PThorpe92
d702e6a80c
Polish checkpointing and fix tests, add documentation
2025-07-30 14:08:53 +03:00
PThorpe92
8ec99a9143
Remove assert for !NO_LOCK_HELD, properly handle writing header if reset
2025-07-30 14:08:51 +03:00
PThorpe92
529cc14e29
Fix wal tests remove unwrap from previous Result return val
2025-07-30 14:08:33 +03:00
PThorpe92
7640535ba4
Fix transaction read0 shortcut in WAL and track whether we have snapshot
2025-07-30 14:08:33 +03:00
PThorpe92
ff1987a45c
Temporarily remove optimization for new read tx to grab read mark 0 and skip db file
2025-07-30 14:08:33 +03:00
PThorpe92
318bfa9590
Change incorrect comments and rename guard
2025-07-30 14:08:33 +03:00
PThorpe92
1490a586b1
Apply suggestions/fixes and add extensive comments to wal chkpt
2025-07-30 14:08:33 +03:00
PThorpe92
3e75444388
Remove panic in cacheflush io.block in pager now that checkpoitns can return busy
2025-07-30 14:08:33 +03:00
PThorpe92
3db72cf111
Just forget Full checkpoint mode for now, comment out compat test
2025-07-30 14:08:33 +03:00
PThorpe92
49f90980d4
Create new header after truncation chkpt
2025-07-30 14:08:33 +03:00
PThorpe92
b214c3dfc8
Add diff chkpt modes to sqlite3 api, finish checkpoint logic and add tests
2025-07-30 14:08:33 +03:00
PThorpe92
eaa6f99fa8
Hold and ensure release of proper locks if we trunc the db file post-checkpoint
2025-07-30 14:08:33 +03:00
PThorpe92
8ca37b71b6
Ensure we properly hold and release read locks in log restart method and fix tests
2025-07-30 14:08:33 +03:00
PThorpe92
9b7e5ed292
Trunc db file after backfilling everything in calling method
2025-07-30 14:08:33 +03:00
PThorpe92
1a9b7ef76e
Add support for truncate, restart and full checkpointing methods
2025-07-30 14:08:31 +03:00
PThorpe92
ad286bb873
Use new wait_for_completion for sync IO
2025-07-30 14:07:04 +03:00
PThorpe92
9c75872827
Make writing wal header able to be sync
2025-07-30 14:05:18 +03:00
PThorpe92
dcf346b9b0
Update apparently very old branch to new api
2025-07-30 13:48:05 +03:00
PThorpe92
edbc29e2a6
Add restart_header and truncate for WAL and tests to prep for truncate checkpoint mode
2025-07-30 13:48:05 +03:00
PThorpe92
272a63b562
Fix truncate impl in JS bindings
2025-07-30 13:48:05 +03:00
PThorpe92
73fe4ffa06
Add truncate method to DatabaseStorage trait
2025-07-30 13:48:05 +03:00
pedrocarlo
a1c5a6efae
begin_sync should return a Completion
2025-07-29 13:40:58 -03:00
pedrocarlo
58b51e036d
read_page should return a Completion
2025-07-29 12:42:36 -03:00
pedrocarlo
6edf65f752
append_frame should return a Completion
2025-07-29 11:49:10 -03:00
pedrocarlo
be3793e9f1
begin_write_wal_header should return completion + wait for completion in new_shared
2025-07-29 11:45:44 -03:00
pedrocarlo
5861145fb4
read_frame should return a Completion
2025-07-29 11:40:53 -03:00
Pekka Enberg
8adc807cd7
Merge 'Change function signatures to return IO Completions' from Pedro Muniz
...
Changes a couple of function signatures to return `Completion`. Also, I
changed `Completion` to be internally `Arc` to abstract the `Arc`
implementation detail, and to be able to attach a `#[must_use]` to the
`Completion` struct, so that cargo check can show us where we are not
tracking completions in the code. I also attached a `#[must_use]` to
`IOResult` so that we can see the places that we are not propagating or
waiting for I/O, demonstrating locations where functions should be
reentrant and are not.
