than once. So, LLADD has to be altered so that it never calls loadPage() more than once per operation (more efficient), and/or pin/unpin need to be implemented. An easy way to do pin/unpin is to have them pull things out of the cache replacement list, and then re-insert them. |
|
More...
|
almost 20 years ago
|
| Continuing work on multi-threading. r/w access to buffer manager getting close, but still buggy. |
|
More...
|
almost 20 years ago
|
| pageCache.c is now re-entrant. |
|
More...
|
almost 20 years ago
|
| pageCache is now re-entrant, in theory. |
|
More...
|
almost 20 years ago
|
| Moved page.h and some of the logging headers out of the public API. |
|
More...
|
almost 20 years ago
|
| Cleaning up bufferManager / page for locking. Want to limit access to the Page struct. |
|
More...
|
almost 20 years ago
|
| page.c is re-entrant (mostly), and now reuses DeRalloced space properly. (For now, BufferManager still is not re-entrant, and also prevents space from being reused.) |
|
More...
|
almost 20 years ago
|
| Made pageWriteLSN static. |
|
More...
|
almost 20 years ago
|
| More documentation fixes. |
|
More...
|
almost 20 years ago
|
| Fixed groupings for LLADD API docs. |
|
More...
|
almost 20 years ago
|
| *** empty log message *** |
|
More...
|
almost 20 years ago
|
| Ported LLADD to Fedora, cleaned up autoconf setup, and numerouse #includes that were problematic. |
|
More...
|
almost 20 years ago
|
| Documentation update, added latch profiling tools. |
|
More...
|
almost 20 years ago
|
| Log truncation. (But no checkpoints, so it doesn't get called... it does pass testing though. :) |
|
More...
|
almost 20 years ago
|
| Logwriter can now handler partial log entries correctly (it ignores them) Buffer Mananger no longer steals pages pre-maturely Alloc is an operation, and correct. Caching broken out into a new module Wrote a utility to translate a log file into human-readable ascii. |
|
More...
|
almost 20 years ago
|
| Cleaned out old cruft, such as commented out code, dead data structures, and dead files. sloccount went from $75,000 to $50,000 on src/lladd. ;) |
|
More...
|
almost 20 years ago
|
| Bugfixes ; blobs pass regression. Next stop: Delete old cruft. |
|
More...
|
almost 20 years ago
|
| Wrote blobmanager, didn't compile it yet. |
|
More...
|
almost 20 years ago
|
| blob manager commit. |
|
More...
|
almost 20 years ago
|
| Preliminary implementation of blobManager |
|
More...
|
almost 20 years ago
|
