Messages in this thread | | | Date | Tue, 28 May 2013 17:48:01 +0100 | From | Matthew Garrett <> | Subject | Re: [PATCH] x86/PCI: setup data may be in highmem |
| |
On Tue, May 28, 2013 at 10:31:51AM -0600, Bjorn Helgaas wrote: > On Tue, May 28, 2013 at 5:36 AM, Matt Fleming <matt@console-pimps.org> wrote: > > On Fri, 24 May, at 08:38:10AM, Bjorn Helgaas wrote: > >> pcibios_add_device() is mostly called at boot-time, when > >> early_ioremap() probably works well. But it's also called when we > >> hot-add devices later, and it looks like early_ioremap() will then > >> generate warnings because "system_state != SYSTEM_BOOTING". > > > > Oops. Good point. Is there any reason we can't use ioremap()? > > I assume Matthew had some reason for avoiding that in the first place, > so I hope he'll chime in.
I have no recollection of why I did it that way. If this is all happening late enough for ioremap() to work then that sounds fine.
-- Matthew Garrett | mjg59@srcf.ucam.org
| |