Re: Use pg_malloc macros in src/fe_utils

Michael Paquier <michael@paquier.xyz>

From: Michael Paquier <michael@paquier.xyz>
To: Andreas Karlsson <andreas@proxel.se>
Cc: Henrik TJ <henrik@0x48.dk>, PostgreSQL Hackers <pgsql-hackers@postgresql.org>
Date: 2026-02-27T10:00:00Z
Lists: pgsql-hackers
On Fri, Feb 27, 2026 at 02:15:46AM +0100, Andreas Karlsson wrote:
> 1. What should we do about when we allocate a an array of characters? Would
> it make sense to use pg_array_alloc() or would that jsut be silly? For
> example:
> 
> -pad = (char *) pg_malloc(l + 1);
> +pad = pg_malloc_array(char, l + 1);

I can see that tar_get_file_name() has been changed in 0001, which is
fine, so I have merged the change from 0002 in
dir_get_file_name()@walmethods.c into 0001, for consistency.  I don't
really have a strong opinion about the rest of 0002, TBH. 

> 2. I found a small and harmless thinko. The buffer in verify_tar_file() is
> actually a char * but for some reason the code did the following:
> 
> buffer = pg_malloc(READ_CHUNK_SIZE * sizeof(uint8));
> 
> What should we do about it? Just skip the "sizof(uint8)"?

This one has already been discussed, see here:
https://www.postgresql.org/message-id/aUJ2zxgPCaVsVi2a@ip-10-97-1-34.eu-west-3.compute.internal
The story is a bit larger than this single allocation, as it impacts
the meaning of the surrounding routines with backup manifests.

And applied 0001 after double-checking it.  Thanks.
--
Michael

Commits

Same data as JSON: GET /api/v1/messages/:b64id/commits the thread's linked commits as JSON, with link sources. API reference →
  1. Use pg_malloc_object() and pg_alloc_array() variants in frontend code

  2. fe_utils: Sprinkle some pg_malloc_object() and pg_malloc_array()