diff options
author | Heikki Linnakangas <heikki.linnakangas@iki.fi> | 2014-11-20 17:56:26 +0200 |
---|---|---|
committer | Heikki Linnakangas <heikki.linnakangas@iki.fi> | 2014-11-20 18:46:41 +0200 |
commit | 2c03216d831160bedd72d45f712601b6f7d03f1c (patch) | |
tree | ab6a03d031ffa605d848b0b7067add15e56e2207 /src/include/access/xlogutils.h | |
parent | 8dc626defec23016dd5988208d8704b858b9d21d (diff) | |
download | postgresql-2c03216d831160bedd72d45f712601b6f7d03f1c.tar.gz postgresql-2c03216d831160bedd72d45f712601b6f7d03f1c.zip |
Revamp the WAL record format.
Each WAL record now carries information about the modified relation and
block(s) in a standardized format. That makes it easier to write tools that
need that information, like pg_rewind, prefetching the blocks to speed up
recovery, etc.
There's a whole new API for building WAL records, replacing the XLogRecData
chains used previously. The new API consists of XLogRegister* functions,
which are called for each buffer and chunk of data that is added to the
record. The new API also gives more control over when a full-page image is
written, by passing flags to the XLogRegisterBuffer function.
This also simplifies the XLogReadBufferForRedo() calls. The function can dig
the relation and block number from the WAL record, so they no longer need to
be passed as arguments.
For the convenience of redo routines, XLogReader now disects each WAL record
after reading it, copying the main data part and the per-block data into
MAXALIGNed buffers. The data chunks are not aligned within the WAL record,
but the redo routines can assume that the pointers returned by XLogRecGet*
functions are. Redo routines are now passed the XLogReaderState, which
contains the record in the already-disected format, instead of the plain
XLogRecord.
The new record format also makes the fixed size XLogRecord header smaller,
by removing the xl_len field. The length of the "main data" portion is now
stored at the end of the WAL record, and there's a separate header after
XLogRecord for it. The alignment padding at the end of XLogRecord is also
removed. This compansates for the fact that the new format would otherwise
be more bulky than the old format.
Reviewed by Andres Freund, Amit Kapila, Michael Paquier, Alvaro Herrera,
Fujii Masao.
Diffstat (limited to 'src/include/access/xlogutils.h')
-rw-r--r-- | src/include/access/xlogutils.h | 21 |
1 files changed, 6 insertions, 15 deletions
diff --git a/src/include/access/xlogutils.h b/src/include/access/xlogutils.h index 8d906967232..68f72cfac6d 100644 --- a/src/include/access/xlogutils.h +++ b/src/include/access/xlogutils.h @@ -11,7 +11,7 @@ #ifndef XLOG_UTILS_H #define XLOG_UTILS_H -#include "access/xlogrecord.h" +#include "access/xlogreader.h" #include "storage/bufmgr.h" @@ -33,26 +33,17 @@ typedef enum * replayed) */ } XLogRedoAction; -extern XLogRedoAction XLogReadBufferForRedo(XLogRecPtr lsn, XLogRecord *record, - int block_index, RelFileNode rnode, BlockNumber blkno, - Buffer *buf); -extern XLogRedoAction XLogReadBufferForRedoExtended(XLogRecPtr lsn, - XLogRecord *record, int block_index, - RelFileNode rnode, ForkNumber forkno, - BlockNumber blkno, +extern XLogRedoAction XLogReadBufferForRedo(XLogReaderState *record, + uint8 buffer_id, Buffer *buf); +extern Buffer XLogInitBufferForRedo(XLogReaderState *record, uint8 block_id); +extern XLogRedoAction XLogReadBufferForRedoExtended(XLogReaderState *record, + uint8 buffer_id, ReadBufferMode mode, bool get_cleanup_lock, Buffer *buf); -extern Buffer XLogReadBuffer(RelFileNode rnode, BlockNumber blkno, bool init); extern Buffer XLogReadBufferExtended(RelFileNode rnode, ForkNumber forknum, BlockNumber blkno, ReadBufferMode mode); -extern Buffer RestoreBackupBlock(XLogRecPtr lsn, XLogRecord *record, - int block_index, - bool get_cleanup_lock, bool keep_buffer); -extern Buffer RestoreBackupBlockContents(XLogRecPtr lsn, BkpBlock bkpb, - char *blk, bool get_cleanup_lock, bool keep_buffer); - extern Relation CreateFakeRelcacheEntry(RelFileNode rnode); extern void FreeFakeRelcacheEntry(Relation fakerel); |