Re: [PATCH v2 1/8] xen: calculate XEN_BUILD_TIME using XEN_BUILD_DATE value

From: Ian Jackson <ian.jackson@citrix.com>
To: Daniel Kiper <daniel.kiper@oracle.com>
Cc: "Stefano Stabellini" <sstabellini@kernel.org>,
	"Wei Liu" <wei.liu2@citrix.com>,
	"George Dunlap" <George.Dunlap@eu.citrix.com>,
	"Andrew Cooper" <andrew.cooper3@citrix.com>,
	"Roger Pau Monné" <royger@FreeBSD.org>,
	"Tim Deegan" <tim@xen.org>, "Julien Grall" <julien.grall@arm.com>,
	"Jan Beulich" <JBeulich@suse.com>,
	"Tamas K Lengyel" <tamas.k.lengyel@gmail.com>,
	xen-devel <xen-devel@lists.xenproject.org>
Subject: Re: [PATCH v2 1/8] xen: calculate XEN_BUILD_TIME using XEN_BUILD_DATE value
Date: Wed, 4 Jul 2018 17:30:54 +0100	[thread overview]
Message-ID: <23356.63038.529901.339878@mariner.uk.xensource.com> (raw)
In-Reply-To: <20180704160743.GF26572@olila.local.net-space.pl>

Daniel Kiper writes ("Re: [PATCH v2 1/8] xen: calculate XEN_BUILD_TIME using XEN_BUILD_DATE value"):
> On Wed, Jul 04, 2018 at 04:41:30PM +0100, Ian Jackson wrote:
> > I don't have a FreeBSD host to hand to test, CC'ing Roger.
> 
> This is not the date command problem. This is more related to SOURCE_DATE_EPOCH
> spec (https://reproducible-builds.org/specs/source-date-epoch/#idm55)
> which clearly says:
>
>   The value MUST be reproducible (deterministic) across different
>   executions of the build, depending only on the source code. It SHOULD
>   be set to the last modification time of the source, incorporating any
>   packaging-specific modifications.
> 
> So, AIUI, above implies that we should not use the date command. Am I right?

No.

That spec is for people *setting* SOURCE_DATE_EPOCH.  Which we are not
doing.

For a *consumer* of SOURCE_DATE_EPOCH, if it is not set, using the
output of date or the current time or something is fine.

> > How do in intend to choose which wrapper script to use ?
> 
> I though that a mechanism to differentiate GNU and BSD systems exists in
> the Xen build system. So, I was going to reuse it somehow.

ISTR this being slightly clumsy.  But good luck :-).

> > This pile of stuff will run on awful lot of times during each make so
> > keeping it of reasonable size is a good plan.
> >
> > Maybe it would be better to have a single shell script which outputs a
> > whole bunch of variable settings for make to $(eval ) ?
> 
> It looks that this variable is referenced once, so, it should not be a problem.

I think make's recursive expansion system may result in these shell
runes being executed multiple times.

Ian.

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xenproject.org
https://lists.xenproject.org/mailman/listinfo/xen-devel