2012-11-28 16:35:01 +01:00
|
|
|
/*-------------------------------------------------------------------------
|
|
|
|
*
|
|
|
|
* gindesc.c
|
|
|
|
* rmgr descriptor routines for access/transam/gin/ginxlog.c
|
|
|
|
*
|
2023-01-02 21:00:37 +01:00
|
|
|
* Portions Copyright (c) 1996-2023, PostgreSQL Global Development Group
|
2012-11-28 16:35:01 +01:00
|
|
|
* Portions Copyright (c) 1994, Regents of the University of California
|
|
|
|
*
|
|
|
|
*
|
|
|
|
* IDENTIFICATION
|
|
|
|
* src/backend/access/rmgrdesc/gindesc.c
|
|
|
|
*
|
|
|
|
*-------------------------------------------------------------------------
|
|
|
|
*/
|
|
|
|
#include "postgres.h"
|
|
|
|
|
2017-02-14 21:37:59 +01:00
|
|
|
#include "access/ginxlog.h"
|
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.
2014-11-20 16:56:26 +01:00
|
|
|
#include "access/xlogutils.h"
|
2012-11-28 16:35:01 +01:00
|
|
|
#include "lib/stringinfo.h"
|
Change internal RelFileNode references to RelFileNumber or RelFileLocator.
We have been using the term RelFileNode to refer to either (1) the
integer that is used to name the sequence of files for a certain relation
within the directory set aside for that tablespace/database combination;
or (2) that value plus the OIDs of the tablespace and database; or
occasionally (3) the whole series of files created for a relation
based on those values. Using the same name for more than one thing is
confusing.
Replace RelFileNode with RelFileNumber when we're talking about just the
single number, i.e. (1) from above, and with RelFileLocator when we're
talking about all the things that are needed to locate a relation's files
on disk, i.e. (2) from above. In the places where we refer to (3) as
a relfilenode, instead refer to "relation storage".
Since there is a ton of SQL code in the world that knows about
pg_class.relfilenode, don't change the name of that column, or of other
SQL-facing things that derive their name from it.
On the other hand, do adjust closely-related internal terminology. For
example, the structure member names dbNode and spcNode appear to be
derived from the fact that the structure itself was called RelFileNode,
so change those to dbOid and spcOid. Likewise, various variables with
names like rnode and relnode get renamed appropriately, according to
how they're being used in context.
Hopefully, this is clearer than before. It is also preparation for
future patches that intend to widen the relfilenumber fields from its
current width of 32 bits. Variables that store a relfilenumber are now
declared as type RelFileNumber rather than type Oid; right now, these
are the same, but that can now more easily be changed.
Dilip Kumar, per an idea from me. Reviewed also by Andres Freund.
I fixed some whitespace issues, changed a couple of words in a
comment, and made one other minor correction.
Discussion: http://postgr.es/m/CA+TgmoamOtXbVAQf9hWFzonUo6bhhjS6toZQd7HZ-pmojtAmag@mail.gmail.com
