public inbox for [email protected]
help / color / mirror / Atom feedFrom: Melanie Plageman <[email protected]>
To: Kirill Reshke <[email protected]>
Cc: Andres Freund <[email protected]>
Cc: Robert Haas <[email protected]>
Cc: Andrey Borodin <[email protected]>
Cc: PostgreSQL Hackers <[email protected]>
Cc: Heikki Linnakangas <[email protected]>
Cc: Chao Li <[email protected]>
Subject: Re: eliminate xl_heap_visible to reduce WAL (and eventually set VM on-access)
Date: Wed, 17 Dec 2025 19:30:01 -0500
Message-ID: <CAAKRu_Yt6EH5aFSJBm-k7PrNM4bTt56fTRbyU7gqYXe4cW+F9g@mail.gmail.com> (raw)
In-Reply-To: <CALdSSPhtPK36_sr9yFo3cN9TXQmjvSX3BZXCF8fVBLX-GETD0Q@mail.gmail.com>
References: <CAAKRu_ZP-3=SaZykpwDBMJOdUKyW3Wm5JZfPFRR3L5Ac8ouq4w@mail.gmail.com>
<CAAKRu_bgkOQqu3K5n4YLRsNBZqJ9Rjg80ROqgKSr2UGz4b5hUg@mail.gmail.com>
<2wk7jo4m4qwh5sn33pfgerdjfujebbccsmmlownybddbh6nawl@mdyyqpqzxjek>
<CAAKRu_YR-COJ9aGnMUQqt5yoWmUBjikqrd4jGNZYouHaXpis9g@mail.gmail.com>
<CALdSSPhiCwJwWwgJP1NmqRmnp9RS2tGOBY0gQrfLCbB+OS5_KQ@mail.gmail.com>
<CAAKRu_YS+Ocm=OzMaZnG4egFiE9v4VYfZ25DXd6jbwegqmGYbQ@mail.gmail.com>
<CAAKRu_ZGZSqhGt-RcmmfiSheC+1fjQdxy6_+oM-1jMn8hyVptQ@mail.gmail.com>
<CALdSSPg+B8RTzTXhJvCcKJBqgzhPZkq0E2oqxQdv74ZNZOMVzg@mail.gmail.com>
<CAAKRu_Zha7HcdQBv8tTtQrcry5332J6kHnOc1X=TT03LzUXDow@mail.gmail.com>
<CAAKRu_amF00f2T_H8N6pbbe75C22EeX1OqA=svpj8LNO1sdUuw@mail.gmail.com>
<mhf4vkmh3j57zx7vuxp4jagtdzwhu3573pgfpmnjwqa6i6yj5y@sy4ymcdtdklo>
<CAAKRu_agCLAQ9OjmrTdJe-X=Xr7QnU4d=cfxdQGwc9jNx9w31w@mail.gmail.com>
<CAAKRu_ayWLg=WDGZZfSPWf0KjPM8u=LBb0D6XaEWyx2_YFFwAQ@mail.gmail.com>
<CALdSSPhtPK36_sr9yFo3cN9TXQmjvSX3BZXCF8fVBLX-GETD0Q@mail.gmail.com>
Thanks for the review!
In addition to addressing your feedback, attached v28 includes a
number of small fixes to comments, commit messages, and other things.
Notably, I've added one new refactoring patch 0009, which reduces the
diff of 0010 -- using the GlobalVisState instead of OldestXmin for
page visibility -- even further.
On Wed, Dec 17, 2025 at 1:27 PM Kirill Reshke <[email protected]> wrote:
>
> > I've done this. I've actually added three such verifications -- one
> > after each step where the VM is expected to change. It shouldn't be
> > very expensive, so I think it is okay. The way the test would fail if
> > the buffer wasn't correctly dirtied is that it would assert out -- so
> > the visibility map test wouldn't even have a chance to fail. But, I
> > think it is also okay to confirm that the expected things are
> > happening with the VM -- it just gives us extra coverage.
>
> +1 on extra coverage. Should we also do sql-level check that the VM
> indeed does not need to set PD_ALL_VISIBLE (check header bytes using
> pageinspect?).
That's an interesting idea. I checked and, AFAICT, there are no tests
currently directly comparing the flags column returned by the
pageinspect page_header() function to one of the flag values. I've
added the following to attached v28.
SELECT (flags & x'0004'::int) <> 0
FROM page_header(get_raw_page('test_vac_unmodified_heap', 0));
But I'm not sure if it is weird/confusing to be comparing the flag
directly to the number 4 like this. I don't really want to bother with
adding another function to pageinspect returning the status of
PD_ALL_VISIBLE (like page_visible() or something).
> v27-0003 & v27-0004: I did not get the exact reason we introduced
> `identify_and_fix_vm_corruption` in 0003 and moved code in 0004 to
> another place. I can see we have this starting v25 of patch set. Well,
> maybe this is not an issue at all...
It's mostly for ease of review. This is a pretty sensitive area of
code, so I thought it would be easier for the reviewer to confirm
correctness if I split it up. Andres had mentioned that the commit was
hard to review because so many different things were happening.
In v27, 0003 moves the VM clear code into a helper. 0004 and 0005
moves all the VM setting/clearing code to
heap_page_prune_and_freeze(). And 0006 actually sets the VM in the
same critical section as pruning/freezing and emits a single WAL
record.
I'm not really sure which commits should stay independent in the final
version I push to master.
> in v27-0005. This patch changes code which is not exercised in
> tests[0]. I spent some time understanding the conditions when we
> entered this. There is a comment about non-finished relation
> extension, but I got no success trying to reproduce this. I ended up
> modifying code to lose PageSetAllVisible in proper places and running
> vacuum. Looks like everything works as expected. I will spend some
> more time on this, maybe I will be successful in writing an
> injection-point-based TAP test which hits this...
Based on the coverage report link you provided, that code is changed
by v27 0007, not 0005. 0005 is about moving an assertion out of
lazy_scan_prune(). 0007 changes lazy_scan_new_or_empty() (the code in
question).
Regarding 0007, it looks like what is uncovered (the orange bits in
the coverage report are uncovered, I assume) is empty pages _without_
PD_ALL_VISIBLE set. I don't see anywhere where PageSetAllVisible() is
called except vacuum and COPY FREEZE.
If I was trying to guess how empty pages with PD_ALL_VISIBLE set are
getting vacuumed, I would think it is due to SKIP_PAGES_THRESHOLD
causing us to vacuum an all-frozen empty page.
Then the question is, why wouldn't we have coverage of the empty page
first being set all-visible/all-frozen? It can't be COPY FREEZE
because the page is empty. And it can't be vacuum, because then we
would have coverage. It's very mysterious.
It would be good to have coverage for this case. I don't think you'll
need an injection point for the main case of "empty page not yet set
all-visible is vacuumed for the first time" (unless I'm
misunderstanding something).
I'm not sure how you'll test the "vacuuming an empty, previously
uninitialized page" case described in this comment, though.
* It's possible that another backend has extended the heap,
* initialized the page, and then failed to WAL-log the page due
* to an ERROR. Since heap extension is not WAL-logged, recovery
* might try to replay our record setting the page all-visible and
* find that the page isn't initialized, which will cause a PANIC.
* To prevent that, check whether the page has been previously
* WAL-logged, and if not, do that now.
You'd want to force an error during relation extension and then vacuum
the page. I don't know if you need an injection point to force the
error -- depends on what kind of error, I think.
So that I know for attribution, did you review 0003-0005?
- Melanie
Attachments:
[text/x-patch] v28-0001-Combine-visibilitymap_set-cases-in-lazy_scan_pru.patch (10.0K, 2-v28-0001-Combine-visibilitymap_set-cases-in-lazy_scan_pru.patch)
download | inline diff:
From ee7c5f860799f195644e2fedf2b63b6789045cbc Mon Sep 17 00:00:00 2001
From: Melanie Plageman <[email protected]>
Date: Mon, 8 Dec 2025 15:49:54 -0500
Subject: [PATCH v28 01/15] Combine visibilitymap_set() cases in
lazy_scan_prune()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
lazy_scan_prune() previously had two separate cases that called
visibilitymap_set() after pruning and freezing. These branches were
nearly identical except that one attempted to avoid dirtying the heap
buffer. However, that situation can never occur — the heap buffer cannot
be clean at that point (and we would hit an assertion if it were).
In lazy_scan_prune(), when we change a previously all-visible page to
all-frozen and the page was recorded as all-visible in the visibility
map by find_next_unskippable_block(), the heap buffer will always be
dirty. Either we have just frozen a tuple and already dirtied the
buffer, or the buffer was modified between find_next_unskippable_block()
and heap_page_prune_and_freeze() and then pruned in
heap_page_prune_and_freeze().
Additionally, XLogRegisterBuffer() asserts that the buffer is dirty, so
attempting to add a clean heap buffer to the WAL chain would assert out
anyway.
Since the “clean heap buffer with already set VM” case is impossible,
the two visibilitymap_set() branches in lazy_scan_prune() can be merged.
Doing so makes the intent clearer and emphasizes that the heap buffer
must always be marked dirty before being added to the WAL chain.
This commit also adds a test case for vacuuming when no heap
modifications are required. Currently this ensures that the heap buffer
is marked dirty before it is added to the WAL chain, but if we later
remove the heap buffer from the VM-set WAL chain or pass it with the
REGBUF_NO_CHANGES flag, this test would guard that behavior.
Author: Melanie Plageman <[email protected]>
Reviewed-by: Chao Li <[email protected]>
Reviewed-by: Srinath Reddy Sadipiralla <[email protected]>
Reviewed-by: Kirill Reshke <[email protected]>
Discussion: https://postgr.es/m/5CEAA162-67B1-44DA-B60D-8B65717E8B05%40gmail.com
---
.../pg_visibility/expected/pg_visibility.out | 44 ++++++++++
contrib/pg_visibility/sql/pg_visibility.sql | 20 +++++
src/backend/access/heap/vacuumlazy.c | 87 ++++---------------
3 files changed, 82 insertions(+), 69 deletions(-)
diff --git a/contrib/pg_visibility/expected/pg_visibility.out b/contrib/pg_visibility/expected/pg_visibility.out
index 09fa5933a35..e10f1706015 100644
--- a/contrib/pg_visibility/expected/pg_visibility.out
+++ b/contrib/pg_visibility/expected/pg_visibility.out
@@ -1,4 +1,5 @@
CREATE EXTENSION pg_visibility;
+CREATE EXTENSION pageinspect;
--
-- recently-dropped table
--
@@ -204,6 +205,49 @@ select pg_truncate_visibility_map('test_partition');
(1 row)
+-- test the case where vacuum phase I does not need to modify the heap buffer
+-- and only needs to set the VM
+create table test_vac_unmodified_heap(a int);
+insert into test_vac_unmodified_heap values (1);
+vacuum (freeze) test_vac_unmodified_heap;
+select pg_visibility_map_summary('test_vac_unmodified_heap');
+ pg_visibility_map_summary
+---------------------------
+ (1,1)
+(1 row)
+
+-- the checkpoint cleans the buffer dirtied by freezing the sole tuple
+checkpoint;
+-- truncating the VM ensures that the next vacuum will need to set it
+select pg_truncate_visibility_map('test_vac_unmodified_heap');
+ pg_truncate_visibility_map
+----------------------------
+
+(1 row)
+
+select pg_visibility_map_summary('test_vac_unmodified_heap');
+ pg_visibility_map_summary
+---------------------------
+ (0,0)
+(1 row)
+
+-- though the VM is truncated, the heap page-level visibility hint,
+-- PD_ALL_VISIBLE should still be set
+SELECT (flags & x'0004'::int) <> 0
+ FROM page_header(get_raw_page('test_vac_unmodified_heap', 0));
+ ?column?
+----------
+ t
+(1 row)
+
+-- vacuum sets the VM
+vacuum test_vac_unmodified_heap;
+select pg_visibility_map_summary('test_vac_unmodified_heap');
+ pg_visibility_map_summary
+---------------------------
+ (1,1)
+(1 row)
+
-- test copy freeze
create table copyfreeze (a int, b char(1500));
-- load all rows via COPY FREEZE and ensure that all pages are set all-visible
diff --git a/contrib/pg_visibility/sql/pg_visibility.sql b/contrib/pg_visibility/sql/pg_visibility.sql
index 5af06ec5b76..57af8a0c5b6 100644
--- a/contrib/pg_visibility/sql/pg_visibility.sql
+++ b/contrib/pg_visibility/sql/pg_visibility.sql
@@ -1,4 +1,5 @@
CREATE EXTENSION pg_visibility;
+CREATE EXTENSION pageinspect;
--
-- recently-dropped table
@@ -94,6 +95,25 @@ select count(*) > 0 from pg_visibility_map_summary('test_partition');
select * from pg_check_frozen('test_partition'); -- hopefully none
select pg_truncate_visibility_map('test_partition');
+-- test the case where vacuum phase I does not need to modify the heap buffer
+-- and only needs to set the VM
+create table test_vac_unmodified_heap(a int);
+insert into test_vac_unmodified_heap values (1);
+vacuum (freeze) test_vac_unmodified_heap;
+select pg_visibility_map_summary('test_vac_unmodified_heap');
+-- the checkpoint cleans the buffer dirtied by freezing the sole tuple
+checkpoint;
+-- truncating the VM ensures that the next vacuum will need to set it
+select pg_truncate_visibility_map('test_vac_unmodified_heap');
+select pg_visibility_map_summary('test_vac_unmodified_heap');
+-- though the VM is truncated, the heap page-level visibility hint,
+-- PD_ALL_VISIBLE should still be set
+SELECT (flags & x'0004'::int) <> 0
+ FROM page_header(get_raw_page('test_vac_unmodified_heap', 0));
+-- vacuum sets the VM
+vacuum test_vac_unmodified_heap;
+select pg_visibility_map_summary('test_vac_unmodified_heap');
+
-- test copy freeze
create table copyfreeze (a int, b char(1500));
diff --git a/src/backend/access/heap/vacuumlazy.c b/src/backend/access/heap/vacuumlazy.c
index 30778a15639..cecba2146ea 100644
--- a/src/backend/access/heap/vacuumlazy.c
+++ b/src/backend/access/heap/vacuumlazy.c
@@ -2093,16 +2093,14 @@ lazy_scan_prune(LVRelState *vacrel,
* of last heap_vac_scan_next_block() call), and from all_visible and
* all_frozen variables
*/
- if (!all_visible_according_to_vm && presult.all_visible)
+ if ((presult.all_visible && !all_visible_according_to_vm) ||
+ (presult.all_frozen && !VM_ALL_FROZEN(rel, blkno, &vmbuffer)))
{
uint8 old_vmbits;
uint8 flags = VISIBILITYMAP_ALL_VISIBLE;
if (presult.all_frozen)
- {
- Assert(!TransactionIdIsValid(presult.vm_conflict_horizon));
flags |= VISIBILITYMAP_ALL_FROZEN;
- }
/*
* It should never be the case that the visibility map page is set
@@ -2110,15 +2108,25 @@ lazy_scan_prune(LVRelState *vacrel,
* checksums are not enabled). Regardless, set both bits so that we
* get back in sync.
*
- * NB: If the heap page is all-visible but the VM bit is not set, we
- * don't need to dirty the heap page. However, if checksums are
- * enabled, we do need to make sure that the heap page is dirtied
- * before passing it to visibilitymap_set(), because it may be logged.
- * Given that this situation should only happen in rare cases after a
- * crash, it is not worth optimizing.
+ * Even if PD_ALL_VISIBLE is already set, we don't need to worry about
+ * unnecessarily dirtying the heap buffer. Nearly the only scenario
+ * where PD_ALL_VISIBLE is set but the VM is not is if the VM was
+ * removed -- and that isn't worth optimizing for. And if we add the
+ * heap buffer to the WAL chain (without passing REGBUF_NO_CHANGES),
+ * it must be marked dirty.
*/
PageSetAllVisible(page);
MarkBufferDirty(buf);
+
+ /*
+ * If the page is being set all-frozen, we pass InvalidTransactionId
+ * as the cutoff_xid, since a snapshot conflict horizon sufficient to
+ * make everything safe for REDO was logged when the page's tuples
+ * were frozen.
+ */
+ Assert(!presult.all_frozen ||
+ !TransactionIdIsValid(presult.vm_conflict_horizon));
+
old_vmbits = visibilitymap_set(vacrel->rel, blkno, buf,
InvalidXLogRecPtr,
vmbuffer, presult.vm_conflict_horizon,
@@ -2190,65 +2198,6 @@ lazy_scan_prune(LVRelState *vacrel,
VISIBILITYMAP_VALID_BITS);
}
- /*
- * If the all-visible page is all-frozen but not marked as such yet, mark
- * it as all-frozen.
- */
- else if (all_visible_according_to_vm && presult.all_frozen &&
- !VM_ALL_FROZEN(vacrel->rel, blkno, &vmbuffer))
- {
- uint8 old_vmbits;
-
- /*
- * Avoid relying on all_visible_according_to_vm as a proxy for the
- * page-level PD_ALL_VISIBLE bit being set, since it might have become
- * stale -- even when all_visible is set
- */
- if (!PageIsAllVisible(page))
- {
- PageSetAllVisible(page);
- MarkBufferDirty(buf);
- }
-
- /*
- * Set the page all-frozen (and all-visible) in the VM.
- *
- * We can pass InvalidTransactionId as our cutoff_xid, since a
- * snapshotConflictHorizon sufficient to make everything safe for REDO
- * was logged when the page's tuples were frozen.
- */
- Assert(!TransactionIdIsValid(presult.vm_conflict_horizon));
- old_vmbits = visibilitymap_set(vacrel->rel, blkno, buf,
- InvalidXLogRecPtr,
- vmbuffer, InvalidTransactionId,
- VISIBILITYMAP_ALL_VISIBLE |
- VISIBILITYMAP_ALL_FROZEN);
-
- /*
- * The page was likely already set all-visible in the VM. However,
- * there is a small chance that it was modified sometime between
- * setting all_visible_according_to_vm and checking the visibility
- * during pruning. Check the return value of old_vmbits anyway to
- * ensure the visibility map counters used for logging are accurate.
- */
- if ((old_vmbits & VISIBILITYMAP_ALL_VISIBLE) == 0)
- {
- vacrel->vm_new_visible_pages++;
- vacrel->vm_new_visible_frozen_pages++;
- *vm_page_frozen = true;
- }
-
- /*
- * We already checked that the page was not set all-frozen in the VM
- * above, so we don't need to test the value of old_vmbits.
- */
- else
- {
- vacrel->vm_new_frozen_pages++;
- *vm_page_frozen = true;
- }
- }
-
return presult.ndeleted;
}
--
2.43.0
[text/x-patch] v28-0002-Eliminate-use-of-cached-VM-value-in-lazy_scan_pr.patch (13.6K, 3-v28-0002-Eliminate-use-of-cached-VM-value-in-lazy_scan_pr.patch)
download | inline diff:
From 0dac7060ae0eddc2617a1919150757a7e63924f3 Mon Sep 17 00:00:00 2001
From: Melanie Plageman <[email protected]>
Date: Thu, 11 Dec 2025 10:48:13 -0500
Subject: [PATCH v28 02/15] Eliminate use of cached VM value in
lazy_scan_prune()
lazy_scan_prune() takes a parameter from lazy_scan_heap() indicating
whether the page was marked all-visible in the VM at the time it was
last checked in find_next_unskippable_block(). This behavior is
historical, dating back to commit 608195a3a365, when we did not pin the
VM page until confirming it was not all-visible. Now that the VM page is
already pinned, there is no meaningful benefit to relying on a cached VM
status.
Removing this cached value simplifies the logic in both lazy_scan_heap()
and lazy_scan_prune(). It also clarifies future work that will set the
visibility map on-access: such paths will not have a cached value
available which would make the logic harder to reason about. Eliminating
it also enables us to detect and repair VM corruption on-access.
Along with removing the cached value and unconditionally checking the
visibility status of the heap page, this commit also moves the VM
corruption handling to occur first. This reordering should have no
performance impact, since the checks are inexpensive and performed only
once per page. It does, however, make the control flow easier to
understand. The new restructuring also makes it possible that after
fixing corruption, the VM could be newly set, if pruning found the page
all-visible.
Author: Melanie Plageman <[email protected]>
Discussion: https://postgr.es/m/5CEAA162-67B1-44DA-B60D-8B65717E8B05%40gmail.com
---
src/backend/access/heap/vacuumlazy.c | 178 ++++++++++++---------------
1 file changed, 79 insertions(+), 99 deletions(-)
diff --git a/src/backend/access/heap/vacuumlazy.c b/src/backend/access/heap/vacuumlazy.c
index cecba2146ea..9d2523a55b2 100644
--- a/src/backend/access/heap/vacuumlazy.c
+++ b/src/backend/access/heap/vacuumlazy.c
@@ -248,13 +248,6 @@ typedef enum
*/
#define EAGER_SCAN_REGION_SIZE 4096
-/*
- * heap_vac_scan_next_block() sets these flags to communicate information
- * about the block it read to the caller.
- */
-#define VAC_BLK_WAS_EAGER_SCANNED (1 << 0)
-#define VAC_BLK_ALL_VISIBLE_ACCORDING_TO_VM (1 << 1)
-
typedef struct LVRelState
{
/* Target heap relation and its indexes */
@@ -358,7 +351,6 @@ typedef struct LVRelState
/* State maintained by heap_vac_scan_next_block() */
BlockNumber current_block; /* last block returned */
BlockNumber next_unskippable_block; /* next unskippable block */
- bool next_unskippable_allvis; /* its visibility status */
bool next_unskippable_eager_scanned; /* if it was eagerly scanned */
Buffer next_unskippable_vmbuffer; /* buffer containing its VM bit */
@@ -432,7 +424,7 @@ static bool lazy_scan_new_or_empty(LVRelState *vacrel, Buffer buf,
bool sharelock, Buffer vmbuffer);
static int lazy_scan_prune(LVRelState *vacrel, Buffer buf,
BlockNumber blkno, Page page,
- Buffer vmbuffer, bool all_visible_according_to_vm,
+ Buffer vmbuffer,
bool *has_lpdead_items, bool *vm_page_frozen);
static bool lazy_scan_noprune(LVRelState *vacrel, Buffer buf,
BlockNumber blkno, Page page,
@@ -1248,7 +1240,6 @@ lazy_scan_heap(LVRelState *vacrel)
/* Initialize for the first heap_vac_scan_next_block() call */
vacrel->current_block = InvalidBlockNumber;
vacrel->next_unskippable_block = InvalidBlockNumber;
- vacrel->next_unskippable_allvis = false;
vacrel->next_unskippable_eager_scanned = false;
vacrel->next_unskippable_vmbuffer = InvalidBuffer;
@@ -1264,13 +1255,13 @@ lazy_scan_heap(LVRelState *vacrel)
MAIN_FORKNUM,
heap_vac_scan_next_block,
vacrel,
- sizeof(uint8));
+ sizeof(bool));
while (true)
{
Buffer buf;
Page page;
- uint8 blk_info = 0;
+ bool was_eager_scanned = false;
int ndeleted = 0;
bool has_lpdead_items;
void *per_buffer_data = NULL;
@@ -1339,13 +1330,13 @@ lazy_scan_heap(LVRelState *vacrel)
if (!BufferIsValid(buf))
break;
- blk_info = *((uint8 *) per_buffer_data);
+ was_eager_scanned = *((bool *) per_buffer_data);
CheckBufferIsPinnedOnce(buf);
page = BufferGetPage(buf);
blkno = BufferGetBlockNumber(buf);
vacrel->scanned_pages++;
- if (blk_info & VAC_BLK_WAS_EAGER_SCANNED)
+ if (was_eager_scanned)
vacrel->eager_scanned_pages++;
/* Report as block scanned, update error traceback information */
@@ -1416,7 +1407,6 @@ lazy_scan_heap(LVRelState *vacrel)
if (got_cleanup_lock)
ndeleted = lazy_scan_prune(vacrel, buf, blkno, page,
vmbuffer,
- blk_info & VAC_BLK_ALL_VISIBLE_ACCORDING_TO_VM,
&has_lpdead_items, &vm_page_frozen);
/*
@@ -1433,8 +1423,7 @@ lazy_scan_heap(LVRelState *vacrel)
* exclude pages skipped due to cleanup lock contention from eager
* freeze algorithm caps.
*/
- if (got_cleanup_lock &&
- (blk_info & VAC_BLK_WAS_EAGER_SCANNED))
+ if (got_cleanup_lock && was_eager_scanned)
{
/* Aggressive vacuums do not eager scan. */
Assert(!vacrel->aggressive);
@@ -1601,7 +1590,6 @@ heap_vac_scan_next_block(ReadStream *stream,
{
BlockNumber next_block;
LVRelState *vacrel = callback_private_data;
- uint8 blk_info = 0;
/* relies on InvalidBlockNumber + 1 overflowing to 0 on first call */
next_block = vacrel->current_block + 1;
@@ -1664,8 +1652,8 @@ heap_vac_scan_next_block(ReadStream *stream,
* otherwise they would've been unskippable.
