Teach nbtree backwards scans to avoid relocking a just-read leaf page to
read its current left sibling link when it isn't truly necessary. This
happened inside _bt_readnextpage whenever _bt_readpage had already
determined that there'll be no further matches to the left (or at least
none for the current primitive index scan, for a scan with array keys).
A new precheck inside _bt_readnextpage is all that we need to avoid
these useless lock acquisitions. Arguably, using a precheck like this
was a missed opportunity for commit 2ed5b87f96, which taught nbtree to
drop leaf page pins early to avoid blocking cleanup by VACUUM. Forwards
scans already managed to avoid relocking the page like this.
The optimization added by this commit is particularly helpful with
backwards scans that use array keys where the scan must perform multiple
primitive index scans. Such backwards scans will now avoid a useless
leaf page re-lock at the end of each primitive index scan.
Note that this commit does not attempt to avoid needlessly re-locking a
leaf page that was just read when the scan must follow the leaf page's
left link. That more ambitious optimization could work by stashing the
left link when the page is first read by a backwards scan, allowing the
subsequent _bt_readnextpage call to optimistically skip re-reading the
original page just to get a new copy of its left link. For now we only
address cases where we don't care about our original page's left link.
Author: Peter Geoghegan <pg@bowt.ie>
Reviewed-By: Matthias van de Meent <boekewurm+postgres@gmail.com>
Discussion: https://postgr.es/m/CAH2-Wz=xgs7PojG=EUvhgadwENzu_mY_riNh-w9wFPsaS717ew@mail.gmail.com
Commit 9e8da0f7 taught nbtree to handle ScalarArrayOpExpr quals
natively. This works by pushing down the full context (the array keys)
to the nbtree index AM, enabling it to execute multiple primitive index
scans that the planner treats as one continuous index scan/index path.
This earlier enhancement enabled nbtree ScalarArrayOp index-only scans.
It also allowed scans with ScalarArrayOp quals to return ordered results
(with some notable restrictions, described further down).
Take this general approach a lot further: teach nbtree SAOP index scans
to decide how to execute ScalarArrayOp scans (when and where to start
the next primitive index scan) based on physical index characteristics.
This can be far more efficient. All SAOP scans will now reliably avoid
duplicative leaf page accesses (just like any other nbtree index scan).
SAOP scans whose array keys are naturally clustered together now require
far fewer index descents, since we'll reliably avoid starting a new
primitive scan just to get to a later offset from the same leaf page.
The scan's arrays now advance using binary searches for the array
element that best matches the next tuple's attribute value. Required
scan key arrays (i.e. arrays from scan keys that can terminate the scan)
ratchet forward in lockstep with the index scan. Non-required arrays
(i.e. arrays from scan keys that can only exclude non-matching tuples)
"advance" without the process ever rolling over to a higher-order array.
Naturally, only required SAOP scan keys trigger skipping over leaf pages
(non-required arrays cannot safely end or start primitive index scans).
Consequently, even index scans of a composite index with a high-order
inequality scan key (which we'll mark required) and a low-order SAOP
scan key (which we won't mark required) now avoid repeating leaf page
accesses -- that benefit isn't limited to simpler equality-only cases.
In general, all nbtree index scans now output tuples as if they were one
continuous index scan -- even scans that mix a high-order inequality
with lower-order SAOP equalities reliably output tuples in index order.
This allows us to remove a couple of special cases that were applied
when building index paths with SAOP clauses during planning.
Bugfix commit 807a40c5 taught the planner to avoid generating unsafe
path keys: path keys on a multicolumn index path, with a SAOP clause on
any attribute beyond the first/most significant attribute. These cases
are now all safe, so we go back to generating path keys without regard
for the presence of SAOP clauses (just like with any other clause type).
Affected queries can now exploit scan output order in all the usual ways
(e.g., certain "ORDER BY ... LIMIT n" queries can now terminate early).
Also undo changes from follow-up bugfix commit a4523c5a, which taught
the planner to produce alternative index paths, with path keys, but
without low-order SAOP index quals (filter quals were used instead).
