rdelaage
/
postgresql
mirror of https://git.postgresql.org/git/postgresql.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity
postgresql/src/include/access/tableam.h

Ignoring revisions in .git-blame-ignore-revs. Click here to bypass and see the normal blame view.

2111 lines
76 KiB
C
Raw Normal View History

tableam: introduce table AM infrastructure. This introduces the concept of table access methods, i.e. CREATE ACCESS METHOD ... TYPE TABLE and CREATE TABLE ... USING (storage-engine). No table access functionality is delegated to table AMs as of this commit, that'll be done in following commits. Subsequent commits will incrementally abstract table access functionality to be routed through table access methods. That change is too large to be reviewed & committed at once, so it'll be done incrementally. Docs will be updated at the end, as adding them incrementally would likely make them less coherent, and definitely is a lot more work, without a lot of benefit. Table access methods are specified similar to index access methods, i.e. pg_am.amhandler returns, as INTERNAL, a pointer to a struct with callbacks. In contrast to index AMs that struct needs to live as long as a backend, typically that's achieved by just returning a pointer to a constant struct. Psql's \d+ now displays a table's access method. That can be disabled with HIDE_TABLEAM=true, which is mainly useful so regression tests can be run against different AMs. It's quite possible that this behaviour still needs to be fine tuned. For now it's not allowed to set a table AM for a partitioned table, as we've not resolved how partitions would inherit that. Disallowing allows us to introduce, if we decide that's the way forward, such a behaviour without a compatibility break. Catversion bumped, to add the heap table AM and references to it. Author: Haribabu Kommi, Andres Freund, Alvaro Herrera, Dimitri Golgov and others Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql https://postgr.es/m/20190107235616.6lur25ph22u5u5av@alap3.anarazel.de https://postgr.es/m/20190304234700.w5tmhducs5wxgzls@alap3.anarazel.de
2019-03-06 18:54:38 +01:00
/*-------------------------------------------------------------------------
*
* tableam.h
* POSTGRES table access method definitions.
*
*
Update copyright for 2023 Backpatch-through: 11
2023-01-02 21:00:37 +01:00
* Portions Copyright (c) 1996-2023, PostgreSQL Global Development Group
tableam: introduce table AM infrastructure. This introduces the concept of table access methods, i.e. CREATE ACCESS METHOD ... TYPE TABLE and CREATE TABLE ... USING (storage-engine). No table access functionality is delegated to table AMs as of this commit, that'll be done in following commits. Subsequent commits will incrementally abstract table access functionality to be routed through table access methods. That change is too large to be reviewed & committed at once, so it'll be done incrementally. Docs will be updated at the end, as adding them incrementally would likely make them less coherent, and definitely is a lot more work, without a lot of benefit. Table access methods are specified similar to index access methods, i.e. pg_am.amhandler returns, as INTERNAL, a pointer to a struct with callbacks. In contrast to index AMs that struct needs to live as long as a backend, typically that's achieved by just returning a pointer to a constant struct. Psql's \d+ now displays a table's access method. That can be disabled with HIDE_TABLEAM=true, which is mainly useful so regression tests can be run against different AMs. It's quite possible that this behaviour still needs to be fine tuned. For now it's not allowed to set a table AM for a partitioned table, as we've not resolved how partitions would inherit that. Disallowing allows us to introduce, if we decide that's the way forward, such a behaviour without a compatibility break. Catversion bumped, to add the heap table AM and references to it. Author: Haribabu Kommi, Andres Freund, Alvaro Herrera, Dimitri Golgov and others Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql https://postgr.es/m/20190107235616.6lur25ph22u5u5av@alap3.anarazel.de https://postgr.es/m/20190304234700.w5tmhducs5wxgzls@alap3.anarazel.de
2019-03-06 18:54:38 +01:00
* Portions Copyright (c) 1994, Regents of the University of California
*
* src/include/access/tableam.h
*
tableam: basic documentation. This adds documentation about the user oriented parts of table access methods (i.e. the default_table_access_method GUC and the USING clause for CREATE TABLE etc), adds a basic chapter about the table access method interface, and adds a note to storage.sgml that it's contents don't necessarily apply for non-builtin AMs. Author: Haribabu Kommi and Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-04-04 02:37:00 +02:00
* NOTES
* See tableam.sgml for higher level documentation.
*
tableam: introduce table AM infrastructure. This introduces the concept of table access methods, i.e. CREATE ACCESS METHOD ... TYPE TABLE and CREATE TABLE ... USING (storage-engine). No table access functionality is delegated to table AMs as of this commit, that'll be done in following commits. Subsequent commits will incrementally abstract table access functionality to be routed through table access methods. That change is too large to be reviewed & committed at once, so it'll be done incrementally. Docs will be updated at the end, as adding them incrementally would likely make them less coherent, and definitely is a lot more work, without a lot of benefit. Table access methods are specified similar to index access methods, i.e. pg_am.amhandler returns, as INTERNAL, a pointer to a struct with callbacks. In contrast to index AMs that struct needs to live as long as a backend, typically that's achieved by just returning a pointer to a constant struct. Psql's \d+ now displays a table's access method. That can be disabled with HIDE_TABLEAM=true, which is mainly useful so regression tests can be run against different AMs. It's quite possible that this behaviour still needs to be fine tuned. For now it's not allowed to set a table AM for a partitioned table, as we've not resolved how partitions would inherit that. Disallowing allows us to introduce, if we decide that's the way forward, such a behaviour without a compatibility break. Catversion bumped, to add the heap table AM and references to it. Author: Haribabu Kommi, Andres Freund, Alvaro Herrera, Dimitri Golgov and others Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql https://postgr.es/m/20190107235616.6lur25ph22u5u5av@alap3.anarazel.de https://postgr.es/m/20190304234700.w5tmhducs5wxgzls@alap3.anarazel.de
2019-03-06 18:54:38 +01:00
*-------------------------------------------------------------------------
*/
#ifndef TABLEAM_H
#define TABLEAM_H
tableam: Add and use scan APIs. Too allow table accesses to be not directly dependent on heap, several new abstractions are needed. Specifically: 1) Heap scans need to be generalized into table scans. Do this by introducing TableScanDesc, which will be the "base class" for individual AMs. This contains the AM independent fields from HeapScanDesc. The previous heap_{beginscan,rescan,endscan} et al. have been replaced with a table_ version. There's no direct replacement for heap_getnext(), as that returned a HeapTuple, which is undesirable for a other AMs. Instead there's table_scan_getnextslot(). But note that heap_getnext() lives on, it's still used widely to access catalog tables. This is achieved by new scan_begin, scan_end, scan_rescan, scan_getnextslot callbacks. 2) The portion of parallel scans that's shared between backends need to be able to do so without the user doing per-AM work. To achieve that new parallelscan_{estimate, initialize, reinitialize} callbacks are introduced, which operate on a new ParallelTableScanDesc, which again can be subclassed by AMs. As it is likely that several AMs are going to be block oriented, block oriented callbacks that can be shared between such AMs are provided and used by heap. table_block_parallelscan_{estimate, intiialize, reinitialize} as callbacks, and table_block_parallelscan_{nextpage, init} for use in AMs. These operate on a ParallelBlockTableScanDesc. 3) Index scans need to be able to access tables to return a tuple, and there needs to be state across individual accesses to the heap to store state like buffers. That's now handled by introducing a sort-of-scan IndexFetchTable, which again is intended to be subclassed by individual AMs (for heap IndexFetchHeap). The relevant callbacks for an AM are index_fetch_{end, begin, reset} to create the necessary state, and index_fetch_tuple to retrieve an indexed tuple. Note that index_fetch_tuple implementations need to be smarter than just blindly fetching the tuples for AMs that have optimizations similar to heap's HOT - the currently alive tuple in the update chain needs to be fetched if appropriate. Similar to table_scan_getnextslot(), it's undesirable to continue to return HeapTuples. Thus index_fetch_heap (might want to rename that later) now accepts a slot as an argument. Core code doesn't have a lot of call sites performing index scans without going through the systable_* API (in contrast to loads of heap_getnext calls and working directly with HeapTuples). Index scans now store the result of a search in IndexScanDesc->xs_heaptid, rather than xs_ctup->t_self. As the target is not generally a HeapTuple anymore that seems cleaner. To be able to sensible adapt code to use the above, two further callbacks have been introduced: a) slot_callbacks returns a TupleTableSlotOps* suitable for creating slots capable of holding a tuple of the AMs type. table_slot_callbacks() and table_slot_create() are based upon that, but have additional logic to deal with views, foreign tables, etc. While this change could have been done separately, nearly all the call sites that needed to be adapted for the rest of this commit also would have been needed to be adapted for table_slot_callbacks(), making separation not worthwhile. b) tuple_satisfies_snapshot checks whether the tuple in a slot is currently visible according to a snapshot. That's required as a few places now don't have a buffer + HeapTuple around, but a slot (which in heap's case internally has that information). Additionally a few infrastructure changes were needed: I) SysScanDesc, as used by systable_{beginscan, getnext} et al. now internally uses a slot to keep track of tuples. While systable_getnext() still returns HeapTuples, and will so for the foreseeable future, the index API (see 1) above) now only deals with slots. The remainder, and largest part, of this commit is then adjusting all scans in postgres to use the new APIs. Author: Andres Freund, Haribabu Kommi, Alvaro Herrera Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-11 20:46:41 +01:00
#include "access/relscan.h"
#include "access/sdir.h"
Implement streaming mode in ReorderBuffer. Instead of serializing the transaction to disk after reaching the logical_decoding_work_mem limit in memory, we consume the changes we have in memory and invoke stream API methods added by commit 45fdc9738b. However, sometimes if we have incomplete toast or speculative insert we spill to the disk because we can't generate the complete tuple and stream. And, as soon as we get the complete tuple we stream the transaction including the serialized changes. We can do this incremental processing thanks to having assignments (associating subxact with toplevel xacts) in WAL right away, and thanks to logging the invalidation messages at each command end. These features are added by commits 0bead9af48 and c55040ccd0 respectively. Now that we can stream in-progress transactions, the concurrent aborts may cause failures when the output plugin consults catalogs (both system and user-defined). We handle such failures by returning ERRCODE_TRANSACTION_ROLLBACK sqlerrcode from system table scan APIs to the backend or WALSender decoding a specific uncommitted transaction. The decoding logic on the receipt of such a sqlerrcode aborts the decoding of the current transaction and continue with the decoding of other transactions. We have ReorderBufferTXN pointer in each ReorderBufferChange by which we know which xact it belongs to. The output plugin can use this to decide which changes to discard in case of stream_abort_cb (e.g. when a subxact gets discarded). We also provide a new option via SQL APIs to fetch the changes being streamed. Author: Dilip Kumar, Tomas Vondra, Amit Kapila, Nikhil Sontakke Reviewed-by: Amit Kapila, Kuntal Ghosh, Ajin Cherian Tested-by: Neha Sharma, Mahendra Singh Thalor and Ajin Cherian Discussion: https://postgr.es/m/688b0b7f-2f6c-d827-c27b-216a8e3ea700@2ndquadrant.com
2020-08-08 04:04:39 +02:00
#include "access/xact.h"
Split up guc.c for better build speed and ease of maintenance. guc.c has grown to be one of our largest .c files, making it a bottleneck for compilation. It's also acquired a bunch of knowledge that'd be better kept elsewhere, because of our not very good habit of putting variable-specific check hooks here. Hence, split it up along these lines: * guc.c itself retains just the core GUC housekeeping mechanisms. * New file guc_funcs.c contains the SET/SHOW interfaces and some SQL-accessible functions for GUC manipulation. * New file guc_tables.c contains the data arrays that define the built-in GUC variables, along with some already-exported constant tables. * GUC check/assign/show hook functions are moved to the variable's home module, whenever that's clearly identifiable. A few hard- to-classify hooks ended up in commands/variable.c, which was already a home for miscellaneous GUC hook functions. To avoid cluttering a lot more header files with #include "guc.h", I also invented a new header file utils/guc_hooks.h and put all the GUC hook functions' declarations there, regardless of their originating module. That allowed removal of #include "guc.h" from some existing headers. The fallout from that (hopefully all caught here) demonstrates clearly why such inclusions are best minimized: there are a lot of files that, for example, were getting array.h at two or more levels of remove, despite not having any connection at all to GUCs in themselves. There is some very minor code beautification here, such as renaming a couple of inconsistently-named hook functions and improving some comments. But mostly this just moves code from point A to point B and deals with the ensuing needs for #include adjustments and exporting a few functions that previously weren't exported. Patch by me, per a suggestion from Andres Freund; thanks also to Michael Paquier for the idea to invent guc_funcs.c. Discussion: https://postgr.es/m/587607.1662836699@sss.pgh.pa.us
2022-09-13 17:05:07 +02:00
#include "executor/tuptable.h"
tableam: Add and use scan APIs. Too allow table accesses to be not directly dependent on heap, several new abstractions are needed. Specifically: 1) Heap scans need to be generalized into table scans. Do this by introducing TableScanDesc, which will be the "base class" for individual AMs. This contains the AM independent fields from HeapScanDesc. The previous heap_{beginscan,rescan,endscan} et al. have been replaced with a table_ version. There's no direct replacement for heap_getnext(), as that returned a HeapTuple, which is undesirable for a other AMs. Instead there's table_scan_getnextslot(). But note that heap_getnext() lives on, it's still used widely to access catalog tables. This is achieved by new scan_begin, scan_end, scan_rescan, scan_getnextslot callbacks. 2) The portion of parallel scans that's shared between backends need to be able to do so without the user doing per-AM work. To achieve that new parallelscan_{estimate, initialize, reinitialize} callbacks are introduced, which operate on a new ParallelTableScanDesc, which again can be subclassed by AMs. As it is likely that several AMs are going to be block oriented, block oriented callbacks that can be shared between such AMs are provided and used by heap. table_block_parallelscan_{estimate, intiialize, reinitialize} as callbacks, and table_block_parallelscan_{nextpage, init} for use in AMs. These operate on a ParallelBlockTableScanDesc. 3) Index scans need to be able to access tables to return a tuple, and there needs to be state across individual accesses to the heap to store state like buffers. That's now handled by introducing a sort-of-scan IndexFetchTable, which again is intended to be subclassed by individual AMs (for heap IndexFetchHeap). The relevant callbacks for an AM are index_fetch_{end, begin, reset} to create the necessary state, and index_fetch_tuple to retrieve an indexed tuple. Note that index_fetch_tuple implementations need to be smarter than just blindly fetching the tuples for AMs that have optimizations similar to heap's HOT - the currently alive tuple in the update chain needs to be fetched if appropriate. Similar to table_scan_getnextslot(), it's undesirable to continue to return HeapTuples. Thus index_fetch_heap (might want to rename that later) now accepts a slot as an argument. Core code doesn't have a lot of call sites performing index scans without going through the systable_* API (in contrast to loads of heap_getnext calls and working directly with HeapTuples). Index scans now store the result of a search in IndexScanDesc->xs_heaptid, rather than xs_ctup->t_self. As the target is not generally a HeapTuple anymore that seems cleaner. To be able to sensible adapt code to use the above, two further callbacks have been introduced: a) slot_callbacks returns a TupleTableSlotOps* suitable for creating slots capable of holding a tuple of the AMs type. table_slot_callbacks() and table_slot_create() are based upon that, but have additional logic to deal with views, foreign tables, etc. While this change could have been done separately, nearly all the call sites that needed to be adapted for the rest of this commit also would have been needed to be adapted for table_slot_callbacks(), making separation not worthwhile. b) tuple_satisfies_snapshot checks whether the tuple in a slot is currently visible according to a snapshot. That's required as a few places now don't have a buffer + HeapTuple around, but a slot (which in heap's case internally has that information). Additionally a few infrastructure changes were needed: I) SysScanDesc, as used by systable_{beginscan, getnext} et al. now internally uses a slot to keep track of tuples. While systable_getnext() still returns HeapTuples, and will so for the foreseeable future, the index API (see 1) above) now only deals with slots. The remainder, and largest part, of this commit is then adjusting all scans in postgres to use the new APIs. Author: Andres Freund, Haribabu Kommi, Alvaro Herrera Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-11 20:46:41 +01:00
#include "utils/rel.h"
#include "utils/snapshot.h"
tableam: introduce table AM infrastructure. This introduces the concept of table access methods, i.e. CREATE ACCESS METHOD ... TYPE TABLE and CREATE TABLE ... USING (storage-engine). No table access functionality is delegated to table AMs as of this commit, that'll be done in following commits. Subsequent commits will incrementally abstract table access functionality to be routed through table access methods. That change is too large to be reviewed & committed at once, so it'll be done incrementally. Docs will be updated at the end, as adding them incrementally would likely make them less coherent, and definitely is a lot more work, without a lot of benefit. Table access methods are specified similar to index access methods, i.e. pg_am.amhandler returns, as INTERNAL, a pointer to a struct with callbacks. In contrast to index AMs that struct needs to live as long as a backend, typically that's achieved by just returning a pointer to a constant struct. Psql's \d+ now displays a table's access method. That can be disabled with HIDE_TABLEAM=true, which is mainly useful so regression tests can be run against different AMs. It's quite possible that this behaviour still needs to be fine tuned. For now it's not allowed to set a table AM for a partitioned table, as we've not resolved how partitions would inherit that. Disallowing allows us to introduce, if we decide that's the way forward, such a behaviour without a compatibility break. Catversion bumped, to add the heap table AM and references to it. Author: Haribabu Kommi, Andres Freund, Alvaro Herrera, Dimitri Golgov and others Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql https://postgr.es/m/20190107235616.6lur25ph22u5u5av@alap3.anarazel.de https://postgr.es/m/20190304234700.w5tmhducs5wxgzls@alap3.anarazel.de
2019-03-06 18:54:38 +01:00
#define DEFAULT_TABLE_ACCESS_METHOD "heap"
tableam: comment and formatting fixes. Author: Heikki Linnakangas Discussion: https://postgr.es/m/9a7fb9cc-2419-5db7-8840-ddc10c93f122@iki.fi
2019-04-08 19:25:16 +02:00
/* GUCs */
Apply PGDLLIMPORT markings broadly. Up until now, we've had a policy of only marking certain variables in the PostgreSQL header files with PGDLLIMPORT, but now we've decided to mark them all. This means that extensions running on Windows should no longer operate at a disadvantage as compared to extensions running on Linux: if the variable is present in a header file, it should be accessible. Discussion: http://postgr.es/m/CA+TgmoYanc1_FSfimhgiWSqVyP5KKmh5NP2BWNwDhO8Pg2vGYQ@mail.gmail.com
2022-04-08 14:16:38 +02:00
extern PGDLLIMPORT char *default_table_access_method;
extern PGDLLIMPORT bool synchronize_seqscans;
tableam: introduce table AM infrastructure. This introduces the concept of table access methods, i.e. CREATE ACCESS METHOD ... TYPE TABLE and CREATE TABLE ... USING (storage-engine). No table access functionality is delegated to table AMs as of this commit, that'll be done in following commits. Subsequent commits will incrementally abstract table access functionality to be routed through table access methods. That change is too large to be reviewed & committed at once, so it'll be done incrementally. Docs will be updated at the end, as adding them incrementally would likely make them less coherent, and definitely is a lot more work, without a lot of benefit. Table access methods are specified similar to index access methods, i.e. pg_am.amhandler returns, as INTERNAL, a pointer to a struct with callbacks. In contrast to index AMs that struct needs to live as long as a backend, typically that's achieved by just returning a pointer to a constant struct. Psql's \d+ now displays a table's access method. That can be disabled with HIDE_TABLEAM=true, which is mainly useful so regression tests can be run against different AMs. It's quite possible that this behaviour still needs to be fine tuned. For now it's not allowed to set a table AM for a partitioned table, as we've not resolved how partitions would inherit that. Disallowing allows us to introduce, if we decide that's the way forward, such a behaviour without a compatibility break. Catversion bumped, to add the heap table AM and references to it. Author: Haribabu Kommi, Andres Freund, Alvaro Herrera, Dimitri Golgov and others Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql https://postgr.es/m/20190107235616.6lur25ph22u5u5av@alap3.anarazel.de https://postgr.es/m/20190304234700.w5tmhducs5wxgzls@alap3.anarazel.de
2019-03-06 18:54:38 +01:00
tableam: Add tuple_{insert, delete, update, lock} and use. This adds new, required, table AM callbacks for insert/delete/update and lock_tuple. To be able to reasonably use those, the EvalPlanQual mechanism had to be adapted, moving more logic into the AM. Previously both delete/update/lock call-sites and the EPQ mechanism had to have awareness of the specific tuple format to be able to fetch the latest version of a tuple. Obviously that needs to be abstracted away. To do so, move the logic that find the latest row version into the AM. lock_tuple has a new flag argument, TUPLE_LOCK_FLAG_FIND_LAST_VERSION, that forces it to lock the last version, rather than the current one. It'd have been possible to do so via a separate callback as well, but finding the last version usually also necessitates locking the newest version, making it sensible to combine the two. This replaces the previous use of EvalPlanQualFetch(). Additionally HeapTupleUpdated, which previously signaled either a concurrent update or delete, is now split into two, to avoid callers needing AM specific knowledge to differentiate. The move of finding the latest row version into tuple_lock means that encountering a row concurrently moved into another partition will now raise an error about "tuple to be locked" rather than "tuple to be updated/deleted" - which is accurate, as that always happens when locking rows. While possible slightly less helpful for users, it seems like an acceptable trade-off. As part of this commit HTSU_Result has been renamed to TM_Result, and its members been expanded to differentiated between updating and deleting. HeapUpdateFailureData has been renamed to TM_FailureData. The interface to speculative insertion is changed so nodeModifyTable.c does not have to set the speculative token itself anymore. Instead there's a version of tuple_insert, tuple_insert_speculative, that performs the speculative insertion (without requiring a flag to signal that fact), and the speculative insertion is either made permanent with table_complete_speculative(succeeded = true) or aborted with succeeded = false). Note that multi_insert is not yet routed through tableam, nor is COPY. Changing multi_insert requires changes to copy.c that are large enough to better be done separately. Similarly, although simpler, CREATE TABLE AS and CREATE MATERIALIZED VIEW are also only going to be adjusted in a later commit. Author: Andres Freund and Haribabu Kommi Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20190313003903.nwvrxi7rw3ywhdel@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-24 03:55:57 +01:00
struct BulkInsertStateData;
tableam: Support for an index build's initial table scan(s). To support building indexes over tables of different AMs, the scans to do so need to be routed through the table AM. While moving a fair amount of code, nearly all the changes are just moving code to below a callback. Currently the range based interface wouldn't make much sense for non block based table AMs. But that seems aceptable for now. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-28 03:59:06 +01:00
struct IndexInfo;
tableam: sample scan. This moves sample scan support to below tableam. It's not optional as there is, in contrast to e.g. bitmap heap scans, no alternative way to perform tablesample queries. If an AM can't deal with the block based API, it will have to throw an ERROR. The tableam callbacks for this are block based, but given the current TsmRoutine interface, that seems to be required. The new interface doesn't require TsmRoutines to perform visibility checks anymore - that requires the TsmRoutine to know details about the AM, which we want to avoid. To continue to allow taking the returned number of tuples account SampleScanState now has a donetuples field (which previously e.g. existed in SystemRowsSamplerData), which is only incremented after the visibility check succeeds. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-31 05:18:53 +02:00
struct SampleScanState;
tableam: bitmap table scan. This moves bitmap heap scan support to below an optional tableam callback. It's optional as the whole concept of bitmap heapscans is fairly block specific. This basically moves the work previously done in bitgetpage() into the new scan_bitmap_next_block callback, and the direct poking into the buffer done in BitmapHeapNext() into the new scan_bitmap_next_tuple() callback. The abstraction is currently somewhat leaky because nodeBitmapHeapscan.c's prefetching and visibilitymap based logic remains - it's likely that we'll later have to move more into the AM. But it's not trivial to do so without introducing a significant amount of code duplication between the AMs, so that's a project for later. Note that now nodeBitmapHeapscan.c and the associated node types are a bit misnamed. But it's not clear whether renaming wouldn't be a cure worse than the disease. Either way, that'd be best done in a separate commit. Author: Andres Freund Reviewed-By: Robert Haas (in an older version) Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-04-01 02:51:49 +02:00
struct TBMIterateResult;
tableam: VACUUM and ANALYZE support. This is a relatively straightforward move of the current implementation to sit below tableam. As the current analyze sampling implementation is pretty inherently block based, the tableam analyze interface is as well. It might make sense to generalize that at some point, but that seems like a larger project that shouldn't be undertaken at the same time as the introduction of tableam. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-31 00:21:09 +01:00
struct VacuumParams;
tableam: Support for an index build's initial table scan(s). To support building indexes over tables of different AMs, the scans to do so need to be routed through the table AM. While moving a fair amount of code, nearly all the changes are just moving code to below a callback. Currently the range based interface wouldn't make much sense for non block based table AMs. But that seems aceptable for now. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-28 03:59:06 +01:00
struct ValidateIndexState;
tableam: Add tuple_{insert, delete, update, lock} and use. This adds new, required, table AM callbacks for insert/delete/update and lock_tuple. To be able to reasonably use those, the EvalPlanQual mechanism had to be adapted, moving more logic into the AM. Previously both delete/update/lock call-sites and the EPQ mechanism had to have awareness of the specific tuple format to be able to fetch the latest version of a tuple. Obviously that needs to be abstracted away. To do so, move the logic that find the latest row version into the AM. lock_tuple has a new flag argument, TUPLE_LOCK_FLAG_FIND_LAST_VERSION, that forces it to lock the last version, rather than the current one. It'd have been possible to do so via a separate callback as well, but finding the last version usually also necessitates locking the newest version, making it sensible to combine the two. This replaces the previous use of EvalPlanQualFetch(). Additionally HeapTupleUpdated, which previously signaled either a concurrent update or delete, is now split into two, to avoid callers needing AM specific knowledge to differentiate. The move of finding the latest row version into tuple_lock means that encountering a row concurrently moved into another partition will now raise an error about "tuple to be locked" rather than "tuple to be updated/deleted" - which is accurate, as that always happens when locking rows. While possible slightly less helpful for users, it seems like an acceptable trade-off. As part of this commit HTSU_Result has been renamed to TM_Result, and its members been expanded to differentiated between updating and deleting. HeapUpdateFailureData has been renamed to TM_FailureData. The interface to speculative insertion is changed so nodeModifyTable.c does not have to set the speculative token itself anymore. Instead there's a version of tuple_insert, tuple_insert_speculative, that performs the speculative insertion (without requiring a flag to signal that fact), and the speculative insertion is either made permanent with table_complete_speculative(succeeded = true) or aborted with succeeded = false). Note that multi_insert is not yet routed through tableam, nor is COPY. Changing multi_insert requires changes to copy.c that are large enough to better be done separately. Similarly, although simpler, CREATE TABLE AS and CREATE MATERIALIZED VIEW are also only going to be adjusted in a later commit. Author: Andres Freund and Haribabu Kommi Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20190313003903.nwvrxi7rw3ywhdel@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-24 03:55:57 +01:00
Don't to predicate lock for analyze scans, refactor scan option passing. Before this commit, when ANALYZE was run on a table and serializable was used (either by virtue of an explicit BEGIN TRANSACTION ISOLATION LEVEL SERIALIZABLE, or default_transaction_isolation being set to serializable) a null pointer dereference lead to a crash. The analyze scan doesn't need a snapshot (nor predicate locking), but before this commit a scan only contained information about being a bitmap or sample scan. Refactor the option passing to the scan_begin callback to use a bitmask instead. Alternatively we could have added a new boolean parameter, but that seems harder to read. Even before this issue various people (Heikki, Tom, Robert) suggested doing so. These changes don't change the scan APIs outside of tableam. The flags argument could be exposed, it's not necessary to fix this problem. Also the wrapper table_beginscan* functions encapsulate most of that complexity. After these changes fixing the bug is trivial, just don't acquire predicate lock for analyze style scans. That was already done for bitmap heap scans. Add an assert that a snapshot is passed when acquiring the predicate lock, so this kind of bug doesn't require running with serializable. Also add a comment about sample scans currently requiring predicate locking the entire relation, that previously wasn't remarked upon. Reported-By: Joe Wildish Author: Andres Freund Discussion: https://postgr.es/m/4EA80A20-E9BF-49F1-9F01-5B66CAB21453@elusive.cx https://postgr.es/m/20190411164947.nkii4gaeilt4bui7@alap3.anarazel.de https://postgr.es/m/20190518203102.g7peu2fianukjuxm@alap3.anarazel.de
2019-05-20 00:10:28 +02:00
/*
* Bitmask values for the flags argument to the scan_begin callback.
*/
typedef enum ScanOptions
{
/* one of SO_TYPE_* may be specified */
SO_TYPE_SEQSCAN = 1 << 0,
SO_TYPE_BITMAPSCAN = 1 << 1,
SO_TYPE_SAMPLESCAN = 1 << 2,
Fix bug in Tid scan. Commit 147e3722f7 changed Tid scan so that it calls table_beginscan() and uses the scan option for seq scan. This change caused two issues. (1) The change caused Tid scan to take a predicate lock on the entire relation in serializable transaction even when relation-level lock is not necessary. This could lead to an unexpected serialization error. (2) The change caused Tid scan to increment the number of seq_scan in pg_stat_*_tables views even though it's not seq scan. This could confuse the users. This commit adds the scan option for Tid scan and makes Tid scan use it, to avoid those issues. Back-patch to v12, where the bug was introduced. Author: Tatsuhito Kasahara Reviewed-by: Kyotaro Horiguchi, Masahiko Sawada, Fujii Masao Discussion: https://postgr.es/m/CAP0=ZVKy+gTbFmB6X_UW0pP3WaeJ-fkUWHoD-pExS=at3CY76g@mail.gmail.com
2020-02-07 14:00:21 +01:00
SO_TYPE_TIDSCAN = 1 << 3,
Add TID Range Scans to support efficient scanning ranges of TIDs This adds a new executor node named TID Range Scan. The query planner will generate paths for TID Range scans when quals are discovered on base relations which search for ranges on the table's ctid column. These ranges may be open at either end. For example, WHERE ctid >= '(10,0)'; will return all tuples on page 10 and over. To support this, two new optional callback functions have been added to table AM. scan_set_tidrange is used to set the scan range to just the given range of TIDs. scan_getnextslot_tidrange fetches the next tuple in the given range. For AMs were scanning ranges of TIDs would not make sense, these functions can be set to NULL in the TableAmRoutine. The query planner won't generate TID Range Scan Paths in that case. Author: Edmund Horner, David Rowley Reviewed-by: David Rowley, Tomas Vondra, Tom Lane, Andres Freund, Zhihong Yu Discussion: https://postgr.es/m/CAMyN-kB-nFTkF=VA_JPwFNo08S0d-Yk0F741S2B7LDmYAi8eyA@mail.gmail.com
2021-02-27 10:59:36 +01:00
SO_TYPE_TIDRANGESCAN = 1 << 4,
SO_TYPE_ANALYZE = 1 << 5,
Don't to predicate lock for analyze scans, refactor scan option passing. Before this commit, when ANALYZE was run on a table and serializable was used (either by virtue of an explicit BEGIN TRANSACTION ISOLATION LEVEL SERIALIZABLE, or default_transaction_isolation being set to serializable) a null pointer dereference lead to a crash. The analyze scan doesn't need a snapshot (nor predicate locking), but before this commit a scan only contained information about being a bitmap or sample scan. Refactor the option passing to the scan_begin callback to use a bitmask instead. Alternatively we could have added a new boolean parameter, but that seems harder to read. Even before this issue various people (Heikki, Tom, Robert) suggested doing so. These changes don't change the scan APIs outside of tableam. The flags argument could be exposed, it's not necessary to fix this problem. Also the wrapper table_beginscan* functions encapsulate most of that complexity. After these changes fixing the bug is trivial, just don't acquire predicate lock for analyze style scans. That was already done for bitmap heap scans. Add an assert that a snapshot is passed when acquiring the predicate lock, so this kind of bug doesn't require running with serializable. Also add a comment about sample scans currently requiring predicate locking the entire relation, that previously wasn't remarked upon. Reported-By: Joe Wildish Author: Andres Freund Discussion: https://postgr.es/m/4EA80A20-E9BF-49F1-9F01-5B66CAB21453@elusive.cx https://postgr.es/m/20190411164947.nkii4gaeilt4bui7@alap3.anarazel.de https://postgr.es/m/20190518203102.g7peu2fianukjuxm@alap3.anarazel.de
2019-05-20 00:10:28 +02:00
/* several of SO_ALLOW_* may be specified */
/* allow or disallow use of access strategy */
Add TID Range Scans to support efficient scanning ranges of TIDs This adds a new executor node named TID Range Scan. The query planner will generate paths for TID Range scans when quals are discovered on base relations which search for ranges on the table's ctid column. These ranges may be open at either end. For example, WHERE ctid >= '(10,0)'; will return all tuples on page 10 and over. To support this, two new optional callback functions have been added to table AM. scan_set_tidrange is used to set the scan range to just the given range of TIDs. scan_getnextslot_tidrange fetches the next tuple in the given range. For AMs were scanning ranges of TIDs would not make sense, these functions can be set to NULL in the TableAmRoutine. The query planner won't generate TID Range Scan Paths in that case. Author: Edmund Horner, David Rowley Reviewed-by: David Rowley, Tomas Vondra, Tom Lane, Andres Freund, Zhihong Yu Discussion: https://postgr.es/m/CAMyN-kB-nFTkF=VA_JPwFNo08S0d-Yk0F741S2B7LDmYAi8eyA@mail.gmail.com
2021-02-27 10:59:36 +01:00
SO_ALLOW_STRAT = 1 << 6,
Don't to predicate lock for analyze scans, refactor scan option passing. Before this commit, when ANALYZE was run on a table and serializable was used (either by virtue of an explicit BEGIN TRANSACTION ISOLATION LEVEL SERIALIZABLE, or default_transaction_isolation being set to serializable) a null pointer dereference lead to a crash. The analyze scan doesn't need a snapshot (nor predicate locking), but before this commit a scan only contained information about being a bitmap or sample scan. Refactor the option passing to the scan_begin callback to use a bitmask instead. Alternatively we could have added a new boolean parameter, but that seems harder to read. Even before this issue various people (Heikki, Tom, Robert) suggested doing so. These changes don't change the scan APIs outside of tableam. The flags argument could be exposed, it's not necessary to fix this problem. Also the wrapper table_beginscan* functions encapsulate most of that complexity. After these changes fixing the bug is trivial, just don't acquire predicate lock for analyze style scans. That was already done for bitmap heap scans. Add an assert that a snapshot is passed when acquiring the predicate lock, so this kind of bug doesn't require running with serializable. Also add a comment about sample scans currently requiring predicate locking the entire relation, that previously wasn't remarked upon. Reported-By: Joe Wildish Author: Andres Freund Discussion: https://postgr.es/m/4EA80A20-E9BF-49F1-9F01-5B66CAB21453@elusive.cx https://postgr.es/m/20190411164947.nkii4gaeilt4bui7@alap3.anarazel.de https://postgr.es/m/20190518203102.g7peu2fianukjuxm@alap3.anarazel.de
2019-05-20 00:10:28 +02:00
/* report location to syncscan logic? */
Add TID Range Scans to support efficient scanning ranges of TIDs This adds a new executor node named TID Range Scan. The query planner will generate paths for TID Range scans when quals are discovered on base relations which search for ranges on the table's ctid column. These ranges may be open at either end. For example, WHERE ctid >= '(10,0)'; will return all tuples on page 10 and over. To support this, two new optional callback functions have been added to table AM. scan_set_tidrange is used to set the scan range to just the given range of TIDs. scan_getnextslot_tidrange fetches the next tuple in the given range. For AMs were scanning ranges of TIDs would not make sense, these functions can be set to NULL in the TableAmRoutine. The query planner won't generate TID Range Scan Paths in that case. Author: Edmund Horner, David Rowley Reviewed-by: David Rowley, Tomas Vondra, Tom Lane, Andres Freund, Zhihong Yu Discussion: https://postgr.es/m/CAMyN-kB-nFTkF=VA_JPwFNo08S0d-Yk0F741S2B7LDmYAi8eyA@mail.gmail.com
2021-02-27 10:59:36 +01:00
SO_ALLOW_SYNC = 1 << 7,
Don't to predicate lock for analyze scans, refactor scan option passing. Before this commit, when ANALYZE was run on a table and serializable was used (either by virtue of an explicit BEGIN TRANSACTION ISOLATION LEVEL SERIALIZABLE, or default_transaction_isolation being set to serializable) a null pointer dereference lead to a crash. The analyze scan doesn't need a snapshot (nor predicate locking), but before this commit a scan only contained information about being a bitmap or sample scan. Refactor the option passing to the scan_begin callback to use a bitmask instead. Alternatively we could have added a new boolean parameter, but that seems harder to read. Even before this issue various people (Heikki, Tom, Robert) suggested doing so. These changes don't change the scan APIs outside of tableam. The flags argument could be exposed, it's not necessary to fix this problem. Also the wrapper table_beginscan* functions encapsulate most of that complexity. After these changes fixing the bug is trivial, just don't acquire predicate lock for analyze style scans. That was already done for bitmap heap scans. Add an assert that a snapshot is passed when acquiring the predicate lock, so this kind of bug doesn't require running with serializable. Also add a comment about sample scans currently requiring predicate locking the entire relation, that previously wasn't remarked upon. Reported-By: Joe Wildish Author: Andres Freund Discussion: https://postgr.es/m/4EA80A20-E9BF-49F1-9F01-5B66CAB21453@elusive.cx https://postgr.es/m/20190411164947.nkii4gaeilt4bui7@alap3.anarazel.de https://postgr.es/m/20190518203102.g7peu2fianukjuxm@alap3.anarazel.de
2019-05-20 00:10:28 +02:00
/* verify visibility page-at-a-time? */
Add TID Range Scans to support efficient scanning ranges of TIDs This adds a new executor node named TID Range Scan. The query planner will generate paths for TID Range scans when quals are discovered on base relations which search for ranges on the table's ctid column. These ranges may be open at either end. For example, WHERE ctid >= '(10,0)'; will return all tuples on page 10 and over. To support this, two new optional callback functions have been added to table AM. scan_set_tidrange is used to set the scan range to just the given range of TIDs. scan_getnextslot_tidrange fetches the next tuple in the given range. For AMs were scanning ranges of TIDs would not make sense, these functions can be set to NULL in the TableAmRoutine. The query planner won't generate TID Range Scan Paths in that case. Author: Edmund Horner, David Rowley Reviewed-by: David Rowley, Tomas Vondra, Tom Lane, Andres Freund, Zhihong Yu Discussion: https://postgr.es/m/CAMyN-kB-nFTkF=VA_JPwFNo08S0d-Yk0F741S2B7LDmYAi8eyA@mail.gmail.com
2021-02-27 10:59:36 +01:00
SO_ALLOW_PAGEMODE = 1 << 8,
Don't to predicate lock for analyze scans, refactor scan option passing. Before this commit, when ANALYZE was run on a table and serializable was used (either by virtue of an explicit BEGIN TRANSACTION ISOLATION LEVEL SERIALIZABLE, or default_transaction_isolation being set to serializable) a null pointer dereference lead to a crash. The analyze scan doesn't need a snapshot (nor predicate locking), but before this commit a scan only contained information about being a bitmap or sample scan. Refactor the option passing to the scan_begin callback to use a bitmask instead. Alternatively we could have added a new boolean parameter, but that seems harder to read. Even before this issue various people (Heikki, Tom, Robert) suggested doing so. These changes don't change the scan APIs outside of tableam. The flags argument could be exposed, it's not necessary to fix this problem. Also the wrapper table_beginscan* functions encapsulate most of that complexity. After these changes fixing the bug is trivial, just don't acquire predicate lock for analyze style scans. That was already done for bitmap heap scans. Add an assert that a snapshot is passed when acquiring the predicate lock, so this kind of bug doesn't require running with serializable. Also add a comment about sample scans currently requiring predicate locking the entire relation, that previously wasn't remarked upon. Reported-By: Joe Wildish Author: Andres Freund Discussion: https://postgr.es/m/4EA80A20-E9BF-49F1-9F01-5B66CAB21453@elusive.cx https://postgr.es/m/20190411164947.nkii4gaeilt4bui7@alap3.anarazel.de https://postgr.es/m/20190518203102.g7peu2fianukjuxm@alap3.anarazel.de
2019-05-20 00:10:28 +02:00
/* unregister snapshot at scan end? */
Add TID Range Scans to support efficient scanning ranges of TIDs This adds a new executor node named TID Range Scan. The query planner will generate paths for TID Range scans when quals are discovered on base relations which search for ranges on the table's ctid column. These ranges may be open at either end. For example, WHERE ctid >= '(10,0)'; will return all tuples on page 10 and over. To support this, two new optional callback functions have been added to table AM. scan_set_tidrange is used to set the scan range to just the given range of TIDs. scan_getnextslot_tidrange fetches the next tuple in the given range. For AMs were scanning ranges of TIDs would not make sense, these functions can be set to NULL in the TableAmRoutine. The query planner won't generate TID Range Scan Paths in that case. Author: Edmund Horner, David Rowley Reviewed-by: David Rowley, Tomas Vondra, Tom Lane, Andres Freund, Zhihong Yu Discussion: https://postgr.es/m/CAMyN-kB-nFTkF=VA_JPwFNo08S0d-Yk0F741S2B7LDmYAi8eyA@mail.gmail.com
2021-02-27 10:59:36 +01:00
SO_TEMP_SNAPSHOT = 1 << 9
Don't to predicate lock for analyze scans, refactor scan option passing. Before this commit, when ANALYZE was run on a table and serializable was used (either by virtue of an explicit BEGIN TRANSACTION ISOLATION LEVEL SERIALIZABLE, or default_transaction_isolation being set to serializable) a null pointer dereference lead to a crash. The analyze scan doesn't need a snapshot (nor predicate locking), but before this commit a scan only contained information about being a bitmap or sample scan. Refactor the option passing to the scan_begin callback to use a bitmask instead. Alternatively we could have added a new boolean parameter, but that seems harder to read. Even before this issue various people (Heikki, Tom, Robert) suggested doing so. These changes don't change the scan APIs outside of tableam. The flags argument could be exposed, it's not necessary to fix this problem. Also the wrapper table_beginscan* functions encapsulate most of that complexity. After these changes fixing the bug is trivial, just don't acquire predicate lock for analyze style scans. That was already done for bitmap heap scans. Add an assert that a snapshot is passed when acquiring the predicate lock, so this kind of bug doesn't require running with serializable. Also add a comment about sample scans currently requiring predicate locking the entire relation, that previously wasn't remarked upon. Reported-By: Joe Wildish Author: Andres Freund Discussion: https://postgr.es/m/4EA80A20-E9BF-49F1-9F01-5B66CAB21453@elusive.cx https://postgr.es/m/20190411164947.nkii4gaeilt4bui7@alap3.anarazel.de https://postgr.es/m/20190518203102.g7peu2fianukjuxm@alap3.anarazel.de
2019-05-20 00:10:28 +02:00
} ScanOptions;
tableam: Add tuple_{insert, delete, update, lock} and use. This adds new, required, table AM callbacks for insert/delete/update and lock_tuple. To be able to reasonably use those, the EvalPlanQual mechanism had to be adapted, moving more logic into the AM. Previously both delete/update/lock call-sites and the EPQ mechanism had to have awareness of the specific tuple format to be able to fetch the latest version of a tuple. Obviously that needs to be abstracted away. To do so, move the logic that find the latest row version into the AM. lock_tuple has a new flag argument, TUPLE_LOCK_FLAG_FIND_LAST_VERSION, that forces it to lock the last version, rather than the current one. It'd have been possible to do so via a separate callback as well, but finding the last version usually also necessitates locking the newest version, making it sensible to combine the two. This replaces the previous use of EvalPlanQualFetch(). Additionally HeapTupleUpdated, which previously signaled either a concurrent update or delete, is now split into two, to avoid callers needing AM specific knowledge to differentiate. The move of finding the latest row version into tuple_lock means that encountering a row concurrently moved into another partition will now raise an error about "tuple to be locked" rather than "tuple to be updated/deleted" - which is accurate, as that always happens when locking rows. While possible slightly less helpful for users, it seems like an acceptable trade-off. As part of this commit HTSU_Result has been renamed to TM_Result, and its members been expanded to differentiated between updating and deleting. HeapUpdateFailureData has been renamed to TM_FailureData. The interface to speculative insertion is changed so nodeModifyTable.c does not have to set the speculative token itself anymore. Instead there's a version of tuple_insert, tuple_insert_speculative, that performs the speculative insertion (without requiring a flag to signal that fact), and the speculative insertion is either made permanent with table_complete_speculative(succeeded = true) or aborted with succeeded = false). Note that multi_insert is not yet routed through tableam, nor is COPY. Changing multi_insert requires changes to copy.c that are large enough to better be done separately. Similarly, although simpler, CREATE TABLE AS and CREATE MATERIALIZED VIEW are also only going to be adjusted in a later commit. Author: Andres Freund and Haribabu Kommi Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20190313003903.nwvrxi7rw3ywhdel@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-24 03:55:57 +01:00
/*
tableam: comment and formatting fixes. Author: Heikki Linnakangas Discussion: https://postgr.es/m/9a7fb9cc-2419-5db7-8840-ddc10c93f122@iki.fi
2019-04-08 19:25:16 +02:00
* Result codes for table_{update,delete,lock_tuple}, and for visibility
tableam: Add tuple_{insert, delete, update, lock} and use. This adds new, required, table AM callbacks for insert/delete/update and lock_tuple. To be able to reasonably use those, the EvalPlanQual mechanism had to be adapted, moving more logic into the AM. Previously both delete/update/lock call-sites and the EPQ mechanism had to have awareness of the specific tuple format to be able to fetch the latest version of a tuple. Obviously that needs to be abstracted away. To do so, move the logic that find the latest row version into the AM. lock_tuple has a new flag argument, TUPLE_LOCK_FLAG_FIND_LAST_VERSION, that forces it to lock the last version, rather than the current one. It'd have been possible to do so via a separate callback as well, but finding the last version usually also necessitates locking the newest version, making it sensible to combine the two. This replaces the previous use of EvalPlanQualFetch(). Additionally HeapTupleUpdated, which previously signaled either a concurrent update or delete, is now split into two, to avoid callers needing AM specific knowledge to differentiate. The move of finding the latest row version into tuple_lock means that encountering a row concurrently moved into another partition will now raise an error about "tuple to be locked" rather than "tuple to be updated/deleted" - which is accurate, as that always happens when locking rows. While possible slightly less helpful for users, it seems like an acceptable trade-off. As part of this commit HTSU_Result has been renamed to TM_Result, and its members been expanded to differentiated between updating and deleting. HeapUpdateFailureData has been renamed to TM_FailureData. The interface to speculative insertion is changed so nodeModifyTable.c does not have to set the speculative token itself anymore. Instead there's a version of tuple_insert, tuple_insert_speculative, that performs the speculative insertion (without requiring a flag to signal that fact), and the speculative insertion is either made permanent with table_complete_speculative(succeeded = true) or aborted with succeeded = false). Note that multi_insert is not yet routed through tableam, nor is COPY. Changing multi_insert requires changes to copy.c that are large enough to better be done separately. Similarly, although simpler, CREATE TABLE AS and CREATE MATERIALIZED VIEW are also only going to be adjusted in a later commit. Author: Andres Freund and Haribabu Kommi Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20190313003903.nwvrxi7rw3ywhdel@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-24 03:55:57 +01:00
* routines inside table AMs.
*/
typedef enum TM_Result
{
/*
* Signals that the action succeeded (i.e. update/delete performed, lock
* was acquired)
*/
TM_Ok,
/* The affected tuple wasn't visible to the relevant snapshot */
TM_Invisible,
/* The affected tuple was already modified by the calling backend */
TM_SelfModified,
/*
* The affected tuple was updated by another transaction. This includes
* the case where tuple was moved to another partition.
*/
TM_Updated,
/* The affected tuple was deleted by another transaction */
TM_Deleted,
/*
* The affected tuple is currently being modified by another session. This
tableam: comment and formatting fixes. Author: Heikki Linnakangas Discussion: https://postgr.es/m/9a7fb9cc-2419-5db7-8840-ddc10c93f122@iki.fi
2019-04-08 19:25:16 +02:00
* will only be returned if table_(update/delete/lock_tuple) are
* instructed not to wait.
tableam: Add tuple_{insert, delete, update, lock} and use. This adds new, required, table AM callbacks for insert/delete/update and lock_tuple. To be able to reasonably use those, the EvalPlanQual mechanism had to be adapted, moving more logic into the AM. Previously both delete/update/lock call-sites and the EPQ mechanism had to have awareness of the specific tuple format to be able to fetch the latest version of a tuple. Obviously that needs to be abstracted away. To do so, move the logic that find the latest row version into the AM. lock_tuple has a new flag argument, TUPLE_LOCK_FLAG_FIND_LAST_VERSION, that forces it to lock the last version, rather than the current one. It'd have been possible to do so via a separate callback as well, but finding the last version usually also necessitates locking the newest version, making it sensible to combine the two. This replaces the previous use of EvalPlanQualFetch(). Additionally HeapTupleUpdated, which previously signaled either a concurrent update or delete, is now split into two, to avoid callers needing AM specific knowledge to differentiate. The move of finding the latest row version into tuple_lock means that encountering a row concurrently moved into another partition will now raise an error about "tuple to be locked" rather than "tuple to be updated/deleted" - which is accurate, as that always happens when locking rows. While possible slightly less helpful for users, it seems like an acceptable trade-off. As part of this commit HTSU_Result has been renamed to TM_Result, and its members been expanded to differentiated between updating and deleting. HeapUpdateFailureData has been renamed to TM_FailureData. The interface to speculative insertion is changed so nodeModifyTable.c does not have to set the speculative token itself anymore. Instead there's a version of tuple_insert, tuple_insert_speculative, that performs the speculative insertion (without requiring a flag to signal that fact), and the speculative insertion is either made permanent with table_complete_speculative(succeeded = true) or aborted with succeeded = false). Note that multi_insert is not yet routed through tableam, nor is COPY. Changing multi_insert requires changes to copy.c that are large enough to better be done separately. Similarly, although simpler, CREATE TABLE AS and CREATE MATERIALIZED VIEW are also only going to be adjusted in a later commit. Author: Andres Freund and Haribabu Kommi Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20190313003903.nwvrxi7rw3ywhdel@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-24 03:55:57 +01:00
*/
TM_BeingModified,
/* lock couldn't be acquired, action skipped. Only used by lock_tuple */
TM_WouldBlock
} TM_Result;
Ignore BRIN indexes when checking for HOT updates When determining whether an index update may be skipped by using HOT, we can ignore attributes indexed by block summarizing indexes without references to individual tuples that need to be cleaned up. A new type TU_UpdateIndexes provides a signal to the executor to determine which indexes to update - no indexes, all indexes, or only the summarizing indexes. This also removes rd_indexattr list, and replaces it with rd_attrsvalid flag. The list was not used anywhere, and a simple flag is sufficient. This was originally committed as 5753d4ee32, but then got reverted by e3fcca0d0d because of correctness issues. Original patch by Josef Simanek, various fixes and improvements by Tomas Vondra and me. Authors: Matthias van de Meent, Josef Simanek, Tomas Vondra Reviewed-by: Tomas Vondra, Alvaro Herrera Discussion: https://postgr.es/m/05ebcb44-f383-86e3-4f31-0a97a55634cf@enterprisedb.com Discussion: https://postgr.es/m/CAFp7QwpMRGcDAQumN7onN9HjrJ3u4X3ZRXdGFT0K5G2JWvnbWg%40mail.gmail.com
2023-03-20 10:34:07 +01:00
/*
* Result codes for table_update(..., update_indexes*..).
* Used to determine which indexes to update.
*/
typedef enum TU_UpdateIndexes
{
/* No indexed columns were updated (incl. TID addressing of tuple) */
TU_None,
/* A non-summarizing indexed column was updated, or the TID has changed */
TU_All,
/* Only summarized columns were updated, TID is unchanged */
TU_Summarizing
} TU_UpdateIndexes;
tableam: Add tuple_{insert, delete, update, lock} and use. This adds new, required, table AM callbacks for insert/delete/update and lock_tuple. To be able to reasonably use those, the EvalPlanQual mechanism had to be adapted, moving more logic into the AM. Previously both delete/update/lock call-sites and the EPQ mechanism had to have awareness of the specific tuple format to be able to fetch the latest version of a tuple. Obviously that needs to be abstracted away. To do so, move the logic that find the latest row version into the AM. lock_tuple has a new flag argument, TUPLE_LOCK_FLAG_FIND_LAST_VERSION, that forces it to lock the last version, rather than the current one. It'd have been possible to do so via a separate callback as well, but finding the last version usually also necessitates locking the newest version, making it sensible to combine the two. This replaces the previous use of EvalPlanQualFetch(). Additionally HeapTupleUpdated, which previously signaled either a concurrent update or delete, is now split into two, to avoid callers needing AM specific knowledge to differentiate. The move of finding the latest row version into tuple_lock means that encountering a row concurrently moved into another partition will now raise an error about "tuple to be locked" rather than "tuple to be updated/deleted" - which is accurate, as that always happens when locking rows. While possible slightly less helpful for users, it seems like an acceptable trade-off. As part of this commit HTSU_Result has been renamed to TM_Result, and its members been expanded to differentiated between updating and deleting. HeapUpdateFailureData has been renamed to TM_FailureData. The interface to speculative insertion is changed so nodeModifyTable.c does not have to set the speculative token itself anymore. Instead there's a version of tuple_insert, tuple_insert_speculative, that performs the speculative insertion (without requiring a flag to signal that fact), and the speculative insertion is either made permanent with table_complete_speculative(succeeded = true) or aborted with succeeded = false). Note that multi_insert is not yet routed through tableam, nor is COPY. Changing multi_insert requires changes to copy.c that are large enough to better be done separately. Similarly, although simpler, CREATE TABLE AS and CREATE MATERIALIZED VIEW are also only going to be adjusted in a later commit. Author: Andres Freund and Haribabu Kommi Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20190313003903.nwvrxi7rw3ywhdel@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-24 03:55:57 +01:00
/*
tableam: Rename wrapper functions to match callback names. Some of the wrapper functions didn't match the callback names. Many of them due to staying "consistent" with historic naming of the wrapped functionality. We decided that for most cases it's more important to be for tableam to be consistent going forward, than with the past. The one exception is beginscan/endscan/... because it'd have looked odd to have systable_beginscan/endscan/... with a different naming scheme, and changing the systable_* APIs would have caused way too much churn (including breaking a lot of external users). Author: Ashwin Agrawal, with some small additions by Andres Freund Reviewed-By: Andres Freund Discussion: https://postgr.es/m/CALfoeiugyrXZfX7n0ORCa4L-m834dzmaE8eFdbNR6PMpetU4Ww@mail.gmail.com
2019-05-24 01:25:48 +02:00
* When table_tuple_update, table_tuple_delete, or table_tuple_lock fail
* because the target tuple is already outdated, they fill in this struct to
* provide information to the caller about what happened.
tableam: comment and formatting fixes. Author: Heikki Linnakangas Discussion: https://postgr.es/m/9a7fb9cc-2419-5db7-8840-ddc10c93f122@iki.fi
2019-04-08 19:25:16 +02:00
*
tableam: Add tuple_{insert, delete, update, lock} and use. This adds new, required, table AM callbacks for insert/delete/update and lock_tuple. To be able to reasonably use those, the EvalPlanQual mechanism had to be adapted, moving more logic into the AM. Previously both delete/update/lock call-sites and the EPQ mechanism had to have awareness of the specific tuple format to be able to fetch the latest version of a tuple. Obviously that needs to be abstracted away. To do so, move the logic that find the latest row version into the AM. lock_tuple has a new flag argument, TUPLE_LOCK_FLAG_FIND_LAST_VERSION, that forces it to lock the last version, rather than the current one. It'd have been possible to do so via a separate callback as well, but finding the last version usually also necessitates locking the newest version, making it sensible to combine the two. This replaces the previous use of EvalPlanQualFetch(). Additionally HeapTupleUpdated, which previously signaled either a concurrent update or delete, is now split into two, to avoid callers needing AM specific knowledge to differentiate. The move of finding the latest row version into tuple_lock means that encountering a row concurrently moved into another partition will now raise an error about "tuple to be locked" rather than "tuple to be updated/deleted" - which is accurate, as that always happens when locking rows. While possible slightly less helpful for users, it seems like an acceptable trade-off. As part of this commit HTSU_Result has been renamed to TM_Result, and its members been expanded to differentiated between updating and deleting. HeapUpdateFailureData has been renamed to TM_FailureData. The interface to speculative insertion is changed so nodeModifyTable.c does not have to set the speculative token itself anymore. Instead there's a version of tuple_insert, tuple_insert_speculative, that performs the speculative insertion (without requiring a flag to signal that fact), and the speculative insertion is either made permanent with table_complete_speculative(succeeded = true) or aborted with succeeded = false). Note that multi_insert is not yet routed through tableam, nor is COPY. Changing multi_insert requires changes to copy.c that are large enough to better be done separately. Similarly, although simpler, CREATE TABLE AS and CREATE MATERIALIZED VIEW are also only going to be adjusted in a later commit. Author: Andres Freund and Haribabu Kommi Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20190313003903.nwvrxi7rw3ywhdel@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-24 03:55:57 +01:00
* ctid is the target's ctid link: it is the same as the target's TID if the
* target was deleted, or the location of the replacement tuple if the target
* was updated.
tableam: comment and formatting fixes. Author: Heikki Linnakangas Discussion: https://postgr.es/m/9a7fb9cc-2419-5db7-8840-ddc10c93f122@iki.fi
2019-04-08 19:25:16 +02:00
*
tableam: Add tuple_{insert, delete, update, lock} and use. This adds new, required, table AM callbacks for insert/delete/update and lock_tuple. To be able to reasonably use those, the EvalPlanQual mechanism had to be adapted, moving more logic into the AM. Previously both delete/update/lock call-sites and the EPQ mechanism had to have awareness of the specific tuple format to be able to fetch the latest version of a tuple. Obviously that needs to be abstracted away. To do so, move the logic that find the latest row version into the AM. lock_tuple has a new flag argument, TUPLE_LOCK_FLAG_FIND_LAST_VERSION, that forces it to lock the last version, rather than the current one. It'd have been possible to do so via a separate callback as well, but finding the last version usually also necessitates locking the newest version, making it sensible to combine the two. This replaces the previous use of EvalPlanQualFetch(). Additionally HeapTupleUpdated, which previously signaled either a concurrent update or delete, is now split into two, to avoid callers needing AM specific knowledge to differentiate. The move of finding the latest row version into tuple_lock means that encountering a row concurrently moved into another partition will now raise an error about "tuple to be locked" rather than "tuple to be updated/deleted" - which is accurate, as that always happens when locking rows. While possible slightly less helpful for users, it seems like an acceptable trade-off. As part of this commit HTSU_Result has been renamed to TM_Result, and its members been expanded to differentiated between updating and deleting. HeapUpdateFailureData has been renamed to TM_FailureData. The interface to speculative insertion is changed so nodeModifyTable.c does not have to set the speculative token itself anymore. Instead there's a version of tuple_insert, tuple_insert_speculative, that performs the speculative insertion (without requiring a flag to signal that fact), and the speculative insertion is either made permanent with table_complete_speculative(succeeded = true) or aborted with succeeded = false). Note that multi_insert is not yet routed through tableam, nor is COPY. Changing multi_insert requires changes to copy.c that are large enough to better be done separately. Similarly, although simpler, CREATE TABLE AS and CREATE MATERIALIZED VIEW are also only going to be adjusted in a later commit. Author: Andres Freund and Haribabu Kommi Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20190313003903.nwvrxi7rw3ywhdel@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-24 03:55:57 +01:00
* xmax is the outdating transaction's XID. If the caller wants to visit the
* replacement tuple, it must check that this matches before believing the
* replacement is really a match.
tableam: comment and formatting fixes. Author: Heikki Linnakangas Discussion: https://postgr.es/m/9a7fb9cc-2419-5db7-8840-ddc10c93f122@iki.fi
2019-04-08 19:25:16 +02:00
*
tableam: Add tuple_{insert, delete, update, lock} and use. This adds new, required, table AM callbacks for insert/delete/update and lock_tuple. To be able to reasonably use those, the EvalPlanQual mechanism had to be adapted, moving more logic into the AM. Previously both delete/update/lock call-sites and the EPQ mechanism had to have awareness of the specific tuple format to be able to fetch the latest version of a tuple. Obviously that needs to be abstracted away. To do so, move the logic that find the latest row version into the AM. lock_tuple has a new flag argument, TUPLE_LOCK_FLAG_FIND_LAST_VERSION, that forces it to lock the last version, rather than the current one. It'd have been possible to do so via a separate callback as well, but finding the last version usually also necessitates locking the newest version, making it sensible to combine the two. This replaces the previous use of EvalPlanQualFetch(). Additionally HeapTupleUpdated, which previously signaled either a concurrent update or delete, is now split into two, to avoid callers needing AM specific knowledge to differentiate. The move of finding the latest row version into tuple_lock means that encountering a row concurrently moved into another partition will now raise an error about "tuple to be locked" rather than "tuple to be updated/deleted" - which is accurate, as that always happens when locking rows. While possible slightly less helpful for users, it seems like an acceptable trade-off. As part of this commit HTSU_Result has been renamed to TM_Result, and its members been expanded to differentiated between updating and deleting. HeapUpdateFailureData has been renamed to TM_FailureData. The interface to speculative insertion is changed so nodeModifyTable.c does not have to set the speculative token itself anymore. Instead there's a version of tuple_insert, tuple_insert_speculative, that performs the speculative insertion (without requiring a flag to signal that fact), and the speculative insertion is either made permanent with table_complete_speculative(succeeded = true) or aborted with succeeded = false). Note that multi_insert is not yet routed through tableam, nor is COPY. Changing multi_insert requires changes to copy.c that are large enough to better be done separately. Similarly, although simpler, CREATE TABLE AS and CREATE MATERIALIZED VIEW are also only going to be adjusted in a later commit. Author: Andres Freund and Haribabu Kommi Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20190313003903.nwvrxi7rw3ywhdel@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-24 03:55:57 +01:00
* cmax is the outdating command's CID, but only when the failure code is
* TM_SelfModified (i.e., something in the current transaction outdated the
* tuple); otherwise cmax is zero. (We make this restriction because
* HeapTupleHeaderGetCmax doesn't work for tuples outdated in other
* transactions.)
*/
typedef struct TM_FailureData
{
ItemPointerData ctid;
TransactionId xmax;
CommandId cmax;
bool traversed;
} TM_FailureData;
Enhance nbtree index tuple deletion. Teach nbtree and heapam to cooperate in order to eagerly remove duplicate tuples representing dead MVCC versions. This is "bottom-up deletion". Each bottom-up deletion pass is triggered lazily in response to a flood of versions on an nbtree leaf page. This usually involves a "logically unchanged index" hint (these are produced by the executor mechanism added by commit 9dc718bd). The immediate goal of bottom-up index deletion is to avoid "unnecessary" page splits caused entirely by version duplicates. It naturally has an even more useful effect, though: it acts as a backstop against accumulating an excessive number of index tuple versions for any given _logical row_. Bottom-up index deletion complements what we might now call "top-down index deletion": index vacuuming performed by VACUUM. Bottom-up index deletion responds to the immediate local needs of queries, while leaving it up to autovacuum to perform infrequent clean sweeps of the index. The overall effect is to avoid certain pathological performance issues related to "version churn" from UPDATEs. The previous tableam interface used by index AMs to perform tuple deletion (the table_compute_xid_horizon_for_tuples() function) has been replaced with a new interface that supports certain new requirements. Many (perhaps all) of the capabilities added to nbtree by this commit could also be extended to other index AMs. That is left as work for a later commit. Extend deletion of LP_DEAD-marked index tuples in nbtree by adding logic to consider extra index tuples (that are not LP_DEAD-marked) for deletion in passing. This increases the number of index tuples deleted significantly in many cases. The LP_DEAD deletion process (which is now called "simple deletion" to clearly distinguish it from bottom-up deletion) won't usually need to visit any extra table blocks to check these extra tuples. We have to visit the same table blocks anyway to generate a latestRemovedXid value (at least in the common case where the index deletion operation's WAL record needs such a value). Testing has shown that the "extra tuples" simple deletion enhancement increases the number of index tuples deleted with almost any workload that has LP_DEAD bits set in leaf pages. That is, it almost never fails to delete at least a few extra index tuples. It helps most of all in cases that happen to naturally have a lot of delete-safe tuples. It's not uncommon for an individual deletion operation to end up deleting an order of magnitude more index tuples compared to the old naive approach (e.g., custom instrumentation of the patch shows that this happens fairly often when the regression tests are run). Add a further enhancement that augments simple deletion and bottom-up deletion in indexes that make use of deduplication: Teach nbtree's _bt_delitems_delete() function to support granular TID deletion in posting list tuples. It is now possible to delete individual TIDs from posting list tuples provided the TIDs have a tableam block number of a table block that gets visited as part of the deletion process (visiting the table block can be triggered directly or indirectly). Setting the LP_DEAD bit of a posting list tuple is still an all-or-nothing thing, but that matters much less now that deletion only needs to start out with the right _general_ idea about which index tuples are deletable. Bump XLOG_PAGE_MAGIC because xl_btree_delete changed. No bump in BTREE_VERSION, since there are no changes to the on-disk representation of nbtree indexes. Indexes built on PostgreSQL 12 or PostgreSQL 13 will automatically benefit from bottom-up index deletion (i.e. no reindexing required) following a pg_upgrade. The enhancement to simple deletion is available with all B-Tree indexes following a pg_upgrade, no matter what PostgreSQL version the user upgrades from. Author: Peter Geoghegan <pg@bowt.ie> Reviewed-By: Heikki Linnakangas <hlinnaka@iki.fi> Reviewed-By: Victor Yegorov <vyegorov@gmail.com> Discussion: https://postgr.es/m/CAH2-Wzm+maE3apHB8NOtmM=p-DO65j2V5GzAWCOEEuy3JZgb2g@mail.gmail.com
2021-01-13 18:21:32 +01:00
/*
* State used when calling table_index_delete_tuples().
*
* Represents the status of table tuples, referenced by table TID and taken by
* index AM from index tuples. State consists of high level parameters of the
* deletion operation, plus two mutable palloc()'d arrays for information
* about the status of individual table tuples. These are conceptually one
* single array. Using two arrays keeps the TM_IndexDelete struct small,
* which makes sorting the first array (the deltids array) fast.
*
* Some index AM callers perform simple index tuple deletion (by specifying
* bottomup = false), and include only known-dead deltids. These known-dead
* entries are all marked knowndeletable = true directly (typically these are
* TIDs from LP_DEAD-marked index tuples), but that isn't strictly required.
*
* Callers that specify bottomup = true are "bottom-up index deletion"
* callers. The considerations for the tableam are more subtle with these
* callers because they ask the tableam to perform highly speculative work,
* and might only expect the tableam to check a small fraction of all entries.
* Caller is not allowed to specify knowndeletable = true for any entry
* because everything is highly speculative. Bottom-up caller provides
* context and hints to tableam -- see comments below for details on how index
* AMs and tableams should coordinate during bottom-up index deletion.
*
* Simple index deletion callers may ask the tableam to perform speculative
* work, too. This is a little like bottom-up deletion, but not too much.
* The tableam will only perform speculative work when it's practically free
* to do so in passing for simple deletion caller (while always performing
Fix duplicate words in comments Remove accidentally duplicated words in code comments. Author: Dagfinn Ilmari Mannsåker <ilmari@ilmari.org> Discussion: https://postgr.es/m/87bl45t0co.fsf@wibble.ilmari.org
2021-10-04 15:12:57 +02:00
* whatever work is needed to enable knowndeletable/LP_DEAD index tuples to
Enhance nbtree index tuple deletion. Teach nbtree and heapam to cooperate in order to eagerly remove duplicate tuples representing dead MVCC versions. This is "bottom-up deletion". Each bottom-up deletion pass is triggered lazily in response to a flood of versions on an nbtree leaf page. This usually involves a "logically unchanged index" hint (these are produced by the executor mechanism added by commit 9dc718bd). The immediate goal of bottom-up index deletion is to avoid "unnecessary" page splits caused entirely by version duplicates. It naturally has an even more useful effect, though: it acts as a backstop against accumulating an excessive number of index tuple versions for any given _logical row_. Bottom-up index deletion complements what we might now call "top-down index deletion": index vacuuming performed by VACUUM. Bottom-up index deletion responds to the immediate local needs of queries, while leaving it up to autovacuum to perform infrequent clean sweeps of the index. The overall effect is to avoid certain pathological performance issues related to "version churn" from UPDATEs. The previous tableam interface used by index AMs to perform tuple deletion (the table_compute_xid_horizon_for_tuples() function) has been replaced with a new interface that supports certain new requirements. Many (perhaps all) of the capabilities added to nbtree by this commit could also be extended to other index AMs. That is left as work for a later commit. Extend deletion of LP_DEAD-marked index tuples in nbtree by adding logic to consider extra index tuples (that are not LP_DEAD-marked) for deletion in passing. This increases the number of index tuples deleted significantly in many cases. The LP_DEAD deletion process (which is now called "simple deletion" to clearly distinguish it from bottom-up deletion) won't usually need to visit any extra table blocks to check these extra tuples. We have to visit the same table blocks anyway to generate a latestRemovedXid value (at least in the common case where the index deletion operation's WAL record needs such a value). Testing has shown that the "extra tuples" simple deletion enhancement increases the number of index tuples deleted with almost any workload that has LP_DEAD bits set in leaf pages. That is, it almost never fails to delete at least a few extra index tuples. It helps most of all in cases that happen to naturally have a lot of delete-safe tuples. It's not uncommon for an individual deletion operation to end up deleting an order of magnitude more index tuples compared to the old naive approach (e.g., custom instrumentation of the patch shows that this happens fairly often when the regression tests are run). Add a further enhancement that augments simple deletion and bottom-up deletion in indexes that make use of deduplication: Teach nbtree's _bt_delitems_delete() function to support granular TID deletion in posting list tuples. It is now possible to delete individual TIDs from posting list tuples provided the TIDs have a tableam block number of a table block that gets visited as part of the deletion process (visiting the table block can be triggered directly or indirectly). Setting the LP_DEAD bit of a posting list tuple is still an all-or-nothing thing, but that matters much less now that deletion only needs to start out with the right _general_ idea about which index tuples are deletable. Bump XLOG_PAGE_MAGIC because xl_btree_delete changed. No bump in BTREE_VERSION, since there are no changes to the on-disk representation of nbtree indexes. Indexes built on PostgreSQL 12 or PostgreSQL 13 will automatically benefit from bottom-up index deletion (i.e. no reindexing required) following a pg_upgrade. The enhancement to simple deletion is available with all B-Tree indexes following a pg_upgrade, no matter what PostgreSQL version the user upgrades from. Author: Peter Geoghegan <pg@bowt.ie> Reviewed-By: Heikki Linnakangas <hlinnaka@iki.fi> Reviewed-By: Victor Yegorov <vyegorov@gmail.com> Discussion: https://postgr.es/m/CAH2-Wzm+maE3apHB8NOtmM=p-DO65j2V5GzAWCOEEuy3JZgb2g@mail.gmail.com
2021-01-13 18:21:32 +01:00
* be deleted within index AM). This is the real reason why it's possible for
* simple index deletion caller to specify knowndeletable = false up front
* (this means "check if it's possible for me to delete corresponding index
* tuple when it's cheap to do so in passing"). The index AM should only
* include "extra" entries for index tuples whose TIDs point to a table block
* that tableam is expected to have to visit anyway (in the event of a block
* orientated tableam). The tableam isn't strictly obligated to check these
* "extra" TIDs, but a block-based AM should always manage to do so in
* practice.
*
* The final contents of the deltids/status arrays are interesting to callers
* that ask tableam to perform speculative work (i.e. when _any_ items have
* knowndeletable set to false up front). These index AM callers will
* naturally need to consult final state to determine which index tuples are
* in fact deletable.
*
* The index AM can keep track of which index tuple relates to which deltid by
* setting idxoffnum (and/or relying on each entry being uniquely identifiable
* using tid), which is important when the final contents of the array will
* need to be interpreted -- the array can shrink from initial size after
* tableam processing and/or have entries in a new order (tableam may sort
* deltids array for its own reasons). Bottom-up callers may find that final
* ndeltids is 0 on return from call to tableam, in which case no index tuple
* deletions are possible. Simple deletion callers can rely on any entries
* they know to be deletable appearing in the final array as deletable.
*/
typedef struct TM_IndexDelete
{
ItemPointerData tid; /* table TID from index tuple */
int16 id; /* Offset into TM_IndexStatus array */
} TM_IndexDelete;
typedef struct TM_IndexStatus
{
OffsetNumber idxoffnum; /* Index am page offset number */
bool knowndeletable; /* Currently known to be deletable? */
/* Bottom-up index deletion specific fields follow */
bool promising; /* Promising (duplicate) index tuple? */
int16 freespace; /* Space freed in index if deleted */
} TM_IndexStatus;
/*
* Index AM/tableam coordination is central to the design of bottom-up index
* deletion. The index AM provides hints about where to look to the tableam
* by marking some entries as "promising". Index AM does this with duplicate
* index tuples that are strongly suspected to be old versions left behind by
* UPDATEs that did not logically modify indexed values. Index AM may find it
* helpful to only mark entries as promising when they're thought to have been
* affected by such an UPDATE in the recent past.
*
* Bottom-up index deletion casts a wide net at first, usually by including
* all TIDs on a target index page. It is up to the tableam to worry about
* the cost of checking transaction status information. The tableam is in
* control, but needs careful guidance from the index AM. Index AM requests
* that bottomupfreespace target be met, while tableam measures progress
* towards that goal by tallying the per-entry freespace value for known
* deletable entries. (All !bottomup callers can just set these space related
* fields to zero.)