*/
vacrel->current_block = next_block;
- blk_info |= VAC_BLK_ALL_VISIBLE_ACCORDING_TO_VM;
- *((uint8 *) per_buffer_data) = blk_info;
+ /* Block was not eager scanned */
+ *((bool *) per_buffer_data) = false;
return vacrel->current_block;
}
else
@@ -1677,11 +1665,7 @@ heap_vac_scan_next_block(ReadStream *stream,
Assert(next_block == vacrel->next_unskippable_block);
vacrel->current_block = next_block;
- if (vacrel->next_unskippable_allvis)
- blk_info |= VAC_BLK_ALL_VISIBLE_ACCORDING_TO_VM;
- if (vacrel->next_unskippable_eager_scanned)
- blk_info |= VAC_BLK_WAS_EAGER_SCANNED;
- *((uint8 *) per_buffer_data) = blk_info;
+ *((bool *) per_buffer_data) = vacrel->next_unskippable_eager_scanned;
return vacrel->current_block;
}
}
@@ -1706,7 +1690,6 @@ find_next_unskippable_block(LVRelState *vacrel, bool *skipsallvis)
BlockNumber next_unskippable_block = vacrel->next_unskippable_block + 1;
Buffer next_unskippable_vmbuffer = vacrel->next_unskippable_vmbuffer;
bool next_unskippable_eager_scanned = false;
- bool next_unskippable_allvis;
*skipsallvis = false;
@@ -1716,7 +1699,6 @@ find_next_unskippable_block(LVRelState *vacrel, bool *skipsallvis)
next_unskippable_block,
&next_unskippable_vmbuffer);
- next_unskippable_allvis = (mapbits & VISIBILITYMAP_ALL_VISIBLE) != 0;
/*
* At the start of each eager scan region, normal vacuums with eager
@@ -1735,7 +1717,7 @@ find_next_unskippable_block(LVRelState *vacrel, bool *skipsallvis)
* A block is unskippable if it is not all visible according to the
* visibility map.
*/
- if (!next_unskippable_allvis)
+ if ((mapbits & VISIBILITYMAP_ALL_VISIBLE) == 0)
{
Assert((mapbits & VISIBILITYMAP_ALL_FROZEN) == 0);
break;
@@ -1792,7 +1774,6 @@ find_next_unskippable_block(LVRelState *vacrel, bool *skipsallvis)
/* write the local variables back to vacrel */
vacrel->next_unskippable_block = next_unskippable_block;
- vacrel->next_unskippable_allvis = next_unskippable_allvis;
vacrel->next_unskippable_eager_scanned = next_unskippable_eager_scanned;
vacrel->next_unskippable_vmbuffer = next_unskippable_vmbuffer;
}
@@ -1953,9 +1934,7 @@ cmpOffsetNumbers(const void *a, const void *b)
* Caller must hold pin and buffer cleanup lock on the buffer.
*
* vmbuffer is the buffer containing the VM block with visibility information
- * for the heap block, blkno. all_visible_according_to_vm is the saved
- * visibility status of the heap block looked up earlier by the caller. We
- * won't rely entirely on this status, as it may be out of date.
+ * for the heap block, blkno.
*
* *has_lpdead_items is set to true or false depending on whether, upon return
* from this function, any LP_DEAD items are still present on the page.
@@ -1972,7 +1951,6 @@ lazy_scan_prune(LVRelState *vacrel,
BlockNumber blkno,
Page page,
Buffer vmbuffer,
- bool all_visible_according_to_vm,
bool *has_lpdead_items,
bool *vm_page_frozen)
{
@@ -1986,6 +1964,8 @@ lazy_scan_prune(LVRelState *vacrel,
.vistest = vacrel->vistest,
.cutoffs = &vacrel->cutoffs,
};
+ uint8 old_vmbits = 0;
+ uint8 new_vmbits = 0;
Assert(BufferGetBlockNumber(buf) == blkno);
@@ -2088,70 +2068,7 @@ lazy_scan_prune(LVRelState *vacrel,
Assert(!presult.all_visible || !(*has_lpdead_items));
Assert(!presult.all_frozen || presult.all_visible);
- /*
- * Handle setting visibility map bit based on information from the VM (as
- * of last heap_vac_scan_next_block() call), and from all_visible and
- * all_frozen variables
- */
- if ((presult.all_visible && !all_visible_according_to_vm) ||
- (presult.all_frozen && !VM_ALL_FROZEN(rel, blkno, &vmbuffer)))
- {
- uint8 old_vmbits;
- uint8 flags = VISIBILITYMAP_ALL_VISIBLE;
-
- if (presult.all_frozen)
- flags |= VISIBILITYMAP_ALL_FROZEN;
-
- /*
- * It should never be the case that the visibility map page is set
- * while the page-level bit is clear, but the reverse is allowed (if
- * checksums are not enabled). Regardless, set both bits so that we
- * get back in sync.
- *
- * Even if PD_ALL_VISIBLE is already set, we don't need to worry about
- * unnecessarily dirtying the heap buffer. Nearly the only scenario
- * where PD_ALL_VISIBLE is set but the VM is not is if the VM was
- * removed -- and that isn't worth optimizing for. And if we add the
- * heap buffer to the WAL chain (without passing REGBUF_NO_CHANGES),
- * it must be marked dirty.
- */
- PageSetAllVisible(page);
- MarkBufferDirty(buf);
-
- /*
- * If the page is being set all-frozen, we pass InvalidTransactionId
- * as the cutoff_xid, since a snapshot conflict horizon sufficient to
- * make everything safe for REDO was logged when the page's tuples
- * were frozen.
- */
- Assert(!presult.all_frozen ||
- !TransactionIdIsValid(presult.vm_conflict_horizon));
-
- old_vmbits = visibilitymap_set(vacrel->rel, blkno, buf,
- InvalidXLogRecPtr,
- vmbuffer, presult.vm_conflict_horizon,
- flags);
-
- /*
- * If the page wasn't already set all-visible and/or all-frozen in the
- * VM, count it as newly set for logging.
- */
- if ((old_vmbits & VISIBILITYMAP_ALL_VISIBLE) == 0)
- {
- vacrel->vm_new_visible_pages++;
- if (presult.all_frozen)
- {
- vacrel->vm_new_visible_frozen_pages++;
- *vm_page_frozen = true;
- }
- }
- else if ((old_vmbits & VISIBILITYMAP_ALL_FROZEN) == 0 &&
- presult.all_frozen)
- {
- vacrel->vm_new_frozen_pages++;
- *vm_page_frozen = true;
- }
- }
+ old_vmbits = visibilitymap_get_status(vacrel->rel, blkno, &vmbuffer);
/*
* As of PostgreSQL 9.2, the visibility map bit should never be set if the
@@ -2159,8 +2076,8 @@ lazy_scan_prune(LVRelState *vacrel,
* cleared after heap_vac_scan_next_block() was called, so we must recheck
* with buffer lock before concluding that the VM is corrupt.
*/
- else if (all_visible_according_to_vm && !PageIsAllVisible(page) &&
- visibilitymap_get_status(vacrel->rel, blkno, &vmbuffer) != 0)
+ if (!PageIsAllVisible(page) &&
+ (old_vmbits & VISIBILITYMAP_ALL_VISIBLE) != 0)
{
ereport(WARNING,
(errcode(ERRCODE_DATA_CORRUPTED),
@@ -2198,6 +2115,69 @@ lazy_scan_prune(LVRelState *vacrel,
VISIBILITYMAP_VALID_BITS);
}
+ if (!presult.all_visible)
+ return presult.ndeleted;
+
+ /* Set the visibility map and page visibility hint */
+ new_vmbits = VISIBILITYMAP_ALL_VISIBLE;
+
+ if (presult.all_frozen)
+ new_vmbits |= VISIBILITYMAP_ALL_FROZEN;
+
+ /* Nothing to do */
+ if (old_vmbits == new_vmbits)
+ return presult.ndeleted;
+
+ Assert(presult.all_visible);
+
+ /*
+ * It should never be the case that the visibility map page is set while
+ * the page-level bit is clear, but the reverse is allowed (if checksums
+ * are not enabled). Regardless, set both bits so that we get back in
+ * sync.
+ *
+ * Even if PD_ALL_VISIBLE is already set, we don't need to worry about
+ * unnecessarily dirtying the heap buffer. Nearly the only scenario where
+ * PD_ALL_VISIBLE is set but the VM is not is if the VM was removed -- and
+ * that isn't worth optimizing for. And if we add the heap buffer to the
+ * WAL chain (without passing REGBUF_NO_CHANGES), it must be marked dirty.
+ */
+ PageSetAllVisible(page);
+ MarkBufferDirty(buf);
+
+ /*
+ * If the page is being set all-frozen, we pass InvalidTransactionId as
+ * the cutoff_xid, since a snapshot conflict horizon sufficient to make
+ * everything safe for REDO was logged when the page's tuples were frozen.
+ */
+ Assert(!presult.all_frozen ||
+ !TransactionIdIsValid(presult.vm_conflict_horizon));
+
+ visibilitymap_set(vacrel->rel, blkno, buf,
+ InvalidXLogRecPtr,
+ vmbuffer, presult.vm_conflict_horizon,
+ new_vmbits);
+
+ /*
+ * If the page wasn't already set all-visible and/or all-frozen in the VM,
+ * count it as newly set for logging.
+ */
+ if ((old_vmbits & VISIBILITYMAP_ALL_VISIBLE) == 0)
+ {
+ vacrel->vm_new_visible_pages++;
+ if (presult.all_frozen)
+ {
+ vacrel->vm_new_visible_frozen_pages++;
+ *vm_page_frozen = true;
+ }
+ }
+ else if ((old_vmbits & VISIBILITYMAP_ALL_FROZEN) == 0 &&
+ presult.all_frozen)
+ {
+ vacrel->vm_new_frozen_pages++;
+ *vm_page_frozen = true;
+ }
+
return presult.ndeleted;
}
--
2.43.0
[text/x-patch] v28-0003-Refactor-lazy_scan_prune-VM-clear-logic-into-hel.patch (6.5K, 4-v28-0003-Refactor-lazy_scan_prune-VM-clear-logic-into-hel.patch)
download | inline diff:
From c23840a12ff14eeffb5116c2cfd34e34e3987b02 Mon Sep 17 00:00:00 2001
From: Melanie Plageman <[email protected]>
Date: Tue, 2 Dec 2025 13:36:39 -0500
Subject: [PATCH v28 03/15] Refactor lazy_scan_prune() VM clear logic into
helper
Encapsulating them in a helper makes the whole function clearer. There
is no functional change other than moving it into a helper.
Reviewed-by: Kirill Reshke <[email protected]>
---
src/backend/access/heap/vacuumlazy.c | 122 +++++++++++++++++----------
1 file changed, 78 insertions(+), 44 deletions(-)
diff --git a/src/backend/access/heap/vacuumlazy.c b/src/backend/access/heap/vacuumlazy.c
index 9d2523a55b2..ff34a99edbd 100644
--- a/src/backend/access/heap/vacuumlazy.c
+++ b/src/backend/access/heap/vacuumlazy.c
@@ -422,6 +422,11 @@ static void find_next_unskippable_block(LVRelState *vacrel, bool *skipsallvis);
static bool lazy_scan_new_or_empty(LVRelState *vacrel, Buffer buf,
BlockNumber blkno, Page page,
bool sharelock, Buffer vmbuffer);
+static void identify_and_fix_vm_corruption(Relation rel, Buffer heap_buffer,
+ BlockNumber heap_blk, Page heap_page,
+ int nlpdead_items,
+ Buffer vmbuffer,
+ uint8 vmbits);
static int lazy_scan_prune(LVRelState *vacrel, Buffer buf,
BlockNumber blkno, Page page,
Buffer vmbuffer,
@@ -1928,6 +1933,77 @@ cmpOffsetNumbers(const void *a, const void *b)
return pg_cmp_u16(*(const OffsetNumber *) a, *(const OffsetNumber *) b);
}
+/*
+ * Helper to correct any corruption detected on an heap page and its
+ * corresponding visibility map page after pruning but before setting the
+ * visibility map. It examines the heap page, the associated VM page, and the
+ * number of dead items previously identified.
+ *
+ * This function must be called while holding an exclusive lock on the heap
+ * buffer, and the dead items must have been discovered under that same lock.
+
+ * The provided vmbits must reflect the current state of the VM block
+ * referenced by vmbuffer. Although we do not hold a lock on the VM buffer, it
+ * is pinned, and the heap buffer is exclusively locked, ensuring that no
+ * other backend can update the VM bits corresponding to this heap page.
+ */
+static void
+identify_and_fix_vm_corruption(Relation rel, Buffer heap_buffer,
+ BlockNumber heap_blk, Page heap_page,
+ int nlpdead_items,
+ Buffer vmbuffer,
+ uint8 vmbits)
+{
+ Assert(visibilitymap_get_status(rel, heap_blk, &vmbuffer) == vmbits);
+
+ Assert(BufferIsLockedByMeInMode(heap_buffer, BUFFER_LOCK_EXCLUSIVE));
+
+ /*
+ * As of PostgreSQL 9.2, the visibility map bit should never be set if the
+ * page-level bit is clear. However, it's possible that the bit got
+ * cleared after heap_vac_scan_next_block() was called, so we must recheck
+ * with buffer lock before concluding that the VM is corrupt.
+ */
+ if (!PageIsAllVisible(heap_page) &&
+ ((vmbits & VISIBILITYMAP_ALL_VISIBLE) != 0))
+ {
+ ereport(WARNING,
+ (errcode(ERRCODE_DATA_CORRUPTED),
+ errmsg("page is not marked all-visible but visibility map bit is set in relation \"%s\" page %u",
+ RelationGetRelationName(rel), heap_blk)));
+
+ visibilitymap_clear(rel, heap_blk, vmbuffer,
+ VISIBILITYMAP_VALID_BITS);
+ }
+
+ /*
+ * It's possible for the value returned by
+ * GetOldestNonRemovableTransactionId() to move backwards, so it's not
+ * wrong for us to see tuples that appear to not be visible to everyone
+ * yet, while PD_ALL_VISIBLE is already set. The real safe xmin value
+ * never moves backwards, but GetOldestNonRemovableTransactionId() is
+ * conservative and sometimes returns a value that's unnecessarily small,
+ * so if we see that contradiction it just means that the tuples that we
+ * think are not visible to everyone yet actually are, and the
+ * PD_ALL_VISIBLE flag is correct.
+ *
+ * There should never be LP_DEAD items on a page with PD_ALL_VISIBLE set,
+ * however.
+ */
+ else if (PageIsAllVisible(heap_page) && nlpdead_items > 0)
+ {
+ ereport(WARNING,
+ (errcode(ERRCODE_DATA_CORRUPTED),
+ errmsg("page containing LP_DEAD items is marked as all-visible in relation \"%s\" page %u",
+ RelationGetRelationName(rel), heap_blk)));
+
+ PageClearAllVisible(heap_page);
+ MarkBufferDirty(heap_buffer);
+ visibilitymap_clear(rel, heap_blk, vmbuffer,
+ VISIBILITYMAP_VALID_BITS);
+ }
+}
+
/*
* lazy_scan_prune() -- lazy_scan_heap() pruning and freezing.
*
@@ -2070,50 +2146,8 @@ lazy_scan_prune(LVRelState *vacrel,
old_vmbits = visibilitymap_get_status(vacrel->rel, blkno, &vmbuffer);
- /*
- * As of PostgreSQL 9.2, the visibility map bit should never be set if the
- * page-level bit is clear. However, it's possible that the bit got
- * cleared after heap_vac_scan_next_block() was called, so we must recheck
- * with buffer lock before concluding that the VM is corrupt.
- */
- if (!PageIsAllVisible(page) &&
- (old_vmbits & VISIBILITYMAP_ALL_VISIBLE) != 0)
- {
- ereport(WARNING,
- (errcode(ERRCODE_DATA_CORRUPTED),
- errmsg("page is not marked all-visible but visibility map bit is set in relation \"%s\" page %u",
- vacrel->relname, blkno)));
-
- visibilitymap_clear(vacrel->rel, blkno, vmbuffer,
- VISIBILITYMAP_VALID_BITS);
- }
-
- /*
- * It's possible for the value returned by
- * GetOldestNonRemovableTransactionId() to move backwards, so it's not
- * wrong for us to see tuples that appear to not be visible to everyone
- * yet, while PD_ALL_VISIBLE is already set. The real safe xmin value
- * never moves backwards, but GetOldestNonRemovableTransactionId() is
- * conservative and sometimes returns a value that's unnecessarily small,
- * so if we see that contradiction it just means that the tuples that we
- * think are not visible to everyone yet actually are, and the
- * PD_ALL_VISIBLE flag is correct.
- *
- * There should never be LP_DEAD items on a page with PD_ALL_VISIBLE set,
- * however.
- */
- else if (presult.lpdead_items > 0 && PageIsAllVisible(page))
- {
- ereport(WARNING,
- (errcode(ERRCODE_DATA_CORRUPTED),
- errmsg("page containing LP_DEAD items is marked as all-visible in relation \"%s\" page %u",
- vacrel->relname, blkno)));
-
- PageClearAllVisible(page);
- MarkBufferDirty(buf);
- visibilitymap_clear(vacrel->rel, blkno, vmbuffer,
- VISIBILITYMAP_VALID_BITS);
- }
+ identify_and_fix_vm_corruption(vacrel->rel, buf, blkno, page,
+ presult.lpdead_items, vmbuffer, old_vmbits);
if (!presult.all_visible)
return presult.ndeleted;
--
2.43.0
[text/x-patch] v28-0004-Set-the-VM-in-heap_page_prune_and_freeze.patch (26.4K, 5-v28-0004-Set-the-VM-in-heap_page_prune_and_freeze.patch)
download | inline diff:
From 014abb83438cf3a3600f34b1060bca430f572275 Mon Sep 17 00:00:00 2001
From: Melanie Plageman <[email protected]>
Date: Tue, 2 Dec 2025 15:07:42 -0500
Subject: [PATCH v28 04/15] Set the VM in heap_page_prune_and_freeze()
This has no independent benefit. It is meant for ease of review. As of
this commit, there is still a separate WAL record emitted for setting
the VM after pruning and freezing. But it is easier to review if moving
the logic into pruneheap.c is separate from setting the VM in the same
WAL record.
---
src/backend/access/heap/pruneheap.c | 302 +++++++++++++++++++++++----
src/backend/access/heap/vacuumlazy.c | 142 +------------
src/include/access/heapam.h | 21 ++
3 files changed, 286 insertions(+), 179 deletions(-)
diff --git a/src/backend/access/heap/pruneheap.c b/src/backend/access/heap/pruneheap.c
index 07aa08cfe14..14d40476be9 100644
--- a/src/backend/access/heap/pruneheap.c
+++ b/src/backend/access/heap/pruneheap.c
@@ -19,7 +19,7 @@
#include "access/htup_details.h"
#include "access/multixact.h"
#include "access/transam.h"
-#include "access/visibilitymapdefs.h"
+#include "access/visibilitymap.h"
#include "access/xlog.h"
#include "access/xloginsert.h"
#include "commands/vacuum.h"
@@ -44,6 +44,8 @@ typedef struct
bool mark_unused_now;
/* whether to attempt freezing tuples */
bool attempt_freeze;
+ /* whether or not to attempt updating the VM */
+ bool attempt_update_vm;
struct VacuumCutoffs *cutoffs;
/*-------------------------------------------------------
@@ -140,16 +142,17 @@ typedef struct
* all_visible and all_frozen indicate if the all-visible and all-frozen
* bits in the visibility map can be set for this page after pruning.
*
- * visibility_cutoff_xid is the newest xmin of live tuples on the page.
- * The caller can use it as the conflict horizon, when setting the VM
- * bits. It is only valid if we froze some tuples, and all_frozen is
- * true.
+ * visibility_cutoff_xid is the newest xmin of live tuples on the page. It
+ * can be used as the conflict horizon when setting the VM or when
+ * freezing all the tuples on the page. It is only valid when all the live
+ * tuples on the page are all-visible.
*
* NOTE: all_visible and all_frozen initially don't include LP_DEAD items.
* That's convenient for heap_page_prune_and_freeze() to use them to
- * decide whether to freeze the page or not. The all_visible and
- * all_frozen values returned to the caller are adjusted to include
- * LP_DEAD items after we determine whether to opportunistically freeze.
+ * decide whether to opportunistically freeze the page or not. The
+ * all_visible and all_frozen values ultimately used to set the VM are
+ * adjusted to include LP_DEAD items after we determine whether or not to
+ * opportunistically freeze.
*/
bool all_visible;
bool all_frozen;
@@ -191,6 +194,17 @@ static void page_verify_redirects(Page page);
static bool heap_page_will_freeze(Relation relation, Buffer buffer,
bool did_tuple_hint_fpi, bool do_prune, bool do_hint_prune,
PruneState *prstate);
+static void identify_and_fix_vm_corruption(Relation rel, Buffer heap_buffer,
+ BlockNumber heap_blk, Page heap_page, int nlpdead_items,
+ Buffer vmbuffer,
+ uint8 vmbits);
+static bool heap_page_will_set_vm(PruneState *prstate,
+ Relation relation,
+ BlockNumber heap_blk, Buffer heap_buffer, Page heap_page,
+ Buffer vmbuffer,
+ int nlpdead_items,
+ uint8 *old_vmbits,
+ uint8 *new_vmbits);
/*
@@ -280,6 +294,7 @@ heap_page_prune_opt(Relation relation, Buffer buffer)
PruneFreezeParams params = {
.relation = relation,
.buffer = buffer,
+ .vmbuffer = InvalidBuffer,
.reason = PRUNE_ON_ACCESS,
.options = 0,
.vistest = vistest,
@@ -341,6 +356,8 @@ prune_freeze_setup(PruneFreezeParams *params,
/* cutoffs must be provided if we will attempt freezing */
Assert(!(params->options & HEAP_PAGE_PRUNE_FREEZE) || params->cutoffs);
prstate->attempt_freeze = (params->options & HEAP_PAGE_PRUNE_FREEZE) != 0;
+ prstate->attempt_update_vm =
+ (params->options & HEAP_PAGE_PRUNE_UPDATE_VM) != 0;
prstate->cutoffs = params->cutoffs;
/*
@@ -396,51 +413,54 @@ prune_freeze_setup(PruneFreezeParams *params,
prstate->frz_conflict_horizon = InvalidTransactionId;
/*
- * Vacuum may update the VM after we're done. We can keep track of
- * whether the page will be all-visible and all-frozen after pruning and
- * freezing to help the caller to do that.
+ * Track whether the page could be marked all-visible and/or all-frozen.
+ * This information is used for opportunistic freezing and for updating
+ * the visibility map (VM) if requested by the caller.
*
- * Currently, only VACUUM sets the VM bits. To save the effort, only do
- * the bookkeeping if the caller needs it. Currently, that's tied to
- * HEAP_PAGE_PRUNE_FREEZE, but it could be a separate flag if you wanted
- * to update the VM bits without also freezing or freeze without also
- * setting the VM bits.
+ * Currently, only VACUUM performs freezing, but other callers may in the
+ * future. Visibility bookkeeping is required not just for setting the VM
+ * bits, but also for opportunistic freezing: we only consider freezing if
+ * the page would become all-frozen, or if it would be all-frozen except
+ * for dead tuples that VACUUM will remove. If attempt_update_vm is false,
+ * we will not set the VM bit even if the page is found to be all-visible.
*
- * In addition to telling the caller whether it can set the VM bit, we
- * also use 'all_visible' and 'all_frozen' for our own decision-making. If
- * the whole page would become frozen, we consider opportunistically
- * freezing tuples. We will not be able to freeze the whole page if there
- * are tuples present that are not visible to everyone or if there are
- * dead tuples which are not yet removable. However, dead tuples which
- * will be removed by the end of vacuuming should not preclude us from
- * opportunistically freezing. Because of that, we do not immediately
- * clear all_visible and all_frozen when we see LP_DEAD items. We fix
- * that after scanning the line pointers. We must correct all_visible and
- * all_frozen before we return them to the caller, so that the caller
- * doesn't set the VM bits incorrectly.
+ * If HEAP_PAGE_PRUNE_UPDATE_VM is passed without HEAP_PAGE_PRUNE_FREEZE,
+ * prstate.all_frozen must be initialized to false, since we will not call
+ * heap_prepare_freeze_tuple() for each tuple.
+ *
+ * Dead tuples that will be removed by the end of vacuum should not
+ * prevent opportunistic freezing. Therefore, we do not clear all_visible
+ * and all_frozen when we encounter LP_DEAD items. Instead, we correct
+ * them after deciding whether to freeze, but before updating the VM, to
+ * avoid setting the VM bits incorrectly.
+ *
+ * If neither freezing nor VM updates are requested, we skip the extra
+ * bookkeeping. In this case, initializing all_visible to false allows
+ * heap_prune_record_unchanged_lp_normal() to bypass unnecessary work.
*/
if (prstate->attempt_freeze)
{
prstate->all_visible = true;
prstate->all_frozen = true;
}
+ else if (prstate->attempt_update_vm)
+ {
+ prstate->all_visible = true;
+ prstate->all_frozen = false;
+ }
else
{
- /*
- * Initializing to false allows skipping the work to update them in
- * heap_prune_record_unchanged_lp_normal().
- */
prstate->all_visible = false;
prstate->all_frozen = false;
}
/*
- * The visibility cutoff xid is the newest xmin of live tuples on the
- * page. In the common case, this will be set as the conflict horizon the
- * caller can use for updating the VM. If, at the end of freezing and
- * pruning, the page is all-frozen, there is no possibility that any
- * running transaction on the standby does not see tuples on the page as
- * all-visible, so the conflict horizon remains InvalidTransactionId.
+ * The visibility cutoff xid is the newest xmin of live, committed tuples
+ * older than OldestXmin on the page. This field is only kept up-to-date
+ * if the page is all-visible. As soon as a tuple is encountered that is
+ * not visible to all, this field is unmaintained. As long as it is
+ * maintained, it can be used to calculate the snapshot conflict horizon
+ * when updating the VM and/or freezing all the tuples on the page.
*/
prstate->visibility_cutoff_xid = InvalidTransactionId;
}
@@ -775,10 +795,134 @@ heap_page_will_freeze(Relation relation, Buffer buffer,
return do_freeze;
}
+/*
+ * Helper to correct any corruption detected on an heap page and its
+ * corresponding visibility map page after pruning but before setting the
+ * visibility map. It examines the heap page, the associated VM page, and the
+ * number of dead items previously identified.
+ *
+ * This function must be called while holding an exclusive lock on the heap
+ * buffer, and the dead items must have been discovered under that same lock.
