From: | Pavel Stehule <pavel(dot)stehule(at)gmail(dot)com> |
---|---|
To: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
Cc: | PostgreSQL Hackers <pgsql-hackers(at)postgresql(dot)org> |
Subject: | Re: Let's get rid of SPI_push/SPI_pop |
Date: | 2016-11-07 04:43:22 |
Message-ID: | CAFj8pRBXFKsM9z795KTgepLs-7F8+0zMw2-ZUVxh9Zd8=vdf7g@mail.gmail.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
2016-11-07 2:16 GMT+01:00 Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>:
> The intent of SPI_push/SPI_pop seems to be to draw a boundary line between
> nested layers of SPI callers. Which is fine, but the SPI_connect and
> SPI_finish calls of the inner layer would suffice for that. AFAICS,
> the only thing that SPI_push/SPI_pop buy for us is the ability to detect
> a missing SPI_connect or SPI_finish in an inner function layer. And
> that seems pretty useless, because any such bug in a function would be
> immediately detected in simple testing that calls it without any outer
> level of SPI calls.
>
> As against that, we have the risk of forgotten SPI_push/SPI_pop calls that
> go undetected for years, as just seen in commit fc8b81a29. We've had that
> type of bug before too, cf 0d4899e44. And then there's the fact that we
> put conditional SPI_push/SPI_pop calls into various places, eg deac9488d,
> which it seems to me largely destroys whatever debugging value the concept
> did have.
>
> So I think we should just delete these functions and adjust SPI_connect
> and SPI_finish so that they just push/pop a context level unconditionally.
> (Which will make them simpler, not more complicated.)
>
> We can provide do-nothing macros by these names to avoid an API break
> for third-party extensions.
>
> Comments, objections?
>
cannot be there some performance impacts?
Regards
Pavel
>
> regards, tom lane
>
>
> --
> Sent via pgsql-hackers mailing list (pgsql-hackers(at)postgresql(dot)org)
> To make changes to your subscription:
> http://www.postgresql.org/mailpref/pgsql-hackers
>
From | Date | Subject | |
---|---|---|---|
Next Message | Tsunakawa, Takayuki | 2016-11-07 04:44:58 | Re: [RFC] Should we fix postmaster to avoid slow shutdown? |
Previous Message | Michael Paquier | 2016-11-07 04:36:45 | Re: pg_hba_file_settings view patch |