We'll no longer generate these alternative paths, since they can no
longer offer any meaningful advantages over standard index qual paths.
Affected queries thereby avoid all of the disadvantages that come from
using filter quals within index scan nodes. They can avoid extra heap
page accesses from using filter quals to exclude non-matching tuples
(index quals will never have that problem). They can also skip over
irrelevant sections of the index in more cases (though only when nbtree
determines that starting another primitive scan actually makes sense).
There is a theoretical risk that removing restrictions on SAOP index
paths from the planner will break compatibility with amcanorder-based
index AMs maintained as extensions. Such an index AM could have the
same limitations around ordered SAOP scans as nbtree had up until now.
Adding a pro forma incompatibility item about the issue to the Postgres
17 release notes seems like a good idea.
Author: Peter Geoghegan <pg@bowt.ie>
Author: Matthias van de Meent <boekewurm+postgres@gmail.com>
Reviewed-By: Heikki Linnakangas <hlinnaka@iki.fi>
Reviewed-By: Matthias van de Meent <boekewurm+postgres@gmail.com>
Reviewed-By: Tomas Vondra <tomas.vondra@enterprisedb.com>
Discussion: https://postgr.es/m/CAH2-Wz=ksvN_sjcnD1+Bt-WtifRA5ok48aDYnq3pkKhxgMQpcw@mail.gmail.com
e0b1ee17dc introduced optimization for matching B-tree scan keys required for
the directional scan. However, it incorrectly assumed that all keys required
for opposite direction scan are satisfied by _bt_first(). It has been
illustrated that with multiple scan keys over the same column, a lesser one
(according to the scan direction) could win leaving the other one unsatisfied.
Instead of relying on _bt_first() this commit introduces code that memorizes
whether there was at least one match on the page. If that's true we know that
keys required for opposite-direction scan are satisfied as soon as
corresponding values are not NULLs.
Also, this commit simplifies the description for the optimization of keys
required for the current direction scan. Now the flag used for this is named
continuescanPrechecked and means exactly that *continuescan flag is known
to be true for the last item on the page.
Reported-by: Peter Geoghegan
Discussion: https://postgr.es/m/CAH2-Wzn0LeLcb1PdBnK0xisz8NpHkxRrMr3NWJ%2BKOK-WZ%2BQtTQ%40mail.gmail.com
Reviewed-by: Pavel Borisov
It's not necessary to keep the firstPage flag as a field of BTScanOpaqueData.
This commit makes it an argument of the _bt_readpage() function. We can easily
distinguish first-time and repeated calls (within the scan) of this function.
Reported-by: Peter Geoghegan
Discussion: https://postgr.es/m/CAH2-Wzk4SOsw%2BtHuTFiz8U9Jqj-R77rYPkhWKODCBb1mdHACXA%40mail.gmail.com
Reviewed-by: Pavel Borisov
Remove comments that supposed that holding a pin was a useful interlock
for _bt_walk_left(). There are times when _bt_walk_left() doesn't hold
either a lock or a pin on any page, so clearly this can't be true.
_bt_walk_left() is even prepared to deal with concurrent deletion of
both the original page and any pages to its left.
Oversight in commit 2ed5b87f96.
Teach _bt_binsrch (and related helper routines like _bt_search and
_bt_compare) about the initial positioning requirements of backward
scans. Routines like _bt_binsrch already know all about "nextkey"
searches, so it seems natural to teach them about "goback"/backward
searches, too. These concepts are closely related, and are much easier
to understand when discussed together.
Now that certain implementation details are hidden from _bt_first, it's
straightforward to add a new optimization: backward scans using the <
strategy now avoid extra leaf page accesses in certain "boundary cases".
