On Mon, Sep 26, 2022 at 04:50:36PM +0900, Michael Paquier wrote:
> You are right. bool is not usually a problem in a signal handler, but
> sig_atomic_t is the type we ought to use. I'll go adjust that.
Done this one. I have scanned the code, but did not notice a similar
mistake. It is worth noting that we have only one remaining "volatile
bool" in the headers now.
--
Michael
From:
Michael Paquier Date: Subject:
Re: Refactor backup related code (was: Is it correct to say, "invalid data in file \"%s\"", BACKUP_LABEL_FILE in do_pg_backup_stop?)