+ *
+ * The provided vmbits must reflect the current state of the VM block
+ * referenced by vmbuffer. Although we do not hold a lock on the VM buffer, it
+ * is pinned, and the heap buffer is exclusively locked, ensuring that no
+ * other backend can update the VM bits corresponding to this heap page.
+ */
+static void
+identify_and_fix_vm_corruption(Relation rel, Buffer heap_buffer,
+ BlockNumber heap_blk, Page heap_page,
+ int nlpdead_items,
+ Buffer vmbuffer,
+ uint8 vmbits)
+{
+ Assert(visibilitymap_get_status(rel, heap_blk, &vmbuffer) == vmbits);
+
+ Assert(BufferIsLockedByMeInMode(heap_buffer, BUFFER_LOCK_EXCLUSIVE));
+
+ /*
+ * As of PostgreSQL 9.2, the visibility map bit should never be set if the
+ * page-level bit is clear. However, it's possible that the bit got
+ * cleared after heap_vac_scan_next_block() was called, so we must recheck
+ * with buffer lock before concluding that the VM is corrupt.
+ */
+ if (!PageIsAllVisible(heap_page) &&
+ ((vmbits & VISIBILITYMAP_ALL_VISIBLE) != 0))
+ {
+ ereport(WARNING,
+ (errcode(ERRCODE_DATA_CORRUPTED),
+ errmsg("page is not marked all-visible but visibility map bit is set in relation \"%s\" page %u",
+ RelationGetRelationName(rel), heap_blk)));
+
+ visibilitymap_clear(rel, heap_blk, vmbuffer,
+ VISIBILITYMAP_VALID_BITS);
+ }
+
+ /*
+ * It's possible for the value returned by
+ * GetOldestNonRemovableTransactionId() to move backwards, so it's not
+ * wrong for us to see tuples that appear to not be visible to everyone
+ * yet, while PD_ALL_VISIBLE is already set. The real safe xmin value
+ * never moves backwards, but GetOldestNonRemovableTransactionId() is
+ * conservative and sometimes returns a value that's unnecessarily small,
+ * so if we see that contradiction it just means that the tuples that we
+ * think are not visible to everyone yet actually are, and the
+ * PD_ALL_VISIBLE flag is correct.
+ *
+ * There should never be LP_DEAD items on a page with PD_ALL_VISIBLE set,
+ * however.
+ */
+ else if (PageIsAllVisible(heap_page) && nlpdead_items > 0)
+ {
+ ereport(WARNING,
+ (errcode(ERRCODE_DATA_CORRUPTED),
+ errmsg("page containing LP_DEAD items is marked as all-visible in relation \"%s\" page %u",
+ RelationGetRelationName(rel), heap_blk)));
+
+ PageClearAllVisible(heap_page);
+ MarkBufferDirty(heap_buffer);
+ visibilitymap_clear(rel, heap_blk, vmbuffer,
+ VISIBILITYMAP_VALID_BITS);
+ }
+}
+
+/*
+ * Decide whether to set the visibility map bits (all-visible and all-frozen)
+ * for heap_blk using information from the PruneState and VM.
+ *
+ * This function does not actually set the VM bits or page-level visibility
+ * hint, PD_ALL_VISIBLE.
+ *
+ * However, if it finds that the page-level visibility hint or VM is
+ * corrupted, it will fix them by clearing the VM bits and visibility hint.
+ * This does not need to be done in a critical section.
+ *
+ * Returns true if one or both VM bits should be set, along with returning the
+ * current value of the VM bits in *old_vmbits and the desired new value of
+ * the VM bits in *new_vmbits.
+ */
+static bool
+heap_page_will_set_vm(PruneState *prstate,
+ Relation relation,
+ BlockNumber heap_blk, Buffer heap_buffer, Page heap_page,
+ Buffer vmbuffer,
+ int nlpdead_items,
+ uint8 *old_vmbits,
+ uint8 *new_vmbits)
+{
+ if (!prstate->attempt_update_vm)
+ return false;
+
+ *old_vmbits = visibilitymap_get_status(relation, heap_blk,
+ &vmbuffer);
+
+ /* We do this even if not all-visible */
+ identify_and_fix_vm_corruption(relation, heap_buffer, heap_blk, heap_page,
+ nlpdead_items, vmbuffer,
+ *old_vmbits);
+
+ if (!prstate->all_visible)
+ return false;
+
+ *new_vmbits = VISIBILITYMAP_ALL_VISIBLE;
+
+ if (prstate->all_frozen)
+ *new_vmbits |= VISIBILITYMAP_ALL_FROZEN;
+
+ if (*new_vmbits == *old_vmbits)
+ {
+ *new_vmbits = 0;
+ return false;
+ }
+
+ return true;
+}
+
/*
* Prune and repair fragmentation and potentially freeze tuples on the
- * specified page.
+ * specified page. If the page's visibility status has changed, update it in
+ * the VM.
*
* Caller must have pin and buffer cleanup lock on the page. Note that we
* don't update the FSM information for page on caller's behalf. Caller might
@@ -793,12 +937,13 @@ heap_page_will_freeze(Relation relation, Buffer buffer,
* tuples if it's required in order to advance relfrozenxid / relminmxid, or
* if it's considered advantageous for overall system performance to do so
* now. The 'params.cutoffs', 'presult', 'new_relfrozen_xid' and
- * 'new_relmin_mxid' arguments are required when freezing. When
- * HEAP_PAGE_PRUNE_FREEZE option is passed, we also set presult->all_visible
- * and presult->all_frozen after determining whether or not to
- * opportunistically freeze, to indicate if the VM bits can be set. They are
- * always set to false when the HEAP_PAGE_PRUNE_FREEZE option is not passed,
- * because at the moment only callers that also freeze need that information.
+ * 'new_relmin_mxid' arguments are required when freezing.
+ *
+ * If HEAP_PAGE_PRUNE_UPDATE_VM is set in params and the visibility status of
+ * the page has changed, we will update the VM at the same time as pruning and
+ * freezing the heap page. We will also update presult->old_vmbits and
+ * presult->new_vmbits with the state of the VM before and after updating it
+ * for the caller to use in bookkeeping.
*
* presult contains output parameters needed by callers, such as the number of
* tuples removed and the offsets of dead items on the page after pruning.
@@ -823,13 +968,19 @@ heap_page_prune_and_freeze(PruneFreezeParams *params,
MultiXactId *new_relmin_mxid)
{
Buffer buffer = params->buffer;
+ Buffer vmbuffer = params->vmbuffer;
Page page = BufferGetPage(buffer);
+ BlockNumber blockno = BufferGetBlockNumber(buffer);
PruneState prstate;
bool do_freeze;
bool do_prune;
bool do_hint_prune;
+ bool do_set_vm;
bool did_tuple_hint_fpi;
int64 fpi_before = pgWalUsage.wal_fpi;
+ uint8 new_vmbits = 0;
+ uint8 old_vmbits = 0;
+
/* Initialize prstate */
prune_freeze_setup(params,
@@ -1011,6 +1162,65 @@ heap_page_prune_and_freeze(PruneFreezeParams *params,
*new_relmin_mxid = prstate.pagefrz.NoFreezePageRelminMxid;
}
}
+
+ /* Now update the visibility map and PD_ALL_VISIBLE hint */
+ Assert(!prstate.all_visible || (prstate.lpdead_items == 0));
+
+ do_set_vm = heap_page_will_set_vm(&prstate,
+ params->relation,
+ blockno,
+ buffer,
+ page,
+ vmbuffer,
+ prstate.lpdead_items,
+ &old_vmbits,
+ &new_vmbits);
+
+ /*
+ * new_vmbits should be 0 regardless of whether or not the page is
+ * all-visible if we do not intend to set the VM.
+ */
+ Assert(do_set_vm || new_vmbits == 0);
+
+ /* Set the visibility map and page visibility hint, if relevant */
+ if (do_set_vm)
+ {
+ Assert(prstate.all_visible);
+
+ /*
+ * It should never be the case that the visibility map page is set
+ * while the page-level bit is clear, but the reverse is allowed (if
+ * checksums are not enabled). Regardless, set both bits so that we
+ * get back in sync.
+ *
+ * Even if PD_ALL_VISIBLE is already set, we don't need to worry about
+ * unnecessarily dirtying the heap buffer. Nearly the only scenario
+ * where PD_ALL_VISIBLE is set but the VM is not is if the VM was
+ * removed -- and that isn't worth optimizing for. And if we add the
+ * heap buffer to the WAL chain (without passing REGBUF_NO_CHANGES),
+ * it must be marked dirty.
+ */
+ PageSetAllVisible(page);
+ MarkBufferDirty(buffer);
+
+ /*
+ * If the page is being set all-frozen, we pass InvalidTransactionId
+ * as the cutoff_xid, since a snapshot conflict horizon sufficient to
+ * make everything safe for REDO was logged when the page's tuples
+ * were frozen.
+ */
+ Assert(!prstate.all_frozen ||
+ !TransactionIdIsValid(presult->vm_conflict_horizon));
+
+ visibilitymap_set(params->relation, blockno, buffer,
+ InvalidXLogRecPtr,
+ vmbuffer, presult->vm_conflict_horizon,
+ new_vmbits);
+ }
+
+ /* Save the vmbits for caller */
+ presult->old_vmbits = old_vmbits;
+ presult->new_vmbits = new_vmbits;
}
@@ -1485,6 +1695,8 @@ heap_prune_record_unchanged_lp_normal(Page page, PruneState *prstate, OffsetNumb
{
TransactionId xmin;
+ Assert(prstate->attempt_update_vm);
+
if (!HeapTupleHeaderXminCommitted(htup))
{
prstate->all_visible = false;
diff --git a/src/backend/access/heap/vacuumlazy.c b/src/backend/access/heap/vacuumlazy.c
index ff34a99edbd..d5c57516785 100644
--- a/src/backend/access/heap/vacuumlazy.c
+++ b/src/backend/access/heap/vacuumlazy.c
@@ -422,11 +422,6 @@ static void find_next_unskippable_block(LVRelState *vacrel, bool *skipsallvis);
static bool lazy_scan_new_or_empty(LVRelState *vacrel, Buffer buf,
BlockNumber blkno, Page page,
bool sharelock, Buffer vmbuffer);
-static void identify_and_fix_vm_corruption(Relation rel, Buffer heap_buffer,
- BlockNumber heap_blk, Page heap_page,
- int nlpdead_items,
- Buffer vmbuffer,
- uint8 vmbits);
static int lazy_scan_prune(LVRelState *vacrel, Buffer buf,
BlockNumber blkno, Page page,
Buffer vmbuffer,
@@ -1933,77 +1928,6 @@ cmpOffsetNumbers(const void *a, const void *b)
return pg_cmp_u16(*(const OffsetNumber *) a, *(const OffsetNumber *) b);
}
-/*
- * Helper to correct any corruption detected on an heap page and its
- * corresponding visibility map page after pruning but before setting the
- * visibility map. It examines the heap page, the associated VM page, and the
- * number of dead items previously identified.
- *
- * This function must be called while holding an exclusive lock on the heap
- * buffer, and the dead items must have been discovered under that same lock.
-
- * The provided vmbits must reflect the current state of the VM block
- * referenced by vmbuffer. Although we do not hold a lock on the VM buffer, it
- * is pinned, and the heap buffer is exclusively locked, ensuring that no
- * other backend can update the VM bits corresponding to this heap page.
- */
-static void
-identify_and_fix_vm_corruption(Relation rel, Buffer heap_buffer,
- BlockNumber heap_blk, Page heap_page,
- int nlpdead_items,
- Buffer vmbuffer,
- uint8 vmbits)
-{
- Assert(visibilitymap_get_status(rel, heap_blk, &vmbuffer) == vmbits);
-
- Assert(BufferIsLockedByMeInMode(heap_buffer, BUFFER_LOCK_EXCLUSIVE));
-
- /*
- * As of PostgreSQL 9.2, the visibility map bit should never be set if the
- * page-level bit is clear. However, it's possible that the bit got
- * cleared after heap_vac_scan_next_block() was called, so we must recheck
- * with buffer lock before concluding that the VM is corrupt.
- */
- if (!PageIsAllVisible(heap_page) &&
- ((vmbits & VISIBILITYMAP_ALL_VISIBLE) != 0))
- {
- ereport(WARNING,
- (errcode(ERRCODE_DATA_CORRUPTED),
- errmsg("page is not marked all-visible but visibility map bit is set in relation \"%s\" page %u",
- RelationGetRelationName(rel), heap_blk)));
-
- visibilitymap_clear(rel, heap_blk, vmbuffer,
- VISIBILITYMAP_VALID_BITS);
- }
-
- /*
- * It's possible for the value returned by
- * GetOldestNonRemovableTransactionId() to move backwards, so it's not
- * wrong for us to see tuples that appear to not be visible to everyone
- * yet, while PD_ALL_VISIBLE is already set. The real safe xmin value
- * never moves backwards, but GetOldestNonRemovableTransactionId() is
- * conservative and sometimes returns a value that's unnecessarily small,
- * so if we see that contradiction it just means that the tuples that we
- * think are not visible to everyone yet actually are, and the
- * PD_ALL_VISIBLE flag is correct.
- *
- * There should never be LP_DEAD items on a page with PD_ALL_VISIBLE set,
- * however.
- */
- else if (PageIsAllVisible(heap_page) && nlpdead_items > 0)
- {
- ereport(WARNING,
- (errcode(ERRCODE_DATA_CORRUPTED),
- errmsg("page containing LP_DEAD items is marked as all-visible in relation \"%s\" page %u",
- RelationGetRelationName(rel), heap_blk)));
-
- PageClearAllVisible(heap_page);
- MarkBufferDirty(heap_buffer);
- visibilitymap_clear(rel, heap_blk, vmbuffer,
- VISIBILITYMAP_VALID_BITS);
- }
-}
-
/*
* lazy_scan_prune() -- lazy_scan_heap() pruning and freezing.
*
@@ -2035,13 +1959,12 @@ lazy_scan_prune(LVRelState *vacrel,
PruneFreezeParams params = {
.relation = rel,
.buffer = buf,
+ .vmbuffer = vmbuffer,
.reason = PRUNE_VACUUM_SCAN,
- .options = HEAP_PAGE_PRUNE_FREEZE,
+ .options = HEAP_PAGE_PRUNE_FREEZE | HEAP_PAGE_PRUNE_UPDATE_VM,
.vistest = vacrel->vistest,
.cutoffs = &vacrel->cutoffs,
};
- uint8 old_vmbits = 0;
- uint8 new_vmbits = 0;
Assert(BufferGetBlockNumber(buf) == blkno);
@@ -2141,73 +2064,24 @@ lazy_scan_prune(LVRelState *vacrel,
/* Did we find LP_DEAD items? */
*has_lpdead_items = (presult.lpdead_items > 0);
- Assert(!presult.all_visible || !(*has_lpdead_items));
- Assert(!presult.all_frozen || presult.all_visible);
-
- old_vmbits = visibilitymap_get_status(vacrel->rel, blkno, &vmbuffer);
-
- identify_and_fix_vm_corruption(vacrel->rel, buf, blkno, page,
- presult.lpdead_items, vmbuffer, old_vmbits);
-
- if (!presult.all_visible)
- return presult.ndeleted;
-
- /* Set the visibility map and page visibility hint */
- new_vmbits = VISIBILITYMAP_ALL_VISIBLE;
-
- if (presult.all_frozen)
- new_vmbits |= VISIBILITYMAP_ALL_FROZEN;
-
- /* Nothing to do */
- if (old_vmbits == new_vmbits)
- return presult.ndeleted;
-
- Assert(presult.all_visible);
-
- /*
- * It should never be the case that the visibility map page is set while
- * the page-level bit is clear, but the reverse is allowed (if checksums
- * are not enabled). Regardless, set both bits so that we get back in
- * sync.
- *
- * Even if PD_ALL_VISIBLE is already set, we don't need to worry about
- * unnecessarily dirtying the heap buffer. Nearly the only scenario where
- * PD_ALL_VISIBLE is set but the VM is not is if the VM was removed -- and
- * that isn't worth optimizing for. And if we add the heap buffer to the
- * WAL chain (without passing REGBUF_NO_CHANGES), it must be marked dirty.
- */
- PageSetAllVisible(page);
- MarkBufferDirty(buf);
-
- /*
- * If the page is being set all-frozen, we pass InvalidTransactionId as
- * the cutoff_xid, since a snapshot conflict horizon sufficient to make
- * everything safe for REDO was logged when the page's tuples were frozen.
- */
- Assert(!presult.all_frozen ||
- !TransactionIdIsValid(presult.vm_conflict_horizon));
-
- visibilitymap_set(vacrel->rel, blkno, buf,
- InvalidXLogRecPtr,
- vmbuffer, presult.vm_conflict_horizon,
- new_vmbits);
-
/*
* If the page wasn't already set all-visible and/or all-frozen in the VM,
* count it as newly set for logging.
*/
- if ((old_vmbits & VISIBILITYMAP_ALL_VISIBLE) == 0)
+ if ((presult.old_vmbits & VISIBILITYMAP_ALL_VISIBLE) == 0 &&
+ (presult.new_vmbits & VISIBILITYMAP_ALL_VISIBLE) != 0)
{
vacrel->vm_new_visible_pages++;
- if (presult.all_frozen)
+ if ((presult.new_vmbits & VISIBILITYMAP_ALL_FROZEN) != 0)
{
vacrel->vm_new_visible_frozen_pages++;
*vm_page_frozen = true;
}
}
- else if ((old_vmbits & VISIBILITYMAP_ALL_FROZEN) == 0 &&
- presult.all_frozen)
+ else if ((presult.old_vmbits & VISIBILITYMAP_ALL_FROZEN) == 0 &&
+ (presult.new_vmbits & VISIBILITYMAP_ALL_FROZEN) != 0)
{
+ Assert((presult.new_vmbits & VISIBILITYMAP_ALL_VISIBLE) != 0);
vacrel->vm_new_frozen_pages++;
*vm_page_frozen = true;
}
diff --git a/src/include/access/heapam.h b/src/include/access/heapam.h
index f7e4ae3843c..ad2af13ec39 100644
--- a/src/include/access/heapam.h
+++ b/src/include/access/heapam.h
@@ -42,6 +42,7 @@
/* "options" flag bits for heap_page_prune_and_freeze */
#define HEAP_PAGE_PRUNE_MARK_UNUSED_NOW (1 << 0)
#define HEAP_PAGE_PRUNE_FREEZE (1 << 1)
+#define HEAP_PAGE_PRUNE_UPDATE_VM (1 << 2)
typedef struct BulkInsertStateData *BulkInsertState;
typedef struct GlobalVisState GlobalVisState;
@@ -238,6 +239,12 @@ typedef struct PruneFreezeParams
Relation relation; /* relation containing buffer to be pruned */
Buffer buffer; /* buffer to be pruned */
+ /*
+ * If we will consider updating the visibility map, vmbuffer should
+ * contain the correct block of the visibility map and be pinned.
+ */
+ Buffer vmbuffer;
+
/*
* The reason pruning was performed. It is used to set the WAL record
* opcode which is used for debugging and analysis purposes.
@@ -252,6 +259,9 @@ typedef struct PruneFreezeParams
*
* HEAP_PAGE_PRUNE_FREEZE indicates that we will also freeze tuples, and
* will return 'all_visible', 'all_frozen' flags to the caller.
+ *
+ * HEAP_PAGE_PRUNE_UPDATE_VM indicates that we will set the page's status
+ * in the VM.
*/
int options;
@@ -299,6 +309,17 @@ typedef struct PruneFreezeResult
bool all_frozen;
TransactionId vm_conflict_horizon;
+ /*
+ * old_vmbits are the state of the all-visible and all-frozen bits in the
+ * visibility map before updating it during phase I of vacuuming.
+ * new_vmbits are the state of those bits after phase I of vacuuming.
+ *
+ * These are only set if the HEAP_PAGE_PRUNE_UPDATE_VM option is set and
+ * we have attempted to update the VM.
+ */
+ uint8 new_vmbits;
+ uint8 old_vmbits;
+
/*
* Whether or not the page makes rel truncation unsafe. This is set to
* 'true', even if the page contains LP_DEAD items. VACUUM will remove
--
2.43.0
[text/x-patch] v28-0005-Move-VM-assert-into-prune-freeze-code.patch (10.9K, 6-v28-0005-Move-VM-assert-into-prune-freeze-code.patch)
download | inline diff:
From 218bcd4dffe014647495a9bba11d8beaeb1465cd Mon Sep 17 00:00:00 2001
From: Melanie Plageman <[email protected]>
Date: Tue, 2 Dec 2025 15:57:34 -0500
Subject: [PATCH v28 05/15] Move VM assert into prune/freeze code
This is a step toward setting the VM in the same WAL record as pruning
and freezing. It moves the check of the heap page into prune/freeze code
before setting the VM. This allows us to remove some fields of the
PruneFreezeResult.
---
src/backend/access/heap/pruneheap.c | 86 ++++++++++++++++++++++------
src/backend/access/heap/vacuumlazy.c | 68 +---------------------
src/include/access/heapam.h | 25 +++-----
3 files changed, 77 insertions(+), 102 deletions(-)
diff --git a/src/backend/access/heap/pruneheap.c b/src/backend/access/heap/pruneheap.c
index 14d40476be9..39149fbba7c 100644
--- a/src/backend/access/heap/pruneheap.c
+++ b/src/backend/access/heap/pruneheap.c
@@ -918,6 +918,31 @@ heap_page_will_set_vm(PruneState *prstate,
return true;
}
+#ifdef USE_ASSERT_CHECKING
+
+/*
+ * Wrapper for heap_page_would_be_all_visible() which can be used for callers
+ * that expect no LP_DEAD on the page. Currently assert-only, but there is no
+ * reason not to use it outside of asserts.
+ */
+static bool
+heap_page_is_all_visible(Relation rel, Buffer buf,
+ TransactionId OldestXmin,
+ bool *all_frozen,
+ TransactionId *visibility_cutoff_xid,
+ OffsetNumber *logging_offnum)
+{
+
+ return heap_page_would_be_all_visible(rel, buf,
+ OldestXmin,
+ NULL, 0,
+ all_frozen,
+ visibility_cutoff_xid,
+ logging_offnum);
+}
+#endif
+
+
/*
* Prune and repair fragmentation and potentially freeze tuples on the
@@ -971,6 +996,7 @@ heap_page_prune_and_freeze(PruneFreezeParams *params,
Buffer vmbuffer = params->vmbuffer;
Page page = BufferGetPage(buffer);
BlockNumber blockno = BufferGetBlockNumber(buffer);
+ TransactionId vm_conflict_horizon = InvalidTransactionId;
PruneState prstate;
bool do_freeze;
bool do_prune;
@@ -1129,23 +1155,8 @@ heap_page_prune_and_freeze(PruneFreezeParams *params,
presult->nfrozen = prstate.nfrozen;
presult->live_tuples = prstate.live_tuples;
presult->recently_dead_tuples = prstate.recently_dead_tuples;
- presult->all_visible = prstate.all_visible;
- presult->all_frozen = prstate.all_frozen;
presult->hastup = prstate.hastup;
- /*
- * For callers planning to update the visibility map, the conflict horizon
- * for that record must be the newest xmin on the page. However, if the
- * page is completely frozen, there can be no conflict and the
- * vm_conflict_horizon should remain InvalidTransactionId. This includes
- * the case that we just froze all the tuples; the prune-freeze record
- * included the conflict XID already so the caller doesn't need it.
- */
- if (presult->all_frozen)
- presult->vm_conflict_horizon = InvalidTransactionId;
- else
- presult->vm_conflict_horizon = prstate.visibility_cutoff_xid;
-
presult->lpdead_items = prstate.lpdead_items;
/* the presult->deadoffsets array was already filled in */
@@ -1163,6 +1174,46 @@ heap_page_prune_and_freeze(PruneFreezeParams *params,
}
}
+ /*
+ * If updating the visibility map, the conflict horizon for that record
+ * must be the newest xmin on the page. However, if the page is
+ * completely frozen, there can be no conflict and the vm_conflict_horizon
+ * should remain InvalidTransactionId. This includes the case that we
+ * just froze all the tuples; the prune-freeze record included the
+ * conflict XID already so we don't need to again.
+ */
+ if (prstate.all_frozen)
+ vm_conflict_horizon = InvalidTransactionId;
+ else
+ vm_conflict_horizon = prstate.visibility_cutoff_xid;
+
+ /*
+ * During its second pass over the heap, VACUUM calls
+ * heap_page_would_be_all_visible() to determine whether a page is
+ * all-visible and all-frozen. The logic here is similar. After completing
+ * pruning and freezing, use an assertion to verify that our results
+ * remain consistent with heap_page_would_be_all_visible().
+ */
+#ifdef USE_ASSERT_CHECKING
+ if (prstate.all_visible)
+ {
+ TransactionId debug_cutoff;
+ bool debug_all_frozen;
+
+ Assert(presult->lpdead_items == 0);
+
+ Assert(heap_page_is_all_visible(params->relation, buffer,
+ prstate.cutoffs->OldestXmin,
+ &debug_all_frozen,
+ &debug_cutoff, off_loc));
+
+ Assert(prstate.all_frozen == debug_all_frozen);
+
+ Assert(!TransactionIdIsValid(debug_cutoff) ||
+ debug_cutoff == vm_conflict_horizon);
+ }
+#endif
+
/* Now update the visibility map and PD_ALL_VISIBLE hint */
Assert(!prstate.all_visible || (prstate.lpdead_items == 0));
@@ -1209,12 +1260,11 @@ heap_page_prune_and_freeze(PruneFreezeParams *params,
* make everything safe for REDO was logged when the page's tuples
* were frozen.