Consider the following example, which uses the tenk1 table (and its
tenk1_hundred index) from the standard regression tests:
SELECT * FROM tenk1 WHERE hundred < 12 ORDER BY hundred DESC LIMIT 1;
Before this commit, nbtree would scan two leaf pages, even though it was
only really necessary to scan one leaf page. We'll now descend straight
to the leaf page containing a (12, -inf) high key instead. The scan
will locate matching non-pivot tuples with "hundred" values starting
from the value 11. The scan won't waste a page access on the right
sibling leaf page, which cannot possibly contain any matching tuples.
You can think of the optimization added by this commit as disabling an
optimization (the _bt_compare "!pivotsearch" behavior that was added to
Postgres 12 in commit dd299df8) for a small subset of cases where it was
always counterproductive.
Equivalently, you can think of the new optimization as extending the
"pivotsearch" behavior that page deletion by VACUUM has long required
(since the aforementioned Postgres 12 commit went in) to other, similar
cases. Obviously, this isn't strictly necessary for these new cases
(unlike VACUUM, _bt_first is prepared to move the scan to the left once
on the leaf level), but the underlying principle is the same.
Author: Peter Geoghegan <pg@bowt.ie>
Reviewed-By: Matthias van de Meent <boekewurm+postgres@gmail.com>
Discussion: https://postgr.es/m/CAH2-Wz=XPzM8HzaLPq278Vms420mVSHfgs9wi5tjFKHcapZCEw@mail.gmail.com
Currently, B-tree code matches every scan key to every item on the page.
Imagine the ordered B-tree scan for the query like this.
SELECT * FROM tbl WHERE col > 'a' AND col < 'b' ORDER BY col;
The (col > 'a') scan key will be always matched once we find the location to
start the scan. The (col < 'b') scan key will match every item on the page
as long as it matches the last item on the page.
This patch implements prechecking of the scan keys required for directional
scan on beginning of page scan. If precheck is successful we can skip this
scan keys check for the items on the page. That could lead to significant
acceleration especially if the comparison operator is expensive.
Idea from patch by Konstantin Knizhnik.
Discussion: https://postgr.es/m/079c3f8e-3371-abe2-e93c-fc8a0ae3f571%40garret.ru
Reviewed-by: Peter Geoghegan, Pavel Borisov
When predicate-locking btrees, we have a special case for completely
empty btrees, since there is no page to lock. This was racy, because,
without buffer lock held, a matching key could be inserted between the
_bt_search() and the PredicateLockRelation() calls.
Fix, by rechecking _bt_search() after taking the relation-level SIREAD
lock, if using SERIALIZABLE isolation and an empty btree is discovered.
Back-patch to all supported releases. Fixes one aspect of bug #17949.
Reported-by: Artem Anisimov <artem.anisimov.255@gmail.com>
Reviewed-by: Dmitry Dolgov <9erthalion6@gmail.com>
Reviewed-by: Heikki Linnakangas <hlinnaka@iki.fi>
Discussion: https://postgr.es/m/17949-a0f17035294a55e2%40postgresql.org
Split nbtree's _bt_getbuf function is two: code that read locks or write
locks existing pages remains in _bt_getbuf, while code that deals with
allocating new pages is moved to a new, dedicated function called
_bt_allocbuf. This simplifies most _bt_getbuf callers, since it is no
longer necessary for them to pass a heaprel argument. Many of the
changes to nbtree from commit 61b313e4 can be reverted. This minimizes
the divergence between HEAD/PostgreSQL 16 and earlier release branches.
_bt_allocbuf replaces the previous nbtree idiom of passing P_NEW to
_bt_getbuf. There are only 3 affected call sites, all of which continue
to pass a heaprel for recovery conflict purposes. Note that nbtree's
use of P_NEW was superficial; nbtree never actually relied on the P_NEW
code paths in bufmgr.c, so this change is strictly mechanical.
GiST already took the same approach; it has a dedicated function for
allocating new pages called gistNewBuffer(). That factor allowed commit
61b313e4 to make much more targeted changes to GiST.
