On Fri, 18 Dec 2009, Yinghai Lu wrote:
>
> ok, please check attached is right or not
>
> +static void __release_child_resources(struct resource *r)
> +{
> + struct resource *tmp, *p;
> + resource_size_t size;
> +
> + p = r->child;
> + r->child = NULL;
> + while (p) {
> + tmp = p;
> + p = p->sibling;
> +
> + tmp->parent = NULL;
> + tmp->sibling = NULL;
> + __release_child_resources(tmp);
> +
> + printk(KERN_DEBUG "release child resource %pR\n", tmp);
> + /* need to restore size, and keep flags */
> + size = resource_size(tmp);
> + tmp->start = 0;
> + tmp->end = size - 1;
> + }
> +}
Ok, this looks mostly right. I do worry about the alignment information:
you lose that thing for any resource that had IORESOURCE_STARTALIGN set
when you do this thing. That's pretty fundamental to the whole resource
code, I suspect we should just finally add a 'alignment' field to the
resource struct, so that alignment doesn't get lost when a resource is
allocated.
(Do a "git grep IORESOURCE_.*ALIGN" to see the kind of stuff I'm talking
about, and look at he PCI 'setup-bus.c' code that sets that STARTALIGN
thing).
So a preliminary ack on the resource.c parts. The rest I'm still a bit
dubious about, and the whole "we've lost alignment on the resources" is
probably indicative of how none of the resource code has ever really been
designed for this kind of "tear down and build back up again" behavior.
Linus
--
To unsubscribe from this list: send the line "unsubscribe linux-pci" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