*/
- Assert(!prstate.all_frozen ||
- !TransactionIdIsValid(presult->vm_conflict_horizon));
+ Assert(!prstate.all_frozen || !TransactionIdIsValid(vm_conflict_horizon));
visibilitymap_set(params->relation, blockno, buffer,
InvalidXLogRecPtr,
- vmbuffer, presult->vm_conflict_horizon,
+ vmbuffer, vm_conflict_horizon,
new_vmbits);
}
diff --git a/src/backend/access/heap/vacuumlazy.c b/src/backend/access/heap/vacuumlazy.c
index d5c57516785..61564aea5fd 100644
--- a/src/backend/access/heap/vacuumlazy.c
+++ b/src/backend/access/heap/vacuumlazy.c
@@ -456,20 +456,6 @@ static void dead_items_add(LVRelState *vacrel, BlockNumber blkno, OffsetNumber *
static void dead_items_reset(LVRelState *vacrel);
static void dead_items_cleanup(LVRelState *vacrel);
-#ifdef USE_ASSERT_CHECKING
-static bool heap_page_is_all_visible(Relation rel, Buffer buf,
- TransactionId OldestXmin,
- bool *all_frozen,
- TransactionId *visibility_cutoff_xid,
- OffsetNumber *logging_offnum);
-#endif
-static bool heap_page_would_be_all_visible(Relation rel, Buffer buf,
- TransactionId OldestXmin,
- OffsetNumber *deadoffsets,
- int ndeadoffsets,
- bool *all_frozen,
- TransactionId *visibility_cutoff_xid,
- OffsetNumber *logging_offnum);
static void update_relstats_all_indexes(LVRelState *vacrel);
static void vacuum_error_callback(void *arg);
static void update_vacuum_error_info(LVRelState *vacrel,
@@ -2005,32 +1991,6 @@ lazy_scan_prune(LVRelState *vacrel,
vacrel->new_frozen_tuple_pages++;
}
- /*
- * VACUUM will call heap_page_is_all_visible() during the second pass over
- * the heap to determine all_visible and all_frozen for the page -- this
- * is a specialized version of the logic from this function. Now that
- * we've finished pruning and freezing, make sure that we're in total
- * agreement with heap_page_is_all_visible() using an assertion.
- */
-#ifdef USE_ASSERT_CHECKING
- if (presult.all_visible)
- {
- TransactionId debug_cutoff;
- bool debug_all_frozen;
-
- Assert(presult.lpdead_items == 0);
-
- Assert(heap_page_is_all_visible(vacrel->rel, buf,
- vacrel->cutoffs.OldestXmin, &debug_all_frozen,
- &debug_cutoff, &vacrel->offnum));
-
- Assert(presult.all_frozen == debug_all_frozen);
-
- Assert(!TransactionIdIsValid(debug_cutoff) ||
- debug_cutoff == presult.vm_conflict_horizon);
- }
-#endif
-
/*
* Now save details of the LP_DEAD items from the page in vacrel
*/
@@ -3487,29 +3447,6 @@ dead_items_cleanup(LVRelState *vacrel)
vacrel->pvs = NULL;
}
-#ifdef USE_ASSERT_CHECKING
-
-/*
- * Wrapper for heap_page_would_be_all_visible() which can be used for callers
- * that expect no LP_DEAD on the page. Currently assert-only, but there is no
- * reason not to use it outside of asserts.
- */
-static bool
-heap_page_is_all_visible(Relation rel, Buffer buf,
- TransactionId OldestXmin,
- bool *all_frozen,
- TransactionId *visibility_cutoff_xid,
- OffsetNumber *logging_offnum)
-{
-
- return heap_page_would_be_all_visible(rel, buf,
- OldestXmin,
- NULL, 0,
- all_frozen,
- visibility_cutoff_xid,
- logging_offnum);
-}
-#endif
/*
* Check whether the heap page in buf is all-visible except for the dead
@@ -3533,15 +3470,12 @@ heap_page_is_all_visible(Relation rel, Buffer buf,
* - *logging_offnum: OffsetNumber of current tuple being processed;
* used by vacuum's error callback system.
*
- * Callers looking to verify that the page is already all-visible can call
- * heap_page_is_all_visible().
- *
* This logic is closely related to heap_prune_record_unchanged_lp_normal().
* If you modify this function, ensure consistency with that code. An
* assertion cross-checks that both remain in agreement. Do not introduce new
* side-effects.
*/
-static bool
+bool
heap_page_would_be_all_visible(Relation rel, Buffer buf,
TransactionId OldestXmin,
OffsetNumber *deadoffsets,
diff --git a/src/include/access/heapam.h b/src/include/access/heapam.h
index ad2af13ec39..bec2f840102 100644
--- a/src/include/access/heapam.h
+++ b/src/include/access/heapam.h
@@ -257,8 +257,7 @@ typedef struct PruneFreezeParams
* HEAP_PAGE_PRUNE_MARK_UNUSED_NOW indicates that dead items can be set
* LP_UNUSED during pruning.
*
- * HEAP_PAGE_PRUNE_FREEZE indicates that we will also freeze tuples, and
- * will return 'all_visible', 'all_frozen' flags to the caller.
+ * HEAP_PAGE_PRUNE_FREEZE indicates that we will also freeze tuples
*
* HEAP_PAGE_PRUNE_UPDATE_VM indicates that we will set the page's status
* in the VM.
@@ -294,21 +293,6 @@ typedef struct PruneFreezeResult
int live_tuples;
int recently_dead_tuples;
- /*
- * all_visible and all_frozen indicate if the all-visible and all-frozen
- * bits in the visibility map can be set for this page, after pruning.
- *
- * vm_conflict_horizon is the newest xmin of live tuples on the page. The
- * caller can use it as the conflict horizon when setting the VM bits. It
- * is only valid if we froze some tuples (nfrozen > 0), and all_frozen is
- * true.
- *
- * These are only set if the HEAP_PRUNE_FREEZE option is set.
- */
- bool all_visible;
- bool all_frozen;
- TransactionId vm_conflict_horizon;
-
/*
* old_vmbits are the state of the all-visible and all-frozen bits in the
* visibility map before updating it during phase I of vacuuming.
@@ -454,6 +438,13 @@ extern void log_heap_prune_and_freeze(Relation relation, Buffer buffer,
/* in heap/vacuumlazy.c */
extern void heap_vacuum_rel(Relation rel,
const VacuumParams params, BufferAccessStrategy bstrategy);
+extern bool heap_page_would_be_all_visible(Relation rel, Buffer buf,
+ TransactionId OldestXmin,
+ OffsetNumber *deadoffsets,
+ int ndeadoffsets,
+ bool *all_frozen,
+ TransactionId *visibility_cutoff_xid,
+ OffsetNumber *logging_offnum);
/* in heap/heapam_visibility.c */
extern bool HeapTupleSatisfiesVisibility(HeapTuple htup, Snapshot snapshot,
--
2.43.0
[text/x-patch] v28-0006-Eliminate-XLOG_HEAP2_VISIBLE-from-vacuum-phase-I.patch (14.4K, 7-v28-0006-Eliminate-XLOG_HEAP2_VISIBLE-from-vacuum-phase-I.patch)
download | inline diff:
From beb1ae3557904962dcec6266b882cdc75a0c7051 Mon Sep 17 00:00:00 2001
From: Melanie Plageman <[email protected]>
Date: Tue, 2 Dec 2025 16:16:22 -0500
Subject: [PATCH v28 06/15] Eliminate XLOG_HEAP2_VISIBLE from vacuum phase I
prune/freeze
Vacuum no longer emits a separate WAL record for each page set
all-visible or all-frozen during phase I. Instead, visibility map
updates are now included in the XLOG_HEAP2_PRUNE_VACUUM_SCAN record that
is already emitted for pruning and freezing.
Previously, heap_page_prune_and_freeze() determined whether a page was
all-visible, but the corresponding VM bits were only set later in
lazy_scan_prune(). Now the VM is updated immediately in
heap_page_prune_and_freeze(), at the same time as the heap
modifications.
This change applies only to vacuum phase I, not to pruning performed
during normal page access.
Author: Melanie Plageman <[email protected]>
Reviewed-by: Andres Freund <[email protected]>
Reviewed-by: Robert Haas <[email protected]>
Reviewed-by: Kirill Reshke <[email protected]>
Reviewed-by: Chao Li <[email protected]>
Discussion: https://postgr.es/m/flat/CAAKRu_ZMw6Npd_qm2KM%2BFwQ3cMOMx1Dh3VMhp8-V7SOLxdK9-g%40mail.gmail.com
---
src/backend/access/heap/pruneheap.c | 277 ++++++++++++++++------------
1 file changed, 158 insertions(+), 119 deletions(-)
diff --git a/src/backend/access/heap/pruneheap.c b/src/backend/access/heap/pruneheap.c
index 39149fbba7c..3521e70b8d0 100644
--- a/src/backend/access/heap/pruneheap.c
+++ b/src/backend/access/heap/pruneheap.c
@@ -205,6 +205,11 @@ static bool heap_page_will_set_vm(PruneState *prstate,
int nlpdead_items,
uint8 *old_vmbits,
uint8 *new_vmbits);
+static TransactionId get_conflict_xid(bool do_prune, bool do_freeze, bool do_set_vm,
+ uint8 old_vmbits, uint8 new_vmbits,
+ TransactionId latest_xid_removed,
+ TransactionId frz_conflict_horizon,
+ TransactionId visibility_cutoff_xid);
/*
@@ -795,6 +800,68 @@ heap_page_will_freeze(Relation relation, Buffer buffer,
return do_freeze;
}
+/*
+ * Calculate the conflict horizon for the whole XLOG_HEAP2_PRUNE_VACUUM_SCAN
+ * or XLOG_HEAP2_PRUNE_ON_ACCESS record.
+ */
+static TransactionId
+get_conflict_xid(bool do_prune, bool do_freeze, bool do_set_vm,
+ uint8 old_vmbits, uint8 new_vmbits,
+ TransactionId latest_xid_removed, TransactionId frz_conflict_horizon,
+ TransactionId visibility_cutoff_xid)
+{
+ TransactionId conflict_xid;
+
+ /*
+ * We can omit the snapshot conflict horizon if we are not pruning or
+ * freezing any tuples and are setting an already all-visible page
+ * all-frozen in the VM. In this case, all of the tuples on the page must
+ * already be visible to all MVCC snapshots on the standby.
+ */
+ if (!do_prune &&
+ !do_freeze &&
+ do_set_vm &&
+ (old_vmbits & VISIBILITYMAP_ALL_VISIBLE) != 0 &&
+ (new_vmbits & VISIBILITYMAP_ALL_FROZEN) != 0)
+ return InvalidTransactionId;
+
+ /*
+ * The snapshotConflictHorizon for the whole record should be the most
+ * conservative of all the horizons calculated for any of the possible
+ * modifications. If this record will prune tuples, any transactions on
+ * the standby older than the youngest xmax of the most recently removed
+ * tuple this record will prune will conflict. If this record will freeze
+ * tuples, any transactions on the standby with xids older than the
+ * youngest tuple this record will freeze will conflict.
+ */
+ conflict_xid = InvalidTransactionId;
+
+ /*
+ * If we are updating the VM, the conflict horizon is almost always the
+ * visibility cutoff XID.
+ *
+ * Separately, if we are freezing any tuples, as an optimization, we can
+ * use the visibility_cutoff_xid as the conflict horizon if the page will
+ * be all-frozen. This is true even if there are LP_DEAD line pointers
+ * because we ignored those when maintaining the visibility_cutoff_xid.
+ * This will have been calculated earlier as the frz_conflict_horizon when
+ * we determined we would freeze.
+ */
+ if (do_set_vm)
+ conflict_xid = visibility_cutoff_xid;
+ else if (do_freeze)
+ conflict_xid = frz_conflict_horizon;
+
+ /*
+ * If we are removing tuples with a younger xmax than our so far
+ * calculated conflict_xid, we must use this as our horizon.
+ */
+ if (TransactionIdFollows(latest_xid_removed, conflict_xid))
+ conflict_xid = latest_xid_removed;
+
+ return conflict_xid;
+}
+
/*
* Helper to correct any corruption detected on an heap page and its
* corresponding visibility map page after pruning but before setting the
@@ -996,7 +1063,6 @@ heap_page_prune_and_freeze(PruneFreezeParams *params,
Buffer vmbuffer = params->vmbuffer;
Page page = BufferGetPage(buffer);
BlockNumber blockno = BufferGetBlockNumber(buffer);
- TransactionId vm_conflict_horizon = InvalidTransactionId;
PruneState prstate;
bool do_freeze;
bool do_prune;
@@ -1004,10 +1070,10 @@ heap_page_prune_and_freeze(PruneFreezeParams *params,
bool do_set_vm;
bool did_tuple_hint_fpi;
int64 fpi_before = pgWalUsage.wal_fpi;
+ TransactionId conflict_xid = InvalidTransactionId;
uint8 new_vmbits = 0;
uint8 old_vmbits = 0;
-
/* Initialize prstate */
prune_freeze_setup(params,
new_relfrozen_xid, new_relmin_mxid,
@@ -1068,6 +1134,37 @@ heap_page_prune_and_freeze(PruneFreezeParams *params,
prstate.all_visible = prstate.all_frozen = false;
Assert(!prstate.all_frozen || prstate.all_visible);
+ Assert(!prstate.all_visible || (prstate.lpdead_items == 0));
+
+ /*
+ * Decide whether to set the VM bits based on information from the VM and
+ * the all_visible/all_frozen flags.
+ */
+ do_set_vm = heap_page_will_set_vm(&prstate,
+ params->relation,
+ blockno,
+ buffer,
+ page,
+ vmbuffer,
+ prstate.lpdead_items,
+ &old_vmbits,
+ &new_vmbits);
+
+ /*
+ * new_vmbits should be 0 regardless of whether or not the page is
+ * all-visible if we do not intend to set the VM.
+ */
+ Assert(do_set_vm || new_vmbits == 0);
+
+ conflict_xid = get_conflict_xid(do_prune, do_freeze, do_set_vm,
+ old_vmbits, new_vmbits,
+ prstate.latest_xid_removed,
+ prstate.frz_conflict_horizon,
+ prstate.visibility_cutoff_xid);
+
+ /* Lock vmbuffer before entering a critical section */
+ if (do_set_vm)
+ LockBuffer(vmbuffer, BUFFER_LOCK_EXCLUSIVE);
/* Any error while applying the changes is critical */
START_CRIT_SECTION();
@@ -1089,14 +1186,17 @@ heap_page_prune_and_freeze(PruneFreezeParams *params,
/*
* If that's all we had to do to the page, this is a non-WAL-logged
- * hint. If we are going to freeze or prune the page, we will mark
- * the buffer dirty below.
+ * hint. If we are going to freeze or prune the page or set
+ * PD_ALL_VISIBLE, we will mark the buffer dirty below.
+ *
+ * Setting PD_ALL_VISIBLE is fully WAL-logged because it is forbidden
+ * for the VM to be set and PD_ALL_VISIBLE to be clear.
*/
- if (!do_freeze && !do_prune)
+ if (!do_freeze && !do_prune && !do_set_vm)
MarkBufferDirtyHint(buffer, true);
}
- if (do_prune || do_freeze)
+ if (do_prune || do_freeze || do_set_vm)
{
/* Apply the planned item changes and repair page fragmentation. */
if (do_prune)
@@ -1110,6 +1210,27 @@ heap_page_prune_and_freeze(PruneFreezeParams *params,
if (do_freeze)
heap_freeze_prepared_tuples(buffer, prstate.frozen, prstate.nfrozen);
+ /* Set the visibility map and page visibility hint */
+ if (do_set_vm)
+ {
+ /*
+ * While it is valid for PD_ALL_VISIBLE to be set when the
+ * corresponding VM bit is clear, we strongly prefer to keep them
+ * in sync.
+ *
+ * Even if we are only setting the VM and PD_ALL_VISIBLE is
+ * already set, we don't need to worry about unnecessarily
+ * dirtying the heap buffer below. Nearly the only scenario where
+ * PD_ALL_VISIBLE is set but the VM is not is if the VM was
+ * removed, and that isn't worth optimizing for. And, if we add
+ * the heap buffer to the WAL chain (without passing
+ * REGBUF_NO_CHANGES), it must be marked dirty.
+ */
+ PageSetAllVisible(page);
+ visibilitymap_set_vmbits(blockno, vmbuffer, new_vmbits,
+ params->relation->rd_locator);
+ }
+
MarkBufferDirty(buffer);
/*
@@ -1117,29 +1238,12 @@ heap_page_prune_and_freeze(PruneFreezeParams *params,
*/
if (RelationNeedsWAL(params->relation))
{
- /*
- * The snapshotConflictHorizon for the whole record should be the
- * most conservative of all the horizons calculated for any of the
- * possible modifications. If this record will prune tuples, any
- * transactions on the standby older than the youngest xmax of the
- * most recently removed tuple this record will prune will
- * conflict. If this record will freeze tuples, any transactions
- * on the standby with xids older than the youngest tuple this
- * record will freeze will conflict.
- */
- TransactionId conflict_xid;
-
- if (TransactionIdFollows(prstate.frz_conflict_horizon,
- prstate.latest_xid_removed))
- conflict_xid = prstate.frz_conflict_horizon;
- else
- conflict_xid = prstate.latest_xid_removed;
-
log_heap_prune_and_freeze(params->relation, buffer,
- InvalidBuffer, /* vmbuffer */
- 0, /* vmflags */
+ do_set_vm ? vmbuffer : InvalidBuffer,
+ do_set_vm ? new_vmbits : 0,
conflict_xid,
- true, params->reason,
+ true, /* cleanup lock */
+ params->reason,
prstate.frozen, prstate.nfrozen,
prstate.redirected, prstate.nredirected,
prstate.nowdead, prstate.ndead,
@@ -1149,43 +1253,8 @@ heap_page_prune_and_freeze(PruneFreezeParams *params,
END_CRIT_SECTION();
- /* Copy information back for caller */
- presult->ndeleted = prstate.ndeleted;
- presult->nnewlpdead = prstate.ndead;
- presult->nfrozen = prstate.nfrozen;
- presult->live_tuples = prstate.live_tuples;
- presult->recently_dead_tuples = prstate.recently_dead_tuples;
- presult->hastup = prstate.hastup;
-
- presult->lpdead_items = prstate.lpdead_items;
- /* the presult->deadoffsets array was already filled in */
-
- if (prstate.attempt_freeze)
- {
- if (presult->nfrozen > 0)
- {
- *new_relfrozen_xid = prstate.pagefrz.FreezePageRelfrozenXid;
- *new_relmin_mxid = prstate.pagefrz.FreezePageRelminMxid;
- }
- else
- {
- *new_relfrozen_xid = prstate.pagefrz.NoFreezePageRelfrozenXid;
- *new_relmin_mxid = prstate.pagefrz.NoFreezePageRelminMxid;
- }
- }
-
- /*
- * If updating the visibility map, the conflict horizon for that record
- * must be the newest xmin on the page. However, if the page is
- * completely frozen, there can be no conflict and the vm_conflict_horizon
- * should remain InvalidTransactionId. This includes the case that we
- * just froze all the tuples; the prune-freeze record included the
- * conflict XID already so we don't need to again.
- */
- if (prstate.all_frozen)
- vm_conflict_horizon = InvalidTransactionId;
- else
- vm_conflict_horizon = prstate.visibility_cutoff_xid;
+ if (do_set_vm)
+ LockBuffer(vmbuffer, BUFFER_LOCK_UNLOCK);
/*
* During its second pass over the heap, VACUUM calls
@@ -1200,7 +1269,8 @@ heap_page_prune_and_freeze(PruneFreezeParams *params,
TransactionId debug_cutoff;
bool debug_all_frozen;
- Assert(presult->lpdead_items == 0);
+ Assert(prstate.lpdead_items == 0);
+ Assert(prstate.cutoffs);
Assert(heap_page_is_all_visible(params->relation, buffer,
prstate.cutoffs->OldestXmin,
@@ -1210,67 +1280,36 @@ heap_page_prune_and_freeze(PruneFreezeParams *params,
Assert(prstate.all_frozen == debug_all_frozen);
Assert(!TransactionIdIsValid(debug_cutoff) ||
- debug_cutoff == vm_conflict_horizon);
+ debug_cutoff == prstate.visibility_cutoff_xid);
}
#endif
- /* Now update the visibility map and PD_ALL_VISIBLE hint */
- Assert(!prstate.all_visible || (prstate.lpdead_items == 0));
-
- do_set_vm = heap_page_will_set_vm(&prstate,
- params->relation,
- blockno,
- buffer,
- page,
- vmbuffer,
- prstate.lpdead_items,
- &old_vmbits,
- &new_vmbits);
+ /* Copy information back for caller */
+ presult->ndeleted = prstate.ndeleted;
+ presult->nnewlpdead = prstate.ndead;
+ presult->nfrozen = prstate.nfrozen;
+ presult->live_tuples = prstate.live_tuples;
+ presult->recently_dead_tuples = prstate.recently_dead_tuples;
+ presult->hastup = prstate.hastup;
+ presult->new_vmbits = new_vmbits;
+ presult->old_vmbits = old_vmbits;
- /*
- * new_vmbits should be 0 regardless of whether or not the page is
- * all-visible if we do not intend to set the VM.
- */
- Assert(do_set_vm || new_vmbits == 0);
+ presult->lpdead_items = prstate.lpdead_items;
+ /* the presult->deadoffsets array was already filled in */
- /* Set the visibility map and page visibility hint, if relevant */
- if (do_set_vm)
+ if (prstate.attempt_freeze)
{
- Assert(prstate.all_visible);
-
- /*
- * It should never be the case that the visibility map page is set
- * while the page-level bit is clear, but the reverse is allowed (if
- * checksums are not enabled). Regardless, set both bits so that we
- * get back in sync.
- *
- * Even if PD_ALL_VISIBLE is already set, we don't need to worry about
- * unnecessarily dirtying the heap buffer. Nearly the only scenario
- * where PD_ALL_VISIBLE is set but the VM is not is if the VM was
- * removed -- and that isn't worth optimizing for. And if we add the
- * heap buffer to the WAL chain (without passing REGBUF_NO_CHANGES),
- * it must be marked dirty.
- */
- PageSetAllVisible(page);
- MarkBufferDirty(buffer);
-
- /*
- * If the page is being set all-frozen, we pass InvalidTransactionId
- * as the cutoff_xid, since a snapshot conflict horizon sufficient to
- * make everything safe for REDO was logged when the page's tuples
- * were frozen.
- */
- Assert(!prstate.all_frozen || !TransactionIdIsValid(vm_conflict_horizon));
-
- visibilitymap_set(params->relation, blockno, buffer,
- InvalidXLogRecPtr,
- vmbuffer, vm_conflict_horizon,
- new_vmbits);
+ if (presult->nfrozen > 0)
+ {
+ *new_relfrozen_xid = prstate.pagefrz.FreezePageRelfrozenXid;
+ *new_relmin_mxid = prstate.pagefrz.FreezePageRelminMxid;
+ }
+ else
+ {
+ *new_relfrozen_xid = prstate.pagefrz.NoFreezePageRelfrozenXid;
+ *new_relmin_mxid = prstate.pagefrz.NoFreezePageRelminMxid;
+ }
}
-
- /* Save the vmbits for caller */
- presult->old_vmbits = old_vmbits;
- presult->new_vmbits = new_vmbits;
}
--
2.43.0
[text/x-patch] v28-0007-Eliminate-XLOG_HEAP2_VISIBLE-from-empty-page-vac.patch (2.6K, 8-v28-0007-Eliminate-XLOG_HEAP2_VISIBLE-from-empty-page-vac.patch)
download | inline diff:
From 3d8444447657a04513e044b4261b5d6334f1bef7 Mon Sep 17 00:00:00 2001
From: Melanie Plageman <[email protected]>
Date: Sat, 27 Sep 2025 11:55:21 -0400
Subject: [PATCH v28 07/15] Eliminate XLOG_HEAP2_VISIBLE from empty-page vacuum
As part of removing XLOG_HEAP2_VISIBLE records, phase I of VACUUM now
marks empty pages all-visible in a XLOG_HEAP2_PRUNE_VACUUM_SCAN record.
Author: Melanie Plageman <[email protected]>
Reviewed-by: Robert Haas <[email protected]>
Reviewed-by: Andres Freund <[email protected]>
Reviewed-by: Chao Li <[email protected]>
---
src/backend/access/heap/vacuumlazy.c | 35 +++++++++++++++++++++++-----
1 file changed, 29 insertions(+), 6 deletions(-)
diff --git a/src/backend/access/heap/vacuumlazy.c b/src/backend/access/heap/vacuumlazy.c
index 61564aea5fd..e311e7d6604 100644
--- a/src/backend/access/heap/vacuumlazy.c
+++ b/src/backend/access/heap/vacuumlazy.c
@@ -1866,9 +1866,12 @@ lazy_scan_new_or_empty(LVRelState *vacrel, Buffer buf, BlockNumber blkno,
*/
if (!PageIsAllVisible(page))
{
+ /* Lock vmbuffer before entering critical section */
+ LockBuffer(vmbuffer, BUFFER_LOCK_EXCLUSIVE);
+
START_CRIT_SECTION();
- /* mark buffer dirty before writing a WAL record */
+ /* Mark buffer dirty before writing any WAL records */
MarkBufferDirty(buf);
/*
@@ -1885,13 +1888,33 @@ lazy_scan_new_or_empty(LVRelState *vacrel, Buffer buf, BlockNumber blkno,
log_newpage_buffer(buf, true);
PageSetAllVisible(page);
- visibilitymap_set(vacrel->rel, blkno, buf,
- InvalidXLogRecPtr,
- vmbuffer, InvalidTransactionId,
- VISIBILITYMAP_ALL_VISIBLE |
- VISIBILITYMAP_ALL_FROZEN);
+ visibilitymap_set_vmbits(blkno,
+ vmbuffer,
+ VISIBILITYMAP_ALL_VISIBLE |
+ VISIBILITYMAP_ALL_FROZEN,
+ vacrel->rel->rd_locator);
+
+ /*
+ * Emit WAL for setting PD_ALL_VISIBLE on the heap page and
+ * setting the VM.