Discussion: http://postgr.es/m/CA+Tgmobp7+7kmi4gkq7Y+4AM9fTvL+O1oQ4-5gFTT+6Ng-dQ=g@mail.gmail.com
Discussion: http://postgr.es/m/CAFiTN-vTe79M8uDH1yprOU64MNFE+R3ODRuA+JWf27JbhY4hJw@mail.gmail.com
2022-07-06 17:39:09 +02:00
|
|
|
#include "storage/relfilelocator.h"
|
2012-11-28 16:35:01 +01:00
|
|
|
|
2014-03-31 14:15:19 +02:00
|
|
|
static void
|
|
|
|
desc_recompress_leaf(StringInfo buf, ginxlogRecompressDataLeaf *insertData)
|
|
|
|
{
|
|
|
|
int i;
|
|
|
|
char *walbuf = ((char *) insertData) + sizeof(ginxlogRecompressDataLeaf);
|
|
|
|
|
|
|
|
appendStringInfo(buf, " %d segments:", (int) insertData->nactions);
|
|
|
|
|
|
|
|
for (i = 0; i < insertData->nactions; i++)
|
|
|
|
{
|
|
|
|
uint8 a_segno = *((uint8 *) (walbuf++));
|
|
|
|
uint8 a_action = *((uint8 *) (walbuf++));
|
|
|
|
uint16 nitems = 0;
|
|
|
|
int newsegsize = 0;
|
|
|
|
|
|
|
|
if (a_action == GIN_SEGMENT_INSERT ||
|
|
|
|
a_action == GIN_SEGMENT_REPLACE)
|
|
|
|
{
|
|
|
|
newsegsize = SizeOfGinPostingList((GinPostingList *) walbuf);
|
|
|
|
walbuf += SHORTALIGN(newsegsize);
|
|
|
|
}
|
|
|
|
|
|
|
|
if (a_action == GIN_SEGMENT_ADDITEMS)
|
|
|
|
{
|
|
|
|
memcpy(&nitems, walbuf, sizeof(uint16));
|
|
|
|
walbuf += sizeof(uint16);
|
|
|
|
walbuf += nitems * sizeof(ItemPointerData);
|
|
|
|
}
|
|
|
|
|
|
|
|
switch (a_action)
|
|
|
|
{
|
|
|
|
case GIN_SEGMENT_ADDITEMS:
|
|
|
|
appendStringInfo(buf, " %d (add %d items)", a_segno, nitems);
|
|
|
|
break;
|
|
|
|
case GIN_SEGMENT_DELETE:
|
|
|
|
appendStringInfo(buf, " %d (delete)", a_segno);
|
|
|
|
break;
|
|
|
|
case GIN_SEGMENT_INSERT:
|
|
|
|
appendStringInfo(buf, " %d (insert)", a_segno);
|
|
|
|
break;
|
|
|
|
case GIN_SEGMENT_REPLACE:
|
|
|
|
appendStringInfo(buf, " %d (replace)", a_segno);
|
|
|
|
break;
|
|
|
|
default:
|
|
|
|
appendStringInfo(buf, " %d unknown action %d ???", a_segno, a_action);
|
|
|
|
/* cannot decode unrecognized actions further */
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2012-11-28 16:35:01 +01:00
|
|
|
void
|
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.
2014-11-20 16:56:26 +01:00
|
|
|
gin_desc(StringInfo buf, XLogReaderState *record)
|
2012-11-28 16:35:01 +01:00
|
|
|
{
|
2014-06-14 09:46:48 +02:00
|
|
|
char *rec = XLogRecGetData(record);
|
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.
2014-11-20 16:56:26 +01:00
|
|
|
uint8 info = XLogRecGetInfo(record) & ~XLR_INFO_MASK;
|
2012-11-28 16:35:01 +01:00
|
|
|
|
|
|
|
switch (info)
|
|
|
|
{
|
|
|
|
case XLOG_GIN_CREATE_PTREE:
|
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.
2014-11-20 16:56:26 +01:00
|
|
|
/* no further information */
|
2012-11-28 16:35:01 +01:00
|
|
|
break;
|
|
|
|
case XLOG_GIN_INSERT:
|
2013-11-27 18:21:23 +01:00
|
|
|
{
|
|
|
|
ginxlogInsert *xlrec = (ginxlogInsert *) rec;
|
|
|
|
|
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.