*/
typedef struct TM_IndexDeleteOp
{
Add hardening to catch invalid TIDs in indexes. Add hardening to the heapam index tuple deletion path to catch TIDs in index pages that point to a heap item that index tuples should never point to. The corruption we're trying to catch here is particularly tricky to detect, since it typically involves "extra" (corrupt) index tuples, as opposed to the absence of required index tuples in the index. For example, a heap TID from an index page that turns out to point to an LP_UNUSED item in the heap page has a good chance of being caught by one of the new checks. There is a decent chance that the recently fixed parallel VACUUM bug (see commit 9bacec15) would have been caught had that particular check been in place for Postgres 14. No backpatch of this extra hardening for now, though. Author: Peter Geoghegan <pg@bowt.ie> Reviewed-By: Andres Freund <andres@anarazel.de> Discussion: https://postgr.es/m/CAH2-Wzk-4_raTzawWGaiqNvkpwDXxv3y1AQhQyUeHfkU=tFCeA@mail.gmail.com
2021-11-05 03:54:05 +01:00
Relation irel; /* Target index relation */
BlockNumber iblknum; /* Index block number (for error reports) */
Enhance nbtree index tuple deletion. Teach nbtree and heapam to cooperate in order to eagerly remove duplicate tuples representing dead MVCC versions. This is "bottom-up deletion". Each bottom-up deletion pass is triggered lazily in response to a flood of versions on an nbtree leaf page. This usually involves a "logically unchanged index" hint (these are produced by the executor mechanism added by commit 9dc718bd). The immediate goal of bottom-up index deletion is to avoid "unnecessary" page splits caused entirely by version duplicates. It naturally has an even more useful effect, though: it acts as a backstop against accumulating an excessive number of index tuple versions for any given _logical row_. Bottom-up index deletion complements what we might now call "top-down index deletion": index vacuuming performed by VACUUM. Bottom-up index deletion responds to the immediate local needs of queries, while leaving it up to autovacuum to perform infrequent clean sweeps of the index. The overall effect is to avoid certain pathological performance issues related to "version churn" from UPDATEs. The previous tableam interface used by index AMs to perform tuple deletion (the table_compute_xid_horizon_for_tuples() function) has been replaced with a new interface that supports certain new requirements. Many (perhaps all) of the capabilities added to nbtree by this commit could also be extended to other index AMs. That is left as work for a later commit. Extend deletion of LP_DEAD-marked index tuples in nbtree by adding logic to consider extra index tuples (that are not LP_DEAD-marked) for deletion in passing. This increases the number of index tuples deleted significantly in many cases. The LP_DEAD deletion process (which is now called "simple deletion" to clearly distinguish it from bottom-up deletion) won't usually need to visit any extra table blocks to check these extra tuples. We have to visit the same table blocks anyway to generate a latestRemovedXid value (at least in the common case where the index deletion operation's WAL record needs such a value). Testing has shown that the "extra tuples" simple deletion enhancement increases the number of index tuples deleted with almost any workload that has LP_DEAD bits set in leaf pages. That is, it almost never fails to delete at least a few extra index tuples. It helps most of all in cases that happen to naturally have a lot of delete-safe tuples. It's not uncommon for an individual deletion operation to end up deleting an order of magnitude more index tuples compared to the old naive approach (e.g., custom instrumentation of the patch shows that this happens fairly often when the regression tests are run). Add a further enhancement that augments simple deletion and bottom-up deletion in indexes that make use of deduplication: Teach nbtree's _bt_delitems_delete() function to support granular TID deletion in posting list tuples. It is now possible to delete individual TIDs from posting list tuples provided the TIDs have a tableam block number of a table block that gets visited as part of the deletion process (visiting the table block can be triggered directly or indirectly). Setting the LP_DEAD bit of a posting list tuple is still an all-or-nothing thing, but that matters much less now that deletion only needs to start out with the right _general_ idea about which index tuples are deletable. Bump XLOG_PAGE_MAGIC because xl_btree_delete changed. No bump in BTREE_VERSION, since there are no changes to the on-disk representation of nbtree indexes. Indexes built on PostgreSQL 12 or PostgreSQL 13 will automatically benefit from bottom-up index deletion (i.e. no reindexing required) following a pg_upgrade. The enhancement to simple deletion is available with all B-Tree indexes following a pg_upgrade, no matter what PostgreSQL version the user upgrades from. Author: Peter Geoghegan <pg@bowt.ie> Reviewed-By: Heikki Linnakangas <hlinnaka@iki.fi> Reviewed-By: Victor Yegorov <vyegorov@gmail.com> Discussion: https://postgr.es/m/CAH2-Wzm+maE3apHB8NOtmM=p-DO65j2V5GzAWCOEEuy3JZgb2g@mail.gmail.com
2021-01-13 18:21:32 +01:00
bool bottomup; /* Bottom-up (not simple) deletion? */
int bottomupfreespace; /* Bottom-up space target */
/* Mutable per-TID information follows (index AM initializes entries) */
int ndeltids; /* Current # of deltids/status elements */
TM_IndexDelete *deltids;
TM_IndexStatus *status;
} TM_IndexDeleteOp;
tableam: Rename wrapper functions to match callback names. Some of the wrapper functions didn't match the callback names. Many of them due to staying "consistent" with historic naming of the wrapped functionality. We decided that for most cases it's more important to be for tableam to be consistent going forward, than with the past. The one exception is beginscan/endscan/... because it'd have looked odd to have systable_beginscan/endscan/... with a different naming scheme, and changing the systable_* APIs would have caused way too much churn (including breaking a lot of external users). Author: Ashwin Agrawal, with some small additions by Andres Freund Reviewed-By: Andres Freund Discussion: https://postgr.es/m/CALfoeiugyrXZfX7n0ORCa4L-m834dzmaE8eFdbNR6PMpetU4Ww@mail.gmail.com
2019-05-24 01:25:48 +02:00
/* "options" flag bits for table_tuple_insert */
Skip WAL for new relfilenodes, under wal_level=minimal. Until now, only selected bulk operations (e.g. COPY) did this. If a given relfilenode received both a WAL-skipping COPY and a WAL-logged operation (e.g. INSERT), recovery could lose tuples from the COPY. See src/backend/access/transam/README section "Skipping WAL for New RelFileNode" for the new coding rules. Maintainers of table access methods should examine that section. To maintain data durability, just before commit, we choose between an fsync of the relfilenode and copying its contents to WAL. A new GUC, wal_skip_threshold, guides that choice. If this change slows a workload that creates small, permanent relfilenodes under wal_level=minimal, try adjusting wal_skip_threshold. Users setting a timeout on COMMIT may need to adjust that timeout, and log_min_duration_statement analysis will reflect time consumption moving to COMMIT from commands like COPY. Internally, this requires a reliable determination of whether RollbackAndReleaseCurrentSubTransaction() would unlink a relation's current relfilenode. Introduce rd_firstRelfilenodeSubid. Amend the specification of rd_createSubid such that the field is zero when a new rel has an old rd_node. Make relcache.c retain entries for certain dropped relations until end of transaction. Bump XLOG_PAGE_MAGIC, since this introduces XLOG_GIST_ASSIGN_LSN. Future servers accept older WAL, so this bump is discretionary. Kyotaro Horiguchi, reviewed (in earlier, similar versions) by Robert Haas. Heikki Linnakangas and Michael Paquier implemented earlier designs that materially clarified the problem. Reviewed, in earlier designs, by Andrew Dunstan, Andres Freund, Alvaro Herrera, Tom Lane, Fujii Masao, and Simon Riggs. Reported by Martijn van Oosterhout. Discussion: https://postgr.es/m/20150702220524.GA9392@svana.org
2020-04-04 21:25:34 +02:00
/* TABLE_INSERT_SKIP_WAL was 0x0001; RelationNeedsWAL() now governs */
tableam: Add tuple_{insert, delete, update, lock} and use. This adds new, required, table AM callbacks for insert/delete/update and lock_tuple. To be able to reasonably use those, the EvalPlanQual mechanism had to be adapted, moving more logic into the AM. Previously both delete/update/lock call-sites and the EPQ mechanism had to have awareness of the specific tuple format to be able to fetch the latest version of a tuple. Obviously that needs to be abstracted away. To do so, move the logic that find the latest row version into the AM. lock_tuple has a new flag argument, TUPLE_LOCK_FLAG_FIND_LAST_VERSION, that forces it to lock the last version, rather than the current one. It'd have been possible to do so via a separate callback as well, but finding the last version usually also necessitates locking the newest version, making it sensible to combine the two. This replaces the previous use of EvalPlanQualFetch(). Additionally HeapTupleUpdated, which previously signaled either a concurrent update or delete, is now split into two, to avoid callers needing AM specific knowledge to differentiate. The move of finding the latest row version into tuple_lock means that encountering a row concurrently moved into another partition will now raise an error about "tuple to be locked" rather than "tuple to be updated/deleted" - which is accurate, as that always happens when locking rows. While possible slightly less helpful for users, it seems like an acceptable trade-off. As part of this commit HTSU_Result has been renamed to TM_Result, and its members been expanded to differentiated between updating and deleting. HeapUpdateFailureData has been renamed to TM_FailureData. The interface to speculative insertion is changed so nodeModifyTable.c does not have to set the speculative token itself anymore. Instead there's a version of tuple_insert, tuple_insert_speculative, that performs the speculative insertion (without requiring a flag to signal that fact), and the speculative insertion is either made permanent with table_complete_speculative(succeeded = true) or aborted with succeeded = false). Note that multi_insert is not yet routed through tableam, nor is COPY. Changing multi_insert requires changes to copy.c that are large enough to better be done separately. Similarly, although simpler, CREATE TABLE AS and CREATE MATERIALIZED VIEW are also only going to be adjusted in a later commit. Author: Andres Freund and Haribabu Kommi Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20190313003903.nwvrxi7rw3ywhdel@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-24 03:55:57 +01:00
#define TABLE_INSERT_SKIP_FSM 0x0002
#define TABLE_INSERT_FROZEN 0x0004
#define TABLE_INSERT_NO_LOGICAL 0x0008
tableam: Rename wrapper functions to match callback names. Some of the wrapper functions didn't match the callback names. Many of them due to staying "consistent" with historic naming of the wrapped functionality. We decided that for most cases it's more important to be for tableam to be consistent going forward, than with the past. The one exception is beginscan/endscan/... because it'd have looked odd to have systable_beginscan/endscan/... with a different naming scheme, and changing the systable_* APIs would have caused way too much churn (including breaking a lot of external users). Author: Ashwin Agrawal, with some small additions by Andres Freund Reviewed-By: Andres Freund Discussion: https://postgr.es/m/CALfoeiugyrXZfX7n0ORCa4L-m834dzmaE8eFdbNR6PMpetU4Ww@mail.gmail.com
2019-05-24 01:25:48 +02:00
/* flag bits for table_tuple_lock */
tableam: Add tuple_{insert, delete, update, lock} and use. This adds new, required, table AM callbacks for insert/delete/update and lock_tuple. To be able to reasonably use those, the EvalPlanQual mechanism had to be adapted, moving more logic into the AM. Previously both delete/update/lock call-sites and the EPQ mechanism had to have awareness of the specific tuple format to be able to fetch the latest version of a tuple. Obviously that needs to be abstracted away. To do so, move the logic that find the latest row version into the AM. lock_tuple has a new flag argument, TUPLE_LOCK_FLAG_FIND_LAST_VERSION, that forces it to lock the last version, rather than the current one. It'd have been possible to do so via a separate callback as well, but finding the last version usually also necessitates locking the newest version, making it sensible to combine the two. This replaces the previous use of EvalPlanQualFetch(). Additionally HeapTupleUpdated, which previously signaled either a concurrent update or delete, is now split into two, to avoid callers needing AM specific knowledge to differentiate. The move of finding the latest row version into tuple_lock means that encountering a row concurrently moved into another partition will now raise an error about "tuple to be locked" rather than "tuple to be updated/deleted" - which is accurate, as that always happens when locking rows. While possible slightly less helpful for users, it seems like an acceptable trade-off. As part of this commit HTSU_Result has been renamed to TM_Result, and its members been expanded to differentiated between updating and deleting. HeapUpdateFailureData has been renamed to TM_FailureData. The interface to speculative insertion is changed so nodeModifyTable.c does not have to set the speculative token itself anymore. Instead there's a version of tuple_insert, tuple_insert_speculative, that performs the speculative insertion (without requiring a flag to signal that fact), and the speculative insertion is either made permanent with table_complete_speculative(succeeded = true) or aborted with succeeded = false). Note that multi_insert is not yet routed through tableam, nor is COPY. Changing multi_insert requires changes to copy.c that are large enough to better be done separately. Similarly, although simpler, CREATE TABLE AS and CREATE MATERIALIZED VIEW are also only going to be adjusted in a later commit. Author: Andres Freund and Haribabu Kommi Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20190313003903.nwvrxi7rw3ywhdel@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-24 03:55:57 +01:00
/* Follow tuples whose update is in progress if lock modes don't conflict */
#define TUPLE_LOCK_FLAG_LOCK_UPDATE_IN_PROGRESS (1 << 0)
tableam: comment and formatting fixes. Author: Heikki Linnakangas Discussion: https://postgr.es/m/9a7fb9cc-2419-5db7-8840-ddc10c93f122@iki.fi
2019-04-08 19:25:16 +02:00
/* Follow update chain and lock latest version of tuple */
tableam: Add tuple_{insert, delete, update, lock} and use. This adds new, required, table AM callbacks for insert/delete/update and lock_tuple. To be able to reasonably use those, the EvalPlanQual mechanism had to be adapted, moving more logic into the AM. Previously both delete/update/lock call-sites and the EPQ mechanism had to have awareness of the specific tuple format to be able to fetch the latest version of a tuple. Obviously that needs to be abstracted away. To do so, move the logic that find the latest row version into the AM. lock_tuple has a new flag argument, TUPLE_LOCK_FLAG_FIND_LAST_VERSION, that forces it to lock the last version, rather than the current one. It'd have been possible to do so via a separate callback as well, but finding the last version usually also necessitates locking the newest version, making it sensible to combine the two. This replaces the previous use of EvalPlanQualFetch(). Additionally HeapTupleUpdated, which previously signaled either a concurrent update or delete, is now split into two, to avoid callers needing AM specific knowledge to differentiate. The move of finding the latest row version into tuple_lock means that encountering a row concurrently moved into another partition will now raise an error about "tuple to be locked" rather than "tuple to be updated/deleted" - which is accurate, as that always happens when locking rows. While possible slightly less helpful for users, it seems like an acceptable trade-off. As part of this commit HTSU_Result has been renamed to TM_Result, and its members been expanded to differentiated between updating and deleting. HeapUpdateFailureData has been renamed to TM_FailureData. The interface to speculative insertion is changed so nodeModifyTable.c does not have to set the speculative token itself anymore. Instead there's a version of tuple_insert, tuple_insert_speculative, that performs the speculative insertion (without requiring a flag to signal that fact), and the speculative insertion is either made permanent with table_complete_speculative(succeeded = true) or aborted with succeeded = false). Note that multi_insert is not yet routed through tableam, nor is COPY. Changing multi_insert requires changes to copy.c that are large enough to better be done separately. Similarly, although simpler, CREATE TABLE AS and CREATE MATERIALIZED VIEW are also only going to be adjusted in a later commit. Author: Andres Freund and Haribabu Kommi Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20190313003903.nwvrxi7rw3ywhdel@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-24 03:55:57 +01:00
#define TUPLE_LOCK_FLAG_FIND_LAST_VERSION (1 << 1)
tableam: Support for an index build's initial table scan(s). To support building indexes over tables of different AMs, the scans to do so need to be routed through the table AM. While moving a fair amount of code, nearly all the changes are just moving code to below a callback. Currently the range based interface wouldn't make much sense for non block based table AMs. But that seems aceptable for now. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-28 03:59:06 +01:00
/* Typedef for callback function for table_index_build_scan */
typedef void (*IndexBuildCallback) (Relation index,
Pass ItemPointer not HeapTuple to IndexBuildCallback. Not all AMs use HeapTuples internally, making it inconvenient to pass a HeapTuple. As the index callbacks really only need the TID, not the full tuple, modify callback to only take ItemPointer. Author: Ashwin Agrawal Reviewed-By: Andres Freund Discussion: https://postgr.es/m/CALfoeis6=8ehuR=VNtHvj3z16cYfCwPdTcpaxU+sfSUJ5QgR3g@mail.gmail.com
2019-11-08 09:44:52 +01:00
ItemPointer tid,
tableam: Support for an index build's initial table scan(s). To support building indexes over tables of different AMs, the scans to do so need to be routed through the table AM. While moving a fair amount of code, nearly all the changes are just moving code to below a callback. Currently the range based interface wouldn't make much sense for non block based table AMs. But that seems aceptable for now. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-28 03:59:06 +01:00
Datum *values,
bool *isnull,
bool tupleIsAlive,
void *state);
tableam: introduce table AM infrastructure. This introduces the concept of table access methods, i.e. CREATE ACCESS METHOD ... TYPE TABLE and CREATE TABLE ... USING (storage-engine). No table access functionality is delegated to table AMs as of this commit, that'll be done in following commits. Subsequent commits will incrementally abstract table access functionality to be routed through table access methods. That change is too large to be reviewed & committed at once, so it'll be done incrementally. Docs will be updated at the end, as adding them incrementally would likely make them less coherent, and definitely is a lot more work, without a lot of benefit. Table access methods are specified similar to index access methods, i.e. pg_am.amhandler returns, as INTERNAL, a pointer to a struct with callbacks. In contrast to index AMs that struct needs to live as long as a backend, typically that's achieved by just returning a pointer to a constant struct. Psql's \d+ now displays a table's access method. That can be disabled with HIDE_TABLEAM=true, which is mainly useful so regression tests can be run against different AMs. It's quite possible that this behaviour still needs to be fine tuned. For now it's not allowed to set a table AM for a partitioned table, as we've not resolved how partitions would inherit that. Disallowing allows us to introduce, if we decide that's the way forward, such a behaviour without a compatibility break. Catversion bumped, to add the heap table AM and references to it. Author: Haribabu Kommi, Andres Freund, Alvaro Herrera, Dimitri Golgov and others Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql https://postgr.es/m/20190107235616.6lur25ph22u5u5av@alap3.anarazel.de https://postgr.es/m/20190304234700.w5tmhducs5wxgzls@alap3.anarazel.de
2019-03-06 18:54:38 +01:00
/*
* API struct for a table AM. Note this must be allocated in a
* server-lifetime manner, typically as a static const struct, which then gets
* returned by FormData_pg_am.amhandler.
tableam: Add and use scan APIs. Too allow table accesses to be not directly dependent on heap, several new abstractions are needed. Specifically: 1) Heap scans need to be generalized into table scans. Do this by introducing TableScanDesc, which will be the "base class" for individual AMs. This contains the AM independent fields from HeapScanDesc. The previous heap_{beginscan,rescan,endscan} et al. have been replaced with a table_ version. There's no direct replacement for heap_getnext(), as that returned a HeapTuple, which is undesirable for a other AMs. Instead there's table_scan_getnextslot(). But note that heap_getnext() lives on, it's still used widely to access catalog tables. This is achieved by new scan_begin, scan_end, scan_rescan, scan_getnextslot callbacks. 2) The portion of parallel scans that's shared between backends need to be able to do so without the user doing per-AM work. To achieve that new parallelscan_{estimate, initialize, reinitialize} callbacks are introduced, which operate on a new ParallelTableScanDesc, which again can be subclassed by AMs. As it is likely that several AMs are going to be block oriented, block oriented callbacks that can be shared between such AMs are provided and used by heap. table_block_parallelscan_{estimate, intiialize, reinitialize} as callbacks, and table_block_parallelscan_{nextpage, init} for use in AMs. These operate on a ParallelBlockTableScanDesc. 3) Index scans need to be able to access tables to return a tuple, and there needs to be state across individual accesses to the heap to store state like buffers. That's now handled by introducing a sort-of-scan IndexFetchTable, which again is intended to be subclassed by individual AMs (for heap IndexFetchHeap). The relevant callbacks for an AM are index_fetch_{end, begin, reset} to create the necessary state, and index_fetch_tuple to retrieve an indexed tuple. Note that index_fetch_tuple implementations need to be smarter than just blindly fetching the tuples for AMs that have optimizations similar to heap's HOT - the currently alive tuple in the update chain needs to be fetched if appropriate. Similar to table_scan_getnextslot(), it's undesirable to continue to return HeapTuples. Thus index_fetch_heap (might want to rename that later) now accepts a slot as an argument. Core code doesn't have a lot of call sites performing index scans without going through the systable_* API (in contrast to loads of heap_getnext calls and working directly with HeapTuples). Index scans now store the result of a search in IndexScanDesc->xs_heaptid, rather than xs_ctup->t_self. As the target is not generally a HeapTuple anymore that seems cleaner. To be able to sensible adapt code to use the above, two further callbacks have been introduced: a) slot_callbacks returns a TupleTableSlotOps* suitable for creating slots capable of holding a tuple of the AMs type. table_slot_callbacks() and table_slot_create() are based upon that, but have additional logic to deal with views, foreign tables, etc. While this change could have been done separately, nearly all the call sites that needed to be adapted for the rest of this commit also would have been needed to be adapted for table_slot_callbacks(), making separation not worthwhile. b) tuple_satisfies_snapshot checks whether the tuple in a slot is currently visible according to a snapshot. That's required as a few places now don't have a buffer + HeapTuple around, but a slot (which in heap's case internally has that information). Additionally a few infrastructure changes were needed: I) SysScanDesc, as used by systable_{beginscan, getnext} et al. now internally uses a slot to keep track of tuples. While systable_getnext() still returns HeapTuples, and will so for the foreseeable future, the index API (see 1) above) now only deals with slots. The remainder, and largest part, of this commit is then adjusting all scans in postgres to use the new APIs. Author: Andres Freund, Haribabu Kommi, Alvaro Herrera Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-11 20:46:41 +01:00
*
tableam: comment and formatting fixes. Author: Heikki Linnakangas Discussion: https://postgr.es/m/9a7fb9cc-2419-5db7-8840-ddc10c93f122@iki.fi
2019-04-08 19:25:16 +02:00
* In most cases it's not appropriate to call the callbacks directly, use the
* table_* wrapper functions instead.
tableam: Add and use scan APIs. Too allow table accesses to be not directly dependent on heap, several new abstractions are needed. Specifically: 1) Heap scans need to be generalized into table scans. Do this by introducing TableScanDesc, which will be the "base class" for individual AMs. This contains the AM independent fields from HeapScanDesc. The previous heap_{beginscan,rescan,endscan} et al. have been replaced with a table_ version. There's no direct replacement for heap_getnext(), as that returned a HeapTuple, which is undesirable for a other AMs. Instead there's table_scan_getnextslot(). But note that heap_getnext() lives on, it's still used widely to access catalog tables. This is achieved by new scan_begin, scan_end, scan_rescan, scan_getnextslot callbacks. 2) The portion of parallel scans that's shared between backends need to be able to do so without the user doing per-AM work. To achieve that new parallelscan_{estimate, initialize, reinitialize} callbacks are introduced, which operate on a new ParallelTableScanDesc, which again can be subclassed by AMs. As it is likely that several AMs are going to be block oriented, block oriented callbacks that can be shared between such AMs are provided and used by heap. table_block_parallelscan_{estimate, intiialize, reinitialize} as callbacks, and table_block_parallelscan_{nextpage, init} for use in AMs. These operate on a ParallelBlockTableScanDesc. 3) Index scans need to be able to access tables to return a tuple, and there needs to be state across individual accesses to the heap to store state like buffers. That's now handled by introducing a sort-of-scan IndexFetchTable, which again is intended to be subclassed by individual AMs (for heap IndexFetchHeap). The relevant callbacks for an AM are index_fetch_{end, begin, reset} to create the necessary state, and index_fetch_tuple to retrieve an indexed tuple. Note that index_fetch_tuple implementations need to be smarter than just blindly fetching the tuples for AMs that have optimizations similar to heap's HOT - the currently alive tuple in the update chain needs to be fetched if appropriate. Similar to table_scan_getnextslot(), it's undesirable to continue to return HeapTuples. Thus index_fetch_heap (might want to rename that later) now accepts a slot as an argument. Core code doesn't have a lot of call sites performing index scans without going through the systable_* API (in contrast to loads of heap_getnext calls and working directly with HeapTuples). Index scans now store the result of a search in IndexScanDesc->xs_heaptid, rather than xs_ctup->t_self. As the target is not generally a HeapTuple anymore that seems cleaner. To be able to sensible adapt code to use the above, two further callbacks have been introduced: a) slot_callbacks returns a TupleTableSlotOps* suitable for creating slots capable of holding a tuple of the AMs type. table_slot_callbacks() and table_slot_create() are based upon that, but have additional logic to deal with views, foreign tables, etc. While this change could have been done separately, nearly all the call sites that needed to be adapted for the rest of this commit also would have been needed to be adapted for table_slot_callbacks(), making separation not worthwhile. b) tuple_satisfies_snapshot checks whether the tuple in a slot is currently visible according to a snapshot. That's required as a few places now don't have a buffer + HeapTuple around, but a slot (which in heap's case internally has that information). Additionally a few infrastructure changes were needed: I) SysScanDesc, as used by systable_{beginscan, getnext} et al. now internally uses a slot to keep track of tuples. While systable_getnext() still returns HeapTuples, and will so for the foreseeable future, the index API (see 1) above) now only deals with slots. The remainder, and largest part, of this commit is then adjusting all scans in postgres to use the new APIs. Author: Andres Freund, Haribabu Kommi, Alvaro Herrera Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-11 20:46:41 +01:00
*
* GetTableAmRoutine() asserts that required callbacks are filled in, remember
* to update when adding a callback.
tableam: introduce table AM infrastructure. This introduces the concept of table access methods, i.e. CREATE ACCESS METHOD ... TYPE TABLE and CREATE TABLE ... USING (storage-engine). No table access functionality is delegated to table AMs as of this commit, that'll be done in following commits. Subsequent commits will incrementally abstract table access functionality to be routed through table access methods. That change is too large to be reviewed & committed at once, so it'll be done incrementally. Docs will be updated at the end, as adding them incrementally would likely make them less coherent, and definitely is a lot more work, without a lot of benefit. Table access methods are specified similar to index access methods, i.e. pg_am.amhandler returns, as INTERNAL, a pointer to a struct with callbacks. In contrast to index AMs that struct needs to live as long as a backend, typically that's achieved by just returning a pointer to a constant struct. Psql's \d+ now displays a table's access method. That can be disabled with HIDE_TABLEAM=true, which is mainly useful so regression tests can be run against different AMs. It's quite possible that this behaviour still needs to be fine tuned. For now it's not allowed to set a table AM for a partitioned table, as we've not resolved how partitions would inherit that. Disallowing allows us to introduce, if we decide that's the way forward, such a behaviour without a compatibility break. Catversion bumped, to add the heap table AM and references to it. Author: Haribabu Kommi, Andres Freund, Alvaro Herrera, Dimitri Golgov and others Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql https://postgr.es/m/20190107235616.6lur25ph22u5u5av@alap3.anarazel.de https://postgr.es/m/20190304234700.w5tmhducs5wxgzls@alap3.anarazel.de
2019-03-06 18:54:38 +01:00
*/
typedef struct TableAmRoutine
{
/* this must be set to T_TableAmRoutine */
NodeTag type;
tableam: Add and use scan APIs. Too allow table accesses to be not directly dependent on heap, several new abstractions are needed. Specifically: 1) Heap scans need to be generalized into table scans. Do this by introducing TableScanDesc, which will be the "base class" for individual AMs. This contains the AM independent fields from HeapScanDesc. The previous heap_{beginscan,rescan,endscan} et al. have been replaced with a table_ version. There's no direct replacement for heap_getnext(), as that returned a HeapTuple, which is undesirable for a other AMs. Instead there's table_scan_getnextslot(). But note that heap_getnext() lives on, it's still used widely to access catalog tables. This is achieved by new scan_begin, scan_end, scan_rescan, scan_getnextslot callbacks. 2) The portion of parallel scans that's shared between backends need to be able to do so without the user doing per-AM work. To achieve that new parallelscan_{estimate, initialize, reinitialize} callbacks are introduced, which operate on a new ParallelTableScanDesc, which again can be subclassed by AMs. As it is likely that several AMs are going to be block oriented, block oriented callbacks that can be shared between such AMs are provided and used by heap. table_block_parallelscan_{estimate, intiialize, reinitialize} as callbacks, and table_block_parallelscan_{nextpage, init} for use in AMs. These operate on a ParallelBlockTableScanDesc. 3) Index scans need to be able to access tables to return a tuple, and there needs to be state across individual accesses to the heap to store state like buffers. That's now handled by introducing a sort-of-scan IndexFetchTable, which again is intended to be subclassed by individual AMs (for heap IndexFetchHeap). The relevant callbacks for an AM are index_fetch_{end, begin, reset} to create the necessary state, and index_fetch_tuple to retrieve an indexed tuple. Note that index_fetch_tuple implementations need to be smarter than just blindly fetching the tuples for AMs that have optimizations similar to heap's HOT - the currently alive tuple in the update chain needs to be fetched if appropriate. Similar to table_scan_getnextslot(), it's undesirable to continue to return HeapTuples. Thus index_fetch_heap (might want to rename that later) now accepts a slot as an argument. Core code doesn't have a lot of call sites performing index scans without going through the systable_* API (in contrast to loads of heap_getnext calls and working directly with HeapTuples). Index scans now store the result of a search in IndexScanDesc->xs_heaptid, rather than xs_ctup->t_self. As the target is not generally a HeapTuple anymore that seems cleaner. To be able to sensible adapt code to use the above, two further callbacks have been introduced: a) slot_callbacks returns a TupleTableSlotOps* suitable for creating slots capable of holding a tuple of the AMs type. table_slot_callbacks() and table_slot_create() are based upon that, but have additional logic to deal with views, foreign tables, etc. While this change could have been done separately, nearly all the call sites that needed to be adapted for the rest of this commit also would have been needed to be adapted for table_slot_callbacks(), making separation not worthwhile. b) tuple_satisfies_snapshot checks whether the tuple in a slot is currently visible according to a snapshot. That's required as a few places now don't have a buffer + HeapTuple around, but a slot (which in heap's case internally has that information). Additionally a few infrastructure changes were needed: I) SysScanDesc, as used by systable_{beginscan, getnext} et al. now internally uses a slot to keep track of tuples. While systable_getnext() still returns HeapTuples, and will so for the foreseeable future, the index API (see 1) above) now only deals with slots. The remainder, and largest part, of this commit is then adjusting all scans in postgres to use the new APIs. Author: Andres Freund, Haribabu Kommi, Alvaro Herrera Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-11 20:46:41 +01:00
/* ------------------------------------------------------------------------
* Slot related callbacks.
* ------------------------------------------------------------------------
*/
/*
* Return slot implementation suitable for storing a tuple of this AM.
*/
const TupleTableSlotOps *(*slot_callbacks) (Relation rel);
/* ------------------------------------------------------------------------
* Table scan callbacks.
* ------------------------------------------------------------------------
*/
/*
* Start a scan of `rel`. The callback has to return a TableScanDesc,
* which will typically be embedded in a larger, AM specific, struct.
*
* If nkeys != 0, the results need to be filtered by those scan keys.
*
* pscan, if not NULL, will have already been initialized with
* parallelscan_initialize(), and has to be for the same relation. Will
* only be set coming from table_beginscan_parallel().
*
Don't to predicate lock for analyze scans, refactor scan option passing. Before this commit, when ANALYZE was run on a table and serializable was used (either by virtue of an explicit BEGIN TRANSACTION ISOLATION LEVEL SERIALIZABLE, or default_transaction_isolation being set to serializable) a null pointer dereference lead to a crash. The analyze scan doesn't need a snapshot (nor predicate locking), but before this commit a scan only contained information about being a bitmap or sample scan. Refactor the option passing to the scan_begin callback to use a bitmask instead. Alternatively we could have added a new boolean parameter, but that seems harder to read. Even before this issue various people (Heikki, Tom, Robert) suggested doing so. These changes don't change the scan APIs outside of tableam. The flags argument could be exposed, it's not necessary to fix this problem. Also the wrapper table_beginscan* functions encapsulate most of that complexity. After these changes fixing the bug is trivial, just don't acquire predicate lock for analyze style scans. That was already done for bitmap heap scans. Add an assert that a snapshot is passed when acquiring the predicate lock, so this kind of bug doesn't require running with serializable. Also add a comment about sample scans currently requiring predicate locking the entire relation, that previously wasn't remarked upon. Reported-By: Joe Wildish Author: Andres Freund Discussion: https://postgr.es/m/4EA80A20-E9BF-49F1-9F01-5B66CAB21453@elusive.cx https://postgr.es/m/20190411164947.nkii4gaeilt4bui7@alap3.anarazel.de https://postgr.es/m/20190518203102.g7peu2fianukjuxm@alap3.anarazel.de
2019-05-20 00:10:28 +02:00
* `flags` is a bitmask indicating the type of scan (ScanOptions's
* SO_TYPE_*, currently only one may be specified), options controlling
* the scan's behaviour (ScanOptions's SO_ALLOW_*, several may be
* specified, an AM may ignore unsupported ones) and whether the snapshot
* needs to be deallocated at scan_end (ScanOptions's SO_TEMP_SNAPSHOT).
tableam: Add and use scan APIs. Too allow table accesses to be not directly dependent on heap, several new abstractions are needed. Specifically: 1) Heap scans need to be generalized into table scans. Do this by introducing TableScanDesc, which will be the "base class" for individual AMs. This contains the AM independent fields from HeapScanDesc. The previous heap_{beginscan,rescan,endscan} et al. have been replaced with a table_ version. There's no direct replacement for heap_getnext(), as that returned a HeapTuple, which is undesirable for a other AMs. Instead there's table_scan_getnextslot(). But note that heap_getnext() lives on, it's still used widely to access catalog tables. This is achieved by new scan_begin, scan_end, scan_rescan, scan_getnextslot callbacks. 2) The portion of parallel scans that's shared between backends need to be able to do so without the user doing per-AM work. To achieve that new parallelscan_{estimate, initialize, reinitialize} callbacks are introduced, which operate on a new ParallelTableScanDesc, which again can be subclassed by AMs. As it is likely that several AMs are going to be block oriented, block oriented callbacks that can be shared between such AMs are provided and used by heap. table_block_parallelscan_{estimate, intiialize, reinitialize} as callbacks, and table_block_parallelscan_{nextpage, init} for use in AMs. These operate on a ParallelBlockTableScanDesc. 3) Index scans need to be able to access tables to return a tuple, and there needs to be state across individual accesses to the heap to store state like buffers. That's now handled by introducing a sort-of-scan IndexFetchTable, which again is intended to be subclassed by individual AMs (for heap IndexFetchHeap). The relevant callbacks for an AM are index_fetch_{end, begin, reset} to create the necessary state, and index_fetch_tuple to retrieve an indexed tuple. Note that index_fetch_tuple implementations need to be smarter than just blindly fetching the tuples for AMs that have optimizations similar to heap's HOT - the currently alive tuple in the update chain needs to be fetched if appropriate. Similar to table_scan_getnextslot(), it's undesirable to continue to return HeapTuples. Thus index_fetch_heap (might want to rename that later) now accepts a slot as an argument. Core code doesn't have a lot of call sites performing index scans without going through the systable_* API (in contrast to loads of heap_getnext calls and working directly with HeapTuples). Index scans now store the result of a search in IndexScanDesc->xs_heaptid, rather than xs_ctup->t_self. As the target is not generally a HeapTuple anymore that seems cleaner. To be able to sensible adapt code to use the above, two further callbacks have been introduced: a) slot_callbacks returns a TupleTableSlotOps* suitable for creating slots capable of holding a tuple of the AMs type. table_slot_callbacks() and table_slot_create() are based upon that, but have additional logic to deal with views, foreign tables, etc. While this change could have been done separately, nearly all the call sites that needed to be adapted for the rest of this commit also would have been needed to be adapted for table_slot_callbacks(), making separation not worthwhile. b) tuple_satisfies_snapshot checks whether the tuple in a slot is currently visible according to a snapshot. That's required as a few places now don't have a buffer + HeapTuple around, but a slot (which in heap's case internally has that information). Additionally a few infrastructure changes were needed: I) SysScanDesc, as used by systable_{beginscan, getnext} et al. now internally uses a slot to keep track of tuples. While systable_getnext() still returns HeapTuples, and will so for the foreseeable future, the index API (see 1) above) now only deals with slots. The remainder, and largest part, of this commit is then adjusting all scans in postgres to use the new APIs. Author: Andres Freund, Haribabu Kommi, Alvaro Herrera Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-11 20:46:41 +01:00
*/
TableScanDesc (*scan_begin) (Relation rel,
Snapshot snapshot,
int nkeys, struct ScanKeyData *key,
ParallelTableScanDesc pscan,
Don't to predicate lock for analyze scans, refactor scan option passing. Before this commit, when ANALYZE was run on a table and serializable was used (either by virtue of an explicit BEGIN TRANSACTION ISOLATION LEVEL SERIALIZABLE, or default_transaction_isolation being set to serializable) a null pointer dereference lead to a crash. The analyze scan doesn't need a snapshot (nor predicate locking), but before this commit a scan only contained information about being a bitmap or sample scan. Refactor the option passing to the scan_begin callback to use a bitmask instead. Alternatively we could have added a new boolean parameter, but that seems harder to read. Even before this issue various people (Heikki, Tom, Robert) suggested doing so. These changes don't change the scan APIs outside of tableam. The flags argument could be exposed, it's not necessary to fix this problem. Also the wrapper table_beginscan* functions encapsulate most of that complexity. After these changes fixing the bug is trivial, just don't acquire predicate lock for analyze style scans. That was already done for bitmap heap scans. Add an assert that a snapshot is passed when acquiring the predicate lock, so this kind of bug doesn't require running with serializable. Also add a comment about sample scans currently requiring predicate locking the entire relation, that previously wasn't remarked upon. Reported-By: Joe Wildish Author: Andres Freund Discussion: https://postgr.es/m/4EA80A20-E9BF-49F1-9F01-5B66CAB21453@elusive.cx https://postgr.es/m/20190411164947.nkii4gaeilt4bui7@alap3.anarazel.de https://postgr.es/m/20190518203102.g7peu2fianukjuxm@alap3.anarazel.de
2019-05-20 00:10:28 +02:00
uint32 flags);
tableam: Add and use scan APIs. Too allow table accesses to be not directly dependent on heap, several new abstractions are needed. Specifically: 1) Heap scans need to be generalized into table scans. Do this by introducing TableScanDesc, which will be the "base class" for individual AMs. This contains the AM independent fields from HeapScanDesc. The previous heap_{beginscan,rescan,endscan} et al. have been replaced with a table_ version. There's no direct replacement for heap_getnext(), as that returned a HeapTuple, which is undesirable for a other AMs. Instead there's table_scan_getnextslot(). But note that heap_getnext() lives on, it's still used widely to access catalog tables. This is achieved by new scan_begin, scan_end, scan_rescan, scan_getnextslot callbacks. 2) The portion of parallel scans that's shared between backends need to be able to do so without the user doing per-AM work. To achieve that new parallelscan_{estimate, initialize, reinitialize} callbacks are introduced, which operate on a new ParallelTableScanDesc, which again can be subclassed by AMs. As it is likely that several AMs are going to be block oriented, block oriented callbacks that can be shared between such AMs are provided and used by heap. table_block_parallelscan_{estimate, intiialize, reinitialize} as callbacks, and table_block_parallelscan_{nextpage, init} for use in AMs. These operate on a ParallelBlockTableScanDesc. 3) Index scans need to be able to access tables to return a tuple, and there needs to be state across individual accesses to the heap to store state like buffers. That's now handled by introducing a sort-of-scan IndexFetchTable, which again is intended to be subclassed by individual AMs (for heap IndexFetchHeap). The relevant callbacks for an AM are index_fetch_{end, begin, reset} to create the necessary state, and index_fetch_tuple to retrieve an indexed tuple. Note that index_fetch_tuple implementations need to be smarter than just blindly fetching the tuples for AMs that have optimizations similar to heap's HOT - the currently alive tuple in the update chain needs to be fetched if appropriate. Similar to table_scan_getnextslot(), it's undesirable to continue to return HeapTuples. Thus index_fetch_heap (might want to rename that later) now accepts a slot as an argument. Core code doesn't have a lot of call sites performing index scans without going through the systable_* API (in contrast to loads of heap_getnext calls and working directly with HeapTuples). Index scans now store the result of a search in IndexScanDesc->xs_heaptid, rather than xs_ctup->t_self. As the target is not generally a HeapTuple anymore that seems cleaner. To be able to sensible adapt code to use the above, two further callbacks have been introduced: a) slot_callbacks returns a TupleTableSlotOps* suitable for creating slots capable of holding a tuple of the AMs type. table_slot_callbacks() and table_slot_create() are based upon that, but have additional logic to deal with views, foreign tables, etc. While this change could have been done separately, nearly all the call sites that needed to be adapted for the rest of this commit also would have been needed to be adapted for table_slot_callbacks(), making separation not worthwhile. b) tuple_satisfies_snapshot checks whether the tuple in a slot is currently visible according to a snapshot. That's required as a few places now don't have a buffer + HeapTuple around, but a slot (which in heap's case internally has that information). Additionally a few infrastructure changes were needed: I) SysScanDesc, as used by systable_{beginscan, getnext} et al. now internally uses a slot to keep track of tuples. While systable_getnext() still returns HeapTuples, and will so for the foreseeable future, the index API (see 1) above) now only deals with slots. The remainder, and largest part, of this commit is then adjusting all scans in postgres to use the new APIs. Author: Andres Freund, Haribabu Kommi, Alvaro Herrera Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-11 20:46:41 +01:00
/*
* Release resources and deallocate scan. If TableScanDesc.temp_snap,
* TableScanDesc.rs_snapshot needs to be unregistered.
*/
void (*scan_end) (TableScanDesc scan);
/*
tableam: comment and formatting fixes. Author: Heikki Linnakangas Discussion: https://postgr.es/m/9a7fb9cc-2419-5db7-8840-ddc10c93f122@iki.fi
2019-04-08 19:25:16 +02:00
* Restart relation scan. If set_params is set to true, allow_{strat,
tableam: Add and use scan APIs. Too allow table accesses to be not directly dependent on heap, several new abstractions are needed. Specifically: 1) Heap scans need to be generalized into table scans. Do this by introducing TableScanDesc, which will be the "base class" for individual AMs. This contains the AM independent fields from HeapScanDesc. The previous heap_{beginscan,rescan,endscan} et al. have been replaced with a table_ version. There's no direct replacement for heap_getnext(), as that returned a HeapTuple, which is undesirable for a other AMs. Instead there's table_scan_getnextslot(). But note that heap_getnext() lives on, it's still used widely to access catalog tables. This is achieved by new scan_begin, scan_end, scan_rescan, scan_getnextslot callbacks. 2) The portion of parallel scans that's shared between backends need to be able to do so without the user doing per-AM work. To achieve that new parallelscan_{estimate, initialize, reinitialize} callbacks are introduced, which operate on a new ParallelTableScanDesc, which again can be subclassed by AMs. As it is likely that several AMs are going to be block oriented, block oriented callbacks that can be shared between such AMs are provided and used by heap. table_block_parallelscan_{estimate, intiialize, reinitialize} as callbacks, and table_block_parallelscan_{nextpage, init} for use in AMs. These operate on a ParallelBlockTableScanDesc. 3) Index scans need to be able to access tables to return a tuple, and there needs to be state across individual accesses to the heap to store state like buffers. That's now handled by introducing a sort-of-scan IndexFetchTable, which again is intended to be subclassed by individual AMs (for heap IndexFetchHeap). The relevant callbacks for an AM are index_fetch_{end, begin, reset} to create the necessary state, and index_fetch_tuple to retrieve an indexed tuple. Note that index_fetch_tuple implementations need to be smarter than just blindly fetching the tuples for AMs that have optimizations similar to heap's HOT - the currently alive tuple in the update chain needs to be fetched if appropriate. Similar to table_scan_getnextslot(), it's undesirable to continue to return HeapTuples. Thus index_fetch_heap (might want to rename that later) now accepts a slot as an argument. Core code doesn't have a lot of call sites performing index scans without going through the systable_* API (in contrast to loads of heap_getnext calls and working directly with HeapTuples). Index scans now store the result of a search in IndexScanDesc->xs_heaptid, rather than xs_ctup->t_self. As the target is not generally a HeapTuple anymore that seems cleaner. To be able to sensible adapt code to use the above, two further callbacks have been introduced: a) slot_callbacks returns a TupleTableSlotOps* suitable for creating slots capable of holding a tuple of the AMs type. table_slot_callbacks() and table_slot_create() are based upon that, but have additional logic to deal with views, foreign tables, etc. While this change could have been done separately, nearly all the call sites that needed to be adapted for the rest of this commit also would have been needed to be adapted for table_slot_callbacks(), making separation not worthwhile. b) tuple_satisfies_snapshot checks whether the tuple in a slot is currently visible according to a snapshot. That's required as a few places now don't have a buffer + HeapTuple around, but a slot (which in heap's case internally has that information). Additionally a few infrastructure changes were needed: I) SysScanDesc, as used by systable_{beginscan, getnext} et al. now internally uses a slot to keep track of tuples. While systable_getnext() still returns HeapTuples, and will so for the foreseeable future, the index API (see 1) above) now only deals with slots. The remainder, and largest part, of this commit is then adjusting all scans in postgres to use the new APIs. Author: Andres Freund, Haribabu Kommi, Alvaro Herrera Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-11 20:46:41 +01:00
* sync, pagemode} (see scan_begin) changes should be taken into account.
*/
tableam: Formatting and other minor cleanups. The superflous heapam_xlog.h includes were reported by Peter Geoghegan.
2019-03-31 05:13:56 +02:00
void (*scan_rescan) (TableScanDesc scan, struct ScanKeyData *key,
bool set_params, bool allow_strat,
bool allow_sync, bool allow_pagemode);
tableam: Add and use scan APIs. Too allow table accesses to be not directly dependent on heap, several new abstractions are needed. Specifically: 1) Heap scans need to be generalized into table scans. Do this by introducing TableScanDesc, which will be the "base class" for individual AMs. This contains the AM independent fields from HeapScanDesc. The previous heap_{beginscan,rescan,endscan} et al. have been replaced with a table_ version. There's no direct replacement for heap_getnext(), as that returned a HeapTuple, which is undesirable for a other AMs. Instead there's table_scan_getnextslot(). But note that heap_getnext() lives on, it's still used widely to access catalog tables. This is achieved by new scan_begin, scan_end, scan_rescan, scan_getnextslot callbacks. 2) The portion of parallel scans that's shared between backends need to be able to do so without the user doing per-AM work. To achieve that new parallelscan_{estimate, initialize, reinitialize} callbacks are introduced, which operate on a new ParallelTableScanDesc, which again can be subclassed by AMs. As it is likely that several AMs are going to be block oriented, block oriented callbacks that can be shared between such AMs are provided and used by heap. table_block_parallelscan_{estimate, intiialize, reinitialize} as callbacks, and table_block_parallelscan_{nextpage, init} for use in AMs. These operate on a ParallelBlockTableScanDesc. 3) Index scans need to be able to access tables to return a tuple, and there needs to be state across individual accesses to the heap to store state like buffers. That's now handled by introducing a sort-of-scan IndexFetchTable, which again is intended to be subclassed by individual AMs (for heap IndexFetchHeap). The relevant callbacks for an AM are index_fetch_{end, begin, reset} to create the necessary state, and index_fetch_tuple to retrieve an indexed tuple. Note that index_fetch_tuple implementations need to be smarter than just blindly fetching the tuples for AMs that have optimizations similar to heap's HOT - the currently alive tuple in the update chain needs to be fetched if appropriate. Similar to table_scan_getnextslot(), it's undesirable to continue to return HeapTuples. Thus index_fetch_heap (might want to rename that later) now accepts a slot as an argument. Core code doesn't have a lot of call sites performing index scans without going through the systable_* API (in contrast to loads of heap_getnext calls and working directly with HeapTuples). Index scans now store the result of a search in IndexScanDesc->xs_heaptid, rather than xs_ctup->t_self. As the target is not generally a HeapTuple anymore that seems cleaner. To be able to sensible adapt code to use the above, two further callbacks have been introduced: a) slot_callbacks returns a TupleTableSlotOps* suitable for creating slots capable of holding a tuple of the AMs type. table_slot_callbacks() and table_slot_create() are based upon that, but have additional logic to deal with views, foreign tables, etc. While this change could have been done separately, nearly all the call sites that needed to be adapted for the rest of this commit also would have been needed to be adapted for table_slot_callbacks(), making separation not worthwhile. b) tuple_satisfies_snapshot checks whether the tuple in a slot is currently visible according to a snapshot. That's required as a few places now don't have a buffer + HeapTuple around, but a slot (which in heap's case internally has that information). Additionally a few infrastructure changes were needed: I) SysScanDesc, as used by systable_{beginscan, getnext} et al. now internally uses a slot to keep track of tuples. While systable_getnext() still returns HeapTuples, and will so for the foreseeable future, the index API (see 1) above) now only deals with slots. The remainder, and largest part, of this commit is then adjusting all scans in postgres to use the new APIs. Author: Andres Freund, Haribabu Kommi, Alvaro Herrera Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-11 20:46:41 +01:00
/*
* Return next tuple from `scan`, store in slot.
*/
bool (*scan_getnextslot) (TableScanDesc scan,
tableam: Formatting and other minor cleanups. The superflous heapam_xlog.h includes were reported by Peter Geoghegan.
2019-03-31 05:13:56 +02:00
ScanDirection direction,
TupleTableSlot *slot);
tableam: Add and use scan APIs. Too allow table accesses to be not directly dependent on heap, several new abstractions are needed. Specifically: 1) Heap scans need to be generalized into table scans. Do this by introducing TableScanDesc, which will be the "base class" for individual AMs. This contains the AM independent fields from HeapScanDesc. The previous heap_{beginscan,rescan,endscan} et al. have been replaced with a table_ version. There's no direct replacement for heap_getnext(), as that returned a HeapTuple, which is undesirable for a other AMs. Instead there's table_scan_getnextslot(). But note that heap_getnext() lives on, it's still used widely to access catalog tables. This is achieved by new scan_begin, scan_end, scan_rescan, scan_getnextslot callbacks. 2) The portion of parallel scans that's shared between backends need to be able to do so without the user doing per-AM work. To achieve that new parallelscan_{estimate, initialize, reinitialize} callbacks are introduced, which operate on a new ParallelTableScanDesc, which again can be subclassed by AMs. As it is likely that several AMs are going to be block oriented, block oriented callbacks that can be shared between such AMs are provided and used by heap. table_block_parallelscan_{estimate, intiialize, reinitialize} as callbacks, and table_block_parallelscan_{nextpage, init} for use in AMs. These operate on a ParallelBlockTableScanDesc. 3) Index scans need to be able to access tables to return a tuple, and there needs to be state across individual accesses to the heap to store state like buffers. That's now handled by introducing a sort-of-scan IndexFetchTable, which again is intended to be subclassed by individual AMs (for heap IndexFetchHeap). The relevant callbacks for an AM are index_fetch_{end, begin, reset} to create the necessary state, and index_fetch_tuple to retrieve an indexed tuple. Note that index_fetch_tuple implementations need to be smarter than just blindly fetching the tuples for AMs that have optimizations similar to heap's HOT - the currently alive tuple in the update chain needs to be fetched if appropriate. Similar to table_scan_getnextslot(), it's undesirable to continue to return HeapTuples. Thus index_fetch_heap (might want to rename that later) now accepts a slot as an argument. Core code doesn't have a lot of call sites performing index scans without going through the systable_* API (in contrast to loads of heap_getnext calls and working directly with HeapTuples). Index scans now store the result of a search in IndexScanDesc->xs_heaptid, rather than xs_ctup->t_self. As the target is not generally a HeapTuple anymore that seems cleaner. To be able to sensible adapt code to use the above, two further callbacks have been introduced: a) slot_callbacks returns a TupleTableSlotOps* suitable for creating slots capable of holding a tuple of the AMs type. table_slot_callbacks() and table_slot_create() are based upon that, but have additional logic to deal with views, foreign tables, etc. While this change could have been done separately, nearly all the call sites that needed to be adapted for the rest of this commit also would have been needed to be adapted for table_slot_callbacks(), making separation not worthwhile. b) tuple_satisfies_snapshot checks whether the tuple in a slot is currently visible according to a snapshot. That's required as a few places now don't have a buffer + HeapTuple around, but a slot (which in heap's case internally has that information). Additionally a few infrastructure changes were needed: I) SysScanDesc, as used by systable_{beginscan, getnext} et al. now internally uses a slot to keep track of tuples. While systable_getnext() still returns HeapTuples, and will so for the foreseeable future, the index API (see 1) above) now only deals with slots. The remainder, and largest part, of this commit is then adjusting all scans in postgres to use the new APIs. Author: Andres Freund, Haribabu Kommi, Alvaro Herrera Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-11 20:46:41 +01:00
Add TID Range Scans to support efficient scanning ranges of TIDs This adds a new executor node named TID Range Scan. The query planner will generate paths for TID Range scans when quals are discovered on base relations which search for ranges on the table's ctid column. These ranges may be open at either end. For example, WHERE ctid >= '(10,0)'; will return all tuples on page 10 and over. To support this, two new optional callback functions have been added to table AM. scan_set_tidrange is used to set the scan range to just the given range of TIDs. scan_getnextslot_tidrange fetches the next tuple in the given range. For AMs were scanning ranges of TIDs would not make sense, these functions can be set to NULL in the TableAmRoutine. The query planner won't generate TID Range Scan Paths in that case. Author: Edmund Horner, David Rowley Reviewed-by: David Rowley, Tomas Vondra, Tom Lane, Andres Freund, Zhihong Yu Discussion: https://postgr.es/m/CAMyN-kB-nFTkF=VA_JPwFNo08S0d-Yk0F741S2B7LDmYAi8eyA@mail.gmail.com
2021-02-27 10:59:36 +01:00
/*-----------
* Optional functions to provide scanning for ranges of ItemPointers.
* Implementations must either provide both of these functions, or neither
* of them.
*
* Implementations of scan_set_tidrange must themselves handle
* ItemPointers of any value. i.e, they must handle each of the following:
*
* 1) mintid or maxtid is beyond the end of the table; and
* 2) mintid is above maxtid; and
* 3) item offset for mintid or maxtid is beyond the maximum offset
* allowed by the AM.
*
* Implementations can assume that scan_set_tidrange is always called
* before can_getnextslot_tidrange or after scan_rescan and before any
* further calls to scan_getnextslot_tidrange.
*/
void (*scan_set_tidrange) (TableScanDesc scan,
ItemPointer mintid,
ItemPointer maxtid);
/*
* Return next tuple from `scan` that's in the range of TIDs defined by
* scan_set_tidrange.
*/
bool (*scan_getnextslot_tidrange) (TableScanDesc scan,
ScanDirection direction,
TupleTableSlot *slot);
tableam: Add and use scan APIs. Too allow table accesses to be not directly dependent on heap, several new abstractions are needed. Specifically: 1) Heap scans need to be generalized into table scans. Do this by introducing TableScanDesc, which will be the "base class" for individual AMs. This contains the AM independent fields from HeapScanDesc. The previous heap_{beginscan,rescan,endscan} et al. have been replaced with a table_ version. There's no direct replacement for heap_getnext(), as that returned a HeapTuple, which is undesirable for a other AMs. Instead there's table_scan_getnextslot(). But note that heap_getnext() lives on, it's still used widely to access catalog tables. This is achieved by new scan_begin, scan_end, scan_rescan, scan_getnextslot callbacks. 2) The portion of parallel scans that's shared between backends need to be able to do so without the user doing per-AM work. To achieve that new parallelscan_{estimate, initialize, reinitialize} callbacks are introduced, which operate on a new ParallelTableScanDesc, which again can be subclassed by AMs. As it is likely that several AMs are going to be block oriented, block oriented callbacks that can be shared between such AMs are provided and used by heap. table_block_parallelscan_{estimate, intiialize, reinitialize} as callbacks, and table_block_parallelscan_{nextpage, init} for use in AMs. These operate on a ParallelBlockTableScanDesc. 3) Index scans need to be able to access tables to return a tuple, and there needs to be state across individual accesses to the heap to store state like buffers. That's now handled by introducing a sort-of-scan IndexFetchTable, which again is intended to be subclassed by individual AMs (for heap IndexFetchHeap). The relevant callbacks for an AM are index_fetch_{end, begin, reset} to create the necessary state, and index_fetch_tuple to retrieve an indexed tuple. Note that index_fetch_tuple implementations need to be smarter than just blindly fetching the tuples for AMs that have optimizations similar to heap's HOT - the currently alive tuple in the update chain needs to be fetched if appropriate. Similar to table_scan_getnextslot(), it's undesirable to continue to return HeapTuples. Thus index_fetch_heap (might want to rename that later) now accepts a slot as an argument. Core code doesn't have a lot of call sites performing index scans without going through the systable_* API (in contrast to loads of heap_getnext calls and working directly with HeapTuples). Index scans now store the result of a search in IndexScanDesc->xs_heaptid, rather than xs_ctup->t_self. As the target is not generally a HeapTuple anymore that seems cleaner. To be able to sensible adapt code to use the above, two further callbacks have been introduced: a) slot_callbacks returns a TupleTableSlotOps* suitable for creating slots capable of holding a tuple of the AMs type. table_slot_callbacks() and table_slot_create() are based upon that, but have additional logic to deal with views, foreign tables, etc. While this change could have been done separately, nearly all the call sites that needed to be adapted for the rest of this commit also would have been needed to be adapted for table_slot_callbacks(), making separation not worthwhile. b) tuple_satisfies_snapshot checks whether the tuple in a slot is currently visible according to a snapshot. That's required as a few places now don't have a buffer + HeapTuple around, but a slot (which in heap's case internally has that information). Additionally a few infrastructure changes were needed: I) SysScanDesc, as used by systable_{beginscan, getnext} et al. now internally uses a slot to keep track of tuples. While systable_getnext() still returns HeapTuples, and will so for the foreseeable future, the index API (see 1) above) now only deals with slots. The remainder, and largest part, of this commit is then adjusting all scans in postgres to use the new APIs. Author: Andres Freund, Haribabu Kommi, Alvaro Herrera Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-11 20:46:41 +01:00
/* ------------------------------------------------------------------------
* Parallel table scan related functions.
* ------------------------------------------------------------------------
*/
/*
* Estimate the size of shared memory needed for a parallel scan of this
* relation. The snapshot does not need to be accounted for.
*/
Size (*parallelscan_estimate) (Relation rel);
/*
* Initialize ParallelTableScanDesc for a parallel scan of this relation.
tableam: comment and formatting fixes. Author: Heikki Linnakangas Discussion: https://postgr.es/m/9a7fb9cc-2419-5db7-8840-ddc10c93f122@iki.fi
2019-04-08 19:25:16 +02:00
* `pscan` will be sized according to parallelscan_estimate() for the same
tableam: Add and use scan APIs. Too allow table accesses to be not directly dependent on heap, several new abstractions are needed. Specifically: 1) Heap scans need to be generalized into table scans. Do this by introducing TableScanDesc, which will be the "base class" for individual AMs. This contains the AM independent fields from HeapScanDesc. The previous heap_{beginscan,rescan,endscan} et al. have been replaced with a table_ version. There's no direct replacement for heap_getnext(), as that returned a HeapTuple, which is undesirable for a other AMs. Instead there's table_scan_getnextslot(). But note that heap_getnext() lives on, it's still used widely to access catalog tables. This is achieved by new scan_begin, scan_end, scan_rescan, scan_getnextslot callbacks. 2) The portion of parallel scans that's shared between backends need to be able to do so without the user doing per-AM work. To achieve that new parallelscan_{estimate, initialize, reinitialize} callbacks are introduced, which operate on a new ParallelTableScanDesc, which again can be subclassed by AMs. As it is likely that several AMs are going to be block oriented, block oriented callbacks that can be shared between such AMs are provided and used by heap. table_block_parallelscan_{estimate, intiialize, reinitialize} as callbacks, and table_block_parallelscan_{nextpage, init} for use in AMs. These operate on a ParallelBlockTableScanDesc. 3) Index scans need to be able to access tables to return a tuple, and there needs to be state across individual accesses to the heap to store state like buffers. That's now handled by introducing a sort-of-scan IndexFetchTable, which again is intended to be subclassed by individual AMs (for heap IndexFetchHeap). The relevant callbacks for an AM are index_fetch_{end, begin, reset} to create the necessary state, and index_fetch_tuple to retrieve an indexed tuple. Note that index_fetch_tuple implementations need to be smarter than just blindly fetching the tuples for AMs that have optimizations similar to heap's HOT - the currently alive tuple in the update chain needs to be fetched if appropriate. Similar to table_scan_getnextslot(), it's undesirable to continue to return HeapTuples. Thus index_fetch_heap (might want to rename that later) now accepts a slot as an argument. Core code doesn't have a lot of call sites performing index scans without going through the systable_* API (in contrast to loads of heap_getnext calls and working directly with HeapTuples). Index scans now store the result of a search in IndexScanDesc->xs_heaptid, rather than xs_ctup->t_self. As the target is not generally a HeapTuple anymore that seems cleaner. To be able to sensible adapt code to use the above, two further callbacks have been introduced: a) slot_callbacks returns a TupleTableSlotOps* suitable for creating slots capable of holding a tuple of the AMs type. table_slot_callbacks() and table_slot_create() are based upon that, but have additional logic to deal with views, foreign tables, etc. While this change could have been done separately, nearly all the call sites that needed to be adapted for the rest of this commit also would have been needed to be adapted for table_slot_callbacks(), making separation not worthwhile. b) tuple_satisfies_snapshot checks whether the tuple in a slot is currently visible according to a snapshot. That's required as a few places now don't have a buffer + HeapTuple around, but a slot (which in heap's case internally has that information). Additionally a few infrastructure changes were needed: I) SysScanDesc, as used by systable_{beginscan, getnext} et al. now internally uses a slot to keep track of tuples. While systable_getnext() still returns HeapTuples, and will so for the foreseeable future, the index API (see 1) above) now only deals with slots. The remainder, and largest part, of this commit is then adjusting all scans in postgres to use the new APIs. Author: Andres Freund, Haribabu Kommi, Alvaro Herrera Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-11 20:46:41 +01:00
* relation.
*/
tableam: Formatting and other minor cleanups. The superflous heapam_xlog.h includes were reported by Peter Geoghegan.
2019-03-31 05:13:56 +02:00
Size (*parallelscan_initialize) (Relation rel,
ParallelTableScanDesc pscan);
tableam: Add and use scan APIs. Too allow table accesses to be not directly dependent on heap, several new abstractions are needed. Specifically: 1) Heap scans need to be generalized into table scans. Do this by introducing TableScanDesc, which will be the "base class" for individual AMs. This contains the AM independent fields from HeapScanDesc. The previous heap_{beginscan,rescan,endscan} et al. have been replaced with a table_ version. There's no direct replacement for heap_getnext(), as that returned a HeapTuple, which is undesirable for a other AMs. Instead there's table_scan_getnextslot(). But note that heap_getnext() lives on, it's still used widely to access catalog tables. This is achieved by new scan_begin, scan_end, scan_rescan, scan_getnextslot callbacks. 2) The portion of parallel scans that's shared between backends need to be able to do so without the user doing per-AM work. To achieve that new parallelscan_{estimate, initialize, reinitialize} callbacks are introduced, which operate on a new ParallelTableScanDesc, which again can be subclassed by AMs. As it is likely that several AMs are going to be block oriented, block oriented callbacks that can be shared between such AMs are provided and used by heap. table_block_parallelscan_{estimate, intiialize, reinitialize} as callbacks, and table_block_parallelscan_{nextpage, init} for use in AMs. These operate on a ParallelBlockTableScanDesc. 3) Index scans need to be able to access tables to return a tuple, and there needs to be state across individual accesses to the heap to store state like buffers. That's now handled by introducing a sort-of-scan IndexFetchTable, which again is intended to be subclassed by individual AMs (for heap IndexFetchHeap). The relevant callbacks for an AM are index_fetch_{end, begin, reset} to create the necessary state, and index_fetch_tuple to retrieve an indexed tuple. Note that index_fetch_tuple implementations need to be smarter than just blindly fetching the tuples for AMs that have optimizations similar to heap's HOT - the currently alive tuple in the update chain needs to be fetched if appropriate. Similar to table_scan_getnextslot(), it's undesirable to continue to return HeapTuples. Thus index_fetch_heap (might want to rename that later) now accepts a slot as an argument. Core code doesn't have a lot of call sites performing index scans without going through the systable_* API (in contrast to loads of heap_getnext calls and working directly with HeapTuples). Index scans now store the result of a search in IndexScanDesc->xs_heaptid, rather than xs_ctup->t_self. As the target is not generally a HeapTuple anymore that seems cleaner. To be able to sensible adapt code to use the above, two further callbacks have been introduced: a) slot_callbacks returns a TupleTableSlotOps* suitable for creating slots capable of holding a tuple of the AMs type. table_slot_callbacks() and table_slot_create() are based upon that, but have additional logic to deal with views, foreign tables, etc. While this change could have been done separately, nearly all the call sites that needed to be adapted for the rest of this commit also would have been needed to be adapted for table_slot_callbacks(), making separation not worthwhile. b) tuple_satisfies_snapshot checks whether the tuple in a slot is currently visible according to a snapshot. That's required as a few places now don't have a buffer + HeapTuple around, but a slot (which in heap's case internally has that information). Additionally a few infrastructure changes were needed: I) SysScanDesc, as used by systable_{beginscan, getnext} et al. now internally uses a slot to keep track of tuples. While systable_getnext() still returns HeapTuples, and will so for the foreseeable future, the index API (see 1) above) now only deals with slots. The remainder, and largest part, of this commit is then adjusting all scans in postgres to use the new APIs. Author: Andres Freund, Haribabu Kommi, Alvaro Herrera Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-11 20:46:41 +01:00
/*
tableam: Comment fixes. Author: Haribabu Kommi Discussion: CAJrrPGeeYOqP3hkZyohDx_8dot4zvPuPMDBmhJ=iC85cTBNeYw@mail.gmail.com
2019-03-29 16:16:35 +01:00
* Reinitialize `pscan` for a new scan. `rel` will be the same relation as
tableam: Add and use scan APIs. Too allow table accesses to be not directly dependent on heap, several new abstractions are needed. Specifically: 1) Heap scans need to be generalized into table scans. Do this by introducing TableScanDesc, which will be the "base class" for individual AMs. This contains the AM independent fields from HeapScanDesc. The previous heap_{beginscan,rescan,endscan} et al. have been replaced with a table_ version. There's no direct replacement for heap_getnext(), as that returned a HeapTuple, which is undesirable for a other AMs. Instead there's table_scan_getnextslot(). But note that heap_getnext() lives on, it's still used widely to access catalog tables. This is achieved by new scan_begin, scan_end, scan_rescan, scan_getnextslot callbacks. 2) The portion of parallel scans that's shared between backends need to be able to do so without the user doing per-AM work. To achieve that new parallelscan_{estimate, initialize, reinitialize} callbacks are introduced, which operate on a new ParallelTableScanDesc, which again can be subclassed by AMs. As it is likely that several AMs are going to be block oriented, block oriented callbacks that can be shared between such AMs are provided and used by heap. table_block_parallelscan_{estimate, intiialize, reinitialize} as callbacks, and table_block_parallelscan_{nextpage, init} for use in AMs. These operate on a ParallelBlockTableScanDesc. 3) Index scans need to be able to access tables to return a tuple, and there needs to be state across individual accesses to the heap to store state like buffers. That's now handled by introducing a sort-of-scan IndexFetchTable, which again is intended to be subclassed by individual AMs (for heap IndexFetchHeap). The relevant callbacks for an AM are index_fetch_{end, begin, reset} to create the necessary state, and index_fetch_tuple to retrieve an indexed tuple. Note that index_fetch_tuple implementations need to be smarter than just blindly fetching the tuples for AMs that have optimizations similar to heap's HOT - the currently alive tuple in the update chain needs to be fetched if appropriate. Similar to table_scan_getnextslot(), it's undesirable to continue to return HeapTuples. Thus index_fetch_heap (might want to rename that later) now accepts a slot as an argument. Core code doesn't have a lot of call sites performing index scans without going through the systable_* API (in contrast to loads of heap_getnext calls and working directly with HeapTuples). Index scans now store the result of a search in IndexScanDesc->xs_heaptid, rather than xs_ctup->t_self. As the target is not generally a HeapTuple anymore that seems cleaner. To be able to sensible adapt code to use the above, two further callbacks have been introduced: a) slot_callbacks returns a TupleTableSlotOps* suitable for creating slots capable of holding a tuple of the AMs type. table_slot_callbacks() and table_slot_create() are based upon that, but have additional logic to deal with views, foreign tables, etc. While this change could have been done separately, nearly all the call sites that needed to be adapted for the rest of this commit also would have been needed to be adapted for table_slot_callbacks(), making separation not worthwhile. b) tuple_satisfies_snapshot checks whether the tuple in a slot is currently visible according to a snapshot. That's required as a few places now don't have a buffer + HeapTuple around, but a slot (which in heap's case internally has that information). Additionally a few infrastructure changes were needed: I) SysScanDesc, as used by systable_{beginscan, getnext} et al. now internally uses a slot to keep track of tuples. While systable_getnext() still returns HeapTuples, and will so for the foreseeable future, the index API (see 1) above) now only deals with slots. The remainder, and largest part, of this commit is then adjusting all scans in postgres to use the new APIs. Author: Andres Freund, Haribabu Kommi, Alvaro Herrera Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-11 20:46:41 +01:00
* when `pscan` was initialized by parallelscan_initialize.
*/
tableam: Formatting and other minor cleanups. The superflous heapam_xlog.h includes were reported by Peter Geoghegan.
2019-03-31 05:13:56 +02:00
void (*parallelscan_reinitialize) (Relation rel,
ParallelTableScanDesc pscan);
tableam: Add and use scan APIs. Too allow table accesses to be not directly dependent on heap, several new abstractions are needed. Specifically: 1) Heap scans need to be generalized into table scans. Do this by introducing TableScanDesc, which will be the "base class" for individual AMs. This contains the AM independent fields from HeapScanDesc. The previous heap_{beginscan,rescan,endscan} et al. have been replaced with a table_ version. There's no direct replacement for heap_getnext(), as that returned a HeapTuple, which is undesirable for a other AMs. Instead there's table_scan_getnextslot(). But note that heap_getnext() lives on, it's still used widely to access catalog tables. This is achieved by new scan_begin, scan_end, scan_rescan, scan_getnextslot callbacks. 2) The portion of parallel scans that's shared between backends need to be able to do so without the user doing per-AM work. To achieve that new parallelscan_{estimate, initialize, reinitialize} callbacks are introduced, which operate on a new ParallelTableScanDesc, which again can be subclassed by AMs. As it is likely that several AMs are going to be block oriented, block oriented callbacks that can be shared between such AMs are provided and used by heap. table_block_parallelscan_{estimate, intiialize, reinitialize} as callbacks, and table_block_parallelscan_{nextpage, init} for use in AMs. These operate on a ParallelBlockTableScanDesc. 3) Index scans need to be able to access tables to return a tuple, and there needs to be state across individual accesses to the heap to store state like buffers. That's now handled by introducing a sort-of-scan IndexFetchTable, which again is intended to be subclassed by individual AMs (for heap IndexFetchHeap). The relevant callbacks for an AM are index_fetch_{end, begin, reset} to create the necessary state, and index_fetch_tuple to retrieve an indexed tuple. Note that index_fetch_tuple implementations need to be smarter than just blindly fetching the tuples for AMs that have optimizations similar to heap's HOT - the currently alive tuple in the update chain needs to be fetched if appropriate. Similar to table_scan_getnextslot(), it's undesirable to continue to return HeapTuples. Thus index_fetch_heap (might want to rename that later) now accepts a slot as an argument. Core code doesn't have a lot of call sites performing index scans without going through the systable_* API (in contrast to loads of heap_getnext calls and working directly with HeapTuples). Index scans now store the result of a search in IndexScanDesc->xs_heaptid, rather than xs_ctup->t_self. As the target is not generally a HeapTuple anymore that seems cleaner. To be able to sensible adapt code to use the above, two further callbacks have been introduced: a) slot_callbacks returns a TupleTableSlotOps* suitable for creating slots capable of holding a tuple of the AMs type. table_slot_callbacks() and table_slot_create() are based upon that, but have additional logic to deal with views, foreign tables, etc. While this change could have been done separately, nearly all the call sites that needed to be adapted for the rest of this commit also would have been needed to be adapted for table_slot_callbacks(), making separation not worthwhile. b) tuple_satisfies_snapshot checks whether the tuple in a slot is currently visible according to a snapshot. That's required as a few places now don't have a buffer + HeapTuple around, but a slot (which in heap's case internally has that information). Additionally a few infrastructure changes were needed: I) SysScanDesc, as used by systable_{beginscan, getnext} et al. now internally uses a slot to keep track of tuples. While systable_getnext() still returns HeapTuples, and will so for the foreseeable future, the index API (see 1) above) now only deals with slots. The remainder, and largest part, of this commit is then adjusting all scans in postgres to use the new APIs. Author: Andres Freund, Haribabu Kommi, Alvaro Herrera Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-11 20:46:41 +01:00
/* ------------------------------------------------------------------------
* Index Scan Callbacks
* ------------------------------------------------------------------------
*/
/*
* Prepare to fetch tuples from the relation, as needed when fetching
tableam: comment and formatting fixes. Author: Heikki Linnakangas Discussion: https://postgr.es/m/9a7fb9cc-2419-5db7-8840-ddc10c93f122@iki.fi
2019-04-08 19:25:16 +02:00
* tuples for an index scan. The callback has to return an
tableam: Add and use scan APIs. Too allow table accesses to be not directly dependent on heap, several new abstractions are needed. Specifically: 1) Heap scans need to be generalized into table scans. Do this by introducing TableScanDesc, which will be the "base class" for individual AMs. This contains the AM independent fields from HeapScanDesc. The previous heap_{beginscan,rescan,endscan} et al. have been replaced with a table_ version. There's no direct replacement for heap_getnext(), as that returned a HeapTuple, which is undesirable for a other AMs. Instead there's table_scan_getnextslot(). But note that heap_getnext() lives on, it's still used widely to access catalog tables. This is achieved by new scan_begin, scan_end, scan_rescan, scan_getnextslot callbacks. 2) The portion of parallel scans that's shared between backends need to be able to do so without the user doing per-AM work. To achieve that new parallelscan_{estimate, initialize, reinitialize} callbacks are introduced, which operate on a new ParallelTableScanDesc, which again can be subclassed by AMs. As it is likely that several AMs are going to be block oriented, block oriented callbacks that can be shared between such AMs are provided and used by heap. table_block_parallelscan_{estimate, intiialize, reinitialize} as callbacks, and table_block_parallelscan_{nextpage, init} for use in AMs. These operate on a ParallelBlockTableScanDesc. 3) Index scans need to be able to access tables to return a tuple, and there needs to be state across individual accesses to the heap to store state like buffers. That's now handled by introducing a sort-of-scan IndexFetchTable, which again is intended to be subclassed by individual AMs (for heap IndexFetchHeap). The relevant callbacks for an AM are index_fetch_{end, begin, reset} to create the necessary state, and index_fetch_tuple to retrieve an indexed tuple. Note that index_fetch_tuple implementations need to be smarter than just blindly fetching the tuples for AMs that have optimizations similar to heap's HOT - the currently alive tuple in the update chain needs to be fetched if appropriate. Similar to table_scan_getnextslot(), it's undesirable to continue to return HeapTuples. Thus index_fetch_heap (might want to rename that later) now accepts a slot as an argument. Core code doesn't have a lot of call sites performing index scans without going through the systable_* API (in contrast to loads of heap_getnext calls and working directly with HeapTuples). Index scans now store the result of a search in IndexScanDesc->xs_heaptid, rather than xs_ctup->t_self. As the target is not generally a HeapTuple anymore that seems cleaner. To be able to sensible adapt code to use the above, two further callbacks have been introduced: a) slot_callbacks returns a TupleTableSlotOps* suitable for creating slots capable of holding a tuple of the AMs type. table_slot_callbacks() and table_slot_create() are based upon that, but have additional logic to deal with views, foreign tables, etc. While this change could have been done separately, nearly all the call sites that needed to be adapted for the rest of this commit also would have been needed to be adapted for table_slot_callbacks(), making separation not worthwhile. b) tuple_satisfies_snapshot checks whether the tuple in a slot is currently visible according to a snapshot. That's required as a few places now don't have a buffer + HeapTuple around, but a slot (which in heap's case internally has that information). Additionally a few infrastructure changes were needed: I) SysScanDesc, as used by systable_{beginscan, getnext} et al. now internally uses a slot to keep track of tuples. While systable_getnext() still returns HeapTuples, and will so for the foreseeable future, the index API (see 1) above) now only deals with slots. The remainder, and largest part, of this commit is then adjusting all scans in postgres to use the new APIs. Author: Andres Freund, Haribabu Kommi, Alvaro Herrera Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-11 20:46:41 +01:00
* IndexFetchTableData, which the AM will typically embed in a larger
* structure with additional information.
*
* Tuples for an index scan can then be fetched via index_fetch_tuple.
*/
struct IndexFetchTableData *(*index_fetch_begin) (Relation rel);
/*
* Reset index fetch. Typically this will release cross index fetch
* resources held in IndexFetchTableData.
*/
void (*index_fetch_reset) (struct IndexFetchTableData *data);
/*
* Release resources and deallocate index fetch.
*/
void (*index_fetch_end) (struct IndexFetchTableData *data);
/*
* Fetch tuple at `tid` into `slot`, after doing a visibility test
* according to `snapshot`. If a tuple was found and passed the visibility
* test, return true, false otherwise.
*
* Note that AMs that do not necessarily update indexes when indexed
tableam: comment and formatting fixes. Author: Heikki Linnakangas Discussion: https://postgr.es/m/9a7fb9cc-2419-5db7-8840-ddc10c93f122@iki.fi
2019-04-08 19:25:16 +02:00
* columns do not change, need to return the current/correct version of
* the tuple that is visible to the snapshot, even if the tid points to an
* older version of the tuple.
tableam: Add and use scan APIs. Too allow table accesses to be not directly dependent on heap, several new abstractions are needed. Specifically: 1) Heap scans need to be generalized into table scans. Do this by introducing TableScanDesc, which will be the "base class" for individual AMs. This contains the AM independent fields from HeapScanDesc. The previous heap_{beginscan,rescan,endscan} et al. have been replaced with a table_ version. There's no direct replacement for heap_getnext(), as that returned a HeapTuple, which is undesirable for a other AMs. Instead there's table_scan_getnextslot(). But note that heap_getnext() lives on, it's still used widely to access catalog tables. This is achieved by new scan_begin, scan_end, scan_rescan, scan_getnextslot callbacks. 2) The portion of parallel scans that's shared between backends need to be able to do so without the user doing per-AM work. To achieve that new parallelscan_{estimate, initialize, reinitialize} callbacks are introduced, which operate on a new ParallelTableScanDesc, which again can be subclassed by AMs. As it is likely that several AMs are going to be block oriented, block oriented callbacks that can be shared between such AMs are provided and used by heap. table_block_parallelscan_{estimate, intiialize, reinitialize} as callbacks, and table_block_parallelscan_{nextpage, init} for use in AMs. These operate on a ParallelBlockTableScanDesc. 3) Index scans need to be able to access tables to return a tuple, and there needs to be state across individual accesses to the heap to store state like buffers. That's now handled by introducing a sort-of-scan IndexFetchTable, which again is intended to be subclassed by individual AMs (for heap IndexFetchHeap). The relevant callbacks for an AM are index_fetch_{end, begin, reset} to create the necessary state, and index_fetch_tuple to retrieve an indexed tuple. Note that index_fetch_tuple implementations need to be smarter than just blindly fetching the tuples for AMs that have optimizations similar to heap's HOT - the currently alive tuple in the update chain needs to be fetched if appropriate. Similar to table_scan_getnextslot(), it's undesirable to continue to return HeapTuples. Thus index_fetch_heap (might want to rename that later) now accepts a slot as an argument. Core code doesn't have a lot of call sites performing index scans without going through the systable_* API (in contrast to loads of heap_getnext calls and working directly with HeapTuples). Index scans now store the result of a search in IndexScanDesc->xs_heaptid, rather than xs_ctup->t_self. As the target is not generally a HeapTuple anymore that seems cleaner. To be able to sensible adapt code to use the above, two further callbacks have been introduced: a) slot_callbacks returns a TupleTableSlotOps* suitable for creating slots capable of holding a tuple of the AMs type. table_slot_callbacks() and table_slot_create() are based upon that, but have additional logic to deal with views, foreign tables, etc. While this change could have been done separately, nearly all the call sites that needed to be adapted for the rest of this commit also would have been needed to be adapted for table_slot_callbacks(), making separation not worthwhile. b) tuple_satisfies_snapshot checks whether the tuple in a slot is currently visible according to a snapshot. That's required as a few places now don't have a buffer + HeapTuple around, but a slot (which in heap's case internally has that information). Additionally a few infrastructure changes were needed: I) SysScanDesc, as used by systable_{beginscan, getnext} et al. now internally uses a slot to keep track of tuples. While systable_getnext() still returns HeapTuples, and will so for the foreseeable future, the index API (see 1) above) now only deals with slots. The remainder, and largest part, of this commit is then adjusting all scans in postgres to use the new APIs. Author: Andres Freund, Haribabu Kommi, Alvaro Herrera Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-11 20:46:41 +01:00
*
* *call_again is false on the first call to index_fetch_tuple for a tid.