Also, while we are with this refactor in progress I want to relax the
Clippy CI lint on unused_variables.
Closes #2309
2025-07-29 12:41:14 +03:00
pedrocarlo
3831e0db39
convert must_use compile warnings to unused_variables to track locations where we need to refactor in the future
2025-07-28 16:09:26 -03:00
pedrocarlo
d30c7d54c8
change all Arc<Completion> to Completion
2025-07-28 15:32:45 -03:00
pedrocarlo
7789c569a0
make Completion implementation contain an inner Arc<CompletionInner> so that we can must_use the Completion struct
2025-07-28 15:31:42 -03:00
pedrocarlo
617254116d
begin_read_page should return completions
2025-07-28 15:31:42 -03:00
pedrocarlo
28d6245440
begin_write_btree_page should return completion
2025-07-28 15:31:42 -03:00
pedrocarlo
3104e3fee5
adjust DatabaseStorage trait to return completions
2025-07-28 15:31:42 -03:00
Diego Reis
0346c65a72
Fix clippy
2025-07-28 14:48:52 -03:00
Pekka Enberg
e2d4cbbe48
Merge 'core: Enforce shared database object per database file' from Pekka Enberg
...
We need to ensures that there is a single, shared `Database` object per
a database file. We need because it is not safe to have multiple
independent WAL files open because coordination happens at process-level
POSIX file advisory locks.
Fixes #2267
Reviewed-by: Pere Diaz Bou <pere-altea@homail.com >
Closes #2299
2025-07-28 19:34:35 +03:00
Pekka Enberg
5b6a30c1df
core/storage: Fix B-Tree test cases to use ":memory:"
...
...otherwise they all share the same `Database` object.
2025-07-28 19:13:53 +03:00
Nikita Sivukhin
09b18f6b6e
add WAL API methods to the rust bindings and extend result of wal_insert_frame method
2025-07-28 17:20:10 +04:00
Jussi Saurio
36e0ca5a9f
pager: remove unnecessary LoadFreelistTrunkPage state
2025-07-28 10:11:57 +03:00
Jussi Saurio
e7b07c1357
pager: reset allocate_page_state in reset_internal_states()
2025-07-28 10:11:57 +03:00
Jussi Saurio
c349a9d689
Ensure underlying payload vec cannot be copied so that raw pointers remain valid
2025-07-28 10:11:57 +03:00
Jussi Saurio
08d5b3b4bc
btree: make fill_cell_payload() re-entrant (overflow pages may require IO)
2025-07-28 09:00:59 +03:00
Jussi Saurio
927aca7857
Fix incorrect autovacuum test
2025-07-28 09:00:59 +03:00
Jussi Saurio
e2e25a48f6
Pager: document origins of BtreePageAllocMode
2025-07-28 09:00:59 +03:00
Jussi Saurio
5ce65bf8e7
btree/pager: reuse freelist pages in allocate_page() to fix UPDATE perf
2025-07-28 09:00:59 +03:00
PThorpe92
b08c465450
Fix writing wal header for async IO
2025-07-27 21:52:13 -04:00
PThorpe92
e6737d923d
Return correct value for pragma checkpoint
2025-07-26 23:09:40 -04:00
PThorpe92
fb611390c0
Update test to use realistic expectations for should_checkpoint in cacheflush
2025-07-26 23:03:51 -04:00
PThorpe92
7c027fed8c
Keep should_checkpoint logic for now until greater checkpointing is fixed
2025-07-26 23:03:51 -04:00
PThorpe92
6644036be4
Stop checkpointing after every write when wal frame size > threshold
2025-07-26 23:03:47 -04:00
Glauber Costa
b8ee38868d
implement the pragma encoding
...
Do not allow setting it. That ship has sailed around 2005.
2025-07-26 19:37:39 -05:00
