Re: C functions, dll, server fall down

From: Merlin Moncure <mmoncure(at)gmail(dot)com>
To: AlexeyChe <alexey(dot)chernushev(at)gmail(dot)com>
Cc: pgsql-general(at)postgresql(dot)org
Subject: Re: C functions, dll, server fall down
Date: 2009-12-09 23:25:42
Message-ID: b42b73150912091525s354e9458n7f20123ba79015d4@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-general

On Wed, Dec 9, 2009 at 8:33 AM, AlexeyChe <alexey(dot)chernushev(at)gmail(dot)com> wrote:
>
> Hi there
> Need help, I just took the example from Postgres help, code:
>
> #include "postgres.h"
> #include "fmgr.h"
>
> #ifdef PG_MODULE_MAGIC
> PG_MODULE_MAGIC;
> #endif
>
> PG_FUNCTION_INFO_V1(add_one);
>
> __declspec (dllexport) Datum add_one(PG_FUNCTION_ARGS)
> {
>    int32   arg = PG_GETARG_INT32(0);
>
>    PG_RETURN_INT32(arg + 1);
> }
>
> PG_FUNCTION_INFO_V1(fo);
>
> __declspec (dllexport) Datum fo(PG_FUNCTION_ARGS)
> {
>    float8   arg = PG_GETARG_FLOAT8(0);
>
>    PG_RETURN_FLOAT8(arg+1);
> }
>
>
>
> CREATE OR REPLACE FUNCTION "public"."add_one_f"(double precision) RETURN
> double precision AS
> 'dll_c.dll','fo'
> LANGUAGE C STRICT
>
> Select add_one_f(99)
>
>
> server closed the connection unexpectedly
>        This probably means the server terminated abnormally
>        before OR while processing the request.
>
> Server falling down in a simple function.
> WinXP, server - Postgres 8.3, binaries for compile dll - 8.3.5
> build in Dev-cpp (mingw)
>
> PS. Function for int32( add_one(integer) ) is working

hm. I wonder if this is confusion between value/reference datums. Can
you do a quick check to see exactly where it's crashing (either before
you get into the function, at the getarg, or in the return)?

you can elog(WARNING, msg) to print out debug info from inside the function.

merlin

In response to

Browse pgsql-general by date

  From Date Subject
Next Message DM 2009-12-09 23:38:01 Install compat-postgresql-libs-debuginfo on postgres 8.3.8 without using RPM
Previous Message Stephen Tyler 2009-12-09 23:21:10 Re: Excessive (and slow) fsync() within single transaction