Proposal for Signal Detection Refactoring

From: Chris Travers <chris(dot)travers(at)adjust(dot)com>
To: pgsql-hackers(at)lists(dot)postgresql(dot)org
Subject: Proposal for Signal Detection Refactoring
Date: 2018-09-20 13:08:34
Message-ID: CAN-RpxD_MDp_wbP2Ob_6TyRdTc+4poufEXbQUnQNVBpAcdY+YQ@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox
Thread:
Lists: pgsql-hackers

After the patch we did for the race condition here, one point which was
brought up which I thought was quite relevant was the fact that checking
global flags feels a little ugly.

I would like to set up a patch which refactors this as follows:

1. We create an enum of cancellation levels:
NO_INTERRUPT
INTERRUPT
QUERY_CANCEL
PROCESS_EXIT

2. We create a macro called PENDING_INTERRUPT_LEVEL() which returns the
highest pending interrupt level.

3. We check on whether the output of this is greater or equal to the level
we need to handle and work on that basis.

This allows us to handle cases where we just want to check the interrupt
level for return or cleanup purposes. It does not apply to cases where we
need to change interrupt handling temporarily as we do in a couple of
cases. I think it is cleaner than checking the QueryCancelPending and
ProcDiePending global flags directly.

So here's a small patch. I will add it for the next commit fest unless
anyone has any reason I shouldn't.

--
Best Regards,
Chris Travers
Head of Database

Tel: +49 162 9037 210 | Skype: einhverfr | www.adjust.com
Saarbrücker Straße 37a, 10405 Berlin

Attachment Content-Type Size
sig_refactor.patch application/octet-stream 2.8 KB

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Jonathan S. Katz 2018-09-20 13:09:58 Re: PostgreSQL 11 {Beta 4, RC1} Release: 2018-09-20
Previous Message Stephen Frost 2018-09-20 13:07:21 Re: Query is over 2x slower with jit=on