Author: Peter Geoghegan <pg@bowt.ie>
Reviewed-By: Heikki Linnakangas <hlinnaka@iki.fi>
Discussion: https://postgr.es/m/CAH2-Wz=8Z9qY58bjm_7TAHgtW6RzZ5Ke62q5emdCEy9BAzwhmg@mail.gmail.com
This is done in preparation for logical decoding on standby, which needs to
include whether visibility affecting WAL records are about a (user) catalog
table. Which is only known for the table, not the indexes.
It's also nice to be able to pass the heap relation to GlobalVisTestFor() in
vacuumRedirectAndPlaceholder().
Author: "Drouvot, Bertrand" <bertranddrouvot.pg@gmail.com>
Discussion: https://postgr.es/m/21b700c3-eecf-2e05-a699-f8c78dd31ec7@gmail.com
This makes the code more consistent with SpGiST, GiST and GIN, that
already use this style, and the idea is to make easier the introduction
of more sanity checks for each of these AM-specific macros. BRIN uses a
different set of macros to get a page's type and flags, so it has no
need for something similar.
Author: Matthias van de Meent
Discussion: https://postgr.es/m/CAEze2WjE3+tGO9Fs9+iZMU+z6mMZKo54W1Zt98WKqbEUHbHOBg@mail.gmail.com
The term "super-exclusive lock" is a synonym for "buffer cleanup lock"
that first appeared in nbtree many years ago. Standardize things by
consistently using the term cleanup lock. This finishes work started by
commit 276db875.
There is no good reason to have two terms. But there is a good reason
to only have one: to avoid confusion around why VACUUM acquires a full
cleanup lock (not just an ordinary exclusive lock) in index AMs, during
ambulkdelete calls. This has nothing to do with protecting the physical
index data structure itself. It is needed to implement a locking
protocol that ensures that TIDs pointing to the heap/table structure
cannot get marked for recycling by VACUUM before it is safe (which is
somewhat similar to how VACUUM uses cleanup locks during its first heap
pass). Note that it isn't strictly necessary for index AMs to implement
this locking protocol -- several index AMs use an MVCC snapshot as their
sole interlock to prevent unsafe TID recycling.
In passing, update the nbtree README. Cleanly separate discussion of
the aforementioned index vacuuming locking protocol from discussion of
the "drop leaf page pin" optimization added by commit 2ed5b87f. We now
structure discussion of the latter by describing how individual index
scans may safely opt out of applying the standard locking protocol (and
so can avoid blocking progress by VACUUM). Also document why the
optimization is not safe to apply during nbtree index-only scans.
Author: Peter Geoghegan <pg@bowt.ie>
Discussion: https://postgr.es/m/CAH2-WzngHgQa92tz6NQihf4nxJwRzCV36yMJO_i8dS+2mgEVKw@mail.gmail.com
Discussion: https://postgr.es/m/CAH2-WzkHPgsBBvGWjz=8PjNhDefy7XRkDKiT5NxMs-n5ZCf2dA@mail.gmail.com
Add more defensive checks around posting list split code. These should
detect corruption involving duplicate table TIDs earlier and more
reliably than any existing check.
Follow up to commit 8f72bbac.
Discussion: https://postgr.es/m/CAH2-WzkrSY_kjyd1_M5xJK1uM0govJXMxPn8JUSvwcUOiHuWVw@mail.gmail.com
Backpatch: 13-, where nbtree deduplication was introduced.
Otherwise we risk "leaking" deleted pages by making them non-recyclable
indefinitely. Commit 6655a729 did the same thing for deleted pages in
GiST indexes. That work was used as a starting point here.
Stop storing an XID indicating the oldest bpto.xact across all deleted
though unrecycled pages in nbtree metapages. There is no longer any
reason to care about that condition/the oldest XID. It only ever made
sense when wraparound was something _bt_vacuum_needs_cleanup() had to
consider.
The btm_oldest_btpo_xact metapage field has been repurposed and renamed.
It is now btm_last_cleanup_num_delpages, which is used to remember how
many non-recycled deleted pages remain from the last VACUUM (in practice
its value is usually the precise number of pages that were _newly
deleted_ during the specific VACUUM operation that last set the field).