* If there potentially is another tuple matching the tid, *call_again
Spelling adjustments
2020-06-07 15:06:51 +02:00
* needs to be set to true by index_fetch_tuple, signaling to the caller
tableam: Add and use scan APIs. Too allow table accesses to be not directly dependent on heap, several new abstractions are needed. Specifically: 1) Heap scans need to be generalized into table scans. Do this by introducing TableScanDesc, which will be the "base class" for individual AMs. This contains the AM independent fields from HeapScanDesc. The previous heap_{beginscan,rescan,endscan} et al. have been replaced with a table_ version. There's no direct replacement for heap_getnext(), as that returned a HeapTuple, which is undesirable for a other AMs. Instead there's table_scan_getnextslot(). But note that heap_getnext() lives on, it's still used widely to access catalog tables. This is achieved by new scan_begin, scan_end, scan_rescan, scan_getnextslot callbacks. 2) The portion of parallel scans that's shared between backends need to be able to do so without the user doing per-AM work. To achieve that new parallelscan_{estimate, initialize, reinitialize} callbacks are introduced, which operate on a new ParallelTableScanDesc, which again can be subclassed by AMs. As it is likely that several AMs are going to be block oriented, block oriented callbacks that can be shared between such AMs are provided and used by heap. table_block_parallelscan_{estimate, intiialize, reinitialize} as callbacks, and table_block_parallelscan_{nextpage, init} for use in AMs. These operate on a ParallelBlockTableScanDesc. 3) Index scans need to be able to access tables to return a tuple, and there needs to be state across individual accesses to the heap to store state like buffers. That's now handled by introducing a sort-of-scan IndexFetchTable, which again is intended to be subclassed by individual AMs (for heap IndexFetchHeap). The relevant callbacks for an AM are index_fetch_{end, begin, reset} to create the necessary state, and index_fetch_tuple to retrieve an indexed tuple. Note that index_fetch_tuple implementations need to be smarter than just blindly fetching the tuples for AMs that have optimizations similar to heap's HOT - the currently alive tuple in the update chain needs to be fetched if appropriate. Similar to table_scan_getnextslot(), it's undesirable to continue to return HeapTuples. Thus index_fetch_heap (might want to rename that later) now accepts a slot as an argument. Core code doesn't have a lot of call sites performing index scans without going through the systable_* API (in contrast to loads of heap_getnext calls and working directly with HeapTuples). Index scans now store the result of a search in IndexScanDesc->xs_heaptid, rather than xs_ctup->t_self. As the target is not generally a HeapTuple anymore that seems cleaner. To be able to sensible adapt code to use the above, two further callbacks have been introduced: a) slot_callbacks returns a TupleTableSlotOps* suitable for creating slots capable of holding a tuple of the AMs type. table_slot_callbacks() and table_slot_create() are based upon that, but have additional logic to deal with views, foreign tables, etc. While this change could have been done separately, nearly all the call sites that needed to be adapted for the rest of this commit also would have been needed to be adapted for table_slot_callbacks(), making separation not worthwhile. b) tuple_satisfies_snapshot checks whether the tuple in a slot is currently visible according to a snapshot. That's required as a few places now don't have a buffer + HeapTuple around, but a slot (which in heap's case internally has that information). Additionally a few infrastructure changes were needed: I) SysScanDesc, as used by systable_{beginscan, getnext} et al. now internally uses a slot to keep track of tuples. While systable_getnext() still returns HeapTuples, and will so for the foreseeable future, the index API (see 1) above) now only deals with slots. The remainder, and largest part, of this commit is then adjusting all scans in postgres to use the new APIs. Author: Andres Freund, Haribabu Kommi, Alvaro Herrera Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-11 20:46:41 +01:00
* that index_fetch_tuple should be called again for the same tid.
*
tableam: comment and formatting fixes. Author: Heikki Linnakangas Discussion: https://postgr.es/m/9a7fb9cc-2419-5db7-8840-ddc10c93f122@iki.fi
2019-04-08 19:25:16 +02:00
* *all_dead, if all_dead is not NULL, should be set to true by
tableam: Add helper for indexes to check if a corresponding table tuples exist. This is, likely exclusively, useful to verify that conflicts detected in a unique index are with live tuples, rather than dead ones. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-26 00:52:55 +01:00
* index_fetch_tuple iff it is guaranteed that no backend needs to see
tableam: comment improvements. Author: Brad DeJong Discussion: https://postgr.es/m/CAJnrtnxDYOQFsDfWz2iri0T_fFL2ZbbzgCOE=4yaMcszgcsf4A@mail.gmail.com Backpatch: 12-
2019-07-18 04:39:54 +02:00
* that tuple. Index AMs can use that to avoid returning that tid in
tableam: Add helper for indexes to check if a corresponding table tuples exist. This is, likely exclusively, useful to verify that conflicts detected in a unique index are with live tuples, rather than dead ones. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-26 00:52:55 +01:00
* future searches.
tableam: Add and use scan APIs. Too allow table accesses to be not directly dependent on heap, several new abstractions are needed. Specifically: 1) Heap scans need to be generalized into table scans. Do this by introducing TableScanDesc, which will be the "base class" for individual AMs. This contains the AM independent fields from HeapScanDesc. The previous heap_{beginscan,rescan,endscan} et al. have been replaced with a table_ version. There's no direct replacement for heap_getnext(), as that returned a HeapTuple, which is undesirable for a other AMs. Instead there's table_scan_getnextslot(). But note that heap_getnext() lives on, it's still used widely to access catalog tables. This is achieved by new scan_begin, scan_end, scan_rescan, scan_getnextslot callbacks. 2) The portion of parallel scans that's shared between backends need to be able to do so without the user doing per-AM work. To achieve that new parallelscan_{estimate, initialize, reinitialize} callbacks are introduced, which operate on a new ParallelTableScanDesc, which again can be subclassed by AMs. As it is likely that several AMs are going to be block oriented, block oriented callbacks that can be shared between such AMs are provided and used by heap. table_block_parallelscan_{estimate, intiialize, reinitialize} as callbacks, and table_block_parallelscan_{nextpage, init} for use in AMs. These operate on a ParallelBlockTableScanDesc. 3) Index scans need to be able to access tables to return a tuple, and there needs to be state across individual accesses to the heap to store state like buffers. That's now handled by introducing a sort-of-scan IndexFetchTable, which again is intended to be subclassed by individual AMs (for heap IndexFetchHeap). The relevant callbacks for an AM are index_fetch_{end, begin, reset} to create the necessary state, and index_fetch_tuple to retrieve an indexed tuple. Note that index_fetch_tuple implementations need to be smarter than just blindly fetching the tuples for AMs that have optimizations similar to heap's HOT - the currently alive tuple in the update chain needs to be fetched if appropriate. Similar to table_scan_getnextslot(), it's undesirable to continue to return HeapTuples. Thus index_fetch_heap (might want to rename that later) now accepts a slot as an argument. Core code doesn't have a lot of call sites performing index scans without going through the systable_* API (in contrast to loads of heap_getnext calls and working directly with HeapTuples). Index scans now store the result of a search in IndexScanDesc->xs_heaptid, rather than xs_ctup->t_self. As the target is not generally a HeapTuple anymore that seems cleaner. To be able to sensible adapt code to use the above, two further callbacks have been introduced: a) slot_callbacks returns a TupleTableSlotOps* suitable for creating slots capable of holding a tuple of the AMs type. table_slot_callbacks() and table_slot_create() are based upon that, but have additional logic to deal with views, foreign tables, etc. While this change could have been done separately, nearly all the call sites that needed to be adapted for the rest of this commit also would have been needed to be adapted for table_slot_callbacks(), making separation not worthwhile. b) tuple_satisfies_snapshot checks whether the tuple in a slot is currently visible according to a snapshot. That's required as a few places now don't have a buffer + HeapTuple around, but a slot (which in heap's case internally has that information). Additionally a few infrastructure changes were needed: I) SysScanDesc, as used by systable_{beginscan, getnext} et al. now internally uses a slot to keep track of tuples. While systable_getnext() still returns HeapTuples, and will so for the foreseeable future, the index API (see 1) above) now only deals with slots. The remainder, and largest part, of this commit is then adjusting all scans in postgres to use the new APIs. Author: Andres Freund, Haribabu Kommi, Alvaro Herrera Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-11 20:46:41 +01:00
*/
bool (*index_fetch_tuple) (struct IndexFetchTableData *scan,
ItemPointer tid,
Snapshot snapshot,
TupleTableSlot *slot,
bool *call_again, bool *all_dead);
tableam: comment and formatting fixes. Author: Heikki Linnakangas Discussion: https://postgr.es/m/9a7fb9cc-2419-5db7-8840-ddc10c93f122@iki.fi
2019-04-08 19:25:16 +02:00
tableam: Add and use scan APIs. Too allow table accesses to be not directly dependent on heap, several new abstractions are needed. Specifically: 1) Heap scans need to be generalized into table scans. Do this by introducing TableScanDesc, which will be the "base class" for individual AMs. This contains the AM independent fields from HeapScanDesc. The previous heap_{beginscan,rescan,endscan} et al. have been replaced with a table_ version. There's no direct replacement for heap_getnext(), as that returned a HeapTuple, which is undesirable for a other AMs. Instead there's table_scan_getnextslot(). But note that heap_getnext() lives on, it's still used widely to access catalog tables. This is achieved by new scan_begin, scan_end, scan_rescan, scan_getnextslot callbacks. 2) The portion of parallel scans that's shared between backends need to be able to do so without the user doing per-AM work. To achieve that new parallelscan_{estimate, initialize, reinitialize} callbacks are introduced, which operate on a new ParallelTableScanDesc, which again can be subclassed by AMs. As it is likely that several AMs are going to be block oriented, block oriented callbacks that can be shared between such AMs are provided and used by heap. table_block_parallelscan_{estimate, intiialize, reinitialize} as callbacks, and table_block_parallelscan_{nextpage, init} for use in AMs. These operate on a ParallelBlockTableScanDesc. 3) Index scans need to be able to access tables to return a tuple, and there needs to be state across individual accesses to the heap to store state like buffers. That's now handled by introducing a sort-of-scan IndexFetchTable, which again is intended to be subclassed by individual AMs (for heap IndexFetchHeap). The relevant callbacks for an AM are index_fetch_{end, begin, reset} to create the necessary state, and index_fetch_tuple to retrieve an indexed tuple. Note that index_fetch_tuple implementations need to be smarter than just blindly fetching the tuples for AMs that have optimizations similar to heap's HOT - the currently alive tuple in the update chain needs to be fetched if appropriate. Similar to table_scan_getnextslot(), it's undesirable to continue to return HeapTuples. Thus index_fetch_heap (might want to rename that later) now accepts a slot as an argument. Core code doesn't have a lot of call sites performing index scans without going through the systable_* API (in contrast to loads of heap_getnext calls and working directly with HeapTuples). Index scans now store the result of a search in IndexScanDesc->xs_heaptid, rather than xs_ctup->t_self. As the target is not generally a HeapTuple anymore that seems cleaner. To be able to sensible adapt code to use the above, two further callbacks have been introduced: a) slot_callbacks returns a TupleTableSlotOps* suitable for creating slots capable of holding a tuple of the AMs type. table_slot_callbacks() and table_slot_create() are based upon that, but have additional logic to deal with views, foreign tables, etc. While this change could have been done separately, nearly all the call sites that needed to be adapted for the rest of this commit also would have been needed to be adapted for table_slot_callbacks(), making separation not worthwhile. b) tuple_satisfies_snapshot checks whether the tuple in a slot is currently visible according to a snapshot. That's required as a few places now don't have a buffer + HeapTuple around, but a slot (which in heap's case internally has that information). Additionally a few infrastructure changes were needed: I) SysScanDesc, as used by systable_{beginscan, getnext} et al. now internally uses a slot to keep track of tuples. While systable_getnext() still returns HeapTuples, and will so for the foreseeable future, the index API (see 1) above) now only deals with slots. The remainder, and largest part, of this commit is then adjusting all scans in postgres to use the new APIs. Author: Andres Freund, Haribabu Kommi, Alvaro Herrera Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-11 20:46:41 +01:00
/* ------------------------------------------------------------------------
* Callbacks for non-modifying operations on individual tuples
* ------------------------------------------------------------------------
*/
tableam: Add and use table_fetch_row_version(). This is essentially the tableam version of heapam_fetch(), i.e. fetching a tuple identified by a tid, performing visibility checks. Note that this different from table_index_fetch_tuple(), which is for index lookups. It therefore has to handle a tid pointing to an earlier version of a tuple if the AM uses an optimization like heap's HOT. Add comments to that end. This commit removes the stats_relation argument from heap_fetch, as it's been unused for a long time. Author: Andres Freund Reviewed-By: Haribabu Kommi Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-25 08:13:42 +01:00
/*
tableam: comment and formatting fixes. Author: Heikki Linnakangas Discussion: https://postgr.es/m/9a7fb9cc-2419-5db7-8840-ddc10c93f122@iki.fi
2019-04-08 19:25:16 +02:00
* Fetch tuple at `tid` into `slot`, after doing a visibility test
tableam: Add and use table_fetch_row_version(). This is essentially the tableam version of heapam_fetch(), i.e. fetching a tuple identified by a tid, performing visibility checks. Note that this different from table_index_fetch_tuple(), which is for index lookups. It therefore has to handle a tid pointing to an earlier version of a tuple if the AM uses an optimization like heap's HOT. Add comments to that end. This commit removes the stats_relation argument from heap_fetch, as it's been unused for a long time. Author: Andres Freund Reviewed-By: Haribabu Kommi Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-25 08:13:42 +01:00
* according to `snapshot`. If a tuple was found and passed the visibility
* test, returns true, false otherwise.
*/
bool (*tuple_fetch_row_version) (Relation rel,
ItemPointer tid,
Snapshot snapshot,
TupleTableSlot *slot);
tableam: Avoid relying on relation size to determine validity of tids. Instead add a tableam callback to do so. To avoid adding per validation overhead, pass a scan to tuple_tid_valid. In heap's case we'd otherwise incurred a RelationGetNumberOfBlocks() call for each tid - which'd have added noticable overhead to nodeTidscan.c. Author: Andres Freund Reviewed-By: Ashwin Agrawal Discussion: https://postgr.es/m/20190515185447.gno2jtqxyktylyvs@alap3.anarazel.de
2019-05-18 03:52:01 +02:00
/*
* Is tid valid for a scan of this relation.
*/
bool (*tuple_tid_valid) (TableScanDesc scan,
ItemPointer tid);
tableam: Add table_get_latest_tid, to wrap heap_get_latest_tid. This primarily is to allow WHERE CURRENT OF to continue to work as it currently does. It's not clear to me that these semantics make sense for every AM, but it works for the in-core heap, and the out of core zheap. We can refine it further at a later point if necessary. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-26 01:14:48 +01:00
/*
* Return the latest version of the tuple at `tid`, by updating `tid` to
* point at the newest version.
*/
tableam: Avoid relying on relation size to determine validity of tids. Instead add a tableam callback to do so. To avoid adding per validation overhead, pass a scan to tuple_tid_valid. In heap's case we'd otherwise incurred a RelationGetNumberOfBlocks() call for each tid - which'd have added noticable overhead to nodeTidscan.c. Author: Andres Freund Reviewed-By: Ashwin Agrawal Discussion: https://postgr.es/m/20190515185447.gno2jtqxyktylyvs@alap3.anarazel.de
2019-05-18 03:52:01 +02:00
void (*tuple_get_latest_tid) (TableScanDesc scan,
tableam: Add table_get_latest_tid, to wrap heap_get_latest_tid. This primarily is to allow WHERE CURRENT OF to continue to work as it currently does. It's not clear to me that these semantics make sense for every AM, but it works for the in-core heap, and the out of core zheap. We can refine it further at a later point if necessary. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-26 01:14:48 +01:00
ItemPointer tid);
tableam: Add and use scan APIs. Too allow table accesses to be not directly dependent on heap, several new abstractions are needed. Specifically: 1) Heap scans need to be generalized into table scans. Do this by introducing TableScanDesc, which will be the "base class" for individual AMs. This contains the AM independent fields from HeapScanDesc. The previous heap_{beginscan,rescan,endscan} et al. have been replaced with a table_ version. There's no direct replacement for heap_getnext(), as that returned a HeapTuple, which is undesirable for a other AMs. Instead there's table_scan_getnextslot(). But note that heap_getnext() lives on, it's still used widely to access catalog tables. This is achieved by new scan_begin, scan_end, scan_rescan, scan_getnextslot callbacks. 2) The portion of parallel scans that's shared between backends need to be able to do so without the user doing per-AM work. To achieve that new parallelscan_{estimate, initialize, reinitialize} callbacks are introduced, which operate on a new ParallelTableScanDesc, which again can be subclassed by AMs. As it is likely that several AMs are going to be block oriented, block oriented callbacks that can be shared between such AMs are provided and used by heap. table_block_parallelscan_{estimate, intiialize, reinitialize} as callbacks, and table_block_parallelscan_{nextpage, init} for use in AMs. These operate on a ParallelBlockTableScanDesc. 3) Index scans need to be able to access tables to return a tuple, and there needs to be state across individual accesses to the heap to store state like buffers. That's now handled by introducing a sort-of-scan IndexFetchTable, which again is intended to be subclassed by individual AMs (for heap IndexFetchHeap). The relevant callbacks for an AM are index_fetch_{end, begin, reset} to create the necessary state, and index_fetch_tuple to retrieve an indexed tuple. Note that index_fetch_tuple implementations need to be smarter than just blindly fetching the tuples for AMs that have optimizations similar to heap's HOT - the currently alive tuple in the update chain needs to be fetched if appropriate. Similar to table_scan_getnextslot(), it's undesirable to continue to return HeapTuples. Thus index_fetch_heap (might want to rename that later) now accepts a slot as an argument. Core code doesn't have a lot of call sites performing index scans without going through the systable_* API (in contrast to loads of heap_getnext calls and working directly with HeapTuples). Index scans now store the result of a search in IndexScanDesc->xs_heaptid, rather than xs_ctup->t_self. As the target is not generally a HeapTuple anymore that seems cleaner. To be able to sensible adapt code to use the above, two further callbacks have been introduced: a) slot_callbacks returns a TupleTableSlotOps* suitable for creating slots capable of holding a tuple of the AMs type. table_slot_callbacks() and table_slot_create() are based upon that, but have additional logic to deal with views, foreign tables, etc. While this change could have been done separately, nearly all the call sites that needed to be adapted for the rest of this commit also would have been needed to be adapted for table_slot_callbacks(), making separation not worthwhile. b) tuple_satisfies_snapshot checks whether the tuple in a slot is currently visible according to a snapshot. That's required as a few places now don't have a buffer + HeapTuple around, but a slot (which in heap's case internally has that information). Additionally a few infrastructure changes were needed: I) SysScanDesc, as used by systable_{beginscan, getnext} et al. now internally uses a slot to keep track of tuples. While systable_getnext() still returns HeapTuples, and will so for the foreseeable future, the index API (see 1) above) now only deals with slots. The remainder, and largest part, of this commit is then adjusting all scans in postgres to use the new APIs. Author: Andres Freund, Haribabu Kommi, Alvaro Herrera Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-11 20:46:41 +01:00
/*
* Does the tuple in `slot` satisfy `snapshot`? The slot needs to be of
* the appropriate type for the AM.
*/
bool (*tuple_satisfies_snapshot) (Relation rel,
TupleTableSlot *slot,
Snapshot snapshot);
Enhance nbtree index tuple deletion. Teach nbtree and heapam to cooperate in order to eagerly remove duplicate tuples representing dead MVCC versions. This is "bottom-up deletion". Each bottom-up deletion pass is triggered lazily in response to a flood of versions on an nbtree leaf page. This usually involves a "logically unchanged index" hint (these are produced by the executor mechanism added by commit 9dc718bd). The immediate goal of bottom-up index deletion is to avoid "unnecessary" page splits caused entirely by version duplicates. It naturally has an even more useful effect, though: it acts as a backstop against accumulating an excessive number of index tuple versions for any given _logical row_. Bottom-up index deletion complements what we might now call "top-down index deletion": index vacuuming performed by VACUUM. Bottom-up index deletion responds to the immediate local needs of queries, while leaving it up to autovacuum to perform infrequent clean sweeps of the index. The overall effect is to avoid certain pathological performance issues related to "version churn" from UPDATEs. The previous tableam interface used by index AMs to perform tuple deletion (the table_compute_xid_horizon_for_tuples() function) has been replaced with a new interface that supports certain new requirements. Many (perhaps all) of the capabilities added to nbtree by this commit could also be extended to other index AMs. That is left as work for a later commit. Extend deletion of LP_DEAD-marked index tuples in nbtree by adding logic to consider extra index tuples (that are not LP_DEAD-marked) for deletion in passing. This increases the number of index tuples deleted significantly in many cases. The LP_DEAD deletion process (which is now called "simple deletion" to clearly distinguish it from bottom-up deletion) won't usually need to visit any extra table blocks to check these extra tuples. We have to visit the same table blocks anyway to generate a latestRemovedXid value (at least in the common case where the index deletion operation's WAL record needs such a value). Testing has shown that the "extra tuples" simple deletion enhancement increases the number of index tuples deleted with almost any workload that has LP_DEAD bits set in leaf pages. That is, it almost never fails to delete at least a few extra index tuples. It helps most of all in cases that happen to naturally have a lot of delete-safe tuples. It's not uncommon for an individual deletion operation to end up deleting an order of magnitude more index tuples compared to the old naive approach (e.g., custom instrumentation of the patch shows that this happens fairly often when the regression tests are run). Add a further enhancement that augments simple deletion and bottom-up deletion in indexes that make use of deduplication: Teach nbtree's _bt_delitems_delete() function to support granular TID deletion in posting list tuples. It is now possible to delete individual TIDs from posting list tuples provided the TIDs have a tableam block number of a table block that gets visited as part of the deletion process (visiting the table block can be triggered directly or indirectly). Setting the LP_DEAD bit of a posting list tuple is still an all-or-nothing thing, but that matters much less now that deletion only needs to start out with the right _general_ idea about which index tuples are deletable. Bump XLOG_PAGE_MAGIC because xl_btree_delete changed. No bump in BTREE_VERSION, since there are no changes to the on-disk representation of nbtree indexes. Indexes built on PostgreSQL 12 or PostgreSQL 13 will automatically benefit from bottom-up index deletion (i.e. no reindexing required) following a pg_upgrade. The enhancement to simple deletion is available with all B-Tree indexes following a pg_upgrade, no matter what PostgreSQL version the user upgrades from. Author: Peter Geoghegan <pg@bowt.ie> Reviewed-By: Heikki Linnakangas <hlinnaka@iki.fi> Reviewed-By: Victor Yegorov <vyegorov@gmail.com> Discussion: https://postgr.es/m/CAH2-Wzm+maE3apHB8NOtmM=p-DO65j2V5GzAWCOEEuy3JZgb2g@mail.gmail.com
2021-01-13 18:21:32 +01:00
/* see table_index_delete_tuples() */
TransactionId (*index_delete_tuples) (Relation rel,
TM_IndexDeleteOp *delstate);
Compute XID horizon for page level index vacuum on primary. Previously the xid horizon was only computed during WAL replay. That had two major problems: 1) It relied on knowing what the table pointed to looks like. That was easy enough before the introducing of tableam (we knew it had to be heap, although some trickery around logging the heap relfilenodes was required). But to properly handle table AMs we need per-database catalog access to look up the AM handler, which recovery doesn't allow. 2) Not knowing the xid horizon also makes it hard to support logical decoding on standbys. When on a catalog table, we need to be able to conflict with slots that have an xid horizon that's too old. But computing the horizon by visiting the heap only works once consistency is reached, but we always need to be able to detect conflicts. There's also a secondary problem, in that the current method performs redundant work on every standby. But that's counterbalanced by potentially computing the value when not necessary (either because there's no standby, or because there's no connected backends). Solve 1) and 2) by moving computation of the xid horizon to the primary and by involving tableam in the computation of the horizon. To address the potentially increased overhead, increase the efficiency of the xid horizon computation for heap by sorting the tids, and eliminating redundant buffer accesses. When prefetching is available, additionally perform prefetching of buffers. As this is more of a maintenance task, rather than something routinely done in every read only query, we add an arbitrary 10 to the effective concurrency - thereby using IO concurrency, when not globally enabled. That's possibly not the perfect formula, but seems good enough for now. Bumps WAL format, as latestRemovedXid is now part of the records, and the heap's relfilenode isn't anymore. Author: Andres Freund, Amit Khandekar, Robert Haas Reviewed-By: Robert Haas Discussion: https://postgr.es/m/20181212204154.nsxf3gzqv3gesl32@alap3.anarazel.de https://postgr.es/m/20181214014235.dal5ogljs3bmlq44@alap3.anarazel.de https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-26 22:41:46 +01:00
tableam: Add tuple_{insert, delete, update, lock} and use. This adds new, required, table AM callbacks for insert/delete/update and lock_tuple. To be able to reasonably use those, the EvalPlanQual mechanism had to be adapted, moving more logic into the AM. Previously both delete/update/lock call-sites and the EPQ mechanism had to have awareness of the specific tuple format to be able to fetch the latest version of a tuple. Obviously that needs to be abstracted away. To do so, move the logic that find the latest row version into the AM. lock_tuple has a new flag argument, TUPLE_LOCK_FLAG_FIND_LAST_VERSION, that forces it to lock the last version, rather than the current one. It'd have been possible to do so via a separate callback as well, but finding the last version usually also necessitates locking the newest version, making it sensible to combine the two. This replaces the previous use of EvalPlanQualFetch(). Additionally HeapTupleUpdated, which previously signaled either a concurrent update or delete, is now split into two, to avoid callers needing AM specific knowledge to differentiate. The move of finding the latest row version into tuple_lock means that encountering a row concurrently moved into another partition will now raise an error about "tuple to be locked" rather than "tuple to be updated/deleted" - which is accurate, as that always happens when locking rows. While possible slightly less helpful for users, it seems like an acceptable trade-off. As part of this commit HTSU_Result has been renamed to TM_Result, and its members been expanded to differentiated between updating and deleting. HeapUpdateFailureData has been renamed to TM_FailureData. The interface to speculative insertion is changed so nodeModifyTable.c does not have to set the speculative token itself anymore. Instead there's a version of tuple_insert, tuple_insert_speculative, that performs the speculative insertion (without requiring a flag to signal that fact), and the speculative insertion is either made permanent with table_complete_speculative(succeeded = true) or aborted with succeeded = false). Note that multi_insert is not yet routed through tableam, nor is COPY. Changing multi_insert requires changes to copy.c that are large enough to better be done separately. Similarly, although simpler, CREATE TABLE AS and CREATE MATERIALIZED VIEW are also only going to be adjusted in a later commit. Author: Andres Freund and Haribabu Kommi Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20190313003903.nwvrxi7rw3ywhdel@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-24 03:55:57 +01:00
/* ------------------------------------------------------------------------
* Manipulations of physical tuples.
* ------------------------------------------------------------------------
*/
tableam: Rename wrapper functions to match callback names. Some of the wrapper functions didn't match the callback names. Many of them due to staying "consistent" with historic naming of the wrapped functionality. We decided that for most cases it's more important to be for tableam to be consistent going forward, than with the past. The one exception is beginscan/endscan/... because it'd have looked odd to have systable_beginscan/endscan/... with a different naming scheme, and changing the systable_* APIs would have caused way too much churn (including breaking a lot of external users). Author: Ashwin Agrawal, with some small additions by Andres Freund Reviewed-By: Andres Freund Discussion: https://postgr.es/m/CALfoeiugyrXZfX7n0ORCa4L-m834dzmaE8eFdbNR6PMpetU4Ww@mail.gmail.com
2019-05-24 01:25:48 +02:00
/* see table_tuple_insert() for reference about parameters */
tableam: Formatting and other minor cleanups. The superflous heapam_xlog.h includes were reported by Peter Geoghegan.
2019-03-31 05:13:56 +02:00
void (*tuple_insert) (Relation rel, TupleTableSlot *slot,
CommandId cid, int options,
struct BulkInsertStateData *bistate);
tableam: Add tuple_{insert, delete, update, lock} and use. This adds new, required, table AM callbacks for insert/delete/update and lock_tuple. To be able to reasonably use those, the EvalPlanQual mechanism had to be adapted, moving more logic into the AM. Previously both delete/update/lock call-sites and the EPQ mechanism had to have awareness of the specific tuple format to be able to fetch the latest version of a tuple. Obviously that needs to be abstracted away. To do so, move the logic that find the latest row version into the AM. lock_tuple has a new flag argument, TUPLE_LOCK_FLAG_FIND_LAST_VERSION, that forces it to lock the last version, rather than the current one. It'd have been possible to do so via a separate callback as well, but finding the last version usually also necessitates locking the newest version, making it sensible to combine the two. This replaces the previous use of EvalPlanQualFetch(). Additionally HeapTupleUpdated, which previously signaled either a concurrent update or delete, is now split into two, to avoid callers needing AM specific knowledge to differentiate. The move of finding the latest row version into tuple_lock means that encountering a row concurrently moved into another partition will now raise an error about "tuple to be locked" rather than "tuple to be updated/deleted" - which is accurate, as that always happens when locking rows. While possible slightly less helpful for users, it seems like an acceptable trade-off. As part of this commit HTSU_Result has been renamed to TM_Result, and its members been expanded to differentiated between updating and deleting. HeapUpdateFailureData has been renamed to TM_FailureData. The interface to speculative insertion is changed so nodeModifyTable.c does not have to set the speculative token itself anymore. Instead there's a version of tuple_insert, tuple_insert_speculative, that performs the speculative insertion (without requiring a flag to signal that fact), and the speculative insertion is either made permanent with table_complete_speculative(succeeded = true) or aborted with succeeded = false). Note that multi_insert is not yet routed through tableam, nor is COPY. Changing multi_insert requires changes to copy.c that are large enough to better be done separately. Similarly, although simpler, CREATE TABLE AS and CREATE MATERIALIZED VIEW are also only going to be adjusted in a later commit. Author: Andres Freund and Haribabu Kommi Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20190313003903.nwvrxi7rw3ywhdel@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-24 03:55:57 +01:00
tableam: Rename wrapper functions to match callback names. Some of the wrapper functions didn't match the callback names. Many of them due to staying "consistent" with historic naming of the wrapped functionality. We decided that for most cases it's more important to be for tableam to be consistent going forward, than with the past. The one exception is beginscan/endscan/... because it'd have looked odd to have systable_beginscan/endscan/... with a different naming scheme, and changing the systable_* APIs would have caused way too much churn (including breaking a lot of external users). Author: Ashwin Agrawal, with some small additions by Andres Freund Reviewed-By: Andres Freund Discussion: https://postgr.es/m/CALfoeiugyrXZfX7n0ORCa4L-m834dzmaE8eFdbNR6PMpetU4Ww@mail.gmail.com
2019-05-24 01:25:48 +02:00
/* see table_tuple_insert_speculative() for reference about parameters */
tableam: Add tuple_{insert, delete, update, lock} and use. This adds new, required, table AM callbacks for insert/delete/update and lock_tuple. To be able to reasonably use those, the EvalPlanQual mechanism had to be adapted, moving more logic into the AM. Previously both delete/update/lock call-sites and the EPQ mechanism had to have awareness of the specific tuple format to be able to fetch the latest version of a tuple. Obviously that needs to be abstracted away. To do so, move the logic that find the latest row version into the AM. lock_tuple has a new flag argument, TUPLE_LOCK_FLAG_FIND_LAST_VERSION, that forces it to lock the last version, rather than the current one. It'd have been possible to do so via a separate callback as well, but finding the last version usually also necessitates locking the newest version, making it sensible to combine the two. This replaces the previous use of EvalPlanQualFetch(). Additionally HeapTupleUpdated, which previously signaled either a concurrent update or delete, is now split into two, to avoid callers needing AM specific knowledge to differentiate. The move of finding the latest row version into tuple_lock means that encountering a row concurrently moved into another partition will now raise an error about "tuple to be locked" rather than "tuple to be updated/deleted" - which is accurate, as that always happens when locking rows. While possible slightly less helpful for users, it seems like an acceptable trade-off. As part of this commit HTSU_Result has been renamed to TM_Result, and its members been expanded to differentiated between updating and deleting. HeapUpdateFailureData has been renamed to TM_FailureData. The interface to speculative insertion is changed so nodeModifyTable.c does not have to set the speculative token itself anymore. Instead there's a version of tuple_insert, tuple_insert_speculative, that performs the speculative insertion (without requiring a flag to signal that fact), and the speculative insertion is either made permanent with table_complete_speculative(succeeded = true) or aborted with succeeded = false). Note that multi_insert is not yet routed through tableam, nor is COPY. Changing multi_insert requires changes to copy.c that are large enough to better be done separately. Similarly, although simpler, CREATE TABLE AS and CREATE MATERIALIZED VIEW are also only going to be adjusted in a later commit. Author: Andres Freund and Haribabu Kommi Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20190313003903.nwvrxi7rw3ywhdel@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-24 03:55:57 +01:00
void (*tuple_insert_speculative) (Relation rel,
TupleTableSlot *slot,
CommandId cid,
int options,
struct BulkInsertStateData *bistate,
uint32 specToken);
tableam: Rename wrapper functions to match callback names. Some of the wrapper functions didn't match the callback names. Many of them due to staying "consistent" with historic naming of the wrapped functionality. We decided that for most cases it's more important to be for tableam to be consistent going forward, than with the past. The one exception is beginscan/endscan/... because it'd have looked odd to have systable_beginscan/endscan/... with a different naming scheme, and changing the systable_* APIs would have caused way too much churn (including breaking a lot of external users). Author: Ashwin Agrawal, with some small additions by Andres Freund Reviewed-By: Andres Freund Discussion: https://postgr.es/m/CALfoeiugyrXZfX7n0ORCa4L-m834dzmaE8eFdbNR6PMpetU4Ww@mail.gmail.com
2019-05-24 01:25:48 +02:00
/* see table_tuple_complete_speculative() for reference about parameters */
tableam: Add tuple_{insert, delete, update, lock} and use. This adds new, required, table AM callbacks for insert/delete/update and lock_tuple. To be able to reasonably use those, the EvalPlanQual mechanism had to be adapted, moving more logic into the AM. Previously both delete/update/lock call-sites and the EPQ mechanism had to have awareness of the specific tuple format to be able to fetch the latest version of a tuple. Obviously that needs to be abstracted away. To do so, move the logic that find the latest row version into the AM. lock_tuple has a new flag argument, TUPLE_LOCK_FLAG_FIND_LAST_VERSION, that forces it to lock the last version, rather than the current one. It'd have been possible to do so via a separate callback as well, but finding the last version usually also necessitates locking the newest version, making it sensible to combine the two. This replaces the previous use of EvalPlanQualFetch(). Additionally HeapTupleUpdated, which previously signaled either a concurrent update or delete, is now split into two, to avoid callers needing AM specific knowledge to differentiate. The move of finding the latest row version into tuple_lock means that encountering a row concurrently moved into another partition will now raise an error about "tuple to be locked" rather than "tuple to be updated/deleted" - which is accurate, as that always happens when locking rows. While possible slightly less helpful for users, it seems like an acceptable trade-off. As part of this commit HTSU_Result has been renamed to TM_Result, and its members been expanded to differentiated between updating and deleting. HeapUpdateFailureData has been renamed to TM_FailureData. The interface to speculative insertion is changed so nodeModifyTable.c does not have to set the speculative token itself anymore. Instead there's a version of tuple_insert, tuple_insert_speculative, that performs the speculative insertion (without requiring a flag to signal that fact), and the speculative insertion is either made permanent with table_complete_speculative(succeeded = true) or aborted with succeeded = false). Note that multi_insert is not yet routed through tableam, nor is COPY. Changing multi_insert requires changes to copy.c that are large enough to better be done separately. Similarly, although simpler, CREATE TABLE AS and CREATE MATERIALIZED VIEW are also only going to be adjusted in a later commit. Author: Andres Freund and Haribabu Kommi Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20190313003903.nwvrxi7rw3ywhdel@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-24 03:55:57 +01:00
void (*tuple_complete_speculative) (Relation rel,
TupleTableSlot *slot,
uint32 specToken,
bool succeeded);
tableam: Add table_multi_insert() and revamp/speed-up COPY FROM buffering. This adds table_multi_insert(), and converts COPY FROM, the only user of heap_multi_insert, to it. A simple conversion of COPY FROM use slots would have yielded a slowdown when inserting into a partitioned table for some workloads. Different partitions might need different slots (both slot types and their descriptors), and dropping / creating slots when there's constant partition changes is measurable. Thus instead revamp the COPY FROM buffering for partitioned tables to allow to buffer inserts into multiple tables, flushing only when limits are reached across all partition buffers. By only dropping slots when there've been inserts into too many different partitions, the aforementioned overhead is gone. By allowing larger batches, even when there are frequent partition changes, we actuall speed such cases up significantly. By using slots COPY of very narrow rows into unlogged / temporary might slow down very slightly (due to the indirect function calls). Author: David Rowley, Andres Freund, Haribabu Kommi Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20190327054923.t3epfuewxfqdt22e@alap3.anarazel.de
2019-04-05 00:47:19 +02:00
/* see table_multi_insert() for reference about parameters */
void (*multi_insert) (Relation rel, TupleTableSlot **slots, int nslots,
CommandId cid, int options, struct BulkInsertStateData *bistate);
tableam: Rename wrapper functions to match callback names. Some of the wrapper functions didn't match the callback names. Many of them due to staying "consistent" with historic naming of the wrapped functionality. We decided that for most cases it's more important to be for tableam to be consistent going forward, than with the past. The one exception is beginscan/endscan/... because it'd have looked odd to have systable_beginscan/endscan/... with a different naming scheme, and changing the systable_* APIs would have caused way too much churn (including breaking a lot of external users). Author: Ashwin Agrawal, with some small additions by Andres Freund Reviewed-By: Andres Freund Discussion: https://postgr.es/m/CALfoeiugyrXZfX7n0ORCa4L-m834dzmaE8eFdbNR6PMpetU4Ww@mail.gmail.com
2019-05-24 01:25:48 +02:00
/* see table_tuple_delete() for reference about parameters */
tableam: Add tuple_{insert, delete, update, lock} and use. This adds new, required, table AM callbacks for insert/delete/update and lock_tuple. To be able to reasonably use those, the EvalPlanQual mechanism had to be adapted, moving more logic into the AM. Previously both delete/update/lock call-sites and the EPQ mechanism had to have awareness of the specific tuple format to be able to fetch the latest version of a tuple. Obviously that needs to be abstracted away. To do so, move the logic that find the latest row version into the AM. lock_tuple has a new flag argument, TUPLE_LOCK_FLAG_FIND_LAST_VERSION, that forces it to lock the last version, rather than the current one. It'd have been possible to do so via a separate callback as well, but finding the last version usually also necessitates locking the newest version, making it sensible to combine the two. This replaces the previous use of EvalPlanQualFetch(). Additionally HeapTupleUpdated, which previously signaled either a concurrent update or delete, is now split into two, to avoid callers needing AM specific knowledge to differentiate. The move of finding the latest row version into tuple_lock means that encountering a row concurrently moved into another partition will now raise an error about "tuple to be locked" rather than "tuple to be updated/deleted" - which is accurate, as that always happens when locking rows. While possible slightly less helpful for users, it seems like an acceptable trade-off. As part of this commit HTSU_Result has been renamed to TM_Result, and its members been expanded to differentiated between updating and deleting. HeapUpdateFailureData has been renamed to TM_FailureData. The interface to speculative insertion is changed so nodeModifyTable.c does not have to set the speculative token itself anymore. Instead there's a version of tuple_insert, tuple_insert_speculative, that performs the speculative insertion (without requiring a flag to signal that fact), and the speculative insertion is either made permanent with table_complete_speculative(succeeded = true) or aborted with succeeded = false). Note that multi_insert is not yet routed through tableam, nor is COPY. Changing multi_insert requires changes to copy.c that are large enough to better be done separately. Similarly, although simpler, CREATE TABLE AS and CREATE MATERIALIZED VIEW are also only going to be adjusted in a later commit. Author: Andres Freund and Haribabu Kommi Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20190313003903.nwvrxi7rw3ywhdel@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-24 03:55:57 +01:00
TM_Result (*tuple_delete) (Relation rel,
ItemPointer tid,
CommandId cid,
Snapshot snapshot,
Snapshot crosscheck,
bool wait,
TM_FailureData *tmfd,
Allow locking updated tuples in tuple_update() and tuple_delete() Currently, in read committed transaction isolation mode (default), we have the following sequence of actions when tuple_update()/tuple_delete() finds the tuple updated by concurrent transaction. 1. Attempt to update/delete tuple with tuple_update()/tuple_delete(), which returns TM_Updated. 2. Lock tuple with tuple_lock(). 3. Re-evaluate plan qual (recheck if we still need to update/delete and calculate the new tuple for update). 4. Second attempt to update/delete tuple with tuple_update()/tuple_delete(). This attempt should be successful, since the tuple was previously locked. This patch eliminates step 2 by taking the lock during first tuple_update()/tuple_delete() call. Heap table access method saves some efforts by checking the updated tuple once instead of twice. Future undo-based table access methods, which will start from the latest row version, can immediately place a lock there. The code in nodeModifyTable.c is simplified by removing the nested switch/case. Discussion: https://postgr.es/m/CAPpHfdua-YFw3XTprfutzGp28xXLigFtzNbuFY8yPhqeq6X5kg%40mail.gmail.com Reviewed-by: Aleksander Alekseev, Pavel Borisov, Vignesh C, Mason Sharp Reviewed-by: Andres Freund, Chris Travers
2023-03-22 22:13:37 +01:00
bool changingPart,
LazyTupleTableSlot *lockedSlot);
tableam: Add tuple_{insert, delete, update, lock} and use. This adds new, required, table AM callbacks for insert/delete/update and lock_tuple. To be able to reasonably use those, the EvalPlanQual mechanism had to be adapted, moving more logic into the AM. Previously both delete/update/lock call-sites and the EPQ mechanism had to have awareness of the specific tuple format to be able to fetch the latest version of a tuple. Obviously that needs to be abstracted away. To do so, move the logic that find the latest row version into the AM. lock_tuple has a new flag argument, TUPLE_LOCK_FLAG_FIND_LAST_VERSION, that forces it to lock the last version, rather than the current one. It'd have been possible to do so via a separate callback as well, but finding the last version usually also necessitates locking the newest version, making it sensible to combine the two. This replaces the previous use of EvalPlanQualFetch(). Additionally HeapTupleUpdated, which previously signaled either a concurrent update or delete, is now split into two, to avoid callers needing AM specific knowledge to differentiate. The move of finding the latest row version into tuple_lock means that encountering a row concurrently moved into another partition will now raise an error about "tuple to be locked" rather than "tuple to be updated/deleted" - which is accurate, as that always happens when locking rows. While possible slightly less helpful for users, it seems like an acceptable trade-off. As part of this commit HTSU_Result has been renamed to TM_Result, and its members been expanded to differentiated between updating and deleting. HeapUpdateFailureData has been renamed to TM_FailureData. The interface to speculative insertion is changed so nodeModifyTable.c does not have to set the speculative token itself anymore. Instead there's a version of tuple_insert, tuple_insert_speculative, that performs the speculative insertion (without requiring a flag to signal that fact), and the speculative insertion is either made permanent with table_complete_speculative(succeeded = true) or aborted with succeeded = false). Note that multi_insert is not yet routed through tableam, nor is COPY. Changing multi_insert requires changes to copy.c that are large enough to better be done separately. Similarly, although simpler, CREATE TABLE AS and CREATE MATERIALIZED VIEW are also only going to be adjusted in a later commit. Author: Andres Freund and Haribabu Kommi Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20190313003903.nwvrxi7rw3ywhdel@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-24 03:55:57 +01:00
tableam: Rename wrapper functions to match callback names. Some of the wrapper functions didn't match the callback names. Many of them due to staying "consistent" with historic naming of the wrapped functionality. We decided that for most cases it's more important to be for tableam to be consistent going forward, than with the past. The one exception is beginscan/endscan/... because it'd have looked odd to have systable_beginscan/endscan/... with a different naming scheme, and changing the systable_* APIs would have caused way too much churn (including breaking a lot of external users). Author: Ashwin Agrawal, with some small additions by Andres Freund Reviewed-By: Andres Freund Discussion: https://postgr.es/m/CALfoeiugyrXZfX7n0ORCa4L-m834dzmaE8eFdbNR6PMpetU4Ww@mail.gmail.com
2019-05-24 01:25:48 +02:00
/* see table_tuple_update() for reference about parameters */
tableam: Add tuple_{insert, delete, update, lock} and use. This adds new, required, table AM callbacks for insert/delete/update and lock_tuple. To be able to reasonably use those, the EvalPlanQual mechanism had to be adapted, moving more logic into the AM. Previously both delete/update/lock call-sites and the EPQ mechanism had to have awareness of the specific tuple format to be able to fetch the latest version of a tuple. Obviously that needs to be abstracted away. To do so, move the logic that find the latest row version into the AM. lock_tuple has a new flag argument, TUPLE_LOCK_FLAG_FIND_LAST_VERSION, that forces it to lock the last version, rather than the current one. It'd have been possible to do so via a separate callback as well, but finding the last version usually also necessitates locking the newest version, making it sensible to combine the two. This replaces the previous use of EvalPlanQualFetch(). Additionally HeapTupleUpdated, which previously signaled either a concurrent update or delete, is now split into two, to avoid callers needing AM specific knowledge to differentiate. The move of finding the latest row version into tuple_lock means that encountering a row concurrently moved into another partition will now raise an error about "tuple to be locked" rather than "tuple to be updated/deleted" - which is accurate, as that always happens when locking rows. While possible slightly less helpful for users, it seems like an acceptable trade-off. As part of this commit HTSU_Result has been renamed to TM_Result, and its members been expanded to differentiated between updating and deleting. HeapUpdateFailureData has been renamed to TM_FailureData. The interface to speculative insertion is changed so nodeModifyTable.c does not have to set the speculative token itself anymore. Instead there's a version of tuple_insert, tuple_insert_speculative, that performs the speculative insertion (without requiring a flag to signal that fact), and the speculative insertion is either made permanent with table_complete_speculative(succeeded = true) or aborted with succeeded = false). Note that multi_insert is not yet routed through tableam, nor is COPY. Changing multi_insert requires changes to copy.c that are large enough to better be done separately. Similarly, although simpler, CREATE TABLE AS and CREATE MATERIALIZED VIEW are also only going to be adjusted in a later commit. Author: Andres Freund and Haribabu Kommi Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20190313003903.nwvrxi7rw3ywhdel@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-24 03:55:57 +01:00
TM_Result (*tuple_update) (Relation rel,
ItemPointer otid,
TupleTableSlot *slot,
CommandId cid,
Snapshot snapshot,
Snapshot crosscheck,
bool wait,
TM_FailureData *tmfd,
LockTupleMode *lockmode,
Allow locking updated tuples in tuple_update() and tuple_delete() Currently, in read committed transaction isolation mode (default), we have the following sequence of actions when tuple_update()/tuple_delete() finds the tuple updated by concurrent transaction. 1. Attempt to update/delete tuple with tuple_update()/tuple_delete(), which returns TM_Updated. 2. Lock tuple with tuple_lock(). 3. Re-evaluate plan qual (recheck if we still need to update/delete and calculate the new tuple for update). 4. Second attempt to update/delete tuple with tuple_update()/tuple_delete(). This attempt should be successful, since the tuple was previously locked. This patch eliminates step 2 by taking the lock during first tuple_update()/tuple_delete() call. Heap table access method saves some efforts by checking the updated tuple once instead of twice. Future undo-based table access methods, which will start from the latest row version, can immediately place a lock there. The code in nodeModifyTable.c is simplified by removing the nested switch/case. Discussion: https://postgr.es/m/CAPpHfdua-YFw3XTprfutzGp28xXLigFtzNbuFY8yPhqeq6X5kg%40mail.gmail.com Reviewed-by: Aleksander Alekseev, Pavel Borisov, Vignesh C, Mason Sharp Reviewed-by: Andres Freund, Chris Travers
2023-03-22 22:13:37 +01:00
TU_UpdateIndexes *update_indexes,
LazyTupleTableSlot *lockedSlot);
tableam: Add tuple_{insert, delete, update, lock} and use. This adds new, required, table AM callbacks for insert/delete/update and lock_tuple. To be able to reasonably use those, the EvalPlanQual mechanism had to be adapted, moving more logic into the AM. Previously both delete/update/lock call-sites and the EPQ mechanism had to have awareness of the specific tuple format to be able to fetch the latest version of a tuple. Obviously that needs to be abstracted away. To do so, move the logic that find the latest row version into the AM. lock_tuple has a new flag argument, TUPLE_LOCK_FLAG_FIND_LAST_VERSION, that forces it to lock the last version, rather than the current one. It'd have been possible to do so via a separate callback as well, but finding the last version usually also necessitates locking the newest version, making it sensible to combine the two. This replaces the previous use of EvalPlanQualFetch(). Additionally HeapTupleUpdated, which previously signaled either a concurrent update or delete, is now split into two, to avoid callers needing AM specific knowledge to differentiate. The move of finding the latest row version into tuple_lock means that encountering a row concurrently moved into another partition will now raise an error about "tuple to be locked" rather than "tuple to be updated/deleted" - which is accurate, as that always happens when locking rows. While possible slightly less helpful for users, it seems like an acceptable trade-off. As part of this commit HTSU_Result has been renamed to TM_Result, and its members been expanded to differentiated between updating and deleting. HeapUpdateFailureData has been renamed to TM_FailureData. The interface to speculative insertion is changed so nodeModifyTable.c does not have to set the speculative token itself anymore. Instead there's a version of tuple_insert, tuple_insert_speculative, that performs the speculative insertion (without requiring a flag to signal that fact), and the speculative insertion is either made permanent with table_complete_speculative(succeeded = true) or aborted with succeeded = false). Note that multi_insert is not yet routed through tableam, nor is COPY. Changing multi_insert requires changes to copy.c that are large enough to better be done separately. Similarly, although simpler, CREATE TABLE AS and CREATE MATERIALIZED VIEW are also only going to be adjusted in a later commit. Author: Andres Freund and Haribabu Kommi Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20190313003903.nwvrxi7rw3ywhdel@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-24 03:55:57 +01:00
tableam: Rename wrapper functions to match callback names. Some of the wrapper functions didn't match the callback names. Many of them due to staying "consistent" with historic naming of the wrapped functionality. We decided that for most cases it's more important to be for tableam to be consistent going forward, than with the past. The one exception is beginscan/endscan/... because it'd have looked odd to have systable_beginscan/endscan/... with a different naming scheme, and changing the systable_* APIs would have caused way too much churn (including breaking a lot of external users). Author: Ashwin Agrawal, with some small additions by Andres Freund Reviewed-By: Andres Freund Discussion: https://postgr.es/m/CALfoeiugyrXZfX7n0ORCa4L-m834dzmaE8eFdbNR6PMpetU4Ww@mail.gmail.com
2019-05-24 01:25:48 +02:00
/* see table_tuple_lock() for reference about parameters */
tableam: Add tuple_{insert, delete, update, lock} and use. This adds new, required, table AM callbacks for insert/delete/update and lock_tuple. To be able to reasonably use those, the EvalPlanQual mechanism had to be adapted, moving more logic into the AM. Previously both delete/update/lock call-sites and the EPQ mechanism had to have awareness of the specific tuple format to be able to fetch the latest version of a tuple. Obviously that needs to be abstracted away. To do so, move the logic that find the latest row version into the AM. lock_tuple has a new flag argument, TUPLE_LOCK_FLAG_FIND_LAST_VERSION, that forces it to lock the last version, rather than the current one. It'd have been possible to do so via a separate callback as well, but finding the last version usually also necessitates locking the newest version, making it sensible to combine the two. This replaces the previous use of EvalPlanQualFetch(). Additionally HeapTupleUpdated, which previously signaled either a concurrent update or delete, is now split into two, to avoid callers needing AM specific knowledge to differentiate. The move of finding the latest row version into tuple_lock means that encountering a row concurrently moved into another partition will now raise an error about "tuple to be locked" rather than "tuple to be updated/deleted" - which is accurate, as that always happens when locking rows. While possible slightly less helpful for users, it seems like an acceptable trade-off. As part of this commit HTSU_Result has been renamed to TM_Result, and its members been expanded to differentiated between updating and deleting. HeapUpdateFailureData has been renamed to TM_FailureData. The interface to speculative insertion is changed so nodeModifyTable.c does not have to set the speculative token itself anymore. Instead there's a version of tuple_insert, tuple_insert_speculative, that performs the speculative insertion (without requiring a flag to signal that fact), and the speculative insertion is either made permanent with table_complete_speculative(succeeded = true) or aborted with succeeded = false). Note that multi_insert is not yet routed through tableam, nor is COPY. Changing multi_insert requires changes to copy.c that are large enough to better be done separately. Similarly, although simpler, CREATE TABLE AS and CREATE MATERIALIZED VIEW are also only going to be adjusted in a later commit. Author: Andres Freund and Haribabu Kommi Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20190313003903.nwvrxi7rw3ywhdel@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-24 03:55:57 +01:00
TM_Result (*tuple_lock) (Relation rel,
ItemPointer tid,
Snapshot snapshot,
TupleTableSlot *slot,
CommandId cid,
LockTupleMode mode,
LockWaitPolicy wait_policy,
uint8 flags,
TM_FailureData *tmfd);
tableam: Add table_finish_bulk_insert(). This replaces the previous calls of heap_sync() in places using bulk-insert. By passing in the flags used for bulk-insert the AM can decide (first at insert time and then during the finish call) which of the optimizations apply to it, and what operations are necessary to finish a bulk insert operation. Also change HEAP_INSERT_* flags to TABLE_INSERT, and rename hi_options to ti_options. These changes are made even in copy.c, which hasn't yet been converted to tableam. There's no harm in doing so. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-04-01 23:41:42 +02:00
/*
* Perform operations necessary to complete insertions made via
Skip WAL for new relfilenodes, under wal_level=minimal. Until now, only selected bulk operations (e.g. COPY) did this. If a given relfilenode received both a WAL-skipping COPY and a WAL-logged operation (e.g. INSERT), recovery could lose tuples from the COPY. See src/backend/access/transam/README section "Skipping WAL for New RelFileNode" for the new coding rules. Maintainers of table access methods should examine that section. To maintain data durability, just before commit, we choose between an fsync of the relfilenode and copying its contents to WAL. A new GUC, wal_skip_threshold, guides that choice. If this change slows a workload that creates small, permanent relfilenodes under wal_level=minimal, try adjusting wal_skip_threshold. Users setting a timeout on COMMIT may need to adjust that timeout, and log_min_duration_statement analysis will reflect time consumption moving to COMMIT from commands like COPY. Internally, this requires a reliable determination of whether RollbackAndReleaseCurrentSubTransaction() would unlink a relation's current relfilenode. Introduce rd_firstRelfilenodeSubid. Amend the specification of rd_createSubid such that the field is zero when a new rel has an old rd_node. Make relcache.c retain entries for certain dropped relations until end of transaction. Bump XLOG_PAGE_MAGIC, since this introduces XLOG_GIST_ASSIGN_LSN. Future servers accept older WAL, so this bump is discretionary. Kyotaro Horiguchi, reviewed (in earlier, similar versions) by Robert Haas. Heikki Linnakangas and Michael Paquier implemented earlier designs that materially clarified the problem. Reviewed, in earlier designs, by Andrew Dunstan, Andres Freund, Alvaro Herrera, Tom Lane, Fujii Masao, and Simon Riggs. Reported by Martijn van Oosterhout. Discussion: https://postgr.es/m/20150702220524.GA9392@svana.org
2020-04-04 21:25:34 +02:00
* tuple_insert and multi_insert with a BulkInsertState specified. In-tree
* access methods ceased to use this.
tableam: Add table_finish_bulk_insert(). This replaces the previous calls of heap_sync() in places using bulk-insert. By passing in the flags used for bulk-insert the AM can decide (first at insert time and then during the finish call) which of the optimizations apply to it, and what operations are necessary to finish a bulk insert operation. Also change HEAP_INSERT_* flags to TABLE_INSERT, and rename hi_options to ti_options. These changes are made even in copy.c, which hasn't yet been converted to tableam. There's no harm in doing so. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-04-01 23:41:42 +02:00
*
* Typically callers of tuple_insert and multi_insert will just pass all
tableam: comment and formatting fixes. Author: Heikki Linnakangas Discussion: https://postgr.es/m/9a7fb9cc-2419-5db7-8840-ddc10c93f122@iki.fi
2019-04-08 19:25:16 +02:00
* the flags that apply to them, and each AM has to decide which of them
* make sense for it, and then only take actions in finish_bulk_insert for
* those flags, and ignore others.
tableam: Add table_finish_bulk_insert(). This replaces the previous calls of heap_sync() in places using bulk-insert. By passing in the flags used for bulk-insert the AM can decide (first at insert time and then during the finish call) which of the optimizations apply to it, and what operations are necessary to finish a bulk insert operation. Also change HEAP_INSERT_* flags to TABLE_INSERT, and rename hi_options to ti_options. These changes are made even in copy.c, which hasn't yet been converted to tableam. There's no harm in doing so. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-04-01 23:41:42 +02:00
*
* Optional callback.
*/
void (*finish_bulk_insert) (Relation rel, int options);
tableam: Support for an index build's initial table scan(s). To support building indexes over tables of different AMs, the scans to do so need to be routed through the table AM. While moving a fair amount of code, nearly all the changes are just moving code to below a callback. Currently the range based interface wouldn't make much sense for non block based table AMs. But that seems aceptable for now. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-28 03:59:06 +01:00
/* ------------------------------------------------------------------------
* DDL related functionality.
* ------------------------------------------------------------------------
*/
tableam: relation creation, VACUUM FULL/CLUSTER, SET TABLESPACE. This moves the responsibility for: - creating the storage necessary for a relation, including creating a new relfilenode for a relation with existing storage - non-transactional truncation of a relation - VACUUM FULL / CLUSTER's rewrite of a table below tableam. This is fairly straight forward, with a bit of complexity smattered in to move the computation of xid / multixid horizons below the AM, as they don't make sense for every table AM. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-29 04:01:14 +01:00
/*
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
* This callback needs to create new relation storage for `rel`, with
tableam: relation creation, VACUUM FULL/CLUSTER, SET TABLESPACE. This moves the responsibility for: - creating the storage necessary for a relation, including creating a new relfilenode for a relation with existing storage - non-transactional truncation of a relation - VACUUM FULL / CLUSTER's rewrite of a table below tableam. This is fairly straight forward, with a bit of complexity smattered in to move the computation of xid / multixid horizons below the AM, as they don't make sense for every table AM. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-29 04:01:14 +01:00
* appropriate durability behaviour for `persistence`.
*
Fix several recently introduced issues around handling new relation forks. Most of these stem from d25f519107 "tableam: relation creation, VACUUM FULL/CLUSTER, SET TABLESPACE.". 1) To pass data to the relation_set_new_filenode() RelationSetNewRelfilenode() was made to update RelationData.rd_rel directly. That's not OK however, as it makes the relcache entries temporarily inconsistent. Which among other scenarios is a problem if a REINDEX targets an index on pg_class - the CatalogTupleUpdate() in RelationSetNewRelfilenode(). Presumably that was introduced because other places in the code do so - while those aren't "good practice" they don't appear to be actively buggy (e.g. because system tables may not be targeted). I (Andres) should have caught this while reviewing and signficantly evolving the code in that commit, mea culpa. Fix that by instead passing in the new RelFileNode as separate argument to relation_set_new_filenode() and rely on the relcache to update the catalog entry. Also revert that the RelationMapUpdateMap() call was changed to immediate, and undo some other more unnecessary changes. 2) Document that the relation_set_new_filenode cannot rely on the whole relcache entry to be valid. It might be worthwhile to refactor the code to never have to rely on that, but given the way heap_create() is currently coded, that'd be a large change. 3) ATExecSetTableSpace() shouldn't do FlushRelationBuffers() itself. A table AM might not use shared buffers at all. Move to index_copy_data() and heapam_relation_copy_data(). 4) heapam_relation_set_new_filenode() previously sometimes accessed rel->rd_rel->relpersistence rather than the `persistence` argument. Code movement mistake. 5) Previously heapam_relation_set_new_filenode() re-opened the smgr relation to create the init for, if necesary. Instead have RelationCreateStorage() return the SMgrRelation and use it to create the init fork. 6) Add a note about the danger of modifying the relcache directly to ATExecSetTableSpace() - it's currently not a bug because there's a check ERRORing for catalog tables. Regression tests and assertion improvements that together trigger the bug described in 1) will be added in a later commit, as there is a related bug on all branches. Reported-By: Michael Paquier Diagnosed-By: Tom Lane and Andres Freund Author: Andres Freund Reviewed-By: Tom Lane Discussion: https://postgr.es/m/20190418011430.GA19133@paquier.xyz
2019-04-30 04:28:05 +02:00
* Note that only the subset of the relcache filled by
* RelationBuildLocalRelation() can be relied upon and that the relation's
Fix few typos and minor wordsmithing in tableam comments. Reported-by: Ashwin Agrawal Author: Ashwin Agrawal Reviewed-by: Amit Kapila Backpatch-through: 12, where it was introduced Discussion: https://postgr.es/m/CALfoeisgdZhYDrJOukaBzvXfJOK2FQ0szVMK7dzmcy6w93iDUA@mail.gmail.com
2019-07-10 04:22:51 +02:00
* catalog entries will either not yet exist (new relation), or will still
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
* reference the old relfilelocator.
Fix several recently introduced issues around handling new relation forks. Most of these stem from d25f519107 "tableam: relation creation, VACUUM FULL/CLUSTER, SET TABLESPACE.". 1) To pass data to the relation_set_new_filenode() RelationSetNewRelfilenode() was made to update RelationData.rd_rel directly. That's not OK however, as it makes the relcache entries temporarily inconsistent. Which among other scenarios is a problem if a REINDEX targets an index on pg_class - the CatalogTupleUpdate() in RelationSetNewRelfilenode(). Presumably that was introduced because other places in the code do so - while those aren't "good practice" they don't appear to be actively buggy (e.g. because system tables may not be targeted). I (Andres) should have caught this while reviewing and signficantly evolving the code in that commit, mea culpa. Fix that by instead passing in the new RelFileNode as separate argument to relation_set_new_filenode() and rely on the relcache to update the catalog entry. Also revert that the RelationMapUpdateMap() call was changed to immediate, and undo some other more unnecessary changes. 2) Document that the relation_set_new_filenode cannot rely on the whole relcache entry to be valid. It might be worthwhile to refactor the code to never have to rely on that, but given the way heap_create() is currently coded, that'd be a large change. 3) ATExecSetTableSpace() shouldn't do FlushRelationBuffers() itself. A table AM might not use shared buffers at all. Move to index_copy_data() and heapam_relation_copy_data(). 4) heapam_relation_set_new_filenode() previously sometimes accessed rel->rd_rel->relpersistence rather than the `persistence` argument. Code movement mistake. 5) Previously heapam_relation_set_new_filenode() re-opened the smgr relation to create the init for, if necesary. Instead have RelationCreateStorage() return the SMgrRelation and use it to create the init fork. 6) Add a note about the danger of modifying the relcache directly to ATExecSetTableSpace() - it's currently not a bug because there's a check ERRORing for catalog tables. Regression tests and assertion improvements that together trigger the bug described in 1) will be added in a later commit, as there is a related bug on all branches. Reported-By: Michael Paquier Diagnosed-By: Tom Lane and Andres Freund Author: Andres Freund Reviewed-By: Tom Lane Discussion: https://postgr.es/m/20190418011430.GA19133@paquier.xyz
2019-04-30 04:28:05 +02:00
*
* As output *freezeXid, *minmulti must be set to the values appropriate
tableam: comment and formatting fixes. Author: Heikki Linnakangas Discussion: https://postgr.es/m/9a7fb9cc-2419-5db7-8840-ddc10c93f122@iki.fi
2019-04-08 19:25:16 +02:00
* for pg_class.{relfrozenxid, relminmxid}. For AMs that don't need those
* fields to be filled they can be set to InvalidTransactionId and
* InvalidMultiXactId, respectively.
tableam: relation creation, VACUUM FULL/CLUSTER, SET TABLESPACE. This moves the responsibility for: - creating the storage necessary for a relation, including creating a new relfilenode for a relation with existing storage - non-transactional truncation of a relation - VACUUM FULL / CLUSTER's rewrite of a table below tableam. This is fairly straight forward, with a bit of complexity smattered in to move the computation of xid / multixid horizons below the AM, as they don't make sense for every table AM. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-29 04:01:14 +01:00
*
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
* See also table_relation_set_new_filelocator().
tableam: relation creation, VACUUM FULL/CLUSTER, SET TABLESPACE. This moves the responsibility for: - creating the storage necessary for a relation, including creating a new relfilenode for a relation with existing storage - non-transactional truncation of a relation - VACUUM FULL / CLUSTER's rewrite of a table below tableam. This is fairly straight forward, with a bit of complexity smattered in to move the computation of xid / multixid horizons below the AM, as they don't make sense for every table AM. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-29 04:01:14 +01:00
*/
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
void (*relation_set_new_filelocator) (Relation rel,
const RelFileLocator *newrlocator,
char persistence,
TransactionId *freezeXid,
MultiXactId *minmulti);
tableam: relation creation, VACUUM FULL/CLUSTER, SET TABLESPACE. This moves the responsibility for: - creating the storage necessary for a relation, including creating a new relfilenode for a relation with existing storage - non-transactional truncation of a relation - VACUUM FULL / CLUSTER's rewrite of a table below tableam. This is fairly straight forward, with a bit of complexity smattered in to move the computation of xid / multixid horizons below the AM, as they don't make sense for every table AM. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-29 04:01:14 +01:00
/*
* This callback needs to remove all contents from `rel`'s current
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
* relfilelocator. No provisions for transactional behaviour need to be
* made. Often this can be implemented by truncating the underlying
* storage to its minimal size.
tableam: relation creation, VACUUM FULL/CLUSTER, SET TABLESPACE. This moves the responsibility for: - creating the storage necessary for a relation, including creating a new relfilenode for a relation with existing storage - non-transactional truncation of a relation - VACUUM FULL / CLUSTER's rewrite of a table below tableam. This is fairly straight forward, with a bit of complexity smattered in to move the computation of xid / multixid horizons below the AM, as they don't make sense for every table AM. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-29 04:01:14 +01:00
*
* See also table_relation_nontransactional_truncate().
*/
void (*relation_nontransactional_truncate) (Relation rel);
/*
* See table_relation_copy_data().
*
* This can typically be implemented by directly copying the underlying
* storage, unless it contains references to the tablespace internally.
*/
Fix several recently introduced issues around handling new relation forks. Most of these stem from d25f519107 "tableam: relation creation, VACUUM FULL/CLUSTER, SET TABLESPACE.". 1) To pass data to the relation_set_new_filenode() RelationSetNewRelfilenode() was made to update RelationData.rd_rel directly. That's not OK however, as it makes the relcache entries temporarily inconsistent. Which among other scenarios is a problem if a REINDEX targets an index on pg_class - the CatalogTupleUpdate() in RelationSetNewRelfilenode(). Presumably that was introduced because other places in the code do so - while those aren't "good practice" they don't appear to be actively buggy (e.g. because system tables may not be targeted). I (Andres) should have caught this while reviewing and signficantly evolving the code in that commit, mea culpa. Fix that by instead passing in the new RelFileNode as separate argument to relation_set_new_filenode() and rely on the relcache to update the catalog entry. Also revert that the RelationMapUpdateMap() call was changed to immediate, and undo some other more unnecessary changes. 2) Document that the relation_set_new_filenode cannot rely on the whole relcache entry to be valid. It might be worthwhile to refactor the code to never have to rely on that, but given the way heap_create() is currently coded, that'd be a large change. 3) ATExecSetTableSpace() shouldn't do FlushRelationBuffers() itself. A table AM might not use shared buffers at all. Move to index_copy_data() and heapam_relation_copy_data(). 4) heapam_relation_set_new_filenode() previously sometimes accessed rel->rd_rel->relpersistence rather than the `persistence` argument. Code movement mistake. 5) Previously heapam_relation_set_new_filenode() re-opened the smgr relation to create the init for, if necesary. Instead have RelationCreateStorage() return the SMgrRelation and use it to create the init fork. 6) Add a note about the danger of modifying the relcache directly to ATExecSetTableSpace() - it's currently not a bug because there's a check ERRORing for catalog tables. Regression tests and assertion improvements that together trigger the bug described in 1) will be added in a later commit, as there is a related bug on all branches. Reported-By: Michael Paquier Diagnosed-By: Tom Lane and Andres Freund Author: Andres Freund Reviewed-By: Tom Lane Discussion: https://postgr.es/m/20190418011430.GA19133@paquier.xyz
2019-04-30 04:28:05 +02:00
void (*relation_copy_data) (Relation rel,
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
const RelFileLocator *newrlocator);
tableam: relation creation, VACUUM FULL/CLUSTER, SET TABLESPACE. This moves the responsibility for: - creating the storage necessary for a relation, including creating a new relfilenode for a relation with existing storage - non-transactional truncation of a relation - VACUUM FULL / CLUSTER's rewrite of a table below tableam. This is fairly straight forward, with a bit of complexity smattered in to move the computation of xid / multixid horizons below the AM, as they don't make sense for every table AM. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-29 04:01:14 +01:00
/* See table_relation_copy_for_cluster() */
Fix some typos and inconsistencies in tableam.h The defined callback definitions have been using references to heap for a couple of variables and comments. This makes the whole interface more consistent by using "table" which is more generic. A variable storing index information was misspelled as well. Author: Michael Paquier Discussion: https://postgr.es/m/20190601190946.GB1905@paquier.xyz
2019-06-04 02:48:25 +02:00
void (*relation_copy_for_cluster) (Relation NewTable,
Relation OldTable,
tableam: Formatting and other minor cleanups. The superflous heapam_xlog.h includes were reported by Peter Geoghegan.
2019-03-31 05:13:56 +02:00
Relation OldIndex,
tableam: relation creation, VACUUM FULL/CLUSTER, SET TABLESPACE. This moves the responsibility for: - creating the storage necessary for a relation, including creating a new relfilenode for a relation with existing storage - non-transactional truncation of a relation - VACUUM FULL / CLUSTER's rewrite of a table below tableam. This is fairly straight forward, with a bit of complexity smattered in to move the computation of xid / multixid horizons below the AM, as they don't make sense for every table AM. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-29 04:01:14 +01:00
bool use_sort,
tableam: Formatting and other minor cleanups. The superflous heapam_xlog.h includes were reported by Peter Geoghegan.
2019-03-31 05:13:56 +02:00
TransactionId OldestXmin,
Allow pg_class xid & multixid horizons to not be set. This allows table AMs that don't need these horizons. This was already documented in the tableam relation_set_new_filenode callback, but an assert prevented if from actually working (the test AM code contained the change itself). Defang the asserts in the general code, and move the stronger ones into heap AM. Relatedly, after CLUSTER/VACUUM, we'd always assign a relfrozenxid / relminmxid. Change the table_relation_copy_for_cluster() interface to allow the AM to overwrite the horizons that get set on the pg_class entry. This'd also in the future allow AMs like heap to compute a relfrozenxid during rewrite that's the table's actual minimum rather than a pre-determined value. Arguably it'd have been better to move the whole computation / setting of those values into the callback, but it seems likely that for other reasons it'd be better to be able to use one value to vacuum/cluster multiple tables (e.g. a toast's horizon shouldn't be different than the table's). Reported-By: Heikki Linnakangas Author: Andres Freund Discussion: https://postgr.es/m/9a7fb9cc-2419-5db7-8840-ddc10c93f122@iki.fi
2019-04-24 06:42:12 +02:00
TransactionId *xid_cutoff,
MultiXactId *multi_cutoff,
tableam: Formatting and other minor cleanups. The superflous heapam_xlog.h includes were reported by Peter Geoghegan.
2019-03-31 05:13:56 +02:00
double *num_tuples,
double *tups_vacuumed,
double *tups_recently_dead);
tableam: relation creation, VACUUM FULL/CLUSTER, SET TABLESPACE. This moves the responsibility for: - creating the storage necessary for a relation, including creating a new relfilenode for a relation with existing storage - non-transactional truncation of a relation - VACUUM FULL / CLUSTER's rewrite of a table below tableam. This is fairly straight forward, with a bit of complexity smattered in to move the computation of xid / multixid horizons below the AM, as they don't make sense for every table AM. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-29 04:01:14 +01:00
tableam: VACUUM and ANALYZE support. This is a relatively straightforward move of the current implementation to sit below tableam. As the current analyze sampling implementation is pretty inherently block based, the tableam analyze interface is as well. It might make sense to generalize that at some point, but that seems like a larger project that shouldn't be undertaken at the same time as the introduction of tableam. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-31 00:21:09 +01:00
/*
tableam: comment improvements. Author: Brad DeJong Discussion: https://postgr.es/m/CAJnrtnxDYOQFsDfWz2iri0T_fFL2ZbbzgCOE=4yaMcszgcsf4A@mail.gmail.com Backpatch: 12-
2019-07-18 04:39:54 +02:00
* React to VACUUM command on the relation. The VACUUM can be triggered by
* a user or by autovacuum. The specific actions performed by the AM will
* depend heavily on the individual AM.
tableam: VACUUM and ANALYZE support. This is a relatively straightforward move of the current implementation to sit below tableam. As the current analyze sampling implementation is pretty inherently block based, the tableam analyze interface is as well. It might make sense to generalize that at some point, but that seems like a larger project that shouldn't be undertaken at the same time as the introduction of tableam. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-31 00:21:09 +01:00
*
* On entry a transaction is already established, and the relation is
* locked with a ShareUpdateExclusive lock.
*
* Note that neither VACUUM FULL (and CLUSTER), nor ANALYZE go through
tableam: comment and formatting fixes. Author: Heikki Linnakangas Discussion: https://postgr.es/m/9a7fb9cc-2419-5db7-8840-ddc10c93f122@iki.fi
2019-04-08 19:25:16 +02:00
* this routine, even if (for ANALYZE) it is part of the same VACUUM
tableam: VACUUM and ANALYZE support. This is a relatively straightforward move of the current implementation to sit below tableam. As the current analyze sampling implementation is pretty inherently block based, the tableam analyze interface is as well. It might make sense to generalize that at some point, but that seems like a larger project that shouldn't be undertaken at the same time as the introduction of tableam. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-31 00:21:09 +01:00
* command.
*
* There probably, in the future, needs to be a separate callback to
* integrate with autovacuum's scheduling.
*/
Simplify state managed by VACUUM. Reorganize the state struct used by VACUUM -- group related items together to make it easier to understand. Also stop relying on stack variables inside lazy_scan_heap() -- move those into the state struct instead. Doing things this way simplifies large groups of related functions whose function signatures had a lot of unnecessary redundancy. Switch over to using int64 for the struct fields used to count things that are reported to the user via log_autovacuum and VACUUM VERBOSE output. We were using double, but that doesn't seem to have any advantages. Using int64 makes it possible to add assertions that verify that the first pass over the heap (pruning) encounters precisely the same number of LP_DEAD items that get deleted from indexes later on, in the second pass over the heap. These assertions will be added in later commits. Finally, adjust the signatures of functions with IndexBulkDeleteResult pointer arguments in cases where there was ambiguity about whether or not the argument relates to a single index or all indexes. Functions now use the idiom that both ambulkdelete() and amvacuumcleanup() have always used (where appropriate): accept a mutable IndexBulkDeleteResult pointer argument, and return a result IndexBulkDeleteResult pointer to caller. Author: Peter Geoghegan <pg@bowt.ie> Reviewed-By: Masahiko Sawada <sawada.mshk@gmail.com> Reviewed-By: Robert Haas <robertmhaas@gmail.com> Discussion: https://postgr.es/m/CAH2-WzkeOSYwC6KNckbhk2b1aNnWum6Yyn0NKP9D-Hq1LGTDPw@mail.gmail.com
2021-04-05 22:21:44 +02:00
void (*relation_vacuum) (Relation rel,
tableam: Formatting and other minor cleanups. The superflous heapam_xlog.h includes were reported by Peter Geoghegan.
2019-03-31 05:13:56 +02:00
struct VacuumParams *params,
tableam: VACUUM and ANALYZE support. This is a relatively straightforward move of the current implementation to sit below tableam. As the current analyze sampling implementation is pretty inherently block based, the tableam analyze interface is as well. It might make sense to generalize that at some point, but that seems like a larger project that shouldn't be undertaken at the same time as the introduction of tableam. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-31 00:21:09 +01:00
BufferAccessStrategy bstrategy);
/*
* Prepare to analyze block `blockno` of `scan`. The scan has been started
* with table_beginscan_analyze(). See also
* table_scan_analyze_next_block().
*
* The callback may acquire resources like locks that are held until
* table_scan_analyze_next_tuple() returns false. It e.g. can make sense
* to hold a lock until all tuples on a block have been analyzed by
* scan_analyze_next_tuple.
*
* The callback can return false if the block is not suitable for
* sampling, e.g. because it's a metapage that could never contain tuples.
