[PATCH] Fix build on hosts that won't allow var = {} initialisation

Andrew Bartlett

unread,

May 21, 2013, 5:47:35 PM5/21/13

to

This build failure is preventing a number of build farm hosts from
building, and in turn is making it harder to do config.h difference
comparisons for bug 8969.

Please review/push.

Thanks,

Andrew Bartlett
--
Andrew Bartlett http://samba.org/~abartlet/
Authentication Developer, Samba Team http://samba.org

0001-smbd-Fix-build-on-platforms-that-will-not-support-va.patch

Richard Sharpe

unread,

May 22, 2013, 12:51:50 AM5/22/13

to

On Tue, May 21, 2013 at 2:47 PM, Andrew Bartlett <abar...@samba.org> wrote:
> This build failure is preventing a number of build farm hosts from
> building, and in turn is making it harder to do config.h difference
> comparisons for bug 8969.
>
> Please review/push.

It looks OK. However, there is no signed-off-by line.

> Thanks,
>
> Andrew Bartlett
> --
> Andrew Bartlett http://samba.org/~abartlet/
> Authentication Developer, Samba Team http://samba.org
>

--
Regards,
Richard Sharpe
(何以解憂？唯有杜康。--曹操)

Andrew Bartlett

unread,

May 22, 2013, 8:11:14 PM5/22/13

to

On Tue, 2013-05-21 at 23:18 -0700, Richard Sharpe wrote:
> On Tue, May 21, 2013 at 10:11 PM, Andrew Bartlett <abar...@samba.org> wrote:

> > On Tue, 2013-05-21 at 21:51 -0700, Richard Sharpe wrote:
> >> On Tue, May 21, 2013 at 2:47 PM, Andrew Bartlett <abar...@samba.org> wrote:
> >> > This build failure is preventing a number of build farm hosts from
> >> > building, and in turn is making it harder to do config.h difference
> >> > comparisons for bug 8969.
> >> >
> >> > Please review/push.
> >>
> >> It looks OK. However, there is no signed-off-by line.
> >

> > https://www.samba.org/samba/devel/copyright-policy.html describes the
> > meaning of Signed-off-by for Samba, but as I own my copyright personally
> > and so have not mailed in that statement (including LGPL permission), it
> > does not apply to me.
> >
> > In consideration of the above, I choose not to add this line to avoid
> > any further confusion as to the licence or circumstances under which I
> > provide my code.
>
> OK. Reviewed and pushed.

I don't see it in master, did something go wrong with this, or was there
a mix-up with my md5 patch perhaps?

Richard Sharpe

unread,

May 23, 2013, 12:38:07 AM5/23/13

to

On Wed, May 22, 2013 at 5:11 PM, Andrew Bartlett <abar...@samba.org> wrote:
> On Tue, 2013-05-21 at 23:18 -0700, Richard Sharpe wrote:
>> On Tue, May 21, 2013 at 10:11 PM, Andrew Bartlett <abar...@samba.org> wrote:
>> > On Tue, 2013-05-21 at 21:51 -0700, Richard Sharpe wrote:
>> >> On Tue, May 21, 2013 at 2:47 PM, Andrew Bartlett <abar...@samba.org> wrote:
>> >> > This build failure is preventing a number of build farm hosts from
>> >> > building, and in turn is making it harder to do config.h difference
>> >> > comparisons for bug 8969.
>> >> >
>> >> > Please review/push.
>> >>
>> >> It looks OK. However, there is no signed-off-by line.
>> >
>> > https://www.samba.org/samba/devel/copyright-policy.html describes the
>> > meaning of Signed-off-by for Samba, but as I own my copyright personally
>> > and so have not mailed in that statement (including LGPL permission), it
>> > does not apply to me.
>> >
>> > In consideration of the above, I choose not to add this line to avoid
>> > any further confusion as to the licence or circumstances under which I
>> > provide my code.
>>
>> OK. Reviewed and pushed.
>
> I don't see it in master, did something go wrong with this, or was there
> a mix-up with my md5 patch perhaps?

Is it not this one?

commit 30a224367595f9e2506e41053fc9c7cb96a90928
Author: Andrew Bartlett <abar...@samba.org>
Date: Wed May 22 00:14:19 2013 +1000

build: Update md5.h detection in waf and autoconf to use sys/md5.h and -lmd

This brings the two build systems in sync, without using md5.h (which is a p

Tested on FreeBSD

Andrew Bartlett

Reviewed-by: Richard Sharpe <rsh...@samba.org>

Autobuild-User(master): Richard Sharpe <sha...@samba.org>
Autobuild-Date(master): Wed May 22 10:06:40 CEST 2013 on sn-devel-104

Andrew Bartlett

unread,

May 23, 2013, 1:42:32 AM5/23/13

to

No, the patch I was proposing is attached. What you pushed is another
patch I was also proposing (I've been busy).

0001-smbd-Fix-build-on-platforms-that-will-not-support-va.patch

Christian Ambach

unread,

May 23, 2013, 11:58:41 AM5/23/13

to

On 05/23/2013 07:42 AM, Andrew Bartlett wrote:
> No, the patch I was proposing is attached. What you pushed is another
> patch I was also proposing (I've been busy).

> - struct smb_acl_wrapper acl_wrapper = {};
> + /* Initialise this to zero, in a portable way */
> + struct smb_acl_wrapper acl_wrapper = {
> + NULL
> + };

Why not use ZERO_STRUCT (as used elsewhere)?

Cheers,
Christian

Volker Lendecke

unread,

May 24, 2013, 2:39:40 AM5/24/13

to

ZERO_STRUCT is an explicit call to memset. With the
explicit initialization the compiler at least has a chance
to do it differently if it chooses so.

Volker

--
SerNet GmbH, Bahnhofsallee 1b, 37081 Göttingen
phone: +49-551-370000-0, fax: +49-551-370000-9
AG Göttingen, HRB 2816, GF: Dr. Johannes Loxen
http://www.sernet.de, mailto:kon...@sernet.de

Andrew Bartlett

unread,

May 24, 2013, 4:09:53 AM5/24/13

to

On Fri, 2013-05-24 at 08:39 +0200, Volker Lendecke wrote:
> On Thu, May 23, 2013 at 05:58:41PM +0200, Christian Ambach wrote:
> > On 05/23/2013 07:42 AM, Andrew Bartlett wrote:
> > >No, the patch I was proposing is attached. What you pushed is another
> > >patch I was also proposing (I've been busy).
> >
> > >- struct smb_acl_wrapper acl_wrapper = {};
> > >+ /* Initialise this to zero, in a portable way */
> > >+ struct smb_acl_wrapper acl_wrapper = {
> > >+ NULL
> > >+ };
> >
> > Why not use ZERO_STRUCT (as used elsewhere)?
>
> ZERO_STRUCT is an explicit call to memset. With the
> explicit initialization the compiler at least has a chance
> to do it differently if it chooses so.

Thanks Volker, that puts it much better than I could express!