Re: Add CHECK_FOR_INTERRUPTS in polling loop code path in XactLockTableWait
От | Xuneng Zhou |
---|---|
Тема | Re: Add CHECK_FOR_INTERRUPTS in polling loop code path in XactLockTableWait |
Дата | |
Msg-id | CABPTF7UU4fBt0HiH+YxXLSD2=e3V-7RNggk90Mg2qeLBdLBhOg@mail.gmail.com обсуждение исходный текст |
Ответ на | Re: Add CHECK_FOR_INTERRUPTS in polling loop code path in XactLockTableWait (Fujii Masao <masao.fujii@oss.nttdata.com>) |
Список | pgsql-hackers |
Hi, > Thanks for the patch! I haven't reviewed it yet, but since this is > a v19 item, please add it to the next CommitFest so we don't lose > track of it. > > Also, I think it would be better to split the addition of the wait event > and the introduction of exponential backoff in XactLockTableWait() into > separate patches. They serve different purposes and can be committed > independently. > I've renamed and created two discussion threads and commitfest entries for these patches to allow independent evaluation. 1) Add new wait event to XactLockTableWait https://commitfest.postgresql.org/patch/5804/ https://www.postgresql.org/message-id/CABPTF7WZODAVPFxtn9ygA9d6zckkJbFG%3DSUtHdvk7ca%3DUTzSFg%40mail.gmail.com 2) Add progressive backoff to XactLockTableWait https://commitfest.postgresql.org/patch/5806/ https://www.postgresql.org/message-id/CABPTF7XmTrBp8S93a%2BzQ5M3FhLB6o8kWn9yQ1YnHJqTPT9dRYA%40mail.gmail.com There's some code overlap between XactLockTableWait() and ConditionalXactLockTableWait() in 2) that would warrant a shared helper function. However, introducing such a helper would create dependencies between the patches, defeating the purpose of keeping them separate. I’m very familiar with splitting patches and submitting them separately, so please let me know if anything here needs improvement. Best regards, Xuneng
В списке pgsql-hackers по дате отправления: