This is an archived copy of the Xen.org mailing list, which we have preserved to ensure that existing links to archives are not broken. The live archive, which contains the latest emails, can be found at http://lists.xen.org/
Home Products Support Community News


[Xen-devel] architecture selection in config

To: <xen-devel@xxxxxxxxxxxxxxxxxxx>
Subject: [Xen-devel] architecture selection in config
From: "Jan Beulich" <JBeulich@xxxxxxxxxx>
Date: Tue, 14 Jun 2005 01:41:47 -0600
Delivery-date: Tue, 14 Jun 2005 07:41:18 +0000
Envelope-to: www-data@xxxxxxxxxxxxxxxxxxx
List-help: <mailto:xen-devel-request@lists.xensource.com?subject=help>
List-id: Xen developer discussion <xen-devel.lists.xensource.com>
List-post: <mailto:xen-devel@lists.xensource.com>
List-subscribe: <http://lists.xensource.com/cgi-bin/mailman/listinfo/xen-devel>, <mailto:xen-devel-request@lists.xensource.com?subject=subscribe>
List-unsubscribe: <http://lists.xensource.com/cgi-bin/mailman/listinfo/xen-devel>, <mailto:xen-devel-request@lists.xensource.com?subject=unsubscribe>
Sender: xen-devel-bounces@xxxxxxxxxxxxxxxxxxx
Trying to integrate the unstable sparse Linux tree into the SuSE one I ran into 
a number of build issues, mostly associated with out-of-source tree building 
(patches later). However, in the course of fixing one particular problem (the 
microcode patch driver not buildable as a module), I changed its i386 Kconfig 
option to bool, resulting in a warning that the x86-64 Kconfig redefines this 
to be a tristate. Changing this option there to be bool obviously circumvents 
the problem (and in the given case is necessary anyway), but this begs a more 
general question: Isn't the way architecture selection is currently done prone 
to cause grief especially when more acritectures get added? Since all 
architectures' Kconfig-s get included from arch/xen/Kconfig, contradicting 
types of config options in architectures more distinct than i386 from x86-64 
are likely in my opinion, and hence are likely to cause problems looking 

And following that, an even more general question (tried to search the list 
archives, but didn't quickly find a discussion of this topic) would be whether 
XEN shouldn't better integrate with the way architecture determination is done 
in the base kernel, namely making use of the ARCH/SUBARCH functionality already 
present for UML (and thus by default deriving the architecture from the host 
one instead of making this a config option, requiring a command line argument 
when a cross config is desired).

Thanks, Jan

Xen-devel mailing list

<Prev in Thread] Current Thread [Next in Thread>
  • [Xen-devel] architecture selection in config, Jan Beulich <=