Обсуждение: C Function Memory Management

Hello


2012/12/12 Martin French <Martin.French@romaxtech.com>:
> Ok, so just a (hopefully) quick question.
>
> I know that I should be using the Postgres function palloc in place of
> malloc, but what about realloc? Is there an equivalent?

yes,

repalloc

Regards

Pavel Stehule

>
> Thanks.
> =============================================
>
> Romax Technology Limited
> Rutherford House
> Nottingham Science & Technology Park
> Nottingham,
> NG7 2PZ
> England
>
> Telephone numbers:
> +44 (0)115 951 88 00 (main)
>
> For other office locations see:
> http://www.romaxtech.com/Contact
> =================================
> ===============
> E-mail: info@romaxtech.com
> Website: www.romaxtech.com
> =================================
>
> ================
> Confidentiality Statement
> This transmission is for the addressee only and contains information that is
> confidential and privileged.
> Unless you are the named addressee, or authorised to receive it on behalf of
> the addressee
> you may not copy or use it, or disclose it to anyone else.
> If you have received this transmission in error please delete from your
> system and contact the sender. Thank you for your cooperation.
> =================================================

Martin French <Martin.French@romaxtech.com> writes:
>> repalloc

> Maybe I should've looked a little harder for that one! Ha Ha!

> I assume it's an exact functional copy of realloc, and see where I get
> with it!

One thing to know about it is that repalloc re-sizes the given chunk
while keeping it in the same memory context it was originally allocated
in (ie, CurrentMemoryContext doesn't matter).  This may not matter much
if you're only dealing with allocations that just live as long as the
current function call --- but if you're trying to preserve data across
calls, it's critical to know.

            regards, tom lane

Martin French <Martin.French@romaxtech.com> writes:
> So out of curiosity, what would be your approach for data preservation
> cross call?

Usually I'd allocate the space in fcinfo->flinfo->fn_mcxt (which is a
query-lifespan context) and keep a pointer to it in
fcinfo->flinfo->fn_extra (which is a void * field that a C function
can do what it pleases with).  There are plenty of examples in the
Postgres source code.  If you are looking to cache data longer than
one query then it takes more thought.

            regards, tom lane