database/postgres
1
0
Fork 0
mirror of https://github.com/postgres/postgres.git synced 2025-07-28 23:42:10 +03:00
Code Activity

Fix core dump in ReorderBufferRestoreChange on alignment-picky platforms.

Browse Source 
When re-reading an update involving both an old tuple and a new tuple from
disk, reorderbuffer.c was careless about whether the new tuple is suitably
aligned for direct access --- in general, it isn't.  We'd missed seeing
this in the buildfarm because the contrib/test_decoding tests exercise this
code path only a few times, and by chance all of those cases have old
tuples with length a multiple of 4, which is usually enough to make the
access to the new tuple's t_len safe.  For some still-not-entirely-clear
reason, however, Debian's sparc build gets a bus error, as reported by
Christoph Berg; perhaps it's assuming 8-byte alignment of the pointer?

The lack of previous field reports is probably because you need all of
these conditions to trigger a crash: an alignment-picky platform (not
Intel), a transaction large enough to spill to disk, an update within
that xact that changes a primary-key field and has an odd-length old tuple,
and of course logical decoding tracing the transaction.

Avoid the alignment assumption by using memcpy instead of fetching t_len
directly, and add a test case that exposes the crash on picky platforms.
Back-patch to 9.4 where the bug was introduced.

Discussion: <20160413094117.GC21485@msg.credativ.de>
This commit is contained in:
Tom Lane
2016-04-14 19:42:21 -04:00
parent c2dc194bdb
commit 6a3d3965d6
3 changed files with 23 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files

11
contrib/test_decoding/expected/ddl.out
View File

@ -233,6 +233,8 @@ SELECT 'tx logical msg' FROM pg_logical_emit_message(true, 'test', 'tx logical m
DELETE FROM tr_etoomuch WHERE id < 5000;
UPDATE tr_etoomuch SET data = - data WHERE id > 5000;
CREATE TABLE tr_oddlength (id text primary key, data text);
INSERT INTO tr_oddlength VALUES('ab', 'foo');
COMMIT;
/* display results, but hide most of the output */
SELECT count(*), min(data), max(data)
@ -244,13 +246,16 @@ ORDER BY 1,2;
1 | BEGIN | BEGIN
1 | COMMIT | COMMIT
1 | message: transactional: 1 prefix: test, sz: 14 content:tx logical msg | message: transactional: 1 prefix: test, sz: 14 content:tx logical msg
1 | table public.tr_oddlength: INSERT: id[text]:'ab' data[text]:'foo' | table public.tr_oddlength: INSERT: id[text]:'ab' data[text]:'foo'
20467 | table public.tr_etoomuch: DELETE: id[integer]:1 | table public.tr_etoomuch: UPDATE: id[integer]:9999 data[integer]:-9999
(4 rows)
(5 rows)
-- check updates of primary keys work correctly
BEGIN;
CREATE TABLE spoolme AS SELECT g.i FROM generate_series(1, 5000) g(i);
UPDATE tr_etoomuch SET id = -id WHERE id = 5000;
UPDATE tr_oddlength SET id = 'x', data = 'quux';
UPDATE tr_oddlength SET id = 'yy', data = 'a';
DELETE FROM spoolme;
DROP TABLE spoolme;
COMMIT;
@ -260,7 +265,9 @@ WHERE data ~ 'UPDATE';
data
-------------------------------------------------------------------------------------------------------------
table public.tr_etoomuch: UPDATE: old-key: id[integer]:5000 new-tuple: id[integer]:-5000 data[integer]:5000
(1 row)
table public.tr_oddlength: UPDATE: old-key: id[text]:'ab' new-tuple: id[text]:'x' data[text]:'quux'
table public.tr_oddlength: UPDATE: old-key: id[text]:'x' new-tuple: id[text]:'yy' data[text]:'a'
(3 rows)
-- check that a large, spooled, upsert works
INSERT INTO tr_etoomuch (id, data)