*
* XXX: This obviously is primarily suited for block-based AMs. It's not
tableam: comment and formatting fixes. Author: Heikki Linnakangas Discussion: https://postgr.es/m/9a7fb9cc-2419-5db7-8840-ddc10c93f122@iki.fi
2019-04-08 19:25:16 +02:00
* clear what a good interface for non block based AMs would be, so there
* isn't one yet.
tableam: VACUUM and ANALYZE support. This is a relatively straightforward move of the current implementation to sit below tableam. As the current analyze sampling implementation is pretty inherently block based, the tableam analyze interface is as well. It might make sense to generalize that at some point, but that seems like a larger project that shouldn't be undertaken at the same time as the introduction of tableam. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-31 00:21:09 +01:00
*/
bool (*scan_analyze_next_block) (TableScanDesc scan,
BlockNumber blockno,
BufferAccessStrategy bstrategy);
/*
* See table_scan_analyze_next_tuple().
*
* Not every AM might have a meaningful concept of dead rows, in which
* case it's OK to not increment *deadrows - but note that that may
* influence autovacuum scheduling (see comment for relation_vacuum
* callback).
*/
bool (*scan_analyze_next_tuple) (TableScanDesc scan,
TransactionId OldestXmin,
double *liverows,
double *deadrows,
TupleTableSlot *slot);
tableam: Support for an index build's initial table scan(s). To support building indexes over tables of different AMs, the scans to do so need to be routed through the table AM. While moving a fair amount of code, nearly all the changes are just moving code to below a callback. Currently the range based interface wouldn't make much sense for non block based table AMs. But that seems aceptable for now. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-28 03:59:06 +01:00
/* see table_index_build_range_scan for reference about parameters */
Fix some typos and inconsistencies in tableam.h The defined callback definitions have been using references to heap for a couple of variables and comments. This makes the whole interface more consistent by using "table" which is more generic. A variable storing index information was misspelled as well. Author: Michael Paquier Discussion: https://postgr.es/m/20190601190946.GB1905@paquier.xyz
2019-06-04 02:48:25 +02:00
double (*index_build_range_scan) (Relation table_rel,
tableam: Support for an index build's initial table scan(s). To support building indexes over tables of different AMs, the scans to do so need to be routed through the table AM. While moving a fair amount of code, nearly all the changes are just moving code to below a callback. Currently the range based interface wouldn't make much sense for non block based table AMs. But that seems aceptable for now. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-28 03:59:06 +01:00
Relation index_rel,
Fix some typos and inconsistencies in tableam.h The defined callback definitions have been using references to heap for a couple of variables and comments. This makes the whole interface more consistent by using "table" which is more generic. A variable storing index information was misspelled as well. Author: Michael Paquier Discussion: https://postgr.es/m/20190601190946.GB1905@paquier.xyz
2019-06-04 02:48:25 +02:00
struct IndexInfo *index_info,
tableam: Support for an index build's initial table scan(s). To support building indexes over tables of different AMs, the scans to do so need to be routed through the table AM. While moving a fair amount of code, nearly all the changes are just moving code to below a callback. Currently the range based interface wouldn't make much sense for non block based table AMs. But that seems aceptable for now. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-28 03:59:06 +01:00
bool allow_sync,
bool anyvisible,
Report progress of CREATE INDEX operations This uses the progress reporting infrastructure added by c16dc1aca5e0, adding support for CREATE INDEX and CREATE INDEX CONCURRENTLY. There are two pieces to this: one is index-AM-agnostic, and the other is AM-specific. The latter is fairly elaborate for btrees, including reportage for parallel index builds and the separate phases that btree index creation uses; other index AMs, which are much simpler in their building procedures, have simplistic reporting only, but that seems sufficient, at least for non-concurrent builds. The index-AM-agnostic part is fairly complete, providing insight into the CONCURRENTLY wait phases as well as block-based progress during the index validation table scan. (The index validation index scan requires patching each AM, which has not been included here.) Reviewers: Rahila Syed, Pavan Deolasee, Tatsuro Yamada Discussion: https://postgr.es/m/20181220220022.mg63bhk26zdpvmcj@alvherre.pgsql
2019-04-02 20:18:08 +02:00
bool progress,
tableam: Support for an index build's initial table scan(s). To support building indexes over tables of different AMs, the scans to do so need to be routed through the table AM. While moving a fair amount of code, nearly all the changes are just moving code to below a callback. Currently the range based interface wouldn't make much sense for non block based table AMs. But that seems aceptable for now. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-28 03:59:06 +01:00
BlockNumber start_blockno,
tableam: Fix index_build_range_scan parameter name. All of the other code thinks that the 8th parameter is the number of blocks, but this declaration thinks that it's the ending block number. Repair this inconsistency. Patch by me, reviewed by Andres Freund. Discussion: http://postgr.es/m/CA+TgmoY49ManQWnJtiwkuytXBkmyTuDFqb74Pr4Zn2Nq9TuNBQ@mail.gmail.com
2019-06-11 02:04:44 +02:00
BlockNumber numblocks,
tableam: Support for an index build's initial table scan(s). To support building indexes over tables of different AMs, the scans to do so need to be routed through the table AM. While moving a fair amount of code, nearly all the changes are just moving code to below a callback. Currently the range based interface wouldn't make much sense for non block based table AMs. But that seems aceptable for now. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-28 03:59:06 +01:00
IndexBuildCallback callback,
void *callback_state,
TableScanDesc scan);
/* see table_index_validate_scan for reference about parameters */
Fix some typos and inconsistencies in tableam.h The defined callback definitions have been using references to heap for a couple of variables and comments. This makes the whole interface more consistent by using "table" which is more generic. A variable storing index information was misspelled as well. Author: Michael Paquier Discussion: https://postgr.es/m/20190601190946.GB1905@paquier.xyz
2019-06-04 02:48:25 +02:00
void (*index_validate_scan) (Relation table_rel,
tableam: Support for an index build's initial table scan(s). To support building indexes over tables of different AMs, the scans to do so need to be routed through the table AM. While moving a fair amount of code, nearly all the changes are just moving code to below a callback. Currently the range based interface wouldn't make much sense for non block based table AMs. But that seems aceptable for now. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-28 03:59:06 +01:00
Relation index_rel,
struct IndexInfo *index_info,
Snapshot snapshot,
struct ValidateIndexState *state);
tableam: Move heap specific logic from estimate_rel_size below tableam. This just moves the table/matview[/toast] determination of relation size to a callback, and uses a copy of the existing logic to implement that callback for heap. It probably would make sense to also move the index specific logic into a callback, so the metapage handling (and probably more) can be index specific. But that's a separate task. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-31 00:40:33 +01:00
tableam: Don't assume that every AM uses md.c style storage. Previously various parts of the code routed size requests through RelationGetNumberOfBlocks[InFork]. That works if md.c is used by the AM, but not otherwise. Add a tableam callback to return the size of the table. As not every AM will use postgres' BLCKSZ, have it return bytes, and have RelationGetNumberOfBlocksInFork() round the byte size up into blocks. To allow code outside of the AM to determine the actual relation size map InvalidForkNumber the total size of a relation, as not every AM might just need the postgres defined forks. A few users of RelationGetNumberOfBlocks() ought to be converted away from that. One case, the use of it to determine whether a tid is valid, will be fixed in a follow up commit. Others will have to wait for v13. Author: Andres Freund Discussion: https://postgr.es/m/20190423225201.3bbv6tbqzkb5w7cw@alap3.anarazel.de
2019-05-18 03:06:18 +02:00
/* ------------------------------------------------------------------------
* Miscellaneous functions.
* ------------------------------------------------------------------------
*/
/*
* See table_relation_size().
*
tableam: Avoid relying on relation size to determine validity of tids. Instead add a tableam callback to do so. To avoid adding per validation overhead, pass a scan to tuple_tid_valid. In heap's case we'd otherwise incurred a RelationGetNumberOfBlocks() call for each tid - which'd have added noticable overhead to nodeTidscan.c. Author: Andres Freund Reviewed-By: Ashwin Agrawal Discussion: https://postgr.es/m/20190515185447.gno2jtqxyktylyvs@alap3.anarazel.de
2019-05-18 03:52:01 +02:00
* Note that currently a few callers use the MAIN_FORKNUM size to figure
* out the range of potentially interesting blocks (brin, analyze). It's
* probable that we'll need to revise the interface for those at some
* point.
tableam: Don't assume that every AM uses md.c style storage. Previously various parts of the code routed size requests through RelationGetNumberOfBlocks[InFork]. That works if md.c is used by the AM, but not otherwise. Add a tableam callback to return the size of the table. As not every AM will use postgres' BLCKSZ, have it return bytes, and have RelationGetNumberOfBlocksInFork() round the byte size up into blocks. To allow code outside of the AM to determine the actual relation size map InvalidForkNumber the total size of a relation, as not every AM might just need the postgres defined forks. A few users of RelationGetNumberOfBlocks() ought to be converted away from that. One case, the use of it to determine whether a tid is valid, will be fixed in a follow up commit. Others will have to wait for v13. Author: Andres Freund Discussion: https://postgr.es/m/20190423225201.3bbv6tbqzkb5w7cw@alap3.anarazel.de
2019-05-18 03:06:18 +02:00
*/
uint64 (*relation_size) (Relation rel, ForkNumber forkNumber);
tableam: Move heap-specific logic from needs_toast_table below tableam. This allows table AMs to completely suppress TOAST table creation, or to modify the conditions under which they are created. Patch by me. Reviewed by Andres Freund. Discussion: http://postgr.es/m/CA+Tgmoa4O2n=yphqD2pERUnYmUO84bH1SqMsA-nSxBGsZ7gWfA@mail.gmail.com
2019-05-21 17:57:13 +02:00
/*
* This callback should return true if the relation requires a TOAST table
* and false if it does not. It may wish to examine the relation's tuple
* descriptor before making a decision, but if it uses some other method
* of storing large values (or if it does not support them) it can simply
* return false.
*/
bool (*relation_needs_toast_table) (Relation rel);
tableam: Allow choice of toast AM. Previously, the toast table had to be implemented by the same AM that was used for the main table, which was bad, because the detoasting code won't work with anything but heap. This commit doesn't fix the latter problem, although there's another patch coming which does, but it does let you pick something that works (i.e. heap, right now). Patch by me, reviewed by Andres Freund. Discussion: http://postgr.es/m/CA+TgmoZv-=2iWM4jcw5ZhJeL18HF96+W1yJeYrnGMYdkFFnEpQ@mail.gmail.com
2020-01-07 20:23:25 +01:00
/*
* This callback should return the OID of the table AM that implements
* TOAST tables for this AM. If the relation_needs_toast_table callback
* always returns false, this callback is not required.
*/
Oid (*relation_toast_am) (Relation rel);
tableam: New callback relation_fetch_toast_slice. Instead of always calling heap_fetch_toast_slice during detoasting, invoke a table AM callback which, when the toast table is a heap table, will be heap_fetch_toast_slice. This makes it possible for a table AM other than heap to be used as a TOAST table. It also completes the series of commits intended to improve the interaction of tableam with TOAST that began with commit 8b94dab06617ef80a0901ab103ebd8754427ef5a; detoast.c is now, hopefully, fully AM-independent. Patch by me, reviewed by Andres Freund and Peter Eisentraut. Discussion: http://postgr.es/m/CA+TgmoZv-=2iWM4jcw5ZhJeL18HF96+W1yJeYrnGMYdkFFnEpQ@mail.gmail.com
2020-01-07 20:35:48 +01:00
/*
* This callback is invoked when detoasting a value stored in a toast
* table implemented by this AM. See table_relation_fetch_toast_slice()
* for more details.
*/
void (*relation_fetch_toast_slice) (Relation toastrel, Oid valueid,
int32 attrsize,
int32 sliceoffset,
int32 slicelength,
struct varlena *result);
tableam: Move heap-specific logic from needs_toast_table below tableam. This allows table AMs to completely suppress TOAST table creation, or to modify the conditions under which they are created. Patch by me. Reviewed by Andres Freund. Discussion: http://postgr.es/m/CA+Tgmoa4O2n=yphqD2pERUnYmUO84bH1SqMsA-nSxBGsZ7gWfA@mail.gmail.com
2019-05-21 17:57:13 +02:00
tableam: Move heap specific logic from estimate_rel_size below tableam. This just moves the table/matview[/toast] determination of relation size to a callback, and uses a copy of the existing logic to implement that callback for heap. It probably would make sense to also move the index specific logic into a callback, so the metapage handling (and probably more) can be index specific. But that's a separate task. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-31 00:40:33 +01:00
/* ------------------------------------------------------------------------
* Planner related functions.
* ------------------------------------------------------------------------
*/
/*
* See table_relation_estimate_size().
*
tableam: comment and formatting fixes. Author: Heikki Linnakangas Discussion: https://postgr.es/m/9a7fb9cc-2419-5db7-8840-ddc10c93f122@iki.fi
2019-04-08 19:25:16 +02:00
* While block oriented, it shouldn't be too hard for an AM that doesn't
Fix few typos and minor wordsmithing in tableam comments. Reported-by: Ashwin Agrawal Author: Ashwin Agrawal Reviewed-by: Amit Kapila Backpatch-through: 12, where it was introduced Discussion: https://postgr.es/m/CALfoeisgdZhYDrJOukaBzvXfJOK2FQ0szVMK7dzmcy6w93iDUA@mail.gmail.com
2019-07-10 04:22:51 +02:00
* internally use blocks to convert into a usable representation.
tableam: Don't assume that every AM uses md.c style storage. Previously various parts of the code routed size requests through RelationGetNumberOfBlocks[InFork]. That works if md.c is used by the AM, but not otherwise. Add a tableam callback to return the size of the table. As not every AM will use postgres' BLCKSZ, have it return bytes, and have RelationGetNumberOfBlocksInFork() round the byte size up into blocks. To allow code outside of the AM to determine the actual relation size map InvalidForkNumber the total size of a relation, as not every AM might just need the postgres defined forks. A few users of RelationGetNumberOfBlocks() ought to be converted away from that. One case, the use of it to determine whether a tid is valid, will be fixed in a follow up commit. Others will have to wait for v13. Author: Andres Freund Discussion: https://postgr.es/m/20190423225201.3bbv6tbqzkb5w7cw@alap3.anarazel.de
2019-05-18 03:06:18 +02:00
*
* This differs from the relation_size callback by returning size
* estimates (both relation size and tuple count) for planning purposes,
* rather than returning a currently correct estimate.
tableam: Move heap specific logic from estimate_rel_size below tableam. This just moves the table/matview[/toast] determination of relation size to a callback, and uses a copy of the existing logic to implement that callback for heap. It probably would make sense to also move the index specific logic into a callback, so the metapage handling (and probably more) can be index specific. But that's a separate task. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-31 00:40:33 +01:00
*/
void (*relation_estimate_size) (Relation rel, int32 *attr_widths,
BlockNumber *pages, double *tuples,
double *allvisfrac);
tableam: sample scan. This moves sample scan support to below tableam. It's not optional as there is, in contrast to e.g. bitmap heap scans, no alternative way to perform tablesample queries. If an AM can't deal with the block based API, it will have to throw an ERROR. The tableam callbacks for this are block based, but given the current TsmRoutine interface, that seems to be required. The new interface doesn't require TsmRoutines to perform visibility checks anymore - that requires the TsmRoutine to know details about the AM, which we want to avoid. To continue to allow taking the returned number of tuples account SampleScanState now has a donetuples field (which previously e.g. existed in SystemRowsSamplerData), which is only incremented after the visibility check succeeds. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-31 05:18:53 +02:00
/* ------------------------------------------------------------------------
* Executor related functions.
* ------------------------------------------------------------------------
*/
/*
tableam: bitmap table scan. This moves bitmap heap scan support to below an optional tableam callback. It's optional as the whole concept of bitmap heapscans is fairly block specific. This basically moves the work previously done in bitgetpage() into the new scan_bitmap_next_block callback, and the direct poking into the buffer done in BitmapHeapNext() into the new scan_bitmap_next_tuple() callback. The abstraction is currently somewhat leaky because nodeBitmapHeapscan.c's prefetching and visibilitymap based logic remains - it's likely that we'll later have to move more into the AM. But it's not trivial to do so without introducing a significant amount of code duplication between the AMs, so that's a project for later. Note that now nodeBitmapHeapscan.c and the associated node types are a bit misnamed. But it's not clear whether renaming wouldn't be a cure worse than the disease. Either way, that'd be best done in a separate commit. Author: Andres Freund Reviewed-By: Robert Haas (in an older version) Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-04-01 02:51:49 +02:00
* Prepare to fetch / check / return tuples from `tbmres->blockno` as part
* of a bitmap table scan. `scan` was started via table_beginscan_bm().
tableam: comment and formatting fixes. Author: Heikki Linnakangas Discussion: https://postgr.es/m/9a7fb9cc-2419-5db7-8840-ddc10c93f122@iki.fi
2019-04-08 19:25:16 +02:00
* Return false if there are no tuples to be found on the page, true
tableam: bitmap table scan. This moves bitmap heap scan support to below an optional tableam callback. It's optional as the whole concept of bitmap heapscans is fairly block specific. This basically moves the work previously done in bitgetpage() into the new scan_bitmap_next_block callback, and the direct poking into the buffer done in BitmapHeapNext() into the new scan_bitmap_next_tuple() callback. The abstraction is currently somewhat leaky because nodeBitmapHeapscan.c's prefetching and visibilitymap based logic remains - it's likely that we'll later have to move more into the AM. But it's not trivial to do so without introducing a significant amount of code duplication between the AMs, so that's a project for later. Note that now nodeBitmapHeapscan.c and the associated node types are a bit misnamed. But it's not clear whether renaming wouldn't be a cure worse than the disease. Either way, that'd be best done in a separate commit. Author: Andres Freund Reviewed-By: Robert Haas (in an older version) Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-04-01 02:51:49 +02:00
* otherwise.
*
* This will typically read and pin the target block, and do the necessary
* work to allow scan_bitmap_next_tuple() to return tuples (e.g. it might
* make sense to perform tuple visibility checks at this time). For some
* AMs it will make more sense to do all the work referencing `tbmres`
* contents here, for others it might be better to defer more work to
* scan_bitmap_next_tuple.
*
* If `tbmres->blockno` is -1, this is a lossy scan and all visible tuples
* on the page have to be returned, otherwise the tuples at offsets in
* `tbmres->offsets` need to be returned.
*
* XXX: Currently this may only be implemented if the AM uses md.c as its
* storage manager, and uses ItemPointer->ip_blkid in a manner that maps
* blockids directly to the underlying storage. nodeBitmapHeapscan.c
* performs prefetching directly using that interface. This probably
* needs to be rectified at a later point.
*
* XXX: Currently this may only be implemented if the AM uses the
* visibilitymap, as nodeBitmapHeapscan.c unconditionally accesses it to
* perform prefetching. This probably needs to be rectified at a later
* point.
*
* Optional callback, but either both scan_bitmap_next_block and
* scan_bitmap_next_tuple need to exist, or neither.
*/
bool (*scan_bitmap_next_block) (TableScanDesc scan,
struct TBMIterateResult *tbmres);
/*
* Fetch the next tuple of a bitmap table scan into `slot` and return true
* if a visible tuple was found, false otherwise.
*
* For some AMs it will make more sense to do all the work referencing
* `tbmres` contents in scan_bitmap_next_block, for others it might be
* better to defer more work to this callback.
*
* Optional callback, but either both scan_bitmap_next_block and
* scan_bitmap_next_tuple need to exist, or neither.
*/
bool (*scan_bitmap_next_tuple) (TableScanDesc scan,
struct TBMIterateResult *tbmres,
TupleTableSlot *slot);
/*
* Prepare to fetch tuples from the next block in a sample scan. Return
* false if the sample scan is finished, true otherwise. `scan` was
* started via table_beginscan_sampling().
tableam: sample scan. This moves sample scan support to below tableam. It's not optional as there is, in contrast to e.g. bitmap heap scans, no alternative way to perform tablesample queries. If an AM can't deal with the block based API, it will have to throw an ERROR. The tableam callbacks for this are block based, but given the current TsmRoutine interface, that seems to be required. The new interface doesn't require TsmRoutines to perform visibility checks anymore - that requires the TsmRoutine to know details about the AM, which we want to avoid. To continue to allow taking the returned number of tuples account SampleScanState now has a donetuples field (which previously e.g. existed in SystemRowsSamplerData), which is only incremented after the visibility check succeeds. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-31 05:18:53 +02:00
*
tableam: comment improvements. Author: Brad DeJong Discussion: https://postgr.es/m/CAJnrtnxDYOQFsDfWz2iri0T_fFL2ZbbzgCOE=4yaMcszgcsf4A@mail.gmail.com Backpatch: 12-
2019-07-18 04:39:54 +02:00
* Typically this will first determine the target block by calling the
tableam: sample scan. This moves sample scan support to below tableam. It's not optional as there is, in contrast to e.g. bitmap heap scans, no alternative way to perform tablesample queries. If an AM can't deal with the block based API, it will have to throw an ERROR. The tableam callbacks for this are block based, but given the current TsmRoutine interface, that seems to be required. The new interface doesn't require TsmRoutines to perform visibility checks anymore - that requires the TsmRoutine to know details about the AM, which we want to avoid. To continue to allow taking the returned number of tuples account SampleScanState now has a donetuples field (which previously e.g. existed in SystemRowsSamplerData), which is only incremented after the visibility check succeeds. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-31 05:18:53 +02:00
* TsmRoutine's NextSampleBlock() callback if not NULL, or alternatively
* perform a sequential scan over all blocks. The determined block is
* then typically read and pinned.
*
* As the TsmRoutine interface is block based, a block needs to be passed
* to NextSampleBlock(). If that's not appropriate for an AM, it
* internally needs to perform mapping between the internal and a block
* based representation.
*
* Note that it's not acceptable to hold deadlock prone resources such as
* lwlocks until scan_sample_next_tuple() has exhausted the tuples on the
* block - the tuple is likely to be returned to an upper query node, and
tableam: comment and formatting fixes. Author: Heikki Linnakangas Discussion: https://postgr.es/m/9a7fb9cc-2419-5db7-8840-ddc10c93f122@iki.fi
2019-04-08 19:25:16 +02:00
* the next call could be off a long while. Holding buffer pins and such
* is obviously OK.
tableam: sample scan. This moves sample scan support to below tableam. It's not optional as there is, in contrast to e.g. bitmap heap scans, no alternative way to perform tablesample queries. If an AM can't deal with the block based API, it will have to throw an ERROR. The tableam callbacks for this are block based, but given the current TsmRoutine interface, that seems to be required. The new interface doesn't require TsmRoutines to perform visibility checks anymore - that requires the TsmRoutine to know details about the AM, which we want to avoid. To continue to allow taking the returned number of tuples account SampleScanState now has a donetuples field (which previously e.g. existed in SystemRowsSamplerData), which is only incremented after the visibility check succeeds. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-31 05:18:53 +02:00
*
* Currently it is required to implement this interface, as there's no
* alternative way (contrary e.g. to bitmap scans) to implement sample
tableam: comment improvements. Author: Brad DeJong Discussion: https://postgr.es/m/CAJnrtnxDYOQFsDfWz2iri0T_fFL2ZbbzgCOE=4yaMcszgcsf4A@mail.gmail.com Backpatch: 12-
2019-07-18 04:39:54 +02:00
* scans. If infeasible to implement, the AM may raise an error.
tableam: sample scan. This moves sample scan support to below tableam. It's not optional as there is, in contrast to e.g. bitmap heap scans, no alternative way to perform tablesample queries. If an AM can't deal with the block based API, it will have to throw an ERROR. The tableam callbacks for this are block based, but given the current TsmRoutine interface, that seems to be required. The new interface doesn't require TsmRoutines to perform visibility checks anymore - that requires the TsmRoutine to know details about the AM, which we want to avoid. To continue to allow taking the returned number of tuples account SampleScanState now has a donetuples field (which previously e.g. existed in SystemRowsSamplerData), which is only incremented after the visibility check succeeds. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-31 05:18:53 +02:00
*/
bool (*scan_sample_next_block) (TableScanDesc scan,
struct SampleScanState *scanstate);
/*
* This callback, only called after scan_sample_next_block has returned
* true, should determine the next tuple to be returned from the selected
* block using the TsmRoutine's NextSampleTuple() callback.
*
* The callback needs to perform visibility checks, and only return
Fix assorted inconsistencies. There were a number of issues in the recent commits which include typos, code and comments mismatch, leftover function declarations. Fix them. Reported-by: Alexander Lakhin Author: Alexander Lakhin, Amit Kapila and Amit Langote Reviewed-by: Amit Kapila Discussion: https://postgr.es/m/ef0c0232-0c1d-3a35-63d4-0ebd06e31387@gmail.com
2019-06-08 04:46:38 +02:00
* visible tuples. That obviously can mean calling NextSampleTuple()
tableam: sample scan. This moves sample scan support to below tableam. It's not optional as there is, in contrast to e.g. bitmap heap scans, no alternative way to perform tablesample queries. If an AM can't deal with the block based API, it will have to throw an ERROR. The tableam callbacks for this are block based, but given the current TsmRoutine interface, that seems to be required. The new interface doesn't require TsmRoutines to perform visibility checks anymore - that requires the TsmRoutine to know details about the AM, which we want to avoid. To continue to allow taking the returned number of tuples account SampleScanState now has a donetuples field (which previously e.g. existed in SystemRowsSamplerData), which is only incremented after the visibility check succeeds. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-31 05:18:53 +02:00
* multiple times.
*
* The TsmRoutine interface assumes that there's a maximum offset on a
* given page, so if that doesn't apply to an AM, it needs to emulate that
* assumption somehow.
*/
bool (*scan_sample_next_tuple) (TableScanDesc scan,
struct SampleScanState *scanstate,
TupleTableSlot *slot);
tableam: introduce table AM infrastructure. This introduces the concept of table access methods, i.e. CREATE ACCESS METHOD ... TYPE TABLE and CREATE TABLE ... USING (storage-engine). No table access functionality is delegated to table AMs as of this commit, that'll be done in following commits. Subsequent commits will incrementally abstract table access functionality to be routed through table access methods. That change is too large to be reviewed & committed at once, so it'll be done incrementally. Docs will be updated at the end, as adding them incrementally would likely make them less coherent, and definitely is a lot more work, without a lot of benefit. Table access methods are specified similar to index access methods, i.e. pg_am.amhandler returns, as INTERNAL, a pointer to a struct with callbacks. In contrast to index AMs that struct needs to live as long as a backend, typically that's achieved by just returning a pointer to a constant struct. Psql's \d+ now displays a table's access method. That can be disabled with HIDE_TABLEAM=true, which is mainly useful so regression tests can be run against different AMs. It's quite possible that this behaviour still needs to be fine tuned. For now it's not allowed to set a table AM for a partitioned table, as we've not resolved how partitions would inherit that. Disallowing allows us to introduce, if we decide that's the way forward, such a behaviour without a compatibility break. Catversion bumped, to add the heap table AM and references to it. Author: Haribabu Kommi, Andres Freund, Alvaro Herrera, Dimitri Golgov and others Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql https://postgr.es/m/20190107235616.6lur25ph22u5u5av@alap3.anarazel.de https://postgr.es/m/20190304234700.w5tmhducs5wxgzls@alap3.anarazel.de
2019-03-06 18:54:38 +01:00
} TableAmRoutine;
tableam: Add and use scan APIs. Too allow table accesses to be not directly dependent on heap, several new abstractions are needed. Specifically: 1) Heap scans need to be generalized into table scans. Do this by introducing TableScanDesc, which will be the "base class" for individual AMs. This contains the AM independent fields from HeapScanDesc. The previous heap_{beginscan,rescan,endscan} et al. have been replaced with a table_ version. There's no direct replacement for heap_getnext(), as that returned a HeapTuple, which is undesirable for a other AMs. Instead there's table_scan_getnextslot(). But note that heap_getnext() lives on, it's still used widely to access catalog tables. This is achieved by new scan_begin, scan_end, scan_rescan, scan_getnextslot callbacks. 2) The portion of parallel scans that's shared between backends need to be able to do so without the user doing per-AM work. To achieve that new parallelscan_{estimate, initialize, reinitialize} callbacks are introduced, which operate on a new ParallelTableScanDesc, which again can be subclassed by AMs. As it is likely that several AMs are going to be block oriented, block oriented callbacks that can be shared between such AMs are provided and used by heap. table_block_parallelscan_{estimate, intiialize, reinitialize} as callbacks, and table_block_parallelscan_{nextpage, init} for use in AMs. These operate on a ParallelBlockTableScanDesc. 3) Index scans need to be able to access tables to return a tuple, and there needs to be state across individual accesses to the heap to store state like buffers. That's now handled by introducing a sort-of-scan IndexFetchTable, which again is intended to be subclassed by individual AMs (for heap IndexFetchHeap). The relevant callbacks for an AM are index_fetch_{end, begin, reset} to create the necessary state, and index_fetch_tuple to retrieve an indexed tuple. Note that index_fetch_tuple implementations need to be smarter than just blindly fetching the tuples for AMs that have optimizations similar to heap's HOT - the currently alive tuple in the update chain needs to be fetched if appropriate. Similar to table_scan_getnextslot(), it's undesirable to continue to return HeapTuples. Thus index_fetch_heap (might want to rename that later) now accepts a slot as an argument. Core code doesn't have a lot of call sites performing index scans without going through the systable_* API (in contrast to loads of heap_getnext calls and working directly with HeapTuples). Index scans now store the result of a search in IndexScanDesc->xs_heaptid, rather than xs_ctup->t_self. As the target is not generally a HeapTuple anymore that seems cleaner. To be able to sensible adapt code to use the above, two further callbacks have been introduced: a) slot_callbacks returns a TupleTableSlotOps* suitable for creating slots capable of holding a tuple of the AMs type. table_slot_callbacks() and table_slot_create() are based upon that, but have additional logic to deal with views, foreign tables, etc. While this change could have been done separately, nearly all the call sites that needed to be adapted for the rest of this commit also would have been needed to be adapted for table_slot_callbacks(), making separation not worthwhile. b) tuple_satisfies_snapshot checks whether the tuple in a slot is currently visible according to a snapshot. That's required as a few places now don't have a buffer + HeapTuple around, but a slot (which in heap's case internally has that information). Additionally a few infrastructure changes were needed: I) SysScanDesc, as used by systable_{beginscan, getnext} et al. now internally uses a slot to keep track of tuples. While systable_getnext() still returns HeapTuples, and will so for the foreseeable future, the index API (see 1) above) now only deals with slots. The remainder, and largest part, of this commit is then adjusting all scans in postgres to use the new APIs. Author: Andres Freund, Haribabu Kommi, Alvaro Herrera Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-11 20:46:41 +01:00
/* ----------------------------------------------------------------------------
* Slot functions.
* ----------------------------------------------------------------------------
*/
/*
* Returns slot callbacks suitable for holding tuples of the appropriate type
* for the relation. Works for tables, views, foreign tables and partitioned
* tables.
*/
Harmonize heapam and tableam parameter names. Make sure that function declarations use names that exactly match the corresponding names from function definitions. Having parameter names that are reliably consistent in this way will make it easier to reason about groups of related C functions from the same translation unit as a module. It will also make certain refactoring tasks easier. Like other recent commits that cleaned up function parameter names, this commit was written with help from clang-tidy. Later commits will do the same for other parts of the codebase. Author: Peter Geoghegan <pg@bowt.ie> Reviewed-By: David Rowley <dgrowleyml@gmail.com> Discussion: https://postgr.es/m/CAH2-WznJt9CMM9KJTMjJh_zbL5hD9oX44qdJ4aqZtjFi-zA3Tg@mail.gmail.com
2022-09-20 01:46:23 +02:00
extern const TupleTableSlotOps *table_slot_callbacks(Relation relation);
tableam: Add and use scan APIs. Too allow table accesses to be not directly dependent on heap, several new abstractions are needed. Specifically: 1) Heap scans need to be generalized into table scans. Do this by introducing TableScanDesc, which will be the "base class" for individual AMs. This contains the AM independent fields from HeapScanDesc. The previous heap_{beginscan,rescan,endscan} et al. have been replaced with a table_ version. There's no direct replacement for heap_getnext(), as that returned a HeapTuple, which is undesirable for a other AMs. Instead there's table_scan_getnextslot(). But note that heap_getnext() lives on, it's still used widely to access catalog tables. This is achieved by new scan_begin, scan_end, scan_rescan, scan_getnextslot callbacks. 2) The portion of parallel scans that's shared between backends need to be able to do so without the user doing per-AM work. To achieve that new parallelscan_{estimate, initialize, reinitialize} callbacks are introduced, which operate on a new ParallelTableScanDesc, which again can be subclassed by AMs. As it is likely that several AMs are going to be block oriented, block oriented callbacks that can be shared between such AMs are provided and used by heap. table_block_parallelscan_{estimate, intiialize, reinitialize} as callbacks, and table_block_parallelscan_{nextpage, init} for use in AMs. These operate on a ParallelBlockTableScanDesc. 3) Index scans need to be able to access tables to return a tuple, and there needs to be state across individual accesses to the heap to store state like buffers. That's now handled by introducing a sort-of-scan IndexFetchTable, which again is intended to be subclassed by individual AMs (for heap IndexFetchHeap). The relevant callbacks for an AM are index_fetch_{end, begin, reset} to create the necessary state, and index_fetch_tuple to retrieve an indexed tuple. Note that index_fetch_tuple implementations need to be smarter than just blindly fetching the tuples for AMs that have optimizations similar to heap's HOT - the currently alive tuple in the update chain needs to be fetched if appropriate. Similar to table_scan_getnextslot(), it's undesirable to continue to return HeapTuples. Thus index_fetch_heap (might want to rename that later) now accepts a slot as an argument. Core code doesn't have a lot of call sites performing index scans without going through the systable_* API (in contrast to loads of heap_getnext calls and working directly with HeapTuples). Index scans now store the result of a search in IndexScanDesc->xs_heaptid, rather than xs_ctup->t_self. As the target is not generally a HeapTuple anymore that seems cleaner. To be able to sensible adapt code to use the above, two further callbacks have been introduced: a) slot_callbacks returns a TupleTableSlotOps* suitable for creating slots capable of holding a tuple of the AMs type. table_slot_callbacks() and table_slot_create() are based upon that, but have additional logic to deal with views, foreign tables, etc. While this change could have been done separately, nearly all the call sites that needed to be adapted for the rest of this commit also would have been needed to be adapted for table_slot_callbacks(), making separation not worthwhile. b) tuple_satisfies_snapshot checks whether the tuple in a slot is currently visible according to a snapshot. That's required as a few places now don't have a buffer + HeapTuple around, but a slot (which in heap's case internally has that information). Additionally a few infrastructure changes were needed: I) SysScanDesc, as used by systable_{beginscan, getnext} et al. now internally uses a slot to keep track of tuples. While systable_getnext() still returns HeapTuples, and will so for the foreseeable future, the index API (see 1) above) now only deals with slots. The remainder, and largest part, of this commit is then adjusting all scans in postgres to use the new APIs. Author: Andres Freund, Haribabu Kommi, Alvaro Herrera Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-11 20:46:41 +01:00
/*
* Returns slot using the callbacks returned by table_slot_callbacks(), and
* registers it on *reglist.
*/
Harmonize heapam and tableam parameter names. Make sure that function declarations use names that exactly match the corresponding names from function definitions. Having parameter names that are reliably consistent in this way will make it easier to reason about groups of related C functions from the same translation unit as a module. It will also make certain refactoring tasks easier. Like other recent commits that cleaned up function parameter names, this commit was written with help from clang-tidy. Later commits will do the same for other parts of the codebase. Author: Peter Geoghegan <pg@bowt.ie> Reviewed-By: David Rowley <dgrowleyml@gmail.com> Discussion: https://postgr.es/m/CAH2-WznJt9CMM9KJTMjJh_zbL5hD9oX44qdJ4aqZtjFi-zA3Tg@mail.gmail.com
2022-09-20 01:46:23 +02:00
extern TupleTableSlot *table_slot_create(Relation relation, List **reglist);
tableam: Add and use scan APIs. Too allow table accesses to be not directly dependent on heap, several new abstractions are needed. Specifically: 1) Heap scans need to be generalized into table scans. Do this by introducing TableScanDesc, which will be the "base class" for individual AMs. This contains the AM independent fields from HeapScanDesc. The previous heap_{beginscan,rescan,endscan} et al. have been replaced with a table_ version. There's no direct replacement for heap_getnext(), as that returned a HeapTuple, which is undesirable for a other AMs. Instead there's table_scan_getnextslot(). But note that heap_getnext() lives on, it's still used widely to access catalog tables. This is achieved by new scan_begin, scan_end, scan_rescan, scan_getnextslot callbacks. 2) The portion of parallel scans that's shared between backends need to be able to do so without the user doing per-AM work. To achieve that new parallelscan_{estimate, initialize, reinitialize} callbacks are introduced, which operate on a new ParallelTableScanDesc, which again can be subclassed by AMs. As it is likely that several AMs are going to be block oriented, block oriented callbacks that can be shared between such AMs are provided and used by heap. table_block_parallelscan_{estimate, intiialize, reinitialize} as callbacks, and table_block_parallelscan_{nextpage, init} for use in AMs. These operate on a ParallelBlockTableScanDesc. 3) Index scans need to be able to access tables to return a tuple, and there needs to be state across individual accesses to the heap to store state like buffers. That's now handled by introducing a sort-of-scan IndexFetchTable, which again is intended to be subclassed by individual AMs (for heap IndexFetchHeap). The relevant callbacks for an AM are index_fetch_{end, begin, reset} to create the necessary state, and index_fetch_tuple to retrieve an indexed tuple. Note that index_fetch_tuple implementations need to be smarter than just blindly fetching the tuples for AMs that have optimizations similar to heap's HOT - the currently alive tuple in the update chain needs to be fetched if appropriate. Similar to table_scan_getnextslot(), it's undesirable to continue to return HeapTuples. Thus index_fetch_heap (might want to rename that later) now accepts a slot as an argument. Core code doesn't have a lot of call sites performing index scans without going through the systable_* API (in contrast to loads of heap_getnext calls and working directly with HeapTuples). Index scans now store the result of a search in IndexScanDesc->xs_heaptid, rather than xs_ctup->t_self. As the target is not generally a HeapTuple anymore that seems cleaner. To be able to sensible adapt code to use the above, two further callbacks have been introduced: a) slot_callbacks returns a TupleTableSlotOps* suitable for creating slots capable of holding a tuple of the AMs type. table_slot_callbacks() and table_slot_create() are based upon that, but have additional logic to deal with views, foreign tables, etc. While this change could have been done separately, nearly all the call sites that needed to be adapted for the rest of this commit also would have been needed to be adapted for table_slot_callbacks(), making separation not worthwhile. b) tuple_satisfies_snapshot checks whether the tuple in a slot is currently visible according to a snapshot. That's required as a few places now don't have a buffer + HeapTuple around, but a slot (which in heap's case internally has that information). Additionally a few infrastructure changes were needed: I) SysScanDesc, as used by systable_{beginscan, getnext} et al. now internally uses a slot to keep track of tuples. While systable_getnext() still returns HeapTuples, and will so for the foreseeable future, the index API (see 1) above) now only deals with slots. The remainder, and largest part, of this commit is then adjusting all scans in postgres to use the new APIs. Author: Andres Freund, Haribabu Kommi, Alvaro Herrera Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-11 20:46:41 +01:00
/* ----------------------------------------------------------------------------
* Table scan functions.
* ----------------------------------------------------------------------------
*/
/*
* Start a scan of `rel`. Returned tuples pass a visibility test of
* `snapshot`, and if nkeys != 0, the results are filtered by those scan keys.
*/
static inline TableScanDesc
table_beginscan(Relation rel, Snapshot snapshot,
int nkeys, struct ScanKeyData *key)
{
Don't to predicate lock for analyze scans, refactor scan option passing. Before this commit, when ANALYZE was run on a table and serializable was used (either by virtue of an explicit BEGIN TRANSACTION ISOLATION LEVEL SERIALIZABLE, or default_transaction_isolation being set to serializable) a null pointer dereference lead to a crash. The analyze scan doesn't need a snapshot (nor predicate locking), but before this commit a scan only contained information about being a bitmap or sample scan. Refactor the option passing to the scan_begin callback to use a bitmask instead. Alternatively we could have added a new boolean parameter, but that seems harder to read. Even before this issue various people (Heikki, Tom, Robert) suggested doing so. These changes don't change the scan APIs outside of tableam. The flags argument could be exposed, it's not necessary to fix this problem. Also the wrapper table_beginscan* functions encapsulate most of that complexity. After these changes fixing the bug is trivial, just don't acquire predicate lock for analyze style scans. That was already done for bitmap heap scans. Add an assert that a snapshot is passed when acquiring the predicate lock, so this kind of bug doesn't require running with serializable. Also add a comment about sample scans currently requiring predicate locking the entire relation, that previously wasn't remarked upon. Reported-By: Joe Wildish Author: Andres Freund Discussion: https://postgr.es/m/4EA80A20-E9BF-49F1-9F01-5B66CAB21453@elusive.cx https://postgr.es/m/20190411164947.nkii4gaeilt4bui7@alap3.anarazel.de https://postgr.es/m/20190518203102.g7peu2fianukjuxm@alap3.anarazel.de
2019-05-20 00:10:28 +02:00
uint32 flags = SO_TYPE_SEQSCAN |
SO_ALLOW_STRAT | SO_ALLOW_SYNC | SO_ALLOW_PAGEMODE;
return rel->rd_tableam->scan_begin(rel, snapshot, nkeys, key, NULL, flags);
tableam: Add and use scan APIs. Too allow table accesses to be not directly dependent on heap, several new abstractions are needed. Specifically: 1) Heap scans need to be generalized into table scans. Do this by introducing TableScanDesc, which will be the "base class" for individual AMs. This contains the AM independent fields from HeapScanDesc. The previous heap_{beginscan,rescan,endscan} et al. have been replaced with a table_ version. There's no direct replacement for heap_getnext(), as that returned a HeapTuple, which is undesirable for a other AMs. Instead there's table_scan_getnextslot(). But note that heap_getnext() lives on, it's still used widely to access catalog tables. This is achieved by new scan_begin, scan_end, scan_rescan, scan_getnextslot callbacks. 2) The portion of parallel scans that's shared between backends need to be able to do so without the user doing per-AM work. To achieve that new parallelscan_{estimate, initialize, reinitialize} callbacks are introduced, which operate on a new ParallelTableScanDesc, which again can be subclassed by AMs. As it is likely that several AMs are going to be block oriented, block oriented callbacks that can be shared between such AMs are provided and used by heap. table_block_parallelscan_{estimate, intiialize, reinitialize} as callbacks, and table_block_parallelscan_{nextpage, init} for use in AMs. These operate on a ParallelBlockTableScanDesc. 3) Index scans need to be able to access tables to return a tuple, and there needs to be state across individual accesses to the heap to store state like buffers. That's now handled by introducing a sort-of-scan IndexFetchTable, which again is intended to be subclassed by individual AMs (for heap IndexFetchHeap). The relevant callbacks for an AM are index_fetch_{end, begin, reset} to create the necessary state, and index_fetch_tuple to retrieve an indexed tuple. Note that index_fetch_tuple implementations need to be smarter than just blindly fetching the tuples for AMs that have optimizations similar to heap's HOT - the currently alive tuple in the update chain needs to be fetched if appropriate. Similar to table_scan_getnextslot(), it's undesirable to continue to return HeapTuples. Thus index_fetch_heap (might want to rename that later) now accepts a slot as an argument. Core code doesn't have a lot of call sites performing index scans without going through the systable_* API (in contrast to loads of heap_getnext calls and working directly with HeapTuples). Index scans now store the result of a search in IndexScanDesc->xs_heaptid, rather than xs_ctup->t_self. As the target is not generally a HeapTuple anymore that seems cleaner. To be able to sensible adapt code to use the above, two further callbacks have been introduced: a) slot_callbacks returns a TupleTableSlotOps* suitable for creating slots capable of holding a tuple of the AMs type. table_slot_callbacks() and table_slot_create() are based upon that, but have additional logic to deal with views, foreign tables, etc. While this change could have been done separately, nearly all the call sites that needed to be adapted for the rest of this commit also would have been needed to be adapted for table_slot_callbacks(), making separation not worthwhile. b) tuple_satisfies_snapshot checks whether the tuple in a slot is currently visible according to a snapshot. That's required as a few places now don't have a buffer + HeapTuple around, but a slot (which in heap's case internally has that information). Additionally a few infrastructure changes were needed: I) SysScanDesc, as used by systable_{beginscan, getnext} et al. now internally uses a slot to keep track of tuples. While systable_getnext() still returns HeapTuples, and will so for the foreseeable future, the index API (see 1) above) now only deals with slots. The remainder, and largest part, of this commit is then adjusting all scans in postgres to use the new APIs. Author: Andres Freund, Haribabu Kommi, Alvaro Herrera Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-11 20:46:41 +01:00
}
/*
* Like table_beginscan(), but for scanning catalog. It'll automatically use a
* snapshot appropriate for scanning catalog relations.
*/
Harmonize heapam and tableam parameter names. Make sure that function declarations use names that exactly match the corresponding names from function definitions. Having parameter names that are reliably consistent in this way will make it easier to reason about groups of related C functions from the same translation unit as a module. It will also make certain refactoring tasks easier. Like other recent commits that cleaned up function parameter names, this commit was written with help from clang-tidy. Later commits will do the same for other parts of the codebase. Author: Peter Geoghegan <pg@bowt.ie> Reviewed-By: David Rowley <dgrowleyml@gmail.com> Discussion: https://postgr.es/m/CAH2-WznJt9CMM9KJTMjJh_zbL5hD9oX44qdJ4aqZtjFi-zA3Tg@mail.gmail.com
2022-09-20 01:46:23 +02:00
extern TableScanDesc table_beginscan_catalog(Relation relation, int nkeys,
tableam: Add and use scan APIs. Too allow table accesses to be not directly dependent on heap, several new abstractions are needed. Specifically: 1) Heap scans need to be generalized into table scans. Do this by introducing TableScanDesc, which will be the "base class" for individual AMs. This contains the AM independent fields from HeapScanDesc. The previous heap_{beginscan,rescan,endscan} et al. have been replaced with a table_ version. There's no direct replacement for heap_getnext(), as that returned a HeapTuple, which is undesirable for a other AMs. Instead there's table_scan_getnextslot(). But note that heap_getnext() lives on, it's still used widely to access catalog tables. This is achieved by new scan_begin, scan_end, scan_rescan, scan_getnextslot callbacks. 2) The portion of parallel scans that's shared between backends need to be able to do so without the user doing per-AM work. To achieve that new parallelscan_{estimate, initialize, reinitialize} callbacks are introduced, which operate on a new ParallelTableScanDesc, which again can be subclassed by AMs. As it is likely that several AMs are going to be block oriented, block oriented callbacks that can be shared between such AMs are provided and used by heap. table_block_parallelscan_{estimate, intiialize, reinitialize} as callbacks, and table_block_parallelscan_{nextpage, init} for use in AMs. These operate on a ParallelBlockTableScanDesc. 3) Index scans need to be able to access tables to return a tuple, and there needs to be state across individual accesses to the heap to store state like buffers. That's now handled by introducing a sort-of-scan IndexFetchTable, which again is intended to be subclassed by individual AMs (for heap IndexFetchHeap). The relevant callbacks for an AM are index_fetch_{end, begin, reset} to create the necessary state, and index_fetch_tuple to retrieve an indexed tuple. Note that index_fetch_tuple implementations need to be smarter than just blindly fetching the tuples for AMs that have optimizations similar to heap's HOT - the currently alive tuple in the update chain needs to be fetched if appropriate. Similar to table_scan_getnextslot(), it's undesirable to continue to return HeapTuples. Thus index_fetch_heap (might want to rename that later) now accepts a slot as an argument. Core code doesn't have a lot of call sites performing index scans without going through the systable_* API (in contrast to loads of heap_getnext calls and working directly with HeapTuples). Index scans now store the result of a search in IndexScanDesc->xs_heaptid, rather than xs_ctup->t_self. As the target is not generally a HeapTuple anymore that seems cleaner. To be able to sensible adapt code to use the above, two further callbacks have been introduced: a) slot_callbacks returns a TupleTableSlotOps* suitable for creating slots capable of holding a tuple of the AMs type. table_slot_callbacks() and table_slot_create() are based upon that, but have additional logic to deal with views, foreign tables, etc. While this change could have been done separately, nearly all the call sites that needed to be adapted for the rest of this commit also would have been needed to be adapted for table_slot_callbacks(), making separation not worthwhile. b) tuple_satisfies_snapshot checks whether the tuple in a slot is currently visible according to a snapshot. That's required as a few places now don't have a buffer + HeapTuple around, but a slot (which in heap's case internally has that information). Additionally a few infrastructure changes were needed: I) SysScanDesc, as used by systable_{beginscan, getnext} et al. now internally uses a slot to keep track of tuples. While systable_getnext() still returns HeapTuples, and will so for the foreseeable future, the index API (see 1) above) now only deals with slots. The remainder, and largest part, of this commit is then adjusting all scans in postgres to use the new APIs. Author: Andres Freund, Haribabu Kommi, Alvaro Herrera Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-11 20:46:41 +01:00
struct ScanKeyData *key);
/*
* Like table_beginscan(), but table_beginscan_strat() offers an extended API
* that lets the caller control whether a nondefault buffer access strategy
* can be used, and whether syncscan can be chosen (possibly resulting in the
* scan not starting from block zero). Both of these default to true with
* plain table_beginscan.
*/
static inline TableScanDesc
table_beginscan_strat(Relation rel, Snapshot snapshot,
int nkeys, struct ScanKeyData *key,
bool allow_strat, bool allow_sync)
{
Don't to predicate lock for analyze scans, refactor scan option passing. Before this commit, when ANALYZE was run on a table and serializable was used (either by virtue of an explicit BEGIN TRANSACTION ISOLATION LEVEL SERIALIZABLE, or default_transaction_isolation being set to serializable) a null pointer dereference lead to a crash. The analyze scan doesn't need a snapshot (nor predicate locking), but before this commit a scan only contained information about being a bitmap or sample scan. Refactor the option passing to the scan_begin callback to use a bitmask instead. Alternatively we could have added a new boolean parameter, but that seems harder to read. Even before this issue various people (Heikki, Tom, Robert) suggested doing so. These changes don't change the scan APIs outside of tableam. The flags argument could be exposed, it's not necessary to fix this problem. Also the wrapper table_beginscan* functions encapsulate most of that complexity. After these changes fixing the bug is trivial, just don't acquire predicate lock for analyze style scans. That was already done for bitmap heap scans. Add an assert that a snapshot is passed when acquiring the predicate lock, so this kind of bug doesn't require running with serializable. Also add a comment about sample scans currently requiring predicate locking the entire relation, that previously wasn't remarked upon. Reported-By: Joe Wildish Author: Andres Freund Discussion: https://postgr.es/m/4EA80A20-E9BF-49F1-9F01-5B66CAB21453@elusive.cx https://postgr.es/m/20190411164947.nkii4gaeilt4bui7@alap3.anarazel.de https://postgr.es/m/20190518203102.g7peu2fianukjuxm@alap3.anarazel.de
2019-05-20 00:10:28 +02:00
uint32 flags = SO_TYPE_SEQSCAN | SO_ALLOW_PAGEMODE;
if (allow_strat)
flags |= SO_ALLOW_STRAT;
if (allow_sync)
flags |= SO_ALLOW_SYNC;
return rel->rd_tableam->scan_begin(rel, snapshot, nkeys, key, NULL, flags);
tableam: Add and use scan APIs. Too allow table accesses to be not directly dependent on heap, several new abstractions are needed. Specifically: 1) Heap scans need to be generalized into table scans. Do this by introducing TableScanDesc, which will be the "base class" for individual AMs. This contains the AM independent fields from HeapScanDesc. The previous heap_{beginscan,rescan,endscan} et al. have been replaced with a table_ version. There's no direct replacement for heap_getnext(), as that returned a HeapTuple, which is undesirable for a other AMs. Instead there's table_scan_getnextslot(). But note that heap_getnext() lives on, it's still used widely to access catalog tables. This is achieved by new scan_begin, scan_end, scan_rescan, scan_getnextslot callbacks. 2) The portion of parallel scans that's shared between backends need to be able to do so without the user doing per-AM work. To achieve that new parallelscan_{estimate, initialize, reinitialize} callbacks are introduced, which operate on a new ParallelTableScanDesc, which again can be subclassed by AMs. As it is likely that several AMs are going to be block oriented, block oriented callbacks that can be shared between such AMs are provided and used by heap. table_block_parallelscan_{estimate, intiialize, reinitialize} as callbacks, and table_block_parallelscan_{nextpage, init} for use in AMs. These operate on a ParallelBlockTableScanDesc. 3) Index scans need to be able to access tables to return a tuple, and there needs to be state across individual accesses to the heap to store state like buffers. That's now handled by introducing a sort-of-scan IndexFetchTable, which again is intended to be subclassed by individual AMs (for heap IndexFetchHeap). The relevant callbacks for an AM are index_fetch_{end, begin, reset} to create the necessary state, and index_fetch_tuple to retrieve an indexed tuple. Note that index_fetch_tuple implementations need to be smarter than just blindly fetching the tuples for AMs that have optimizations similar to heap's HOT - the currently alive tuple in the update chain needs to be fetched if appropriate. Similar to table_scan_getnextslot(), it's undesirable to continue to return HeapTuples. Thus index_fetch_heap (might want to rename that later) now accepts a slot as an argument. Core code doesn't have a lot of call sites performing index scans without going through the systable_* API (in contrast to loads of heap_getnext calls and working directly with HeapTuples). Index scans now store the result of a search in IndexScanDesc->xs_heaptid, rather than xs_ctup->t_self. As the target is not generally a HeapTuple anymore that seems cleaner. To be able to sensible adapt code to use the above, two further callbacks have been introduced: a) slot_callbacks returns a TupleTableSlotOps* suitable for creating slots capable of holding a tuple of the AMs type. table_slot_callbacks() and table_slot_create() are based upon that, but have additional logic to deal with views, foreign tables, etc. While this change could have been done separately, nearly all the call sites that needed to be adapted for the rest of this commit also would have been needed to be adapted for table_slot_callbacks(), making separation not worthwhile. b) tuple_satisfies_snapshot checks whether the tuple in a slot is currently visible according to a snapshot. That's required as a few places now don't have a buffer + HeapTuple around, but a slot (which in heap's case internally has that information). Additionally a few infrastructure changes were needed: I) SysScanDesc, as used by systable_{beginscan, getnext} et al. now internally uses a slot to keep track of tuples. While systable_getnext() still returns HeapTuples, and will so for the foreseeable future, the index API (see 1) above) now only deals with slots. The remainder, and largest part, of this commit is then adjusting all scans in postgres to use the new APIs. Author: Andres Freund, Haribabu Kommi, Alvaro Herrera Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-11 20:46:41 +01:00
}
/*
* table_beginscan_bm is an alternative entry point for setting up a
* TableScanDesc for a bitmap heap scan. Although that scan technology is
* really quite unlike a standard seqscan, there is just enough commonality to
* make it worth using the same data structure.
*/
static inline TableScanDesc
table_beginscan_bm(Relation rel, Snapshot snapshot,
int nkeys, struct ScanKeyData *key)
{
Don't to predicate lock for analyze scans, refactor scan option passing. Before this commit, when ANALYZE was run on a table and serializable was used (either by virtue of an explicit BEGIN TRANSACTION ISOLATION LEVEL SERIALIZABLE, or default_transaction_isolation being set to serializable) a null pointer dereference lead to a crash. The analyze scan doesn't need a snapshot (nor predicate locking), but before this commit a scan only contained information about being a bitmap or sample scan. Refactor the option passing to the scan_begin callback to use a bitmask instead. Alternatively we could have added a new boolean parameter, but that seems harder to read. Even before this issue various people (Heikki, Tom, Robert) suggested doing so. These changes don't change the scan APIs outside of tableam. The flags argument could be exposed, it's not necessary to fix this problem. Also the wrapper table_beginscan* functions encapsulate most of that complexity. After these changes fixing the bug is trivial, just don't acquire predicate lock for analyze style scans. That was already done for bitmap heap scans. Add an assert that a snapshot is passed when acquiring the predicate lock, so this kind of bug doesn't require running with serializable. Also add a comment about sample scans currently requiring predicate locking the entire relation, that previously wasn't remarked upon. Reported-By: Joe Wildish Author: Andres Freund Discussion: https://postgr.es/m/4EA80A20-E9BF-49F1-9F01-5B66CAB21453@elusive.cx https://postgr.es/m/20190411164947.nkii4gaeilt4bui7@alap3.anarazel.de https://postgr.es/m/20190518203102.g7peu2fianukjuxm@alap3.anarazel.de
2019-05-20 00:10:28 +02:00
uint32 flags = SO_TYPE_BITMAPSCAN | SO_ALLOW_PAGEMODE;
return rel->rd_tableam->scan_begin(rel, snapshot, nkeys, key, NULL, flags);
tableam: Add and use scan APIs. Too allow table accesses to be not directly dependent on heap, several new abstractions are needed. Specifically: 1) Heap scans need to be generalized into table scans. Do this by introducing TableScanDesc, which will be the "base class" for individual AMs. This contains the AM independent fields from HeapScanDesc. The previous heap_{beginscan,rescan,endscan} et al. have been replaced with a table_ version. There's no direct replacement for heap_getnext(), as that returned a HeapTuple, which is undesirable for a other AMs. Instead there's table_scan_getnextslot(). But note that heap_getnext() lives on, it's still used widely to access catalog tables. This is achieved by new scan_begin, scan_end, scan_rescan, scan_getnextslot callbacks. 2) The portion of parallel scans that's shared between backends need to be able to do so without the user doing per-AM work. To achieve that new parallelscan_{estimate, initialize, reinitialize} callbacks are introduced, which operate on a new ParallelTableScanDesc, which again can be subclassed by AMs. As it is likely that several AMs are going to be block oriented, block oriented callbacks that can be shared between such AMs are provided and used by heap. table_block_parallelscan_{estimate, intiialize, reinitialize} as callbacks, and table_block_parallelscan_{nextpage, init} for use in AMs. These operate on a ParallelBlockTableScanDesc. 3) Index scans need to be able to access tables to return a tuple, and there needs to be state across individual accesses to the heap to store state like buffers. That's now handled by introducing a sort-of-scan IndexFetchTable, which again is intended to be subclassed by individual AMs (for heap IndexFetchHeap). The relevant callbacks for an AM are index_fetch_{end, begin, reset} to create the necessary state, and index_fetch_tuple to retrieve an indexed tuple. Note that index_fetch_tuple implementations need to be smarter than just blindly fetching the tuples for AMs that have optimizations similar to heap's HOT - the currently alive tuple in the update chain needs to be fetched if appropriate. Similar to table_scan_getnextslot(), it's undesirable to continue to return HeapTuples. Thus index_fetch_heap (might want to rename that later) now accepts a slot as an argument. Core code doesn't have a lot of call sites performing index scans without going through the systable_* API (in contrast to loads of heap_getnext calls and working directly with HeapTuples). Index scans now store the result of a search in IndexScanDesc->xs_heaptid, rather than xs_ctup->t_self. As the target is not generally a HeapTuple anymore that seems cleaner. To be able to sensible adapt code to use the above, two further callbacks have been introduced: a) slot_callbacks returns a TupleTableSlotOps* suitable for creating slots capable of holding a tuple of the AMs type. table_slot_callbacks() and table_slot_create() are based upon that, but have additional logic to deal with views, foreign tables, etc. While this change could have been done separately, nearly all the call sites that needed to be adapted for the rest of this commit also would have been needed to be adapted for table_slot_callbacks(), making separation not worthwhile. b) tuple_satisfies_snapshot checks whether the tuple in a slot is currently visible according to a snapshot. That's required as a few places now don't have a buffer + HeapTuple around, but a slot (which in heap's case internally has that information). Additionally a few infrastructure changes were needed: I) SysScanDesc, as used by systable_{beginscan, getnext} et al. now internally uses a slot to keep track of tuples. While systable_getnext() still returns HeapTuples, and will so for the foreseeable future, the index API (see 1) above) now only deals with slots. The remainder, and largest part, of this commit is then adjusting all scans in postgres to use the new APIs. Author: Andres Freund, Haribabu Kommi, Alvaro Herrera Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-11 20:46:41 +01:00
}
/*
* table_beginscan_sampling is an alternative entry point for setting up a
* TableScanDesc for a TABLESAMPLE scan. As with bitmap scans, it's worth
* using the same data structure although the behavior is rather different.
* In addition to the options offered by table_beginscan_strat, this call
* also allows control of whether page-mode visibility checking is used.
*/
static inline TableScanDesc
table_beginscan_sampling(Relation rel, Snapshot snapshot,
int nkeys, struct ScanKeyData *key,
tableam: Formatting and other minor cleanups. The superflous heapam_xlog.h includes were reported by Peter Geoghegan.
2019-03-31 05:13:56 +02:00
bool allow_strat, bool allow_sync,
bool allow_pagemode)
tableam: Add and use scan APIs. Too allow table accesses to be not directly dependent on heap, several new abstractions are needed. Specifically: 1) Heap scans need to be generalized into table scans. Do this by introducing TableScanDesc, which will be the "base class" for individual AMs. This contains the AM independent fields from HeapScanDesc. The previous heap_{beginscan,rescan,endscan} et al. have been replaced with a table_ version. There's no direct replacement for heap_getnext(), as that returned a HeapTuple, which is undesirable for a other AMs. Instead there's table_scan_getnextslot(). But note that heap_getnext() lives on, it's still used widely to access catalog tables. This is achieved by new scan_begin, scan_end, scan_rescan, scan_getnextslot callbacks. 2) The portion of parallel scans that's shared between backends need to be able to do so without the user doing per-AM work. To achieve that new parallelscan_{estimate, initialize, reinitialize} callbacks are introduced, which operate on a new ParallelTableScanDesc, which again can be subclassed by AMs. As it is likely that several AMs are going to be block oriented, block oriented callbacks that can be shared between such AMs are provided and used by heap. table_block_parallelscan_{estimate, intiialize, reinitialize} as callbacks, and table_block_parallelscan_{nextpage, init} for use in AMs. These operate on a ParallelBlockTableScanDesc. 3) Index scans need to be able to access tables to return a tuple, and there needs to be state across individual accesses to the heap to store state like buffers. That's now handled by introducing a sort-of-scan IndexFetchTable, which again is intended to be subclassed by individual AMs (for heap IndexFetchHeap). The relevant callbacks for an AM are index_fetch_{end, begin, reset} to create the necessary state, and index_fetch_tuple to retrieve an indexed tuple. Note that index_fetch_tuple implementations need to be smarter than just blindly fetching the tuples for AMs that have optimizations similar to heap's HOT - the currently alive tuple in the update chain needs to be fetched if appropriate. Similar to table_scan_getnextslot(), it's undesirable to continue to return HeapTuples. Thus index_fetch_heap (might want to rename that later) now accepts a slot as an argument. Core code doesn't have a lot of call sites performing index scans without going through the systable_* API (in contrast to loads of heap_getnext calls and working directly with HeapTuples). Index scans now store the result of a search in IndexScanDesc->xs_heaptid, rather than xs_ctup->t_self. As the target is not generally a HeapTuple anymore that seems cleaner. To be able to sensible adapt code to use the above, two further callbacks have been introduced: a) slot_callbacks returns a TupleTableSlotOps* suitable for creating slots capable of holding a tuple of the AMs type. table_slot_callbacks() and table_slot_create() are based upon that, but have additional logic to deal with views, foreign tables, etc. While this change could have been done separately, nearly all the call sites that needed to be adapted for the rest of this commit also would have been needed to be adapted for table_slot_callbacks(), making separation not worthwhile. b) tuple_satisfies_snapshot checks whether the tuple in a slot is currently visible according to a snapshot. That's required as a few places now don't have a buffer + HeapTuple around, but a slot (which in heap's case internally has that information). Additionally a few infrastructure changes were needed: I) SysScanDesc, as used by systable_{beginscan, getnext} et al. now internally uses a slot to keep track of tuples. While systable_getnext() still returns HeapTuples, and will so for the foreseeable future, the index API (see 1) above) now only deals with slots. The remainder, and largest part, of this commit is then adjusting all scans in postgres to use the new APIs. Author: Andres Freund, Haribabu Kommi, Alvaro Herrera Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-11 20:46:41 +01:00
{
Don't to predicate lock for analyze scans, refactor scan option passing. Before this commit, when ANALYZE was run on a table and serializable was used (either by virtue of an explicit BEGIN TRANSACTION ISOLATION LEVEL SERIALIZABLE, or default_transaction_isolation being set to serializable) a null pointer dereference lead to a crash. The analyze scan doesn't need a snapshot (nor predicate locking), but before this commit a scan only contained information about being a bitmap or sample scan. Refactor the option passing to the scan_begin callback to use a bitmask instead. Alternatively we could have added a new boolean parameter, but that seems harder to read. Even before this issue various people (Heikki, Tom, Robert) suggested doing so. These changes don't change the scan APIs outside of tableam. The flags argument could be exposed, it's not necessary to fix this problem. Also the wrapper table_beginscan* functions encapsulate most of that complexity. After these changes fixing the bug is trivial, just don't acquire predicate lock for analyze style scans. That was already done for bitmap heap scans. Add an assert that a snapshot is passed when acquiring the predicate lock, so this kind of bug doesn't require running with serializable. Also add a comment about sample scans currently requiring predicate locking the entire relation, that previously wasn't remarked upon. Reported-By: Joe Wildish Author: Andres Freund Discussion: https://postgr.es/m/4EA80A20-E9BF-49F1-9F01-5B66CAB21453@elusive.cx https://postgr.es/m/20190411164947.nkii4gaeilt4bui7@alap3.anarazel.de https://postgr.es/m/20190518203102.g7peu2fianukjuxm@alap3.anarazel.de
2019-05-20 00:10:28 +02:00
uint32 flags = SO_TYPE_SAMPLESCAN;
if (allow_strat)
flags |= SO_ALLOW_STRAT;
if (allow_sync)
flags |= SO_ALLOW_SYNC;
if (allow_pagemode)
flags |= SO_ALLOW_PAGEMODE;
return rel->rd_tableam->scan_begin(rel, snapshot, nkeys, key, NULL, flags);
tableam: Add and use scan APIs. Too allow table accesses to be not directly dependent on heap, several new abstractions are needed. Specifically: 1) Heap scans need to be generalized into table scans. Do this by introducing TableScanDesc, which will be the "base class" for individual AMs. This contains the AM independent fields from HeapScanDesc. The previous heap_{beginscan,rescan,endscan} et al. have been replaced with a table_ version. There's no direct replacement for heap_getnext(), as that returned a HeapTuple, which is undesirable for a other AMs. Instead there's table_scan_getnextslot(). But note that heap_getnext() lives on, it's still used widely to access catalog tables. This is achieved by new scan_begin, scan_end, scan_rescan, scan_getnextslot callbacks. 2) The portion of parallel scans that's shared between backends need to be able to do so without the user doing per-AM work. To achieve that new parallelscan_{estimate, initialize, reinitialize} callbacks are introduced, which operate on a new ParallelTableScanDesc, which again can be subclassed by AMs. As it is likely that several AMs are going to be block oriented, block oriented callbacks that can be shared between such AMs are provided and used by heap. table_block_parallelscan_{estimate, intiialize, reinitialize} as callbacks, and table_block_parallelscan_{nextpage, init} for use in AMs. These operate on a ParallelBlockTableScanDesc. 3) Index scans need to be able to access tables to return a tuple, and there needs to be state across individual accesses to the heap to store state like buffers. That's now handled by introducing a sort-of-scan IndexFetchTable, which again is intended to be subclassed by individual AMs (for heap IndexFetchHeap). The relevant callbacks for an AM are index_fetch_{end, begin, reset} to create the necessary state, and index_fetch_tuple to retrieve an indexed tuple. Note that index_fetch_tuple implementations need to be smarter than just blindly fetching the tuples for AMs that have optimizations similar to heap's HOT - the currently alive tuple in the update chain needs to be fetched if appropriate. Similar to table_scan_getnextslot(), it's undesirable to continue to return HeapTuples. Thus index_fetch_heap (might want to rename that later) now accepts a slot as an argument. Core code doesn't have a lot of call sites performing index scans without going through the systable_* API (in contrast to loads of heap_getnext calls and working directly with HeapTuples). Index scans now store the result of a search in IndexScanDesc->xs_heaptid, rather than xs_ctup->t_self. As the target is not generally a HeapTuple anymore that seems cleaner. To be able to sensible adapt code to use the above, two further callbacks have been introduced: a) slot_callbacks returns a TupleTableSlotOps* suitable for creating slots capable of holding a tuple of the AMs type. table_slot_callbacks() and table_slot_create() are based upon that, but have additional logic to deal with views, foreign tables, etc. While this change could have been done separately, nearly all the call sites that needed to be adapted for the rest of this commit also would have been needed to be adapted for table_slot_callbacks(), making separation not worthwhile. b) tuple_satisfies_snapshot checks whether the tuple in a slot is currently visible according to a snapshot. That's required as a few places now don't have a buffer + HeapTuple around, but a slot (which in heap's case internally has that information). Additionally a few infrastructure changes were needed: I) SysScanDesc, as used by systable_{beginscan, getnext} et al. now internally uses a slot to keep track of tuples. While systable_getnext() still returns HeapTuples, and will so for the foreseeable future, the index API (see 1) above) now only deals with slots. The remainder, and largest part, of this commit is then adjusting all scans in postgres to use the new APIs. Author: Andres Freund, Haribabu Kommi, Alvaro Herrera Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-11 20:46:41 +01:00
}
Fix bug in Tid scan. Commit 147e3722f7 changed Tid scan so that it calls table_beginscan() and uses the scan option for seq scan. This change caused two issues. (1) The change caused Tid scan to take a predicate lock on the entire relation in serializable transaction even when relation-level lock is not necessary. This could lead to an unexpected serialization error. (2) The change caused Tid scan to increment the number of seq_scan in pg_stat_*_tables views even though it's not seq scan. This could confuse the users. This commit adds the scan option for Tid scan and makes Tid scan use it, to avoid those issues. Back-patch to v12, where the bug was introduced. Author: Tatsuhito Kasahara Reviewed-by: Kyotaro Horiguchi, Masahiko Sawada, Fujii Masao Discussion: https://postgr.es/m/CAP0=ZVKy+gTbFmB6X_UW0pP3WaeJ-fkUWHoD-pExS=at3CY76g@mail.gmail.com
2020-02-07 14:00:21 +01:00
/*
* table_beginscan_tid is an alternative entry point for setting up a
* TableScanDesc for a Tid scan. As with bitmap scans, it's worth using
* the same data structure although the behavior is rather different.
*/
static inline TableScanDesc
table_beginscan_tid(Relation rel, Snapshot snapshot)
{
uint32 flags = SO_TYPE_TIDSCAN;
return rel->rd_tableam->scan_begin(rel, snapshot, 0, NULL, NULL, flags);
}
tableam: Add and use scan APIs. Too allow table accesses to be not directly dependent on heap, several new abstractions are needed. Specifically: 1) Heap scans need to be generalized into table scans. Do this by introducing TableScanDesc, which will be the "base class" for individual AMs. This contains the AM independent fields from HeapScanDesc. The previous heap_{beginscan,rescan,endscan} et al. have been replaced with a table_ version. There's no direct replacement for heap_getnext(), as that returned a HeapTuple, which is undesirable for a other AMs. Instead there's table_scan_getnextslot(). But note that heap_getnext() lives on, it's still used widely to access catalog tables. This is achieved by new scan_begin, scan_end, scan_rescan, scan_getnextslot callbacks. 2) The portion of parallel scans that's shared between backends need to be able to do so without the user doing per-AM work. To achieve that new parallelscan_{estimate, initialize, reinitialize} callbacks are introduced, which operate on a new ParallelTableScanDesc, which again can be subclassed by AMs. As it is likely that several AMs are going to be block oriented, block oriented callbacks that can be shared between such AMs are provided and used by heap. table_block_parallelscan_{estimate, intiialize, reinitialize} as callbacks, and table_block_parallelscan_{nextpage, init} for use in AMs. These operate on a ParallelBlockTableScanDesc. 3) Index scans need to be able to access tables to return a tuple, and there needs to be state across individual accesses to the heap to store state like buffers. That's now handled by introducing a sort-of-scan IndexFetchTable, which again is intended to be subclassed by individual AMs (for heap IndexFetchHeap). The relevant callbacks for an AM are index_fetch_{end, begin, reset} to create the necessary state, and index_fetch_tuple to retrieve an indexed tuple. Note that index_fetch_tuple implementations need to be smarter than just blindly fetching the tuples for AMs that have optimizations similar to heap's HOT - the currently alive tuple in the update chain needs to be fetched if appropriate. Similar to table_scan_getnextslot(), it's undesirable to continue to return HeapTuples. Thus index_fetch_heap (might want to rename that later) now accepts a slot as an argument. Core code doesn't have a lot of call sites performing index scans without going through the systable_* API (in contrast to loads of heap_getnext calls and working directly with HeapTuples). Index scans now store the result of a search in IndexScanDesc->xs_heaptid, rather than xs_ctup->t_self. As the target is not generally a HeapTuple anymore that seems cleaner. To be able to sensible adapt code to use the above, two further callbacks have been introduced: a) slot_callbacks returns a TupleTableSlotOps* suitable for creating slots capable of holding a tuple of the AMs type. table_slot_callbacks() and table_slot_create() are based upon that, but have additional logic to deal with views, foreign tables, etc. While this change could have been done separately, nearly all the call sites that needed to be adapted for the rest of this commit also would have been needed to be adapted for table_slot_callbacks(), making separation not worthwhile. b) tuple_satisfies_snapshot checks whether the tuple in a slot is currently visible according to a snapshot. That's required as a few places now don't have a buffer + HeapTuple around, but a slot (which in heap's case internally has that information). Additionally a few infrastructure changes were needed: I) SysScanDesc, as used by systable_{beginscan, getnext} et al. now internally uses a slot to keep track of tuples. While systable_getnext() still returns HeapTuples, and will so for the foreseeable future, the index API (see 1) above) now only deals with slots. The remainder, and largest part, of this commit is then adjusting all scans in postgres to use the new APIs. Author: Andres Freund, Haribabu Kommi, Alvaro Herrera Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-11 20:46:41 +01:00
/*
* table_beginscan_analyze is an alternative entry point for setting up a
* TableScanDesc for an ANALYZE scan. As with bitmap scans, it's worth using
* the same data structure although the behavior is rather different.
*/
static inline TableScanDesc
table_beginscan_analyze(Relation rel)
{
Don't to predicate lock for analyze scans, refactor scan option passing. Before this commit, when ANALYZE was run on a table and serializable was used (either by virtue of an explicit BEGIN TRANSACTION ISOLATION LEVEL SERIALIZABLE, or default_transaction_isolation being set to serializable) a null pointer dereference lead to a crash. The analyze scan doesn't need a snapshot (nor predicate locking), but before this commit a scan only contained information about being a bitmap or sample scan. Refactor the option passing to the scan_begin callback to use a bitmask instead. Alternatively we could have added a new boolean parameter, but that seems harder to read. Even before this issue various people (Heikki, Tom, Robert) suggested doing so. These changes don't change the scan APIs outside of tableam. The flags argument could be exposed, it's not necessary to fix this problem. Also the wrapper table_beginscan* functions encapsulate most of that complexity. After these changes fixing the bug is trivial, just don't acquire predicate lock for analyze style scans. That was already done for bitmap heap scans. Add an assert that a snapshot is passed when acquiring the predicate lock, so this kind of bug doesn't require running with serializable. Also add a comment about sample scans currently requiring predicate locking the entire relation, that previously wasn't remarked upon. Reported-By: Joe Wildish Author: Andres Freund Discussion: https://postgr.es/m/4EA80A20-E9BF-49F1-9F01-5B66CAB21453@elusive.cx https://postgr.es/m/20190411164947.nkii4gaeilt4bui7@alap3.anarazel.de https://postgr.es/m/20190518203102.g7peu2fianukjuxm@alap3.anarazel.de
2019-05-20 00:10:28 +02:00
uint32 flags = SO_TYPE_ANALYZE;
return rel->rd_tableam->scan_begin(rel, NULL, 0, NULL, NULL, flags);
tableam: Add and use scan APIs. Too allow table accesses to be not directly dependent on heap, several new abstractions are needed. Specifically: 1) Heap scans need to be generalized into table scans. Do this by introducing TableScanDesc, which will be the "base class" for individual AMs. This contains the AM independent fields from HeapScanDesc. The previous heap_{beginscan,rescan,endscan} et al. have been replaced with a table_ version. There's no direct replacement for heap_getnext(), as that returned a HeapTuple, which is undesirable for a other AMs. Instead there's table_scan_getnextslot(). But note that heap_getnext() lives on, it's still used widely to access catalog tables. This is achieved by new scan_begin, scan_end, scan_rescan, scan_getnextslot callbacks. 2) The portion of parallel scans that's shared between backends need to be able to do so without the user doing per-AM work. To achieve that new parallelscan_{estimate, initialize, reinitialize} callbacks are introduced, which operate on a new ParallelTableScanDesc, which again can be subclassed by AMs. As it is likely that several AMs are going to be block oriented, block oriented callbacks that can be shared between such AMs are provided and used by heap. table_block_parallelscan_{estimate, intiialize, reinitialize} as callbacks, and table_block_parallelscan_{nextpage, init} for use in AMs. These operate on a ParallelBlockTableScanDesc. 3) Index scans need to be able to access tables to return a tuple, and there needs to be state across individual accesses to the heap to store state like buffers. That's now handled by introducing a sort-of-scan IndexFetchTable, which again is intended to be subclassed by individual AMs (for heap IndexFetchHeap). The relevant callbacks for an AM are index_fetch_{end, begin, reset} to create the necessary state, and index_fetch_tuple to retrieve an indexed tuple. Note that index_fetch_tuple implementations need to be smarter than just blindly fetching the tuples for AMs that have optimizations similar to heap's HOT - the currently alive tuple in the update chain needs to be fetched if appropriate. Similar to table_scan_getnextslot(), it's undesirable to continue to return HeapTuples. Thus index_fetch_heap (might want to rename that later) now accepts a slot as an argument. Core code doesn't have a lot of call sites performing index scans without going through the systable_* API (in contrast to loads of heap_getnext calls and working directly with HeapTuples). Index scans now store the result of a search in IndexScanDesc->xs_heaptid, rather than xs_ctup->t_self. As the target is not generally a HeapTuple anymore that seems cleaner. To be able to sensible adapt code to use the above, two further callbacks have been introduced: a) slot_callbacks returns a TupleTableSlotOps* suitable for creating slots capable of holding a tuple of the AMs type. table_slot_callbacks() and table_slot_create() are based upon that, but have additional logic to deal with views, foreign tables, etc. While this change could have been done separately, nearly all the call sites that needed to be adapted for the rest of this commit also would have been needed to be adapted for table_slot_callbacks(), making separation not worthwhile. b) tuple_satisfies_snapshot checks whether the tuple in a slot is currently visible according to a snapshot. That's required as a few places now don't have a buffer + HeapTuple around, but a slot (which in heap's case internally has that information). Additionally a few infrastructure changes were needed: I) SysScanDesc, as used by systable_{beginscan, getnext} et al. now internally uses a slot to keep track of tuples. While systable_getnext() still returns HeapTuples, and will so for the foreseeable future, the index API (see 1) above) now only deals with slots. The remainder, and largest part, of this commit is then adjusting all scans in postgres to use the new APIs. Author: Andres Freund, Haribabu Kommi, Alvaro Herrera Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-11 20:46:41 +01:00
}
/*
* End relation scan.