The general idea behind storing btm_last_cleanup_num_delpages is to use
it to give _some_ consideration to non-recycled deleted pages inside
_bt_vacuum_needs_cleanup() -- though never too much. We only really
need to avoid leaving a truly excessive number of deleted pages in an
unrecycled state forever. We only do this to cover certain narrow cases
where no other factor makes VACUUM do a full scan, and yet the index
continues to grow (and so actually misses out on recycling existing
deleted pages).
These metapage changes result in a clear user-visible benefit: We no
longer trigger full index scans during VACUUM operations solely due to
the presence of only 1 or 2 known deleted (though unrecycled) blocks
from a very large index. All that matters now is keeping the costs and
benefits in balance over time.
Fix an issue that has been around since commit 857f9c36, which added the
"skip full scan of index" mechanism (i.e. the _bt_vacuum_needs_cleanup()
logic). The accuracy of btm_last_cleanup_num_heap_tuples accidentally
hinged upon _when_ the source value gets stored. We now always store
btm_last_cleanup_num_heap_tuples in btvacuumcleanup(). This fixes the
issue because IndexVacuumInfo.num_heap_tuples (the source field) is
expected to accurately indicate the state of the table _after_ the
VACUUM completes inside btvacuumcleanup().
A backpatchable fix cannot easily be extracted from this commit. A
targeted fix for the issue will follow in a later commit, though that
won't happen today.
I (pgeoghegan) have chosen to remove any mention of deleted pages in the
documentation of the vacuum_cleanup_index_scale_factor GUC/param, since
the presence of deleted (though unrecycled) pages is no longer of much
concern to users. The vacuum_cleanup_index_scale_factor description in
the docs now seems rather unclear in any case, and it should probably be
rewritten in the near future. Perhaps some passing mention of page
deletion will be added back at the same time.
Bump XLOG_PAGE_MAGIC due to nbtree WAL records using full XIDs now.
Author: Peter Geoghegan <pg@bowt.ie>
Reviewed-By: Masahiko Sawada <sawada.mshk@gmail.com>
Discussion: https://postgr.es/m/CAH2-WznpdHvujGUwYZ8sihX=d5u-tRYhi-F4wnV2uN2zHpMUXw@mail.gmail.com
For parallel btree scan to work for array of scan keys, it should reach
BTPARALLEL_DONE state once for every distinct combination of array keys.
This is required to ensure that the parallel workers don't try to seize
blocks at the same time for different scan keys. We missed to update this
state when we discovered that the scan keys can't be satisfied.
Author: James Hunter
Reviewed-by: Amit Kapila
Tested-by: Justin Pryzby
Backpatch-through: 10, where it was introduced
Discussion: https://postgr.es/m/4248CABC-25E3-4809-B4D0-128E1BAABC3C@amazon.com
Holding just a buffer pin (with no buffer lock) on an nbtree buffer/page
provides very weak guarantees, especially compared to heapam, where it's
often safe to read a page while only holding a buffer pin. This commit
has Valgrind enforce the following rule: it is never okay to access an
nbtree buffer without holding both a pin and a lock on the buffer.
A draft version of this patch detected questionable code that was
cleaned up by commits fa7ff642 and 7154aa16. The code in question used
to access an nbtree buffer page's special/opaque area with no buffer
lock (only a buffer pin). This practice (which isn't obviously unsafe)
is hereby formally disallowed in nbtree. There doesn't seem to be any
reason to allow it, and banning it keeps things simple for Valgrind.
The new checks are implemented by adding custom nbtree client requests
(located in LockBuffer() wrapper functions); these requests are
"superimposed" on top of the generic bufmgr.c Valgrind client requests
added by commit 1e0dfd16. No custom resource management cleanup code is
needed to undo the effects of marking buffers as non-accessible under
this scheme.
