Re: RFC: ACPI/scsi/libata integration and hotswap

[Christoph Hellwig <hch@infradead.org>]

On Fri, Dec 09, 2005 at 11:50:09AM +0000, Matthew Garrett wrote:
> On Fri, Dec 09, 2005 at 11:42:46AM +0000, Christoph Hellwig wrote:
> 
> > As a concept it's _much_ better.  Although it should be platform_scsi_init
> > and every architecture would provide an, in most cases noop, implementation.
> 
> How about
> 
> if (platform_scsi_init)
> 	platform_scsi_init(&scsi_bus_type);
> 
> ? This is similar to how the platform_notify callback code is handled. 
> Making it per-arch isn't quite ideal, since x86 can be ACPI or APM and 
> kernels need support for both. On the other hand, I can't think of any 
> way that APM could do anything useful with the information, so per-arch 
> may be reasonable.

I think a per-arch hook is better, if an architecture needs different
backend implementations it can dispatch internally.  And the above won't
work unless platform_scsi_init is a function pointer which would be quite
ugly.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/