*/
static inline void
table_endscan(TableScanDesc scan)
{
scan->rs_rd->rd_tableam->scan_end(scan);
}
/*
* Restart a relation scan.
*/
static inline void
table_rescan(TableScanDesc scan,
struct ScanKeyData *key)
{
scan->rs_rd->rd_tableam->scan_rescan(scan, key, false, false, false, false);
}
/*
* Restart a relation scan after changing params.
*
* This call allows changing the buffer strategy, syncscan, and pagemode
* options before starting a fresh scan. Note that although the actual use of
* syncscan might change (effectively, enabling or disabling reporting), the
* previously selected startblock will be kept.
*/
static inline void
table_rescan_set_params(TableScanDesc scan, struct ScanKeyData *key,
bool allow_strat, bool allow_sync, bool allow_pagemode)
{
scan->rs_rd->rd_tableam->scan_rescan(scan, key, true,
allow_strat, allow_sync,
allow_pagemode);
}
/*
* Update snapshot used by the scan.
*/
extern void table_scan_update_snapshot(TableScanDesc scan, Snapshot snapshot);
/*
* Return next tuple from `scan`, store in slot.
*/
static inline bool
table_scan_getnextslot(TableScanDesc sscan, ScanDirection direction, TupleTableSlot *slot)
{
slot->tts_tableOid = RelationGetRelid(sscan->rs_rd);
Implement streaming mode in ReorderBuffer. Instead of serializing the transaction to disk after reaching the logical_decoding_work_mem limit in memory, we consume the changes we have in memory and invoke stream API methods added by commit 45fdc9738b. However, sometimes if we have incomplete toast or speculative insert we spill to the disk because we can't generate the complete tuple and stream. And, as soon as we get the complete tuple we stream the transaction including the serialized changes. We can do this incremental processing thanks to having assignments (associating subxact with toplevel xacts) in WAL right away, and thanks to logging the invalidation messages at each command end. These features are added by commits 0bead9af48 and c55040ccd0 respectively. Now that we can stream in-progress transactions, the concurrent aborts may cause failures when the output plugin consults catalogs (both system and user-defined). We handle such failures by returning ERRCODE_TRANSACTION_ROLLBACK sqlerrcode from system table scan APIs to the backend or WALSender decoding a specific uncommitted transaction. The decoding logic on the receipt of such a sqlerrcode aborts the decoding of the current transaction and continue with the decoding of other transactions. We have ReorderBufferTXN pointer in each ReorderBufferChange by which we know which xact it belongs to. The output plugin can use this to decide which changes to discard in case of stream_abort_cb (e.g. when a subxact gets discarded). We also provide a new option via SQL APIs to fetch the changes being streamed. Author: Dilip Kumar, Tomas Vondra, Amit Kapila, Nikhil Sontakke Reviewed-by: Amit Kapila, Kuntal Ghosh, Ajin Cherian Tested-by: Neha Sharma, Mahendra Singh Thalor and Ajin Cherian Discussion: https://postgr.es/m/688b0b7f-2f6c-d827-c27b-216a8e3ea700@2ndquadrant.com
2020-08-08 04:04:39 +02:00
Remove dead NoMovementScanDirection code Here remove some dead code from heapgettup() and heapgettup_pagemode() which was trying to support NoMovementScanDirection scans. This code can never be reached as standard_ExecutorRun() never calls ExecutePlan with NoMovementScanDirection. Additionally, plans which were scanning an unordered index would use NoMovementScanDirection rather than ForwardScanDirection. There was no real need for this, so here we adjust this so we use ForwardScanDirection for unordered index scans. A comment in pathnodes.h claimed that NoMovementScanDirection was used for PathKey reasons, but if that was true, it no longer is, per code in build_index_paths(). This does change the non-text format of the EXPLAIN output so that unordered index scans now have a "Forward" scan direction rather than "NoMovement". The text format of EXPLAIN has not changed. Author: Melanie Plageman Reviewed-by: Tom Lane, David Rowley Discussion: https://postgr.es/m/CAAKRu_bvkhka0CZQun28KTqhuUh5ZqY=_T8QEqZqOL02rpi2bw@mail.gmail.com
2023-01-31 22:52:41 +01:00
/* We don't expect actual scans using NoMovementScanDirection */
Assert(direction == ForwardScanDirection ||
direction == BackwardScanDirection);
Implement streaming mode in ReorderBuffer. Instead of serializing the transaction to disk after reaching the logical_decoding_work_mem limit in memory, we consume the changes we have in memory and invoke stream API methods added by commit 45fdc9738b. However, sometimes if we have incomplete toast or speculative insert we spill to the disk because we can't generate the complete tuple and stream. And, as soon as we get the complete tuple we stream the transaction including the serialized changes. We can do this incremental processing thanks to having assignments (associating subxact with toplevel xacts) in WAL right away, and thanks to logging the invalidation messages at each command end. These features are added by commits 0bead9af48 and c55040ccd0 respectively. Now that we can stream in-progress transactions, the concurrent aborts may cause failures when the output plugin consults catalogs (both system and user-defined). We handle such failures by returning ERRCODE_TRANSACTION_ROLLBACK sqlerrcode from system table scan APIs to the backend or WALSender decoding a specific uncommitted transaction. The decoding logic on the receipt of such a sqlerrcode aborts the decoding of the current transaction and continue with the decoding of other transactions. We have ReorderBufferTXN pointer in each ReorderBufferChange by which we know which xact it belongs to. The output plugin can use this to decide which changes to discard in case of stream_abort_cb (e.g. when a subxact gets discarded). We also provide a new option via SQL APIs to fetch the changes being streamed. Author: Dilip Kumar, Tomas Vondra, Amit Kapila, Nikhil Sontakke Reviewed-by: Amit Kapila, Kuntal Ghosh, Ajin Cherian Tested-by: Neha Sharma, Mahendra Singh Thalor and Ajin Cherian Discussion: https://postgr.es/m/688b0b7f-2f6c-d827-c27b-216a8e3ea700@2ndquadrant.com
2020-08-08 04:04:39 +02:00
/*
* We don't expect direct calls to table_scan_getnextslot with valid
* CheckXidAlive for catalog or regular tables. See detailed comments in
* xact.c where these variables are declared.
*/
if (unlikely(TransactionIdIsValid(CheckXidAlive) && !bsysscan))
elog(ERROR, "unexpected table_scan_getnextslot call during logical decoding");
tableam: Add and use scan APIs. Too allow table accesses to be not directly dependent on heap, several new abstractions are needed. Specifically: 1) Heap scans need to be generalized into table scans. Do this by introducing TableScanDesc, which will be the "base class" for individual AMs. This contains the AM independent fields from HeapScanDesc. The previous heap_{beginscan,rescan,endscan} et al. have been replaced with a table_ version. There's no direct replacement for heap_getnext(), as that returned a HeapTuple, which is undesirable for a other AMs. Instead there's table_scan_getnextslot(). But note that heap_getnext() lives on, it's still used widely to access catalog tables. This is achieved by new scan_begin, scan_end, scan_rescan, scan_getnextslot callbacks. 2) The portion of parallel scans that's shared between backends need to be able to do so without the user doing per-AM work. To achieve that new parallelscan_{estimate, initialize, reinitialize} callbacks are introduced, which operate on a new ParallelTableScanDesc, which again can be subclassed by AMs. As it is likely that several AMs are going to be block oriented, block oriented callbacks that can be shared between such AMs are provided and used by heap. table_block_parallelscan_{estimate, intiialize, reinitialize} as callbacks, and table_block_parallelscan_{nextpage, init} for use in AMs. These operate on a ParallelBlockTableScanDesc. 3) Index scans need to be able to access tables to return a tuple, and there needs to be state across individual accesses to the heap to store state like buffers. That's now handled by introducing a sort-of-scan IndexFetchTable, which again is intended to be subclassed by individual AMs (for heap IndexFetchHeap). The relevant callbacks for an AM are index_fetch_{end, begin, reset} to create the necessary state, and index_fetch_tuple to retrieve an indexed tuple. Note that index_fetch_tuple implementations need to be smarter than just blindly fetching the tuples for AMs that have optimizations similar to heap's HOT - the currently alive tuple in the update chain needs to be fetched if appropriate. Similar to table_scan_getnextslot(), it's undesirable to continue to return HeapTuples. Thus index_fetch_heap (might want to rename that later) now accepts a slot as an argument. Core code doesn't have a lot of call sites performing index scans without going through the systable_* API (in contrast to loads of heap_getnext calls and working directly with HeapTuples). Index scans now store the result of a search in IndexScanDesc->xs_heaptid, rather than xs_ctup->t_self. As the target is not generally a HeapTuple anymore that seems cleaner. To be able to sensible adapt code to use the above, two further callbacks have been introduced: a) slot_callbacks returns a TupleTableSlotOps* suitable for creating slots capable of holding a tuple of the AMs type. table_slot_callbacks() and table_slot_create() are based upon that, but have additional logic to deal with views, foreign tables, etc. While this change could have been done separately, nearly all the call sites that needed to be adapted for the rest of this commit also would have been needed to be adapted for table_slot_callbacks(), making separation not worthwhile. b) tuple_satisfies_snapshot checks whether the tuple in a slot is currently visible according to a snapshot. That's required as a few places now don't have a buffer + HeapTuple around, but a slot (which in heap's case internally has that information). Additionally a few infrastructure changes were needed: I) SysScanDesc, as used by systable_{beginscan, getnext} et al. now internally uses a slot to keep track of tuples. While systable_getnext() still returns HeapTuples, and will so for the foreseeable future, the index API (see 1) above) now only deals with slots. The remainder, and largest part, of this commit is then adjusting all scans in postgres to use the new APIs. Author: Andres Freund, Haribabu Kommi, Alvaro Herrera Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-11 20:46:41 +01:00
return sscan->rs_rd->rd_tableam->scan_getnextslot(sscan, direction, slot);
}
Add TID Range Scans to support efficient scanning ranges of TIDs This adds a new executor node named TID Range Scan. The query planner will generate paths for TID Range scans when quals are discovered on base relations which search for ranges on the table's ctid column. These ranges may be open at either end. For example, WHERE ctid >= '(10,0)'; will return all tuples on page 10 and over. To support this, two new optional callback functions have been added to table AM. scan_set_tidrange is used to set the scan range to just the given range of TIDs. scan_getnextslot_tidrange fetches the next tuple in the given range. For AMs were scanning ranges of TIDs would not make sense, these functions can be set to NULL in the TableAmRoutine. The query planner won't generate TID Range Scan Paths in that case. Author: Edmund Horner, David Rowley Reviewed-by: David Rowley, Tomas Vondra, Tom Lane, Andres Freund, Zhihong Yu Discussion: https://postgr.es/m/CAMyN-kB-nFTkF=VA_JPwFNo08S0d-Yk0F741S2B7LDmYAi8eyA@mail.gmail.com
2021-02-27 10:59:36 +01:00
/* ----------------------------------------------------------------------------
* TID Range scanning related functions.
* ----------------------------------------------------------------------------
*/
/*
* table_beginscan_tidrange is the entry point for setting up a TableScanDesc
* for a TID range scan.
*/
static inline TableScanDesc
table_beginscan_tidrange(Relation rel, Snapshot snapshot,
ItemPointer mintid,
ItemPointer maxtid)
{
TableScanDesc sscan;
uint32 flags = SO_TYPE_TIDRANGESCAN | SO_ALLOW_PAGEMODE;
sscan = rel->rd_tableam->scan_begin(rel, snapshot, 0, NULL, NULL, flags);
/* Set the range of TIDs to scan */
sscan->rs_rd->rd_tableam->scan_set_tidrange(sscan, mintid, maxtid);
return sscan;
}
/*
* table_rescan_tidrange resets the scan position and sets the minimum and
* maximum TID range to scan for a TableScanDesc created by
* table_beginscan_tidrange.
*/
static inline void
table_rescan_tidrange(TableScanDesc sscan, ItemPointer mintid,
ItemPointer maxtid)
{
/* Ensure table_beginscan_tidrange() was used. */
Assert((sscan->rs_flags & SO_TYPE_TIDRANGESCAN) != 0);
sscan->rs_rd->rd_tableam->scan_rescan(sscan, NULL, false, false, false, false);
sscan->rs_rd->rd_tableam->scan_set_tidrange(sscan, mintid, maxtid);
}
/*
* Fetch the next tuple from `sscan` for a TID range scan created by
* table_beginscan_tidrange(). Stores the tuple in `slot` and returns true,
* or returns false if no more tuples exist in the range.
*/
static inline bool
table_scan_getnextslot_tidrange(TableScanDesc sscan, ScanDirection direction,
TupleTableSlot *slot)
{
/* Ensure table_beginscan_tidrange() was used. */
Assert((sscan->rs_flags & SO_TYPE_TIDRANGESCAN) != 0);
Remove dead NoMovementScanDirection code Here remove some dead code from heapgettup() and heapgettup_pagemode() which was trying to support NoMovementScanDirection scans. This code can never be reached as standard_ExecutorRun() never calls ExecutePlan with NoMovementScanDirection. Additionally, plans which were scanning an unordered index would use NoMovementScanDirection rather than ForwardScanDirection. There was no real need for this, so here we adjust this so we use ForwardScanDirection for unordered index scans. A comment in pathnodes.h claimed that NoMovementScanDirection was used for PathKey reasons, but if that was true, it no longer is, per code in build_index_paths(). This does change the non-text format of the EXPLAIN output so that unordered index scans now have a "Forward" scan direction rather than "NoMovement". The text format of EXPLAIN has not changed. Author: Melanie Plageman Reviewed-by: Tom Lane, David Rowley Discussion: https://postgr.es/m/CAAKRu_bvkhka0CZQun28KTqhuUh5ZqY=_T8QEqZqOL02rpi2bw@mail.gmail.com
2023-01-31 22:52:41 +01:00
/* We don't expect actual scans using NoMovementScanDirection */
Assert(direction == ForwardScanDirection ||
direction == BackwardScanDirection);
Add TID Range Scans to support efficient scanning ranges of TIDs This adds a new executor node named TID Range Scan. The query planner will generate paths for TID Range scans when quals are discovered on base relations which search for ranges on the table's ctid column. These ranges may be open at either end. For example, WHERE ctid >= '(10,0)'; will return all tuples on page 10 and over. To support this, two new optional callback functions have been added to table AM. scan_set_tidrange is used to set the scan range to just the given range of TIDs. scan_getnextslot_tidrange fetches the next tuple in the given range. For AMs were scanning ranges of TIDs would not make sense, these functions can be set to NULL in the TableAmRoutine. The query planner won't generate TID Range Scan Paths in that case. Author: Edmund Horner, David Rowley Reviewed-by: David Rowley, Tomas Vondra, Tom Lane, Andres Freund, Zhihong Yu Discussion: https://postgr.es/m/CAMyN-kB-nFTkF=VA_JPwFNo08S0d-Yk0F741S2B7LDmYAi8eyA@mail.gmail.com
2021-02-27 10:59:36 +01:00
return sscan->rs_rd->rd_tableam->scan_getnextslot_tidrange(sscan,
direction,
slot);
}
tableam: Add and use scan APIs. Too allow table accesses to be not directly dependent on heap, several new abstractions are needed. Specifically: 1) Heap scans need to be generalized into table scans. Do this by introducing TableScanDesc, which will be the "base class" for individual AMs. This contains the AM independent fields from HeapScanDesc. The previous heap_{beginscan,rescan,endscan} et al. have been replaced with a table_ version. There's no direct replacement for heap_getnext(), as that returned a HeapTuple, which is undesirable for a other AMs. Instead there's table_scan_getnextslot(). But note that heap_getnext() lives on, it's still used widely to access catalog tables. This is achieved by new scan_begin, scan_end, scan_rescan, scan_getnextslot callbacks. 2) The portion of parallel scans that's shared between backends need to be able to do so without the user doing per-AM work. To achieve that new parallelscan_{estimate, initialize, reinitialize} callbacks are introduced, which operate on a new ParallelTableScanDesc, which again can be subclassed by AMs. As it is likely that several AMs are going to be block oriented, block oriented callbacks that can be shared between such AMs are provided and used by heap. table_block_parallelscan_{estimate, intiialize, reinitialize} as callbacks, and table_block_parallelscan_{nextpage, init} for use in AMs. These operate on a ParallelBlockTableScanDesc. 3) Index scans need to be able to access tables to return a tuple, and there needs to be state across individual accesses to the heap to store state like buffers. That's now handled by introducing a sort-of-scan IndexFetchTable, which again is intended to be subclassed by individual AMs (for heap IndexFetchHeap). The relevant callbacks for an AM are index_fetch_{end, begin, reset} to create the necessary state, and index_fetch_tuple to retrieve an indexed tuple. Note that index_fetch_tuple implementations need to be smarter than just blindly fetching the tuples for AMs that have optimizations similar to heap's HOT - the currently alive tuple in the update chain needs to be fetched if appropriate. Similar to table_scan_getnextslot(), it's undesirable to continue to return HeapTuples. Thus index_fetch_heap (might want to rename that later) now accepts a slot as an argument. Core code doesn't have a lot of call sites performing index scans without going through the systable_* API (in contrast to loads of heap_getnext calls and working directly with HeapTuples). Index scans now store the result of a search in IndexScanDesc->xs_heaptid, rather than xs_ctup->t_self. As the target is not generally a HeapTuple anymore that seems cleaner. To be able to sensible adapt code to use the above, two further callbacks have been introduced: a) slot_callbacks returns a TupleTableSlotOps* suitable for creating slots capable of holding a tuple of the AMs type. table_slot_callbacks() and table_slot_create() are based upon that, but have additional logic to deal with views, foreign tables, etc. While this change could have been done separately, nearly all the call sites that needed to be adapted for the rest of this commit also would have been needed to be adapted for table_slot_callbacks(), making separation not worthwhile. b) tuple_satisfies_snapshot checks whether the tuple in a slot is currently visible according to a snapshot. That's required as a few places now don't have a buffer + HeapTuple around, but a slot (which in heap's case internally has that information). Additionally a few infrastructure changes were needed: I) SysScanDesc, as used by systable_{beginscan, getnext} et al. now internally uses a slot to keep track of tuples. While systable_getnext() still returns HeapTuples, and will so for the foreseeable future, the index API (see 1) above) now only deals with slots. The remainder, and largest part, of this commit is then adjusting all scans in postgres to use the new APIs. Author: Andres Freund, Haribabu Kommi, Alvaro Herrera Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-11 20:46:41 +01:00
/* ----------------------------------------------------------------------------
* Parallel table scan related functions.
* ----------------------------------------------------------------------------
*/
/*
* Estimate the size of shared memory needed for a parallel scan of this
* relation.
*/
extern Size table_parallelscan_estimate(Relation rel, Snapshot snapshot);
/*
* Initialize ParallelTableScanDesc for a parallel scan of this
* relation. `pscan` needs to be sized according to parallelscan_estimate()
* for the same relation. Call this just once in the leader process; then,
* individual workers attach via table_beginscan_parallel.
*/
tableam: Formatting and other minor cleanups. The superflous heapam_xlog.h includes were reported by Peter Geoghegan.
2019-03-31 05:13:56 +02:00
extern void table_parallelscan_initialize(Relation rel,
ParallelTableScanDesc pscan,
Snapshot snapshot);
tableam: Add and use scan APIs. Too allow table accesses to be not directly dependent on heap, several new abstractions are needed. Specifically: 1) Heap scans need to be generalized into table scans. Do this by introducing TableScanDesc, which will be the "base class" for individual AMs. This contains the AM independent fields from HeapScanDesc. The previous heap_{beginscan,rescan,endscan} et al. have been replaced with a table_ version. There's no direct replacement for heap_getnext(), as that returned a HeapTuple, which is undesirable for a other AMs. Instead there's table_scan_getnextslot(). But note that heap_getnext() lives on, it's still used widely to access catalog tables. This is achieved by new scan_begin, scan_end, scan_rescan, scan_getnextslot callbacks. 2) The portion of parallel scans that's shared between backends need to be able to do so without the user doing per-AM work. To achieve that new parallelscan_{estimate, initialize, reinitialize} callbacks are introduced, which operate on a new ParallelTableScanDesc, which again can be subclassed by AMs. As it is likely that several AMs are going to be block oriented, block oriented callbacks that can be shared between such AMs are provided and used by heap. table_block_parallelscan_{estimate, intiialize, reinitialize} as callbacks, and table_block_parallelscan_{nextpage, init} for use in AMs. These operate on a ParallelBlockTableScanDesc. 3) Index scans need to be able to access tables to return a tuple, and there needs to be state across individual accesses to the heap to store state like buffers. That's now handled by introducing a sort-of-scan IndexFetchTable, which again is intended to be subclassed by individual AMs (for heap IndexFetchHeap). The relevant callbacks for an AM are index_fetch_{end, begin, reset} to create the necessary state, and index_fetch_tuple to retrieve an indexed tuple. Note that index_fetch_tuple implementations need to be smarter than just blindly fetching the tuples for AMs that have optimizations similar to heap's HOT - the currently alive tuple in the update chain needs to be fetched if appropriate. Similar to table_scan_getnextslot(), it's undesirable to continue to return HeapTuples. Thus index_fetch_heap (might want to rename that later) now accepts a slot as an argument. Core code doesn't have a lot of call sites performing index scans without going through the systable_* API (in contrast to loads of heap_getnext calls and working directly with HeapTuples). Index scans now store the result of a search in IndexScanDesc->xs_heaptid, rather than xs_ctup->t_self. As the target is not generally a HeapTuple anymore that seems cleaner. To be able to sensible adapt code to use the above, two further callbacks have been introduced: a) slot_callbacks returns a TupleTableSlotOps* suitable for creating slots capable of holding a tuple of the AMs type. table_slot_callbacks() and table_slot_create() are based upon that, but have additional logic to deal with views, foreign tables, etc. While this change could have been done separately, nearly all the call sites that needed to be adapted for the rest of this commit also would have been needed to be adapted for table_slot_callbacks(), making separation not worthwhile. b) tuple_satisfies_snapshot checks whether the tuple in a slot is currently visible according to a snapshot. That's required as a few places now don't have a buffer + HeapTuple around, but a slot (which in heap's case internally has that information). Additionally a few infrastructure changes were needed: I) SysScanDesc, as used by systable_{beginscan, getnext} et al. now internally uses a slot to keep track of tuples. While systable_getnext() still returns HeapTuples, and will so for the foreseeable future, the index API (see 1) above) now only deals with slots. The remainder, and largest part, of this commit is then adjusting all scans in postgres to use the new APIs. Author: Andres Freund, Haribabu Kommi, Alvaro Herrera Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-11 20:46:41 +01:00
/*
* Begin a parallel scan. `pscan` needs to have been initialized with
* table_parallelscan_initialize(), for the same relation. The initialization
* does not need to have happened in this backend.
*
tableam: comment and formatting fixes. Author: Heikki Linnakangas Discussion: https://postgr.es/m/9a7fb9cc-2419-5db7-8840-ddc10c93f122@iki.fi
2019-04-08 19:25:16 +02:00
* Caller must hold a suitable lock on the relation.
tableam: Add and use scan APIs. Too allow table accesses to be not directly dependent on heap, several new abstractions are needed. Specifically: 1) Heap scans need to be generalized into table scans. Do this by introducing TableScanDesc, which will be the "base class" for individual AMs. This contains the AM independent fields from HeapScanDesc. The previous heap_{beginscan,rescan,endscan} et al. have been replaced with a table_ version. There's no direct replacement for heap_getnext(), as that returned a HeapTuple, which is undesirable for a other AMs. Instead there's table_scan_getnextslot(). But note that heap_getnext() lives on, it's still used widely to access catalog tables. This is achieved by new scan_begin, scan_end, scan_rescan, scan_getnextslot callbacks. 2) The portion of parallel scans that's shared between backends need to be able to do so without the user doing per-AM work. To achieve that new parallelscan_{estimate, initialize, reinitialize} callbacks are introduced, which operate on a new ParallelTableScanDesc, which again can be subclassed by AMs. As it is likely that several AMs are going to be block oriented, block oriented callbacks that can be shared between such AMs are provided and used by heap. table_block_parallelscan_{estimate, intiialize, reinitialize} as callbacks, and table_block_parallelscan_{nextpage, init} for use in AMs. These operate on a ParallelBlockTableScanDesc. 3) Index scans need to be able to access tables to return a tuple, and there needs to be state across individual accesses to the heap to store state like buffers. That's now handled by introducing a sort-of-scan IndexFetchTable, which again is intended to be subclassed by individual AMs (for heap IndexFetchHeap). The relevant callbacks for an AM are index_fetch_{end, begin, reset} to create the necessary state, and index_fetch_tuple to retrieve an indexed tuple. Note that index_fetch_tuple implementations need to be smarter than just blindly fetching the tuples for AMs that have optimizations similar to heap's HOT - the currently alive tuple in the update chain needs to be fetched if appropriate. Similar to table_scan_getnextslot(), it's undesirable to continue to return HeapTuples. Thus index_fetch_heap (might want to rename that later) now accepts a slot as an argument. Core code doesn't have a lot of call sites performing index scans without going through the systable_* API (in contrast to loads of heap_getnext calls and working directly with HeapTuples). Index scans now store the result of a search in IndexScanDesc->xs_heaptid, rather than xs_ctup->t_self. As the target is not generally a HeapTuple anymore that seems cleaner. To be able to sensible adapt code to use the above, two further callbacks have been introduced: a) slot_callbacks returns a TupleTableSlotOps* suitable for creating slots capable of holding a tuple of the AMs type. table_slot_callbacks() and table_slot_create() are based upon that, but have additional logic to deal with views, foreign tables, etc. While this change could have been done separately, nearly all the call sites that needed to be adapted for the rest of this commit also would have been needed to be adapted for table_slot_callbacks(), making separation not worthwhile. b) tuple_satisfies_snapshot checks whether the tuple in a slot is currently visible according to a snapshot. That's required as a few places now don't have a buffer + HeapTuple around, but a slot (which in heap's case internally has that information). Additionally a few infrastructure changes were needed: I) SysScanDesc, as used by systable_{beginscan, getnext} et al. now internally uses a slot to keep track of tuples. While systable_getnext() still returns HeapTuples, and will so for the foreseeable future, the index API (see 1) above) now only deals with slots. The remainder, and largest part, of this commit is then adjusting all scans in postgres to use the new APIs. Author: Andres Freund, Haribabu Kommi, Alvaro Herrera Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-11 20:46:41 +01:00
*/
Harmonize heapam and tableam parameter names. Make sure that function declarations use names that exactly match the corresponding names from function definitions. Having parameter names that are reliably consistent in this way will make it easier to reason about groups of related C functions from the same translation unit as a module. It will also make certain refactoring tasks easier. Like other recent commits that cleaned up function parameter names, this commit was written with help from clang-tidy. Later commits will do the same for other parts of the codebase. Author: Peter Geoghegan <pg@bowt.ie> Reviewed-By: David Rowley <dgrowleyml@gmail.com> Discussion: https://postgr.es/m/CAH2-WznJt9CMM9KJTMjJh_zbL5hD9oX44qdJ4aqZtjFi-zA3Tg@mail.gmail.com
2022-09-20 01:46:23 +02:00
extern TableScanDesc table_beginscan_parallel(Relation relation,
tableam: Formatting and other minor cleanups. The superflous heapam_xlog.h includes were reported by Peter Geoghegan.
2019-03-31 05:13:56 +02:00
ParallelTableScanDesc pscan);
tableam: Add and use scan APIs. Too allow table accesses to be not directly dependent on heap, several new abstractions are needed. Specifically: 1) Heap scans need to be generalized into table scans. Do this by introducing TableScanDesc, which will be the "base class" for individual AMs. This contains the AM independent fields from HeapScanDesc. The previous heap_{beginscan,rescan,endscan} et al. have been replaced with a table_ version. There's no direct replacement for heap_getnext(), as that returned a HeapTuple, which is undesirable for a other AMs. Instead there's table_scan_getnextslot(). But note that heap_getnext() lives on, it's still used widely to access catalog tables. This is achieved by new scan_begin, scan_end, scan_rescan, scan_getnextslot callbacks. 2) The portion of parallel scans that's shared between backends need to be able to do so without the user doing per-AM work. To achieve that new parallelscan_{estimate, initialize, reinitialize} callbacks are introduced, which operate on a new ParallelTableScanDesc, which again can be subclassed by AMs. As it is likely that several AMs are going to be block oriented, block oriented callbacks that can be shared between such AMs are provided and used by heap. table_block_parallelscan_{estimate, intiialize, reinitialize} as callbacks, and table_block_parallelscan_{nextpage, init} for use in AMs. These operate on a ParallelBlockTableScanDesc. 3) Index scans need to be able to access tables to return a tuple, and there needs to be state across individual accesses to the heap to store state like buffers. That's now handled by introducing a sort-of-scan IndexFetchTable, which again is intended to be subclassed by individual AMs (for heap IndexFetchHeap). The relevant callbacks for an AM are index_fetch_{end, begin, reset} to create the necessary state, and index_fetch_tuple to retrieve an indexed tuple. Note that index_fetch_tuple implementations need to be smarter than just blindly fetching the tuples for AMs that have optimizations similar to heap's HOT - the currently alive tuple in the update chain needs to be fetched if appropriate. Similar to table_scan_getnextslot(), it's undesirable to continue to return HeapTuples. Thus index_fetch_heap (might want to rename that later) now accepts a slot as an argument. Core code doesn't have a lot of call sites performing index scans without going through the systable_* API (in contrast to loads of heap_getnext calls and working directly with HeapTuples). Index scans now store the result of a search in IndexScanDesc->xs_heaptid, rather than xs_ctup->t_self. As the target is not generally a HeapTuple anymore that seems cleaner. To be able to sensible adapt code to use the above, two further callbacks have been introduced: a) slot_callbacks returns a TupleTableSlotOps* suitable for creating slots capable of holding a tuple of the AMs type. table_slot_callbacks() and table_slot_create() are based upon that, but have additional logic to deal with views, foreign tables, etc. While this change could have been done separately, nearly all the call sites that needed to be adapted for the rest of this commit also would have been needed to be adapted for table_slot_callbacks(), making separation not worthwhile. b) tuple_satisfies_snapshot checks whether the tuple in a slot is currently visible according to a snapshot. That's required as a few places now don't have a buffer + HeapTuple around, but a slot (which in heap's case internally has that information). Additionally a few infrastructure changes were needed: I) SysScanDesc, as used by systable_{beginscan, getnext} et al. now internally uses a slot to keep track of tuples. While systable_getnext() still returns HeapTuples, and will so for the foreseeable future, the index API (see 1) above) now only deals with slots. The remainder, and largest part, of this commit is then adjusting all scans in postgres to use the new APIs. Author: Andres Freund, Haribabu Kommi, Alvaro Herrera Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-11 20:46:41 +01:00
/*
* Restart a parallel scan. Call this in the leader process. Caller is
* responsible for making sure that all workers have finished the scan
* beforehand.
*/
static inline void
table_parallelscan_reinitialize(Relation rel, ParallelTableScanDesc pscan)
{
Remove spurious return. Per buildfarm member anole. Author: Andres Freund
2019-03-11 23:03:27 +01:00
rel->rd_tableam->parallelscan_reinitialize(rel, pscan);
tableam: Add and use scan APIs. Too allow table accesses to be not directly dependent on heap, several new abstractions are needed. Specifically: 1) Heap scans need to be generalized into table scans. Do this by introducing TableScanDesc, which will be the "base class" for individual AMs. This contains the AM independent fields from HeapScanDesc. The previous heap_{beginscan,rescan,endscan} et al. have been replaced with a table_ version. There's no direct replacement for heap_getnext(), as that returned a HeapTuple, which is undesirable for a other AMs. Instead there's table_scan_getnextslot(). But note that heap_getnext() lives on, it's still used widely to access catalog tables. This is achieved by new scan_begin, scan_end, scan_rescan, scan_getnextslot callbacks. 2) The portion of parallel scans that's shared between backends need to be able to do so without the user doing per-AM work. To achieve that new parallelscan_{estimate, initialize, reinitialize} callbacks are introduced, which operate on a new ParallelTableScanDesc, which again can be subclassed by AMs. As it is likely that several AMs are going to be block oriented, block oriented callbacks that can be shared between such AMs are provided and used by heap. table_block_parallelscan_{estimate, intiialize, reinitialize} as callbacks, and table_block_parallelscan_{nextpage, init} for use in AMs. These operate on a ParallelBlockTableScanDesc. 3) Index scans need to be able to access tables to return a tuple, and there needs to be state across individual accesses to the heap to store state like buffers. That's now handled by introducing a sort-of-scan IndexFetchTable, which again is intended to be subclassed by individual AMs (for heap IndexFetchHeap). The relevant callbacks for an AM are index_fetch_{end, begin, reset} to create the necessary state, and index_fetch_tuple to retrieve an indexed tuple. Note that index_fetch_tuple implementations need to be smarter than just blindly fetching the tuples for AMs that have optimizations similar to heap's HOT - the currently alive tuple in the update chain needs to be fetched if appropriate. Similar to table_scan_getnextslot(), it's undesirable to continue to return HeapTuples. Thus index_fetch_heap (might want to rename that later) now accepts a slot as an argument. Core code doesn't have a lot of call sites performing index scans without going through the systable_* API (in contrast to loads of heap_getnext calls and working directly with HeapTuples). Index scans now store the result of a search in IndexScanDesc->xs_heaptid, rather than xs_ctup->t_self. As the target is not generally a HeapTuple anymore that seems cleaner. To be able to sensible adapt code to use the above, two further callbacks have been introduced: a) slot_callbacks returns a TupleTableSlotOps* suitable for creating slots capable of holding a tuple of the AMs type. table_slot_callbacks() and table_slot_create() are based upon that, but have additional logic to deal with views, foreign tables, etc. While this change could have been done separately, nearly all the call sites that needed to be adapted for the rest of this commit also would have been needed to be adapted for table_slot_callbacks(), making separation not worthwhile. b) tuple_satisfies_snapshot checks whether the tuple in a slot is currently visible according to a snapshot. That's required as a few places now don't have a buffer + HeapTuple around, but a slot (which in heap's case internally has that information). Additionally a few infrastructure changes were needed: I) SysScanDesc, as used by systable_{beginscan, getnext} et al. now internally uses a slot to keep track of tuples. While systable_getnext() still returns HeapTuples, and will so for the foreseeable future, the index API (see 1) above) now only deals with slots. The remainder, and largest part, of this commit is then adjusting all scans in postgres to use the new APIs. Author: Andres Freund, Haribabu Kommi, Alvaro Herrera Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-11 20:46:41 +01:00
}
/* ----------------------------------------------------------------------------
* Index scan related functions.
* ----------------------------------------------------------------------------
*/
/*
* Prepare to fetch tuples from the relation, as needed when fetching tuples
* for an index scan.
*
* Tuples for an index scan can then be fetched via table_index_fetch_tuple().
*/
static inline IndexFetchTableData *
table_index_fetch_begin(Relation rel)
{
return rel->rd_tableam->index_fetch_begin(rel);
}
/*
* Reset index fetch. Typically this will release cross index fetch resources
* held in IndexFetchTableData.
*/
static inline void
table_index_fetch_reset(struct IndexFetchTableData *scan)
{
scan->rel->rd_tableam->index_fetch_reset(scan);
}
/*
* Release resources and deallocate index fetch.
*/
static inline void
table_index_fetch_end(struct IndexFetchTableData *scan)
{
scan->rel->rd_tableam->index_fetch_end(scan);
}
/*
tableam: Add and use table_fetch_row_version(). This is essentially the tableam version of heapam_fetch(), i.e. fetching a tuple identified by a tid, performing visibility checks. Note that this different from table_index_fetch_tuple(), which is for index lookups. It therefore has to handle a tid pointing to an earlier version of a tuple if the AM uses an optimization like heap's HOT. Add comments to that end. This commit removes the stats_relation argument from heap_fetch, as it's been unused for a long time. Author: Andres Freund Reviewed-By: Haribabu Kommi Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-25 08:13:42 +01:00
* Fetches, as part of an index scan, tuple at `tid` into `slot`, after doing
* a visibility test according to `snapshot`. If a tuple was found and passed
Fix misuse of table_index_fetch_tuple_check(). Commit 0d861bbb, which added deduplication to nbtree, had _bt_check_unique() pass a TID to table_index_fetch_tuple_check() that isn't safe to mutate. table_index_fetch_tuple_check()'s tid argument is modified when the TID in question is not the latest visible tuple in a hot chain, though this wasn't documented. To fix, go back to using a local copy of the TID in _bt_check_unique(), and update comments above table_index_fetch_tuple_check(). Backpatch: 13-, where B-Tree deduplication was introduced.
2020-06-25 19:55:28 +02:00
* the visibility test, returns true, false otherwise. Note that *tid may be
* modified when we return true (see later remarks on multiple row versions
* reachable via a single index entry).
tableam: Add and use scan APIs. Too allow table accesses to be not directly dependent on heap, several new abstractions are needed. Specifically: 1) Heap scans need to be generalized into table scans. Do this by introducing TableScanDesc, which will be the "base class" for individual AMs. This contains the AM independent fields from HeapScanDesc. The previous heap_{beginscan,rescan,endscan} et al. have been replaced with a table_ version. There's no direct replacement for heap_getnext(), as that returned a HeapTuple, which is undesirable for a other AMs. Instead there's table_scan_getnextslot(). But note that heap_getnext() lives on, it's still used widely to access catalog tables. This is achieved by new scan_begin, scan_end, scan_rescan, scan_getnextslot callbacks. 2) The portion of parallel scans that's shared between backends need to be able to do so without the user doing per-AM work. To achieve that new parallelscan_{estimate, initialize, reinitialize} callbacks are introduced, which operate on a new ParallelTableScanDesc, which again can be subclassed by AMs. As it is likely that several AMs are going to be block oriented, block oriented callbacks that can be shared between such AMs are provided and used by heap. table_block_parallelscan_{estimate, intiialize, reinitialize} as callbacks, and table_block_parallelscan_{nextpage, init} for use in AMs. These operate on a ParallelBlockTableScanDesc. 3) Index scans need to be able to access tables to return a tuple, and there needs to be state across individual accesses to the heap to store state like buffers. That's now handled by introducing a sort-of-scan IndexFetchTable, which again is intended to be subclassed by individual AMs (for heap IndexFetchHeap). The relevant callbacks for an AM are index_fetch_{end, begin, reset} to create the necessary state, and index_fetch_tuple to retrieve an indexed tuple. Note that index_fetch_tuple implementations need to be smarter than just blindly fetching the tuples for AMs that have optimizations similar to heap's HOT - the currently alive tuple in the update chain needs to be fetched if appropriate. Similar to table_scan_getnextslot(), it's undesirable to continue to return HeapTuples. Thus index_fetch_heap (might want to rename that later) now accepts a slot as an argument. Core code doesn't have a lot of call sites performing index scans without going through the systable_* API (in contrast to loads of heap_getnext calls and working directly with HeapTuples). Index scans now store the result of a search in IndexScanDesc->xs_heaptid, rather than xs_ctup->t_self. As the target is not generally a HeapTuple anymore that seems cleaner. To be able to sensible adapt code to use the above, two further callbacks have been introduced: a) slot_callbacks returns a TupleTableSlotOps* suitable for creating slots capable of holding a tuple of the AMs type. table_slot_callbacks() and table_slot_create() are based upon that, but have additional logic to deal with views, foreign tables, etc. While this change could have been done separately, nearly all the call sites that needed to be adapted for the rest of this commit also would have been needed to be adapted for table_slot_callbacks(), making separation not worthwhile. b) tuple_satisfies_snapshot checks whether the tuple in a slot is currently visible according to a snapshot. That's required as a few places now don't have a buffer + HeapTuple around, but a slot (which in heap's case internally has that information). Additionally a few infrastructure changes were needed: I) SysScanDesc, as used by systable_{beginscan, getnext} et al. now internally uses a slot to keep track of tuples. While systable_getnext() still returns HeapTuples, and will so for the foreseeable future, the index API (see 1) above) now only deals with slots. The remainder, and largest part, of this commit is then adjusting all scans in postgres to use the new APIs. Author: Andres Freund, Haribabu Kommi, Alvaro Herrera Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-11 20:46:41 +01:00
*
* *call_again needs to be false on the first call to table_index_fetch_tuple() for
* a tid. If there potentially is another tuple matching the tid, *call_again
Spelling adjustments
2020-06-07 15:06:51 +02:00
* will be set to true, signaling that table_index_fetch_tuple() should be called
tableam: Add and use scan APIs. Too allow table accesses to be not directly dependent on heap, several new abstractions are needed. Specifically: 1) Heap scans need to be generalized into table scans. Do this by introducing TableScanDesc, which will be the "base class" for individual AMs. This contains the AM independent fields from HeapScanDesc. The previous heap_{beginscan,rescan,endscan} et al. have been replaced with a table_ version. There's no direct replacement for heap_getnext(), as that returned a HeapTuple, which is undesirable for a other AMs. Instead there's table_scan_getnextslot(). But note that heap_getnext() lives on, it's still used widely to access catalog tables. This is achieved by new scan_begin, scan_end, scan_rescan, scan_getnextslot callbacks. 2) The portion of parallel scans that's shared between backends need to be able to do so without the user doing per-AM work. To achieve that new parallelscan_{estimate, initialize, reinitialize} callbacks are introduced, which operate on a new ParallelTableScanDesc, which again can be subclassed by AMs. As it is likely that several AMs are going to be block oriented, block oriented callbacks that can be shared between such AMs are provided and used by heap. table_block_parallelscan_{estimate, intiialize, reinitialize} as callbacks, and table_block_parallelscan_{nextpage, init} for use in AMs. These operate on a ParallelBlockTableScanDesc. 3) Index scans need to be able to access tables to return a tuple, and there needs to be state across individual accesses to the heap to store state like buffers. That's now handled by introducing a sort-of-scan IndexFetchTable, which again is intended to be subclassed by individual AMs (for heap IndexFetchHeap). The relevant callbacks for an AM are index_fetch_{end, begin, reset} to create the necessary state, and index_fetch_tuple to retrieve an indexed tuple. Note that index_fetch_tuple implementations need to be smarter than just blindly fetching the tuples for AMs that have optimizations similar to heap's HOT - the currently alive tuple in the update chain needs to be fetched if appropriate. Similar to table_scan_getnextslot(), it's undesirable to continue to return HeapTuples. Thus index_fetch_heap (might want to rename that later) now accepts a slot as an argument. Core code doesn't have a lot of call sites performing index scans without going through the systable_* API (in contrast to loads of heap_getnext calls and working directly with HeapTuples). Index scans now store the result of a search in IndexScanDesc->xs_heaptid, rather than xs_ctup->t_self. As the target is not generally a HeapTuple anymore that seems cleaner. To be able to sensible adapt code to use the above, two further callbacks have been introduced: a) slot_callbacks returns a TupleTableSlotOps* suitable for creating slots capable of holding a tuple of the AMs type. table_slot_callbacks() and table_slot_create() are based upon that, but have additional logic to deal with views, foreign tables, etc. While this change could have been done separately, nearly all the call sites that needed to be adapted for the rest of this commit also would have been needed to be adapted for table_slot_callbacks(), making separation not worthwhile. b) tuple_satisfies_snapshot checks whether the tuple in a slot is currently visible according to a snapshot. That's required as a few places now don't have a buffer + HeapTuple around, but a slot (which in heap's case internally has that information). Additionally a few infrastructure changes were needed: I) SysScanDesc, as used by systable_{beginscan, getnext} et al. now internally uses a slot to keep track of tuples. While systable_getnext() still returns HeapTuples, and will so for the foreseeable future, the index API (see 1) above) now only deals with slots. The remainder, and largest part, of this commit is then adjusting all scans in postgres to use the new APIs. Author: Andres Freund, Haribabu Kommi, Alvaro Herrera Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-11 20:46:41 +01:00
* again for the same tid.
*
tableam: Add helper for indexes to check if a corresponding table tuples exist. This is, likely exclusively, useful to verify that conflicts detected in a unique index are with live tuples, rather than dead ones. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-26 00:52:55 +01:00
* *all_dead, if all_dead is not NULL, will be set to true by
* table_index_fetch_tuple() iff it is guaranteed that no backend needs to see
Fix few typos and minor wordsmithing in tableam comments. Reported-by: Ashwin Agrawal Author: Ashwin Agrawal Reviewed-by: Amit Kapila Backpatch-through: 12, where it was introduced Discussion: https://postgr.es/m/CALfoeisgdZhYDrJOukaBzvXfJOK2FQ0szVMK7dzmcy6w93iDUA@mail.gmail.com
2019-07-10 04:22:51 +02:00
* that tuple. Index AMs can use that to avoid returning that tid in future
tableam: Add helper for indexes to check if a corresponding table tuples exist. This is, likely exclusively, useful to verify that conflicts detected in a unique index are with live tuples, rather than dead ones. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-26 00:52:55 +01:00
* searches.
tableam: Add and use table_fetch_row_version(). This is essentially the tableam version of heapam_fetch(), i.e. fetching a tuple identified by a tid, performing visibility checks. Note that this different from table_index_fetch_tuple(), which is for index lookups. It therefore has to handle a tid pointing to an earlier version of a tuple if the AM uses an optimization like heap's HOT. Add comments to that end. This commit removes the stats_relation argument from heap_fetch, as it's been unused for a long time. Author: Andres Freund Reviewed-By: Haribabu Kommi Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-25 08:13:42 +01:00
*
Fix comments related to table AMs Incorrect function names were referenced. As this fixes some portions of tableam.h, that is mentioned in the docs as something to look at when implementing a table AM, backpatch down to 12 where this has been introduced. Author: Hironobu Suzuki Discussion: https://postgr.es/m/8fe6d672-28dd-3f1d-7aed-ac2f6d599d3f@interdb.jp Backpatch-through: 12
2020-07-14 06:17:11 +02:00
* The difference between this function and table_tuple_fetch_row_version()
* is that this function returns the currently visible version of a row if
* the AM supports storing multiple row versions reachable via a single index
* entry (like heap's HOT). Whereas table_tuple_fetch_row_version() only
* evaluates the tuple exactly at `tid`. Outside of index entry ->table tuple
* lookups, table_tuple_fetch_row_version() is what's usually needed.
tableam: Add and use scan APIs. Too allow table accesses to be not directly dependent on heap, several new abstractions are needed. Specifically: 1) Heap scans need to be generalized into table scans. Do this by introducing TableScanDesc, which will be the "base class" for individual AMs. This contains the AM independent fields from HeapScanDesc. The previous heap_{beginscan,rescan,endscan} et al. have been replaced with a table_ version. There's no direct replacement for heap_getnext(), as that returned a HeapTuple, which is undesirable for a other AMs. Instead there's table_scan_getnextslot(). But note that heap_getnext() lives on, it's still used widely to access catalog tables. This is achieved by new scan_begin, scan_end, scan_rescan, scan_getnextslot callbacks. 2) The portion of parallel scans that's shared between backends need to be able to do so without the user doing per-AM work. To achieve that new parallelscan_{estimate, initialize, reinitialize} callbacks are introduced, which operate on a new ParallelTableScanDesc, which again can be subclassed by AMs. As it is likely that several AMs are going to be block oriented, block oriented callbacks that can be shared between such AMs are provided and used by heap. table_block_parallelscan_{estimate, intiialize, reinitialize} as callbacks, and table_block_parallelscan_{nextpage, init} for use in AMs. These operate on a ParallelBlockTableScanDesc. 3) Index scans need to be able to access tables to return a tuple, and there needs to be state across individual accesses to the heap to store state like buffers. That's now handled by introducing a sort-of-scan IndexFetchTable, which again is intended to be subclassed by individual AMs (for heap IndexFetchHeap). The relevant callbacks for an AM are index_fetch_{end, begin, reset} to create the necessary state, and index_fetch_tuple to retrieve an indexed tuple. Note that index_fetch_tuple implementations need to be smarter than just blindly fetching the tuples for AMs that have optimizations similar to heap's HOT - the currently alive tuple in the update chain needs to be fetched if appropriate. Similar to table_scan_getnextslot(), it's undesirable to continue to return HeapTuples. Thus index_fetch_heap (might want to rename that later) now accepts a slot as an argument. Core code doesn't have a lot of call sites performing index scans without going through the systable_* API (in contrast to loads of heap_getnext calls and working directly with HeapTuples). Index scans now store the result of a search in IndexScanDesc->xs_heaptid, rather than xs_ctup->t_self. As the target is not generally a HeapTuple anymore that seems cleaner. To be able to sensible adapt code to use the above, two further callbacks have been introduced: a) slot_callbacks returns a TupleTableSlotOps* suitable for creating slots capable of holding a tuple of the AMs type. table_slot_callbacks() and table_slot_create() are based upon that, but have additional logic to deal with views, foreign tables, etc. While this change could have been done separately, nearly all the call sites that needed to be adapted for the rest of this commit also would have been needed to be adapted for table_slot_callbacks(), making separation not worthwhile. b) tuple_satisfies_snapshot checks whether the tuple in a slot is currently visible according to a snapshot. That's required as a few places now don't have a buffer + HeapTuple around, but a slot (which in heap's case internally has that information). Additionally a few infrastructure changes were needed: I) SysScanDesc, as used by systable_{beginscan, getnext} et al. now internally uses a slot to keep track of tuples. While systable_getnext() still returns HeapTuples, and will so for the foreseeable future, the index API (see 1) above) now only deals with slots. The remainder, and largest part, of this commit is then adjusting all scans in postgres to use the new APIs. Author: Andres Freund, Haribabu Kommi, Alvaro Herrera Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-11 20:46:41 +01:00
*/
static inline bool
table_index_fetch_tuple(struct IndexFetchTableData *scan,
ItemPointer tid,
Snapshot snapshot,
TupleTableSlot *slot,
bool *call_again, bool *all_dead)
{
Implement streaming mode in ReorderBuffer. Instead of serializing the transaction to disk after reaching the logical_decoding_work_mem limit in memory, we consume the changes we have in memory and invoke stream API methods added by commit 45fdc9738b. However, sometimes if we have incomplete toast or speculative insert we spill to the disk because we can't generate the complete tuple and stream. And, as soon as we get the complete tuple we stream the transaction including the serialized changes. We can do this incremental processing thanks to having assignments (associating subxact with toplevel xacts) in WAL right away, and thanks to logging the invalidation messages at each command end. These features are added by commits 0bead9af48 and c55040ccd0 respectively. Now that we can stream in-progress transactions, the concurrent aborts may cause failures when the output plugin consults catalogs (both system and user-defined). We handle such failures by returning ERRCODE_TRANSACTION_ROLLBACK sqlerrcode from system table scan APIs to the backend or WALSender decoding a specific uncommitted transaction. The decoding logic on the receipt of such a sqlerrcode aborts the decoding of the current transaction and continue with the decoding of other transactions. We have ReorderBufferTXN pointer in each ReorderBufferChange by which we know which xact it belongs to. The output plugin can use this to decide which changes to discard in case of stream_abort_cb (e.g. when a subxact gets discarded). We also provide a new option via SQL APIs to fetch the changes being streamed. Author: Dilip Kumar, Tomas Vondra, Amit Kapila, Nikhil Sontakke Reviewed-by: Amit Kapila, Kuntal Ghosh, Ajin Cherian Tested-by: Neha Sharma, Mahendra Singh Thalor and Ajin Cherian Discussion: https://postgr.es/m/688b0b7f-2f6c-d827-c27b-216a8e3ea700@2ndquadrant.com
2020-08-08 04:04:39 +02:00
/*
* We don't expect direct calls to table_index_fetch_tuple with valid
* CheckXidAlive for catalog or regular tables. See detailed comments in
* xact.c where these variables are declared.
*/
if (unlikely(TransactionIdIsValid(CheckXidAlive) && !bsysscan))
elog(ERROR, "unexpected table_index_fetch_tuple call during logical decoding");
tableam: Add and use scan APIs. Too allow table accesses to be not directly dependent on heap, several new abstractions are needed. Specifically: 1) Heap scans need to be generalized into table scans. Do this by introducing TableScanDesc, which will be the "base class" for individual AMs. This contains the AM independent fields from HeapScanDesc. The previous heap_{beginscan,rescan,endscan} et al. have been replaced with a table_ version. There's no direct replacement for heap_getnext(), as that returned a HeapTuple, which is undesirable for a other AMs. Instead there's table_scan_getnextslot(). But note that heap_getnext() lives on, it's still used widely to access catalog tables. This is achieved by new scan_begin, scan_end, scan_rescan, scan_getnextslot callbacks. 2) The portion of parallel scans that's shared between backends need to be able to do so without the user doing per-AM work. To achieve that new parallelscan_{estimate, initialize, reinitialize} callbacks are introduced, which operate on a new ParallelTableScanDesc, which again can be subclassed by AMs. As it is likely that several AMs are going to be block oriented, block oriented callbacks that can be shared between such AMs are provided and used by heap. table_block_parallelscan_{estimate, intiialize, reinitialize} as callbacks, and table_block_parallelscan_{nextpage, init} for use in AMs. These operate on a ParallelBlockTableScanDesc. 3) Index scans need to be able to access tables to return a tuple, and there needs to be state across individual accesses to the heap to store state like buffers. That's now handled by introducing a sort-of-scan IndexFetchTable, which again is intended to be subclassed by individual AMs (for heap IndexFetchHeap). The relevant callbacks for an AM are index_fetch_{end, begin, reset} to create the necessary state, and index_fetch_tuple to retrieve an indexed tuple. Note that index_fetch_tuple implementations need to be smarter than just blindly fetching the tuples for AMs that have optimizations similar to heap's HOT - the currently alive tuple in the update chain needs to be fetched if appropriate. Similar to table_scan_getnextslot(), it's undesirable to continue to return HeapTuples. Thus index_fetch_heap (might want to rename that later) now accepts a slot as an argument. Core code doesn't have a lot of call sites performing index scans without going through the systable_* API (in contrast to loads of heap_getnext calls and working directly with HeapTuples). Index scans now store the result of a search in IndexScanDesc->xs_heaptid, rather than xs_ctup->t_self. As the target is not generally a HeapTuple anymore that seems cleaner. To be able to sensible adapt code to use the above, two further callbacks have been introduced: a) slot_callbacks returns a TupleTableSlotOps* suitable for creating slots capable of holding a tuple of the AMs type. table_slot_callbacks() and table_slot_create() are based upon that, but have additional logic to deal with views, foreign tables, etc. While this change could have been done separately, nearly all the call sites that needed to be adapted for the rest of this commit also would have been needed to be adapted for table_slot_callbacks(), making separation not worthwhile. b) tuple_satisfies_snapshot checks whether the tuple in a slot is currently visible according to a snapshot. That's required as a few places now don't have a buffer + HeapTuple around, but a slot (which in heap's case internally has that information). Additionally a few infrastructure changes were needed: I) SysScanDesc, as used by systable_{beginscan, getnext} et al. now internally uses a slot to keep track of tuples. While systable_getnext() still returns HeapTuples, and will so for the foreseeable future, the index API (see 1) above) now only deals with slots. The remainder, and largest part, of this commit is then adjusting all scans in postgres to use the new APIs. Author: Andres Freund, Haribabu Kommi, Alvaro Herrera Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-11 20:46:41 +01:00
return scan->rel->rd_tableam->index_fetch_tuple(scan, tid, snapshot,
slot, call_again,
all_dead);
}
tableam: Add helper for indexes to check if a corresponding table tuples exist. This is, likely exclusively, useful to verify that conflicts detected in a unique index are with live tuples, rather than dead ones. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-26 00:52:55 +01:00
/*
* This is a convenience wrapper around table_index_fetch_tuple() which
* returns whether there are table tuple items corresponding to an index
* entry. This likely is only useful to verify if there's a conflict in a
* unique index.
*/
extern bool table_index_fetch_tuple_check(Relation rel,
ItemPointer tid,
Snapshot snapshot,
bool *all_dead);
tableam: Add and use scan APIs. Too allow table accesses to be not directly dependent on heap, several new abstractions are needed. Specifically: 1) Heap scans need to be generalized into table scans. Do this by introducing TableScanDesc, which will be the "base class" for individual AMs. This contains the AM independent fields from HeapScanDesc. The previous heap_{beginscan,rescan,endscan} et al. have been replaced with a table_ version. There's no direct replacement for heap_getnext(), as that returned a HeapTuple, which is undesirable for a other AMs. Instead there's table_scan_getnextslot(). But note that heap_getnext() lives on, it's still used widely to access catalog tables. This is achieved by new scan_begin, scan_end, scan_rescan, scan_getnextslot callbacks. 2) The portion of parallel scans that's shared between backends need to be able to do so without the user doing per-AM work. To achieve that new parallelscan_{estimate, initialize, reinitialize} callbacks are introduced, which operate on a new ParallelTableScanDesc, which again can be subclassed by AMs. As it is likely that several AMs are going to be block oriented, block oriented callbacks that can be shared between such AMs are provided and used by heap. table_block_parallelscan_{estimate, intiialize, reinitialize} as callbacks, and table_block_parallelscan_{nextpage, init} for use in AMs. These operate on a ParallelBlockTableScanDesc. 3) Index scans need to be able to access tables to return a tuple, and there needs to be state across individual accesses to the heap to store state like buffers. That's now handled by introducing a sort-of-scan IndexFetchTable, which again is intended to be subclassed by individual AMs (for heap IndexFetchHeap). The relevant callbacks for an AM are index_fetch_{end, begin, reset} to create the necessary state, and index_fetch_tuple to retrieve an indexed tuple. Note that index_fetch_tuple implementations need to be smarter than just blindly fetching the tuples for AMs that have optimizations similar to heap's HOT - the currently alive tuple in the update chain needs to be fetched if appropriate. Similar to table_scan_getnextslot(), it's undesirable to continue to return HeapTuples. Thus index_fetch_heap (might want to rename that later) now accepts a slot as an argument. Core code doesn't have a lot of call sites performing index scans without going through the systable_* API (in contrast to loads of heap_getnext calls and working directly with HeapTuples). Index scans now store the result of a search in IndexScanDesc->xs_heaptid, rather than xs_ctup->t_self. As the target is not generally a HeapTuple anymore that seems cleaner. To be able to sensible adapt code to use the above, two further callbacks have been introduced: a) slot_callbacks returns a TupleTableSlotOps* suitable for creating slots capable of holding a tuple of the AMs type. table_slot_callbacks() and table_slot_create() are based upon that, but have additional logic to deal with views, foreign tables, etc. While this change could have been done separately, nearly all the call sites that needed to be adapted for the rest of this commit also would have been needed to be adapted for table_slot_callbacks(), making separation not worthwhile. b) tuple_satisfies_snapshot checks whether the tuple in a slot is currently visible according to a snapshot. That's required as a few places now don't have a buffer + HeapTuple around, but a slot (which in heap's case internally has that information). Additionally a few infrastructure changes were needed: I) SysScanDesc, as used by systable_{beginscan, getnext} et al. now internally uses a slot to keep track of tuples. While systable_getnext() still returns HeapTuples, and will so for the foreseeable future, the index API (see 1) above) now only deals with slots. The remainder, and largest part, of this commit is then adjusting all scans in postgres to use the new APIs. Author: Andres Freund, Haribabu Kommi, Alvaro Herrera Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-11 20:46:41 +01:00
/* ------------------------------------------------------------------------
* Functions for non-modifying operations on individual tuples
* ------------------------------------------------------------------------
*/
tableam: introduce table AM infrastructure. This introduces the concept of table access methods, i.e. CREATE ACCESS METHOD ... TYPE TABLE and CREATE TABLE ... USING (storage-engine). No table access functionality is delegated to table AMs as of this commit, that'll be done in following commits. Subsequent commits will incrementally abstract table access functionality to be routed through table access methods. That change is too large to be reviewed & committed at once, so it'll be done incrementally. Docs will be updated at the end, as adding them incrementally would likely make them less coherent, and definitely is a lot more work, without a lot of benefit. Table access methods are specified similar to index access methods, i.e. pg_am.amhandler returns, as INTERNAL, a pointer to a struct with callbacks. In contrast to index AMs that struct needs to live as long as a backend, typically that's achieved by just returning a pointer to a constant struct. Psql's \d+ now displays a table's access method. That can be disabled with HIDE_TABLEAM=true, which is mainly useful so regression tests can be run against different AMs. It's quite possible that this behaviour still needs to be fine tuned. For now it's not allowed to set a table AM for a partitioned table, as we've not resolved how partitions would inherit that. Disallowing allows us to introduce, if we decide that's the way forward, such a behaviour without a compatibility break. Catversion bumped, to add the heap table AM and references to it. Author: Haribabu Kommi, Andres Freund, Alvaro Herrera, Dimitri Golgov and others Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql https://postgr.es/m/20190107235616.6lur25ph22u5u5av@alap3.anarazel.de https://postgr.es/m/20190304234700.w5tmhducs5wxgzls@alap3.anarazel.de
2019-03-06 18:54:38 +01:00
tableam: Add and use table_fetch_row_version(). This is essentially the tableam version of heapam_fetch(), i.e. fetching a tuple identified by a tid, performing visibility checks. Note that this different from table_index_fetch_tuple(), which is for index lookups. It therefore has to handle a tid pointing to an earlier version of a tuple if the AM uses an optimization like heap's HOT. Add comments to that end. This commit removes the stats_relation argument from heap_fetch, as it's been unused for a long time. Author: Andres Freund Reviewed-By: Haribabu Kommi Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-25 08:13:42 +01:00
/*
tableam: comment and formatting fixes. Author: Heikki Linnakangas Discussion: https://postgr.es/m/9a7fb9cc-2419-5db7-8840-ddc10c93f122@iki.fi
2019-04-08 19:25:16 +02:00
* Fetch tuple at `tid` into `slot`, after doing a visibility test according to
tableam: Add and use table_fetch_row_version(). This is essentially the tableam version of heapam_fetch(), i.e. fetching a tuple identified by a tid, performing visibility checks. Note that this different from table_index_fetch_tuple(), which is for index lookups. It therefore has to handle a tid pointing to an earlier version of a tuple if the AM uses an optimization like heap's HOT. Add comments to that end. This commit removes the stats_relation argument from heap_fetch, as it's been unused for a long time. Author: Andres Freund Reviewed-By: Haribabu Kommi Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-25 08:13:42 +01:00
* `snapshot`. If a tuple was found and passed the visibility test, returns
* true, false otherwise.
*
* See table_index_fetch_tuple's comment about what the difference between
Fix few typos and minor wordsmithing in tableam comments. Reported-by: Ashwin Agrawal Author: Ashwin Agrawal Reviewed-by: Amit Kapila Backpatch-through: 12, where it was introduced Discussion: https://postgr.es/m/CALfoeisgdZhYDrJOukaBzvXfJOK2FQ0szVMK7dzmcy6w93iDUA@mail.gmail.com
2019-07-10 04:22:51 +02:00
* these functions is. It is correct to use this function outside of index
* entry->table tuple lookups.
tableam: Add and use table_fetch_row_version(). This is essentially the tableam version of heapam_fetch(), i.e. fetching a tuple identified by a tid, performing visibility checks. Note that this different from table_index_fetch_tuple(), which is for index lookups. It therefore has to handle a tid pointing to an earlier version of a tuple if the AM uses an optimization like heap's HOT. Add comments to that end. This commit removes the stats_relation argument from heap_fetch, as it's been unused for a long time. Author: Andres Freund Reviewed-By: Haribabu Kommi Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-25 08:13:42 +01:00
*/
static inline bool
tableam: Rename wrapper functions to match callback names. Some of the wrapper functions didn't match the callback names. Many of them due to staying "consistent" with historic naming of the wrapped functionality. We decided that for most cases it's more important to be for tableam to be consistent going forward, than with the past. The one exception is beginscan/endscan/... because it'd have looked odd to have systable_beginscan/endscan/... with a different naming scheme, and changing the systable_* APIs would have caused way too much churn (including breaking a lot of external users). Author: Ashwin Agrawal, with some small additions by Andres Freund Reviewed-By: Andres Freund Discussion: https://postgr.es/m/CALfoeiugyrXZfX7n0ORCa4L-m834dzmaE8eFdbNR6PMpetU4Ww@mail.gmail.com
2019-05-24 01:25:48 +02:00
table_tuple_fetch_row_version(Relation rel,
ItemPointer tid,
Snapshot snapshot,
TupleTableSlot *slot)
tableam: Add and use table_fetch_row_version(). This is essentially the tableam version of heapam_fetch(), i.e. fetching a tuple identified by a tid, performing visibility checks. Note that this different from table_index_fetch_tuple(), which is for index lookups. It therefore has to handle a tid pointing to an earlier version of a tuple if the AM uses an optimization like heap's HOT. Add comments to that end. This commit removes the stats_relation argument from heap_fetch, as it's been unused for a long time. Author: Andres Freund Reviewed-By: Haribabu Kommi Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-25 08:13:42 +01:00
{
Implement streaming mode in ReorderBuffer. Instead of serializing the transaction to disk after reaching the logical_decoding_work_mem limit in memory, we consume the changes we have in memory and invoke stream API methods added by commit 45fdc9738b. However, sometimes if we have incomplete toast or speculative insert we spill to the disk because we can't generate the complete tuple and stream. And, as soon as we get the complete tuple we stream the transaction including the serialized changes. We can do this incremental processing thanks to having assignments (associating subxact with toplevel xacts) in WAL right away, and thanks to logging the invalidation messages at each command end. These features are added by commits 0bead9af48 and c55040ccd0 respectively. Now that we can stream in-progress transactions, the concurrent aborts may cause failures when the output plugin consults catalogs (both system and user-defined). We handle such failures by returning ERRCODE_TRANSACTION_ROLLBACK sqlerrcode from system table scan APIs to the backend or WALSender decoding a specific uncommitted transaction. The decoding logic on the receipt of such a sqlerrcode aborts the decoding of the current transaction and continue with the decoding of other transactions. We have ReorderBufferTXN pointer in each ReorderBufferChange by which we know which xact it belongs to. The output plugin can use this to decide which changes to discard in case of stream_abort_cb (e.g. when a subxact gets discarded). We also provide a new option via SQL APIs to fetch the changes being streamed. Author: Dilip Kumar, Tomas Vondra, Amit Kapila, Nikhil Sontakke Reviewed-by: Amit Kapila, Kuntal Ghosh, Ajin Cherian Tested-by: Neha Sharma, Mahendra Singh Thalor and Ajin Cherian Discussion: https://postgr.es/m/688b0b7f-2f6c-d827-c27b-216a8e3ea700@2ndquadrant.com
2020-08-08 04:04:39 +02:00
/*
* We don't expect direct calls to table_tuple_fetch_row_version with
* valid CheckXidAlive for catalog or regular tables. See detailed
* comments in xact.c where these variables are declared.
*/
if (unlikely(TransactionIdIsValid(CheckXidAlive) && !bsysscan))
elog(ERROR, "unexpected table_tuple_fetch_row_version call during logical decoding");
tableam: Add and use table_fetch_row_version(). This is essentially the tableam version of heapam_fetch(), i.e. fetching a tuple identified by a tid, performing visibility checks. Note that this different from table_index_fetch_tuple(), which is for index lookups. It therefore has to handle a tid pointing to an earlier version of a tuple if the AM uses an optimization like heap's HOT. Add comments to that end. This commit removes the stats_relation argument from heap_fetch, as it's been unused for a long time. Author: Andres Freund Reviewed-By: Haribabu Kommi Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-25 08:13:42 +01:00
return rel->rd_tableam->tuple_fetch_row_version(rel, tid, snapshot, slot);
}
tableam: Add table_get_latest_tid, to wrap heap_get_latest_tid. This primarily is to allow WHERE CURRENT OF to continue to work as it currently does. It's not clear to me that these semantics make sense for every AM, but it works for the in-core heap, and the out of core zheap. We can refine it further at a later point if necessary. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-26 01:14:48 +01:00
/*
tableam: Avoid relying on relation size to determine validity of tids. Instead add a tableam callback to do so. To avoid adding per validation overhead, pass a scan to tuple_tid_valid. In heap's case we'd otherwise incurred a RelationGetNumberOfBlocks() call for each tid - which'd have added noticable overhead to nodeTidscan.c. Author: Andres Freund Reviewed-By: Ashwin Agrawal Discussion: https://postgr.es/m/20190515185447.gno2jtqxyktylyvs@alap3.anarazel.de
2019-05-18 03:52:01 +02:00
* Verify that `tid` is a potentially valid tuple identifier. That doesn't
* mean that the pointed to row needs to exist or be visible, but that
Fix comments related to table AMs Incorrect function names were referenced. As this fixes some portions of tableam.h, that is mentioned in the docs as something to look at when implementing a table AM, backpatch down to 12 where this has been introduced. Author: Hironobu Suzuki Discussion: https://postgr.es/m/8fe6d672-28dd-3f1d-7aed-ac2f6d599d3f@interdb.jp Backpatch-through: 12
2020-07-14 06:17:11 +02:00
* attempting to fetch the row (e.g. with table_tuple_get_latest_tid() or
* table_tuple_fetch_row_version()) should not error out if called with that
* tid.
tableam: Avoid relying on relation size to determine validity of tids. Instead add a tableam callback to do so. To avoid adding per validation overhead, pass a scan to tuple_tid_valid. In heap's case we'd otherwise incurred a RelationGetNumberOfBlocks() call for each tid - which'd have added noticable overhead to nodeTidscan.c. Author: Andres Freund Reviewed-By: Ashwin Agrawal Discussion: https://postgr.es/m/20190515185447.gno2jtqxyktylyvs@alap3.anarazel.de
2019-05-18 03:52:01 +02:00
*
* `scan` needs to have been started via table_beginscan().
tableam: Add table_get_latest_tid, to wrap heap_get_latest_tid. This primarily is to allow WHERE CURRENT OF to continue to work as it currently does. It's not clear to me that these semantics make sense for every AM, but it works for the in-core heap, and the out of core zheap. We can refine it further at a later point if necessary. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-26 01:14:48 +01:00
*/
tableam: Avoid relying on relation size to determine validity of tids. Instead add a tableam callback to do so. To avoid adding per validation overhead, pass a scan to tuple_tid_valid. In heap's case we'd otherwise incurred a RelationGetNumberOfBlocks() call for each tid - which'd have added noticable overhead to nodeTidscan.c. Author: Andres Freund Reviewed-By: Ashwin Agrawal Discussion: https://postgr.es/m/20190515185447.gno2jtqxyktylyvs@alap3.anarazel.de
2019-05-18 03:52:01 +02:00
static inline bool
table_tuple_tid_valid(TableScanDesc scan, ItemPointer tid)
tableam: Add table_get_latest_tid, to wrap heap_get_latest_tid. This primarily is to allow WHERE CURRENT OF to continue to work as it currently does. It's not clear to me that these semantics make sense for every AM, but it works for the in-core heap, and the out of core zheap. We can refine it further at a later point if necessary. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-26 01:14:48 +01:00
{
tableam: Avoid relying on relation size to determine validity of tids. Instead add a tableam callback to do so. To avoid adding per validation overhead, pass a scan to tuple_tid_valid. In heap's case we'd otherwise incurred a RelationGetNumberOfBlocks() call for each tid - which'd have added noticable overhead to nodeTidscan.c. Author: Andres Freund Reviewed-By: Ashwin Agrawal Discussion: https://postgr.es/m/20190515185447.gno2jtqxyktylyvs@alap3.anarazel.de
2019-05-18 03:52:01 +02:00
return scan->rs_rd->rd_tableam->tuple_tid_valid(scan, tid);
tableam: Add table_get_latest_tid, to wrap heap_get_latest_tid. This primarily is to allow WHERE CURRENT OF to continue to work as it currently does. It's not clear to me that these semantics make sense for every AM, but it works for the in-core heap, and the out of core zheap. We can refine it further at a later point if necessary. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-26 01:14:48 +01:00
}
tableam: Avoid relying on relation size to determine validity of tids. Instead add a tableam callback to do so. To avoid adding per validation overhead, pass a scan to tuple_tid_valid. In heap's case we'd otherwise incurred a RelationGetNumberOfBlocks() call for each tid - which'd have added noticable overhead to nodeTidscan.c. Author: Andres Freund Reviewed-By: Ashwin Agrawal Discussion: https://postgr.es/m/20190515185447.gno2jtqxyktylyvs@alap3.anarazel.de
2019-05-18 03:52:01 +02:00
/*
* Return the latest version of the tuple at `tid`, by updating `tid` to
* point at the newest version.
*/
tableam: Rename wrapper functions to match callback names. Some of the wrapper functions didn't match the callback names. Many of them due to staying "consistent" with historic naming of the wrapped functionality. We decided that for most cases it's more important to be for tableam to be consistent going forward, than with the past. The one exception is beginscan/endscan/... because it'd have looked odd to have systable_beginscan/endscan/... with a different naming scheme, and changing the systable_* APIs would have caused way too much churn (including breaking a lot of external users). Author: Ashwin Agrawal, with some small additions by Andres Freund Reviewed-By: Andres Freund Discussion: https://postgr.es/m/CALfoeiugyrXZfX7n0ORCa4L-m834dzmaE8eFdbNR6PMpetU4Ww@mail.gmail.com
2019-05-24 01:25:48 +02:00
extern void table_tuple_get_latest_tid(TableScanDesc scan, ItemPointer tid);
tableam: Avoid relying on relation size to determine validity of tids. Instead add a tableam callback to do so. To avoid adding per validation overhead, pass a scan to tuple_tid_valid. In heap's case we'd otherwise incurred a RelationGetNumberOfBlocks() call for each tid - which'd have added noticable overhead to nodeTidscan.c. Author: Andres Freund Reviewed-By: Ashwin Agrawal Discussion: https://postgr.es/m/20190515185447.gno2jtqxyktylyvs@alap3.anarazel.de
2019-05-18 03:52:01 +02:00
tableam: introduce table AM infrastructure. This introduces the concept of table access methods, i.e. CREATE ACCESS METHOD ... TYPE TABLE and CREATE TABLE ... USING (storage-engine). No table access functionality is delegated to table AMs as of this commit, that'll be done in following commits. Subsequent commits will incrementally abstract table access functionality to be routed through table access methods. That change is too large to be reviewed & committed at once, so it'll be done incrementally. Docs will be updated at the end, as adding them incrementally would likely make them less coherent, and definitely is a lot more work, without a lot of benefit. Table access methods are specified similar to index access methods, i.e. pg_am.amhandler returns, as INTERNAL, a pointer to a struct with callbacks. In contrast to index AMs that struct needs to live as long as a backend, typically that's achieved by just returning a pointer to a constant struct. Psql's \d+ now displays a table's access method. That can be disabled with HIDE_TABLEAM=true, which is mainly useful so regression tests can be run against different AMs. It's quite possible that this behaviour still needs to be fine tuned. For now it's not allowed to set a table AM for a partitioned table, as we've not resolved how partitions would inherit that. Disallowing allows us to introduce, if we decide that's the way forward, such a behaviour without a compatibility break. Catversion bumped, to add the heap table AM and references to it. Author: Haribabu Kommi, Andres Freund, Alvaro Herrera, Dimitri Golgov and others Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql https://postgr.es/m/20190107235616.6lur25ph22u5u5av@alap3.anarazel.de https://postgr.es/m/20190304234700.w5tmhducs5wxgzls@alap3.anarazel.de
2019-03-06 18:54:38 +01:00
/*
tableam: Add and use scan APIs. Too allow table accesses to be not directly dependent on heap, several new abstractions are needed. Specifically: 1) Heap scans need to be generalized into table scans. Do this by introducing TableScanDesc, which will be the "base class" for individual AMs. This contains the AM independent fields from HeapScanDesc. The previous heap_{beginscan,rescan,endscan} et al. have been replaced with a table_ version. There's no direct replacement for heap_getnext(), as that returned a HeapTuple, which is undesirable for a other AMs. Instead there's table_scan_getnextslot(). But note that heap_getnext() lives on, it's still used widely to access catalog tables. This is achieved by new scan_begin, scan_end, scan_rescan, scan_getnextslot callbacks. 2) The portion of parallel scans that's shared between backends need to be able to do so without the user doing per-AM work. To achieve that new parallelscan_{estimate, initialize, reinitialize} callbacks are introduced, which operate on a new ParallelTableScanDesc, which again can be subclassed by AMs. As it is likely that several AMs are going to be block oriented, block oriented callbacks that can be shared between such AMs are provided and used by heap. table_block_parallelscan_{estimate, intiialize, reinitialize} as callbacks, and table_block_parallelscan_{nextpage, init} for use in AMs. These operate on a ParallelBlockTableScanDesc. 3) Index scans need to be able to access tables to return a tuple, and there needs to be state across individual accesses to the heap to store state like buffers. That's now handled by introducing a sort-of-scan IndexFetchTable, which again is intended to be subclassed by individual AMs (for heap IndexFetchHeap). The relevant callbacks for an AM are index_fetch_{end, begin, reset} to create the necessary state, and index_fetch_tuple to retrieve an indexed tuple. Note that index_fetch_tuple implementations need to be smarter than just blindly fetching the tuples for AMs that have optimizations similar to heap's HOT - the currently alive tuple in the update chain needs to be fetched if appropriate. Similar to table_scan_getnextslot(), it's undesirable to continue to return HeapTuples. Thus index_fetch_heap (might want to rename that later) now accepts a slot as an argument. Core code doesn't have a lot of call sites performing index scans without going through the systable_* API (in contrast to loads of heap_getnext calls and working directly with HeapTuples). Index scans now store the result of a search in IndexScanDesc->xs_heaptid, rather than xs_ctup->t_self. As the target is not generally a HeapTuple anymore that seems cleaner. To be able to sensible adapt code to use the above, two further callbacks have been introduced: a) slot_callbacks returns a TupleTableSlotOps* suitable for creating slots capable of holding a tuple of the AMs type. table_slot_callbacks() and table_slot_create() are based upon that, but have additional logic to deal with views, foreign tables, etc. While this change could have been done separately, nearly all the call sites that needed to be adapted for the rest of this commit also would have been needed to be adapted for table_slot_callbacks(), making separation not worthwhile. b) tuple_satisfies_snapshot checks whether the tuple in a slot is currently visible according to a snapshot. That's required as a few places now don't have a buffer + HeapTuple around, but a slot (which in heap's case internally has that information). Additionally a few infrastructure changes were needed: I) SysScanDesc, as used by systable_{beginscan, getnext} et al. now internally uses a slot to keep track of tuples. While systable_getnext() still returns HeapTuples, and will so for the foreseeable future, the index API (see 1) above) now only deals with slots. The remainder, and largest part, of this commit is then adjusting all scans in postgres to use the new APIs. Author: Andres Freund, Haribabu Kommi, Alvaro Herrera Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-11 20:46:41 +01:00
* Return true iff tuple in slot satisfies the snapshot.
