Fix WaitEventSet resource leak in WaitLatchOrSocket().
authorEtsuro Fujita <[email protected]>
Thu, 11 Apr 2024 10:05:07 +0000 (19:05 +0900)
committerEtsuro Fujita <[email protected]>
Thu, 11 Apr 2024 10:05:07 +0000 (19:05 +0900)
commit0341d4b10e786dcbc4c63da9d21af842ab31e118
tree867462b18f3bfa3ec192434f4913aef5728f299c
parent5e9d8bed0094b146d64013d417c51eb0f69aeafd
Fix WaitEventSet resource leak in WaitLatchOrSocket().

This function would have the same issue we solved in commit 501cfd07d:
If an error is thrown after calling CreateWaitEventSet(), the file
descriptor (on epoll- or kqueue-based systems) or handles (on Windows)
that the WaitEventSet contains are leaked.

Like that commit, use PG_TRY-PG_FINALLY (PG_TRY-PG_CATCH in v12) to make
sure the WaitEventSet is freed properly.

Back-patch to all supported versions, but as we do not have this issue
in HEAD (cf. commit 50c67c201), no need to apply this patch to it.

Discussion: https://postgr.es/m/CAPmGK16MqdDoD8oatp8SQWaEa4vS3nfQqDN_Sj9YRuu5J3Lj9g%40mail.gmail.com
src/backend/storage/ipc/latch.c