+ */
+ if (RelationNeedsWAL(vacrel->rel))
+ log_heap_prune_and_freeze(vacrel->rel, buf,
+ vmbuffer,
+ VISIBILITYMAP_ALL_VISIBLE |
+ VISIBILITYMAP_ALL_FROZEN,
+ InvalidTransactionId, /* conflict xid */
+ false, /* cleanup lock */
+ PRUNE_VACUUM_SCAN, /* reason */
+ NULL, 0,
+ NULL, 0,
+ NULL, 0,
+ NULL, 0);
+
END_CRIT_SECTION();
+ LockBuffer(vmbuffer, BUFFER_LOCK_UNLOCK);
+
/* Count the newly all-frozen pages for logging */
vacrel->vm_new_visible_pages++;
vacrel->vm_new_visible_frozen_pages++;
--
2.43.0
[text/x-patch] v28-0008-Remove-XLOG_HEAP2_VISIBLE-entirely.patch (24.7K, 9-v28-0008-Remove-XLOG_HEAP2_VISIBLE-entirely.patch)
download | inline diff:
From 1acb98f855c1c28993bbd9ccb90a2250e4d64980 Mon Sep 17 00:00:00 2001
From: Melanie Plageman <[email protected]>
Date: Sat, 27 Sep 2025 11:55:36 -0400
Subject: [PATCH v28 08/15] Remove XLOG_HEAP2_VISIBLE entirely
As no remaining users emit XLOG_HEAP2_VISIBLE records.
This includes deleting the xl_heap_visible struct and all functions
responsible for emitting or replaying XLOG_HEAP2_VISIBLE records.
This changes the visibility map API, so any external users/consumers of
the VM-only WAL record will need to change.
Author: Melanie Plageman <[email protected]>
Reviewed-by: Andrey Borodin <[email protected]>
Reviewed-by: Andres Freund <[email protected]>
Reviewed-by: Chao Li <[email protected]>
---
src/backend/access/common/bufmask.c | 4 +-
src/backend/access/heap/heapam.c | 54 +-------
src/backend/access/heap/heapam_xlog.c | 155 ++---------------------
src/backend/access/heap/pruneheap.c | 4 +-
src/backend/access/heap/vacuumlazy.c | 16 +--
src/backend/access/heap/visibilitymap.c | 112 +---------------
src/backend/access/rmgrdesc/heapdesc.c | 10 --
src/backend/replication/logical/decode.c | 1 -
src/backend/storage/ipc/standby.c | 12 +-
src/include/access/heapam_xlog.h | 20 ---
src/include/access/visibilitymap.h | 13 +-
src/include/access/visibilitymapdefs.h | 9 --
src/tools/pgindent/typedefs.list | 1 -
13 files changed, 38 insertions(+), 373 deletions(-)
diff --git a/src/backend/access/common/bufmask.c b/src/backend/access/common/bufmask.c
index bb260cffa68..5f07f179415 100644
--- a/src/backend/access/common/bufmask.c
+++ b/src/backend/access/common/bufmask.c
@@ -56,8 +56,8 @@ mask_page_hint_bits(Page page)
/*
* During replay, if the page LSN has advanced past our XLOG record's LSN,
- * we don't mark the page all-visible. See heap_xlog_visible() for
- * details.
+ * we don't mark the page all-visible. See heap_xlog_prune_and_freeze()
+ * for more details.
*/
PageClearAllVisible(page);
}
diff --git a/src/backend/access/heap/heapam.c b/src/backend/access/heap/heapam.c
index 6daf4a87dec..fb7a7548aa0 100644
--- a/src/backend/access/heap/heapam.c
+++ b/src/backend/access/heap/heapam.c
@@ -2539,11 +2539,11 @@ heap_multi_insert(Relation relation, TupleTableSlot **slots, int ntuples,
else if (all_frozen_set)
{
PageSetAllVisible(page);
- visibilitymap_set_vmbits(BufferGetBlockNumber(buffer),
- vmbuffer,
- VISIBILITYMAP_ALL_VISIBLE |
- VISIBILITYMAP_ALL_FROZEN,
- relation->rd_locator);
+ visibilitymap_set(BufferGetBlockNumber(buffer),
+ vmbuffer,
+ VISIBILITYMAP_ALL_VISIBLE |
+ VISIBILITYMAP_ALL_FROZEN,
+ relation->rd_locator);
}
/*
@@ -8813,50 +8813,6 @@ bottomup_sort_and_shrink(TM_IndexDeleteOp *delstate)
return nblocksfavorable;
}
-/*
- * Perform XLogInsert for a heap-visible operation. 'block' is the block
- * being marked all-visible, and vm_buffer is the buffer containing the
- * corresponding visibility map block. Both should have already been modified
- * and dirtied.
- *
- * snapshotConflictHorizon comes from the largest xmin on the page being
- * marked all-visible. REDO routine uses it to generate recovery conflicts.
- *
- * If checksums or wal_log_hints are enabled, we may also generate a full-page
- * image of heap_buffer. Otherwise, we optimize away the FPI (by specifying
- * REGBUF_NO_IMAGE for the heap buffer), in which case the caller should *not*
- * update the heap page's LSN.
- */
-XLogRecPtr
-log_heap_visible(Relation rel, Buffer heap_buffer, Buffer vm_buffer,
- TransactionId snapshotConflictHorizon, uint8 vmflags)
-{
- xl_heap_visible xlrec;
- XLogRecPtr recptr;
- uint8 flags;
-
- Assert(BufferIsValid(heap_buffer));
- Assert(BufferIsValid(vm_buffer));
-
- xlrec.snapshotConflictHorizon = snapshotConflictHorizon;
- xlrec.flags = vmflags;
- if (RelationIsAccessibleInLogicalDecoding(rel))
- xlrec.flags |= VISIBILITYMAP_XLOG_CATALOG_REL;
- XLogBeginInsert();
- XLogRegisterData(&xlrec, SizeOfHeapVisible);
-
- XLogRegisterBuffer(0, vm_buffer, 0);
-
- flags = REGBUF_STANDARD;
- if (!XLogHintBitIsNeeded())
- flags |= REGBUF_NO_IMAGE;
- XLogRegisterBuffer(1, heap_buffer, flags);
-
- recptr = XLogInsert(RM_HEAP2_ID, XLOG_HEAP2_VISIBLE);
-
- return recptr;
-}
-
/*
* Perform XLogInsert for a heap-update operation. Caller must already
* have modified the buffer(s) and marked them dirty.
diff --git a/src/backend/access/heap/heapam_xlog.c b/src/backend/access/heap/heapam_xlog.c
index 1823feff298..47d2479415e 100644
--- a/src/backend/access/heap/heapam_xlog.c
+++ b/src/backend/access/heap/heapam_xlog.c
@@ -236,7 +236,7 @@ heap_xlog_prune_freeze(XLogReaderState *record)
if (PageIsNew(vmpage))
PageInit(vmpage, BLCKSZ, 0);
- visibilitymap_set_vmbits(blkno, vmbuffer, vmflags, rlocator);
+ visibilitymap_set(blkno, vmbuffer, vmflags, rlocator);
Assert(BufferIsDirty(vmbuffer));
PageSetLSN(vmpage, lsn);
@@ -249,142 +249,6 @@ heap_xlog_prune_freeze(XLogReaderState *record)
XLogRecordPageWithFreeSpace(rlocator, blkno, freespace);
}
-/*
- * Replay XLOG_HEAP2_VISIBLE records.
- *
- * The critical integrity requirement here is that we must never end up with
- * a situation where the visibility map bit is set, and the page-level
- * PD_ALL_VISIBLE bit is clear. If that were to occur, then a subsequent
- * page modification would fail to clear the visibility map bit.
- */
-static void
-heap_xlog_visible(XLogReaderState *record)
-{
- XLogRecPtr lsn = record->EndRecPtr;
- xl_heap_visible *xlrec = (xl_heap_visible *) XLogRecGetData(record);
- Buffer vmbuffer = InvalidBuffer;
- Buffer buffer;
- Page page;
- RelFileLocator rlocator;
- BlockNumber blkno;
- XLogRedoAction action;
-
- Assert((xlrec->flags & VISIBILITYMAP_XLOG_VALID_BITS) == xlrec->flags);
-
- XLogRecGetBlockTag(record, 1, &rlocator, NULL, &blkno);
-
- /*
- * If there are any Hot Standby transactions running that have an xmin
- * horizon old enough that this page isn't all-visible for them, they
- * might incorrectly decide that an index-only scan can skip a heap fetch.
- *
- * NB: It might be better to throw some kind of "soft" conflict here that
- * forces any index-only scan that is in flight to perform heap fetches,
- * rather than killing the transaction outright.
- */
- if (InHotStandby)
- ResolveRecoveryConflictWithSnapshot(xlrec->snapshotConflictHorizon,
- xlrec->flags & VISIBILITYMAP_XLOG_CATALOG_REL,
- rlocator);
-
- /*
- * Read the heap page, if it still exists. If the heap file has dropped or
- * truncated later in recovery, we don't need to update the page, but we'd
- * better still update the visibility map.
- */
- action = XLogReadBufferForRedo(record, 1, &buffer);
- if (action == BLK_NEEDS_REDO)
- {
- /*
- * We don't bump the LSN of the heap page when setting the visibility
- * map bit (unless checksums or wal_hint_bits is enabled, in which
- * case we must). This exposes us to torn page hazards, but since
- * we're not inspecting the existing page contents in any way, we
- * don't care.
- */
- page = BufferGetPage(buffer);
-
- PageSetAllVisible(page);
-
- if (XLogHintBitIsNeeded())
- PageSetLSN(page, lsn);
-
- MarkBufferDirty(buffer);
- }
- else if (action == BLK_RESTORED)
- {
- /*
- * If heap block was backed up, we already restored it and there's
- * nothing more to do. (This can only happen with checksums or
- * wal_log_hints enabled.)
- */
- }
-
- if (BufferIsValid(buffer))
- {
- Size space = PageGetFreeSpace(BufferGetPage(buffer));
-
- UnlockReleaseBuffer(buffer);
-
- /*
- * Since FSM is not WAL-logged and only updated heuristically, it
- * easily becomes stale in standbys. If the standby is later promoted
- * and runs VACUUM, it will skip updating individual free space
- * figures for pages that became all-visible (or all-frozen, depending
- * on the vacuum mode,) which is troublesome when FreeSpaceMapVacuum
- * propagates too optimistic free space values to upper FSM layers;
- * later inserters try to use such pages only to find out that they
- * are unusable. This can cause long stalls when there are many such
- * pages.
- *
- * Forestall those problems by updating FSM's idea about a page that
- * is becoming all-visible or all-frozen.
- *
- * Do this regardless of a full-page image being applied, since the
- * FSM data is not in the page anyway.
- */
- if (xlrec->flags & VISIBILITYMAP_VALID_BITS)
- XLogRecordPageWithFreeSpace(rlocator, blkno, space);
- }
-
- /*
- * Even if we skipped the heap page update due to the LSN interlock, it's
- * still safe to update the visibility map. Any WAL record that clears
- * the visibility map bit does so before checking the page LSN, so any
- * bits that need to be cleared will still be cleared.
- */
- if (XLogReadBufferForRedoExtended(record, 0, RBM_ZERO_ON_ERROR, false,
- &vmbuffer) == BLK_NEEDS_REDO)
- {
- Page vmpage = BufferGetPage(vmbuffer);
- Relation reln;
- uint8 vmbits;
-
- /* initialize the page if it was read as zeros */
- if (PageIsNew(vmpage))
- PageInit(vmpage, BLCKSZ, 0);
-
- /* remove VISIBILITYMAP_XLOG_* */
- vmbits = xlrec->flags & VISIBILITYMAP_VALID_BITS;
-
- /*
- * XLogReadBufferForRedoExtended locked the buffer. But
- * visibilitymap_set will handle locking itself.
- */
- LockBuffer(vmbuffer, BUFFER_LOCK_UNLOCK);
-
- reln = CreateFakeRelcacheEntry(rlocator);
-
- visibilitymap_set(reln, blkno, InvalidBuffer, lsn, vmbuffer,
- xlrec->snapshotConflictHorizon, vmbits);
-
- ReleaseBuffer(vmbuffer);
- FreeFakeRelcacheEntry(reln);
- }
- else if (BufferIsValid(vmbuffer))
- UnlockReleaseBuffer(vmbuffer);
-}
-
/*
* Given an "infobits" field from an XLog record, set the correct bits in the
* given infomask and infomask2 for the tuple touched by the record.
@@ -762,8 +626,8 @@ heap_xlog_multi_insert(XLogReaderState *record)
*
* During recovery, however, no concurrent writers exist. Therefore,
* updating the VM without holding the heap page lock is safe enough. This
- * same approach is taken when replaying xl_heap_visible records (see
- * heap_xlog_visible()).
+ * same approach is taken when replaying XLOG_HEAP2_PRUNE* records (see
+ * heap_xlog_prune_and_freeze()).
*/
if ((xlrec->flags & XLH_INSERT_ALL_FROZEN_SET) &&
XLogReadBufferForRedoExtended(record, 1, RBM_ZERO_ON_ERROR, false,
@@ -775,11 +639,11 @@ heap_xlog_multi_insert(XLogReaderState *record)
if (PageIsNew(vmpage))
PageInit(vmpage, BLCKSZ, 0);
- visibilitymap_set_vmbits(blkno,
- vmbuffer,
- VISIBILITYMAP_ALL_VISIBLE |
- VISIBILITYMAP_ALL_FROZEN,
- rlocator);
+ visibilitymap_set(blkno,
+ vmbuffer,
+ VISIBILITYMAP_ALL_VISIBLE |
+ VISIBILITYMAP_ALL_FROZEN,
+ rlocator);
Assert(BufferIsDirty(vmbuffer));
PageSetLSN(vmpage, lsn);
@@ -1360,9 +1224,6 @@ heap2_redo(XLogReaderState *record)
case XLOG_HEAP2_PRUNE_VACUUM_CLEANUP:
heap_xlog_prune_freeze(record);
break;
- case XLOG_HEAP2_VISIBLE:
- heap_xlog_visible(record);
- break;
case XLOG_HEAP2_MULTI_INSERT:
heap_xlog_multi_insert(record);
break;
diff --git a/src/backend/access/heap/pruneheap.c b/src/backend/access/heap/pruneheap.c
index 3521e70b8d0..86de3613f5e 100644
--- a/src/backend/access/heap/pruneheap.c
+++ b/src/backend/access/heap/pruneheap.c
@@ -1227,8 +1227,8 @@ heap_page_prune_and_freeze(PruneFreezeParams *params,
* REGBUF_NO_CHANGES), it must be marked dirty.
*/
PageSetAllVisible(page);
- visibilitymap_set_vmbits(blockno, vmbuffer, new_vmbits,
- params->relation->rd_locator);
+ visibilitymap_set(blockno, vmbuffer, new_vmbits,
+ params->relation->rd_locator);
}
MarkBufferDirty(buffer);
diff --git a/src/backend/access/heap/vacuumlazy.c b/src/backend/access/heap/vacuumlazy.c
index e311e7d6604..9dec4875e3a 100644
--- a/src/backend/access/heap/vacuumlazy.c
+++ b/src/backend/access/heap/vacuumlazy.c
@@ -1888,11 +1888,11 @@ lazy_scan_new_or_empty(LVRelState *vacrel, Buffer buf, BlockNumber blkno,
log_newpage_buffer(buf, true);
PageSetAllVisible(page);
- visibilitymap_set_vmbits(blkno,
- vmbuffer,
- VISIBILITYMAP_ALL_VISIBLE |
- VISIBILITYMAP_ALL_FROZEN,
- vacrel->rel->rd_locator);
+ visibilitymap_set(blkno,
+ vmbuffer,
+ VISIBILITYMAP_ALL_VISIBLE |
+ VISIBILITYMAP_ALL_FROZEN,
+ vacrel->rel->rd_locator);
/*
* Emit WAL for setting PD_ALL_VISIBLE on the heap page and
@@ -2769,9 +2769,9 @@ lazy_vacuum_heap_page(LVRelState *vacrel, BlockNumber blkno, Buffer buffer,
* set PD_ALL_VISIBLE.
*/
PageSetAllVisible(page);
- visibilitymap_set_vmbits(blkno,
- vmbuffer, vmflags,
- vacrel->rel->rd_locator);
+ visibilitymap_set(blkno,
+ vmbuffer, vmflags,
+ vacrel->rel->rd_locator);
conflict_xid = visibility_cutoff_xid;
}
diff --git a/src/backend/access/heap/visibilitymap.c b/src/backend/access/heap/visibilitymap.c
index d14588e92ae..7997e926872 100644
--- a/src/backend/access/heap/visibilitymap.c
+++ b/src/backend/access/heap/visibilitymap.c
@@ -14,8 +14,7 @@
* visibilitymap_clear - clear bits for one page in the visibility map
* visibilitymap_pin - pin a map page for setting a bit
* visibilitymap_pin_ok - check whether correct map page is already pinned
- * visibilitymap_set - set bit(s) in a previously pinned page and log
- * visibilitymap_set_vmbits - set bit(s) in a pinned page
+ * visibilitymap_set - set bit(s) in a previously pinned page
* visibilitymap_get_status - get status of bits
* visibilitymap_count - count number of bits set in visibility map
* visibilitymap_prepare_truncate -
@@ -220,109 +219,6 @@ visibilitymap_pin_ok(BlockNumber heapBlk, Buffer vmbuf)
return BufferIsValid(vmbuf) && BufferGetBlockNumber(vmbuf) == mapBlock;
}
-/*
- * visibilitymap_set - set bit(s) on a previously pinned page
- *
- * recptr is the LSN of the XLOG record we're replaying, if we're in recovery,
- * or InvalidXLogRecPtr in normal running. The VM page LSN is advanced to the
- * one provided; in normal running, we generate a new XLOG record and set the
- * page LSN to that value (though the heap page's LSN may *not* be updated;
- * see below). cutoff_xid is the largest xmin on the page being marked
- * all-visible; it is needed for Hot Standby, and can be InvalidTransactionId
- * if the page contains no tuples. It can also be set to InvalidTransactionId
- * when a page that is already all-visible is being marked all-frozen.
- *
- * Caller is expected to set the heap page's PD_ALL_VISIBLE bit before calling
- * this function. Except in recovery, caller should also pass the heap
- * buffer. When checksums are enabled and we're not in recovery, we must add
- * the heap buffer to the WAL chain to protect it from being torn.
- *
- * You must pass a buffer containing the correct map page to this function.
- * Call visibilitymap_pin first to pin the right one. This function doesn't do
- * any I/O.
- *
- * Returns the state of the page's VM bits before setting flags.
- */
-uint8
-visibilitymap_set(Relation rel, BlockNumber heapBlk, Buffer heapBuf,
- XLogRecPtr recptr, Buffer vmBuf, TransactionId cutoff_xid,
- uint8 flags)
-{
- BlockNumber mapBlock = HEAPBLK_TO_MAPBLOCK(heapBlk);
- uint32 mapByte = HEAPBLK_TO_MAPBYTE(heapBlk);
- uint8 mapOffset = HEAPBLK_TO_OFFSET(heapBlk);
- Page page;
- uint8 *map;
- uint8 status;
-
-#ifdef TRACE_VISIBILITYMAP
- elog(DEBUG1, "vm_set flags 0x%02X for %s %d",
- flags, RelationGetRelationName(rel), heapBlk);
-#endif
-
- Assert(InRecovery || !XLogRecPtrIsValid(recptr));
- Assert(InRecovery || PageIsAllVisible(BufferGetPage(heapBuf)));
- Assert((flags & VISIBILITYMAP_VALID_BITS) == flags);
-
- /* Must never set all_frozen bit without also setting all_visible bit */
- Assert(flags != VISIBILITYMAP_ALL_FROZEN);
-
- /* Check that we have the right heap page pinned, if present */
- if (BufferIsValid(heapBuf) && BufferGetBlockNumber(heapBuf) != heapBlk)
- elog(ERROR, "wrong heap buffer passed to visibilitymap_set");
-
- Assert(!BufferIsValid(heapBuf) ||
- BufferIsLockedByMeInMode(heapBuf, BUFFER_LOCK_EXCLUSIVE));
-
- /* Check that we have the right VM page pinned */
- if (!BufferIsValid(vmBuf) || BufferGetBlockNumber(vmBuf) != mapBlock)
- elog(ERROR, "wrong VM buffer passed to visibilitymap_set");
-
- page = BufferGetPage(vmBuf);
- map = (uint8 *) PageGetContents(page);
- LockBuffer(vmBuf, BUFFER_LOCK_EXCLUSIVE);
-
- status = (map[mapByte] >> mapOffset) & VISIBILITYMAP_VALID_BITS;
- if (flags != status)
- {
- START_CRIT_SECTION();
-
- map[mapByte] |= (flags << mapOffset);
- MarkBufferDirty(vmBuf);
-
- if (RelationNeedsWAL(rel))
- {
- if (!XLogRecPtrIsValid(recptr))
- {
- Assert(!InRecovery);
- recptr = log_heap_visible(rel, heapBuf, vmBuf, cutoff_xid, flags);
-
- /*
- * If data checksums are enabled (or wal_log_hints=on), we
- * need to protect the heap page from being torn.
- *
- * If not, then we must *not* update the heap page's LSN. In
- * this case, the FPI for the heap page was omitted from the
- * WAL record inserted above, so it would be incorrect to
- * update the heap page's LSN.
- */
- if (XLogHintBitIsNeeded())
- {
- Page heapPage = BufferGetPage(heapBuf);
-
- PageSetLSN(heapPage, recptr);
- }
- }
- PageSetLSN(page, recptr);
- }
-
- END_CRIT_SECTION();
- }
-
- LockBuffer(vmBuf, BUFFER_LOCK_UNLOCK);
- return status;
-}
-
/*
* Set VM (visibility map) flags in the VM block in vmBuf.
*
@@ -344,9 +240,9 @@ visibilitymap_set(Relation rel, BlockNumber heapBlk, Buffer heapBuf,
* rlocator is used only for debugging messages.
*/
uint8
-visibilitymap_set_vmbits(BlockNumber heapBlk,
- Buffer vmBuf, uint8 flags,
- const RelFileLocator rlocator)
+visibilitymap_set(BlockNumber heapBlk,
+ Buffer vmBuf, uint8 flags,
+ const RelFileLocator rlocator)
{
BlockNumber mapBlock = HEAPBLK_TO_MAPBLOCK(heapBlk);
uint32 mapByte = HEAPBLK_TO_MAPBYTE(heapBlk);
diff --git a/src/backend/access/rmgrdesc/heapdesc.c b/src/backend/access/rmgrdesc/heapdesc.c
index ca26d1f0ed1..08461fdf593 100644
--- a/src/backend/access/rmgrdesc/heapdesc.c
+++ b/src/backend/access/rmgrdesc/heapdesc.c
@@ -349,13 +349,6 @@ heap2_desc(StringInfo buf, XLogReaderState *record)
}
}
}
- else if (info == XLOG_HEAP2_VISIBLE)
- {
- xl_heap_visible *xlrec = (xl_heap_visible *) rec;
-
- appendStringInfo(buf, "snapshotConflictHorizon: %u, flags: 0x%02X",
- xlrec->snapshotConflictHorizon, xlrec->flags);
- }
else if (info == XLOG_HEAP2_MULTI_INSERT)
{
xl_heap_multi_insert *xlrec = (xl_heap_multi_insert *) rec;
@@ -461,9 +454,6 @@ heap2_identify(uint8 info)
case XLOG_HEAP2_PRUNE_VACUUM_CLEANUP:
id = "PRUNE_VACUUM_CLEANUP";
break;
- case XLOG_HEAP2_VISIBLE:
- id = "VISIBLE";
- break;
case XLOG_HEAP2_MULTI_INSERT:
id = "MULTI_INSERT";
break;
diff --git a/src/backend/replication/logical/decode.c b/src/backend/replication/logical/decode.c
index 5e15cb1825e..c0cac7ea1c3 100644
--- a/src/backend/replication/logical/decode.c
+++ b/src/backend/replication/logical/decode.c
@@ -454,7 +454,6 @@ heap2_decode(LogicalDecodingContext *ctx, XLogRecordBuffer *buf)
case XLOG_HEAP2_PRUNE_ON_ACCESS:
case XLOG_HEAP2_PRUNE_VACUUM_SCAN:
case XLOG_HEAP2_PRUNE_VACUUM_CLEANUP:
- case XLOG_HEAP2_VISIBLE:
case XLOG_HEAP2_LOCK_UPDATED:
break;
default:
diff --git a/src/backend/storage/ipc/standby.c b/src/backend/storage/ipc/standby.c
index fc45d72c79b..3655358ed6b 100644
--- a/src/backend/storage/ipc/standby.c
+++ b/src/backend/storage/ipc/standby.c
@@ -475,12 +475,12 @@ ResolveRecoveryConflictWithSnapshot(TransactionId snapshotConflictHorizon,
* If we get passed InvalidTransactionId then we do nothing (no conflict).
*
* This can happen when replaying already-applied WAL records after a
- * standby crash or restart, or when replaying an XLOG_HEAP2_VISIBLE
- * record that marks as frozen a page which was already all-visible. It's
- * also quite common with records generated during index deletion
- * (original execution of the deletion can reason that a recovery conflict
- * which is sufficient for the deletion operation must take place before
- * replay of the deletion record itself).
+ * standby crash or restart, or when replaying a record that marks as
+ * frozen a page which was already marked all-visible in the visibility
+ * map. It's also quite common with records generated during index
+ * deletion (original execution of the deletion can reason that a recovery
+ * conflict which is sufficient for the deletion operation must take place
+ * before replay of the deletion record itself).