Author: Peter Geoghegan
Reviewed-By: Anastasia Lubennikova, Georgios Kokolatos
Discussion: https://postgr.es/m/CAH2-WzkLgyN3zBvRZ1pkNJThC=xi_0gpWRUb_45eexLH1+k2_Q@mail.gmail.com
Make some of the variable names in _bt_search() consistent with
corresponding variables within _bt_getstackbuf(). This naming scheme is
clearer because the variable names always express a relationship between
the currently locked buffer/page and some other page.
Includes some manual cleanup of places that pgindent messed up,
most of which weren't per project style anyway.
Notably, it seems some people didn't absorb the style rules of
commit c9d297751, because there were a bunch of new occurrences
of function calls with a newline just after the left paren, all
with faulty expectations about how the rest of the call would get
indented.
Deduplication reduces the storage overhead of duplicates in indexes that
use the standard nbtree index access method. The deduplication process
is applied lazily, after the point where opportunistic deletion of
LP_DEAD-marked index tuples occurs. Deduplication is only applied at
the point where a leaf page split would otherwise be required. New
posting list tuples are formed by merging together existing duplicate
tuples. The physical representation of the items on an nbtree leaf page
is made more space efficient by deduplication, but the logical contents
of the page are not changed. Even unique indexes make use of
deduplication as a way of controlling bloat from duplicates whose TIDs
point to different versions of the same logical table row.
The lazy approach taken by nbtree has significant advantages over a GIN
style eager approach. Most individual inserts of index tuples have
exactly the same overhead as before. The extra overhead of
deduplication is amortized across insertions, just like the overhead of
page splits. The key space of indexes works in the same way as it has
since commit dd299df8 (the commit that made heap TID a tiebreaker
column).
Testing has shown that nbtree deduplication can generally make indexes
with about 10 or 15 tuples for each distinct key value about 2.5X - 4X
smaller, even with single column integer indexes (e.g., an index on a
referencing column that accompanies a foreign key). The final size of
single column nbtree indexes comes close to the final size of a similar
contrib/btree_gin index, at least in cases where GIN's posting list
compression isn't very effective. This can significantly improve
transaction throughput, and significantly reduce the cost of vacuuming
indexes.
A new index storage parameter (deduplicate_items) controls the use of
deduplication. The default setting is 'on', so all new B-Tree indexes
automatically use deduplication where possible. This decision will be
reviewed at the end of the Postgres 13 beta period.
There is a regression of approximately 2% of transaction throughput with
synthetic workloads that consist of append-only inserts into a table
with several non-unique indexes, where all indexes have few or no
repeated values. The underlying issue is that cycles are wasted on
unsuccessful attempts at deduplicating items in non-unique indexes.
There doesn't seem to be a way around it short of disabling
deduplication entirely. Note that deduplication of items in unique
indexes is fairly well targeted in general, which avoids the problem
there (we can use a special heuristic to trigger deduplication passes in
unique indexes, since we're specifically targeting "version bloat").
Bump XLOG_PAGE_MAGIC because xl_btree_vacuum changed.
No bump in BTREE_VERSION, since the representation of posting list
tuples works in a way that's backwards compatible with version 4 indexes
(i.e. indexes built on PostgreSQL 12). However, users must still
REINDEX a pg_upgrade'd index to use deduplication, regardless of the
Postgres version they've upgraded from. This is the only way to set the
new nbtree metapage flag indicating that deduplication is generally
safe.
Author: Anastasia Lubennikova, Peter Geoghegan
Reviewed-By: Peter Geoghegan, Heikki Linnakangas
Discussion:
https://postgr.es/m/55E4051B.7020209@postgrespro.ruhttps://postgr.es/m/4ab6e2db-bcee-f4cf-0916-3a06e6ccbb55@postgrespro.ru
btbuild() has nothing to say about how NULL values compare in nbtree.
Besides, there are _bt_compare() header comments that describe how NULL
values are handled.
Assert that _bt_binsrch() binary searches with scantid set in insertion
scankey cannot be performed on leaf pages. Leaf-level binary searches
where scantid is set must use _bt_binsrch_insert() instead.
