From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
---|---|
To: | pgsql-hackers(at)postgreSQL(dot)org |
Subject: | pgdump/parallel.c: "aborting" flag is dead code |
Date: | 2016-05-28 19:06:58 |
Message-ID: | 15583.1464462418@sss.pgh.pa.us |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
parallel.c has an "aborting" flag that's entirely useless, as it's
set only during archive_close_connection(), and tested only in places
that are not reached after that. I've confirmed this both by code
reading and by testing. It appears from some of the comments that
there was once an intent to shut down workers by sending an explicit
TERMINATE message, rather than just closing the pipe, and in that
case maybe there would have been some value in this flag. As is,
though, it just sows confusion, so I propose removing it as attached.
regards, tom lane
Attachment | Content-Type | Size |
---|---|---|
remove-useless-aborting-flag.patch | text/x-diff | 2.7 KB |
From | Date | Subject | |
---|---|---|---|
Next Message | Tom Lane | 2016-05-28 20:01:57 | Misdesigned command/status APIs for parallel dump/restore |
Previous Message | Joe Conway | 2016-05-28 16:39:45 | Re: Does people favor to have matrix data type? |