*/
if (!TransactionIdIsValid(snapshotConflictHorizon))
return;
diff --git a/src/include/access/heapam_xlog.h b/src/include/access/heapam_xlog.h
index 16c2b2e3c9c..b27fcdfb345 100644
--- a/src/include/access/heapam_xlog.h
+++ b/src/include/access/heapam_xlog.h
@@ -60,7 +60,6 @@
#define XLOG_HEAP2_PRUNE_ON_ACCESS 0x10
#define XLOG_HEAP2_PRUNE_VACUUM_SCAN 0x20
#define XLOG_HEAP2_PRUNE_VACUUM_CLEANUP 0x30
-#define XLOG_HEAP2_VISIBLE 0x40
#define XLOG_HEAP2_MULTI_INSERT 0x50
#define XLOG_HEAP2_LOCK_UPDATED 0x60
#define XLOG_HEAP2_NEW_CID 0x70
@@ -443,20 +442,6 @@ typedef struct xl_heap_inplace
#define MinSizeOfHeapInplace (offsetof(xl_heap_inplace, nmsgs) + sizeof(int))
-/*
- * This is what we need to know about setting a visibility map bit
- *
- * Backup blk 0: visibility map buffer
- * Backup blk 1: heap buffer
- */
-typedef struct xl_heap_visible
-{
- TransactionId snapshotConflictHorizon;
- uint8 flags;
-} xl_heap_visible;
-
-#define SizeOfHeapVisible (offsetof(xl_heap_visible, flags) + sizeof(uint8))
-
typedef struct xl_heap_new_cid
{
/*
@@ -500,11 +485,6 @@ extern void heap2_desc(StringInfo buf, XLogReaderState *record);
extern const char *heap2_identify(uint8 info);
extern void heap_xlog_logical_rewrite(XLogReaderState *r);
-extern XLogRecPtr log_heap_visible(Relation rel, Buffer heap_buffer,
- Buffer vm_buffer,
- TransactionId snapshotConflictHorizon,
- uint8 vmflags);
-
/* in heapdesc.c, so it can be shared between frontend/backend code */
extern void heap_xlog_deserialize_prune_and_freeze(char *cursor, uint16 flags,
int *nplans, xlhp_freeze_plan **plans,
diff --git a/src/include/access/visibilitymap.h b/src/include/access/visibilitymap.h
index c6fa37be968..05ba6786b47 100644
--- a/src/include/access/visibilitymap.h
+++ b/src/include/access/visibilitymap.h
@@ -15,7 +15,6 @@
#define VISIBILITYMAP_H
#include "access/visibilitymapdefs.h"
-#include "access/xlogdefs.h"
#include "storage/block.h"
#include "storage/buf.h"
#include "storage/relfilelocator.h"
@@ -32,15 +31,9 @@ extern bool visibilitymap_clear(Relation rel, BlockNumber heapBlk,
extern void visibilitymap_pin(Relation rel, BlockNumber heapBlk,
Buffer *vmbuf);
extern bool visibilitymap_pin_ok(BlockNumber heapBlk, Buffer vmbuf);
-extern uint8 visibilitymap_set(Relation rel,
- BlockNumber heapBlk, Buffer heapBuf,
- XLogRecPtr recptr,
- Buffer vmBuf,
- TransactionId cutoff_xid,
- uint8 flags);
-extern uint8 visibilitymap_set_vmbits(BlockNumber heapBlk,
- Buffer vmBuf, uint8 flags,
- const RelFileLocator rlocator);
+extern uint8 visibilitymap_set(BlockNumber heapBlk,
+ Buffer vmBuf, uint8 flags,
+ const RelFileLocator rlocator);
extern uint8 visibilitymap_get_status(Relation rel, BlockNumber heapBlk, Buffer *vmbuf);
extern void visibilitymap_count(Relation rel, BlockNumber *all_visible, BlockNumber *all_frozen);
extern BlockNumber visibilitymap_prepare_truncate(Relation rel,
diff --git a/src/include/access/visibilitymapdefs.h b/src/include/access/visibilitymapdefs.h
index 5ad5c020877..e01bce4c99f 100644
--- a/src/include/access/visibilitymapdefs.h
+++ b/src/include/access/visibilitymapdefs.h
@@ -21,14 +21,5 @@
#define VISIBILITYMAP_ALL_FROZEN 0x02
#define VISIBILITYMAP_VALID_BITS 0x03 /* OR of all valid visibilitymap
* flags bits */
-/*
- * To detect recovery conflicts during logical decoding on a standby, we need
- * to know if a table is a user catalog table. For that we add an additional
- * bit into xl_heap_visible.flags, in addition to the above.
- *
- * NB: VISIBILITYMAP_XLOG_* may not be passed to visibilitymap_set().
- */
-#define VISIBILITYMAP_XLOG_CATALOG_REL 0x04
-#define VISIBILITYMAP_XLOG_VALID_BITS (VISIBILITYMAP_VALID_BITS | VISIBILITYMAP_XLOG_CATALOG_REL)
#endif /* VISIBILITYMAPDEFS_H */
diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list
index 04845d5e680..6505628120c 100644
--- a/src/tools/pgindent/typedefs.list
+++ b/src/tools/pgindent/typedefs.list
@@ -4331,7 +4331,6 @@ xl_heap_prune
xl_heap_rewrite_mapping
xl_heap_truncate
xl_heap_update
-xl_heap_visible
xl_invalid_page
xl_invalid_page_key
xl_invalidations
--
2.43.0
[text/x-patch] v28-0009-Simplify-heap_page_would_be_all_visible-visibili.patch (2.2K, 10-v28-0009-Simplify-heap_page_would_be_all_visible-visibili.patch)
download | inline diff:
From 8ba420f2ac77650d22905ba7b4660dc70dad9383 Mon Sep 17 00:00:00 2001
From: Melanie Plageman <[email protected]>
Date: Wed, 17 Dec 2025 13:57:16 -0500
Subject: [PATCH v28 09/15] Simplify heap_page_would_be_all_visible visibility
check
heap_page_would_be_all_visible() doesn't care about the distinction
between HEAPTUPLE_RECENTLY_DEAD and HEAPTUPLE_DEAD tuples -- any tuple
that is not HEAPTUPLE_LIVE means the page is not all-visible and causes
us to return false.
Therefore, we don't need to call HeapTupleSatisfiesVacuum(), which
includes an extra step to distinguish between dead and recently dead
tuples using OldestXmin. Replace it with the more minimal
HeapTupleSatisfiesVacuumHorizon().
This has the added benefit of making it easier to replace uses of
OldestXmin in heap_page_would_be_all_visible() in the future.
---
src/backend/access/heap/vacuumlazy.c | 8 +++++++-
1 file changed, 7 insertions(+), 1 deletion(-)
diff --git a/src/backend/access/heap/vacuumlazy.c b/src/backend/access/heap/vacuumlazy.c
index 9dec4875e3a..441b4883d89 100644
--- a/src/backend/access/heap/vacuumlazy.c
+++ b/src/backend/access/heap/vacuumlazy.c
@@ -3535,6 +3535,7 @@ heap_page_would_be_all_visible(Relation rel, Buffer buf,
{
ItemId itemid;
HeapTupleData tuple;
+ TransactionId dead_after = InvalidTransactionId;
/*
* Set the offset number so that we can display it along with any
@@ -3574,12 +3575,14 @@ heap_page_would_be_all_visible(Relation rel, Buffer buf,
/* Visibility checks may do IO or allocate memory */
Assert(CritSectionCount == 0);
- switch (HeapTupleSatisfiesVacuum(&tuple, OldestXmin, buf))
+ switch (HeapTupleSatisfiesVacuumHorizon(&tuple, buf, &dead_after))
{
case HEAPTUPLE_LIVE:
{
TransactionId xmin;
+ Assert(!TransactionIdIsValid(dead_after));
+
/* Check comments in lazy_scan_prune. */
if (!HeapTupleHeaderXminCommitted(tuple.t_data))
{
@@ -3614,6 +3617,9 @@ heap_page_would_be_all_visible(Relation rel, Buffer buf,
case HEAPTUPLE_DEAD:
case HEAPTUPLE_RECENTLY_DEAD:
+ Assert(TransactionIdIsValid(dead_after));
+ /* FALLTHROUGH */
+
case HEAPTUPLE_INSERT_IN_PROGRESS:
case HEAPTUPLE_DELETE_IN_PROGRESS:
{
--
2.43.0
[text/x-patch] v28-0010-Use-GlobalVisState-in-vacuum-to-determine-page-l.patch (11.9K, 11-v28-0010-Use-GlobalVisState-in-vacuum-to-determine-page-l.patch)
download | inline diff:
From 7af3b6f670fd8e4d0bc2141d8d11d54696bc459c Mon Sep 17 00:00:00 2001
From: Melanie Plageman <[email protected]>
Date: Wed, 17 Dec 2025 16:51:05 -0500
Subject: [PATCH v28 10/15] Use GlobalVisState in vacuum to determine page
level visibility
During vacuum's first and third phases, we examine tuples' visibility
to determine if we can set the page all-visible in the visibility map.
Previously, this check compared tuple xmins against a single XID chosen at
the start of vacuum (OldestXmin). We now use GlobalVisState, which also
enables future work to set the VM during on-access pruning, since ordinary
queries have access to GlobalVisState but not OldestXmin.
This also benefits vacuum: in some cases, GlobalVisState may advance
during a vacuum, allowing more pages to become considered all-visible.
And, in the future, we could easily add a heuristic to update
GlobalVisState more frequently during vacuums of large tables.
OldestXmin is still used for freezing and as a backstop to ensure we
don't freeze a dead tuple that wasn't yet prunable according to
GlobalVisState in the rare occurrences where GlobalVisState moves
backwards.
Because comparing a transaction ID against GlobalVisState is more
expensive than comparing against a single XID, we defer this check until
after scanning all tuples on the page. If visibility_cutoff_xid was
maintained, we perform the GlobalVisState check only once per page.
This is safe because visibility_cutoff_xid records the newest xmin on
the page; if it is globally visible, then the entire page is all-visible.
This approach may result in examining more tuple xmins than before,
since with OldestXmin we could sometimes rule out the page being
all-visible earlier. However, profiling shows the additional cost is not
significant.
Reviewed-by: Andres Freund <[email protected]>
Reviewed-by: Chao Li <[email protected]>
---
src/backend/access/heap/heapam_visibility.c | 22 +++++++++
src/backend/access/heap/pruneheap.c | 53 ++++++++++-----------
src/backend/access/heap/vacuumlazy.c | 38 ++++++++++-----
src/include/access/heapam.h | 4 +-
4 files changed, 76 insertions(+), 41 deletions(-)
diff --git a/src/backend/access/heap/heapam_visibility.c b/src/backend/access/heap/heapam_visibility.c
index 05f6946fe60..f4ab1c13169 100644
--- a/src/backend/access/heap/heapam_visibility.c
+++ b/src/backend/access/heap/heapam_visibility.c
@@ -1189,6 +1189,28 @@ HeapTupleSatisfiesVacuum(HeapTuple htup, TransactionId OldestXmin,
return res;
}
+/*
+ * Wrapper around GlobalVisTestIsRemovableXid() for use when examining live
+ * tuples. Returns true if the given XID may be considered running by at least
+ * one snapshot.
+ *
+ * This function alone is insufficient to determine tuple visibility; callers
+ * must also consider the XID's commit status. Its purpose is purely semantic:
+ * when applied to live tuples, GlobalVisTestIsRemovableXid() is checking
+ * whether the inserting transaction is still considered running, not whether
+ * the tuple is removable. Live tuples are, by definition, not removable, but
+ * the snapshot criteria for “transaction still running” are identical to
+ * those used for removal XIDs.
+ *
+ * See the comment above GlobalVisTestIsRemovable[Full]Xid() for details on the
+ * required preconditions for calling this function.
+ */
+bool
+GlobalVisTestXidMaybeRunning(GlobalVisState *state, TransactionId xid)
+{
+ return !GlobalVisTestIsRemovableXid(state, xid);
+}
+
/*
* Work horse for HeapTupleSatisfiesVacuum and similar routines.
*
diff --git a/src/backend/access/heap/pruneheap.c b/src/backend/access/heap/pruneheap.c
index 86de3613f5e..e0b19b3e669 100644
--- a/src/backend/access/heap/pruneheap.c
+++ b/src/backend/access/heap/pruneheap.c
@@ -461,11 +461,12 @@ prune_freeze_setup(PruneFreezeParams *params,
/*
* The visibility cutoff xid is the newest xmin of live, committed tuples
- * older than OldestXmin on the page. This field is only kept up-to-date
- * if the page is all-visible. As soon as a tuple is encountered that is
- * not visible to all, this field is unmaintained. As long as it is
- * maintained, it can be used to calculate the snapshot conflict horizon
- * when updating the VM and/or freezing all the tuples on the page.
+ * on the page older than the visibility horizon represented in the
+ * GlobalVisState. This field is only kept up-to-date if the page is
+ * all-visible. As soon as a tuple is encountered that is not visible to
+ * all, this field is unmaintained. As long as it is maintained, it can be
+ * used to calculate the snapshot conflict horizon when updating the VM
+ * and/or freezing all the tuples on the page.
*/
prstate->visibility_cutoff_xid = InvalidTransactionId;
}
@@ -994,14 +995,14 @@ heap_page_will_set_vm(PruneState *prstate,
*/
static bool
heap_page_is_all_visible(Relation rel, Buffer buf,
- TransactionId OldestXmin,
+ GlobalVisState *vistest,
bool *all_frozen,
TransactionId *visibility_cutoff_xid,
OffsetNumber *logging_offnum)
{
return heap_page_would_be_all_visible(rel, buf,
- OldestXmin,
+ vistest,
NULL, 0,
all_frozen,
visibility_cutoff_xid,
@@ -1088,6 +1089,16 @@ heap_page_prune_and_freeze(PruneFreezeParams *params,
prune_freeze_plan(RelationGetRelid(params->relation),
buffer, &prstate, off_loc);
+ /*
+ * After processing all the live tuples on the page, if the newest xmin
+ * amongst them may be considered running by any snapshot, the page cannot
+ * be all-visible.
+ */
+ if (prstate.all_visible &&
+ TransactionIdIsNormal(prstate.visibility_cutoff_xid) &&
+ GlobalVisTestXidMaybeRunning(prstate.vistest, prstate.visibility_cutoff_xid))
+ prstate.all_visible = prstate.all_frozen = false;
+
/*
* If checksums are enabled, calling heap_prune_satisfies_vacuum() while
* checking tuple visibility information in prune_freeze_plan() may have
@@ -1270,10 +1281,9 @@ heap_page_prune_and_freeze(PruneFreezeParams *params,
bool debug_all_frozen;
Assert(prstate.lpdead_items == 0);
- Assert(prstate.cutoffs);
Assert(heap_page_is_all_visible(params->relation, buffer,
- prstate.cutoffs->OldestXmin,
+ prstate.vistest,
&debug_all_frozen,
&debug_cutoff, off_loc));
@@ -1794,28 +1804,15 @@ heap_prune_record_unchanged_lp_normal(Page page, PruneState *prstate, OffsetNumb
}
/*
- * The inserter definitely committed. But is it old enough
- * that everyone sees it as committed? A FrozenTransactionId
- * is seen as committed to everyone. Otherwise, we check if
- * there is a snapshot that considers this xid to still be
- * running, and if so, we don't consider the page all-visible.
+ * The inserter definitely committed. But we don't know if it
+ * is old enough that everyone sees it as committed. Later,
+ * after processing all the tuples on the page, we'll check if
+ * there is any snapshot that still consider the newest xid on
+ * the page to be running. If so, we don't consider the page
+ * all-visible.
*/
xmin = HeapTupleHeaderGetXmin(htup);
- /*
- * For now always use prstate->cutoffs for this test, because
- * we only update 'all_visible' and 'all_frozen' when freezing
- * is requested. We could use GlobalVisTestIsRemovableXid
- * instead, if a non-freezing caller wanted to set the VM bit.
- */
- Assert(prstate->cutoffs);
- if (!TransactionIdPrecedes(xmin, prstate->cutoffs->OldestXmin))
- {
- prstate->all_visible = false;
- prstate->all_frozen = false;
- break;
- }
-
/* Track newest xmin on page. */
if (TransactionIdFollows(xmin, prstate->visibility_cutoff_xid) &&
TransactionIdIsNormal(xmin))
diff --git a/src/backend/access/heap/vacuumlazy.c b/src/backend/access/heap/vacuumlazy.c
index 441b4883d89..082cdbc5de8 100644
--- a/src/backend/access/heap/vacuumlazy.c
+++ b/src/backend/access/heap/vacuumlazy.c
@@ -2723,7 +2723,7 @@ lazy_vacuum_heap_page(LVRelState *vacrel, BlockNumber blkno, Buffer buffer,
* done outside the critical section.
*/
if (heap_page_would_be_all_visible(vacrel->rel, buffer,
- vacrel->cutoffs.OldestXmin,
+ vacrel->vistest,
deadoffsets, num_offsets,
&all_frozen, &visibility_cutoff_xid,
&vacrel->offnum))
@@ -3484,7 +3484,7 @@ dead_items_cleanup(LVRelState *vacrel)
* Returns true if the page is all-visible other than the provided
* deadoffsets and false otherwise.
*
- * OldestXmin is used to determine visibility.
+ * vistest is used to determine visibility.
*
* Output parameters:
*
@@ -3500,7 +3500,7 @@ dead_items_cleanup(LVRelState *vacrel)
*/
bool
heap_page_would_be_all_visible(Relation rel, Buffer buf,
- TransactionId OldestXmin,
+ GlobalVisState *vistest,
OffsetNumber *deadoffsets,
int ndeadoffsets,
bool *all_frozen,
@@ -3583,7 +3583,7 @@ heap_page_would_be_all_visible(Relation rel, Buffer buf,
Assert(!TransactionIdIsValid(dead_after));
- /* Check comments in lazy_scan_prune. */
+ /* Check heap_prune_record_unchanged_lp_normal comments */
if (!HeapTupleHeaderXminCommitted(tuple.t_data))
{
all_visible = false;
@@ -3592,16 +3592,17 @@ heap_page_would_be_all_visible(Relation rel, Buffer buf,
}
/*
- * The inserter definitely committed. But is it old enough
- * that everyone sees it as committed?
+ * The inserter definitely committed. But we don't know if
+ * it is old enough that everyone sees it as committed.
+ * Don't check that now.
+ *
+ * If we scan all tuples without finding one that prevents
+ * the page from being all-visible, we then check whether
+ * any snapshot still considers the newest XID on the page
+ * to be running. In that case, the page is not considered
+ * all-visible.
*/
xmin = HeapTupleHeaderGetXmin(tuple.t_data);
- if (!TransactionIdPrecedes(xmin, OldestXmin))
- {
- all_visible = false;
- *all_frozen = false;
- break;
- }
/* Track newest xmin on page. */
if (TransactionIdFollows(xmin, *visibility_cutoff_xid) &&
@@ -3633,6 +3634,19 @@ heap_page_would_be_all_visible(Relation rel, Buffer buf,
}
} /* scan along page */
+ /*
+ * After processing all the live tuples on the page, if the newest xmin
+ * among them may still be considered running by any snapshot, the page
+ * cannot be all-visible.
+ */
+ if (all_visible &&
+ TransactionIdIsNormal(*visibility_cutoff_xid) &&
+ GlobalVisTestXidMaybeRunning(vistest, *visibility_cutoff_xid))
+ {
+ all_visible = false;
+ *all_frozen = false;
+ }
+
/* Clear the offset information once we have processed the given page. */
*logging_offnum = InvalidOffsetNumber;
diff --git a/src/include/access/heapam.h b/src/include/access/heapam.h
index bec2f840102..1625b107575 100644
--- a/src/include/access/heapam.h
+++ b/src/include/access/heapam.h
@@ -439,7 +439,7 @@ extern void log_heap_prune_and_freeze(Relation relation, Buffer buffer,
extern void heap_vacuum_rel(Relation rel,
const VacuumParams params, BufferAccessStrategy bstrategy);
extern bool heap_page_would_be_all_visible(Relation rel, Buffer buf,
- TransactionId OldestXmin,
+ GlobalVisState *vistest,
OffsetNumber *deadoffsets,
int ndeadoffsets,
bool *all_frozen,
@@ -453,6 +453,8 @@ extern TM_Result HeapTupleSatisfiesUpdate(HeapTuple htup, CommandId curcid,
Buffer buffer);
extern HTSV_Result HeapTupleSatisfiesVacuum(HeapTuple htup, TransactionId OldestXmin,
Buffer buffer);
+
+extern bool GlobalVisTestXidMaybeRunning(GlobalVisState *state, TransactionId xid);
extern HTSV_Result HeapTupleSatisfiesVacuumHorizon(HeapTuple htup, Buffer buffer,
TransactionId *dead_after);
extern void HeapTupleSetHintBits(HeapTupleHeader tuple, Buffer buffer,
--
2.43.0
[text/x-patch] v28-0011-Unset-all_visible-sooner-if-not-freezing.patch (2.5K, 12-v28-0011-Unset-all_visible-sooner-if-not-freezing.patch)
download | inline diff:
From 11d439fcc4ae35a31be204c2eb8a36d52c162d08 Mon Sep 17 00:00:00 2001
From: Melanie Plageman <[email protected]>
Date: Tue, 14 Oct 2025 15:22:35 -0400
Subject: [PATCH v28 11/15] Unset all_visible sooner if not freezing
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
In the prune/freeze path, we currently delay clearing all_visible and
all_frozen in the presence of dead items to allow opportunistic
freezing.
However, if no freezing will be attempted, there’s no need to delay.
Clearing the flags earlier avoids extra bookkeeping in
heap_prune_record_unchanged_lp_normal(). This currently has no runtime
effect because all callers that consider setting the VM also prepare
freeze plans, but upcoming changes will allow on-access pruning to set
the VM without freezing. The extra bookkeeping was noticeable in a
profile of on-access VM setting.
Reviewed-by: Chao Li <[email protected]>
---
src/backend/access/heap/pruneheap.c | 15 +++++++++++++--
1 file changed, 13 insertions(+), 2 deletions(-)
diff --git a/src/backend/access/heap/pruneheap.c b/src/backend/access/heap/pruneheap.c
index e0b19b3e669..1fa72e19f0d 100644
--- a/src/backend/access/heap/pruneheap.c
+++ b/src/backend/access/heap/pruneheap.c
@@ -1669,8 +1669,13 @@ heap_prune_record_dead(PruneState *prstate, OffsetNumber offnum,
/*
* Deliberately delay unsetting all_visible and all_frozen until later
* during pruning. Removable dead tuples shouldn't preclude freezing the
- * page.
+ * page. If we won't attempt freezing, just unset all-visible now, though.
*/
+ if (!prstate->attempt_freeze)
+ {
+ prstate->all_visible = false;
+ prstate->all_frozen = false;
+ }
/* Record the dead offset for vacuum */
prstate->deadoffsets[prstate->lpdead_items++] = offnum;
@@ -1930,8 +1935,14 @@ heap_prune_record_unchanged_lp_dead(Page page, PruneState *prstate, OffsetNumber
* Similarly, don't unset all_visible and all_frozen until later, at the
* end of heap_page_prune_and_freeze(). This will allow us to attempt to
* freeze the page after pruning. As long as we unset it before updating
- * the visibility map, this will be correct.
+ * the visibility map, this will be correct. If we won't attempt freezing,
+ * though, just unset all-visible now.
*/
+ if (!prstate->attempt_freeze)
+ {
+ prstate->all_visible = false;
+ prstate->all_frozen = false;
+ }
/* Record the dead offset for vacuum */
prstate->deadoffsets[prstate->lpdead_items++] = offnum;
--
2.43.0
[text/x-patch] v28-0012-Track-which-relations-are-modified-by-a-query.patch (2.5K, 13-v28-0012-Track-which-relations-are-modified-by-a-query.patch)
download | inline diff:
From afc918c60fd21be6339d88aa4024f908daeab8d3 Mon Sep 17 00:00:00 2001
From: Melanie Plageman <[email protected]>
Date: Wed, 3 Dec 2025 15:07:24 -0500
Subject: [PATCH v28 12/15] Track which relations are modified by a query
Save the relids in a bitmap in the estate. A later commit will pass this
information down to scan nodes to control whether or not the scan allows
setting the visibility map while on-access pruning. We don't want to set
the visibility map if the query is just going to modify the page
immediately after.