_bt_binsrch_insert() is likely to have additional responsibilities in
the future, such as searching within GIN-style posting lists using
scantid. It seems like a good idea to tighten things up now.
The initial value of the nbtree stack downlink block number field
recorded during an initial descent of the tree wasn't actually used.
Both _bt_getstackbuf() callers overwrote the value with their own value.
Remove the block number field from the stack struct, and add a child
block number argument to _bt_getstackbuf() in its place. This makes the
overall design of _bt_getstackbuf() clearer.
Author: Peter Geoghegan
Reviewed-By: Anastasia Lubennikova
Discussion: https://postgr.es/m/CAH2-Wzmx+UbXt2YNOUCZ-a04VdXU=S=OHuAuD7Z8uQq-PXTYUg@mail.gmail.com
This is still using the 2.0 version of pg_bsd_indent.
I thought it would be good to commit this separately,
so as to document the differences between 2.0 and 2.1 behavior.
Discussion: https://postgr.es/m/16296.1558103386@sss.pgh.pa.us
Commit dd299df8 made nbtree treat heap TID as a tiebreaker column,
establishing the principle that there is only one correct location (page
and page offset number) for every index tuple, no matter what.
Insertions of tuples into non-unique indexes proceed as if heap TID
(scan key's scantid) is just another user-attribute value, but
insertions into unique indexes are more delicate. The TID value in
scantid must initially be omitted to ensure that the unique index
insertion visits every leaf page that duplicates could be on. The
scantid is set once again after unique checking finishes successfully,
which can force _bt_findinsertloc() to step right one or more times, to
locate the leaf page that the new tuple must be inserted on.
Stepping right within _bt_findinsertloc() was assumed to occur no more
frequently than stepping right within _bt_check_unique(), but there was
one important case where that assumption was incorrect: inserting a
"duplicate" with NULL values. Since _bt_check_unique() didn't do any
real work in this case, it wasn't appropriate for _bt_findinsertloc() to
behave as if it was finishing off a conventional unique insertion, where
any existing physical duplicate must be dead or recently dead.
_bt_findinsertloc() might have to grovel through a substantial portion
of all of the leaf pages in the index to insert a single tuple, even
when there were no dead tuples.
To fix, treat insertions of tuples with NULLs into a unique index as if
they were insertions into a non-unique index: never unset scantid before
calling _bt_search() to descend the tree, and bypass _bt_check_unique()
entirely. _bt_check_unique() is no longer responsible for incoming
tuples with NULL values.
Discussion: https://postgr.es/m/CAH2-Wzm08nr+JPx4jMOa9CGqxWYDQ-_D4wtPBiKghXAUiUy-nQ@mail.gmail.com
Teach nbtree forward index scans to check the high key before moving to
the right sibling page in the hope of finding that it isn't actually
necessary to do so. The new check may indicate that the scan definitely
cannot find matching tuples to the right, ending the scan immediately.
We already opportunistically force a similar "continuescan orientated"
key check of the final non-pivot tuple when it's clear that it cannot be
returned to the scan due to being dead-to-all. The new high key check
is complementary.
The new approach for forward scans is more effective than checking the
final non-pivot tuple, especially with composite indexes and non-unique
indexes. The improvements to the logic for picking a split point added
by commit fab25024 make it likely that relatively dissimilar high keys
will appear on a page. A distinguishing key value that can only appear
on non-pivot tuples on the right sibling page will often be present in
leaf page high keys.
Since forcing the final item to be key checked no longer makes any
difference in the case of forward scans, the existing extra key check is
now only used for backwards scans. Backward scans continue to
opportunistically check the final non-pivot tuple, which is actually the
first non-pivot tuple on the page (not the last).
Note that even pg_upgrade'd v3 indexes make use of this optimization.
Author: Peter Geoghegan, Heikki Linnakangas
Reviewed-By: Heikki Linnakangas
Discussion: https://postgr.es/m/CAH2-WzkOmUduME31QnuTFpimejuQoiZ-HOf0pOWeFZNhTMctvA@mail.gmail.com
Make nbtree treat all index tuples as having a heap TID attribute.