2014-11-20 16:56:26 +01:00
|
|
|
appendStringInfo(buf, "isdata: %c isleaf: %c",
|
2013-11-27 18:21:23 +01:00
|
|
|
(xlrec->flags & GIN_INSERT_ISDATA) ? 'T' : 'F',
|
|
|
|
(xlrec->flags & GIN_INSERT_ISLEAF) ? 'T' : 'F');
|
|
|
|
if (!(xlrec->flags & GIN_INSERT_ISLEAF))
|
|
|
|
{
|
2016-12-05 12:29:41 +01:00
|
|
|
char *payload = rec + sizeof(ginxlogInsert);
|
2013-11-27 18:21:23 +01:00
|
|
|
BlockNumber leftChildBlkno;
|
|
|
|
BlockNumber rightChildBlkno;
|
|
|
|
|
2013-11-28 20:54:49 +01:00
|
|
|
leftChildBlkno = BlockIdGetBlockNumber((BlockId) payload);
|
|
|
|
payload += sizeof(BlockIdData);
|
|
|
|
rightChildBlkno = BlockIdGetBlockNumber((BlockId) payload);
|
2013-11-27 18:21:23 +01:00
|
|
|
payload += sizeof(BlockNumber);
|
|
|
|
appendStringInfo(buf, " children: %u/%u",
|
|
|
|
leftChildBlkno, rightChildBlkno);
|
|
|
|
}
|
2016-12-05 12:29:41 +01:00
|
|
|
if (XLogRecHasBlockImage(record, 0))
|
2017-02-08 21:45:30 +01:00
|
|
|
{
|
|
|
|
if (XLogRecBlockImageApply(record, 0))
|
|
|
|
appendStringInfoString(buf, " (full page image)");
|
|
|
|
else
|
|
|
|
appendStringInfoString(buf, " (full page image, for WAL verification)");
|
|
|
|
}
|
2016-12-05 12:29:41 +01:00
|
|
|
else
|
2014-01-22 17:51:48 +01:00
|
|
|
{
|
2016-12-05 12:29:41 +01:00
|
|
|
char *payload = XLogRecGetBlockData(record, 0, NULL);
|
2014-01-22 17:51:48 +01:00
|
|
|
|
2016-12-05 12:29:41 +01:00
|
|
|
if (!(xlrec->flags & GIN_INSERT_ISDATA))
|
|
|
|
appendStringInfo(buf, " isdelete: %c",
|
|
|
|
(((ginxlogInsertEntry *) payload)->isDelete) ? 'T' : 'F');
|
|
|
|
else if (xlrec->flags & GIN_INSERT_ISLEAF)
|
|
|
|
desc_recompress_leaf(buf, (ginxlogRecompressDataLeaf *) payload);
|
2014-03-31 14:15:19 +02:00
|
|
|
else
|
2016-12-05 12:29:41 +01:00
|
|
|
{
|
|
|
|
ginxlogInsertDataInternal *insertData =
|
2023-05-19 23:24:48 +02:00
|
|
|
(ginxlogInsertDataInternal *) payload;
|
2014-05-06 18:12:18 +02:00
|
|
|
|
2016-12-05 12:29:41 +01:00
|
|
|
appendStringInfo(buf, " pitem: %u-%u/%u",
|
|
|
|
PostingItemGetBlockNumber(&insertData->newitem),
|
|
|
|
ItemPointerGetBlockNumber(&insertData->newitem.key),
|
|
|
|
ItemPointerGetOffsetNumber(&insertData->newitem.key));
|
|
|
|
}
|
2014-01-22 17:51:48 +01:00
|
|
|
}
|
2013-11-27 18:21:23 +01:00
|
|
|
}
|
2012-11-28 16:35:01 +01:00
|
|
|
break;
|
|
|
|
case XLOG_GIN_SPLIT:
|
2014-01-22 17:51:48 +01:00
|
|
|
{
|
|
|
|
ginxlogSplit *xlrec = (ginxlogSplit *) rec;
|
|
|
|
|
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.
2014-11-20 16:56:26 +01:00
|
|
|
appendStringInfo(buf, "isrootsplit: %c",
|
|
|
|
(((ginxlogSplit *) rec)->flags & GIN_SPLIT_ROOT) ? 'T' : 'F');
|
2014-01-22 17:51:48 +01:00
|
|
|
appendStringInfo(buf, " isdata: %c isleaf: %c",
|
|
|
|
(xlrec->flags & GIN_INSERT_ISDATA) ? 'T' : 'F',
|
|
|
|
(xlrec->flags & GIN_INSERT_ISLEAF) ? 'T' : 'F');
|
|
|
|
}
|
2012-11-28 16:35:01 +01:00
|
|
|
break;
|
|
|
|
case XLOG_GIN_VACUUM_PAGE:
|
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.
2014-11-20 16:56:26 +01:00
|
|
|
/* no further information */
|
2012-11-28 16:35:01 +01:00
|
|
|
break;
|
2014-01-22 17:51:48 +01:00
|
|
|
case XLOG_GIN_VACUUM_DATA_LEAF_PAGE:
|
|
|
|
{
|
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.