---
src/backend/executor/execMain.c | 4 ++++
src/backend/executor/execUtils.c | 2 ++
src/include/nodes/execnodes.h | 6 ++++++
3 files changed, 12 insertions(+)
diff --git a/src/backend/executor/execMain.c b/src/backend/executor/execMain.c
index 797d8b1ca1c..5b299ef81aa 100644
--- a/src/backend/executor/execMain.c
+++ b/src/backend/executor/execMain.c
@@ -916,6 +916,10 @@ InitPlan(QueryDesc *queryDesc, int eflags)
break;
}
+ /* If it has a rowmark, the relation is modified */
+ estate->es_modified_relids = bms_add_member(estate->es_modified_relids,
+ rc->rti);
+
/* Check that relation is a legal target for marking */
if (relation)
CheckValidRowMarkRel(relation, rc->markType);
diff --git a/src/backend/executor/execUtils.c b/src/backend/executor/execUtils.c
index fdc65c2b42b..28a06dcd244 100644
--- a/src/backend/executor/execUtils.c
+++ b/src/backend/executor/execUtils.c
@@ -893,6 +893,8 @@ ExecInitResultRelation(EState *estate, ResultRelInfo *resultRelInfo,
estate->es_result_relations = (ResultRelInfo **)
palloc0(estate->es_range_table_size * sizeof(ResultRelInfo *));
estate->es_result_relations[rti - 1] = resultRelInfo;
+ estate->es_modified_relids = bms_add_member(estate->es_modified_relids,
+ rti);
/*
* Saving in the list allows to avoid needlessly traversing the whole
diff --git a/src/include/nodes/execnodes.h b/src/include/nodes/execnodes.h
index 3968429f991..d8c385216e0 100644
--- a/src/include/nodes/execnodes.h
+++ b/src/include/nodes/execnodes.h
@@ -676,6 +676,12 @@ typedef struct EState
* ExecDoInitialPruning() */
const char *es_sourceText; /* Source text from QueryDesc */
+ /*
+ * RT indexes of relations modified by the query either through
+ * UPDATE/DELETE/INSERT/MERGE or SELECT FOR UPDATE
+ */
+ Bitmapset *es_modified_relids;
+
JunkFilter *es_junkFilter; /* top-level junk filter, if any */
/* If query can insert/delete tuples, the command ID to mark them with */
--
2.43.0
[text/x-patch] v28-0013-Pass-down-information-on-table-modification-to-s.patch (23.7K, 14-v28-0013-Pass-down-information-on-table-modification-to-s.patch)
download | inline diff:
From a64995852e82d8132f23c1bb361e0f3c8080389a Mon Sep 17 00:00:00 2001
From: Melanie Plageman <[email protected]>
Date: Wed, 3 Dec 2025 15:12:18 -0500
Subject: [PATCH v28 13/15] Pass down information on table modification to scan
node
Pass down information to sequential scan, index scan, and bitmap table
scan nodes on whether or not the query modifies the relation being
scanned. A later commit will use this information to update the VM
during on-access pruning only if the relation is not modified by the
query.
---
contrib/pgrowlocks/pgrowlocks.c | 2 +-
src/backend/access/brin/brin.c | 3 ++-
src/backend/access/gin/gininsert.c | 3 ++-
src/backend/access/heap/heapam_handler.c | 7 +++---
src/backend/access/index/genam.c | 4 ++--
src/backend/access/index/indexam.c | 6 +++---
src/backend/access/nbtree/nbtsort.c | 2 +-
src/backend/access/table/tableam.c | 7 +++---
src/backend/commands/constraint.c | 2 +-
src/backend/commands/copyto.c | 2 +-
src/backend/commands/tablecmds.c | 8 +++----
src/backend/commands/typecmds.c | 4 ++--
src/backend/executor/execIndexing.c | 2 +-
src/backend/executor/execReplication.c | 8 +++----
src/backend/executor/nodeBitmapHeapscan.c | 9 +++++++-
src/backend/executor/nodeIndexonlyscan.c | 2 +-
src/backend/executor/nodeIndexscan.c | 11 ++++++++--
src/backend/executor/nodeSeqscan.c | 26 ++++++++++++++++++++---
src/backend/partitioning/partbounds.c | 2 +-
src/backend/utils/adt/selfuncs.c | 2 +-
src/include/access/genam.h | 2 +-
src/include/access/heapam.h | 6 ++++++
src/include/access/tableam.h | 19 ++++++++++-------
23 files changed, 93 insertions(+), 46 deletions(-)
diff --git a/contrib/pgrowlocks/pgrowlocks.c b/contrib/pgrowlocks/pgrowlocks.c
index f88269332b6..27f01d8055f 100644
--- a/contrib/pgrowlocks/pgrowlocks.c
+++ b/contrib/pgrowlocks/pgrowlocks.c
@@ -114,7 +114,7 @@ pgrowlocks(PG_FUNCTION_ARGS)
RelationGetRelationName(rel));
/* Scan the relation */
- scan = table_beginscan(rel, GetActiveSnapshot(), 0, NULL);
+ scan = table_beginscan(rel, GetActiveSnapshot(), 0, NULL, 0);
hscan = (HeapScanDesc) scan;
attinmeta = TupleDescGetAttInMetadata(rsinfo->setDesc);
diff --git a/src/backend/access/brin/brin.c b/src/backend/access/brin/brin.c
index 26cb75058d1..4ad8941c60a 100644
--- a/src/backend/access/brin/brin.c
+++ b/src/backend/access/brin/brin.c
@@ -2842,7 +2842,8 @@ _brin_parallel_scan_and_build(BrinBuildState *state,
indexInfo->ii_Concurrent = brinshared->isconcurrent;
scan = table_beginscan_parallel(heap,
- ParallelTableScanFromBrinShared(brinshared));
+ ParallelTableScanFromBrinShared(brinshared),
+ 0);
reltuples = table_index_build_scan(heap, index, indexInfo, true, true,
brinbuildCallbackParallel, state, scan);
diff --git a/src/backend/access/gin/gininsert.c b/src/backend/access/gin/gininsert.c
index df30dcc0228..aaa5401b731 100644
--- a/src/backend/access/gin/gininsert.c
+++ b/src/backend/access/gin/gininsert.c
@@ -2058,7 +2058,8 @@ _gin_parallel_scan_and_build(GinBuildState *state,
indexInfo->ii_Concurrent = ginshared->isconcurrent;
scan = table_beginscan_parallel(heap,
- ParallelTableScanFromGinBuildShared(ginshared));
+ ParallelTableScanFromGinBuildShared(ginshared),
+ 0);
reltuples = table_index_build_scan(heap, index, indexInfo, true, progress,
ginBuildCallbackParallel, state, scan);
diff --git a/src/backend/access/heap/heapam_handler.c b/src/backend/access/heap/heapam_handler.c
index dd4fe6bf62f..6c2e4e08b16 100644
--- a/src/backend/access/heap/heapam_handler.c
+++ b/src/backend/access/heap/heapam_handler.c
@@ -79,12 +79,13 @@ heapam_slot_callbacks(Relation relation)
*/
static IndexFetchTableData *
-heapam_index_fetch_begin(Relation rel)
+heapam_index_fetch_begin(Relation rel, uint32 flags)
{
IndexFetchHeapData *hscan = palloc0_object(IndexFetchHeapData);
hscan->xs_base.rel = rel;
hscan->xs_cbuf = InvalidBuffer;
+ hscan->modifies_base_rel = !(flags & SO_HINT_REL_READ_ONLY);
return &hscan->xs_base;
}
@@ -753,7 +754,7 @@ heapam_relation_copy_for_cluster(Relation OldHeap, Relation NewHeap,
tableScan = NULL;
heapScan = NULL;
- indexScan = index_beginscan(OldHeap, OldIndex, SnapshotAny, NULL, 0, 0);
+ indexScan = index_beginscan(OldHeap, OldIndex, SnapshotAny, NULL, 0, 0, 0);
index_rescan(indexScan, NULL, 0, NULL, 0);
}
else
@@ -762,7 +763,7 @@ heapam_relation_copy_for_cluster(Relation OldHeap, Relation NewHeap,
pgstat_progress_update_param(PROGRESS_CLUSTER_PHASE,
PROGRESS_CLUSTER_PHASE_SEQ_SCAN_HEAP);
- tableScan = table_beginscan(OldHeap, SnapshotAny, 0, (ScanKey) NULL);
+ tableScan = table_beginscan(OldHeap, SnapshotAny, 0, (ScanKey) NULL, 0);
heapScan = (HeapScanDesc) tableScan;
indexScan = NULL;
diff --git a/src/backend/access/index/genam.c b/src/backend/access/index/genam.c
index b7f10a1aed0..15f9cc11582 100644
--- a/src/backend/access/index/genam.c
+++ b/src/backend/access/index/genam.c
@@ -447,7 +447,7 @@ systable_beginscan(Relation heapRelation,
}
sysscan->iscan = index_beginscan(heapRelation, irel,
- snapshot, NULL, nkeys, 0);
+ snapshot, NULL, nkeys, 0, 0);
index_rescan(sysscan->iscan, idxkey, nkeys, NULL, 0);
sysscan->scan = NULL;
@@ -708,7 +708,7 @@ systable_beginscan_ordered(Relation heapRelation,
}
sysscan->iscan = index_beginscan(heapRelation, indexRelation,
- snapshot, NULL, nkeys, 0);
+ snapshot, NULL, nkeys, 0, 0);
index_rescan(sysscan->iscan, idxkey, nkeys, NULL, 0);
sysscan->scan = NULL;
diff --git a/src/backend/access/index/indexam.c b/src/backend/access/index/indexam.c
index 0492d92d23b..b5523cf2ab1 100644
--- a/src/backend/access/index/indexam.c
+++ b/src/backend/access/index/indexam.c
@@ -257,7 +257,7 @@ index_beginscan(Relation heapRelation,
Relation indexRelation,
Snapshot snapshot,
IndexScanInstrumentation *instrument,
- int nkeys, int norderbys)
+ int nkeys, int norderbys, uint32 flags)
{
IndexScanDesc scan;
@@ -284,7 +284,7 @@ index_beginscan(Relation heapRelation,
scan->instrument = instrument;
/* prepare to fetch index matches from table */
- scan->xs_heapfetch = table_index_fetch_begin(heapRelation);
+ scan->xs_heapfetch = table_index_fetch_begin(heapRelation, flags);
return scan;
}
@@ -615,7 +615,7 @@ index_beginscan_parallel(Relation heaprel, Relation indexrel,
scan->instrument = instrument;
/* prepare to fetch index matches from table */
- scan->xs_heapfetch = table_index_fetch_begin(heaprel);
+ scan->xs_heapfetch = table_index_fetch_begin(heaprel, 0);
return scan;
}
diff --git a/src/backend/access/nbtree/nbtsort.c b/src/backend/access/nbtree/nbtsort.c
index d7695dc1108..7bdbc7e5fa7 100644
--- a/src/backend/access/nbtree/nbtsort.c
+++ b/src/backend/access/nbtree/nbtsort.c
@@ -1925,7 +1925,7 @@ _bt_parallel_scan_and_sort(BTSpool *btspool, BTSpool *btspool2,
indexInfo = BuildIndexInfo(btspool->index);
indexInfo->ii_Concurrent = btshared->isconcurrent;
scan = table_beginscan_parallel(btspool->heap,
- ParallelTableScanFromBTShared(btshared));
+ ParallelTableScanFromBTShared(btshared), 0);
reltuples = table_index_build_scan(btspool->heap, btspool->index, indexInfo,
true, progress, _bt_build_callback,
&buildstate, scan);
diff --git a/src/backend/access/table/tableam.c b/src/backend/access/table/tableam.c
index 73ebc01a08f..a00bdfdf822 100644
--- a/src/backend/access/table/tableam.c
+++ b/src/backend/access/table/tableam.c
@@ -163,10 +163,11 @@ table_parallelscan_initialize(Relation rel, ParallelTableScanDesc pscan,
}
TableScanDesc
-table_beginscan_parallel(Relation relation, ParallelTableScanDesc pscan)
+table_beginscan_parallel(Relation relation, ParallelTableScanDesc pscan, uint32 flags)
{
Snapshot snapshot;
- uint32 flags = SO_TYPE_SEQSCAN |
+
+ flags |= SO_TYPE_SEQSCAN |
SO_ALLOW_STRAT | SO_ALLOW_SYNC | SO_ALLOW_PAGEMODE;
Assert(RelFileLocatorEquals(relation->rd_locator, pscan->phs_locator));
@@ -248,7 +249,7 @@ table_index_fetch_tuple_check(Relation rel,
bool found;
slot = table_slot_create(rel, NULL);
- scan = table_index_fetch_begin(rel);
+ scan = table_index_fetch_begin(rel, 0);
found = table_index_fetch_tuple(scan, tid, snapshot, slot, &call_again,
all_dead);
table_index_fetch_end(scan);
diff --git a/src/backend/commands/constraint.c b/src/backend/commands/constraint.c
index 3497a8221f2..97c8278e36d 100644
--- a/src/backend/commands/constraint.c
+++ b/src/backend/commands/constraint.c
@@ -106,7 +106,7 @@ unique_key_recheck(PG_FUNCTION_ARGS)
*/
tmptid = checktid;
{
- IndexFetchTableData *scan = table_index_fetch_begin(trigdata->tg_relation);
+ IndexFetchTableData *scan = table_index_fetch_begin(trigdata->tg_relation, 0);
bool call_again = false;
if (!table_index_fetch_tuple(scan, &tmptid, SnapshotSelf, slot,
diff --git a/src/backend/commands/copyto.c b/src/backend/commands/copyto.c
index dae91630ac3..1957bb0f1a2 100644
--- a/src/backend/commands/copyto.c
+++ b/src/backend/commands/copyto.c
@@ -1156,7 +1156,7 @@ CopyRelationTo(CopyToState cstate, Relation rel, Relation root_rel, uint64 *proc
AttrMap *map = NULL;
TupleTableSlot *root_slot = NULL;
- scandesc = table_beginscan(rel, GetActiveSnapshot(), 0, NULL);
+ scandesc = table_beginscan(rel, GetActiveSnapshot(), 0, NULL, 0);
slot = table_slot_create(rel, NULL);
/*
diff --git a/src/backend/commands/tablecmds.c b/src/backend/commands/tablecmds.c
index 6b1a00ed477..130a670d266 100644
--- a/src/backend/commands/tablecmds.c
+++ b/src/backend/commands/tablecmds.c
@@ -6377,7 +6377,7 @@ ATRewriteTable(AlteredTableInfo *tab, Oid OIDNewHeap)
* checking all the constraints.
*/
snapshot = RegisterSnapshot(GetLatestSnapshot());
- scan = table_beginscan(oldrel, snapshot, 0, NULL);
+ scan = table_beginscan(oldrel, snapshot, 0, NULL, 0);
/*
* Switch to per-tuple memory context and reset it for each tuple
@@ -13766,7 +13766,7 @@ validateForeignKeyConstraint(char *conname,
*/
snapshot = RegisterSnapshot(GetLatestSnapshot());
slot = table_slot_create(rel, NULL);
- scan = table_beginscan(rel, snapshot, 0, NULL);
+ scan = table_beginscan(rel, snapshot, 0, NULL, 0);
perTupCxt = AllocSetContextCreate(CurrentMemoryContext,
"validateForeignKeyConstraint",
@@ -22623,7 +22623,7 @@ MergePartitionsMoveRows(List **wqueue, List *mergingPartitions, Relation newPart
/* Scan through the rows. */
snapshot = RegisterSnapshot(GetLatestSnapshot());
- scan = table_beginscan(mergingPartition, snapshot, 0, NULL);
+ scan = table_beginscan(mergingPartition, snapshot, 0, NULL, 0);
/*
* Switch to per-tuple memory context and reset it for each tuple
@@ -23087,7 +23087,7 @@ SplitPartitionMoveRows(List **wqueue, Relation rel, Relation splitRel,
/* Scan through the rows. */
snapshot = RegisterSnapshot(GetLatestSnapshot());
- scan = table_beginscan(splitRel, snapshot, 0, NULL);
+ scan = table_beginscan(splitRel, snapshot, 0, NULL, 0);
/*
* Switch to per-tuple memory context and reset it for each tuple
diff --git a/src/backend/commands/typecmds.c b/src/backend/commands/typecmds.c
index be6ffd6ddb0..2921f68c1c3 100644
--- a/src/backend/commands/typecmds.c
+++ b/src/backend/commands/typecmds.c
@@ -3157,7 +3157,7 @@ validateDomainNotNullConstraint(Oid domainoid)
/* Scan all tuples in this relation */
snapshot = RegisterSnapshot(GetLatestSnapshot());
- scan = table_beginscan(testrel, snapshot, 0, NULL);
+ scan = table_beginscan(testrel, snapshot, 0, NULL, 0);
slot = table_slot_create(testrel, NULL);
while (table_scan_getnextslot(scan, ForwardScanDirection, slot))
{
@@ -3238,7 +3238,7 @@ validateDomainCheckConstraint(Oid domainoid, const char *ccbin, LOCKMODE lockmod
/* Scan all tuples in this relation */
snapshot = RegisterSnapshot(GetLatestSnapshot());
- scan = table_beginscan(testrel, snapshot, 0, NULL);
+ scan = table_beginscan(testrel, snapshot, 0, NULL, 0);
slot = table_slot_create(testrel, NULL);
while (table_scan_getnextslot(scan, ForwardScanDirection, slot))
{
diff --git a/src/backend/executor/execIndexing.c b/src/backend/executor/execIndexing.c
index 0b3a31f1703..74262a34819 100644
--- a/src/backend/executor/execIndexing.c
+++ b/src/backend/executor/execIndexing.c
@@ -816,7 +816,7 @@ check_exclusion_or_unique_constraint(Relation heap, Relation index,
retry:
conflict = false;
found_self = false;
- index_scan = index_beginscan(heap, index, &DirtySnapshot, NULL, indnkeyatts, 0);
+ index_scan = index_beginscan(heap, index, &DirtySnapshot, NULL, indnkeyatts, 0, 0);
index_rescan(index_scan, scankeys, indnkeyatts, NULL, 0);
while (index_getnext_slot(index_scan, ForwardScanDirection, existing_slot))
diff --git a/src/backend/executor/execReplication.c b/src/backend/executor/execReplication.c
index 860f79f9cc1..6e49ea5c5d8 100644
--- a/src/backend/executor/execReplication.c
+++ b/src/backend/executor/execReplication.c
@@ -204,7 +204,7 @@ RelationFindReplTupleByIndex(Relation rel, Oid idxoid,
skey_attoff = build_replindex_scan_key(skey, rel, idxrel, searchslot);
/* Start an index scan. */
- scan = index_beginscan(rel, idxrel, &snap, NULL, skey_attoff, 0);
+ scan = index_beginscan(rel, idxrel, &snap, NULL, skey_attoff, 0, 0);
retry:
found = false;
@@ -382,7 +382,7 @@ RelationFindReplTupleSeq(Relation rel, LockTupleMode lockmode,
/* Start a heap scan. */
InitDirtySnapshot(snap);
- scan = table_beginscan(rel, &snap, 0, NULL);
+ scan = table_beginscan(rel, &snap, 0, NULL, 0);
scanslot = table_slot_create(rel, NULL);
retry:
@@ -601,7 +601,7 @@ RelationFindDeletedTupleInfoSeq(Relation rel, TupleTableSlot *searchslot,
* not yet committed or those just committed prior to the scan are
* excluded in update_most_recent_deletion_info().
*/
- scan = table_beginscan(rel, SnapshotAny, 0, NULL);
+ scan = table_beginscan(rel, SnapshotAny, 0, NULL, 0);
scanslot = table_slot_create(rel, NULL);
table_rescan(scan, NULL);
@@ -665,7 +665,7 @@ RelationFindDeletedTupleInfoByIndex(Relation rel, Oid idxoid,
* not yet committed or those just committed prior to the scan are
* excluded in update_most_recent_deletion_info().
*/
- scan = index_beginscan(rel, idxrel, SnapshotAny, NULL, skey_attoff, 0);
+ scan = index_beginscan(rel, idxrel, SnapshotAny, NULL, skey_attoff, 0, 0);
index_rescan(scan, skey, skey_attoff, NULL, 0);
diff --git a/src/backend/executor/nodeBitmapHeapscan.c b/src/backend/executor/nodeBitmapHeapscan.c
index bf24f3d7fe0..0d854db51a1 100644
--- a/src/backend/executor/nodeBitmapHeapscan.c
+++ b/src/backend/executor/nodeBitmapHeapscan.c
@@ -105,11 +105,18 @@ BitmapTableScanSetup(BitmapHeapScanState *node)
*/
if (!node->ss.ss_currentScanDesc)
{
+ uint32 flags = 0;
+
+ if (!bms_is_member(((Scan *) node->ss.ps.plan)->scanrelid,
+ node->ss.ps.state->es_modified_relids))
+ flags = SO_HINT_REL_READ_ONLY;
+
node->ss.ss_currentScanDesc =
table_beginscan_bm(node->ss.ss_currentRelation,
node->ss.ps.state->es_snapshot,
0,
- NULL);
+ NULL,
+ flags);
}
node->ss.ss_currentScanDesc->st.rs_tbmiterator = tbmiterator;
diff --git a/src/backend/executor/nodeIndexonlyscan.c b/src/backend/executor/nodeIndexonlyscan.c
index 6bea42f128f..2c87ba5f767 100644
--- a/src/backend/executor/nodeIndexonlyscan.c
+++ b/src/backend/executor/nodeIndexonlyscan.c
@@ -94,7 +94,7 @@ IndexOnlyNext(IndexOnlyScanState *node)
estate->es_snapshot,
&node->ioss_Instrument,
node->ioss_NumScanKeys,
- node->ioss_NumOrderByKeys);
+ node->ioss_NumOrderByKeys, 0);
node->ioss_ScanDesc = scandesc;
diff --git a/src/backend/executor/nodeIndexscan.c b/src/backend/executor/nodeIndexscan.c
index 72b135e5dcf..92674441c6d 100644
--- a/src/backend/executor/nodeIndexscan.c
+++ b/src/backend/executor/nodeIndexscan.c
@@ -102,6 +102,12 @@ IndexNext(IndexScanState *node)
if (scandesc == NULL)
{
+ uint32 flags = 0;
+
+ if (!bms_is_member(((Scan *) node->ss.ps.plan)->scanrelid,
+ estate->es_modified_relids))
+ flags = SO_HINT_REL_READ_ONLY;
+
/*
* We reach here if the index scan is not parallel, or if we're
* serially executing an index scan that was planned to be parallel.
@@ -111,7 +117,8 @@ IndexNext(IndexScanState *node)
estate->es_snapshot,
&node->iss_Instrument,
node->iss_NumScanKeys,
- node->iss_NumOrderByKeys);
+ node->iss_NumOrderByKeys,
+ flags);
node->iss_ScanDesc = scandesc;
@@ -207,7 +214,7 @@ IndexNextWithReorder(IndexScanState *node)
estate->es_snapshot,
&node->iss_Instrument,
node->iss_NumScanKeys,
- node->iss_NumOrderByKeys);
+ node->iss_NumOrderByKeys, 0);
node->iss_ScanDesc = scandesc;
diff --git a/src/backend/executor/nodeSeqscan.c b/src/backend/executor/nodeSeqscan.c
index 94047d29430..4d0cbb9dee4 100644
--- a/src/backend/executor/nodeSeqscan.c
+++ b/src/backend/executor/nodeSeqscan.c
@@ -65,13 +65,20 @@ SeqNext(SeqScanState *node)
if (scandesc == NULL)
{
+ uint32 flags = 0;
+
+ if (!bms_is_member(((Scan *) node->ss.ps.plan)->scanrelid,
+ estate->es_modified_relids))
+ flags = SO_HINT_REL_READ_ONLY;
+
/*
* We reach here if the scan is not parallel, or if we're serially
* executing a scan that was planned to be parallel.
*/
scandesc = table_beginscan(node->ss.ss_currentRelation,
estate->es_snapshot,
- 0, NULL);
+ 0, NULL, flags);
+
node->ss.ss_currentScanDesc = scandesc;
}
@@ -367,14 +374,20 @@ ExecSeqScanInitializeDSM(SeqScanState *node,
{
EState *estate = node->ss.ps.state;
ParallelTableScanDesc pscan;
+ uint32 flags = 0;
pscan = shm_toc_allocate(pcxt->toc, node->pscan_len);
table_parallelscan_initialize(node->ss.ss_currentRelation,
pscan,
estate->es_snapshot);
shm_toc_insert(pcxt->toc, node->ss.ps.plan->plan_node_id, pscan);
+ if (!bms_is_member(((Scan *) node->ss.ps.plan)->scanrelid,
+ estate->es_modified_relids))
+ flags = SO_HINT_REL_READ_ONLY;
+
node->ss.ss_currentScanDesc =
- table_beginscan_parallel(node->ss.ss_currentRelation, pscan);
+ table_beginscan_parallel(node->ss.ss_currentRelation, pscan,
+ flags);
}
/* ----------------------------------------------------------------
@@ -404,8 +417,15 @@ ExecSeqScanInitializeWorker(SeqScanState *node,
ParallelWorkerContext *pwcxt)
{
ParallelTableScanDesc pscan;
+ uint32 flags = 0;
+
+ if (!bms_is_member(((Scan *) node->ss.ps.plan)->scanrelid,
+ node->ss.ps.state->es_modified_relids))
+ flags = SO_HINT_REL_READ_ONLY;
pscan = shm_toc_lookup(pwcxt->toc, node->ss.ps.plan->plan_node_id, false);
node->ss.ss_currentScanDesc =
- table_beginscan_parallel(node->ss.ss_currentRelation, pscan);
+ table_beginscan_parallel(node->ss.ss_currentRelation,
+ pscan,
+ flags);
}
diff --git a/src/backend/partitioning/partbounds.c b/src/backend/partitioning/partbounds.c
index 16b0adc172c..91acf1ee2d7 100644
--- a/src/backend/partitioning/partbounds.c
+++ b/src/backend/partitioning/partbounds.c
@@ -3362,7 +3362,7 @@ check_default_partition_contents(Relation parent, Relation default_rel,
econtext = GetPerTupleExprContext(estate);
snapshot = RegisterSnapshot(GetLatestSnapshot());
tupslot = table_slot_create(part_rel, &estate->es_tupleTable);
- scan = table_beginscan(part_rel, snapshot, 0, NULL);
+ scan = table_beginscan(part_rel, snapshot, 0, NULL, 0);
/*
* Switch to per-tuple memory context and reset it for each tuple
diff --git a/src/backend/utils/adt/selfuncs.c b/src/backend/utils/adt/selfuncs.c
index c760b19db55..ec0def0d1e2 100644
--- a/src/backend/utils/adt/selfuncs.c
+++ b/src/backend/utils/adt/selfuncs.c
@@ -7100,7 +7100,7 @@ get_actual_variable_endpoint(Relation heapRel,
index_scan = index_beginscan(heapRel, indexRel,
&SnapshotNonVacuumable, NULL,
- 1, 0);
+ 1, 0, 0);
/* Set it up for index-only scan */
index_scan->xs_want_itup = true;
index_rescan(index_scan, scankeys, 1, NULL, 0);
diff --git a/src/include/access/genam.h b/src/include/access/genam.h
index 9200a22bd9f..d29d9e905fc 100644
--- a/src/include/access/genam.h
+++ b/src/include/access/genam.h
@@ -177,7 +177,7 @@ extern IndexScanDesc index_beginscan(Relation heapRelation,
Relation indexRelation,
Snapshot snapshot,
IndexScanInstrumentation *instrument,
- int nkeys, int norderbys);
+ int nkeys, int norderbys, uint32 flags);
extern IndexScanDesc index_beginscan_bitmap(Relation indexRelation,
Snapshot snapshot,
IndexScanInstrumentation *instrument,
diff --git a/src/include/access/heapam.h b/src/include/access/heapam.h
index 1625b107575..0bfe2366e1a 100644
--- a/src/include/access/heapam.h
+++ b/src/include/access/heapam.h
@@ -119,6 +119,12 @@ typedef struct IndexFetchHeapData
Buffer xs_cbuf; /* current heap buffer in scan, if any */
/* NB: if xs_cbuf is not InvalidBuffer, we hold a pin on that buffer */
+
+ /*
+ * Some optimizations can only be performed if the query does not modify
+ * the underlying relation. Track that here.