*
* This assumes the slot's tuple is valid, and of the appropriate type for the
* AM.
*
* Some AMs might modify the data underlying the tuple as a side-effect. If so
* they ought to mark the relevant buffer dirty.
*/
static inline bool
tableam: Formatting and other minor cleanups. The superflous heapam_xlog.h includes were reported by Peter Geoghegan.
2019-03-31 05:13:56 +02:00
table_tuple_satisfies_snapshot(Relation rel, TupleTableSlot *slot,
Snapshot snapshot)
tableam: Add and use scan APIs. Too allow table accesses to be not directly dependent on heap, several new abstractions are needed. Specifically: 1) Heap scans need to be generalized into table scans. Do this by introducing TableScanDesc, which will be the "base class" for individual AMs. This contains the AM independent fields from HeapScanDesc. The previous heap_{beginscan,rescan,endscan} et al. have been replaced with a table_ version. There's no direct replacement for heap_getnext(), as that returned a HeapTuple, which is undesirable for a other AMs. Instead there's table_scan_getnextslot(). But note that heap_getnext() lives on, it's still used widely to access catalog tables. This is achieved by new scan_begin, scan_end, scan_rescan, scan_getnextslot callbacks. 2) The portion of parallel scans that's shared between backends need to be able to do so without the user doing per-AM work. To achieve that new parallelscan_{estimate, initialize, reinitialize} callbacks are introduced, which operate on a new ParallelTableScanDesc, which again can be subclassed by AMs. As it is likely that several AMs are going to be block oriented, block oriented callbacks that can be shared between such AMs are provided and used by heap. table_block_parallelscan_{estimate, intiialize, reinitialize} as callbacks, and table_block_parallelscan_{nextpage, init} for use in AMs. These operate on a ParallelBlockTableScanDesc. 3) Index scans need to be able to access tables to return a tuple, and there needs to be state across individual accesses to the heap to store state like buffers. That's now handled by introducing a sort-of-scan IndexFetchTable, which again is intended to be subclassed by individual AMs (for heap IndexFetchHeap). The relevant callbacks for an AM are index_fetch_{end, begin, reset} to create the necessary state, and index_fetch_tuple to retrieve an indexed tuple. Note that index_fetch_tuple implementations need to be smarter than just blindly fetching the tuples for AMs that have optimizations similar to heap's HOT - the currently alive tuple in the update chain needs to be fetched if appropriate. Similar to table_scan_getnextslot(), it's undesirable to continue to return HeapTuples. Thus index_fetch_heap (might want to rename that later) now accepts a slot as an argument. Core code doesn't have a lot of call sites performing index scans without going through the systable_* API (in contrast to loads of heap_getnext calls and working directly with HeapTuples). Index scans now store the result of a search in IndexScanDesc->xs_heaptid, rather than xs_ctup->t_self. As the target is not generally a HeapTuple anymore that seems cleaner. To be able to sensible adapt code to use the above, two further callbacks have been introduced: a) slot_callbacks returns a TupleTableSlotOps* suitable for creating slots capable of holding a tuple of the AMs type. table_slot_callbacks() and table_slot_create() are based upon that, but have additional logic to deal with views, foreign tables, etc. While this change could have been done separately, nearly all the call sites that needed to be adapted for the rest of this commit also would have been needed to be adapted for table_slot_callbacks(), making separation not worthwhile. b) tuple_satisfies_snapshot checks whether the tuple in a slot is currently visible according to a snapshot. That's required as a few places now don't have a buffer + HeapTuple around, but a slot (which in heap's case internally has that information). Additionally a few infrastructure changes were needed: I) SysScanDesc, as used by systable_{beginscan, getnext} et al. now internally uses a slot to keep track of tuples. While systable_getnext() still returns HeapTuples, and will so for the foreseeable future, the index API (see 1) above) now only deals with slots. The remainder, and largest part, of this commit is then adjusting all scans in postgres to use the new APIs. Author: Andres Freund, Haribabu Kommi, Alvaro Herrera Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-11 20:46:41 +01:00
{
return rel->rd_tableam->tuple_satisfies_snapshot(rel, slot, snapshot);
}
Compute XID horizon for page level index vacuum on primary. Previously the xid horizon was only computed during WAL replay. That had two major problems: 1) It relied on knowing what the table pointed to looks like. That was easy enough before the introducing of tableam (we knew it had to be heap, although some trickery around logging the heap relfilenodes was required). But to properly handle table AMs we need per-database catalog access to look up the AM handler, which recovery doesn't allow. 2) Not knowing the xid horizon also makes it hard to support logical decoding on standbys. When on a catalog table, we need to be able to conflict with slots that have an xid horizon that's too old. But computing the horizon by visiting the heap only works once consistency is reached, but we always need to be able to detect conflicts. There's also a secondary problem, in that the current method performs redundant work on every standby. But that's counterbalanced by potentially computing the value when not necessary (either because there's no standby, or because there's no connected backends). Solve 1) and 2) by moving computation of the xid horizon to the primary and by involving tableam in the computation of the horizon. To address the potentially increased overhead, increase the efficiency of the xid horizon computation for heap by sorting the tids, and eliminating redundant buffer accesses. When prefetching is available, additionally perform prefetching of buffers. As this is more of a maintenance task, rather than something routinely done in every read only query, we add an arbitrary 10 to the effective concurrency - thereby using IO concurrency, when not globally enabled. That's possibly not the perfect formula, but seems good enough for now. Bumps WAL format, as latestRemovedXid is now part of the records, and the heap's relfilenode isn't anymore. Author: Andres Freund, Amit Khandekar, Robert Haas Reviewed-By: Robert Haas Discussion: https://postgr.es/m/20181212204154.nsxf3gzqv3gesl32@alap3.anarazel.de https://postgr.es/m/20181214014235.dal5ogljs3bmlq44@alap3.anarazel.de https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-26 22:41:46 +01:00
/*
Enhance nbtree index tuple deletion. Teach nbtree and heapam to cooperate in order to eagerly remove duplicate tuples representing dead MVCC versions. This is "bottom-up deletion". Each bottom-up deletion pass is triggered lazily in response to a flood of versions on an nbtree leaf page. This usually involves a "logically unchanged index" hint (these are produced by the executor mechanism added by commit 9dc718bd). The immediate goal of bottom-up index deletion is to avoid "unnecessary" page splits caused entirely by version duplicates. It naturally has an even more useful effect, though: it acts as a backstop against accumulating an excessive number of index tuple versions for any given _logical row_. Bottom-up index deletion complements what we might now call "top-down index deletion": index vacuuming performed by VACUUM. Bottom-up index deletion responds to the immediate local needs of queries, while leaving it up to autovacuum to perform infrequent clean sweeps of the index. The overall effect is to avoid certain pathological performance issues related to "version churn" from UPDATEs. The previous tableam interface used by index AMs to perform tuple deletion (the table_compute_xid_horizon_for_tuples() function) has been replaced with a new interface that supports certain new requirements. Many (perhaps all) of the capabilities added to nbtree by this commit could also be extended to other index AMs. That is left as work for a later commit. Extend deletion of LP_DEAD-marked index tuples in nbtree by adding logic to consider extra index tuples (that are not LP_DEAD-marked) for deletion in passing. This increases the number of index tuples deleted significantly in many cases. The LP_DEAD deletion process (which is now called "simple deletion" to clearly distinguish it from bottom-up deletion) won't usually need to visit any extra table blocks to check these extra tuples. We have to visit the same table blocks anyway to generate a latestRemovedXid value (at least in the common case where the index deletion operation's WAL record needs such a value). Testing has shown that the "extra tuples" simple deletion enhancement increases the number of index tuples deleted with almost any workload that has LP_DEAD bits set in leaf pages. That is, it almost never fails to delete at least a few extra index tuples. It helps most of all in cases that happen to naturally have a lot of delete-safe tuples. It's not uncommon for an individual deletion operation to end up deleting an order of magnitude more index tuples compared to the old naive approach (e.g., custom instrumentation of the patch shows that this happens fairly often when the regression tests are run). Add a further enhancement that augments simple deletion and bottom-up deletion in indexes that make use of deduplication: Teach nbtree's _bt_delitems_delete() function to support granular TID deletion in posting list tuples. It is now possible to delete individual TIDs from posting list tuples provided the TIDs have a tableam block number of a table block that gets visited as part of the deletion process (visiting the table block can be triggered directly or indirectly). Setting the LP_DEAD bit of a posting list tuple is still an all-or-nothing thing, but that matters much less now that deletion only needs to start out with the right _general_ idea about which index tuples are deletable. Bump XLOG_PAGE_MAGIC because xl_btree_delete changed. No bump in BTREE_VERSION, since there are no changes to the on-disk representation of nbtree indexes. Indexes built on PostgreSQL 12 or PostgreSQL 13 will automatically benefit from bottom-up index deletion (i.e. no reindexing required) following a pg_upgrade. The enhancement to simple deletion is available with all B-Tree indexes following a pg_upgrade, no matter what PostgreSQL version the user upgrades from. Author: Peter Geoghegan <pg@bowt.ie> Reviewed-By: Heikki Linnakangas <hlinnaka@iki.fi> Reviewed-By: Victor Yegorov <vyegorov@gmail.com> Discussion: https://postgr.es/m/CAH2-Wzm+maE3apHB8NOtmM=p-DO65j2V5GzAWCOEEuy3JZgb2g@mail.gmail.com
2021-01-13 18:21:32 +01:00
* Determine which index tuples are safe to delete based on their table TID.
*
* Determines which entries from index AM caller's TM_IndexDeleteOp state
* point to vacuumable table tuples. Entries that are found by tableam to be
* vacuumable are naturally safe for index AM to delete, and so get directly
* marked as deletable. See comments above TM_IndexDelete and comments above
* TM_IndexDeleteOp for full details.
*
Standardize rmgrdesc recovery conflict XID output. Standardize on the name snapshotConflictHorizon for all XID fields from WAL records that generate recovery conflicts when in hot standby mode. This supersedes the previous latestRemovedXid naming convention. The new naming convention places emphasis on how the values are actually used by REDO routines. How the values are generated during original execution (details of which vary by record type) is deemphasized. Users of tools like pg_waldump can now grep for snapshotConflictHorizon to see all potential sources of recovery conflicts in a standardized way, without necessarily having to consider which specific record types might be involved. Also bring a couple of WAL record types that didn't follow any kind of naming convention into line. These are heapam's VISIBLE record type and SP-GiST's VACUUM_REDIRECT record type. Now every WAL record whose REDO routine calls ResolveRecoveryConflictWithSnapshot() passes through the snapshotConflictHorizon field from its WAL record. This is follow-up work to the refactoring from commit 9e540599 that made FREEZE_PAGE WAL records use a standard snapshotConflictHorizon style XID cutoff. No bump in XLOG_PAGE_MAGIC, since the underlying format of affected WAL records doesn't change. Author: Peter Geoghegan <pg@bowt.ie> Reviewed-By: Andres Freund <andres@anarazel.de> Discussion: https://postgr.es/m/CAH2-Wzm2CQUmViUq7Opgk=McVREHSOorYaAjR1ZpLYkRN7_dPw@mail.gmail.com
2022-11-17 23:55:08 +01:00
* Returns a snapshotConflictHorizon transaction ID that caller places in
Enhance nbtree index tuple deletion. Teach nbtree and heapam to cooperate in order to eagerly remove duplicate tuples representing dead MVCC versions. This is "bottom-up deletion". Each bottom-up deletion pass is triggered lazily in response to a flood of versions on an nbtree leaf page. This usually involves a "logically unchanged index" hint (these are produced by the executor mechanism added by commit 9dc718bd). The immediate goal of bottom-up index deletion is to avoid "unnecessary" page splits caused entirely by version duplicates. It naturally has an even more useful effect, though: it acts as a backstop against accumulating an excessive number of index tuple versions for any given _logical row_. Bottom-up index deletion complements what we might now call "top-down index deletion": index vacuuming performed by VACUUM. Bottom-up index deletion responds to the immediate local needs of queries, while leaving it up to autovacuum to perform infrequent clean sweeps of the index. The overall effect is to avoid certain pathological performance issues related to "version churn" from UPDATEs. The previous tableam interface used by index AMs to perform tuple deletion (the table_compute_xid_horizon_for_tuples() function) has been replaced with a new interface that supports certain new requirements. Many (perhaps all) of the capabilities added to nbtree by this commit could also be extended to other index AMs. That is left as work for a later commit. Extend deletion of LP_DEAD-marked index tuples in nbtree by adding logic to consider extra index tuples (that are not LP_DEAD-marked) for deletion in passing. This increases the number of index tuples deleted significantly in many cases. The LP_DEAD deletion process (which is now called "simple deletion" to clearly distinguish it from bottom-up deletion) won't usually need to visit any extra table blocks to check these extra tuples. We have to visit the same table blocks anyway to generate a latestRemovedXid value (at least in the common case where the index deletion operation's WAL record needs such a value). Testing has shown that the "extra tuples" simple deletion enhancement increases the number of index tuples deleted with almost any workload that has LP_DEAD bits set in leaf pages. That is, it almost never fails to delete at least a few extra index tuples. It helps most of all in cases that happen to naturally have a lot of delete-safe tuples. It's not uncommon for an individual deletion operation to end up deleting an order of magnitude more index tuples compared to the old naive approach (e.g., custom instrumentation of the patch shows that this happens fairly often when the regression tests are run). Add a further enhancement that augments simple deletion and bottom-up deletion in indexes that make use of deduplication: Teach nbtree's _bt_delitems_delete() function to support granular TID deletion in posting list tuples. It is now possible to delete individual TIDs from posting list tuples provided the TIDs have a tableam block number of a table block that gets visited as part of the deletion process (visiting the table block can be triggered directly or indirectly). Setting the LP_DEAD bit of a posting list tuple is still an all-or-nothing thing, but that matters much less now that deletion only needs to start out with the right _general_ idea about which index tuples are deletable. Bump XLOG_PAGE_MAGIC because xl_btree_delete changed. No bump in BTREE_VERSION, since there are no changes to the on-disk representation of nbtree indexes. Indexes built on PostgreSQL 12 or PostgreSQL 13 will automatically benefit from bottom-up index deletion (i.e. no reindexing required) following a pg_upgrade. The enhancement to simple deletion is available with all B-Tree indexes following a pg_upgrade, no matter what PostgreSQL version the user upgrades from. Author: Peter Geoghegan <pg@bowt.ie> Reviewed-By: Heikki Linnakangas <hlinnaka@iki.fi> Reviewed-By: Victor Yegorov <vyegorov@gmail.com> Discussion: https://postgr.es/m/CAH2-Wzm+maE3apHB8NOtmM=p-DO65j2V5GzAWCOEEuy3JZgb2g@mail.gmail.com
2021-01-13 18:21:32 +01:00
* its index deletion WAL record. This might be used during subsequent REDO
* of the WAL record when in Hot Standby mode -- a recovery conflict for the
* index deletion operation might be required on the standby.
Compute XID horizon for page level index vacuum on primary. Previously the xid horizon was only computed during WAL replay. That had two major problems: 1) It relied on knowing what the table pointed to looks like. That was easy enough before the introducing of tableam (we knew it had to be heap, although some trickery around logging the heap relfilenodes was required). But to properly handle table AMs we need per-database catalog access to look up the AM handler, which recovery doesn't allow. 2) Not knowing the xid horizon also makes it hard to support logical decoding on standbys. When on a catalog table, we need to be able to conflict with slots that have an xid horizon that's too old. But computing the horizon by visiting the heap only works once consistency is reached, but we always need to be able to detect conflicts. There's also a secondary problem, in that the current method performs redundant work on every standby. But that's counterbalanced by potentially computing the value when not necessary (either because there's no standby, or because there's no connected backends). Solve 1) and 2) by moving computation of the xid horizon to the primary and by involving tableam in the computation of the horizon. To address the potentially increased overhead, increase the efficiency of the xid horizon computation for heap by sorting the tids, and eliminating redundant buffer accesses. When prefetching is available, additionally perform prefetching of buffers. As this is more of a maintenance task, rather than something routinely done in every read only query, we add an arbitrary 10 to the effective concurrency - thereby using IO concurrency, when not globally enabled. That's possibly not the perfect formula, but seems good enough for now. Bumps WAL format, as latestRemovedXid is now part of the records, and the heap's relfilenode isn't anymore. Author: Andres Freund, Amit Khandekar, Robert Haas Reviewed-By: Robert Haas Discussion: https://postgr.es/m/20181212204154.nsxf3gzqv3gesl32@alap3.anarazel.de https://postgr.es/m/20181214014235.dal5ogljs3bmlq44@alap3.anarazel.de https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-26 22:41:46 +01:00
*/
static inline TransactionId
Enhance nbtree index tuple deletion. Teach nbtree and heapam to cooperate in order to eagerly remove duplicate tuples representing dead MVCC versions. This is "bottom-up deletion". Each bottom-up deletion pass is triggered lazily in response to a flood of versions on an nbtree leaf page. This usually involves a "logically unchanged index" hint (these are produced by the executor mechanism added by commit 9dc718bd). The immediate goal of bottom-up index deletion is to avoid "unnecessary" page splits caused entirely by version duplicates. It naturally has an even more useful effect, though: it acts as a backstop against accumulating an excessive number of index tuple versions for any given _logical row_. Bottom-up index deletion complements what we might now call "top-down index deletion": index vacuuming performed by VACUUM. Bottom-up index deletion responds to the immediate local needs of queries, while leaving it up to autovacuum to perform infrequent clean sweeps of the index. The overall effect is to avoid certain pathological performance issues related to "version churn" from UPDATEs. The previous tableam interface used by index AMs to perform tuple deletion (the table_compute_xid_horizon_for_tuples() function) has been replaced with a new interface that supports certain new requirements. Many (perhaps all) of the capabilities added to nbtree by this commit could also be extended to other index AMs. That is left as work for a later commit. Extend deletion of LP_DEAD-marked index tuples in nbtree by adding logic to consider extra index tuples (that are not LP_DEAD-marked) for deletion in passing. This increases the number of index tuples deleted significantly in many cases. The LP_DEAD deletion process (which is now called "simple deletion" to clearly distinguish it from bottom-up deletion) won't usually need to visit any extra table blocks to check these extra tuples. We have to visit the same table blocks anyway to generate a latestRemovedXid value (at least in the common case where the index deletion operation's WAL record needs such a value). Testing has shown that the "extra tuples" simple deletion enhancement increases the number of index tuples deleted with almost any workload that has LP_DEAD bits set in leaf pages. That is, it almost never fails to delete at least a few extra index tuples. It helps most of all in cases that happen to naturally have a lot of delete-safe tuples. It's not uncommon for an individual deletion operation to end up deleting an order of magnitude more index tuples compared to the old naive approach (e.g., custom instrumentation of the patch shows that this happens fairly often when the regression tests are run). Add a further enhancement that augments simple deletion and bottom-up deletion in indexes that make use of deduplication: Teach nbtree's _bt_delitems_delete() function to support granular TID deletion in posting list tuples. It is now possible to delete individual TIDs from posting list tuples provided the TIDs have a tableam block number of a table block that gets visited as part of the deletion process (visiting the table block can be triggered directly or indirectly). Setting the LP_DEAD bit of a posting list tuple is still an all-or-nothing thing, but that matters much less now that deletion only needs to start out with the right _general_ idea about which index tuples are deletable. Bump XLOG_PAGE_MAGIC because xl_btree_delete changed. No bump in BTREE_VERSION, since there are no changes to the on-disk representation of nbtree indexes. Indexes built on PostgreSQL 12 or PostgreSQL 13 will automatically benefit from bottom-up index deletion (i.e. no reindexing required) following a pg_upgrade. The enhancement to simple deletion is available with all B-Tree indexes following a pg_upgrade, no matter what PostgreSQL version the user upgrades from. Author: Peter Geoghegan <pg@bowt.ie> Reviewed-By: Heikki Linnakangas <hlinnaka@iki.fi> Reviewed-By: Victor Yegorov <vyegorov@gmail.com> Discussion: https://postgr.es/m/CAH2-Wzm+maE3apHB8NOtmM=p-DO65j2V5GzAWCOEEuy3JZgb2g@mail.gmail.com
2021-01-13 18:21:32 +01:00
table_index_delete_tuples(Relation rel, TM_IndexDeleteOp *delstate)
Compute XID horizon for page level index vacuum on primary. Previously the xid horizon was only computed during WAL replay. That had two major problems: 1) It relied on knowing what the table pointed to looks like. That was easy enough before the introducing of tableam (we knew it had to be heap, although some trickery around logging the heap relfilenodes was required). But to properly handle table AMs we need per-database catalog access to look up the AM handler, which recovery doesn't allow. 2) Not knowing the xid horizon also makes it hard to support logical decoding on standbys. When on a catalog table, we need to be able to conflict with slots that have an xid horizon that's too old. But computing the horizon by visiting the heap only works once consistency is reached, but we always need to be able to detect conflicts. There's also a secondary problem, in that the current method performs redundant work on every standby. But that's counterbalanced by potentially computing the value when not necessary (either because there's no standby, or because there's no connected backends). Solve 1) and 2) by moving computation of the xid horizon to the primary and by involving tableam in the computation of the horizon. To address the potentially increased overhead, increase the efficiency of the xid horizon computation for heap by sorting the tids, and eliminating redundant buffer accesses. When prefetching is available, additionally perform prefetching of buffers. As this is more of a maintenance task, rather than something routinely done in every read only query, we add an arbitrary 10 to the effective concurrency - thereby using IO concurrency, when not globally enabled. That's possibly not the perfect formula, but seems good enough for now. Bumps WAL format, as latestRemovedXid is now part of the records, and the heap's relfilenode isn't anymore. Author: Andres Freund, Amit Khandekar, Robert Haas Reviewed-By: Robert Haas Discussion: https://postgr.es/m/20181212204154.nsxf3gzqv3gesl32@alap3.anarazel.de https://postgr.es/m/20181214014235.dal5ogljs3bmlq44@alap3.anarazel.de https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-26 22:41:46 +01:00
{
Enhance nbtree index tuple deletion. Teach nbtree and heapam to cooperate in order to eagerly remove duplicate tuples representing dead MVCC versions. This is "bottom-up deletion". Each bottom-up deletion pass is triggered lazily in response to a flood of versions on an nbtree leaf page. This usually involves a "logically unchanged index" hint (these are produced by the executor mechanism added by commit 9dc718bd). The immediate goal of bottom-up index deletion is to avoid "unnecessary" page splits caused entirely by version duplicates. It naturally has an even more useful effect, though: it acts as a backstop against accumulating an excessive number of index tuple versions for any given _logical row_. Bottom-up index deletion complements what we might now call "top-down index deletion": index vacuuming performed by VACUUM. Bottom-up index deletion responds to the immediate local needs of queries, while leaving it up to autovacuum to perform infrequent clean sweeps of the index. The overall effect is to avoid certain pathological performance issues related to "version churn" from UPDATEs. The previous tableam interface used by index AMs to perform tuple deletion (the table_compute_xid_horizon_for_tuples() function) has been replaced with a new interface that supports certain new requirements. Many (perhaps all) of the capabilities added to nbtree by this commit could also be extended to other index AMs. That is left as work for a later commit. Extend deletion of LP_DEAD-marked index tuples in nbtree by adding logic to consider extra index tuples (that are not LP_DEAD-marked) for deletion in passing. This increases the number of index tuples deleted significantly in many cases. The LP_DEAD deletion process (which is now called "simple deletion" to clearly distinguish it from bottom-up deletion) won't usually need to visit any extra table blocks to check these extra tuples. We have to visit the same table blocks anyway to generate a latestRemovedXid value (at least in the common case where the index deletion operation's WAL record needs such a value). Testing has shown that the "extra tuples" simple deletion enhancement increases the number of index tuples deleted with almost any workload that has LP_DEAD bits set in leaf pages. That is, it almost never fails to delete at least a few extra index tuples. It helps most of all in cases that happen to naturally have a lot of delete-safe tuples. It's not uncommon for an individual deletion operation to end up deleting an order of magnitude more index tuples compared to the old naive approach (e.g., custom instrumentation of the patch shows that this happens fairly often when the regression tests are run). Add a further enhancement that augments simple deletion and bottom-up deletion in indexes that make use of deduplication: Teach nbtree's _bt_delitems_delete() function to support granular TID deletion in posting list tuples. It is now possible to delete individual TIDs from posting list tuples provided the TIDs have a tableam block number of a table block that gets visited as part of the deletion process (visiting the table block can be triggered directly or indirectly). Setting the LP_DEAD bit of a posting list tuple is still an all-or-nothing thing, but that matters much less now that deletion only needs to start out with the right _general_ idea about which index tuples are deletable. Bump XLOG_PAGE_MAGIC because xl_btree_delete changed. No bump in BTREE_VERSION, since there are no changes to the on-disk representation of nbtree indexes. Indexes built on PostgreSQL 12 or PostgreSQL 13 will automatically benefit from bottom-up index deletion (i.e. no reindexing required) following a pg_upgrade. The enhancement to simple deletion is available with all B-Tree indexes following a pg_upgrade, no matter what PostgreSQL version the user upgrades from. Author: Peter Geoghegan <pg@bowt.ie> Reviewed-By: Heikki Linnakangas <hlinnaka@iki.fi> Reviewed-By: Victor Yegorov <vyegorov@gmail.com> Discussion: https://postgr.es/m/CAH2-Wzm+maE3apHB8NOtmM=p-DO65j2V5GzAWCOEEuy3JZgb2g@mail.gmail.com
2021-01-13 18:21:32 +01:00
return rel->rd_tableam->index_delete_tuples(rel, delstate);
Compute XID horizon for page level index vacuum on primary. Previously the xid horizon was only computed during WAL replay. That had two major problems: 1) It relied on knowing what the table pointed to looks like. That was easy enough before the introducing of tableam (we knew it had to be heap, although some trickery around logging the heap relfilenodes was required). But to properly handle table AMs we need per-database catalog access to look up the AM handler, which recovery doesn't allow. 2) Not knowing the xid horizon also makes it hard to support logical decoding on standbys. When on a catalog table, we need to be able to conflict with slots that have an xid horizon that's too old. But computing the horizon by visiting the heap only works once consistency is reached, but we always need to be able to detect conflicts. There's also a secondary problem, in that the current method performs redundant work on every standby. But that's counterbalanced by potentially computing the value when not necessary (either because there's no standby, or because there's no connected backends). Solve 1) and 2) by moving computation of the xid horizon to the primary and by involving tableam in the computation of the horizon. To address the potentially increased overhead, increase the efficiency of the xid horizon computation for heap by sorting the tids, and eliminating redundant buffer accesses. When prefetching is available, additionally perform prefetching of buffers. As this is more of a maintenance task, rather than something routinely done in every read only query, we add an arbitrary 10 to the effective concurrency - thereby using IO concurrency, when not globally enabled. That's possibly not the perfect formula, but seems good enough for now. Bumps WAL format, as latestRemovedXid is now part of the records, and the heap's relfilenode isn't anymore. Author: Andres Freund, Amit Khandekar, Robert Haas Reviewed-By: Robert Haas Discussion: https://postgr.es/m/20181212204154.nsxf3gzqv3gesl32@alap3.anarazel.de https://postgr.es/m/20181214014235.dal5ogljs3bmlq44@alap3.anarazel.de https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-26 22:41:46 +01:00
}
tableam: Add and use scan APIs. Too allow table accesses to be not directly dependent on heap, several new abstractions are needed. Specifically: 1) Heap scans need to be generalized into table scans. Do this by introducing TableScanDesc, which will be the "base class" for individual AMs. This contains the AM independent fields from HeapScanDesc. The previous heap_{beginscan,rescan,endscan} et al. have been replaced with a table_ version. There's no direct replacement for heap_getnext(), as that returned a HeapTuple, which is undesirable for a other AMs. Instead there's table_scan_getnextslot(). But note that heap_getnext() lives on, it's still used widely to access catalog tables. This is achieved by new scan_begin, scan_end, scan_rescan, scan_getnextslot callbacks. 2) The portion of parallel scans that's shared between backends need to be able to do so without the user doing per-AM work. To achieve that new parallelscan_{estimate, initialize, reinitialize} callbacks are introduced, which operate on a new ParallelTableScanDesc, which again can be subclassed by AMs. As it is likely that several AMs are going to be block oriented, block oriented callbacks that can be shared between such AMs are provided and used by heap. table_block_parallelscan_{estimate, intiialize, reinitialize} as callbacks, and table_block_parallelscan_{nextpage, init} for use in AMs. These operate on a ParallelBlockTableScanDesc. 3) Index scans need to be able to access tables to return a tuple, and there needs to be state across individual accesses to the heap to store state like buffers. That's now handled by introducing a sort-of-scan IndexFetchTable, which again is intended to be subclassed by individual AMs (for heap IndexFetchHeap). The relevant callbacks for an AM are index_fetch_{end, begin, reset} to create the necessary state, and index_fetch_tuple to retrieve an indexed tuple. Note that index_fetch_tuple implementations need to be smarter than just blindly fetching the tuples for AMs that have optimizations similar to heap's HOT - the currently alive tuple in the update chain needs to be fetched if appropriate. Similar to table_scan_getnextslot(), it's undesirable to continue to return HeapTuples. Thus index_fetch_heap (might want to rename that later) now accepts a slot as an argument. Core code doesn't have a lot of call sites performing index scans without going through the systable_* API (in contrast to loads of heap_getnext calls and working directly with HeapTuples). Index scans now store the result of a search in IndexScanDesc->xs_heaptid, rather than xs_ctup->t_self. As the target is not generally a HeapTuple anymore that seems cleaner. To be able to sensible adapt code to use the above, two further callbacks have been introduced: a) slot_callbacks returns a TupleTableSlotOps* suitable for creating slots capable of holding a tuple of the AMs type. table_slot_callbacks() and table_slot_create() are based upon that, but have additional logic to deal with views, foreign tables, etc. While this change could have been done separately, nearly all the call sites that needed to be adapted for the rest of this commit also would have been needed to be adapted for table_slot_callbacks(), making separation not worthwhile. b) tuple_satisfies_snapshot checks whether the tuple in a slot is currently visible according to a snapshot. That's required as a few places now don't have a buffer + HeapTuple around, but a slot (which in heap's case internally has that information). Additionally a few infrastructure changes were needed: I) SysScanDesc, as used by systable_{beginscan, getnext} et al. now internally uses a slot to keep track of tuples. While systable_getnext() still returns HeapTuples, and will so for the foreseeable future, the index API (see 1) above) now only deals with slots. The remainder, and largest part, of this commit is then adjusting all scans in postgres to use the new APIs. Author: Andres Freund, Haribabu Kommi, Alvaro Herrera Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-11 20:46:41 +01:00
tableam: Add tuple_{insert, delete, update, lock} and use. This adds new, required, table AM callbacks for insert/delete/update and lock_tuple. To be able to reasonably use those, the EvalPlanQual mechanism had to be adapted, moving more logic into the AM. Previously both delete/update/lock call-sites and the EPQ mechanism had to have awareness of the specific tuple format to be able to fetch the latest version of a tuple. Obviously that needs to be abstracted away. To do so, move the logic that find the latest row version into the AM. lock_tuple has a new flag argument, TUPLE_LOCK_FLAG_FIND_LAST_VERSION, that forces it to lock the last version, rather than the current one. It'd have been possible to do so via a separate callback as well, but finding the last version usually also necessitates locking the newest version, making it sensible to combine the two. This replaces the previous use of EvalPlanQualFetch(). Additionally HeapTupleUpdated, which previously signaled either a concurrent update or delete, is now split into two, to avoid callers needing AM specific knowledge to differentiate. The move of finding the latest row version into tuple_lock means that encountering a row concurrently moved into another partition will now raise an error about "tuple to be locked" rather than "tuple to be updated/deleted" - which is accurate, as that always happens when locking rows. While possible slightly less helpful for users, it seems like an acceptable trade-off. As part of this commit HTSU_Result has been renamed to TM_Result, and its members been expanded to differentiated between updating and deleting. HeapUpdateFailureData has been renamed to TM_FailureData. The interface to speculative insertion is changed so nodeModifyTable.c does not have to set the speculative token itself anymore. Instead there's a version of tuple_insert, tuple_insert_speculative, that performs the speculative insertion (without requiring a flag to signal that fact), and the speculative insertion is either made permanent with table_complete_speculative(succeeded = true) or aborted with succeeded = false). Note that multi_insert is not yet routed through tableam, nor is COPY. Changing multi_insert requires changes to copy.c that are large enough to better be done separately. Similarly, although simpler, CREATE TABLE AS and CREATE MATERIALIZED VIEW are also only going to be adjusted in a later commit. Author: Andres Freund and Haribabu Kommi Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20190313003903.nwvrxi7rw3ywhdel@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-24 03:55:57 +01:00
/* ----------------------------------------------------------------------------
* Functions for manipulations of physical tuples.
* ----------------------------------------------------------------------------
*/
/*
* Insert a tuple from a slot into table AM routine.
*
tableam: comment improvements. Author: Brad DeJong Discussion: https://postgr.es/m/CAJnrtnxDYOQFsDfWz2iri0T_fFL2ZbbzgCOE=4yaMcszgcsf4A@mail.gmail.com Backpatch: 12-
2019-07-18 04:39:54 +02:00
* The options bitmask allows the caller to specify options that may change the
* behaviour of the AM. The AM will ignore options that it does not support.
tableam: Add tuple_{insert, delete, update, lock} and use. This adds new, required, table AM callbacks for insert/delete/update and lock_tuple. To be able to reasonably use those, the EvalPlanQual mechanism had to be adapted, moving more logic into the AM. Previously both delete/update/lock call-sites and the EPQ mechanism had to have awareness of the specific tuple format to be able to fetch the latest version of a tuple. Obviously that needs to be abstracted away. To do so, move the logic that find the latest row version into the AM. lock_tuple has a new flag argument, TUPLE_LOCK_FLAG_FIND_LAST_VERSION, that forces it to lock the last version, rather than the current one. It'd have been possible to do so via a separate callback as well, but finding the last version usually also necessitates locking the newest version, making it sensible to combine the two. This replaces the previous use of EvalPlanQualFetch(). Additionally HeapTupleUpdated, which previously signaled either a concurrent update or delete, is now split into two, to avoid callers needing AM specific knowledge to differentiate. The move of finding the latest row version into tuple_lock means that encountering a row concurrently moved into another partition will now raise an error about "tuple to be locked" rather than "tuple to be updated/deleted" - which is accurate, as that always happens when locking rows. While possible slightly less helpful for users, it seems like an acceptable trade-off. As part of this commit HTSU_Result has been renamed to TM_Result, and its members been expanded to differentiated between updating and deleting. HeapUpdateFailureData has been renamed to TM_FailureData. The interface to speculative insertion is changed so nodeModifyTable.c does not have to set the speculative token itself anymore. Instead there's a version of tuple_insert, tuple_insert_speculative, that performs the speculative insertion (without requiring a flag to signal that fact), and the speculative insertion is either made permanent with table_complete_speculative(succeeded = true) or aborted with succeeded = false). Note that multi_insert is not yet routed through tableam, nor is COPY. Changing multi_insert requires changes to copy.c that are large enough to better be done separately. Similarly, although simpler, CREATE TABLE AS and CREATE MATERIALIZED VIEW are also only going to be adjusted in a later commit. Author: Andres Freund and Haribabu Kommi Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20190313003903.nwvrxi7rw3ywhdel@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-24 03:55:57 +01:00
*
* If the TABLE_INSERT_SKIP_FSM option is specified, AMs are free to not reuse
* free space in the relation. This can save some cycles when we know the
tableam: comment improvements. Author: Brad DeJong Discussion: https://postgr.es/m/CAJnrtnxDYOQFsDfWz2iri0T_fFL2ZbbzgCOE=4yaMcszgcsf4A@mail.gmail.com Backpatch: 12-
2019-07-18 04:39:54 +02:00
* relation is new and doesn't contain useful amounts of free space.
* TABLE_INSERT_SKIP_FSM is commonly passed directly to
* RelationGetBufferForTuple. See that method for more information.
tableam: Add tuple_{insert, delete, update, lock} and use. This adds new, required, table AM callbacks for insert/delete/update and lock_tuple. To be able to reasonably use those, the EvalPlanQual mechanism had to be adapted, moving more logic into the AM. Previously both delete/update/lock call-sites and the EPQ mechanism had to have awareness of the specific tuple format to be able to fetch the latest version of a tuple. Obviously that needs to be abstracted away. To do so, move the logic that find the latest row version into the AM. lock_tuple has a new flag argument, TUPLE_LOCK_FLAG_FIND_LAST_VERSION, that forces it to lock the last version, rather than the current one. It'd have been possible to do so via a separate callback as well, but finding the last version usually also necessitates locking the newest version, making it sensible to combine the two. This replaces the previous use of EvalPlanQualFetch(). Additionally HeapTupleUpdated, which previously signaled either a concurrent update or delete, is now split into two, to avoid callers needing AM specific knowledge to differentiate. The move of finding the latest row version into tuple_lock means that encountering a row concurrently moved into another partition will now raise an error about "tuple to be locked" rather than "tuple to be updated/deleted" - which is accurate, as that always happens when locking rows. While possible slightly less helpful for users, it seems like an acceptable trade-off. As part of this commit HTSU_Result has been renamed to TM_Result, and its members been expanded to differentiated between updating and deleting. HeapUpdateFailureData has been renamed to TM_FailureData. The interface to speculative insertion is changed so nodeModifyTable.c does not have to set the speculative token itself anymore. Instead there's a version of tuple_insert, tuple_insert_speculative, that performs the speculative insertion (without requiring a flag to signal that fact), and the speculative insertion is either made permanent with table_complete_speculative(succeeded = true) or aborted with succeeded = false). Note that multi_insert is not yet routed through tableam, nor is COPY. Changing multi_insert requires changes to copy.c that are large enough to better be done separately. Similarly, although simpler, CREATE TABLE AS and CREATE MATERIALIZED VIEW are also only going to be adjusted in a later commit. Author: Andres Freund and Haribabu Kommi Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20190313003903.nwvrxi7rw3ywhdel@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-24 03:55:57 +01:00
*
* TABLE_INSERT_FROZEN should only be specified for inserts into
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
* relation storage created during the current subtransaction and when
tableam: Add tuple_{insert, delete, update, lock} and use. This adds new, required, table AM callbacks for insert/delete/update and lock_tuple. To be able to reasonably use those, the EvalPlanQual mechanism had to be adapted, moving more logic into the AM. Previously both delete/update/lock call-sites and the EPQ mechanism had to have awareness of the specific tuple format to be able to fetch the latest version of a tuple. Obviously that needs to be abstracted away. To do so, move the logic that find the latest row version into the AM. lock_tuple has a new flag argument, TUPLE_LOCK_FLAG_FIND_LAST_VERSION, that forces it to lock the last version, rather than the current one. It'd have been possible to do so via a separate callback as well, but finding the last version usually also necessitates locking the newest version, making it sensible to combine the two. This replaces the previous use of EvalPlanQualFetch(). Additionally HeapTupleUpdated, which previously signaled either a concurrent update or delete, is now split into two, to avoid callers needing AM specific knowledge to differentiate. The move of finding the latest row version into tuple_lock means that encountering a row concurrently moved into another partition will now raise an error about "tuple to be locked" rather than "tuple to be updated/deleted" - which is accurate, as that always happens when locking rows. While possible slightly less helpful for users, it seems like an acceptable trade-off. As part of this commit HTSU_Result has been renamed to TM_Result, and its members been expanded to differentiated between updating and deleting. HeapUpdateFailureData has been renamed to TM_FailureData. The interface to speculative insertion is changed so nodeModifyTable.c does not have to set the speculative token itself anymore. Instead there's a version of tuple_insert, tuple_insert_speculative, that performs the speculative insertion (without requiring a flag to signal that fact), and the speculative insertion is either made permanent with table_complete_speculative(succeeded = true) or aborted with succeeded = false). Note that multi_insert is not yet routed through tableam, nor is COPY. Changing multi_insert requires changes to copy.c that are large enough to better be done separately. Similarly, although simpler, CREATE TABLE AS and CREATE MATERIALIZED VIEW are also only going to be adjusted in a later commit. Author: Andres Freund and Haribabu Kommi Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20190313003903.nwvrxi7rw3ywhdel@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-24 03:55:57 +01:00
* there are no prior snapshots or pre-existing portals open.
* This causes rows to be frozen, which is an MVCC violation and
* requires explicit options chosen by user.
*
* TABLE_INSERT_NO_LOGICAL force-disables the emitting of logical decoding
* information for the tuple. This should solely be used during table rewrites
* where RelationIsLogicallyLogged(relation) is not yet accurate for the new
* relation.
*
* Note that most of these options will be applied when inserting into the
tableam: comment and formatting fixes. Author: Heikki Linnakangas Discussion: https://postgr.es/m/9a7fb9cc-2419-5db7-8840-ddc10c93f122@iki.fi
2019-04-08 19:25:16 +02:00
* heap's TOAST table, too, if the tuple requires any out-of-line data.
tableam: Add tuple_{insert, delete, update, lock} and use. This adds new, required, table AM callbacks for insert/delete/update and lock_tuple. To be able to reasonably use those, the EvalPlanQual mechanism had to be adapted, moving more logic into the AM. Previously both delete/update/lock call-sites and the EPQ mechanism had to have awareness of the specific tuple format to be able to fetch the latest version of a tuple. Obviously that needs to be abstracted away. To do so, move the logic that find the latest row version into the AM. lock_tuple has a new flag argument, TUPLE_LOCK_FLAG_FIND_LAST_VERSION, that forces it to lock the last version, rather than the current one. It'd have been possible to do so via a separate callback as well, but finding the last version usually also necessitates locking the newest version, making it sensible to combine the two. This replaces the previous use of EvalPlanQualFetch(). Additionally HeapTupleUpdated, which previously signaled either a concurrent update or delete, is now split into two, to avoid callers needing AM specific knowledge to differentiate. The move of finding the latest row version into tuple_lock means that encountering a row concurrently moved into another partition will now raise an error about "tuple to be locked" rather than "tuple to be updated/deleted" - which is accurate, as that always happens when locking rows. While possible slightly less helpful for users, it seems like an acceptable trade-off. As part of this commit HTSU_Result has been renamed to TM_Result, and its members been expanded to differentiated between updating and deleting. HeapUpdateFailureData has been renamed to TM_FailureData. The interface to speculative insertion is changed so nodeModifyTable.c does not have to set the speculative token itself anymore. Instead there's a version of tuple_insert, tuple_insert_speculative, that performs the speculative insertion (without requiring a flag to signal that fact), and the speculative insertion is either made permanent with table_complete_speculative(succeeded = true) or aborted with succeeded = false). Note that multi_insert is not yet routed through tableam, nor is COPY. Changing multi_insert requires changes to copy.c that are large enough to better be done separately. Similarly, although simpler, CREATE TABLE AS and CREATE MATERIALIZED VIEW are also only going to be adjusted in a later commit. Author: Andres Freund and Haribabu Kommi Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20190313003903.nwvrxi7rw3ywhdel@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-24 03:55:57 +01:00
*
* The BulkInsertState object (if any; bistate can be NULL for default
tableam: Add table_finish_bulk_insert(). This replaces the previous calls of heap_sync() in places using bulk-insert. By passing in the flags used for bulk-insert the AM can decide (first at insert time and then during the finish call) which of the optimizations apply to it, and what operations are necessary to finish a bulk insert operation. Also change HEAP_INSERT_* flags to TABLE_INSERT, and rename hi_options to ti_options. These changes are made even in copy.c, which hasn't yet been converted to tableam. There's no harm in doing so. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-04-01 23:41:42 +02:00
* behavior) is also just passed through to RelationGetBufferForTuple. If
* `bistate` is provided, table_finish_bulk_insert() needs to be called.
tableam: Add tuple_{insert, delete, update, lock} and use. This adds new, required, table AM callbacks for insert/delete/update and lock_tuple. To be able to reasonably use those, the EvalPlanQual mechanism had to be adapted, moving more logic into the AM. Previously both delete/update/lock call-sites and the EPQ mechanism had to have awareness of the specific tuple format to be able to fetch the latest version of a tuple. Obviously that needs to be abstracted away. To do so, move the logic that find the latest row version into the AM. lock_tuple has a new flag argument, TUPLE_LOCK_FLAG_FIND_LAST_VERSION, that forces it to lock the last version, rather than the current one. It'd have been possible to do so via a separate callback as well, but finding the last version usually also necessitates locking the newest version, making it sensible to combine the two. This replaces the previous use of EvalPlanQualFetch(). Additionally HeapTupleUpdated, which previously signaled either a concurrent update or delete, is now split into two, to avoid callers needing AM specific knowledge to differentiate. The move of finding the latest row version into tuple_lock means that encountering a row concurrently moved into another partition will now raise an error about "tuple to be locked" rather than "tuple to be updated/deleted" - which is accurate, as that always happens when locking rows. While possible slightly less helpful for users, it seems like an acceptable trade-off. As part of this commit HTSU_Result has been renamed to TM_Result, and its members been expanded to differentiated between updating and deleting. HeapUpdateFailureData has been renamed to TM_FailureData. The interface to speculative insertion is changed so nodeModifyTable.c does not have to set the speculative token itself anymore. Instead there's a version of tuple_insert, tuple_insert_speculative, that performs the speculative insertion (without requiring a flag to signal that fact), and the speculative insertion is either made permanent with table_complete_speculative(succeeded = true) or aborted with succeeded = false). Note that multi_insert is not yet routed through tableam, nor is COPY. Changing multi_insert requires changes to copy.c that are large enough to better be done separately. Similarly, although simpler, CREATE TABLE AS and CREATE MATERIALIZED VIEW are also only going to be adjusted in a later commit. Author: Andres Freund and Haribabu Kommi Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20190313003903.nwvrxi7rw3ywhdel@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-24 03:55:57 +01:00
*
* On return the slot's tts_tid and tts_tableOid are updated to reflect the
* insertion. But note that any toasting of fields within the slot is NOT
* reflected in the slots contents.
*/
static inline void
tableam: Rename wrapper functions to match callback names. Some of the wrapper functions didn't match the callback names. Many of them due to staying "consistent" with historic naming of the wrapped functionality. We decided that for most cases it's more important to be for tableam to be consistent going forward, than with the past. The one exception is beginscan/endscan/... because it'd have looked odd to have systable_beginscan/endscan/... with a different naming scheme, and changing the systable_* APIs would have caused way too much churn (including breaking a lot of external users). Author: Ashwin Agrawal, with some small additions by Andres Freund Reviewed-By: Andres Freund Discussion: https://postgr.es/m/CALfoeiugyrXZfX7n0ORCa4L-m834dzmaE8eFdbNR6PMpetU4Ww@mail.gmail.com
2019-05-24 01:25:48 +02:00
table_tuple_insert(Relation rel, TupleTableSlot *slot, CommandId cid,
int options, struct BulkInsertStateData *bistate)
tableam: Add tuple_{insert, delete, update, lock} and use. This adds new, required, table AM callbacks for insert/delete/update and lock_tuple. To be able to reasonably use those, the EvalPlanQual mechanism had to be adapted, moving more logic into the AM. Previously both delete/update/lock call-sites and the EPQ mechanism had to have awareness of the specific tuple format to be able to fetch the latest version of a tuple. Obviously that needs to be abstracted away. To do so, move the logic that find the latest row version into the AM. lock_tuple has a new flag argument, TUPLE_LOCK_FLAG_FIND_LAST_VERSION, that forces it to lock the last version, rather than the current one. It'd have been possible to do so via a separate callback as well, but finding the last version usually also necessitates locking the newest version, making it sensible to combine the two. This replaces the previous use of EvalPlanQualFetch(). Additionally HeapTupleUpdated, which previously signaled either a concurrent update or delete, is now split into two, to avoid callers needing AM specific knowledge to differentiate. The move of finding the latest row version into tuple_lock means that encountering a row concurrently moved into another partition will now raise an error about "tuple to be locked" rather than "tuple to be updated/deleted" - which is accurate, as that always happens when locking rows. While possible slightly less helpful for users, it seems like an acceptable trade-off. As part of this commit HTSU_Result has been renamed to TM_Result, and its members been expanded to differentiated between updating and deleting. HeapUpdateFailureData has been renamed to TM_FailureData. The interface to speculative insertion is changed so nodeModifyTable.c does not have to set the speculative token itself anymore. Instead there's a version of tuple_insert, tuple_insert_speculative, that performs the speculative insertion (without requiring a flag to signal that fact), and the speculative insertion is either made permanent with table_complete_speculative(succeeded = true) or aborted with succeeded = false). Note that multi_insert is not yet routed through tableam, nor is COPY. Changing multi_insert requires changes to copy.c that are large enough to better be done separately. Similarly, although simpler, CREATE TABLE AS and CREATE MATERIALIZED VIEW are also only going to be adjusted in a later commit. Author: Andres Freund and Haribabu Kommi Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20190313003903.nwvrxi7rw3ywhdel@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-24 03:55:57 +01:00
{
rel->rd_tableam->tuple_insert(rel, slot, cid, options,
bistate);
}
/*
* Perform a "speculative insertion". These can be backed out afterwards
* without aborting the whole transaction. Other sessions can wait for the
* speculative insertion to be confirmed, turning it into a regular tuple, or
* aborted, as if it never existed. Speculatively inserted tuples behave as
* "value locks" of short duration, used to implement INSERT .. ON CONFLICT.
*
* A transaction having performed a speculative insertion has to either abort,
* or finish the speculative insertion with
tableam: Rename wrapper functions to match callback names. Some of the wrapper functions didn't match the callback names. Many of them due to staying "consistent" with historic naming of the wrapped functionality. We decided that for most cases it's more important to be for tableam to be consistent going forward, than with the past. The one exception is beginscan/endscan/... because it'd have looked odd to have systable_beginscan/endscan/... with a different naming scheme, and changing the systable_* APIs would have caused way too much churn (including breaking a lot of external users). Author: Ashwin Agrawal, with some small additions by Andres Freund Reviewed-By: Andres Freund Discussion: https://postgr.es/m/CALfoeiugyrXZfX7n0ORCa4L-m834dzmaE8eFdbNR6PMpetU4Ww@mail.gmail.com
2019-05-24 01:25:48 +02:00
* table_tuple_complete_speculative(succeeded = ...).
tableam: Add tuple_{insert, delete, update, lock} and use. This adds new, required, table AM callbacks for insert/delete/update and lock_tuple. To be able to reasonably use those, the EvalPlanQual mechanism had to be adapted, moving more logic into the AM. Previously both delete/update/lock call-sites and the EPQ mechanism had to have awareness of the specific tuple format to be able to fetch the latest version of a tuple. Obviously that needs to be abstracted away. To do so, move the logic that find the latest row version into the AM. lock_tuple has a new flag argument, TUPLE_LOCK_FLAG_FIND_LAST_VERSION, that forces it to lock the last version, rather than the current one. It'd have been possible to do so via a separate callback as well, but finding the last version usually also necessitates locking the newest version, making it sensible to combine the two. This replaces the previous use of EvalPlanQualFetch(). Additionally HeapTupleUpdated, which previously signaled either a concurrent update or delete, is now split into two, to avoid callers needing AM specific knowledge to differentiate. The move of finding the latest row version into tuple_lock means that encountering a row concurrently moved into another partition will now raise an error about "tuple to be locked" rather than "tuple to be updated/deleted" - which is accurate, as that always happens when locking rows. While possible slightly less helpful for users, it seems like an acceptable trade-off. As part of this commit HTSU_Result has been renamed to TM_Result, and its members been expanded to differentiated between updating and deleting. HeapUpdateFailureData has been renamed to TM_FailureData. The interface to speculative insertion is changed so nodeModifyTable.c does not have to set the speculative token itself anymore. Instead there's a version of tuple_insert, tuple_insert_speculative, that performs the speculative insertion (without requiring a flag to signal that fact), and the speculative insertion is either made permanent with table_complete_speculative(succeeded = true) or aborted with succeeded = false). Note that multi_insert is not yet routed through tableam, nor is COPY. Changing multi_insert requires changes to copy.c that are large enough to better be done separately. Similarly, although simpler, CREATE TABLE AS and CREATE MATERIALIZED VIEW are also only going to be adjusted in a later commit. Author: Andres Freund and Haribabu Kommi Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20190313003903.nwvrxi7rw3ywhdel@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-24 03:55:57 +01:00
*/
static inline void
tableam: Rename wrapper functions to match callback names. Some of the wrapper functions didn't match the callback names. Many of them due to staying "consistent" with historic naming of the wrapped functionality. We decided that for most cases it's more important to be for tableam to be consistent going forward, than with the past. The one exception is beginscan/endscan/... because it'd have looked odd to have systable_beginscan/endscan/... with a different naming scheme, and changing the systable_* APIs would have caused way too much churn (including breaking a lot of external users). Author: Ashwin Agrawal, with some small additions by Andres Freund Reviewed-By: Andres Freund Discussion: https://postgr.es/m/CALfoeiugyrXZfX7n0ORCa4L-m834dzmaE8eFdbNR6PMpetU4Ww@mail.gmail.com
2019-05-24 01:25:48 +02:00
table_tuple_insert_speculative(Relation rel, TupleTableSlot *slot,
CommandId cid, int options,
struct BulkInsertStateData *bistate,
uint32 specToken)
tableam: Add tuple_{insert, delete, update, lock} and use. This adds new, required, table AM callbacks for insert/delete/update and lock_tuple. To be able to reasonably use those, the EvalPlanQual mechanism had to be adapted, moving more logic into the AM. Previously both delete/update/lock call-sites and the EPQ mechanism had to have awareness of the specific tuple format to be able to fetch the latest version of a tuple. Obviously that needs to be abstracted away. To do so, move the logic that find the latest row version into the AM. lock_tuple has a new flag argument, TUPLE_LOCK_FLAG_FIND_LAST_VERSION, that forces it to lock the last version, rather than the current one. It'd have been possible to do so via a separate callback as well, but finding the last version usually also necessitates locking the newest version, making it sensible to combine the two. This replaces the previous use of EvalPlanQualFetch(). Additionally HeapTupleUpdated, which previously signaled either a concurrent update or delete, is now split into two, to avoid callers needing AM specific knowledge to differentiate. The move of finding the latest row version into tuple_lock means that encountering a row concurrently moved into another partition will now raise an error about "tuple to be locked" rather than "tuple to be updated/deleted" - which is accurate, as that always happens when locking rows. While possible slightly less helpful for users, it seems like an acceptable trade-off. As part of this commit HTSU_Result has been renamed to TM_Result, and its members been expanded to differentiated between updating and deleting. HeapUpdateFailureData has been renamed to TM_FailureData. The interface to speculative insertion is changed so nodeModifyTable.c does not have to set the speculative token itself anymore. Instead there's a version of tuple_insert, tuple_insert_speculative, that performs the speculative insertion (without requiring a flag to signal that fact), and the speculative insertion is either made permanent with table_complete_speculative(succeeded = true) or aborted with succeeded = false). Note that multi_insert is not yet routed through tableam, nor is COPY. Changing multi_insert requires changes to copy.c that are large enough to better be done separately. Similarly, although simpler, CREATE TABLE AS and CREATE MATERIALIZED VIEW are also only going to be adjusted in a later commit. Author: Andres Freund and Haribabu Kommi Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20190313003903.nwvrxi7rw3ywhdel@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-24 03:55:57 +01:00
{
rel->rd_tableam->tuple_insert_speculative(rel, slot, cid, options,
bistate, specToken);
}
/*
* Complete "speculative insertion" started in the same transaction. If
* succeeded is true, the tuple is fully inserted, if false, it's removed.
*/
static inline void
tableam: Rename wrapper functions to match callback names. Some of the wrapper functions didn't match the callback names. Many of them due to staying "consistent" with historic naming of the wrapped functionality. We decided that for most cases it's more important to be for tableam to be consistent going forward, than with the past. The one exception is beginscan/endscan/... because it'd have looked odd to have systable_beginscan/endscan/... with a different naming scheme, and changing the systable_* APIs would have caused way too much churn (including breaking a lot of external users). Author: Ashwin Agrawal, with some small additions by Andres Freund Reviewed-By: Andres Freund Discussion: https://postgr.es/m/CALfoeiugyrXZfX7n0ORCa4L-m834dzmaE8eFdbNR6PMpetU4Ww@mail.gmail.com
2019-05-24 01:25:48 +02:00
table_tuple_complete_speculative(Relation rel, TupleTableSlot *slot,
uint32 specToken, bool succeeded)
tableam: Add tuple_{insert, delete, update, lock} and use. This adds new, required, table AM callbacks for insert/delete/update and lock_tuple. To be able to reasonably use those, the EvalPlanQual mechanism had to be adapted, moving more logic into the AM. Previously both delete/update/lock call-sites and the EPQ mechanism had to have awareness of the specific tuple format to be able to fetch the latest version of a tuple. Obviously that needs to be abstracted away. To do so, move the logic that find the latest row version into the AM. lock_tuple has a new flag argument, TUPLE_LOCK_FLAG_FIND_LAST_VERSION, that forces it to lock the last version, rather than the current one. It'd have been possible to do so via a separate callback as well, but finding the last version usually also necessitates locking the newest version, making it sensible to combine the two. This replaces the previous use of EvalPlanQualFetch(). Additionally HeapTupleUpdated, which previously signaled either a concurrent update or delete, is now split into two, to avoid callers needing AM specific knowledge to differentiate. The move of finding the latest row version into tuple_lock means that encountering a row concurrently moved into another partition will now raise an error about "tuple to be locked" rather than "tuple to be updated/deleted" - which is accurate, as that always happens when locking rows. While possible slightly less helpful for users, it seems like an acceptable trade-off. As part of this commit HTSU_Result has been renamed to TM_Result, and its members been expanded to differentiated between updating and deleting. HeapUpdateFailureData has been renamed to TM_FailureData. The interface to speculative insertion is changed so nodeModifyTable.c does not have to set the speculative token itself anymore. Instead there's a version of tuple_insert, tuple_insert_speculative, that performs the speculative insertion (without requiring a flag to signal that fact), and the speculative insertion is either made permanent with table_complete_speculative(succeeded = true) or aborted with succeeded = false). Note that multi_insert is not yet routed through tableam, nor is COPY. Changing multi_insert requires changes to copy.c that are large enough to better be done separately. Similarly, although simpler, CREATE TABLE AS and CREATE MATERIALIZED VIEW are also only going to be adjusted in a later commit. Author: Andres Freund and Haribabu Kommi Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20190313003903.nwvrxi7rw3ywhdel@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-24 03:55:57 +01:00
{
Remove spurious return. Per buildfarm member anole. Author: Andres Freund
2019-03-24 05:09:39 +01:00
rel->rd_tableam->tuple_complete_speculative(rel, slot, specToken,
succeeded);
tableam: Add tuple_{insert, delete, update, lock} and use. This adds new, required, table AM callbacks for insert/delete/update and lock_tuple. To be able to reasonably use those, the EvalPlanQual mechanism had to be adapted, moving more logic into the AM. Previously both delete/update/lock call-sites and the EPQ mechanism had to have awareness of the specific tuple format to be able to fetch the latest version of a tuple. Obviously that needs to be abstracted away. To do so, move the logic that find the latest row version into the AM. lock_tuple has a new flag argument, TUPLE_LOCK_FLAG_FIND_LAST_VERSION, that forces it to lock the last version, rather than the current one. It'd have been possible to do so via a separate callback as well, but finding the last version usually also necessitates locking the newest version, making it sensible to combine the two. This replaces the previous use of EvalPlanQualFetch(). Additionally HeapTupleUpdated, which previously signaled either a concurrent update or delete, is now split into two, to avoid callers needing AM specific knowledge to differentiate. The move of finding the latest row version into tuple_lock means that encountering a row concurrently moved into another partition will now raise an error about "tuple to be locked" rather than "tuple to be updated/deleted" - which is accurate, as that always happens when locking rows. While possible slightly less helpful for users, it seems like an acceptable trade-off. As part of this commit HTSU_Result has been renamed to TM_Result, and its members been expanded to differentiated between updating and deleting. HeapUpdateFailureData has been renamed to TM_FailureData. The interface to speculative insertion is changed so nodeModifyTable.c does not have to set the speculative token itself anymore. Instead there's a version of tuple_insert, tuple_insert_speculative, that performs the speculative insertion (without requiring a flag to signal that fact), and the speculative insertion is either made permanent with table_complete_speculative(succeeded = true) or aborted with succeeded = false). Note that multi_insert is not yet routed through tableam, nor is COPY. Changing multi_insert requires changes to copy.c that are large enough to better be done separately. Similarly, although simpler, CREATE TABLE AS and CREATE MATERIALIZED VIEW are also only going to be adjusted in a later commit. Author: Andres Freund and Haribabu Kommi Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20190313003903.nwvrxi7rw3ywhdel@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-24 03:55:57 +01:00
}
tableam: Add table_multi_insert() and revamp/speed-up COPY FROM buffering. This adds table_multi_insert(), and converts COPY FROM, the only user of heap_multi_insert, to it. A simple conversion of COPY FROM use slots would have yielded a slowdown when inserting into a partitioned table for some workloads. Different partitions might need different slots (both slot types and their descriptors), and dropping / creating slots when there's constant partition changes is measurable. Thus instead revamp the COPY FROM buffering for partitioned tables to allow to buffer inserts into multiple tables, flushing only when limits are reached across all partition buffers. By only dropping slots when there've been inserts into too many different partitions, the aforementioned overhead is gone. By allowing larger batches, even when there are frequent partition changes, we actuall speed such cases up significantly. By using slots COPY of very narrow rows into unlogged / temporary might slow down very slightly (due to the indirect function calls). Author: David Rowley, Andres Freund, Haribabu Kommi Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20190327054923.t3epfuewxfqdt22e@alap3.anarazel.de
2019-04-05 00:47:19 +02:00
/*
tableam: comment and formatting fixes. Author: Heikki Linnakangas Discussion: https://postgr.es/m/9a7fb9cc-2419-5db7-8840-ddc10c93f122@iki.fi
2019-04-08 19:25:16 +02:00
* Insert multiple tuples into a table.
tableam: Add table_multi_insert() and revamp/speed-up COPY FROM buffering. This adds table_multi_insert(), and converts COPY FROM, the only user of heap_multi_insert, to it. A simple conversion of COPY FROM use slots would have yielded a slowdown when inserting into a partitioned table for some workloads. Different partitions might need different slots (both slot types and their descriptors), and dropping / creating slots when there's constant partition changes is measurable. Thus instead revamp the COPY FROM buffering for partitioned tables to allow to buffer inserts into multiple tables, flushing only when limits are reached across all partition buffers. By only dropping slots when there've been inserts into too many different partitions, the aforementioned overhead is gone. By allowing larger batches, even when there are frequent partition changes, we actuall speed such cases up significantly. By using slots COPY of very narrow rows into unlogged / temporary might slow down very slightly (due to the indirect function calls). Author: David Rowley, Andres Freund, Haribabu Kommi Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20190327054923.t3epfuewxfqdt22e@alap3.anarazel.de
2019-04-05 00:47:19 +02:00
*
Fix comments related to table AMs Incorrect function names were referenced. As this fixes some portions of tableam.h, that is mentioned in the docs as something to look at when implementing a table AM, backpatch down to 12 where this has been introduced. Author: Hironobu Suzuki Discussion: https://postgr.es/m/8fe6d672-28dd-3f1d-7aed-ac2f6d599d3f@interdb.jp Backpatch-through: 12
2020-07-14 06:17:11 +02:00
* This is like table_tuple_insert(), but inserts multiple tuples in one
* operation. That's often faster than calling table_tuple_insert() in a loop,
tableam: Add table_multi_insert() and revamp/speed-up COPY FROM buffering. This adds table_multi_insert(), and converts COPY FROM, the only user of heap_multi_insert, to it. A simple conversion of COPY FROM use slots would have yielded a slowdown when inserting into a partitioned table for some workloads. Different partitions might need different slots (both slot types and their descriptors), and dropping / creating slots when there's constant partition changes is measurable. Thus instead revamp the COPY FROM buffering for partitioned tables to allow to buffer inserts into multiple tables, flushing only when limits are reached across all partition buffers. By only dropping slots when there've been inserts into too many different partitions, the aforementioned overhead is gone. By allowing larger batches, even when there are frequent partition changes, we actuall speed such cases up significantly. By using slots COPY of very narrow rows into unlogged / temporary might slow down very slightly (due to the indirect function calls). Author: David Rowley, Andres Freund, Haribabu Kommi Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20190327054923.t3epfuewxfqdt22e@alap3.anarazel.de
2019-04-05 00:47:19 +02:00
* because e.g. the AM can reduce WAL logging and page locking overhead.
*
Fix typos. Reported-by: Justin Pryzby Author: Justin Pryzby Discussion: https://postgr.es/m/20200206021432.GA24549@telsasoft.com
2020-02-10 05:01:18 +01:00
* Except for taking `nslots` tuples as input, and an array of TupleTableSlots
tableam: Add table_multi_insert() and revamp/speed-up COPY FROM buffering. This adds table_multi_insert(), and converts COPY FROM, the only user of heap_multi_insert, to it. A simple conversion of COPY FROM use slots would have yielded a slowdown when inserting into a partitioned table for some workloads. Different partitions might need different slots (both slot types and their descriptors), and dropping / creating slots when there's constant partition changes is measurable. Thus instead revamp the COPY FROM buffering for partitioned tables to allow to buffer inserts into multiple tables, flushing only when limits are reached across all partition buffers. By only dropping slots when there've been inserts into too many different partitions, the aforementioned overhead is gone. By allowing larger batches, even when there are frequent partition changes, we actuall speed such cases up significantly. By using slots COPY of very narrow rows into unlogged / temporary might slow down very slightly (due to the indirect function calls). Author: David Rowley, Andres Freund, Haribabu Kommi Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20190327054923.t3epfuewxfqdt22e@alap3.anarazel.de
2019-04-05 00:47:19 +02:00
* in `slots`, the parameters for table_multi_insert() are the same as for
tableam: Rename wrapper functions to match callback names. Some of the wrapper functions didn't match the callback names. Many of them due to staying "consistent" with historic naming of the wrapped functionality. We decided that for most cases it's more important to be for tableam to be consistent going forward, than with the past. The one exception is beginscan/endscan/... because it'd have looked odd to have systable_beginscan/endscan/... with a different naming scheme, and changing the systable_* APIs would have caused way too much churn (including breaking a lot of external users). Author: Ashwin Agrawal, with some small additions by Andres Freund Reviewed-By: Andres Freund Discussion: https://postgr.es/m/CALfoeiugyrXZfX7n0ORCa4L-m834dzmaE8eFdbNR6PMpetU4Ww@mail.gmail.com
2019-05-24 01:25:48 +02:00
* table_tuple_insert().
tableam: Add table_multi_insert() and revamp/speed-up COPY FROM buffering. This adds table_multi_insert(), and converts COPY FROM, the only user of heap_multi_insert, to it. A simple conversion of COPY FROM use slots would have yielded a slowdown when inserting into a partitioned table for some workloads. Different partitions might need different slots (both slot types and their descriptors), and dropping / creating slots when there's constant partition changes is measurable. Thus instead revamp the COPY FROM buffering for partitioned tables to allow to buffer inserts into multiple tables, flushing only when limits are reached across all partition buffers. By only dropping slots when there've been inserts into too many different partitions, the aforementioned overhead is gone. By allowing larger batches, even when there are frequent partition changes, we actuall speed such cases up significantly. By using slots COPY of very narrow rows into unlogged / temporary might slow down very slightly (due to the indirect function calls). Author: David Rowley, Andres Freund, Haribabu Kommi Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20190327054923.t3epfuewxfqdt22e@alap3.anarazel.de
2019-04-05 00:47:19 +02:00
*
* Note: this leaks memory into the current memory context. You can create a
* temporary context before calling this, if that's a problem.
*/
static inline void
table_multi_insert(Relation rel, TupleTableSlot **slots, int nslots,
CommandId cid, int options, struct BulkInsertStateData *bistate)
{
rel->rd_tableam->multi_insert(rel, slots, nslots,
cid, options, bistate);
}
tableam: Add tuple_{insert, delete, update, lock} and use. This adds new, required, table AM callbacks for insert/delete/update and lock_tuple. To be able to reasonably use those, the EvalPlanQual mechanism had to be adapted, moving more logic into the AM. Previously both delete/update/lock call-sites and the EPQ mechanism had to have awareness of the specific tuple format to be able to fetch the latest version of a tuple. Obviously that needs to be abstracted away. To do so, move the logic that find the latest row version into the AM. lock_tuple has a new flag argument, TUPLE_LOCK_FLAG_FIND_LAST_VERSION, that forces it to lock the last version, rather than the current one. It'd have been possible to do so via a separate callback as well, but finding the last version usually also necessitates locking the newest version, making it sensible to combine the two. This replaces the previous use of EvalPlanQualFetch(). Additionally HeapTupleUpdated, which previously signaled either a concurrent update or delete, is now split into two, to avoid callers needing AM specific knowledge to differentiate. The move of finding the latest row version into tuple_lock means that encountering a row concurrently moved into another partition will now raise an error about "tuple to be locked" rather than "tuple to be updated/deleted" - which is accurate, as that always happens when locking rows. While possible slightly less helpful for users, it seems like an acceptable trade-off. As part of this commit HTSU_Result has been renamed to TM_Result, and its members been expanded to differentiated between updating and deleting. HeapUpdateFailureData has been renamed to TM_FailureData. The interface to speculative insertion is changed so nodeModifyTable.c does not have to set the speculative token itself anymore. Instead there's a version of tuple_insert, tuple_insert_speculative, that performs the speculative insertion (without requiring a flag to signal that fact), and the speculative insertion is either made permanent with table_complete_speculative(succeeded = true) or aborted with succeeded = false). Note that multi_insert is not yet routed through tableam, nor is COPY. Changing multi_insert requires changes to copy.c that are large enough to better be done separately. Similarly, although simpler, CREATE TABLE AS and CREATE MATERIALIZED VIEW are also only going to be adjusted in a later commit. Author: Andres Freund and Haribabu Kommi Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20190313003903.nwvrxi7rw3ywhdel@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-24 03:55:57 +01:00
/*
Allow locking updated tuples in tuple_update() and tuple_delete() Currently, in read committed transaction isolation mode (default), we have the following sequence of actions when tuple_update()/tuple_delete() finds the tuple updated by concurrent transaction. 1. Attempt to update/delete tuple with tuple_update()/tuple_delete(), which returns TM_Updated. 2. Lock tuple with tuple_lock(). 3. Re-evaluate plan qual (recheck if we still need to update/delete and calculate the new tuple for update). 4. Second attempt to update/delete tuple with tuple_update()/tuple_delete(). This attempt should be successful, since the tuple was previously locked. This patch eliminates step 2 by taking the lock during first tuple_update()/tuple_delete() call. Heap table access method saves some efforts by checking the updated tuple once instead of twice. Future undo-based table access methods, which will start from the latest row version, can immediately place a lock there. The code in nodeModifyTable.c is simplified by removing the nested switch/case. Discussion: https://postgr.es/m/CAPpHfdua-YFw3XTprfutzGp28xXLigFtzNbuFY8yPhqeq6X5kg%40mail.gmail.com Reviewed-by: Aleksander Alekseev, Pavel Borisov, Vignesh C, Mason Sharp Reviewed-by: Andres Freund, Chris Travers
2023-03-22 22:13:37 +01:00
* Delete a tuple (or lock last tuple version if lockedSlot is given).
tableam: Add tuple_{insert, delete, update, lock} and use. This adds new, required, table AM callbacks for insert/delete/update and lock_tuple. To be able to reasonably use those, the EvalPlanQual mechanism had to be adapted, moving more logic into the AM. Previously both delete/update/lock call-sites and the EPQ mechanism had to have awareness of the specific tuple format to be able to fetch the latest version of a tuple. Obviously that needs to be abstracted away. To do so, move the logic that find the latest row version into the AM. lock_tuple has a new flag argument, TUPLE_LOCK_FLAG_FIND_LAST_VERSION, that forces it to lock the last version, rather than the current one. It'd have been possible to do so via a separate callback as well, but finding the last version usually also necessitates locking the newest version, making it sensible to combine the two. This replaces the previous use of EvalPlanQualFetch(). Additionally HeapTupleUpdated, which previously signaled either a concurrent update or delete, is now split into two, to avoid callers needing AM specific knowledge to differentiate. The move of finding the latest row version into tuple_lock means that encountering a row concurrently moved into another partition will now raise an error about "tuple to be locked" rather than "tuple to be updated/deleted" - which is accurate, as that always happens when locking rows. While possible slightly less helpful for users, it seems like an acceptable trade-off. As part of this commit HTSU_Result has been renamed to TM_Result, and its members been expanded to differentiated between updating and deleting. HeapUpdateFailureData has been renamed to TM_FailureData. The interface to speculative insertion is changed so nodeModifyTable.c does not have to set the speculative token itself anymore. Instead there's a version of tuple_insert, tuple_insert_speculative, that performs the speculative insertion (without requiring a flag to signal that fact), and the speculative insertion is either made permanent with table_complete_speculative(succeeded = true) or aborted with succeeded = false). Note that multi_insert is not yet routed through tableam, nor is COPY. Changing multi_insert requires changes to copy.c that are large enough to better be done separately. Similarly, although simpler, CREATE TABLE AS and CREATE MATERIALIZED VIEW are also only going to be adjusted in a later commit. Author: Andres Freund and Haribabu Kommi Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20190313003903.nwvrxi7rw3ywhdel@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-24 03:55:57 +01:00
*
* NB: do not call this directly unless prepared to deal with
tableam: Rename wrapper functions to match callback names. Some of the wrapper functions didn't match the callback names. Many of them due to staying "consistent" with historic naming of the wrapped functionality. We decided that for most cases it's more important to be for tableam to be consistent going forward, than with the past. The one exception is beginscan/endscan/... because it'd have looked odd to have systable_beginscan/endscan/... with a different naming scheme, and changing the systable_* APIs would have caused way too much churn (including breaking a lot of external users). Author: Ashwin Agrawal, with some small additions by Andres Freund Reviewed-By: Andres Freund Discussion: https://postgr.es/m/CALfoeiugyrXZfX7n0ORCa4L-m834dzmaE8eFdbNR6PMpetU4Ww@mail.gmail.com
2019-05-24 01:25:48 +02:00
* concurrent-update conditions. Use simple_table_tuple_delete instead.
tableam: Add tuple_{insert, delete, update, lock} and use. This adds new, required, table AM callbacks for insert/delete/update and lock_tuple. To be able to reasonably use those, the EvalPlanQual mechanism had to be adapted, moving more logic into the AM. Previously both delete/update/lock call-sites and the EPQ mechanism had to have awareness of the specific tuple format to be able to fetch the latest version of a tuple. Obviously that needs to be abstracted away. To do so, move the logic that find the latest row version into the AM. lock_tuple has a new flag argument, TUPLE_LOCK_FLAG_FIND_LAST_VERSION, that forces it to lock the last version, rather than the current one. It'd have been possible to do so via a separate callback as well, but finding the last version usually also necessitates locking the newest version, making it sensible to combine the two. This replaces the previous use of EvalPlanQualFetch(). Additionally HeapTupleUpdated, which previously signaled either a concurrent update or delete, is now split into two, to avoid callers needing AM specific knowledge to differentiate. The move of finding the latest row version into tuple_lock means that encountering a row concurrently moved into another partition will now raise an error about "tuple to be locked" rather than "tuple to be updated/deleted" - which is accurate, as that always happens when locking rows. While possible slightly less helpful for users, it seems like an acceptable trade-off. As part of this commit HTSU_Result has been renamed to TM_Result, and its members been expanded to differentiated between updating and deleting. HeapUpdateFailureData has been renamed to TM_FailureData. The interface to speculative insertion is changed so nodeModifyTable.c does not have to set the speculative token itself anymore. Instead there's a version of tuple_insert, tuple_insert_speculative, that performs the speculative insertion (without requiring a flag to signal that fact), and the speculative insertion is either made permanent with table_complete_speculative(succeeded = true) or aborted with succeeded = false). Note that multi_insert is not yet routed through tableam, nor is COPY. Changing multi_insert requires changes to copy.c that are large enough to better be done separately. Similarly, although simpler, CREATE TABLE AS and CREATE MATERIALIZED VIEW are also only going to be adjusted in a later commit. Author: Andres Freund and Haribabu Kommi Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20190313003903.nwvrxi7rw3ywhdel@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-24 03:55:57 +01:00
*
* Input parameters:
* relation - table to be modified (caller must hold suitable lock)
* tid - TID of tuple to be deleted
* cid - delete command ID (used for visibility test, and stored into
* cmax if successful)
* crosscheck - if not InvalidSnapshot, also check tuple against this
* wait - true if should wait for any conflicting update to commit/abort
* Output parameters:
* tmfd - filled in failure cases (see below)
* changingPart - true iff the tuple is being moved to another partition
* table due to an update of the partition key. Otherwise, false.