2014-11-20 16:56:26 +01:00
|
|
|
if (XLogRecHasBlockImage(record, 0))
|
2017-02-08 21:45:30 +01:00
|
|
|
{
|
|
|
|
if (XLogRecBlockImageApply(record, 0))
|
|
|
|
appendStringInfoString(buf, " (full page image)");
|
|
|
|
else
|
|
|
|
appendStringInfoString(buf, " (full page image, for WAL verification)");
|
|
|
|
}
|
2014-03-31 14:15:19 +02:00
|
|
|
else
|
2016-12-05 12:29:41 +01:00
|
|
|
{
|
|
|
|
ginxlogVacuumDataLeafPage *xlrec =
|
2023-05-19 23:24:48 +02:00
|
|
|
(ginxlogVacuumDataLeafPage *) XLogRecGetBlockData(record, 0, NULL);
|
2016-12-05 12:29:41 +01:00
|
|
|
|
2014-03-31 14:15:19 +02:00
|
|
|
desc_recompress_leaf(buf, &xlrec->data);
|
2016-12-05 12:29:41 +01:00
|
|
|
}
|
2014-01-22 17:51:48 +01:00
|
|
|
}
|
|
|
|
break;
|
2012-11-28 16:35:01 +01:00
|
|
|
case XLOG_GIN_DELETE_PAGE:
|
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.
2014-11-20 16:56:26 +01:00
|
|
|
/* no further information */
|
2012-11-28 16:35:01 +01:00
|
|
|
break;
|
|
|
|
case XLOG_GIN_UPDATE_META_PAGE:
|
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.
2014-11-20 16:56:26 +01:00
|
|
|
/* no further information */
|
2012-11-28 16:35:01 +01:00
|
|
|
break;
|
|
|
|
case XLOG_GIN_INSERT_LISTPAGE:
|
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.
2014-11-20 16:56:26 +01:00
|
|
|
/* no further information */
|
2012-11-28 16:35:01 +01:00
|
|
|
break;
|
|
|
|
case XLOG_GIN_DELETE_LISTPAGE:
|
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.
2014-11-20 16:56:26 +01:00
|
|
|
appendStringInfo(buf, "ndeleted: %d",
|
|
|
|
((ginxlogDeleteListPages *) rec)->ndeleted);
|
2012-11-28 16:35:01 +01:00
|
|
|
break;
|
2014-09-19 15:17:12 +02:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
const char *
|
|
|
|
gin_identify(uint8 info)
|
|
|
|
{
|
|
|
|
const char *id = NULL;
|
|
|
|
|
2014-09-22 16:48:14 +02:00
|
|
|
switch (info & ~XLR_INFO_MASK)
|
2014-09-19 15:17:12 +02:00
|
|
|
{
|
|
|
|
case XLOG_GIN_CREATE_PTREE:
|
|
|
|
id = "CREATE_PTREE";
|
|
|
|
break;
|
|
|
|
case XLOG_GIN_INSERT:
|
|
|
|
id = "INSERT";
|
|
|
|
break;
|
|
|
|
case XLOG_GIN_SPLIT:
|
|
|
|
id = "SPLIT";
|
|
|
|
break;
|
|
|
|
case XLOG_GIN_VACUUM_PAGE:
|
|
|
|
id = "VACUUM_PAGE";
|
|
|
|
break;
|
|
|
|
case XLOG_GIN_VACUUM_DATA_LEAF_PAGE:
|
|
|
|
id = "VACUUM_DATA_LEAF_PAGE";
|
|
|
|
break;
|
|
|
|
case XLOG_GIN_DELETE_PAGE:
|
|
|
|
id = "DELETE_PAGE";
|
|
|
|
break;
|
|
|
|
case XLOG_GIN_UPDATE_META_PAGE:
|
|
|
|
id = "UPDATE_META_PAGE";
|
|
|
|
break;
|
|
|
|
case XLOG_GIN_INSERT_LISTPAGE:
|
|
|
|
id = "INSERT_LISTPAGE";
|
|
|
|
break;
|
|
|
|
case XLOG_GIN_DELETE_LISTPAGE:
|
|
|
|
id = "DELETE_LISTPAGE";
|
2012-11-28 16:35:01 +01:00
|
|
|
break;
|
|
|
|
}
|
2014-09-19 15:17:12 +02:00
|
|
|
|
|
|
|
return id;
|
2012-11-28 16:35:01 +01:00
|
|
|
}
|