+ */
+ bool modifies_base_rel;
} IndexFetchHeapData;
/* Result codes for HeapTupleSatisfiesVacuum */
diff --git a/src/include/access/tableam.h b/src/include/access/tableam.h
index 2fa790b6bf5..d10b1b03cdb 100644
--- a/src/include/access/tableam.h
+++ b/src/include/access/tableam.h
@@ -63,6 +63,8 @@ typedef enum ScanOptions
/* unregister snapshot at scan end? */
SO_TEMP_SNAPSHOT = 1 << 9,
+ /* set if the query doesn't modify the rel */
+ SO_HINT_REL_READ_ONLY = 1 << 10,
} ScanOptions;
/*
@@ -420,7 +422,7 @@ typedef struct TableAmRoutine
*
* Tuples for an index scan can then be fetched via index_fetch_tuple.
*/
- struct IndexFetchTableData *(*index_fetch_begin) (Relation rel);
+ struct IndexFetchTableData *(*index_fetch_begin) (Relation rel, uint32 flags);
/*
* Reset index fetch. Typically this will release cross index fetch
@@ -874,9 +876,9 @@ extern TupleTableSlot *table_slot_create(Relation relation, List **reglist);
*/
static inline TableScanDesc
table_beginscan(Relation rel, Snapshot snapshot,
- int nkeys, ScanKeyData *key)
+ int nkeys, ScanKeyData *key, uint32 flags)
{
- uint32 flags = SO_TYPE_SEQSCAN |
+ 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);
@@ -919,9 +921,9 @@ table_beginscan_strat(Relation rel, Snapshot snapshot,
*/
static inline TableScanDesc
table_beginscan_bm(Relation rel, Snapshot snapshot,
- int nkeys, ScanKeyData *key)
+ int nkeys, ScanKeyData *key, uint32 flags)
{
- uint32 flags = SO_TYPE_BITMAPSCAN | SO_ALLOW_PAGEMODE;
+ flags |= SO_TYPE_BITMAPSCAN | SO_ALLOW_PAGEMODE;
return rel->rd_tableam->scan_begin(rel, snapshot, nkeys, key,
NULL, flags);
@@ -1128,7 +1130,8 @@ extern void table_parallelscan_initialize(Relation rel,
* Caller must hold a suitable lock on the relation.
*/
extern TableScanDesc table_beginscan_parallel(Relation relation,
- ParallelTableScanDesc pscan);
+ ParallelTableScanDesc pscan,
+ uint32 flags);
/*
* Begin a parallel tid range scan. `pscan` needs to have been initialized
@@ -1164,9 +1167,9 @@ table_parallelscan_reinitialize(Relation rel, ParallelTableScanDesc pscan)
* Tuples for an index scan can then be fetched via table_index_fetch_tuple().
*/
static inline IndexFetchTableData *
-table_index_fetch_begin(Relation rel)
+table_index_fetch_begin(Relation rel, uint32 flags)
{
- return rel->rd_tableam->index_fetch_begin(rel);
+ return rel->rd_tableam->index_fetch_begin(rel, flags);
}
/*
--
2.43.0
[text/x-patch] v28-0014-Allow-on-access-pruning-to-set-pages-all-visible.patch (11.0K, 15-v28-0014-Allow-on-access-pruning-to-set-pages-all-visible.patch)
download | inline diff:
From 28ccc95ff7620ef07aa3f225d3a7c05aa8c05909 Mon Sep 17 00:00:00 2001
From: Melanie Plageman <[email protected]>
Date: Wed, 3 Dec 2025 15:24:08 -0500
Subject: [PATCH v28 14/15] Allow on-access pruning to set pages all-visible
Many queries do not modify the underlying relation. For such queries, if
on-access pruning occurs during the scan, we can check whether the page
has become all-visible and update the visibility map accordingly.
Previously, only vacuum and COPY FREEZE marked pages as all-visible or
all-frozen.
This commit implements on-access VM setting for sequential scans as well
as for the underlying heap relation in index scans and bitmap heap
scans.
Author: Melanie Plageman <[email protected]>
Reviewed-by: Andres Freund <[email protected]>
Reviewed-by: Kirill Reshke <[email protected]>
Discussion: https://postgr.es/m/flat/CAAKRu_ZMw6Npd_qm2KM%2BFwQ3cMOMx1Dh3VMhp8-V7SOLxdK9-g%40mail.gmail.com
---
src/backend/access/heap/heapam.c | 15 ++++++-
src/backend/access/heap/heapam_handler.c | 15 ++++++-
src/backend/access/heap/pruneheap.c | 40 ++++++++++++++++++-
src/include/access/heapam.h | 24 +++++++++--
.../t/035_standby_logical_decoding.pl | 3 +-
5 files changed, 89 insertions(+), 8 deletions(-)
diff --git a/src/backend/access/heap/heapam.c b/src/backend/access/heap/heapam.c
index fb7a7548aa0..d9dc79f4a96 100644
--- a/src/backend/access/heap/heapam.c
+++ b/src/backend/access/heap/heapam.c
@@ -570,6 +570,7 @@ heap_prepare_pagescan(TableScanDesc sscan)
Buffer buffer = scan->rs_cbuf;
BlockNumber block = scan->rs_cblock;
Snapshot snapshot;
+ Buffer *vmbuffer = NULL;
Page page;
int lines;
bool all_visible;
@@ -584,7 +585,9 @@ heap_prepare_pagescan(TableScanDesc sscan)
/*
* Prune and repair fragmentation for the whole page, if possible.
*/
- heap_page_prune_opt(scan->rs_base.rs_rd, buffer);
+ if (sscan->rs_flags & SO_HINT_REL_READ_ONLY)
+ vmbuffer = &scan->rs_vmbuffer;
+ heap_page_prune_opt(scan->rs_base.rs_rd, buffer, vmbuffer);
/*
* We must hold share lock on the buffer content while examining tuple
@@ -1261,6 +1264,7 @@ heap_beginscan(Relation relation, Snapshot snapshot,
sizeof(TBMIterateResult));
}
+ scan->rs_vmbuffer = InvalidBuffer;
return (TableScanDesc) scan;
}
@@ -1299,6 +1303,12 @@ heap_rescan(TableScanDesc sscan, ScanKey key, bool set_params,
scan->rs_cbuf = InvalidBuffer;
}
+ if (BufferIsValid(scan->rs_vmbuffer))
+ {
+ ReleaseBuffer(scan->rs_vmbuffer);
+ scan->rs_vmbuffer = InvalidBuffer;
+ }
+
/*
* SO_TYPE_BITMAPSCAN would be cleaned up here, but it does not hold any
* additional data vs a normal HeapScan
@@ -1331,6 +1341,9 @@ heap_endscan(TableScanDesc sscan)
if (BufferIsValid(scan->rs_cbuf))
ReleaseBuffer(scan->rs_cbuf);
+ if (BufferIsValid(scan->rs_vmbuffer))
+ ReleaseBuffer(scan->rs_vmbuffer);
+
/*
* Must free the read stream before freeing the BufferAccessStrategy.
*/
diff --git a/src/backend/access/heap/heapam_handler.c b/src/backend/access/heap/heapam_handler.c
index 6c2e4e08b16..2cb98e58956 100644
--- a/src/backend/access/heap/heapam_handler.c
+++ b/src/backend/access/heap/heapam_handler.c
@@ -85,6 +85,7 @@ heapam_index_fetch_begin(Relation rel, uint32 flags)
hscan->xs_base.rel = rel;
hscan->xs_cbuf = InvalidBuffer;
+ hscan->xs_vmbuffer = InvalidBuffer;
hscan->modifies_base_rel = !(flags & SO_HINT_REL_READ_ONLY);
return &hscan->xs_base;
@@ -100,6 +101,12 @@ heapam_index_fetch_reset(IndexFetchTableData *scan)
ReleaseBuffer(hscan->xs_cbuf);
hscan->xs_cbuf = InvalidBuffer;
}
+
+ if (BufferIsValid(hscan->xs_vmbuffer))
+ {
+ ReleaseBuffer(hscan->xs_vmbuffer);
+ hscan->xs_vmbuffer = InvalidBuffer;
+ }
}
static void
@@ -139,7 +146,8 @@ heapam_index_fetch_tuple(struct IndexFetchTableData *scan,
* Prune page, but only if we weren't already on this page
*/
if (prev_buf != hscan->xs_cbuf)
- heap_page_prune_opt(hscan->xs_base.rel, hscan->xs_cbuf);
+ heap_page_prune_opt(hscan->xs_base.rel, hscan->xs_cbuf,
+ hscan->modifies_base_rel ? NULL : &hscan->xs_vmbuffer);
}
/* Obtain share-lock on the buffer so we can examine visibility */
@@ -2472,6 +2480,7 @@ BitmapHeapScanNextBlock(TableScanDesc scan,
TBMIterateResult *tbmres;
OffsetNumber offsets[TBM_MAX_TUPLES_PER_PAGE];
int noffsets = -1;
+ Buffer *vmbuffer = NULL;
Assert(scan->rs_flags & SO_TYPE_BITMAPSCAN);
Assert(hscan->rs_read_stream);
@@ -2518,7 +2527,9 @@ BitmapHeapScanNextBlock(TableScanDesc scan,
/*
* Prune and repair fragmentation for the whole page, if possible.
*/
- heap_page_prune_opt(scan->rs_rd, buffer);
+ if (scan->rs_flags & SO_HINT_REL_READ_ONLY)
+ vmbuffer = &hscan->rs_vmbuffer;
+ heap_page_prune_opt(scan->rs_rd, buffer, vmbuffer);
/*
* We must hold share lock on the buffer content while examining tuple
diff --git a/src/backend/access/heap/pruneheap.c b/src/backend/access/heap/pruneheap.c
index 1fa72e19f0d..c9821a5830c 100644
--- a/src/backend/access/heap/pruneheap.c
+++ b/src/backend/access/heap/pruneheap.c
@@ -202,6 +202,8 @@ static bool heap_page_will_set_vm(PruneState *prstate,
Relation relation,
BlockNumber heap_blk, Buffer heap_buffer, Page heap_page,
Buffer vmbuffer,
+ PruneReason reason,
+ bool do_prune, bool do_freeze,
int nlpdead_items,
uint8 *old_vmbits,
uint8 *new_vmbits);
@@ -223,9 +225,13 @@ static TransactionId get_conflict_xid(bool do_prune, bool do_freeze, bool do_set
* if there's not any use in pruning.
*
* Caller must have pin on the buffer, and must *not* have a lock on it.
+ *
+ * If vmbuffer is not NULL, it is okay for pruning to set the visibility map if
+ * the page is all-visible. We will take care of pinning and, if needed,
+ * reading in the page of the visibility map.
*/
void
-heap_page_prune_opt(Relation relation, Buffer buffer)
+heap_page_prune_opt(Relation relation, Buffer buffer, Buffer *vmbuffer)
{
Page page = BufferGetPage(buffer);
TransactionId prune_xid;
@@ -306,6 +312,13 @@ heap_page_prune_opt(Relation relation, Buffer buffer)
.cutoffs = NULL,
};
+ if (vmbuffer)
+ {
+ visibilitymap_pin(relation, BufferGetBlockNumber(buffer), vmbuffer);
+ params.options |= HEAP_PAGE_PRUNE_UPDATE_VM;
+ params.vmbuffer = *vmbuffer;
+ }
+
heap_page_prune_and_freeze(¶ms, &presult, &dummy_off_loc,
NULL, NULL);
@@ -945,6 +958,9 @@ identify_and_fix_vm_corruption(Relation rel, Buffer heap_buffer,
* corrupted, it will fix them by clearing the VM bits and visibility hint.
* This does not need to be done in a critical section.
*
+ * This should be called only after do_freeze has been decided (and do_prune
+ * has been set), as these factor into our heuristic-based decision.
+ *
* Returns true if one or both VM bits should be set, along with returning the
* current value of the VM bits in *old_vmbits and the desired new value of
* the VM bits in *new_vmbits.
@@ -954,6 +970,8 @@ heap_page_will_set_vm(PruneState *prstate,
Relation relation,
BlockNumber heap_blk, Buffer heap_buffer, Page heap_page,
Buffer vmbuffer,
+ PruneReason reason,
+ bool do_prune, bool do_freeze,
int nlpdead_items,
uint8 *old_vmbits,
uint8 *new_vmbits)
@@ -961,6 +979,24 @@ heap_page_will_set_vm(PruneState *prstate,
if (!prstate->attempt_update_vm)
return false;
+ /*
+ * If this is an on-access call and we're not actually pruning, avoid
+ * setting the visibility map if it would newly dirty the heap page or, if
+ * the page is already dirty, if doing so would require including a
+ * full-page image (FPI) of the heap page in the WAL. This situation
+ * should be rare, as on-access pruning is only attempted when
+ * pd_prune_xid is valid.
+ */
+ if (reason == PRUNE_ON_ACCESS &&
+ prstate->all_visible &&
+ !do_prune && !do_freeze &&
+ (!BufferIsDirty(heap_buffer) || XLogCheckBufferNeedsBackup(heap_buffer)))
+ {
+ prstate->all_visible = false;
+ prstate->all_frozen = false;
+ return false;
+ }
+
*old_vmbits = visibilitymap_get_status(relation, heap_blk,
&vmbuffer);
@@ -1157,6 +1193,8 @@ heap_page_prune_and_freeze(PruneFreezeParams *params,
buffer,
page,
vmbuffer,
+ params->reason,
+ do_prune, do_freeze,
prstate.lpdead_items,
&old_vmbits,
&new_vmbits);
diff --git a/src/include/access/heapam.h b/src/include/access/heapam.h
index 0bfe2366e1a..3328f56c101 100644
--- a/src/include/access/heapam.h
+++ b/src/include/access/heapam.h
@@ -95,6 +95,13 @@ typedef struct HeapScanDescData
*/
ParallelBlockTableScanWorkerData *rs_parallelworkerdata;
+ /*
+ * For sequential scans and bitmap heap scans. If the relation is not
+ * being modified, on-access pruning may read in the current heap page's
+ * corresponding VM block to this buffer.
+ */
+ Buffer rs_vmbuffer;
+
/* these fields only used in page-at-a-time mode and for bitmap scans */
uint32 rs_cindex; /* current tuple's index in vistuples */
uint32 rs_ntuples; /* number of visible tuples on page */
@@ -117,8 +124,18 @@ typedef struct IndexFetchHeapData
{
IndexFetchTableData xs_base; /* AM independent part of the descriptor */
- Buffer xs_cbuf; /* current heap buffer in scan, if any */
- /* NB: if xs_cbuf is not InvalidBuffer, we hold a pin on that buffer */
+ /*
+ * Current heap buffer in scan, if any. NB: if xs_cbuf is not
+ * InvalidBuffer, we hold a pin on that buffer.
+ */
+ Buffer xs_cbuf;
+
+ /*
+ * For index scans that do not modify the underlying heap table, on-access
+ * pruning may read in the current heap page's corresponding VM block to
+ * this buffer.
+ */
+ Buffer xs_vmbuffer;
/*
* Some optimizations can only be performed if the query does not modify
@@ -420,7 +437,8 @@ extern TransactionId heap_index_delete_tuples(Relation rel,
TM_IndexDeleteOp *delstate);
/* in heap/pruneheap.c */
-extern void heap_page_prune_opt(Relation relation, Buffer buffer);
+extern void heap_page_prune_opt(Relation relation, Buffer buffer,
+ Buffer *vmbuffer);
extern void heap_page_prune_and_freeze(PruneFreezeParams *params,
PruneFreezeResult *presult,
OffsetNumber *off_loc,
diff --git a/src/test/recovery/t/035_standby_logical_decoding.pl b/src/test/recovery/t/035_standby_logical_decoding.pl
index ebe2fae1789..bdd9f0a62cd 100644
--- a/src/test/recovery/t/035_standby_logical_decoding.pl
+++ b/src/test/recovery/t/035_standby_logical_decoding.pl
@@ -296,6 +296,7 @@ wal_level = 'logical'
max_replication_slots = 4
max_wal_senders = 4
autovacuum = off
+hot_standby_feedback = on
});
$node_primary->dump_info;
$node_primary->start;
@@ -748,7 +749,7 @@ check_pg_recvlogical_stderr($handle,
$logstart = -s $node_standby->logfile;
reactive_slots_change_hfs_and_wait_for_xmins('shared_row_removal_',
- 'no_conflict_', 0, 1);
+ 'no_conflict_', 1, 0);
# This should not trigger a conflict
wait_until_vacuum_can_remove(
--
2.43.0
[text/x-patch] v28-0015-Set-pd_prune_xid-on-insert.patch (6.7K, 16-v28-0015-Set-pd_prune_xid-on-insert.patch)
download | inline diff:
From 1df98a32c9b32458621c003f895a71c008c085d6 Mon Sep 17 00:00:00 2001
From: Melanie Plageman <[email protected]>
Date: Tue, 29 Jul 2025 16:12:56 -0400
Subject: [PATCH v28 15/15] Set pd_prune_xid on insert
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
Now that visibility map (VM) updates can occur during read-only queries,
it makes sense to also set the page’s pd_prune_xid hint during inserts.
This enables heap_page_prune_and_freeze() to run after a page is
filled with newly inserted tuples the first time it is read.
This change also addresses a long-standing note in heap_insert() and
heap_multi_insert(), which observed that setting pd_prune_xid would
help clean up aborted insertions sooner. Without it, such tuples might
linger until VACUUM, whereas now they can be pruned earlier.
Setting pd_prune_xid on insert can cause a page to be dirtied and
written out when it previously would not have been, affetcting the
reported number of hits in the index-killtuples isolation test. It is
unclear if this is a bug in the way hits are tracked, a faulty test
expectation, or if simply updating the test's expected output is
sufficient remediation.
---
src/backend/access/heap/heapam.c | 25 +++++++++++++------
src/backend/access/heap/heapam_xlog.c | 15 ++++++++++-
.../modules/index/expected/killtuples.out | 6 ++---
3 files changed, 34 insertions(+), 12 deletions(-)
diff --git a/src/backend/access/heap/heapam.c b/src/backend/access/heap/heapam.c
index d9dc79f4a96..ccebc1f244b 100644
--- a/src/backend/access/heap/heapam.c
+++ b/src/backend/access/heap/heapam.c
@@ -2119,6 +2119,7 @@ heap_insert(Relation relation, HeapTuple tup, CommandId cid,
TransactionId xid = GetCurrentTransactionId();
HeapTuple heaptup;
Buffer buffer;
+ Page page;
Buffer vmbuffer = InvalidBuffer;
bool all_visible_cleared = false;
@@ -2178,15 +2179,19 @@ heap_insert(Relation relation, HeapTuple tup, CommandId cid,
}
/*
- * XXX Should we set PageSetPrunable on this page ?
+ * Set pd_prune_xid to trigger heap_page_prune_and_freeze() once the page
+ * is full so that we can set the page all-visible in the VM.
*
- * The inserting transaction may eventually abort thus making this tuple
- * DEAD and hence available for pruning. Though we don't want to optimize
- * for aborts, if no other tuple in this page is UPDATEd/DELETEd, the
- * aborted tuple will never be pruned until next vacuum is triggered.
+ * Setting pd_prune_xid is also handy if the inserting transaction
+ * eventually aborts making this tuple DEAD and hence available for
+ * pruning. If no other tuple in this page is UPDATEd/DELETEd, the aborted
+ * tuple would never otherwise be pruned until next vacuum is triggered.
*
- * If you do add PageSetPrunable here, add it in heap_xlog_insert too.
+ * Don't set it if we are in bootstrap mode, though.
*/
+ page = BufferGetPage(buffer);
+ if (TransactionIdIsNormal(xid))
+ PageSetPrunable(page, xid);
MarkBufferDirty(buffer);
@@ -2196,7 +2201,6 @@ heap_insert(Relation relation, HeapTuple tup, CommandId cid,
xl_heap_insert xlrec;
xl_heap_header xlhdr;
XLogRecPtr recptr;
- Page page = BufferGetPage(buffer);
uint8 info = XLOG_HEAP_INSERT;
int bufflags = 0;
@@ -2560,8 +2564,13 @@ heap_multi_insert(Relation relation, TupleTableSlot **slots, int ntuples,
}
/*
- * XXX Should we set PageSetPrunable on this page ? See heap_insert()
+ * Set pd_prune_xid. See heap_insert() for more on why we do this when
+ * inserting tuples. This only makes sense if we aren't already
+ * setting the page frozen in the VM. We also don't set it in
+ * bootstrap mode.
*/
+ if (!all_frozen_set && TransactionIdIsNormal(xid))
+ PageSetPrunable(page, xid);
MarkBufferDirty(buffer);
diff --git a/src/backend/access/heap/heapam_xlog.c b/src/backend/access/heap/heapam_xlog.c
index 47d2479415e..ab2db931aac 100644
--- a/src/backend/access/heap/heapam_xlog.c
+++ b/src/backend/access/heap/heapam_xlog.c
@@ -447,6 +447,12 @@ heap_xlog_insert(XLogReaderState *record)
freespace = PageGetHeapFreeSpace(page); /* needed to update FSM below */
+ /*
+ * Set the page prunable to trigger on-access pruning later which may
+ * set the page all-visible in the VM.
+ */
+ PageSetPrunable(page, XLogRecGetXid(record));
+
PageSetLSN(page, lsn);
if (xlrec->flags & XLH_INSERT_ALL_VISIBLE_CLEARED)
@@ -596,9 +602,16 @@ heap_xlog_multi_insert(XLogReaderState *record)
if (xlrec->flags & XLH_INSERT_ALL_VISIBLE_CLEARED)
PageClearAllVisible(page);
- /* XLH_INSERT_ALL_FROZEN_SET implies that all tuples are visible */
+ /*
+ * XLH_INSERT_ALL_FROZEN_SET implies that all tuples are visible. If
+ * we are not setting the page frozen, then set the page's prunable
+ * hint so that we trigger on-access pruning later which may set the
+ * page all-visible in the VM.
+ */
if (xlrec->flags & XLH_INSERT_ALL_FROZEN_SET)
PageSetAllVisible(page);
+ else
+ PageSetPrunable(page, XLogRecGetXid(record));
MarkBufferDirty(buffer);
}
diff --git a/src/test/modules/index/expected/killtuples.out b/src/test/modules/index/expected/killtuples.out
index be7ddd756ef..b29f2434b00 100644
--- a/src/test/modules/index/expected/killtuples.out
+++ b/src/test/modules/index/expected/killtuples.out
@@ -54,7 +54,7 @@ step flush: SELECT FROM pg_stat_force_next_flush();
step result: SELECT heap_blks_read + heap_blks_hit - counter.heap_accesses AS new_heap_accesses FROM counter, pg_statio_all_tables WHERE relname = 'kill_prior_tuple';
new_heap_accesses
-----------------
- 1
+ 2
(1 row)
step measure: UPDATE counter SET heap_accesses = (SELECT heap_blks_read + heap_blks_hit FROM pg_statio_all_tables WHERE relname = 'kill_prior_tuple');
@@ -130,7 +130,7 @@ step flush: SELECT FROM pg_stat_force_next_flush();
step result: SELECT heap_blks_read + heap_blks_hit - counter.heap_accesses AS new_heap_accesses FROM counter, pg_statio_all_tables WHERE relname = 'kill_prior_tuple';
new_heap_accesses
-----------------
- 1
+ 2
(1 row)
step measure: UPDATE counter SET heap_accesses = (SELECT heap_blks_read + heap_blks_hit FROM pg_statio_all_tables WHERE relname = 'kill_prior_tuple');
@@ -283,7 +283,7 @@ step flush: SELECT FROM pg_stat_force_next_flush();
step result: SELECT heap_blks_read + heap_blks_hit - counter.heap_accesses AS new_heap_accesses FROM counter, pg_statio_all_tables WHERE relname = 'kill_prior_tuple';
new_heap_accesses
-----------------
- 1
+ 2
(1 row)
step measure: UPDATE counter SET heap_accesses = (SELECT heap_blks_read + heap_blks_hit FROM pg_statio_all_tables WHERE relname = 'kill_prior_tuple');
--
2.43.0
view thread (143+ messages) latest in thread
reply
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Reply to all the recipients using the --to and --cc options:
reply via email
To: [email protected]
Cc: [email protected], [email protected], [email protected], [email protected], [email protected], [email protected], [email protected], [email protected]
Subject: Re: eliminate xl_heap_visible to reduce WAL (and eventually set VM on-access)
In-Reply-To: <CAAKRu_Yt6EH5aFSJBm-k7PrNM4bTt56fTRbyU7gqYXe4cW+F9g@mail.gmail.com>
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
This inbox is served by agora; see mirroring instructions
for how to clone and mirror all data and code used for this inbox