Allow locking updated tuples in tuple_update() and tuple_delete() Currently, in read committed transaction isolation mode (default), we have the following sequence of actions when tuple_update()/tuple_delete() finds the tuple updated by concurrent transaction. 1. Attempt to update/delete tuple with tuple_update()/tuple_delete(), which returns TM_Updated. 2. Lock tuple with tuple_lock(). 3. Re-evaluate plan qual (recheck if we still need to update/delete and calculate the new tuple for update). 4. Second attempt to update/delete tuple with tuple_update()/tuple_delete(). This attempt should be successful, since the tuple was previously locked. This patch eliminates step 2 by taking the lock during first tuple_update()/tuple_delete() call. Heap table access method saves some efforts by checking the updated tuple once instead of twice. Future undo-based table access methods, which will start from the latest row version, can immediately place a lock there. The code in nodeModifyTable.c is simplified by removing the nested switch/case. Discussion: https://postgr.es/m/CAPpHfdua-YFw3XTprfutzGp28xXLigFtzNbuFY8yPhqeq6X5kg%40mail.gmail.com Reviewed-by: Aleksander Alekseev, Pavel Borisov, Vignesh C, Mason Sharp Reviewed-by: Andres Freund, Chris Travers
2023-03-22 22:13:37 +01:00
* lockedSlot - lazy slot to save the locked tuple if should lock the last
* row version during the concurrent update. NULL if not needed.
tableam: Add tuple_{insert, delete, update, lock} and use. This adds new, required, table AM callbacks for insert/delete/update and lock_tuple. To be able to reasonably use those, the EvalPlanQual mechanism had to be adapted, moving more logic into the AM. Previously both delete/update/lock call-sites and the EPQ mechanism had to have awareness of the specific tuple format to be able to fetch the latest version of a tuple. Obviously that needs to be abstracted away. To do so, move the logic that find the latest row version into the AM. lock_tuple has a new flag argument, TUPLE_LOCK_FLAG_FIND_LAST_VERSION, that forces it to lock the last version, rather than the current one. It'd have been possible to do so via a separate callback as well, but finding the last version usually also necessitates locking the newest version, making it sensible to combine the two. This replaces the previous use of EvalPlanQualFetch(). Additionally HeapTupleUpdated, which previously signaled either a concurrent update or delete, is now split into two, to avoid callers needing AM specific knowledge to differentiate. The move of finding the latest row version into tuple_lock means that encountering a row concurrently moved into another partition will now raise an error about "tuple to be locked" rather than "tuple to be updated/deleted" - which is accurate, as that always happens when locking rows. While possible slightly less helpful for users, it seems like an acceptable trade-off. As part of this commit HTSU_Result has been renamed to TM_Result, and its members been expanded to differentiated between updating and deleting. HeapUpdateFailureData has been renamed to TM_FailureData. The interface to speculative insertion is changed so nodeModifyTable.c does not have to set the speculative token itself anymore. Instead there's a version of tuple_insert, tuple_insert_speculative, that performs the speculative insertion (without requiring a flag to signal that fact), and the speculative insertion is either made permanent with table_complete_speculative(succeeded = true) or aborted with succeeded = false). Note that multi_insert is not yet routed through tableam, nor is COPY. Changing multi_insert requires changes to copy.c that are large enough to better be done separately. Similarly, although simpler, CREATE TABLE AS and CREATE MATERIALIZED VIEW are also only going to be adjusted in a later commit. Author: Andres Freund and Haribabu Kommi Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20190313003903.nwvrxi7rw3ywhdel@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-24 03:55:57 +01:00
*
tableam: comment and formatting fixes. Author: Heikki Linnakangas Discussion: https://postgr.es/m/9a7fb9cc-2419-5db7-8840-ddc10c93f122@iki.fi
2019-04-08 19:25:16 +02:00
* Normal, successful return value is TM_Ok, which means we did actually
* delete it. Failure return codes are TM_SelfModified, TM_Updated, and
* TM_BeingModified (the last only possible if wait == false).
tableam: Add tuple_{insert, delete, update, lock} and use. This adds new, required, table AM callbacks for insert/delete/update and lock_tuple. To be able to reasonably use those, the EvalPlanQual mechanism had to be adapted, moving more logic into the AM. Previously both delete/update/lock call-sites and the EPQ mechanism had to have awareness of the specific tuple format to be able to fetch the latest version of a tuple. Obviously that needs to be abstracted away. To do so, move the logic that find the latest row version into the AM. lock_tuple has a new flag argument, TUPLE_LOCK_FLAG_FIND_LAST_VERSION, that forces it to lock the last version, rather than the current one. It'd have been possible to do so via a separate callback as well, but finding the last version usually also necessitates locking the newest version, making it sensible to combine the two. This replaces the previous use of EvalPlanQualFetch(). Additionally HeapTupleUpdated, which previously signaled either a concurrent update or delete, is now split into two, to avoid callers needing AM specific knowledge to differentiate. The move of finding the latest row version into tuple_lock means that encountering a row concurrently moved into another partition will now raise an error about "tuple to be locked" rather than "tuple to be updated/deleted" - which is accurate, as that always happens when locking rows. While possible slightly less helpful for users, it seems like an acceptable trade-off. As part of this commit HTSU_Result has been renamed to TM_Result, and its members been expanded to differentiated between updating and deleting. HeapUpdateFailureData has been renamed to TM_FailureData. The interface to speculative insertion is changed so nodeModifyTable.c does not have to set the speculative token itself anymore. Instead there's a version of tuple_insert, tuple_insert_speculative, that performs the speculative insertion (without requiring a flag to signal that fact), and the speculative insertion is either made permanent with table_complete_speculative(succeeded = true) or aborted with succeeded = false). Note that multi_insert is not yet routed through tableam, nor is COPY. Changing multi_insert requires changes to copy.c that are large enough to better be done separately. Similarly, although simpler, CREATE TABLE AS and CREATE MATERIALIZED VIEW are also only going to be adjusted in a later commit. Author: Andres Freund and Haribabu Kommi Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20190313003903.nwvrxi7rw3ywhdel@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-24 03:55:57 +01:00
*
* In the failure cases, the routine fills *tmfd with the tuple's t_ctid,
* t_xmax, and, if possible, and, if possible, t_cmax. See comments for
* struct TM_FailureData for additional info.
*/
static inline TM_Result
tableam: Rename wrapper functions to match callback names. Some of the wrapper functions didn't match the callback names. Many of them due to staying "consistent" with historic naming of the wrapped functionality. We decided that for most cases it's more important to be for tableam to be consistent going forward, than with the past. The one exception is beginscan/endscan/... because it'd have looked odd to have systable_beginscan/endscan/... with a different naming scheme, and changing the systable_* APIs would have caused way too much churn (including breaking a lot of external users). Author: Ashwin Agrawal, with some small additions by Andres Freund Reviewed-By: Andres Freund Discussion: https://postgr.es/m/CALfoeiugyrXZfX7n0ORCa4L-m834dzmaE8eFdbNR6PMpetU4Ww@mail.gmail.com
2019-05-24 01:25:48 +02:00
table_tuple_delete(Relation rel, ItemPointer tid, CommandId cid,
Snapshot snapshot, Snapshot crosscheck, bool wait,
Allow locking updated tuples in tuple_update() and tuple_delete() Currently, in read committed transaction isolation mode (default), we have the following sequence of actions when tuple_update()/tuple_delete() finds the tuple updated by concurrent transaction. 1. Attempt to update/delete tuple with tuple_update()/tuple_delete(), which returns TM_Updated. 2. Lock tuple with tuple_lock(). 3. Re-evaluate plan qual (recheck if we still need to update/delete and calculate the new tuple for update). 4. Second attempt to update/delete tuple with tuple_update()/tuple_delete(). This attempt should be successful, since the tuple was previously locked. This patch eliminates step 2 by taking the lock during first tuple_update()/tuple_delete() call. Heap table access method saves some efforts by checking the updated tuple once instead of twice. Future undo-based table access methods, which will start from the latest row version, can immediately place a lock there. The code in nodeModifyTable.c is simplified by removing the nested switch/case. Discussion: https://postgr.es/m/CAPpHfdua-YFw3XTprfutzGp28xXLigFtzNbuFY8yPhqeq6X5kg%40mail.gmail.com Reviewed-by: Aleksander Alekseev, Pavel Borisov, Vignesh C, Mason Sharp Reviewed-by: Andres Freund, Chris Travers
2023-03-22 22:13:37 +01:00
TM_FailureData *tmfd, bool changingPart,
LazyTupleTableSlot *lockedSlot)
tableam: Add tuple_{insert, delete, update, lock} and use. This adds new, required, table AM callbacks for insert/delete/update and lock_tuple. To be able to reasonably use those, the EvalPlanQual mechanism had to be adapted, moving more logic into the AM. Previously both delete/update/lock call-sites and the EPQ mechanism had to have awareness of the specific tuple format to be able to fetch the latest version of a tuple. Obviously that needs to be abstracted away. To do so, move the logic that find the latest row version into the AM. lock_tuple has a new flag argument, TUPLE_LOCK_FLAG_FIND_LAST_VERSION, that forces it to lock the last version, rather than the current one. It'd have been possible to do so via a separate callback as well, but finding the last version usually also necessitates locking the newest version, making it sensible to combine the two. This replaces the previous use of EvalPlanQualFetch(). Additionally HeapTupleUpdated, which previously signaled either a concurrent update or delete, is now split into two, to avoid callers needing AM specific knowledge to differentiate. The move of finding the latest row version into tuple_lock means that encountering a row concurrently moved into another partition will now raise an error about "tuple to be locked" rather than "tuple to be updated/deleted" - which is accurate, as that always happens when locking rows. While possible slightly less helpful for users, it seems like an acceptable trade-off. As part of this commit HTSU_Result has been renamed to TM_Result, and its members been expanded to differentiated between updating and deleting. HeapUpdateFailureData has been renamed to TM_FailureData. The interface to speculative insertion is changed so nodeModifyTable.c does not have to set the speculative token itself anymore. Instead there's a version of tuple_insert, tuple_insert_speculative, that performs the speculative insertion (without requiring a flag to signal that fact), and the speculative insertion is either made permanent with table_complete_speculative(succeeded = true) or aborted with succeeded = false). Note that multi_insert is not yet routed through tableam, nor is COPY. Changing multi_insert requires changes to copy.c that are large enough to better be done separately. Similarly, although simpler, CREATE TABLE AS and CREATE MATERIALIZED VIEW are also only going to be adjusted in a later commit. Author: Andres Freund and Haribabu Kommi Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20190313003903.nwvrxi7rw3ywhdel@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-24 03:55:57 +01:00
{
return rel->rd_tableam->tuple_delete(rel, tid, cid,
snapshot, crosscheck,
Allow locking updated tuples in tuple_update() and tuple_delete() Currently, in read committed transaction isolation mode (default), we have the following sequence of actions when tuple_update()/tuple_delete() finds the tuple updated by concurrent transaction. 1. Attempt to update/delete tuple with tuple_update()/tuple_delete(), which returns TM_Updated. 2. Lock tuple with tuple_lock(). 3. Re-evaluate plan qual (recheck if we still need to update/delete and calculate the new tuple for update). 4. Second attempt to update/delete tuple with tuple_update()/tuple_delete(). This attempt should be successful, since the tuple was previously locked. This patch eliminates step 2 by taking the lock during first tuple_update()/tuple_delete() call. Heap table access method saves some efforts by checking the updated tuple once instead of twice. Future undo-based table access methods, which will start from the latest row version, can immediately place a lock there. The code in nodeModifyTable.c is simplified by removing the nested switch/case. Discussion: https://postgr.es/m/CAPpHfdua-YFw3XTprfutzGp28xXLigFtzNbuFY8yPhqeq6X5kg%40mail.gmail.com Reviewed-by: Aleksander Alekseev, Pavel Borisov, Vignesh C, Mason Sharp Reviewed-by: Andres Freund, Chris Travers
2023-03-22 22:13:37 +01:00
wait, tmfd, changingPart,
lockedSlot);
tableam: Add tuple_{insert, delete, update, lock} and use. This adds new, required, table AM callbacks for insert/delete/update and lock_tuple. To be able to reasonably use those, the EvalPlanQual mechanism had to be adapted, moving more logic into the AM. Previously both delete/update/lock call-sites and the EPQ mechanism had to have awareness of the specific tuple format to be able to fetch the latest version of a tuple. Obviously that needs to be abstracted away. To do so, move the logic that find the latest row version into the AM. lock_tuple has a new flag argument, TUPLE_LOCK_FLAG_FIND_LAST_VERSION, that forces it to lock the last version, rather than the current one. It'd have been possible to do so via a separate callback as well, but finding the last version usually also necessitates locking the newest version, making it sensible to combine the two. This replaces the previous use of EvalPlanQualFetch(). Additionally HeapTupleUpdated, which previously signaled either a concurrent update or delete, is now split into two, to avoid callers needing AM specific knowledge to differentiate. The move of finding the latest row version into tuple_lock means that encountering a row concurrently moved into another partition will now raise an error about "tuple to be locked" rather than "tuple to be updated/deleted" - which is accurate, as that always happens when locking rows. While possible slightly less helpful for users, it seems like an acceptable trade-off. As part of this commit HTSU_Result has been renamed to TM_Result, and its members been expanded to differentiated between updating and deleting. HeapUpdateFailureData has been renamed to TM_FailureData. The interface to speculative insertion is changed so nodeModifyTable.c does not have to set the speculative token itself anymore. Instead there's a version of tuple_insert, tuple_insert_speculative, that performs the speculative insertion (without requiring a flag to signal that fact), and the speculative insertion is either made permanent with table_complete_speculative(succeeded = true) or aborted with succeeded = false). Note that multi_insert is not yet routed through tableam, nor is COPY. Changing multi_insert requires changes to copy.c that are large enough to better be done separately. Similarly, although simpler, CREATE TABLE AS and CREATE MATERIALIZED VIEW are also only going to be adjusted in a later commit. Author: Andres Freund and Haribabu Kommi Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20190313003903.nwvrxi7rw3ywhdel@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-24 03:55:57 +01:00
}
/*
Allow locking updated tuples in tuple_update() and tuple_delete() Currently, in read committed transaction isolation mode (default), we have the following sequence of actions when tuple_update()/tuple_delete() finds the tuple updated by concurrent transaction. 1. Attempt to update/delete tuple with tuple_update()/tuple_delete(), which returns TM_Updated. 2. Lock tuple with tuple_lock(). 3. Re-evaluate plan qual (recheck if we still need to update/delete and calculate the new tuple for update). 4. Second attempt to update/delete tuple with tuple_update()/tuple_delete(). This attempt should be successful, since the tuple was previously locked. This patch eliminates step 2 by taking the lock during first tuple_update()/tuple_delete() call. Heap table access method saves some efforts by checking the updated tuple once instead of twice. Future undo-based table access methods, which will start from the latest row version, can immediately place a lock there. The code in nodeModifyTable.c is simplified by removing the nested switch/case. Discussion: https://postgr.es/m/CAPpHfdua-YFw3XTprfutzGp28xXLigFtzNbuFY8yPhqeq6X5kg%40mail.gmail.com Reviewed-by: Aleksander Alekseev, Pavel Borisov, Vignesh C, Mason Sharp Reviewed-by: Andres Freund, Chris Travers
2023-03-22 22:13:37 +01:00
* Update a tuple (or lock last tuple version if lockedSlot is given).
tableam: Add tuple_{insert, delete, update, lock} and use. This adds new, required, table AM callbacks for insert/delete/update and lock_tuple. To be able to reasonably use those, the EvalPlanQual mechanism had to be adapted, moving more logic into the AM. Previously both delete/update/lock call-sites and the EPQ mechanism had to have awareness of the specific tuple format to be able to fetch the latest version of a tuple. Obviously that needs to be abstracted away. To do so, move the logic that find the latest row version into the AM. lock_tuple has a new flag argument, TUPLE_LOCK_FLAG_FIND_LAST_VERSION, that forces it to lock the last version, rather than the current one. It'd have been possible to do so via a separate callback as well, but finding the last version usually also necessitates locking the newest version, making it sensible to combine the two. This replaces the previous use of EvalPlanQualFetch(). Additionally HeapTupleUpdated, which previously signaled either a concurrent update or delete, is now split into two, to avoid callers needing AM specific knowledge to differentiate. The move of finding the latest row version into tuple_lock means that encountering a row concurrently moved into another partition will now raise an error about "tuple to be locked" rather than "tuple to be updated/deleted" - which is accurate, as that always happens when locking rows. While possible slightly less helpful for users, it seems like an acceptable trade-off. As part of this commit HTSU_Result has been renamed to TM_Result, and its members been expanded to differentiated between updating and deleting. HeapUpdateFailureData has been renamed to TM_FailureData. The interface to speculative insertion is changed so nodeModifyTable.c does not have to set the speculative token itself anymore. Instead there's a version of tuple_insert, tuple_insert_speculative, that performs the speculative insertion (without requiring a flag to signal that fact), and the speculative insertion is either made permanent with table_complete_speculative(succeeded = true) or aborted with succeeded = false). Note that multi_insert is not yet routed through tableam, nor is COPY. Changing multi_insert requires changes to copy.c that are large enough to better be done separately. Similarly, although simpler, CREATE TABLE AS and CREATE MATERIALIZED VIEW are also only going to be adjusted in a later commit. Author: Andres Freund and Haribabu Kommi Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20190313003903.nwvrxi7rw3ywhdel@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-24 03:55:57 +01:00
*
* NB: do not call this directly unless you are prepared to deal with
tableam: Rename wrapper functions to match callback names. Some of the wrapper functions didn't match the callback names. Many of them due to staying "consistent" with historic naming of the wrapped functionality. We decided that for most cases it's more important to be for tableam to be consistent going forward, than with the past. The one exception is beginscan/endscan/... because it'd have looked odd to have systable_beginscan/endscan/... with a different naming scheme, and changing the systable_* APIs would have caused way too much churn (including breaking a lot of external users). Author: Ashwin Agrawal, with some small additions by Andres Freund Reviewed-By: Andres Freund Discussion: https://postgr.es/m/CALfoeiugyrXZfX7n0ORCa4L-m834dzmaE8eFdbNR6PMpetU4Ww@mail.gmail.com
2019-05-24 01:25:48 +02:00
* concurrent-update conditions. Use simple_table_tuple_update instead.
tableam: Add tuple_{insert, delete, update, lock} and use. This adds new, required, table AM callbacks for insert/delete/update and lock_tuple. To be able to reasonably use those, the EvalPlanQual mechanism had to be adapted, moving more logic into the AM. Previously both delete/update/lock call-sites and the EPQ mechanism had to have awareness of the specific tuple format to be able to fetch the latest version of a tuple. Obviously that needs to be abstracted away. To do so, move the logic that find the latest row version into the AM. lock_tuple has a new flag argument, TUPLE_LOCK_FLAG_FIND_LAST_VERSION, that forces it to lock the last version, rather than the current one. It'd have been possible to do so via a separate callback as well, but finding the last version usually also necessitates locking the newest version, making it sensible to combine the two. This replaces the previous use of EvalPlanQualFetch(). Additionally HeapTupleUpdated, which previously signaled either a concurrent update or delete, is now split into two, to avoid callers needing AM specific knowledge to differentiate. The move of finding the latest row version into tuple_lock means that encountering a row concurrently moved into another partition will now raise an error about "tuple to be locked" rather than "tuple to be updated/deleted" - which is accurate, as that always happens when locking rows. While possible slightly less helpful for users, it seems like an acceptable trade-off. As part of this commit HTSU_Result has been renamed to TM_Result, and its members been expanded to differentiated between updating and deleting. HeapUpdateFailureData has been renamed to TM_FailureData. The interface to speculative insertion is changed so nodeModifyTable.c does not have to set the speculative token itself anymore. Instead there's a version of tuple_insert, tuple_insert_speculative, that performs the speculative insertion (without requiring a flag to signal that fact), and the speculative insertion is either made permanent with table_complete_speculative(succeeded = true) or aborted with succeeded = false). Note that multi_insert is not yet routed through tableam, nor is COPY. Changing multi_insert requires changes to copy.c that are large enough to better be done separately. Similarly, although simpler, CREATE TABLE AS and CREATE MATERIALIZED VIEW are also only going to be adjusted in a later commit. Author: Andres Freund and Haribabu Kommi Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20190313003903.nwvrxi7rw3ywhdel@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-24 03:55:57 +01:00
*
* Input parameters:
* relation - table to be modified (caller must hold suitable lock)
* otid - TID of old tuple to be replaced
tableam: Comment fixes. Author: Haribabu Kommi Discussion: CAJrrPGeeYOqP3hkZyohDx_8dot4zvPuPMDBmhJ=iC85cTBNeYw@mail.gmail.com
2019-03-29 16:16:35 +01:00
* slot - newly constructed tuple data to store
tableam: Add tuple_{insert, delete, update, lock} and use. This adds new, required, table AM callbacks for insert/delete/update and lock_tuple. To be able to reasonably use those, the EvalPlanQual mechanism had to be adapted, moving more logic into the AM. Previously both delete/update/lock call-sites and the EPQ mechanism had to have awareness of the specific tuple format to be able to fetch the latest version of a tuple. Obviously that needs to be abstracted away. To do so, move the logic that find the latest row version into the AM. lock_tuple has a new flag argument, TUPLE_LOCK_FLAG_FIND_LAST_VERSION, that forces it to lock the last version, rather than the current one. It'd have been possible to do so via a separate callback as well, but finding the last version usually also necessitates locking the newest version, making it sensible to combine the two. This replaces the previous use of EvalPlanQualFetch(). Additionally HeapTupleUpdated, which previously signaled either a concurrent update or delete, is now split into two, to avoid callers needing AM specific knowledge to differentiate. The move of finding the latest row version into tuple_lock means that encountering a row concurrently moved into another partition will now raise an error about "tuple to be locked" rather than "tuple to be updated/deleted" - which is accurate, as that always happens when locking rows. While possible slightly less helpful for users, it seems like an acceptable trade-off. As part of this commit HTSU_Result has been renamed to TM_Result, and its members been expanded to differentiated between updating and deleting. HeapUpdateFailureData has been renamed to TM_FailureData. The interface to speculative insertion is changed so nodeModifyTable.c does not have to set the speculative token itself anymore. Instead there's a version of tuple_insert, tuple_insert_speculative, that performs the speculative insertion (without requiring a flag to signal that fact), and the speculative insertion is either made permanent with table_complete_speculative(succeeded = true) or aborted with succeeded = false). Note that multi_insert is not yet routed through tableam, nor is COPY. Changing multi_insert requires changes to copy.c that are large enough to better be done separately. Similarly, although simpler, CREATE TABLE AS and CREATE MATERIALIZED VIEW are also only going to be adjusted in a later commit. Author: Andres Freund and Haribabu Kommi Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20190313003903.nwvrxi7rw3ywhdel@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-24 03:55:57 +01:00
* cid - update command ID (used for visibility test, and stored into
* cmax/cmin if successful)
* crosscheck - if not InvalidSnapshot, also check old tuple against this
* wait - true if should wait for any conflicting update to commit/abort
* Output parameters:
* tmfd - filled in failure cases (see below)
* lockmode - filled with lock mode acquired on tuple
* update_indexes - in success cases this is set to true if new index entries
* are required for this tuple
Allow locking updated tuples in tuple_update() and tuple_delete() Currently, in read committed transaction isolation mode (default), we have the following sequence of actions when tuple_update()/tuple_delete() finds the tuple updated by concurrent transaction. 1. Attempt to update/delete tuple with tuple_update()/tuple_delete(), which returns TM_Updated. 2. Lock tuple with tuple_lock(). 3. Re-evaluate plan qual (recheck if we still need to update/delete and calculate the new tuple for update). 4. Second attempt to update/delete tuple with tuple_update()/tuple_delete(). This attempt should be successful, since the tuple was previously locked. This patch eliminates step 2 by taking the lock during first tuple_update()/tuple_delete() call. Heap table access method saves some efforts by checking the updated tuple once instead of twice. Future undo-based table access methods, which will start from the latest row version, can immediately place a lock there. The code in nodeModifyTable.c is simplified by removing the nested switch/case. Discussion: https://postgr.es/m/CAPpHfdua-YFw3XTprfutzGp28xXLigFtzNbuFY8yPhqeq6X5kg%40mail.gmail.com Reviewed-by: Aleksander Alekseev, Pavel Borisov, Vignesh C, Mason Sharp Reviewed-by: Andres Freund, Chris Travers
2023-03-22 22:13:37 +01:00
* lockedSlot - lazy slot to save the locked tuple if should lock the last
* row version during the concurrent update. NULL if not needed.
tableam: comment and formatting fixes. Author: Heikki Linnakangas Discussion: https://postgr.es/m/9a7fb9cc-2419-5db7-8840-ddc10c93f122@iki.fi
2019-04-08 19:25:16 +02:00
* Normal, successful return value is TM_Ok, which means we did actually
* update it. Failure return codes are TM_SelfModified, TM_Updated, and
* TM_BeingModified (the last only possible if wait == false).
tableam: Add tuple_{insert, delete, update, lock} and use. This adds new, required, table AM callbacks for insert/delete/update and lock_tuple. To be able to reasonably use those, the EvalPlanQual mechanism had to be adapted, moving more logic into the AM. Previously both delete/update/lock call-sites and the EPQ mechanism had to have awareness of the specific tuple format to be able to fetch the latest version of a tuple. Obviously that needs to be abstracted away. To do so, move the logic that find the latest row version into the AM. lock_tuple has a new flag argument, TUPLE_LOCK_FLAG_FIND_LAST_VERSION, that forces it to lock the last version, rather than the current one. It'd have been possible to do so via a separate callback as well, but finding the last version usually also necessitates locking the newest version, making it sensible to combine the two. This replaces the previous use of EvalPlanQualFetch(). Additionally HeapTupleUpdated, which previously signaled either a concurrent update or delete, is now split into two, to avoid callers needing AM specific knowledge to differentiate. The move of finding the latest row version into tuple_lock means that encountering a row concurrently moved into another partition will now raise an error about "tuple to be locked" rather than "tuple to be updated/deleted" - which is accurate, as that always happens when locking rows. While possible slightly less helpful for users, it seems like an acceptable trade-off. As part of this commit HTSU_Result has been renamed to TM_Result, and its members been expanded to differentiated between updating and deleting. HeapUpdateFailureData has been renamed to TM_FailureData. The interface to speculative insertion is changed so nodeModifyTable.c does not have to set the speculative token itself anymore. Instead there's a version of tuple_insert, tuple_insert_speculative, that performs the speculative insertion (without requiring a flag to signal that fact), and the speculative insertion is either made permanent with table_complete_speculative(succeeded = true) or aborted with succeeded = false). Note that multi_insert is not yet routed through tableam, nor is COPY. Changing multi_insert requires changes to copy.c that are large enough to better be done separately. Similarly, although simpler, CREATE TABLE AS and CREATE MATERIALIZED VIEW are also only going to be adjusted in a later commit. Author: Andres Freund and Haribabu Kommi Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20190313003903.nwvrxi7rw3ywhdel@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-24 03:55:57 +01:00
*
tableam: Comment fixes. Author: Haribabu Kommi Discussion: CAJrrPGeeYOqP3hkZyohDx_8dot4zvPuPMDBmhJ=iC85cTBNeYw@mail.gmail.com
2019-03-29 16:16:35 +01:00
* On success, the slot's tts_tid and tts_tableOid are updated to match the new
* stored tuple; in particular, slot->tts_tid is set to the TID where the
tableam: Add tuple_{insert, delete, update, lock} and use. This adds new, required, table AM callbacks for insert/delete/update and lock_tuple. To be able to reasonably use those, the EvalPlanQual mechanism had to be adapted, moving more logic into the AM. Previously both delete/update/lock call-sites and the EPQ mechanism had to have awareness of the specific tuple format to be able to fetch the latest version of a tuple. Obviously that needs to be abstracted away. To do so, move the logic that find the latest row version into the AM. lock_tuple has a new flag argument, TUPLE_LOCK_FLAG_FIND_LAST_VERSION, that forces it to lock the last version, rather than the current one. It'd have been possible to do so via a separate callback as well, but finding the last version usually also necessitates locking the newest version, making it sensible to combine the two. This replaces the previous use of EvalPlanQualFetch(). Additionally HeapTupleUpdated, which previously signaled either a concurrent update or delete, is now split into two, to avoid callers needing AM specific knowledge to differentiate. The move of finding the latest row version into tuple_lock means that encountering a row concurrently moved into another partition will now raise an error about "tuple to be locked" rather than "tuple to be updated/deleted" - which is accurate, as that always happens when locking rows. While possible slightly less helpful for users, it seems like an acceptable trade-off. As part of this commit HTSU_Result has been renamed to TM_Result, and its members been expanded to differentiated between updating and deleting. HeapUpdateFailureData has been renamed to TM_FailureData. The interface to speculative insertion is changed so nodeModifyTable.c does not have to set the speculative token itself anymore. Instead there's a version of tuple_insert, tuple_insert_speculative, that performs the speculative insertion (without requiring a flag to signal that fact), and the speculative insertion is either made permanent with table_complete_speculative(succeeded = true) or aborted with succeeded = false). Note that multi_insert is not yet routed through tableam, nor is COPY. Changing multi_insert requires changes to copy.c that are large enough to better be done separately. Similarly, although simpler, CREATE TABLE AS and CREATE MATERIALIZED VIEW are also only going to be adjusted in a later commit. Author: Andres Freund and Haribabu Kommi Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20190313003903.nwvrxi7rw3ywhdel@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-24 03:55:57 +01:00
* new tuple was inserted, and its HEAP_ONLY_TUPLE flag is set iff a HOT
* update was done. However, any TOAST changes in the new tuple's
* data are not reflected into *newtup.
*
* In the failure cases, the routine fills *tmfd with the tuple's t_ctid,
* t_xmax, and, if possible, t_cmax. See comments for struct TM_FailureData
* for additional info.
*/
static inline TM_Result
tableam: Rename wrapper functions to match callback names. Some of the wrapper functions didn't match the callback names. Many of them due to staying "consistent" with historic naming of the wrapped functionality. We decided that for most cases it's more important to be for tableam to be consistent going forward, than with the past. The one exception is beginscan/endscan/... because it'd have looked odd to have systable_beginscan/endscan/... with a different naming scheme, and changing the systable_* APIs would have caused way too much churn (including breaking a lot of external users). Author: Ashwin Agrawal, with some small additions by Andres Freund Reviewed-By: Andres Freund Discussion: https://postgr.es/m/CALfoeiugyrXZfX7n0ORCa4L-m834dzmaE8eFdbNR6PMpetU4Ww@mail.gmail.com
2019-05-24 01:25:48 +02:00
table_tuple_update(Relation rel, ItemPointer otid, TupleTableSlot *slot,
CommandId cid, Snapshot snapshot, Snapshot crosscheck,
bool wait, TM_FailureData *tmfd, LockTupleMode *lockmode,
Allow locking updated tuples in tuple_update() and tuple_delete() Currently, in read committed transaction isolation mode (default), we have the following sequence of actions when tuple_update()/tuple_delete() finds the tuple updated by concurrent transaction. 1. Attempt to update/delete tuple with tuple_update()/tuple_delete(), which returns TM_Updated. 2. Lock tuple with tuple_lock(). 3. Re-evaluate plan qual (recheck if we still need to update/delete and calculate the new tuple for update). 4. Second attempt to update/delete tuple with tuple_update()/tuple_delete(). This attempt should be successful, since the tuple was previously locked. This patch eliminates step 2 by taking the lock during first tuple_update()/tuple_delete() call. Heap table access method saves some efforts by checking the updated tuple once instead of twice. Future undo-based table access methods, which will start from the latest row version, can immediately place a lock there. The code in nodeModifyTable.c is simplified by removing the nested switch/case. Discussion: https://postgr.es/m/CAPpHfdua-YFw3XTprfutzGp28xXLigFtzNbuFY8yPhqeq6X5kg%40mail.gmail.com Reviewed-by: Aleksander Alekseev, Pavel Borisov, Vignesh C, Mason Sharp Reviewed-by: Andres Freund, Chris Travers
2023-03-22 22:13:37 +01:00
TU_UpdateIndexes *update_indexes,
LazyTupleTableSlot *lockedSlot)
tableam: Add tuple_{insert, delete, update, lock} and use. This adds new, required, table AM callbacks for insert/delete/update and lock_tuple. To be able to reasonably use those, the EvalPlanQual mechanism had to be adapted, moving more logic into the AM. Previously both delete/update/lock call-sites and the EPQ mechanism had to have awareness of the specific tuple format to be able to fetch the latest version of a tuple. Obviously that needs to be abstracted away. To do so, move the logic that find the latest row version into the AM. lock_tuple has a new flag argument, TUPLE_LOCK_FLAG_FIND_LAST_VERSION, that forces it to lock the last version, rather than the current one. It'd have been possible to do so via a separate callback as well, but finding the last version usually also necessitates locking the newest version, making it sensible to combine the two. This replaces the previous use of EvalPlanQualFetch(). Additionally HeapTupleUpdated, which previously signaled either a concurrent update or delete, is now split into two, to avoid callers needing AM specific knowledge to differentiate. The move of finding the latest row version into tuple_lock means that encountering a row concurrently moved into another partition will now raise an error about "tuple to be locked" rather than "tuple to be updated/deleted" - which is accurate, as that always happens when locking rows. While possible slightly less helpful for users, it seems like an acceptable trade-off. As part of this commit HTSU_Result has been renamed to TM_Result, and its members been expanded to differentiated between updating and deleting. HeapUpdateFailureData has been renamed to TM_FailureData. The interface to speculative insertion is changed so nodeModifyTable.c does not have to set the speculative token itself anymore. Instead there's a version of tuple_insert, tuple_insert_speculative, that performs the speculative insertion (without requiring a flag to signal that fact), and the speculative insertion is either made permanent with table_complete_speculative(succeeded = true) or aborted with succeeded = false). Note that multi_insert is not yet routed through tableam, nor is COPY. Changing multi_insert requires changes to copy.c that are large enough to better be done separately. Similarly, although simpler, CREATE TABLE AS and CREATE MATERIALIZED VIEW are also only going to be adjusted in a later commit. Author: Andres Freund and Haribabu Kommi Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20190313003903.nwvrxi7rw3ywhdel@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-24 03:55:57 +01:00
{
return rel->rd_tableam->tuple_update(rel, otid, slot,
cid, snapshot, crosscheck,
wait, tmfd,
Allow locking updated tuples in tuple_update() and tuple_delete() Currently, in read committed transaction isolation mode (default), we have the following sequence of actions when tuple_update()/tuple_delete() finds the tuple updated by concurrent transaction. 1. Attempt to update/delete tuple with tuple_update()/tuple_delete(), which returns TM_Updated. 2. Lock tuple with tuple_lock(). 3. Re-evaluate plan qual (recheck if we still need to update/delete and calculate the new tuple for update). 4. Second attempt to update/delete tuple with tuple_update()/tuple_delete(). This attempt should be successful, since the tuple was previously locked. This patch eliminates step 2 by taking the lock during first tuple_update()/tuple_delete() call. Heap table access method saves some efforts by checking the updated tuple once instead of twice. Future undo-based table access methods, which will start from the latest row version, can immediately place a lock there. The code in nodeModifyTable.c is simplified by removing the nested switch/case. Discussion: https://postgr.es/m/CAPpHfdua-YFw3XTprfutzGp28xXLigFtzNbuFY8yPhqeq6X5kg%40mail.gmail.com Reviewed-by: Aleksander Alekseev, Pavel Borisov, Vignesh C, Mason Sharp Reviewed-by: Andres Freund, Chris Travers
2023-03-22 22:13:37 +01:00
lockmode, update_indexes,
lockedSlot);
tableam: Add tuple_{insert, delete, update, lock} and use. This adds new, required, table AM callbacks for insert/delete/update and lock_tuple. To be able to reasonably use those, the EvalPlanQual mechanism had to be adapted, moving more logic into the AM. Previously both delete/update/lock call-sites and the EPQ mechanism had to have awareness of the specific tuple format to be able to fetch the latest version of a tuple. Obviously that needs to be abstracted away. To do so, move the logic that find the latest row version into the AM. lock_tuple has a new flag argument, TUPLE_LOCK_FLAG_FIND_LAST_VERSION, that forces it to lock the last version, rather than the current one. It'd have been possible to do so via a separate callback as well, but finding the last version usually also necessitates locking the newest version, making it sensible to combine the two. This replaces the previous use of EvalPlanQualFetch(). Additionally HeapTupleUpdated, which previously signaled either a concurrent update or delete, is now split into two, to avoid callers needing AM specific knowledge to differentiate. The move of finding the latest row version into tuple_lock means that encountering a row concurrently moved into another partition will now raise an error about "tuple to be locked" rather than "tuple to be updated/deleted" - which is accurate, as that always happens when locking rows. While possible slightly less helpful for users, it seems like an acceptable trade-off. As part of this commit HTSU_Result has been renamed to TM_Result, and its members been expanded to differentiated between updating and deleting. HeapUpdateFailureData has been renamed to TM_FailureData. The interface to speculative insertion is changed so nodeModifyTable.c does not have to set the speculative token itself anymore. Instead there's a version of tuple_insert, tuple_insert_speculative, that performs the speculative insertion (without requiring a flag to signal that fact), and the speculative insertion is either made permanent with table_complete_speculative(succeeded = true) or aborted with succeeded = false). Note that multi_insert is not yet routed through tableam, nor is COPY. Changing multi_insert requires changes to copy.c that are large enough to better be done separately. Similarly, although simpler, CREATE TABLE AS and CREATE MATERIALIZED VIEW are also only going to be adjusted in a later commit. Author: Andres Freund and Haribabu Kommi Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20190313003903.nwvrxi7rw3ywhdel@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-24 03:55:57 +01:00
}
/*
* Lock a tuple in the specified mode.
*
* Input parameters:
* relation: relation containing tuple (caller must hold suitable lock)
* tid: TID of tuple to lock
* snapshot: snapshot to use for visibility determinations
* cid: current command ID (used for visibility test, and stored into
* tuple's cmax if lock is successful)
* mode: lock mode desired
* wait_policy: what to do if tuple lock is not available
* flags:
* If TUPLE_LOCK_FLAG_LOCK_UPDATE_IN_PROGRESS, follow the update chain to
* also lock descendant tuples if lock modes don't conflict.
tableam: comment and formatting fixes. Author: Heikki Linnakangas Discussion: https://postgr.es/m/9a7fb9cc-2419-5db7-8840-ddc10c93f122@iki.fi
2019-04-08 19:25:16 +02:00
* If TUPLE_LOCK_FLAG_FIND_LAST_VERSION, follow the update chain and lock
* latest version.
tableam: Add tuple_{insert, delete, update, lock} and use. This adds new, required, table AM callbacks for insert/delete/update and lock_tuple. To be able to reasonably use those, the EvalPlanQual mechanism had to be adapted, moving more logic into the AM. Previously both delete/update/lock call-sites and the EPQ mechanism had to have awareness of the specific tuple format to be able to fetch the latest version of a tuple. Obviously that needs to be abstracted away. To do so, move the logic that find the latest row version into the AM. lock_tuple has a new flag argument, TUPLE_LOCK_FLAG_FIND_LAST_VERSION, that forces it to lock the last version, rather than the current one. It'd have been possible to do so via a separate callback as well, but finding the last version usually also necessitates locking the newest version, making it sensible to combine the two. This replaces the previous use of EvalPlanQualFetch(). Additionally HeapTupleUpdated, which previously signaled either a concurrent update or delete, is now split into two, to avoid callers needing AM specific knowledge to differentiate. The move of finding the latest row version into tuple_lock means that encountering a row concurrently moved into another partition will now raise an error about "tuple to be locked" rather than "tuple to be updated/deleted" - which is accurate, as that always happens when locking rows. While possible slightly less helpful for users, it seems like an acceptable trade-off. As part of this commit HTSU_Result has been renamed to TM_Result, and its members been expanded to differentiated between updating and deleting. HeapUpdateFailureData has been renamed to TM_FailureData. The interface to speculative insertion is changed so nodeModifyTable.c does not have to set the speculative token itself anymore. Instead there's a version of tuple_insert, tuple_insert_speculative, that performs the speculative insertion (without requiring a flag to signal that fact), and the speculative insertion is either made permanent with table_complete_speculative(succeeded = true) or aborted with succeeded = false). Note that multi_insert is not yet routed through tableam, nor is COPY. Changing multi_insert requires changes to copy.c that are large enough to better be done separately. Similarly, although simpler, CREATE TABLE AS and CREATE MATERIALIZED VIEW are also only going to be adjusted in a later commit. Author: Andres Freund and Haribabu Kommi Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20190313003903.nwvrxi7rw3ywhdel@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-24 03:55:57 +01:00
*
* Output parameters:
* *slot: contains the target tuple
* *tmfd: filled in failure cases (see below)
*
* Function result may be:
* TM_Ok: lock was successfully acquired
* TM_Invisible: lock failed because tuple was never visible to us
* TM_SelfModified: lock failed because tuple updated by self
* TM_Updated: lock failed because tuple updated by other xact
* TM_Deleted: lock failed because tuple deleted by other xact
* TM_WouldBlock: lock couldn't be acquired and wait_policy is skip
*
Fix a number of issues around modifying a previously updated row. This commit fixes three, unfortunately related, issues: 1) Since 5db6df0c01, the introduction of DML via tableam, it was possible to trigger "ERROR: unexpected table_lock_tuple status: 1" when updating a row that was previously updated in the same transaction - but only when the previously updated row was before updated in a concurrent transaction (and READ COMMITTED was used). The reason for that was that that case simply wasn't expected. Fixing that lead to: 2) Even before the above commit, there were error checks (introduced in 6868ed7491b7) preventing a row being updated by different commands within the same statement (say in a function called by an UPDATE) - but that check wasn't performed when the row was first updated in a concurrent transaction - instead the second update was silently skipped in that case. After this change we throw the same error as we'd without the concurrent transaction. 3) The error messages (introduced in 6868ed7491b7) preventing such updates emitted the same error message for both DELETE and UPDATE ("tuple to be updated was already modified by an operation triggered by the current command"). While that could be changed separately, it made it hard to write tests that verify the correct correct behavior of the code. This commit changes heap's implementation of table_lock_tuple() to return TM_SelfModified instead of TM_Invisible (previously loosely modeled after EvalPlanQualFetch), and teaches nodeModifyTable.c to handle that in response to table_lock_tuple() and not just in response to table_(delete|update). Additionally it fixes the wrong error message (see 3 above). The comment for table_lock_tuple() is also adjusted to state that TM_Deleted won't return information in TM_FailureData - it'll not always be available. This also adds tests to ensure that DELETE/UPDATE correctly error out when affecting a row that concurrently was modified by another transaction. Author: Andres Freund Reported-By: Tom Lane, when investigating a bug bug fix to another bug by Amit Langote Discussion: https://postgr.es/m/19321.1554567786@sss.pgh.pa.us
2019-04-08 07:14:47 +02:00
* In the failure cases other than TM_Invisible and TM_Deleted, the routine
* fills *tmfd with the tuple's t_ctid, t_xmax, and, if possible, t_cmax. See
* comments for struct TM_FailureData for additional info.
tableam: Add tuple_{insert, delete, update, lock} and use. This adds new, required, table AM callbacks for insert/delete/update and lock_tuple. To be able to reasonably use those, the EvalPlanQual mechanism had to be adapted, moving more logic into the AM. Previously both delete/update/lock call-sites and the EPQ mechanism had to have awareness of the specific tuple format to be able to fetch the latest version of a tuple. Obviously that needs to be abstracted away. To do so, move the logic that find the latest row version into the AM. lock_tuple has a new flag argument, TUPLE_LOCK_FLAG_FIND_LAST_VERSION, that forces it to lock the last version, rather than the current one. It'd have been possible to do so via a separate callback as well, but finding the last version usually also necessitates locking the newest version, making it sensible to combine the two. This replaces the previous use of EvalPlanQualFetch(). Additionally HeapTupleUpdated, which previously signaled either a concurrent update or delete, is now split into two, to avoid callers needing AM specific knowledge to differentiate. The move of finding the latest row version into tuple_lock means that encountering a row concurrently moved into another partition will now raise an error about "tuple to be locked" rather than "tuple to be updated/deleted" - which is accurate, as that always happens when locking rows. While possible slightly less helpful for users, it seems like an acceptable trade-off. As part of this commit HTSU_Result has been renamed to TM_Result, and its members been expanded to differentiated between updating and deleting. HeapUpdateFailureData has been renamed to TM_FailureData. The interface to speculative insertion is changed so nodeModifyTable.c does not have to set the speculative token itself anymore. Instead there's a version of tuple_insert, tuple_insert_speculative, that performs the speculative insertion (without requiring a flag to signal that fact), and the speculative insertion is either made permanent with table_complete_speculative(succeeded = true) or aborted with succeeded = false). Note that multi_insert is not yet routed through tableam, nor is COPY. Changing multi_insert requires changes to copy.c that are large enough to better be done separately. Similarly, although simpler, CREATE TABLE AS and CREATE MATERIALIZED VIEW are also only going to be adjusted in a later commit. Author: Andres Freund and Haribabu Kommi Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20190313003903.nwvrxi7rw3ywhdel@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-24 03:55:57 +01:00
*/
static inline TM_Result
tableam: Rename wrapper functions to match callback names. Some of the wrapper functions didn't match the callback names. Many of them due to staying "consistent" with historic naming of the wrapped functionality. We decided that for most cases it's more important to be for tableam to be consistent going forward, than with the past. The one exception is beginscan/endscan/... because it'd have looked odd to have systable_beginscan/endscan/... with a different naming scheme, and changing the systable_* APIs would have caused way too much churn (including breaking a lot of external users). Author: Ashwin Agrawal, with some small additions by Andres Freund Reviewed-By: Andres Freund Discussion: https://postgr.es/m/CALfoeiugyrXZfX7n0ORCa4L-m834dzmaE8eFdbNR6PMpetU4Ww@mail.gmail.com
2019-05-24 01:25:48 +02:00
table_tuple_lock(Relation rel, ItemPointer tid, Snapshot snapshot,
tableam: Add tuple_{insert, delete, update, lock} and use. This adds new, required, table AM callbacks for insert/delete/update and lock_tuple. To be able to reasonably use those, the EvalPlanQual mechanism had to be adapted, moving more logic into the AM. Previously both delete/update/lock call-sites and the EPQ mechanism had to have awareness of the specific tuple format to be able to fetch the latest version of a tuple. Obviously that needs to be abstracted away. To do so, move the logic that find the latest row version into the AM. lock_tuple has a new flag argument, TUPLE_LOCK_FLAG_FIND_LAST_VERSION, that forces it to lock the last version, rather than the current one. It'd have been possible to do so via a separate callback as well, but finding the last version usually also necessitates locking the newest version, making it sensible to combine the two. This replaces the previous use of EvalPlanQualFetch(). Additionally HeapTupleUpdated, which previously signaled either a concurrent update or delete, is now split into two, to avoid callers needing AM specific knowledge to differentiate. The move of finding the latest row version into tuple_lock means that encountering a row concurrently moved into another partition will now raise an error about "tuple to be locked" rather than "tuple to be updated/deleted" - which is accurate, as that always happens when locking rows. While possible slightly less helpful for users, it seems like an acceptable trade-off. As part of this commit HTSU_Result has been renamed to TM_Result, and its members been expanded to differentiated between updating and deleting. HeapUpdateFailureData has been renamed to TM_FailureData. The interface to speculative insertion is changed so nodeModifyTable.c does not have to set the speculative token itself anymore. Instead there's a version of tuple_insert, tuple_insert_speculative, that performs the speculative insertion (without requiring a flag to signal that fact), and the speculative insertion is either made permanent with table_complete_speculative(succeeded = true) or aborted with succeeded = false). Note that multi_insert is not yet routed through tableam, nor is COPY. Changing multi_insert requires changes to copy.c that are large enough to better be done separately. Similarly, although simpler, CREATE TABLE AS and CREATE MATERIALIZED VIEW are also only going to be adjusted in a later commit. Author: Andres Freund and Haribabu Kommi Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20190313003903.nwvrxi7rw3ywhdel@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-24 03:55:57 +01:00
TupleTableSlot *slot, CommandId cid, LockTupleMode mode,
LockWaitPolicy wait_policy, uint8 flags,
TM_FailureData *tmfd)
{
return rel->rd_tableam->tuple_lock(rel, tid, snapshot, slot,
cid, mode, wait_policy,
flags, tmfd);
}
tableam: Add table_finish_bulk_insert(). This replaces the previous calls of heap_sync() in places using bulk-insert. By passing in the flags used for bulk-insert the AM can decide (first at insert time and then during the finish call) which of the optimizations apply to it, and what operations are necessary to finish a bulk insert operation. Also change HEAP_INSERT_* flags to TABLE_INSERT, and rename hi_options to ti_options. These changes are made even in copy.c, which hasn't yet been converted to tableam. There's no harm in doing so. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-04-01 23:41:42 +02:00
/*
* Perform operations necessary to complete insertions made via
Skip WAL for new relfilenodes, under wal_level=minimal. Until now, only selected bulk operations (e.g. COPY) did this. If a given relfilenode received both a WAL-skipping COPY and a WAL-logged operation (e.g. INSERT), recovery could lose tuples from the COPY. See src/backend/access/transam/README section "Skipping WAL for New RelFileNode" for the new coding rules. Maintainers of table access methods should examine that section. To maintain data durability, just before commit, we choose between an fsync of the relfilenode and copying its contents to WAL. A new GUC, wal_skip_threshold, guides that choice. If this change slows a workload that creates small, permanent relfilenodes under wal_level=minimal, try adjusting wal_skip_threshold. Users setting a timeout on COMMIT may need to adjust that timeout, and log_min_duration_statement analysis will reflect time consumption moving to COMMIT from commands like COPY. Internally, this requires a reliable determination of whether RollbackAndReleaseCurrentSubTransaction() would unlink a relation's current relfilenode. Introduce rd_firstRelfilenodeSubid. Amend the specification of rd_createSubid such that the field is zero when a new rel has an old rd_node. Make relcache.c retain entries for certain dropped relations until end of transaction. Bump XLOG_PAGE_MAGIC, since this introduces XLOG_GIST_ASSIGN_LSN. Future servers accept older WAL, so this bump is discretionary. Kyotaro Horiguchi, reviewed (in earlier, similar versions) by Robert Haas. Heikki Linnakangas and Michael Paquier implemented earlier designs that materially clarified the problem. Reviewed, in earlier designs, by Andrew Dunstan, Andres Freund, Alvaro Herrera, Tom Lane, Fujii Masao, and Simon Riggs. Reported by Martijn van Oosterhout. Discussion: https://postgr.es/m/20150702220524.GA9392@svana.org
2020-04-04 21:25:34 +02:00
* tuple_insert and multi_insert with a BulkInsertState specified.
tableam: Add table_finish_bulk_insert(). This replaces the previous calls of heap_sync() in places using bulk-insert. By passing in the flags used for bulk-insert the AM can decide (first at insert time and then during the finish call) which of the optimizations apply to it, and what operations are necessary to finish a bulk insert operation. Also change HEAP_INSERT_* flags to TABLE_INSERT, and rename hi_options to ti_options. These changes are made even in copy.c, which hasn't yet been converted to tableam. There's no harm in doing so. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-04-01 23:41:42 +02:00
*/
static inline void
table_finish_bulk_insert(Relation rel, int options)
{
/* optional callback */
if (rel->rd_tableam && rel->rd_tableam->finish_bulk_insert)
rel->rd_tableam->finish_bulk_insert(rel, options);
}
tableam: Add tuple_{insert, delete, update, lock} and use. This adds new, required, table AM callbacks for insert/delete/update and lock_tuple. To be able to reasonably use those, the EvalPlanQual mechanism had to be adapted, moving more logic into the AM. Previously both delete/update/lock call-sites and the EPQ mechanism had to have awareness of the specific tuple format to be able to fetch the latest version of a tuple. Obviously that needs to be abstracted away. To do so, move the logic that find the latest row version into the AM. lock_tuple has a new flag argument, TUPLE_LOCK_FLAG_FIND_LAST_VERSION, that forces it to lock the last version, rather than the current one. It'd have been possible to do so via a separate callback as well, but finding the last version usually also necessitates locking the newest version, making it sensible to combine the two. This replaces the previous use of EvalPlanQualFetch(). Additionally HeapTupleUpdated, which previously signaled either a concurrent update or delete, is now split into two, to avoid callers needing AM specific knowledge to differentiate. The move of finding the latest row version into tuple_lock means that encountering a row concurrently moved into another partition will now raise an error about "tuple to be locked" rather than "tuple to be updated/deleted" - which is accurate, as that always happens when locking rows. While possible slightly less helpful for users, it seems like an acceptable trade-off. As part of this commit HTSU_Result has been renamed to TM_Result, and its members been expanded to differentiated between updating and deleting. HeapUpdateFailureData has been renamed to TM_FailureData. The interface to speculative insertion is changed so nodeModifyTable.c does not have to set the speculative token itself anymore. Instead there's a version of tuple_insert, tuple_insert_speculative, that performs the speculative insertion (without requiring a flag to signal that fact), and the speculative insertion is either made permanent with table_complete_speculative(succeeded = true) or aborted with succeeded = false). Note that multi_insert is not yet routed through tableam, nor is COPY. Changing multi_insert requires changes to copy.c that are large enough to better be done separately. Similarly, although simpler, CREATE TABLE AS and CREATE MATERIALIZED VIEW are also only going to be adjusted in a later commit. Author: Andres Freund and Haribabu Kommi Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20190313003903.nwvrxi7rw3ywhdel@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-24 03:55:57 +01:00
tableam: Support for an index build's initial table scan(s). To support building indexes over tables of different AMs, the scans to do so need to be routed through the table AM. While moving a fair amount of code, nearly all the changes are just moving code to below a callback. Currently the range based interface wouldn't make much sense for non block based table AMs. But that seems aceptable for now. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-28 03:59:06 +01:00
/* ------------------------------------------------------------------------
* DDL related functionality.
* ------------------------------------------------------------------------
*/
tableam: relation creation, VACUUM FULL/CLUSTER, SET TABLESPACE. This moves the responsibility for: - creating the storage necessary for a relation, including creating a new relfilenode for a relation with existing storage - non-transactional truncation of a relation - VACUUM FULL / CLUSTER's rewrite of a table below tableam. This is fairly straight forward, with a bit of complexity smattered in to move the computation of xid / multixid horizons below the AM, as they don't make sense for every table AM. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-29 04:01:14 +01:00
/*
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
* Create storage for `rel` in `newrlocator`, with persistence set to
tableam: relation creation, VACUUM FULL/CLUSTER, SET TABLESPACE. This moves the responsibility for: - creating the storage necessary for a relation, including creating a new relfilenode for a relation with existing storage - non-transactional truncation of a relation - VACUUM FULL / CLUSTER's rewrite of a table below tableam. This is fairly straight forward, with a bit of complexity smattered in to move the computation of xid / multixid horizons below the AM, as they don't make sense for every table AM. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-29 04:01:14 +01:00
* `persistence`.
*
* This is used both during relation creation and various DDL operations to
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
* create new rel storage that can be filled from scratch. When creating
* new storage for an existing relfilelocator, this should be called before the
Fix several recently introduced issues around handling new relation forks. Most of these stem from d25f519107 "tableam: relation creation, VACUUM FULL/CLUSTER, SET TABLESPACE.". 1) To pass data to the relation_set_new_filenode() RelationSetNewRelfilenode() was made to update RelationData.rd_rel directly. That's not OK however, as it makes the relcache entries temporarily inconsistent. Which among other scenarios is a problem if a REINDEX targets an index on pg_class - the CatalogTupleUpdate() in RelationSetNewRelfilenode(). Presumably that was introduced because other places in the code do so - while those aren't "good practice" they don't appear to be actively buggy (e.g. because system tables may not be targeted). I (Andres) should have caught this while reviewing and signficantly evolving the code in that commit, mea culpa. Fix that by instead passing in the new RelFileNode as separate argument to relation_set_new_filenode() and rely on the relcache to update the catalog entry. Also revert that the RelationMapUpdateMap() call was changed to immediate, and undo some other more unnecessary changes. 2) Document that the relation_set_new_filenode cannot rely on the whole relcache entry to be valid. It might be worthwhile to refactor the code to never have to rely on that, but given the way heap_create() is currently coded, that'd be a large change. 3) ATExecSetTableSpace() shouldn't do FlushRelationBuffers() itself. A table AM might not use shared buffers at all. Move to index_copy_data() and heapam_relation_copy_data(). 4) heapam_relation_set_new_filenode() previously sometimes accessed rel->rd_rel->relpersistence rather than the `persistence` argument. Code movement mistake. 5) Previously heapam_relation_set_new_filenode() re-opened the smgr relation to create the init for, if necesary. Instead have RelationCreateStorage() return the SMgrRelation and use it to create the init fork. 6) Add a note about the danger of modifying the relcache directly to ATExecSetTableSpace() - it's currently not a bug because there's a check ERRORing for catalog tables. Regression tests and assertion improvements that together trigger the bug described in 1) will be added in a later commit, as there is a related bug on all branches. Reported-By: Michael Paquier Diagnosed-By: Tom Lane and Andres Freund Author: Andres Freund Reviewed-By: Tom Lane Discussion: https://postgr.es/m/20190418011430.GA19133@paquier.xyz
2019-04-30 04:28:05 +02:00
* relcache entry has been updated.
tableam: relation creation, VACUUM FULL/CLUSTER, SET TABLESPACE. This moves the responsibility for: - creating the storage necessary for a relation, including creating a new relfilenode for a relation with existing storage - non-transactional truncation of a relation - VACUUM FULL / CLUSTER's rewrite of a table below tableam. This is fairly straight forward, with a bit of complexity smattered in to move the computation of xid / multixid horizons below the AM, as they don't make sense for every table AM. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-29 04:01:14 +01:00
*
* *freezeXid, *minmulti are set to the xid / multixact horizon for the table
* that pg_class.{relfrozenxid, relminmxid} have to be set to.
*/
static inline void
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
table_relation_set_new_filelocator(Relation rel,
const RelFileLocator *newrlocator,
char persistence,
TransactionId *freezeXid,
MultiXactId *minmulti)
tableam: relation creation, VACUUM FULL/CLUSTER, SET TABLESPACE. This moves the responsibility for: - creating the storage necessary for a relation, including creating a new relfilenode for a relation with existing storage - non-transactional truncation of a relation - VACUUM FULL / CLUSTER's rewrite of a table below tableam. This is fairly straight forward, with a bit of complexity smattered in to move the computation of xid / multixid horizons below the AM, as they don't make sense for every table AM. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-29 04:01:14 +01:00
{
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
rel->rd_tableam->relation_set_new_filelocator(rel, newrlocator,
persistence, freezeXid,
minmulti);
tableam: relation creation, VACUUM FULL/CLUSTER, SET TABLESPACE. This moves the responsibility for: - creating the storage necessary for a relation, including creating a new relfilenode for a relation with existing storage - non-transactional truncation of a relation - VACUUM FULL / CLUSTER's rewrite of a table below tableam. This is fairly straight forward, with a bit of complexity smattered in to move the computation of xid / multixid horizons below the AM, as they don't make sense for every table AM. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-29 04:01:14 +01:00
}
/*
* Remove all table contents from `rel`, in a non-transactional manner.
* Non-transactional meaning that there's no need to support rollbacks. This
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
* commonly only is used to perform truncations for relation storage created in
* the current transaction.
tableam: relation creation, VACUUM FULL/CLUSTER, SET TABLESPACE. This moves the responsibility for: - creating the storage necessary for a relation, including creating a new relfilenode for a relation with existing storage - non-transactional truncation of a relation - VACUUM FULL / CLUSTER's rewrite of a table below tableam. This is fairly straight forward, with a bit of complexity smattered in to move the computation of xid / multixid horizons below the AM, as they don't make sense for every table AM. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-29 04:01:14 +01:00
*/
static inline void
table_relation_nontransactional_truncate(Relation rel)
{
rel->rd_tableam->relation_nontransactional_truncate(rel);
}
/*
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
* Copy data from `rel` into the new relfilelocator `newrlocator`. The new
* relfilelocator may not have storage associated before this function is
tableam: relation creation, VACUUM FULL/CLUSTER, SET TABLESPACE. This moves the responsibility for: - creating the storage necessary for a relation, including creating a new relfilenode for a relation with existing storage - non-transactional truncation of a relation - VACUUM FULL / CLUSTER's rewrite of a table below tableam. This is fairly straight forward, with a bit of complexity smattered in to move the computation of xid / multixid horizons below the AM, as they don't make sense for every table AM. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-29 04:01:14 +01:00
* called. This is only supposed to be used for low level operations like
* changing a relation's tablespace.
*/
static inline void
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
table_relation_copy_data(Relation rel, const RelFileLocator *newrlocator)
tableam: relation creation, VACUUM FULL/CLUSTER, SET TABLESPACE. This moves the responsibility for: - creating the storage necessary for a relation, including creating a new relfilenode for a relation with existing storage - non-transactional truncation of a relation - VACUUM FULL / CLUSTER's rewrite of a table below tableam. This is fairly straight forward, with a bit of complexity smattered in to move the computation of xid / multixid horizons below the AM, as they don't make sense for every table AM. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-29 04:01:14 +01:00
{
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
rel->rd_tableam->relation_copy_data(rel, newrlocator);
tableam: relation creation, VACUUM FULL/CLUSTER, SET TABLESPACE. This moves the responsibility for: - creating the storage necessary for a relation, including creating a new relfilenode for a relation with existing storage - non-transactional truncation of a relation - VACUUM FULL / CLUSTER's rewrite of a table below tableam. This is fairly straight forward, with a bit of complexity smattered in to move the computation of xid / multixid horizons below the AM, as they don't make sense for every table AM. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-29 04:01:14 +01:00
}
/*
Fix some typos and inconsistencies in tableam.h The defined callback definitions have been using references to heap for a couple of variables and comments. This makes the whole interface more consistent by using "table" which is more generic. A variable storing index information was misspelled as well. Author: Michael Paquier Discussion: https://postgr.es/m/20190601190946.GB1905@paquier.xyz
2019-06-04 02:48:25 +02:00
* Copy data from `OldTable` into `NewTable`, as part of a CLUSTER or VACUUM
tableam: relation creation, VACUUM FULL/CLUSTER, SET TABLESPACE. This moves the responsibility for: - creating the storage necessary for a relation, including creating a new relfilenode for a relation with existing storage - non-transactional truncation of a relation - VACUUM FULL / CLUSTER's rewrite of a table below tableam. This is fairly straight forward, with a bit of complexity smattered in to move the computation of xid / multixid horizons below the AM, as they don't make sense for every table AM. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-29 04:01:14 +01:00
* FULL.
*
Allow pg_class xid & multixid horizons to not be set. This allows table AMs that don't need these horizons. This was already documented in the tableam relation_set_new_filenode callback, but an assert prevented if from actually working (the test AM code contained the change itself). Defang the asserts in the general code, and move the stronger ones into heap AM. Relatedly, after CLUSTER/VACUUM, we'd always assign a relfrozenxid / relminmxid. Change the table_relation_copy_for_cluster() interface to allow the AM to overwrite the horizons that get set on the pg_class entry. This'd also in the future allow AMs like heap to compute a relfrozenxid during rewrite that's the table's actual minimum rather than a pre-determined value. Arguably it'd have been better to move the whole computation / setting of those values into the callback, but it seems likely that for other reasons it'd be better to be able to use one value to vacuum/cluster multiple tables (e.g. a toast's horizon shouldn't be different than the table's). Reported-By: Heikki Linnakangas Author: Andres Freund Discussion: https://postgr.es/m/9a7fb9cc-2419-5db7-8840-ddc10c93f122@iki.fi
2019-04-24 06:42:12 +02:00
* Additional Input parameters:
* - use_sort - if true, the table contents are sorted appropriate for
* `OldIndex`; if false and OldIndex is not InvalidOid, the data is copied
tableam: comment improvements. Author: Brad DeJong Discussion: https://postgr.es/m/CAJnrtnxDYOQFsDfWz2iri0T_fFL2ZbbzgCOE=4yaMcszgcsf4A@mail.gmail.com Backpatch: 12-
2019-07-18 04:39:54 +02:00
* in that index's order; if false and OldIndex is InvalidOid, no sorting is
Allow pg_class xid & multixid horizons to not be set. This allows table AMs that don't need these horizons. This was already documented in the tableam relation_set_new_filenode callback, but an assert prevented if from actually working (the test AM code contained the change itself). Defang the asserts in the general code, and move the stronger ones into heap AM. Relatedly, after CLUSTER/VACUUM, we'd always assign a relfrozenxid / relminmxid. Change the table_relation_copy_for_cluster() interface to allow the AM to overwrite the horizons that get set on the pg_class entry. This'd also in the future allow AMs like heap to compute a relfrozenxid during rewrite that's the table's actual minimum rather than a pre-determined value. Arguably it'd have been better to move the whole computation / setting of those values into the callback, but it seems likely that for other reasons it'd be better to be able to use one value to vacuum/cluster multiple tables (e.g. a toast's horizon shouldn't be different than the table's). Reported-By: Heikki Linnakangas Author: Andres Freund Discussion: https://postgr.es/m/9a7fb9cc-2419-5db7-8840-ddc10c93f122@iki.fi
2019-04-24 06:42:12 +02:00
* performed
tableam: comment improvements. Author: Brad DeJong Discussion: https://postgr.es/m/CAJnrtnxDYOQFsDfWz2iri0T_fFL2ZbbzgCOE=4yaMcszgcsf4A@mail.gmail.com Backpatch: 12-
2019-07-18 04:39:54 +02:00
* - OldIndex - see use_sort
Refactor how VACUUM passes around its XID cutoffs. Use a dedicated struct for the XID/MXID cutoffs used by VACUUM, such as FreezeLimit and OldestXmin. This state is initialized in vacuum.c, and then passed around by code from vacuumlazy.c to heapam.c freezing related routines. The new convention is that everybody works off of the same cutoff state, which is passed around via pointers to const. Also simplify some of the logic for dealing with frozen xmin in heap_prepare_freeze_tuple: add dedicated "xmin_already_frozen" state to clearly distinguish xmin XIDs that we're going to freeze from those that were already frozen from before. That way the routine's xmin handling code is symmetrical with the existing xmax handling code. This is preparation for an upcoming commit that will add page level freezing. Also refactor the control flow within FreezeMultiXactId(), while adding stricter sanity checks. We now test OldestXmin directly, instead of using FreezeLimit as an inexact proxy for OldestXmin. This is further preparation for the page level freezing work, which will make the function's caller cede control of page level freezing to the function where appropriate (where heap_prepare_freeze_tuple sees a tuple that happens to contain a MultiXactId in its xmax). Author: Peter Geoghegan <pg@bowt.ie> Reviewed-By: Jeff Davis <pgsql@j-davis.com> Discussion: https://postgr.es/m/CAH2-WznS9TxXmz2_=SY+SyJyDFbiOftKofM9=aDo68BbXNBUMA@mail.gmail.com
2022-12-22 18:37:59 +01:00
* - OldestXmin - computed by vacuum_get_cutoffs(), even when
Allow pg_class xid & multixid horizons to not be set. This allows table AMs that don't need these horizons. This was already documented in the tableam relation_set_new_filenode callback, but an assert prevented if from actually working (the test AM code contained the change itself). Defang the asserts in the general code, and move the stronger ones into heap AM. Relatedly, after CLUSTER/VACUUM, we'd always assign a relfrozenxid / relminmxid. Change the table_relation_copy_for_cluster() interface to allow the AM to overwrite the horizons that get set on the pg_class entry. This'd also in the future allow AMs like heap to compute a relfrozenxid during rewrite that's the table's actual minimum rather than a pre-determined value. Arguably it'd have been better to move the whole computation / setting of those values into the callback, but it seems likely that for other reasons it'd be better to be able to use one value to vacuum/cluster multiple tables (e.g. a toast's horizon shouldn't be different than the table's). Reported-By: Heikki Linnakangas Author: Andres Freund Discussion: https://postgr.es/m/9a7fb9cc-2419-5db7-8840-ddc10c93f122@iki.fi
2019-04-24 06:42:12 +02:00
* not needed for the relation's AM
tableam: comment improvements. Author: Brad DeJong Discussion: https://postgr.es/m/CAJnrtnxDYOQFsDfWz2iri0T_fFL2ZbbzgCOE=4yaMcszgcsf4A@mail.gmail.com Backpatch: 12-
2019-07-18 04:39:54 +02:00
* - *xid_cutoff - ditto
* - *multi_cutoff - ditto
tableam: relation creation, VACUUM FULL/CLUSTER, SET TABLESPACE. This moves the responsibility for: - creating the storage necessary for a relation, including creating a new relfilenode for a relation with existing storage - non-transactional truncation of a relation - VACUUM FULL / CLUSTER's rewrite of a table below tableam. This is fairly straight forward, with a bit of complexity smattered in to move the computation of xid / multixid horizons below the AM, as they don't make sense for every table AM. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-29 04:01:14 +01:00
*
Allow pg_class xid & multixid horizons to not be set. This allows table AMs that don't need these horizons. This was already documented in the tableam relation_set_new_filenode callback, but an assert prevented if from actually working (the test AM code contained the change itself). Defang the asserts in the general code, and move the stronger ones into heap AM. Relatedly, after CLUSTER/VACUUM, we'd always assign a relfrozenxid / relminmxid. Change the table_relation_copy_for_cluster() interface to allow the AM to overwrite the horizons that get set on the pg_class entry. This'd also in the future allow AMs like heap to compute a relfrozenxid during rewrite that's the table's actual minimum rather than a pre-determined value. Arguably it'd have been better to move the whole computation / setting of those values into the callback, but it seems likely that for other reasons it'd be better to be able to use one value to vacuum/cluster multiple tables (e.g. a toast's horizon shouldn't be different than the table's). Reported-By: Heikki Linnakangas Author: Andres Freund Discussion: https://postgr.es/m/9a7fb9cc-2419-5db7-8840-ddc10c93f122@iki.fi
2019-04-24 06:42:12 +02:00
* Output parameters:
* - *xid_cutoff - rel's new relfrozenxid value, may be invalid
* - *multi_cutoff - rel's new relminmxid value, may be invalid
* - *tups_vacuumed - stats, for logging, if appropriate for AM
* - *tups_recently_dead - stats, for logging, if appropriate for AM
tableam: relation creation, VACUUM FULL/CLUSTER, SET TABLESPACE. This moves the responsibility for: - creating the storage necessary for a relation, including creating a new relfilenode for a relation with existing storage - non-transactional truncation of a relation - VACUUM FULL / CLUSTER's rewrite of a table below tableam. This is fairly straight forward, with a bit of complexity smattered in to move the computation of xid / multixid horizons below the AM, as they don't make sense for every table AM. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-29 04:01:14 +01:00
*/
static inline void
Fix some typos and inconsistencies in tableam.h The defined callback definitions have been using references to heap for a couple of variables and comments. This makes the whole interface more consistent by using "table" which is more generic. A variable storing index information was misspelled as well. Author: Michael Paquier Discussion: https://postgr.es/m/20190601190946.GB1905@paquier.xyz
2019-06-04 02:48:25 +02:00
table_relation_copy_for_cluster(Relation OldTable, Relation NewTable,
tableam: relation creation, VACUUM FULL/CLUSTER, SET TABLESPACE. This moves the responsibility for: - creating the storage necessary for a relation, including creating a new relfilenode for a relation with existing storage - non-transactional truncation of a relation - VACUUM FULL / CLUSTER's rewrite of a table below tableam. This is fairly straight forward, with a bit of complexity smattered in to move the computation of xid / multixid horizons below the AM, as they don't make sense for every table AM. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-29 04:01:14 +01:00
Relation OldIndex,
bool use_sort,
TransactionId OldestXmin,
Allow pg_class xid & multixid horizons to not be set. This allows table AMs that don't need these horizons. This was already documented in the tableam relation_set_new_filenode callback, but an assert prevented if from actually working (the test AM code contained the change itself). Defang the asserts in the general code, and move the stronger ones into heap AM. Relatedly, after CLUSTER/VACUUM, we'd always assign a relfrozenxid / relminmxid. Change the table_relation_copy_for_cluster() interface to allow the AM to overwrite the horizons that get set on the pg_class entry. This'd also in the future allow AMs like heap to compute a relfrozenxid during rewrite that's the table's actual minimum rather than a pre-determined value. Arguably it'd have been better to move the whole computation / setting of those values into the callback, but it seems likely that for other reasons it'd be better to be able to use one value to vacuum/cluster multiple tables (e.g. a toast's horizon shouldn't be different than the table's). Reported-By: Heikki Linnakangas Author: Andres Freund Discussion: https://postgr.es/m/9a7fb9cc-2419-5db7-8840-ddc10c93f122@iki.fi
2019-04-24 06:42:12 +02:00
TransactionId *xid_cutoff,
MultiXactId *multi_cutoff,
tableam: relation creation, VACUUM FULL/CLUSTER, SET TABLESPACE. This moves the responsibility for: - creating the storage necessary for a relation, including creating a new relfilenode for a relation with existing storage - non-transactional truncation of a relation - VACUUM FULL / CLUSTER's rewrite of a table below tableam. This is fairly straight forward, with a bit of complexity smattered in to move the computation of xid / multixid horizons below the AM, as they don't make sense for every table AM. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-29 04:01:14 +01:00
double *num_tuples,
double *tups_vacuumed,
double *tups_recently_dead)
{
Fix some typos and inconsistencies in tableam.h The defined callback definitions have been using references to heap for a couple of variables and comments. This makes the whole interface more consistent by using "table" which is more generic. A variable storing index information was misspelled as well. Author: Michael Paquier Discussion: https://postgr.es/m/20190601190946.GB1905@paquier.xyz
2019-06-04 02:48:25 +02:00
OldTable->rd_tableam->relation_copy_for_cluster(OldTable, NewTable, OldIndex,
use_sort, OldestXmin,
xid_cutoff, multi_cutoff,
num_tuples, tups_vacuumed,
tups_recently_dead);
tableam: relation creation, VACUUM FULL/CLUSTER, SET TABLESPACE. This moves the responsibility for: - creating the storage necessary for a relation, including creating a new relfilenode for a relation with existing storage - non-transactional truncation of a relation - VACUUM FULL / CLUSTER's rewrite of a table below tableam. This is fairly straight forward, with a bit of complexity smattered in to move the computation of xid / multixid horizons below the AM, as they don't make sense for every table AM. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-29 04:01:14 +01:00
}
tableam: VACUUM and ANALYZE support. This is a relatively straightforward move of the current implementation to sit below tableam. As the current analyze sampling implementation is pretty inherently block based, the tableam analyze interface is as well. It might make sense to generalize that at some point, but that seems like a larger project that shouldn't be undertaken at the same time as the introduction of tableam. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-31 00:21:09 +01:00
/*
tableam: comment improvements. Author: Brad DeJong Discussion: https://postgr.es/m/CAJnrtnxDYOQFsDfWz2iri0T_fFL2ZbbzgCOE=4yaMcszgcsf4A@mail.gmail.com Backpatch: 12-
2019-07-18 04:39:54 +02:00
* Perform VACUUM on the relation. The VACUUM can be triggered by a user or by
tableam: VACUUM and ANALYZE support. This is a relatively straightforward move of the current implementation to sit below tableam. As the current analyze sampling implementation is pretty inherently block based, the tableam analyze interface is as well. It might make sense to generalize that at some point, but that seems like a larger project that shouldn't be undertaken at the same time as the introduction of tableam. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-31 00:21:09 +01:00
* autovacuum. The specific actions performed by the AM will depend heavily on
* the individual AM.
tableam: comment improvements. Author: Brad DeJong Discussion: https://postgr.es/m/CAJnrtnxDYOQFsDfWz2iri0T_fFL2ZbbzgCOE=4yaMcszgcsf4A@mail.gmail.com Backpatch: 12-
2019-07-18 04:39:54 +02:00
*
tableam: VACUUM and ANALYZE support. This is a relatively straightforward move of the current implementation to sit below tableam. As the current analyze sampling implementation is pretty inherently block based, the tableam analyze interface is as well. It might make sense to generalize that at some point, but that seems like a larger project that shouldn't be undertaken at the same time as the introduction of tableam. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-31 00:21:09 +01:00
* On entry a transaction needs to already been established, and the
tableam: comment and formatting fixes. Author: Heikki Linnakangas Discussion: https://postgr.es/m/9a7fb9cc-2419-5db7-8840-ddc10c93f122@iki.fi
2019-04-08 19:25:16 +02:00
* table is locked with a ShareUpdateExclusive lock.
tableam: VACUUM and ANALYZE support. This is a relatively straightforward move of the current implementation to sit below tableam. As the current analyze sampling implementation is pretty inherently block based, the tableam analyze interface is as well. It might make sense to generalize that at some point, but that seems like a larger project that shouldn't be undertaken at the same time as the introduction of tableam. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-31 00:21:09 +01:00
*
* Note that neither VACUUM FULL (and CLUSTER), nor ANALYZE go through this
tableam: comment and formatting fixes. Author: Heikki Linnakangas Discussion: https://postgr.es/m/9a7fb9cc-2419-5db7-8840-ddc10c93f122@iki.fi
2019-04-08 19:25:16 +02:00
* routine, even if (for ANALYZE) it is part of the same VACUUM command.
tableam: VACUUM and ANALYZE support. This is a relatively straightforward move of the current implementation to sit below tableam. As the current analyze sampling implementation is pretty inherently block based, the tableam analyze interface is as well. It might make sense to generalize that at some point, but that seems like a larger project that shouldn't be undertaken at the same time as the introduction of tableam. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-31 00:21:09 +01:00
*/
static inline void
table_relation_vacuum(Relation rel, struct VacuumParams *params,
BufferAccessStrategy bstrategy)
{
rel->rd_tableam->relation_vacuum(rel, params, bstrategy);
}
/*
* Prepare to analyze block `blockno` of `scan`. The scan needs to have been
* started with table_beginscan_analyze(). Note that this routine might
* acquire resources like locks that are held until
* table_scan_analyze_next_tuple() returns false.
