Re: Move OpenSSL random under USE_OPENSSL_RANDOM

Magnus Hagander <magnus@hagander.net>

From: Magnus Hagander <magnus@hagander.net>
To: Michael Paquier <michael@paquier.xyz>
Cc: Daniel Gustafsson <daniel@yesql.se>, Tom Lane <tgl@sss.pgh.pa.us>, Postgres hackers <pgsql-hackers@lists.postgresql.org>
Date: 2020-11-20T12:59:18Z
Lists: pgsql-hackers
On Fri, Nov 20, 2020 at 3:31 AM Michael Paquier <michael@paquier.xyz> wrote:
>
> On Thu, Nov 19, 2020 at 09:49:05PM +0100, Magnus Hagander wrote:
> > Ugh, that's pretty terrible.
>
> I have spent some time testing this part this morning, and I can see
> that genhtml does not complain with your patch.  It looks like in the
> case of 2771fce the tool got confused because the same function was
> getting compiled twice for the backend and the frontend, but here you
> only get one code path compiled depending on the option used.
>
> +#else /* not OPENSSL or WIN32 */
> I think you mean USE_OPENSSL or just OpenSSL here, but not "OPENSSL".

Yeah. Well, I either meant "OpenSSL or Win32" or "USE_OPENSSL or
WIN32", and ended up with some incorrect mix :) Thanks, fixed.


> pg_strong_random.c needs a pgindent run, there are two inconsistent
> diffs.  Looks fine except for those nits.

I saw only one after this, but maybe I ended up auto-fixing it whenI
changed that define.

That said, pgindent now run, and patch pushed.

Thanks!


--
 Magnus Hagander
 Me: https://www.hagander.net/
 Work: https://www.redpill-linpro.com/



Commits

  1. Remove ability to independently select random number generator

  2. Add pg_strong_random_init function to initialize random number generator