diff options
| author | Tom Lane <tgl@sss.pgh.pa.us> | 2025-01-22 11:58:20 -0500 |
|---|---|---|
| committer | Tom Lane <tgl@sss.pgh.pa.us> | 2025-01-22 11:58:20 -0500 |
| commit | 001c09d8b72ee67ea670a1e9a534fa96948829dd (patch) | |
| tree | 37040417d33d021479cb457a69e9a9fbf297e88b /contrib/lo/sql | |
| parent | 6397819c8af5018663b665dabe2d5bbba56fae9e (diff) | |
Repair incorrect handling of AfterTriggerSharedData.ats_modifiedcols.
This patch fixes two distinct errors that both ultimately trace
to commit 71d60e2aa, which added the ats_modifiedcols field.
The more severe error is that ats_modifiedcols wasn't accounted for
in afterTriggerAddEvent's scanning loop that looks for a pre-existing
duplicate AfterTriggerSharedData. Thus, a new event could be
incorrectly matched to an AfterTriggerSharedData that has a different
value of ats_modifiedcols, resulting in the wrong tg_updatedcols
bitmap getting passed to the trigger whenever it finally gets fired.
We'd not noticed because (a) few triggers consult tg_updatedcols,
and (b) we had no tests exercising a case where such a trigger was
called as an AFTER trigger. In the test case added by this commit,
contrib/lo's trigger fails to remove a large object when expected
because (without this fix) it thinks the LO OID column hasn't changed.
The other problem was introduced by commit ce5aaea8c, which copied the
modified-columns bitmap into trigger-related storage. It made a copy
for every trigger event, whereas what we really want is to make a new
copy only when we make a new AfterTriggerSharedData entry. (We could
imagine adding extra logic to reduce the number of bitmap copies still
more, but it doesn't look worthwhile at the moment.) In a simple test
of an UPDATE of 10000000 rows with a single AFTER trigger, this thinko
roughly tripled the amount of memory consumed by the pending-triggers
data structures, from 160446744 to 480443440 bytes.
Fixing the first problem requires introducing a bms_equal() call into
afterTriggerAddEvent's scanning loop, which is slightly annoying from
a speed perspective. However, getting rid of the excessive bms_copy()
calls from the second problem balances that out; overall speed of
trigger operations is the same or slightly better, in my tests.
Discussion: https://postgr.es/m/3496294.1737501591@sss.pgh.pa.us
Backpatch-through: 13
Diffstat (limited to 'contrib/lo/sql')
| -rw-r--r-- | contrib/lo/sql/lo.sql | 40 |
1 files changed, 40 insertions, 0 deletions
diff --git a/contrib/lo/sql/lo.sql b/contrib/lo/sql/lo.sql index 77039509245..99c3bd1fab3 100644 --- a/contrib/lo/sql/lo.sql +++ b/contrib/lo/sql/lo.sql @@ -27,4 +27,44 @@ DELETE FROM image; SELECT lo_get(43214); +-- Now let's try it with an AFTER trigger + +DROP TRIGGER t_raster ON image; + +CREATE CONSTRAINT TRIGGER t_raster AFTER UPDATE OR DELETE ON image + DEFERRABLE INITIALLY DEFERRED + FOR EACH ROW EXECUTE PROCEDURE lo_manage(raster); + +SELECT lo_create(43223); +SELECT lo_create(43224); +SELECT lo_create(43225); + +INSERT INTO image (title, raster) VALUES ('beautiful image', 43223); + +SELECT lo_get(43223); + +UPDATE image SET raster = 43224 WHERE title = 'beautiful image'; + +SELECT lo_get(43223); -- gone +SELECT lo_get(43224); + +-- test updating of unrelated column +UPDATE image SET title = 'beautiful picture' WHERE title = 'beautiful image'; + +SELECT lo_get(43224); + +-- this case used to be buggy +BEGIN; +UPDATE image SET title = 'beautiful image' WHERE title = 'beautiful picture'; +UPDATE image SET raster = 43225 WHERE title = 'beautiful image'; +SELECT lo_get(43224); +COMMIT; + +SELECT lo_get(43224); -- gone +SELECT lo_get(43225); + +DELETE FROM image; + +SELECT lo_get(43225); -- gone + DROP TABLE image; |