*
* Returns false if block is unsuitable for sampling, true otherwise.
*/
static inline bool
table_scan_analyze_next_block(TableScanDesc scan, BlockNumber blockno,
BufferAccessStrategy bstrategy)
{
return scan->rs_rd->rd_tableam->scan_analyze_next_block(scan, blockno,
bstrategy);
}
/*
tableam: comment and formatting fixes. Author: Heikki Linnakangas Discussion: https://postgr.es/m/9a7fb9cc-2419-5db7-8840-ddc10c93f122@iki.fi
2019-04-08 19:25:16 +02:00
* Iterate over tuples in the block selected with
tableam: VACUUM and ANALYZE support. This is a relatively straightforward move of the current implementation to sit below tableam. As the current analyze sampling implementation is pretty inherently block based, the tableam analyze interface is as well. It might make sense to generalize that at some point, but that seems like a larger project that shouldn't be undertaken at the same time as the introduction of tableam. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-31 00:21:09 +01:00
* table_scan_analyze_next_block() (which needs to have returned true, and
* this routine may not have returned false for the same block before). If a
* tuple that's suitable for sampling is found, true is returned and a tuple
* is stored in `slot`.
*
* *liverows and *deadrows are incremented according to the encountered
* tuples.
*/
static inline bool
table_scan_analyze_next_tuple(TableScanDesc scan, TransactionId OldestXmin,
double *liverows, double *deadrows,
TupleTableSlot *slot)
{
return scan->rs_rd->rd_tableam->scan_analyze_next_tuple(scan, OldestXmin,
liverows, deadrows,
slot);
}
tableam: Support for an index build's initial table scan(s). To support building indexes over tables of different AMs, the scans to do so need to be routed through the table AM. While moving a fair amount of code, nearly all the changes are just moving code to below a callback. Currently the range based interface wouldn't make much sense for non block based table AMs. But that seems aceptable for now. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-28 03:59:06 +01:00
/*
tableam: comment and formatting fixes. Author: Heikki Linnakangas Discussion: https://postgr.es/m/9a7fb9cc-2419-5db7-8840-ddc10c93f122@iki.fi
2019-04-08 19:25:16 +02:00
* table_index_build_scan - scan the table to find tuples to be indexed
tableam: Support for an index build's initial table scan(s). To support building indexes over tables of different AMs, the scans to do so need to be routed through the table AM. While moving a fair amount of code, nearly all the changes are just moving code to below a callback. Currently the range based interface wouldn't make much sense for non block based table AMs. But that seems aceptable for now. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-28 03:59:06 +01:00
*
* This is called back from an access-method-specific index build procedure
Fix some typos and inconsistencies in tableam.h The defined callback definitions have been using references to heap for a couple of variables and comments. This makes the whole interface more consistent by using "table" which is more generic. A variable storing index information was misspelled as well. Author: Michael Paquier Discussion: https://postgr.es/m/20190601190946.GB1905@paquier.xyz
2019-06-04 02:48:25 +02:00
* after the AM has done whatever setup it needs. The parent table relation
tableam: Support for an index build's initial table scan(s). To support building indexes over tables of different AMs, the scans to do so need to be routed through the table AM. While moving a fair amount of code, nearly all the changes are just moving code to below a callback. Currently the range based interface wouldn't make much sense for non block based table AMs. But that seems aceptable for now. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-28 03:59:06 +01:00
* is scanned to find tuples that should be entered into the index. Each
* such tuple is passed to the AM's callback routine, which does the right
* things to add it to the new index. After we return, the AM's index
* build procedure does whatever cleanup it needs.
*
* The total count of live tuples is returned. This is for updating pg_class
* statistics. (It's annoying not to be able to do that here, but we want to
* merge that update with others; see index_update_stats.) Note that the
* index AM itself must keep track of the number of index tuples; we don't do
* so here because the AM might reject some of the tuples for its own reasons,
* such as being unable to store NULLs.
*
Report progress of CREATE INDEX operations This uses the progress reporting infrastructure added by c16dc1aca5e0, adding support for CREATE INDEX and CREATE INDEX CONCURRENTLY. There are two pieces to this: one is index-AM-agnostic, and the other is AM-specific. The latter is fairly elaborate for btrees, including reportage for parallel index builds and the separate phases that btree index creation uses; other index AMs, which are much simpler in their building procedures, have simplistic reporting only, but that seems sufficient, at least for non-concurrent builds. The index-AM-agnostic part is fairly complete, providing insight into the CONCURRENTLY wait phases as well as block-based progress during the index validation table scan. (The index validation index scan requires patching each AM, which has not been included here.) Reviewers: Rahila Syed, Pavan Deolasee, Tatsuro Yamada Discussion: https://postgr.es/m/20181220220022.mg63bhk26zdpvmcj@alvherre.pgsql
2019-04-02 20:18:08 +02:00
* If 'progress', the PROGRESS_SCAN_BLOCKS_TOTAL counter is updated when
* starting the scan, and PROGRESS_SCAN_BLOCKS_DONE is updated as we go along.
tableam: Support for an index build's initial table scan(s). To support building indexes over tables of different AMs, the scans to do so need to be routed through the table AM. While moving a fair amount of code, nearly all the changes are just moving code to below a callback. Currently the range based interface wouldn't make much sense for non block based table AMs. But that seems aceptable for now. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-28 03:59:06 +01:00
*
* A side effect is to set indexInfo->ii_BrokenHotChain to true if we detect
* any potentially broken HOT chains. Currently, we set this if there are any
* RECENTLY_DEAD or DELETE_IN_PROGRESS entries in a HOT chain, without trying
* very hard to detect whether they're really incompatible with the chain tip.
* This only really makes sense for heap AM, it might need to be generalized
* for other AMs later.
*/
static inline double
Fix some typos and inconsistencies in tableam.h The defined callback definitions have been using references to heap for a couple of variables and comments. This makes the whole interface more consistent by using "table" which is more generic. A variable storing index information was misspelled as well. Author: Michael Paquier Discussion: https://postgr.es/m/20190601190946.GB1905@paquier.xyz
2019-06-04 02:48:25 +02:00
table_index_build_scan(Relation table_rel,
tableam: Support for an index build's initial table scan(s). To support building indexes over tables of different AMs, the scans to do so need to be routed through the table AM. While moving a fair amount of code, nearly all the changes are just moving code to below a callback. Currently the range based interface wouldn't make much sense for non block based table AMs. But that seems aceptable for now. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-28 03:59:06 +01:00
Relation index_rel,
Fix some typos and inconsistencies in tableam.h The defined callback definitions have been using references to heap for a couple of variables and comments. This makes the whole interface more consistent by using "table" which is more generic. A variable storing index information was misspelled as well. Author: Michael Paquier Discussion: https://postgr.es/m/20190601190946.GB1905@paquier.xyz
2019-06-04 02:48:25 +02:00
struct IndexInfo *index_info,
tableam: Support for an index build's initial table scan(s). To support building indexes over tables of different AMs, the scans to do so need to be routed through the table AM. While moving a fair amount of code, nearly all the changes are just moving code to below a callback. Currently the range based interface wouldn't make much sense for non block based table AMs. But that seems aceptable for now. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-28 03:59:06 +01:00
bool allow_sync,
Report progress of CREATE INDEX operations This uses the progress reporting infrastructure added by c16dc1aca5e0, adding support for CREATE INDEX and CREATE INDEX CONCURRENTLY. There are two pieces to this: one is index-AM-agnostic, and the other is AM-specific. The latter is fairly elaborate for btrees, including reportage for parallel index builds and the separate phases that btree index creation uses; other index AMs, which are much simpler in their building procedures, have simplistic reporting only, but that seems sufficient, at least for non-concurrent builds. The index-AM-agnostic part is fairly complete, providing insight into the CONCURRENTLY wait phases as well as block-based progress during the index validation table scan. (The index validation index scan requires patching each AM, which has not been included here.) Reviewers: Rahila Syed, Pavan Deolasee, Tatsuro Yamada Discussion: https://postgr.es/m/20181220220022.mg63bhk26zdpvmcj@alvherre.pgsql
2019-04-02 20:18:08 +02:00
bool progress,
tableam: Support for an index build's initial table scan(s). To support building indexes over tables of different AMs, the scans to do so need to be routed through the table AM. While moving a fair amount of code, nearly all the changes are just moving code to below a callback. Currently the range based interface wouldn't make much sense for non block based table AMs. But that seems aceptable for now. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-28 03:59:06 +01:00
IndexBuildCallback callback,
void *callback_state,
TableScanDesc scan)
{
Fix some typos and inconsistencies in tableam.h The defined callback definitions have been using references to heap for a couple of variables and comments. This makes the whole interface more consistent by using "table" which is more generic. A variable storing index information was misspelled as well. Author: Michael Paquier Discussion: https://postgr.es/m/20190601190946.GB1905@paquier.xyz
2019-06-04 02:48:25 +02:00
return table_rel->rd_tableam->index_build_range_scan(table_rel,
index_rel,
index_info,
allow_sync,
false,
progress,
0,
InvalidBlockNumber,
callback,
callback_state,
scan);
tableam: Support for an index build's initial table scan(s). To support building indexes over tables of different AMs, the scans to do so need to be routed through the table AM. While moving a fair amount of code, nearly all the changes are just moving code to below a callback. Currently the range based interface wouldn't make much sense for non block based table AMs. But that seems aceptable for now. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-28 03:59:06 +01:00
}
/*
* As table_index_build_scan(), except that instead of scanning the complete
* table, only the given number of blocks are scanned. Scan to end-of-rel can
Spelling adjustments
2020-06-07 15:06:51 +02:00
* be signaled by passing InvalidBlockNumber as numblocks. Note that
tableam: Support for an index build's initial table scan(s). To support building indexes over tables of different AMs, the scans to do so need to be routed through the table AM. While moving a fair amount of code, nearly all the changes are just moving code to below a callback. Currently the range based interface wouldn't make much sense for non block based table AMs. But that seems aceptable for now. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-28 03:59:06 +01:00
* restricting the range to scan cannot be done when requesting syncscan.
*
* When "anyvisible" mode is requested, all tuples visible to any transaction
* are indexed and counted as live, including those inserted or deleted by
* transactions that are still in progress.
*/
static inline double
Fix some typos and inconsistencies in tableam.h The defined callback definitions have been using references to heap for a couple of variables and comments. This makes the whole interface more consistent by using "table" which is more generic. A variable storing index information was misspelled as well. Author: Michael Paquier Discussion: https://postgr.es/m/20190601190946.GB1905@paquier.xyz
2019-06-04 02:48:25 +02:00
table_index_build_range_scan(Relation table_rel,
tableam: Support for an index build's initial table scan(s). To support building indexes over tables of different AMs, the scans to do so need to be routed through the table AM. While moving a fair amount of code, nearly all the changes are just moving code to below a callback. Currently the range based interface wouldn't make much sense for non block based table AMs. But that seems aceptable for now. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-28 03:59:06 +01:00
Relation index_rel,
Fix some typos and inconsistencies in tableam.h The defined callback definitions have been using references to heap for a couple of variables and comments. This makes the whole interface more consistent by using "table" which is more generic. A variable storing index information was misspelled as well. Author: Michael Paquier Discussion: https://postgr.es/m/20190601190946.GB1905@paquier.xyz
2019-06-04 02:48:25 +02:00
struct IndexInfo *index_info,
tableam: Support for an index build's initial table scan(s). To support building indexes over tables of different AMs, the scans to do so need to be routed through the table AM. While moving a fair amount of code, nearly all the changes are just moving code to below a callback. Currently the range based interface wouldn't make much sense for non block based table AMs. But that seems aceptable for now. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-28 03:59:06 +01:00
bool allow_sync,
bool anyvisible,
Report progress of CREATE INDEX operations This uses the progress reporting infrastructure added by c16dc1aca5e0, adding support for CREATE INDEX and CREATE INDEX CONCURRENTLY. There are two pieces to this: one is index-AM-agnostic, and the other is AM-specific. The latter is fairly elaborate for btrees, including reportage for parallel index builds and the separate phases that btree index creation uses; other index AMs, which are much simpler in their building procedures, have simplistic reporting only, but that seems sufficient, at least for non-concurrent builds. The index-AM-agnostic part is fairly complete, providing insight into the CONCURRENTLY wait phases as well as block-based progress during the index validation table scan. (The index validation index scan requires patching each AM, which has not been included here.) Reviewers: Rahila Syed, Pavan Deolasee, Tatsuro Yamada Discussion: https://postgr.es/m/20181220220022.mg63bhk26zdpvmcj@alvherre.pgsql
2019-04-02 20:18:08 +02:00
bool progress,
tableam: Support for an index build's initial table scan(s). To support building indexes over tables of different AMs, the scans to do so need to be routed through the table AM. While moving a fair amount of code, nearly all the changes are just moving code to below a callback. Currently the range based interface wouldn't make much sense for non block based table AMs. But that seems aceptable for now. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-28 03:59:06 +01:00
BlockNumber start_blockno,
BlockNumber numblocks,
IndexBuildCallback callback,
void *callback_state,
TableScanDesc scan)
{
Fix some typos and inconsistencies in tableam.h The defined callback definitions have been using references to heap for a couple of variables and comments. This makes the whole interface more consistent by using "table" which is more generic. A variable storing index information was misspelled as well. Author: Michael Paquier Discussion: https://postgr.es/m/20190601190946.GB1905@paquier.xyz
2019-06-04 02:48:25 +02:00
return table_rel->rd_tableam->index_build_range_scan(table_rel,
index_rel,
index_info,
allow_sync,
anyvisible,
progress,
start_blockno,
numblocks,
callback,
callback_state,
scan);
tableam: Support for an index build's initial table scan(s). To support building indexes over tables of different AMs, the scans to do so need to be routed through the table AM. While moving a fair amount of code, nearly all the changes are just moving code to below a callback. Currently the range based interface wouldn't make much sense for non block based table AMs. But that seems aceptable for now. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-28 03:59:06 +01:00
}
/*
* table_index_validate_scan - second table scan for concurrent index build
*
* See validate_index() for an explanation.
*/
static inline void
Fix some typos and inconsistencies in tableam.h The defined callback definitions have been using references to heap for a couple of variables and comments. This makes the whole interface more consistent by using "table" which is more generic. A variable storing index information was misspelled as well. Author: Michael Paquier Discussion: https://postgr.es/m/20190601190946.GB1905@paquier.xyz
2019-06-04 02:48:25 +02:00
table_index_validate_scan(Relation table_rel,
tableam: Support for an index build's initial table scan(s). To support building indexes over tables of different AMs, the scans to do so need to be routed through the table AM. While moving a fair amount of code, nearly all the changes are just moving code to below a callback. Currently the range based interface wouldn't make much sense for non block based table AMs. But that seems aceptable for now. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-28 03:59:06 +01:00
Relation index_rel,
struct IndexInfo *index_info,
Snapshot snapshot,
struct ValidateIndexState *state)
{
Fix some typos and inconsistencies in tableam.h The defined callback definitions have been using references to heap for a couple of variables and comments. This makes the whole interface more consistent by using "table" which is more generic. A variable storing index information was misspelled as well. Author: Michael Paquier Discussion: https://postgr.es/m/20190601190946.GB1905@paquier.xyz
2019-06-04 02:48:25 +02:00
table_rel->rd_tableam->index_validate_scan(table_rel,
index_rel,
index_info,
snapshot,
state);
tableam: Support for an index build's initial table scan(s). To support building indexes over tables of different AMs, the scans to do so need to be routed through the table AM. While moving a fair amount of code, nearly all the changes are just moving code to below a callback. Currently the range based interface wouldn't make much sense for non block based table AMs. But that seems aceptable for now. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-28 03:59:06 +01:00
}
tableam: Don't assume that every AM uses md.c style storage. Previously various parts of the code routed size requests through RelationGetNumberOfBlocks[InFork]. That works if md.c is used by the AM, but not otherwise. Add a tableam callback to return the size of the table. As not every AM will use postgres' BLCKSZ, have it return bytes, and have RelationGetNumberOfBlocksInFork() round the byte size up into blocks. To allow code outside of the AM to determine the actual relation size map InvalidForkNumber the total size of a relation, as not every AM might just need the postgres defined forks. A few users of RelationGetNumberOfBlocks() ought to be converted away from that. One case, the use of it to determine whether a tid is valid, will be fixed in a follow up commit. Others will have to wait for v13. Author: Andres Freund Discussion: https://postgr.es/m/20190423225201.3bbv6tbqzkb5w7cw@alap3.anarazel.de
2019-05-18 03:06:18 +02:00
/* ----------------------------------------------------------------------------
* Miscellaneous functionality
* ----------------------------------------------------------------------------
*/
/*
* Return the current size of `rel` in bytes. If `forkNumber` is
* InvalidForkNumber, return the relation's overall size, otherwise the size
* for the indicated fork.
*
* Note that the overall size might not be the equivalent of the sum of sizes
* for the individual forks for some AMs, e.g. because the AMs storage does
* not neatly map onto the builtin types of forks.
*/
static inline uint64
table_relation_size(Relation rel, ForkNumber forkNumber)
{
return rel->rd_tableam->relation_size(rel, forkNumber);
}
tableam: Move heap-specific logic from needs_toast_table below tableam. This allows table AMs to completely suppress TOAST table creation, or to modify the conditions under which they are created. Patch by me. Reviewed by Andres Freund. Discussion: http://postgr.es/m/CA+Tgmoa4O2n=yphqD2pERUnYmUO84bH1SqMsA-nSxBGsZ7gWfA@mail.gmail.com
2019-05-21 17:57:13 +02:00
/*
Fix assorted inconsistencies. There were a number of issues in the recent commits which include typos, code and comments mismatch, leftover function declarations. Fix them. Reported-by: Alexander Lakhin Author: Alexander Lakhin, Amit Kapila and Amit Langote Reviewed-by: Amit Kapila Discussion: https://postgr.es/m/ef0c0232-0c1d-3a35-63d4-0ebd06e31387@gmail.com
2019-06-08 04:46:38 +02:00
* table_relation_needs_toast_table - does this relation need a toast table?
tableam: Move heap-specific logic from needs_toast_table below tableam. This allows table AMs to completely suppress TOAST table creation, or to modify the conditions under which they are created. Patch by me. Reviewed by Andres Freund. Discussion: http://postgr.es/m/CA+Tgmoa4O2n=yphqD2pERUnYmUO84bH1SqMsA-nSxBGsZ7gWfA@mail.gmail.com
2019-05-21 17:57:13 +02:00
*/
static inline bool
table_relation_needs_toast_table(Relation rel)
{
return rel->rd_tableam->relation_needs_toast_table(rel);
}
tableam: Allow choice of toast AM. Previously, the toast table had to be implemented by the same AM that was used for the main table, which was bad, because the detoasting code won't work with anything but heap. This commit doesn't fix the latter problem, although there's another patch coming which does, but it does let you pick something that works (i.e. heap, right now). Patch by me, reviewed by Andres Freund. Discussion: http://postgr.es/m/CA+TgmoZv-=2iWM4jcw5ZhJeL18HF96+W1yJeYrnGMYdkFFnEpQ@mail.gmail.com
2020-01-07 20:23:25 +01:00
/*
* Return the OID of the AM that should be used to implement the TOAST table
* for this relation.
*/
static inline Oid
table_relation_toast_am(Relation rel)
{
return rel->rd_tableam->relation_toast_am(rel);
}
tableam: New callback relation_fetch_toast_slice. Instead of always calling heap_fetch_toast_slice during detoasting, invoke a table AM callback which, when the toast table is a heap table, will be heap_fetch_toast_slice. This makes it possible for a table AM other than heap to be used as a TOAST table. It also completes the series of commits intended to improve the interaction of tableam with TOAST that began with commit 8b94dab06617ef80a0901ab103ebd8754427ef5a; detoast.c is now, hopefully, fully AM-independent. Patch by me, reviewed by Andres Freund and Peter Eisentraut. Discussion: http://postgr.es/m/CA+TgmoZv-=2iWM4jcw5ZhJeL18HF96+W1yJeYrnGMYdkFFnEpQ@mail.gmail.com
2020-01-07 20:35:48 +01:00
/*
* Fetch all or part of a TOAST value from a TOAST table.
*
* If this AM is never used to implement a TOAST table, then this callback
* is not needed. But, if toasted values are ever stored in a table of this
* type, then you will need this callback.
*
* toastrel is the relation in which the toasted value is stored.
*
* valueid identifes which toast value is to be fetched. For the heap,
* this corresponds to the values stored in the chunk_id column.
*
* attrsize is the total size of the toast value to be fetched.
*
* sliceoffset is the offset within the toast value of the first byte that
* should be fetched.
*
* slicelength is the number of bytes from the toast value that should be
* fetched.
*
* result is caller-allocated space into which the fetched bytes should be
* stored.
*/
static inline void
table_relation_fetch_toast_slice(Relation toastrel, Oid valueid,
int32 attrsize, int32 sliceoffset,
int32 slicelength, struct varlena *result)
{
Remove bogus 'return'. Per the buildfarm, via Michael Paquier. Discussion: http://postgr.es/m/20200108032648.GE3413@paquier.xyz
2020-01-09 15:01:37 +01:00
toastrel->rd_tableam->relation_fetch_toast_slice(toastrel, valueid,
attrsize,
sliceoffset, slicelength,
result);
tableam: New callback relation_fetch_toast_slice. Instead of always calling heap_fetch_toast_slice during detoasting, invoke a table AM callback which, when the toast table is a heap table, will be heap_fetch_toast_slice. This makes it possible for a table AM other than heap to be used as a TOAST table. It also completes the series of commits intended to improve the interaction of tableam with TOAST that began with commit 8b94dab06617ef80a0901ab103ebd8754427ef5a; detoast.c is now, hopefully, fully AM-independent. Patch by me, reviewed by Andres Freund and Peter Eisentraut. Discussion: http://postgr.es/m/CA+TgmoZv-=2iWM4jcw5ZhJeL18HF96+W1yJeYrnGMYdkFFnEpQ@mail.gmail.com
2020-01-07 20:35:48 +01:00
}
tableam: Move heap-specific logic from needs_toast_table below tableam. This allows table AMs to completely suppress TOAST table creation, or to modify the conditions under which they are created. Patch by me. Reviewed by Andres Freund. Discussion: http://postgr.es/m/CA+Tgmoa4O2n=yphqD2pERUnYmUO84bH1SqMsA-nSxBGsZ7gWfA@mail.gmail.com
2019-05-21 17:57:13 +02:00
tableam: Move heap specific logic from estimate_rel_size below tableam. This just moves the table/matview[/toast] determination of relation size to a callback, and uses a copy of the existing logic to implement that callback for heap. It probably would make sense to also move the index specific logic into a callback, so the metapage handling (and probably more) can be index specific. But that's a separate task. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-31 00:40:33 +01:00
/* ----------------------------------------------------------------------------
* Planner related functionality
* ----------------------------------------------------------------------------
*/
/*
* Estimate the current size of the relation, as an AM specific workhorse for
* estimate_rel_size(). Look there for an explanation of the parameters.
*/
static inline void
table_relation_estimate_size(Relation rel, int32 *attr_widths,
BlockNumber *pages, double *tuples,
double *allvisfrac)
{
rel->rd_tableam->relation_estimate_size(rel, attr_widths, pages, tuples,
allvisfrac);
}
tableam: sample scan. This moves sample scan support to below tableam. It's not optional as there is, in contrast to e.g. bitmap heap scans, no alternative way to perform tablesample queries. If an AM can't deal with the block based API, it will have to throw an ERROR. The tableam callbacks for this are block based, but given the current TsmRoutine interface, that seems to be required. The new interface doesn't require TsmRoutines to perform visibility checks anymore - that requires the TsmRoutine to know details about the AM, which we want to avoid. To continue to allow taking the returned number of tuples account SampleScanState now has a donetuples field (which previously e.g. existed in SystemRowsSamplerData), which is only incremented after the visibility check succeeds. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-31 05:18:53 +02:00
/* ----------------------------------------------------------------------------
* Executor related functionality
* ----------------------------------------------------------------------------
*/
/*
tableam: bitmap table scan. This moves bitmap heap scan support to below an optional tableam callback. It's optional as the whole concept of bitmap heapscans is fairly block specific. This basically moves the work previously done in bitgetpage() into the new scan_bitmap_next_block callback, and the direct poking into the buffer done in BitmapHeapNext() into the new scan_bitmap_next_tuple() callback. The abstraction is currently somewhat leaky because nodeBitmapHeapscan.c's prefetching and visibilitymap based logic remains - it's likely that we'll later have to move more into the AM. But it's not trivial to do so without introducing a significant amount of code duplication between the AMs, so that's a project for later. Note that now nodeBitmapHeapscan.c and the associated node types are a bit misnamed. But it's not clear whether renaming wouldn't be a cure worse than the disease. Either way, that'd be best done in a separate commit. Author: Andres Freund Reviewed-By: Robert Haas (in an older version) Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-04-01 02:51:49 +02:00
* Prepare to fetch / check / return tuples from `tbmres->blockno` as part of
* a bitmap table scan. `scan` needs to have been started via
tableam: comment and formatting fixes. Author: Heikki Linnakangas Discussion: https://postgr.es/m/9a7fb9cc-2419-5db7-8840-ddc10c93f122@iki.fi
2019-04-08 19:25:16 +02:00
* table_beginscan_bm(). Returns false if there are no tuples to be found on
* the page, true otherwise.
tableam: bitmap table scan. This moves bitmap heap scan support to below an optional tableam callback. It's optional as the whole concept of bitmap heapscans is fairly block specific. This basically moves the work previously done in bitgetpage() into the new scan_bitmap_next_block callback, and the direct poking into the buffer done in BitmapHeapNext() into the new scan_bitmap_next_tuple() callback. The abstraction is currently somewhat leaky because nodeBitmapHeapscan.c's prefetching and visibilitymap based logic remains - it's likely that we'll later have to move more into the AM. But it's not trivial to do so without introducing a significant amount of code duplication between the AMs, so that's a project for later. Note that now nodeBitmapHeapscan.c and the associated node types are a bit misnamed. But it's not clear whether renaming wouldn't be a cure worse than the disease. Either way, that'd be best done in a separate commit. Author: Andres Freund Reviewed-By: Robert Haas (in an older version) Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-04-01 02:51:49 +02:00
*
* Note, this is an optionally implemented function, therefore should only be
* used after verifying the presence (at plan time or such).
*/
static inline bool
table_scan_bitmap_next_block(TableScanDesc scan,
struct TBMIterateResult *tbmres)
{
Implement streaming mode in ReorderBuffer. Instead of serializing the transaction to disk after reaching the logical_decoding_work_mem limit in memory, we consume the changes we have in memory and invoke stream API methods added by commit 45fdc9738b. However, sometimes if we have incomplete toast or speculative insert we spill to the disk because we can't generate the complete tuple and stream. And, as soon as we get the complete tuple we stream the transaction including the serialized changes. We can do this incremental processing thanks to having assignments (associating subxact with toplevel xacts) in WAL right away, and thanks to logging the invalidation messages at each command end. These features are added by commits 0bead9af48 and c55040ccd0 respectively. Now that we can stream in-progress transactions, the concurrent aborts may cause failures when the output plugin consults catalogs (both system and user-defined). We handle such failures by returning ERRCODE_TRANSACTION_ROLLBACK sqlerrcode from system table scan APIs to the backend or WALSender decoding a specific uncommitted transaction. The decoding logic on the receipt of such a sqlerrcode aborts the decoding of the current transaction and continue with the decoding of other transactions. We have ReorderBufferTXN pointer in each ReorderBufferChange by which we know which xact it belongs to. The output plugin can use this to decide which changes to discard in case of stream_abort_cb (e.g. when a subxact gets discarded). We also provide a new option via SQL APIs to fetch the changes being streamed. Author: Dilip Kumar, Tomas Vondra, Amit Kapila, Nikhil Sontakke Reviewed-by: Amit Kapila, Kuntal Ghosh, Ajin Cherian Tested-by: Neha Sharma, Mahendra Singh Thalor and Ajin Cherian Discussion: https://postgr.es/m/688b0b7f-2f6c-d827-c27b-216a8e3ea700@2ndquadrant.com
2020-08-08 04:04:39 +02:00
/*
* We don't expect direct calls to table_scan_bitmap_next_block with valid
* CheckXidAlive for catalog or regular tables. See detailed comments in
* xact.c where these variables are declared.
*/
if (unlikely(TransactionIdIsValid(CheckXidAlive) && !bsysscan))
elog(ERROR, "unexpected table_scan_bitmap_next_block call during logical decoding");
tableam: bitmap table scan. This moves bitmap heap scan support to below an optional tableam callback. It's optional as the whole concept of bitmap heapscans is fairly block specific. This basically moves the work previously done in bitgetpage() into the new scan_bitmap_next_block callback, and the direct poking into the buffer done in BitmapHeapNext() into the new scan_bitmap_next_tuple() callback. The abstraction is currently somewhat leaky because nodeBitmapHeapscan.c's prefetching and visibilitymap based logic remains - it's likely that we'll later have to move more into the AM. But it's not trivial to do so without introducing a significant amount of code duplication between the AMs, so that's a project for later. Note that now nodeBitmapHeapscan.c and the associated node types are a bit misnamed. But it's not clear whether renaming wouldn't be a cure worse than the disease. Either way, that'd be best done in a separate commit. Author: Andres Freund Reviewed-By: Robert Haas (in an older version) Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-04-01 02:51:49 +02:00
return scan->rs_rd->rd_tableam->scan_bitmap_next_block(scan,
tbmres);
}
/*
* Fetch the next tuple of a bitmap table scan into `slot` and return true if
* a visible tuple was found, false otherwise.
* table_scan_bitmap_next_block() needs to previously have selected a
* block (i.e. returned true), and no previous
* table_scan_bitmap_next_tuple() for the same block may have
* returned false.
*/
static inline bool
table_scan_bitmap_next_tuple(TableScanDesc scan,
struct TBMIterateResult *tbmres,
TupleTableSlot *slot)
{
Implement streaming mode in ReorderBuffer. Instead of serializing the transaction to disk after reaching the logical_decoding_work_mem limit in memory, we consume the changes we have in memory and invoke stream API methods added by commit 45fdc9738b. However, sometimes if we have incomplete toast or speculative insert we spill to the disk because we can't generate the complete tuple and stream. And, as soon as we get the complete tuple we stream the transaction including the serialized changes. We can do this incremental processing thanks to having assignments (associating subxact with toplevel xacts) in WAL right away, and thanks to logging the invalidation messages at each command end. These features are added by commits 0bead9af48 and c55040ccd0 respectively. Now that we can stream in-progress transactions, the concurrent aborts may cause failures when the output plugin consults catalogs (both system and user-defined). We handle such failures by returning ERRCODE_TRANSACTION_ROLLBACK sqlerrcode from system table scan APIs to the backend or WALSender decoding a specific uncommitted transaction. The decoding logic on the receipt of such a sqlerrcode aborts the decoding of the current transaction and continue with the decoding of other transactions. We have ReorderBufferTXN pointer in each ReorderBufferChange by which we know which xact it belongs to. The output plugin can use this to decide which changes to discard in case of stream_abort_cb (e.g. when a subxact gets discarded). We also provide a new option via SQL APIs to fetch the changes being streamed. Author: Dilip Kumar, Tomas Vondra, Amit Kapila, Nikhil Sontakke Reviewed-by: Amit Kapila, Kuntal Ghosh, Ajin Cherian Tested-by: Neha Sharma, Mahendra Singh Thalor and Ajin Cherian Discussion: https://postgr.es/m/688b0b7f-2f6c-d827-c27b-216a8e3ea700@2ndquadrant.com
2020-08-08 04:04:39 +02:00
/*
* We don't expect direct calls to table_scan_bitmap_next_tuple with valid
* CheckXidAlive for catalog or regular tables. See detailed comments in
* xact.c where these variables are declared.
*/
if (unlikely(TransactionIdIsValid(CheckXidAlive) && !bsysscan))
elog(ERROR, "unexpected table_scan_bitmap_next_tuple call during logical decoding");
tableam: bitmap table scan. This moves bitmap heap scan support to below an optional tableam callback. It's optional as the whole concept of bitmap heapscans is fairly block specific. This basically moves the work previously done in bitgetpage() into the new scan_bitmap_next_block callback, and the direct poking into the buffer done in BitmapHeapNext() into the new scan_bitmap_next_tuple() callback. The abstraction is currently somewhat leaky because nodeBitmapHeapscan.c's prefetching and visibilitymap based logic remains - it's likely that we'll later have to move more into the AM. But it's not trivial to do so without introducing a significant amount of code duplication between the AMs, so that's a project for later. Note that now nodeBitmapHeapscan.c and the associated node types are a bit misnamed. But it's not clear whether renaming wouldn't be a cure worse than the disease. Either way, that'd be best done in a separate commit. Author: Andres Freund Reviewed-By: Robert Haas (in an older version) Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-04-01 02:51:49 +02:00
return scan->rs_rd->rd_tableam->scan_bitmap_next_tuple(scan,
tbmres,
slot);
}
/*
* Prepare to fetch tuples from the next block in a sample scan. Returns false
* if the sample scan is finished, true otherwise. `scan` needs to have been
* started via table_beginscan_sampling().
tableam: sample scan. This moves sample scan support to below tableam. It's not optional as there is, in contrast to e.g. bitmap heap scans, no alternative way to perform tablesample queries. If an AM can't deal with the block based API, it will have to throw an ERROR. The tableam callbacks for this are block based, but given the current TsmRoutine interface, that seems to be required. The new interface doesn't require TsmRoutines to perform visibility checks anymore - that requires the TsmRoutine to know details about the AM, which we want to avoid. To continue to allow taking the returned number of tuples account SampleScanState now has a donetuples field (which previously e.g. existed in SystemRowsSamplerData), which is only incremented after the visibility check succeeds. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-31 05:18:53 +02:00
*
* This will call the TsmRoutine's NextSampleBlock() callback if necessary
* (i.e. NextSampleBlock is not NULL), or perform a sequential scan over the
* underlying relation.
*/
static inline bool
table_scan_sample_next_block(TableScanDesc scan,
struct SampleScanState *scanstate)
{
Implement streaming mode in ReorderBuffer. Instead of serializing the transaction to disk after reaching the logical_decoding_work_mem limit in memory, we consume the changes we have in memory and invoke stream API methods added by commit 45fdc9738b. However, sometimes if we have incomplete toast or speculative insert we spill to the disk because we can't generate the complete tuple and stream. And, as soon as we get the complete tuple we stream the transaction including the serialized changes. We can do this incremental processing thanks to having assignments (associating subxact with toplevel xacts) in WAL right away, and thanks to logging the invalidation messages at each command end. These features are added by commits 0bead9af48 and c55040ccd0 respectively. Now that we can stream in-progress transactions, the concurrent aborts may cause failures when the output plugin consults catalogs (both system and user-defined). We handle such failures by returning ERRCODE_TRANSACTION_ROLLBACK sqlerrcode from system table scan APIs to the backend or WALSender decoding a specific uncommitted transaction. The decoding logic on the receipt of such a sqlerrcode aborts the decoding of the current transaction and continue with the decoding of other transactions. We have ReorderBufferTXN pointer in each ReorderBufferChange by which we know which xact it belongs to. The output plugin can use this to decide which changes to discard in case of stream_abort_cb (e.g. when a subxact gets discarded). We also provide a new option via SQL APIs to fetch the changes being streamed. Author: Dilip Kumar, Tomas Vondra, Amit Kapila, Nikhil Sontakke Reviewed-by: Amit Kapila, Kuntal Ghosh, Ajin Cherian Tested-by: Neha Sharma, Mahendra Singh Thalor and Ajin Cherian Discussion: https://postgr.es/m/688b0b7f-2f6c-d827-c27b-216a8e3ea700@2ndquadrant.com
2020-08-08 04:04:39 +02:00
/*
* We don't expect direct calls to table_scan_sample_next_block with valid
* CheckXidAlive for catalog or regular tables. See detailed comments in
* xact.c where these variables are declared.
*/
if (unlikely(TransactionIdIsValid(CheckXidAlive) && !bsysscan))
elog(ERROR, "unexpected table_scan_sample_next_block call during logical decoding");
tableam: sample scan. This moves sample scan support to below tableam. It's not optional as there is, in contrast to e.g. bitmap heap scans, no alternative way to perform tablesample queries. If an AM can't deal with the block based API, it will have to throw an ERROR. The tableam callbacks for this are block based, but given the current TsmRoutine interface, that seems to be required. The new interface doesn't require TsmRoutines to perform visibility checks anymore - that requires the TsmRoutine to know details about the AM, which we want to avoid. To continue to allow taking the returned number of tuples account SampleScanState now has a donetuples field (which previously e.g. existed in SystemRowsSamplerData), which is only incremented after the visibility check succeeds. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-31 05:18:53 +02:00
return scan->rs_rd->rd_tableam->scan_sample_next_block(scan, scanstate);
}
/*
* Fetch the next sample tuple into `slot` and return true if a visible tuple
* was found, false otherwise. table_scan_sample_next_block() needs to
tableam: bitmap table scan. This moves bitmap heap scan support to below an optional tableam callback. It's optional as the whole concept of bitmap heapscans is fairly block specific. This basically moves the work previously done in bitgetpage() into the new scan_bitmap_next_block callback, and the direct poking into the buffer done in BitmapHeapNext() into the new scan_bitmap_next_tuple() callback. The abstraction is currently somewhat leaky because nodeBitmapHeapscan.c's prefetching and visibilitymap based logic remains - it's likely that we'll later have to move more into the AM. But it's not trivial to do so without introducing a significant amount of code duplication between the AMs, so that's a project for later. Note that now nodeBitmapHeapscan.c and the associated node types are a bit misnamed. But it's not clear whether renaming wouldn't be a cure worse than the disease. Either way, that'd be best done in a separate commit. Author: Andres Freund Reviewed-By: Robert Haas (in an older version) Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-04-01 02:51:49 +02:00
* previously have selected a block (i.e. returned true), and no previous
* table_scan_sample_next_tuple() for the same block may have returned false.
tableam: sample scan. This moves sample scan support to below tableam. It's not optional as there is, in contrast to e.g. bitmap heap scans, no alternative way to perform tablesample queries. If an AM can't deal with the block based API, it will have to throw an ERROR. The tableam callbacks for this are block based, but given the current TsmRoutine interface, that seems to be required. The new interface doesn't require TsmRoutines to perform visibility checks anymore - that requires the TsmRoutine to know details about the AM, which we want to avoid. To continue to allow taking the returned number of tuples account SampleScanState now has a donetuples field (which previously e.g. existed in SystemRowsSamplerData), which is only incremented after the visibility check succeeds. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-31 05:18:53 +02:00
*
* This will call the TsmRoutine's NextSampleTuple() callback.
*/
static inline bool
table_scan_sample_next_tuple(TableScanDesc scan,
struct SampleScanState *scanstate,
TupleTableSlot *slot)
{
Implement streaming mode in ReorderBuffer. Instead of serializing the transaction to disk after reaching the logical_decoding_work_mem limit in memory, we consume the changes we have in memory and invoke stream API methods added by commit 45fdc9738b. However, sometimes if we have incomplete toast or speculative insert we spill to the disk because we can't generate the complete tuple and stream. And, as soon as we get the complete tuple we stream the transaction including the serialized changes. We can do this incremental processing thanks to having assignments (associating subxact with toplevel xacts) in WAL right away, and thanks to logging the invalidation messages at each command end. These features are added by commits 0bead9af48 and c55040ccd0 respectively. Now that we can stream in-progress transactions, the concurrent aborts may cause failures when the output plugin consults catalogs (both system and user-defined). We handle such failures by returning ERRCODE_TRANSACTION_ROLLBACK sqlerrcode from system table scan APIs to the backend or WALSender decoding a specific uncommitted transaction. The decoding logic on the receipt of such a sqlerrcode aborts the decoding of the current transaction and continue with the decoding of other transactions. We have ReorderBufferTXN pointer in each ReorderBufferChange by which we know which xact it belongs to. The output plugin can use this to decide which changes to discard in case of stream_abort_cb (e.g. when a subxact gets discarded). We also provide a new option via SQL APIs to fetch the changes being streamed. Author: Dilip Kumar, Tomas Vondra, Amit Kapila, Nikhil Sontakke Reviewed-by: Amit Kapila, Kuntal Ghosh, Ajin Cherian Tested-by: Neha Sharma, Mahendra Singh Thalor and Ajin Cherian Discussion: https://postgr.es/m/688b0b7f-2f6c-d827-c27b-216a8e3ea700@2ndquadrant.com
2020-08-08 04:04:39 +02:00
/*
* We don't expect direct calls to table_scan_sample_next_tuple with valid
* CheckXidAlive for catalog or regular tables. See detailed comments in
* xact.c where these variables are declared.
*/
if (unlikely(TransactionIdIsValid(CheckXidAlive) && !bsysscan))
elog(ERROR, "unexpected table_scan_sample_next_tuple call during logical decoding");
tableam: sample scan. This moves sample scan support to below tableam. It's not optional as there is, in contrast to e.g. bitmap heap scans, no alternative way to perform tablesample queries. If an AM can't deal with the block based API, it will have to throw an ERROR. The tableam callbacks for this are block based, but given the current TsmRoutine interface, that seems to be required. The new interface doesn't require TsmRoutines to perform visibility checks anymore - that requires the TsmRoutine to know details about the AM, which we want to avoid. To continue to allow taking the returned number of tuples account SampleScanState now has a donetuples field (which previously e.g. existed in SystemRowsSamplerData), which is only incremented after the visibility check succeeds. Author: Andres Freund Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
2019-03-31 05:18:53 +02:00
return scan->rs_rd->rd_tableam->scan_sample_next_tuple(scan, scanstate,
slot);
}
tableam: Add tuple_{insert, delete, update, lock} and use. This adds new, required, table AM callbacks for insert/delete/update and lock_tuple. To be able to reasonably use those, the EvalPlanQual mechanism had to be adapted, moving more logic into the AM. Previously both delete/update/lock call-sites and the EPQ mechanism had to have awareness of the specific tuple format to be able to fetch the latest version of a tuple. Obviously that needs to be abstracted away. To do so, move the logic that find the latest row version into the AM. lock_tuple has a new flag argument, TUPLE_LOCK_FLAG_FIND_LAST_VERSION, that forces it to lock the last version, rather than the current one. It'd have been possible to do so via a separate callback as well, but finding the last version usually also necessitates locking the newest version, making it sensible to combine the two. This replaces the previous use of EvalPlanQualFetch(). Additionally HeapTupleUpdated, which previously signaled either a concurrent update or delete, is now split into two, to avoid callers needing AM specific knowledge to differentiate. The move of finding the latest row version into tuple_lock means that encountering a row concurrently moved into another partition will now raise an error about "tuple to be locked" rather than "tuple to be updated/deleted" - which is accurate, as that always happens when locking rows. While possible slightly less helpful for users, it seems like an acceptable trade-off. As part of this commit HTSU_Result has been renamed to TM_Result, and its members been expanded to differentiated between updating and deleting. HeapUpdateFailureData has been renamed to TM_FailureData. The interface to speculative insertion is changed so nodeModifyTable.c does not have to set the speculative token itself anymore. Instead there's a version of tuple_insert, tuple_insert_speculative, that performs the speculative insertion (without requiring a flag to signal that fact), and the speculative insertion is either made permanent with table_complete_speculative(succeeded = true) or aborted with succeeded = false). Note that multi_insert is not yet routed through tableam, nor is COPY. Changing multi_insert requires changes to copy.c that are large enough to better be done separately. Similarly, although simpler, CREATE TABLE AS and CREATE MATERIALIZED VIEW are also only going to be adjusted in a later commit. Author: Andres Freund and Haribabu Kommi Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20190313003903.nwvrxi7rw3ywhdel@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-24 03:55:57 +01:00
/* ----------------------------------------------------------------------------
* Functions to make modifications a bit simpler.
* ----------------------------------------------------------------------------
*/
tableam: Rename wrapper functions to match callback names. Some of the wrapper functions didn't match the callback names. Many of them due to staying "consistent" with historic naming of the wrapped functionality. We decided that for most cases it's more important to be for tableam to be consistent going forward, than with the past. The one exception is beginscan/endscan/... because it'd have looked odd to have systable_beginscan/endscan/... with a different naming scheme, and changing the systable_* APIs would have caused way too much churn (including breaking a lot of external users). Author: Ashwin Agrawal, with some small additions by Andres Freund Reviewed-By: Andres Freund Discussion: https://postgr.es/m/CALfoeiugyrXZfX7n0ORCa4L-m834dzmaE8eFdbNR6PMpetU4Ww@mail.gmail.com
2019-05-24 01:25:48 +02:00
extern void simple_table_tuple_insert(Relation rel, TupleTableSlot *slot);
extern void simple_table_tuple_delete(Relation rel, ItemPointer tid,
Snapshot snapshot);
extern void simple_table_tuple_update(Relation rel, ItemPointer otid,
TupleTableSlot *slot, Snapshot snapshot,
Ignore BRIN indexes when checking for HOT updates When determining whether an index update may be skipped by using HOT, we can ignore attributes indexed by block summarizing indexes without references to individual tuples that need to be cleaned up. A new type TU_UpdateIndexes provides a signal to the executor to determine which indexes to update - no indexes, all indexes, or only the summarizing indexes. This also removes rd_indexattr list, and replaces it with rd_attrsvalid flag. The list was not used anywhere, and a simple flag is sufficient. This was originally committed as 5753d4ee32, but then got reverted by e3fcca0d0d because of correctness issues. Original patch by Josef Simanek, various fixes and improvements by Tomas Vondra and me. Authors: Matthias van de Meent, Josef Simanek, Tomas Vondra Reviewed-by: Tomas Vondra, Alvaro Herrera Discussion: https://postgr.es/m/05ebcb44-f383-86e3-4f31-0a97a55634cf@enterprisedb.com Discussion: https://postgr.es/m/CAFp7QwpMRGcDAQumN7onN9HjrJ3u4X3ZRXdGFT0K5G2JWvnbWg%40mail.gmail.com
2023-03-20 10:34:07 +01:00
TU_UpdateIndexes *update_indexes);
tableam: Add tuple_{insert, delete, update, lock} and use. This adds new, required, table AM callbacks for insert/delete/update and lock_tuple. To be able to reasonably use those, the EvalPlanQual mechanism had to be adapted, moving more logic into the AM. Previously both delete/update/lock call-sites and the EPQ mechanism had to have awareness of the specific tuple format to be able to fetch the latest version of a tuple. Obviously that needs to be abstracted away. To do so, move the logic that find the latest row version into the AM. lock_tuple has a new flag argument, TUPLE_LOCK_FLAG_FIND_LAST_VERSION, that forces it to lock the last version, rather than the current one. It'd have been possible to do so via a separate callback as well, but finding the last version usually also necessitates locking the newest version, making it sensible to combine the two. This replaces the previous use of EvalPlanQualFetch(). Additionally HeapTupleUpdated, which previously signaled either a concurrent update or delete, is now split into two, to avoid callers needing AM specific knowledge to differentiate. The move of finding the latest row version into tuple_lock means that encountering a row concurrently moved into another partition will now raise an error about "tuple to be locked" rather than "tuple to be updated/deleted" - which is accurate, as that always happens when locking rows. While possible slightly less helpful for users, it seems like an acceptable trade-off. As part of this commit HTSU_Result has been renamed to TM_Result, and its members been expanded to differentiated between updating and deleting. HeapUpdateFailureData has been renamed to TM_FailureData. The interface to speculative insertion is changed so nodeModifyTable.c does not have to set the speculative token itself anymore. Instead there's a version of tuple_insert, tuple_insert_speculative, that performs the speculative insertion (without requiring a flag to signal that fact), and the speculative insertion is either made permanent with table_complete_speculative(succeeded = true) or aborted with succeeded = false). Note that multi_insert is not yet routed through tableam, nor is COPY. Changing multi_insert requires changes to copy.c that are large enough to better be done separately. Similarly, although simpler, CREATE TABLE AS and CREATE MATERIALIZED VIEW are also only going to be adjusted in a later commit. Author: Andres Freund and Haribabu Kommi Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20190313003903.nwvrxi7rw3ywhdel@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-24 03:55:57 +01:00
tableam: Add and use scan APIs. Too allow table accesses to be not directly dependent on heap, several new abstractions are needed. Specifically: 1) Heap scans need to be generalized into table scans. Do this by introducing TableScanDesc, which will be the "base class" for individual AMs. This contains the AM independent fields from HeapScanDesc. The previous heap_{beginscan,rescan,endscan} et al. have been replaced with a table_ version. There's no direct replacement for heap_getnext(), as that returned a HeapTuple, which is undesirable for a other AMs. Instead there's table_scan_getnextslot(). But note that heap_getnext() lives on, it's still used widely to access catalog tables. This is achieved by new scan_begin, scan_end, scan_rescan, scan_getnextslot callbacks. 2) The portion of parallel scans that's shared between backends need to be able to do so without the user doing per-AM work. To achieve that new parallelscan_{estimate, initialize, reinitialize} callbacks are introduced, which operate on a new ParallelTableScanDesc, which again can be subclassed by AMs. As it is likely that several AMs are going to be block oriented, block oriented callbacks that can be shared between such AMs are provided and used by heap. table_block_parallelscan_{estimate, intiialize, reinitialize} as callbacks, and table_block_parallelscan_{nextpage, init} for use in AMs. These operate on a ParallelBlockTableScanDesc. 3) Index scans need to be able to access tables to return a tuple, and there needs to be state across individual accesses to the heap to store state like buffers. That's now handled by introducing a sort-of-scan IndexFetchTable, which again is intended to be subclassed by individual AMs (for heap IndexFetchHeap). The relevant callbacks for an AM are index_fetch_{end, begin, reset} to create the necessary state, and index_fetch_tuple to retrieve an indexed tuple. Note that index_fetch_tuple implementations need to be smarter than just blindly fetching the tuples for AMs that have optimizations similar to heap's HOT - the currently alive tuple in the update chain needs to be fetched if appropriate. Similar to table_scan_getnextslot(), it's undesirable to continue to return HeapTuples. Thus index_fetch_heap (might want to rename that later) now accepts a slot as an argument. Core code doesn't have a lot of call sites performing index scans without going through the systable_* API (in contrast to loads of heap_getnext calls and working directly with HeapTuples). Index scans now store the result of a search in IndexScanDesc->xs_heaptid, rather than xs_ctup->t_self. As the target is not generally a HeapTuple anymore that seems cleaner. To be able to sensible adapt code to use the above, two further callbacks have been introduced: a) slot_callbacks returns a TupleTableSlotOps* suitable for creating slots capable of holding a tuple of the AMs type. table_slot_callbacks() and table_slot_create() are based upon that, but have additional logic to deal with views, foreign tables, etc. While this change could have been done separately, nearly all the call sites that needed to be adapted for the rest of this commit also would have been needed to be adapted for table_slot_callbacks(), making separation not worthwhile. b) tuple_satisfies_snapshot checks whether the tuple in a slot is currently visible according to a snapshot. That's required as a few places now don't have a buffer + HeapTuple around, but a slot (which in heap's case internally has that information). Additionally a few infrastructure changes were needed: I) SysScanDesc, as used by systable_{beginscan, getnext} et al. now internally uses a slot to keep track of tuples. While systable_getnext() still returns HeapTuples, and will so for the foreseeable future, the index API (see 1) above) now only deals with slots. The remainder, and largest part, of this commit is then adjusting all scans in postgres to use the new APIs. Author: Andres Freund, Haribabu Kommi, Alvaro Herrera Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-11 20:46:41 +01:00
/* ----------------------------------------------------------------------------
* Helper functions to implement parallel scans for block oriented AMs.
* ----------------------------------------------------------------------------
*/
extern Size table_block_parallelscan_estimate(Relation rel);
extern Size table_block_parallelscan_initialize(Relation rel,
ParallelTableScanDesc pscan);
tableam: Formatting and other minor cleanups. The superflous heapam_xlog.h includes were reported by Peter Geoghegan.
2019-03-31 05:13:56 +02:00
extern void table_block_parallelscan_reinitialize(Relation rel,
ParallelTableScanDesc pscan);
extern BlockNumber table_block_parallelscan_nextpage(Relation rel,
Allocate consecutive blocks during parallel seqscans Previously we would allocate blocks to parallel workers during a parallel sequential scan 1 block at a time. Since other workers were likely to request a block before a worker returns for another block number to work on, this could lead to non-sequential I/O patterns in each worker which could cause the operating system's readahead to perform poorly or not at all. Here we change things so that we allocate consecutive "chunks" of blocks to workers and have them work on those until they're done, at which time we allocate another chunk for the worker. The size of these chunks is based on the size of the relation. Initial patch here was by Thomas Munro which showed some good improvements just having a fixed chunk size of 64 blocks with a simple ramp-down near the end of the scan. The revisions of the patch to make the chunk size based on the relation size and the adjusted ramp-down in powers of two was done by me, along with quite extensive benchmarking to determine the optimal chunk sizes. For the most part, benchmarks have shown significant performance improvements for large parallel sequential scans on Linux, FreeBSD and Windows using SSDs. It's less clear how this affects the performance of cloud providers. Tests done so far are unable to obtain stable enough performance to provide meaningful benchmark results. It is possible that this could cause some performance regressions on more obscure filesystems, so we may need to later provide users with some ability to get something closer to the old behavior. For now, let's leave that until we see that it's really required. Author: Thomas Munro, David Rowley Reviewed-by: Ranier Vilela, Soumyadeep Chakraborty, Robert Haas Reviewed-by: Amit Kapila, Kirk Jamison Discussion: https://postgr.es/m/CA+hUKGJ_EErDv41YycXcbMbCBkztA34+z1ts9VQH+ACRuvpxig@mail.gmail.com
2020-07-26 11:02:45 +02:00
ParallelBlockTableScanWorker pbscanwork,
tableam: Formatting and other minor cleanups. The superflous heapam_xlog.h includes were reported by Peter Geoghegan.
2019-03-31 05:13:56 +02:00
ParallelBlockTableScanDesc pbscan);
extern void table_block_parallelscan_startblock_init(Relation rel,
Allocate consecutive blocks during parallel seqscans Previously we would allocate blocks to parallel workers during a parallel sequential scan 1 block at a time. Since other workers were likely to request a block before a worker returns for another block number to work on, this could lead to non-sequential I/O patterns in each worker which could cause the operating system's readahead to perform poorly or not at all. Here we change things so that we allocate consecutive "chunks" of blocks to workers and have them work on those until they're done, at which time we allocate another chunk for the worker. The size of these chunks is based on the size of the relation. Initial patch here was by Thomas Munro which showed some good improvements just having a fixed chunk size of 64 blocks with a simple ramp-down near the end of the scan. The revisions of the patch to make the chunk size based on the relation size and the adjusted ramp-down in powers of two was done by me, along with quite extensive benchmarking to determine the optimal chunk sizes. For the most part, benchmarks have shown significant performance improvements for large parallel sequential scans on Linux, FreeBSD and Windows using SSDs. It's less clear how this affects the performance of cloud providers. Tests done so far are unable to obtain stable enough performance to provide meaningful benchmark results. It is possible that this could cause some performance regressions on more obscure filesystems, so we may need to later provide users with some ability to get something closer to the old behavior. For now, let's leave that until we see that it's really required. Author: Thomas Munro, David Rowley Reviewed-by: Ranier Vilela, Soumyadeep Chakraborty, Robert Haas Reviewed-by: Amit Kapila, Kirk Jamison Discussion: https://postgr.es/m/CA+hUKGJ_EErDv41YycXcbMbCBkztA34+z1ts9VQH+ACRuvpxig@mail.gmail.com
2020-07-26 11:02:45 +02:00
ParallelBlockTableScanWorker pbscanwork,
tableam: Formatting and other minor cleanups. The superflous heapam_xlog.h includes were reported by Peter Geoghegan.
2019-03-31 05:13:56 +02:00
ParallelBlockTableScanDesc pbscan);
tableam: Add and use scan APIs. Too allow table accesses to be not directly dependent on heap, several new abstractions are needed. Specifically: 1) Heap scans need to be generalized into table scans. Do this by introducing TableScanDesc, which will be the "base class" for individual AMs. This contains the AM independent fields from HeapScanDesc. The previous heap_{beginscan,rescan,endscan} et al. have been replaced with a table_ version. There's no direct replacement for heap_getnext(), as that returned a HeapTuple, which is undesirable for a other AMs. Instead there's table_scan_getnextslot(). But note that heap_getnext() lives on, it's still used widely to access catalog tables. This is achieved by new scan_begin, scan_end, scan_rescan, scan_getnextslot callbacks. 2) The portion of parallel scans that's shared between backends need to be able to do so without the user doing per-AM work. To achieve that new parallelscan_{estimate, initialize, reinitialize} callbacks are introduced, which operate on a new ParallelTableScanDesc, which again can be subclassed by AMs. As it is likely that several AMs are going to be block oriented, block oriented callbacks that can be shared between such AMs are provided and used by heap. table_block_parallelscan_{estimate, intiialize, reinitialize} as callbacks, and table_block_parallelscan_{nextpage, init} for use in AMs. These operate on a ParallelBlockTableScanDesc. 3) Index scans need to be able to access tables to return a tuple, and there needs to be state across individual accesses to the heap to store state like buffers. That's now handled by introducing a sort-of-scan IndexFetchTable, which again is intended to be subclassed by individual AMs (for heap IndexFetchHeap). The relevant callbacks for an AM are index_fetch_{end, begin, reset} to create the necessary state, and index_fetch_tuple to retrieve an indexed tuple. Note that index_fetch_tuple implementations need to be smarter than just blindly fetching the tuples for AMs that have optimizations similar to heap's HOT - the currently alive tuple in the update chain needs to be fetched if appropriate. Similar to table_scan_getnextslot(), it's undesirable to continue to return HeapTuples. Thus index_fetch_heap (might want to rename that later) now accepts a slot as an argument. Core code doesn't have a lot of call sites performing index scans without going through the systable_* API (in contrast to loads of heap_getnext calls and working directly with HeapTuples). Index scans now store the result of a search in IndexScanDesc->xs_heaptid, rather than xs_ctup->t_self. As the target is not generally a HeapTuple anymore that seems cleaner. To be able to sensible adapt code to use the above, two further callbacks have been introduced: a) slot_callbacks returns a TupleTableSlotOps* suitable for creating slots capable of holding a tuple of the AMs type. table_slot_callbacks() and table_slot_create() are based upon that, but have additional logic to deal with views, foreign tables, etc. While this change could have been done separately, nearly all the call sites that needed to be adapted for the rest of this commit also would have been needed to be adapted for table_slot_callbacks(), making separation not worthwhile. b) tuple_satisfies_snapshot checks whether the tuple in a slot is currently visible according to a snapshot. That's required as a few places now don't have a buffer + HeapTuple around, but a slot (which in heap's case internally has that information). Additionally a few infrastructure changes were needed: I) SysScanDesc, as used by systable_{beginscan, getnext} et al. now internally uses a slot to keep track of tuples. While systable_getnext() still returns HeapTuples, and will so for the foreseeable future, the index API (see 1) above) now only deals with slots. The remainder, and largest part, of this commit is then adjusting all scans in postgres to use the new APIs. Author: Andres Freund, Haribabu Kommi, Alvaro Herrera Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-11 20:46:41 +01:00
tableam: Provide helper functions for relation sizing. Most block-based table AMs will need the exact same implementation of the relation_size callback as the heap, and if they use a standard page layout, they will likely need an implementation of the relation_estimate_size callback that is very similar to that of the heap. Rearrange to facilitate code reuse. Patch by me, reviewed by Michael Paquier, Daniel Gustafsson, and Álvaro Herrera. Discussion: http://postgr.es/m/CA+TgmoZ6DBPnP1E-vRpQZUJQijJFD54F+SR_pxGiAAS-MyrigA@mail.gmail.com
2019-07-08 20:51:53 +02:00
/* ----------------------------------------------------------------------------
* Helper functions to implement relation sizing for block oriented AMs.
* ----------------------------------------------------------------------------
*/
extern uint64 table_block_relation_size(Relation rel, ForkNumber forkNumber);
extern void table_block_relation_estimate_size(Relation rel,
int32 *attr_widths,
BlockNumber *pages,
double *tuples,
double *allvisfrac,
Size overhead_bytes_per_tuple,
Size usable_bytes_per_page);
tableam: Add and use scan APIs. Too allow table accesses to be not directly dependent on heap, several new abstractions are needed. Specifically: 1) Heap scans need to be generalized into table scans. Do this by introducing TableScanDesc, which will be the "base class" for individual AMs. This contains the AM independent fields from HeapScanDesc. The previous heap_{beginscan,rescan,endscan} et al. have been replaced with a table_ version. There's no direct replacement for heap_getnext(), as that returned a HeapTuple, which is undesirable for a other AMs. Instead there's table_scan_getnextslot(). But note that heap_getnext() lives on, it's still used widely to access catalog tables. This is achieved by new scan_begin, scan_end, scan_rescan, scan_getnextslot callbacks. 2) The portion of parallel scans that's shared between backends need to be able to do so without the user doing per-AM work. To achieve that new parallelscan_{estimate, initialize, reinitialize} callbacks are introduced, which operate on a new ParallelTableScanDesc, which again can be subclassed by AMs. As it is likely that several AMs are going to be block oriented, block oriented callbacks that can be shared between such AMs are provided and used by heap. table_block_parallelscan_{estimate, intiialize, reinitialize} as callbacks, and table_block_parallelscan_{nextpage, init} for use in AMs. These operate on a ParallelBlockTableScanDesc. 3) Index scans need to be able to access tables to return a tuple, and there needs to be state across individual accesses to the heap to store state like buffers. That's now handled by introducing a sort-of-scan IndexFetchTable, which again is intended to be subclassed by individual AMs (for heap IndexFetchHeap). The relevant callbacks for an AM are index_fetch_{end, begin, reset} to create the necessary state, and index_fetch_tuple to retrieve an indexed tuple. Note that index_fetch_tuple implementations need to be smarter than just blindly fetching the tuples for AMs that have optimizations similar to heap's HOT - the currently alive tuple in the update chain needs to be fetched if appropriate. Similar to table_scan_getnextslot(), it's undesirable to continue to return HeapTuples. Thus index_fetch_heap (might want to rename that later) now accepts a slot as an argument. Core code doesn't have a lot of call sites performing index scans without going through the systable_* API (in contrast to loads of heap_getnext calls and working directly with HeapTuples). Index scans now store the result of a search in IndexScanDesc->xs_heaptid, rather than xs_ctup->t_self. As the target is not generally a HeapTuple anymore that seems cleaner. To be able to sensible adapt code to use the above, two further callbacks have been introduced: a) slot_callbacks returns a TupleTableSlotOps* suitable for creating slots capable of holding a tuple of the AMs type. table_slot_callbacks() and table_slot_create() are based upon that, but have additional logic to deal with views, foreign tables, etc. While this change could have been done separately, nearly all the call sites that needed to be adapted for the rest of this commit also would have been needed to be adapted for table_slot_callbacks(), making separation not worthwhile. b) tuple_satisfies_snapshot checks whether the tuple in a slot is currently visible according to a snapshot. That's required as a few places now don't have a buffer + HeapTuple around, but a slot (which in heap's case internally has that information). Additionally a few infrastructure changes were needed: I) SysScanDesc, as used by systable_{beginscan, getnext} et al. now internally uses a slot to keep track of tuples. While systable_getnext() still returns HeapTuples, and will so for the foreseeable future, the index API (see 1) above) now only deals with slots. The remainder, and largest part, of this commit is then adjusting all scans in postgres to use the new APIs. Author: Andres Freund, Haribabu Kommi, Alvaro Herrera Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-11 20:46:41 +01:00
/* ----------------------------------------------------------------------------
tableam: introduce table AM infrastructure. This introduces the concept of table access methods, i.e. CREATE ACCESS METHOD ... TYPE TABLE and CREATE TABLE ... USING (storage-engine). No table access functionality is delegated to table AMs as of this commit, that'll be done in following commits. Subsequent commits will incrementally abstract table access functionality to be routed through table access methods. That change is too large to be reviewed & committed at once, so it'll be done incrementally. Docs will be updated at the end, as adding them incrementally would likely make them less coherent, and definitely is a lot more work, without a lot of benefit. Table access methods are specified similar to index access methods, i.e. pg_am.amhandler returns, as INTERNAL, a pointer to a struct with callbacks. In contrast to index AMs that struct needs to live as long as a backend, typically that's achieved by just returning a pointer to a constant struct. Psql's \d+ now displays a table's access method. That can be disabled with HIDE_TABLEAM=true, which is mainly useful so regression tests can be run against different AMs. It's quite possible that this behaviour still needs to be fine tuned. For now it's not allowed to set a table AM for a partitioned table, as we've not resolved how partitions would inherit that. Disallowing allows us to introduce, if we decide that's the way forward, such a behaviour without a compatibility break. Catversion bumped, to add the heap table AM and references to it. Author: Haribabu Kommi, Andres Freund, Alvaro Herrera, Dimitri Golgov and others Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql https://postgr.es/m/20190107235616.6lur25ph22u5u5av@alap3.anarazel.de https://postgr.es/m/20190304234700.w5tmhducs5wxgzls@alap3.anarazel.de
2019-03-06 18:54:38 +01:00
* Functions in tableamapi.c
tableam: Add and use scan APIs. Too allow table accesses to be not directly dependent on heap, several new abstractions are needed. Specifically: 1) Heap scans need to be generalized into table scans. Do this by introducing TableScanDesc, which will be the "base class" for individual AMs. This contains the AM independent fields from HeapScanDesc. The previous heap_{beginscan,rescan,endscan} et al. have been replaced with a table_ version. There's no direct replacement for heap_getnext(), as that returned a HeapTuple, which is undesirable for a other AMs. Instead there's table_scan_getnextslot(). But note that heap_getnext() lives on, it's still used widely to access catalog tables. This is achieved by new scan_begin, scan_end, scan_rescan, scan_getnextslot callbacks. 2) The portion of parallel scans that's shared between backends need to be able to do so without the user doing per-AM work. To achieve that new parallelscan_{estimate, initialize, reinitialize} callbacks are introduced, which operate on a new ParallelTableScanDesc, which again can be subclassed by AMs. As it is likely that several AMs are going to be block oriented, block oriented callbacks that can be shared between such AMs are provided and used by heap. table_block_parallelscan_{estimate, intiialize, reinitialize} as callbacks, and table_block_parallelscan_{nextpage, init} for use in AMs. These operate on a ParallelBlockTableScanDesc. 3) Index scans need to be able to access tables to return a tuple, and there needs to be state across individual accesses to the heap to store state like buffers. That's now handled by introducing a sort-of-scan IndexFetchTable, which again is intended to be subclassed by individual AMs (for heap IndexFetchHeap). The relevant callbacks for an AM are index_fetch_{end, begin, reset} to create the necessary state, and index_fetch_tuple to retrieve an indexed tuple. Note that index_fetch_tuple implementations need to be smarter than just blindly fetching the tuples for AMs that have optimizations similar to heap's HOT - the currently alive tuple in the update chain needs to be fetched if appropriate. Similar to table_scan_getnextslot(), it's undesirable to continue to return HeapTuples. Thus index_fetch_heap (might want to rename that later) now accepts a slot as an argument. Core code doesn't have a lot of call sites performing index scans without going through the systable_* API (in contrast to loads of heap_getnext calls and working directly with HeapTuples). Index scans now store the result of a search in IndexScanDesc->xs_heaptid, rather than xs_ctup->t_self. As the target is not generally a HeapTuple anymore that seems cleaner. To be able to sensible adapt code to use the above, two further callbacks have been introduced: a) slot_callbacks returns a TupleTableSlotOps* suitable for creating slots capable of holding a tuple of the AMs type. table_slot_callbacks() and table_slot_create() are based upon that, but have additional logic to deal with views, foreign tables, etc. While this change could have been done separately, nearly all the call sites that needed to be adapted for the rest of this commit also would have been needed to be adapted for table_slot_callbacks(), making separation not worthwhile. b) tuple_satisfies_snapshot checks whether the tuple in a slot is currently visible according to a snapshot. That's required as a few places now don't have a buffer + HeapTuple around, but a slot (which in heap's case internally has that information). Additionally a few infrastructure changes were needed: I) SysScanDesc, as used by systable_{beginscan, getnext} et al. now internally uses a slot to keep track of tuples. While systable_getnext() still returns HeapTuples, and will so for the foreseeable future, the index API (see 1) above) now only deals with slots. The remainder, and largest part, of this commit is then adjusting all scans in postgres to use the new APIs. Author: Andres Freund, Haribabu Kommi, Alvaro Herrera Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-11 20:46:41 +01:00
* ----------------------------------------------------------------------------
tableam: introduce table AM infrastructure. This introduces the concept of table access methods, i.e. CREATE ACCESS METHOD ... TYPE TABLE and CREATE TABLE ... USING (storage-engine). No table access functionality is delegated to table AMs as of this commit, that'll be done in following commits. Subsequent commits will incrementally abstract table access functionality to be routed through table access methods. That change is too large to be reviewed & committed at once, so it'll be done incrementally. Docs will be updated at the end, as adding them incrementally would likely make them less coherent, and definitely is a lot more work, without a lot of benefit. Table access methods are specified similar to index access methods, i.e. pg_am.amhandler returns, as INTERNAL, a pointer to a struct with callbacks. In contrast to index AMs that struct needs to live as long as a backend, typically that's achieved by just returning a pointer to a constant struct. Psql's \d+ now displays a table's access method. That can be disabled with HIDE_TABLEAM=true, which is mainly useful so regression tests can be run against different AMs. It's quite possible that this behaviour still needs to be fine tuned. For now it's not allowed to set a table AM for a partitioned table, as we've not resolved how partitions would inherit that. Disallowing allows us to introduce, if we decide that's the way forward, such a behaviour without a compatibility break. Catversion bumped, to add the heap table AM and references to it. Author: Haribabu Kommi, Andres Freund, Alvaro Herrera, Dimitri Golgov and others Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql https://postgr.es/m/20190107235616.6lur25ph22u5u5av@alap3.anarazel.de https://postgr.es/m/20190304234700.w5tmhducs5wxgzls@alap3.anarazel.de
2019-03-06 18:54:38 +01:00
*/
tableam: Add and use scan APIs. Too allow table accesses to be not directly dependent on heap, several new abstractions are needed. Specifically: 1) Heap scans need to be generalized into table scans. Do this by introducing TableScanDesc, which will be the "base class" for individual AMs. This contains the AM independent fields from HeapScanDesc. The previous heap_{beginscan,rescan,endscan} et al. have been replaced with a table_ version. There's no direct replacement for heap_getnext(), as that returned a HeapTuple, which is undesirable for a other AMs. Instead there's table_scan_getnextslot(). But note that heap_getnext() lives on, it's still used widely to access catalog tables. This is achieved by new scan_begin, scan_end, scan_rescan, scan_getnextslot callbacks. 2) The portion of parallel scans that's shared between backends need to be able to do so without the user doing per-AM work. To achieve that new parallelscan_{estimate, initialize, reinitialize} callbacks are introduced, which operate on a new ParallelTableScanDesc, which again can be subclassed by AMs. As it is likely that several AMs are going to be block oriented, block oriented callbacks that can be shared between such AMs are provided and used by heap. table_block_parallelscan_{estimate, intiialize, reinitialize} as callbacks, and table_block_parallelscan_{nextpage, init} for use in AMs. These operate on a ParallelBlockTableScanDesc. 3) Index scans need to be able to access tables to return a tuple, and there needs to be state across individual accesses to the heap to store state like buffers. That's now handled by introducing a sort-of-scan IndexFetchTable, which again is intended to be subclassed by individual AMs (for heap IndexFetchHeap). The relevant callbacks for an AM are index_fetch_{end, begin, reset} to create the necessary state, and index_fetch_tuple to retrieve an indexed tuple. Note that index_fetch_tuple implementations need to be smarter than just blindly fetching the tuples for AMs that have optimizations similar to heap's HOT - the currently alive tuple in the update chain needs to be fetched if appropriate. Similar to table_scan_getnextslot(), it's undesirable to continue to return HeapTuples. Thus index_fetch_heap (might want to rename that later) now accepts a slot as an argument. Core code doesn't have a lot of call sites performing index scans without going through the systable_* API (in contrast to loads of heap_getnext calls and working directly with HeapTuples). Index scans now store the result of a search in IndexScanDesc->xs_heaptid, rather than xs_ctup->t_self. As the target is not generally a HeapTuple anymore that seems cleaner. To be able to sensible adapt code to use the above, two further callbacks have been introduced: a) slot_callbacks returns a TupleTableSlotOps* suitable for creating slots capable of holding a tuple of the AMs type. table_slot_callbacks() and table_slot_create() are based upon that, but have additional logic to deal with views, foreign tables, etc. While this change could have been done separately, nearly all the call sites that needed to be adapted for the rest of this commit also would have been needed to be adapted for table_slot_callbacks(), making separation not worthwhile. b) tuple_satisfies_snapshot checks whether the tuple in a slot is currently visible according to a snapshot. That's required as a few places now don't have a buffer + HeapTuple around, but a slot (which in heap's case internally has that information). Additionally a few infrastructure changes were needed: I) SysScanDesc, as used by systable_{beginscan, getnext} et al. now internally uses a slot to keep track of tuples. While systable_getnext() still returns HeapTuples, and will so for the foreseeable future, the index API (see 1) above) now only deals with slots. The remainder, and largest part, of this commit is then adjusting all scans in postgres to use the new APIs. Author: Andres Freund, Haribabu Kommi, Alvaro Herrera Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
2019-03-11 20:46:41 +01:00
tableam: introduce table AM infrastructure. This introduces the concept of table access methods, i.e. CREATE ACCESS METHOD ... TYPE TABLE and CREATE TABLE ... USING (storage-engine). No table access functionality is delegated to table AMs as of this commit, that'll be done in following commits. Subsequent commits will incrementally abstract table access functionality to be routed through table access methods. That change is too large to be reviewed & committed at once, so it'll be done incrementally. Docs will be updated at the end, as adding them incrementally would likely make them less coherent, and definitely is a lot more work, without a lot of benefit. Table access methods are specified similar to index access methods, i.e. pg_am.amhandler returns, as INTERNAL, a pointer to a struct with callbacks. In contrast to index AMs that struct needs to live as long as a backend, typically that's achieved by just returning a pointer to a constant struct. Psql's \d+ now displays a table's access method. That can be disabled with HIDE_TABLEAM=true, which is mainly useful so regression tests can be run against different AMs. It's quite possible that this behaviour still needs to be fine tuned. For now it's not allowed to set a table AM for a partitioned table, as we've not resolved how partitions would inherit that. Disallowing allows us to introduce, if we decide that's the way forward, such a behaviour without a compatibility break. Catversion bumped, to add the heap table AM and references to it. Author: Haribabu Kommi, Andres Freund, Alvaro Herrera, Dimitri Golgov and others Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql https://postgr.es/m/20190107235616.6lur25ph22u5u5av@alap3.anarazel.de https://postgr.es/m/20190304234700.w5tmhducs5wxgzls@alap3.anarazel.de
2019-03-06 18:54:38 +01:00
extern const TableAmRoutine *GetTableAmRoutine(Oid amhandler);
extern const TableAmRoutine *GetHeapamTableAmRoutine(void);
#endif /* TABLEAM_H */