Re: [Xen-devel] [PATCH RFC 07/25] xen/common/Makefile: introduce HAS_CPUFREQ, HAS_PCI, HAS_PASSTHROUGH, HAS_NS16550

[Ian Campbell <Ian.Campbell@xxxxxxxxxx>]

On Fri, 2011-12-09 at 12:05 +0000, Stefano Stabellini wrote:
> On Wed, 7 Dec 2011, Jan Beulich wrote:
> > >>> On 07.12.11 at 12:24, Stefano Stabellini 
> > >>> <stefano.stabellini@xxxxxxxxxxxxx>
> > wrote:
> > > On Wed, 7 Dec 2011, Jan Beulich wrote:
> > >> >>> On 06.12.11 at 19:19, <stefano.stabellini@xxxxxxxxxxxxx> wrote:
> > >> > From: Stefano Stabellini <stefano.stabellini@xxxxxxxxxxxxx>
> > >> > 
> > >> > - make the compilation of ns16550.c depend upon HAS_NS16550;
> > >> > 
> > >> > - make the compilation of cpufreq depend upon HAS_CPUFREQ;
> > >> > 
> > >> > - make the compilation of pci depend upon HAS_PCI;
> > >> > 
> > >> > - make the compilation of passthrough depend upon HAS_PASSTHROUGH;
> > >> > 
> > >> > - make the compilation of kexec.o depend on HAS_ACPI.
> > >> 
> > >> How are kexec and ACPI connected to one another?
> > > 
> > > Today kexec implementation is dependent on ACPI.
> > 
> > The only dependency I can spot is the call to acpi_dmar_reinstate(),
> > which could clearly get an inline stub on platforms that don't have
> > ACPI (or pass-through).
> 
> It is better than I thought. I am compiling kexec too now.

Even if it can be made to compile it can't work unless someone writes
the actual moving parts which is a fair bit of work. I suggest we leave
this to one side for now and define HAS_KEXEC as False.

Ian.



_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-devel