Index searches can distinguish duplicates by heap TID, since heap TID is
always guaranteed to be unique. This general approach has numerous
benefits for performance, and is prerequisite to teaching VACUUM to
perform "retail index tuple deletion".
Naively adding a new attribute to every pivot tuple has unacceptable
overhead (it bloats internal pages), so suffix truncation of pivot
tuples is added. This will usually truncate away the "extra" heap TID
attribute from pivot tuples during a leaf page split, and may also
truncate away additional user attributes. This can increase fan-out,
especially in a multi-column index. Truncation can only occur at the
attribute granularity, which isn't particularly effective, but works
well enough for now. A future patch may add support for truncating
"within" text attributes by generating truncated key values using new
opclass infrastructure.
Only new indexes (BTREE_VERSION 4 indexes) will have insertions that
treat heap TID as a tiebreaker attribute, or will have pivot tuples
undergo suffix truncation during a leaf page split (on-disk
compatibility with versions 2 and 3 is preserved). Upgrades to version
4 cannot be performed on-the-fly, unlike upgrades from version 2 to
version 3. contrib/amcheck continues to work with version 2 and 3
indexes, while also enforcing stricter invariants when verifying version
4 indexes. These stricter invariants are the same invariants described
by "3.1.12 Sequencing" from the Lehman and Yao paper.
A later patch will enhance the logic used by nbtree to pick a split
point. This patch is likely to negatively impact performance without
smarter choices around the precise point to split leaf pages at. Making
these two mostly-distinct sets of enhancements into distinct commits
seems like it might clarify their design, even though neither commit is
particularly useful on its own.
The maximum allowed size of new tuples is reduced by an amount equal to
the space required to store an extra MAXALIGN()'d TID in a new high key
during leaf page splits. The user-facing definition of the "1/3 of a
page" restriction is already imprecise, and so does not need to be
revised. However, there should be a compatibility note in the v12
release notes.
Author: Peter Geoghegan
Reviewed-By: Heikki Linnakangas, Alexander Korotkov
Discussion: https://postgr.es/m/CAH2-WzkVb0Kom=R+88fDFb=JSxZMFvbHVC6Mn9LJ2n=X=kS-Uw@mail.gmail.com
Use dedicated struct to represent nbtree insertion scan keys. Having a
dedicated struct makes the difference between search type scankeys and
insertion scankeys a lot clearer, and simplifies the signature of
several related functions. This is based on a suggestion by Andrey
Lepikhov.
Streamline how unique index insertions cache binary search progress.
Cache the state of in-progress binary searches within _bt_check_unique()
for later instead of having callers avoid repeating the binary search in
an ad-hoc manner. This makes it easy to add a new optimization:
_bt_check_unique() now falls out of its loop immediately in the common
case where it's already clear that there couldn't possibly be a
duplicate.
The new _bt_check_unique() scheme makes it a lot easier to manage cached
binary search effort afterwards, from within _bt_findinsertloc(). This
is needed for the upcoming patch to make nbtree tuples unique by
treating heap TID as a final tiebreaker column. Unique key binary
searches need to restore lower and upper bounds. They cannot simply
continue to use the >= lower bound as the offset to insert at, because
the heap TID tiebreaker column must be used in comparisons for the
restored binary search (unlike the original _bt_check_unique() binary
search, where scankey's heap TID column must be omitted).
Author: Peter Geoghegan, Heikki Linnakangas
Reviewed-By: Heikki Linnakangas, Andrey Lepikhov
Discussion: https://postgr.es/m/CAH2-WzmE6AhUdk9NdWBf4K3HjWXZBX3+umC7mH7+WDrKcRtsOw@mail.gmail.com
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.dehttps://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
Most of these had been obsoleted by 568d4138c / the SnapshotNow
removal.
This is is preparation for moving most of tqual.[ch] into either
snapmgr.h or heapam.h, which in turn is in preparation for pluggable
table AMs.
Author: Andres Freund
Discussion: https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
