Re: [BUG] Re-entering malloc problem when use --enable-nlsbuildpostgresql - Mailing list pgsql-bugs

From 158306855
Subject Re: [BUG] Re-entering malloc problem when use --enable-nlsbuildpostgresql
Date
Msg-id tencent_31D85ACEFC44E35C77680F58477027A56706@qq.com
Whole thread Raw
In response to Re: [BUG] Re-entering malloc problem when use --enable-nlsbuildpostgresql  (Andres Freund <andres@anarazel.de>)
List pgsql-bugs
> Can't we just reuse errmsg_internal etc?

Yes, the *_internal function can be reused. This is my modified patch.
Please take a look.


Zeng Wenjing

------------------ Original ------------------
From:  "Andres Freund"<andres@anarazel.de>;
Date:  Wed, May 23, 2018 05:11 AM
To:  "158306855"<anderson2013@qq.com>;
Cc:  "Tom Lane"<tgl@sss.pgh.pa.us>; "pgsql-bugs"<pgsql-bugs@lists.postgresql.org>;
Subject:  Re: [BUG] Re-entering malloc problem when use --enable-nlsbuildpostgresql

On 2018-05-21 10:49:53 +0800, 158306855 wrote:
> > It'd be fairly unproblematic to write an untranslated message out. There
> > we can make sure to either only use plain syscalls or use memory from
> > the preallocated context.  I think it'd be ok to not to translate in
> > that situation.
>
>
> I wrote a patch to improve this problem.

> +extern int errmsg_no_translateit(const char *fmt,...) pg_attribute_printf(1, 2);
> +extern int errdetail_no_translateit(const char *fmt,...) pg_attribute_printf(1, 2);
> +extern int errhint_no_translateit(const char *fmt,...) pg_attribute_printf(1, 2);
> +
>  /*
>   * errcontext() is typically called in error context callback functions, not
>   * within an ereport() invocation. The callback function can be in a different

Can't we just reuse errmsg_internal etc?

Greetings,

Andres Freund

Attachment

pgsql-bugs by date:

Previous
From: Kyotaro HORIGUCHI
Date:
Subject: Re: postgres_fdw misbehaviour using "DELETE ... RETURNING *"
Next
From: PG Bug reporting form
Date:
Subject: BUG #15207: In crash recovery,SyncDataDirectory() will fall into recursion trap