diff options
| author | Thomas Munro <tmunro@postgresql.org> | 2023-01-26 14:50:07 +1300 |
|---|---|---|
| committer | Thomas Munro <tmunro@postgresql.org> | 2023-01-26 14:54:17 +1300 |
| commit | 2f65b84683b7e755b4c44dc949f567e6d5d50132 (patch) | |
| tree | a2164c231d9c09eb3a85908384fb71b9da3aa3ae | |
| parent | e9774151c27868ebec08259f3698ebdb321138c4 (diff) | |
Fix rare sharedtuplestore.c corruption.
If the final chunk of an oversized tuple being written out to disk was
exactly 32760 bytes, it would be corrupted due to a fencepost bug.
Bug #17619. Back-patch to 11 where the code arrived.
While testing that (see test module in archives), I (tmunro) noticed
that the per-participant page counter was not initialized to zero as it
should have been; that wasn't a live bug when it was written since DSM
memory was originally always zeroed, but since 14
min_dynamic_shared_memory might be configured and it supplies non-zeroed
memory, so that is also fixed here.
Author: Dmitry Astapov <dastapov@gmail.com>
Discussion: https://postgr.es/m/17619-0de62ceda812b8b5%40postgresql.org
| -rw-r--r-- | src/backend/utils/sort/sharedtuplestore.c | 5 |
1 files changed, 3 insertions, 2 deletions
diff --git a/src/backend/utils/sort/sharedtuplestore.c b/src/backend/utils/sort/sharedtuplestore.c index 57e35db4f8d..fd29ce9609e 100644 --- a/src/backend/utils/sort/sharedtuplestore.c +++ b/src/backend/utils/sort/sharedtuplestore.c @@ -158,6 +158,7 @@ sts_initialize(SharedTuplestore *sts, int participants, LWLockInitialize(&sts->participants[i].lock, LWTRANCHE_SHARED_TUPLESTORE); sts->participants[i].read_page = 0; + sts->participants[i].npages = 0; sts->participants[i].writing = false; } @@ -319,7 +320,7 @@ sts_puttuple(SharedTuplestoreAccessor *accessor, void *meta_data, /* Do we have space? */ size = accessor->sts->meta_data_size + tuple->t_len; - if (accessor->write_pointer + size >= accessor->write_end) + if (accessor->write_pointer + size > accessor->write_end) { if (accessor->write_chunk == NULL) { @@ -339,7 +340,7 @@ sts_puttuple(SharedTuplestoreAccessor *accessor, void *meta_data, } /* It may still not be enough in the case of a gigantic tuple. */ - if (accessor->write_pointer + size >= accessor->write_end) + if (accessor->write_pointer + size > accessor->write_end) { size_t written; |