[PATCH v2] ARM: vectors: use gate_vma for vectors user mapping

[Will Deacon]

Hi Russell,

On Thu, Feb 23, 2012 at 09:01:52AM +0000, Russell King - ARM Linux wrote:
> On Tue, Jan 17, 2012 at 02:11:14PM -0500, Nicolas Pitre wrote:
> > On Tue, 17 Jan 2012, Will Deacon wrote:
> > 
> > > The current user mapping for the vectors page is inserted as a `horrible
> > > hack vma' into each task via arch_setup_additional_pages. This causes
> > > problems with the MM subsystem and vm_normal_page, as described here:
> > > 
> > > https://lkml.org/lkml/2012/1/14/55
> > > 
> > > Following the suggestion from Hugh in the above thread, this patch uses
> > > the gate_vma for the vectors user mapping, therefore consolidating
> > > the horrible hack VMAs into one.
> > > 
> > > Cc: Nicolas Pitre <nico at fluxnic.net>
> > > Signed-off-by: Will Deacon <will.deacon at arm.com>
> > 
> > Acked-by: Nicolas Pitre <nico at linaro.org>
> > Tested-by: Nicolas Pitre <nico at linaro.org>
> 
> Al Viro has spotted this patch, and commented on it last night:
> | static struct vm_area_struct gate_vma;
> | static int __init gate_vma_init(void)
> | {
> |         gate_vma.vm_start       = 0xffff0000;
> |         gate_vma.vm_end         = 0xffff0000 + PAGE_SIZE;
> |         gate_vma.vm_page_prot   = PAGE_READONLY_EXEC;
> |         gate_vma.vm_flags       = VM_READ | VM_EXEC |
> |                                   VM_MAYREAD | VM_MAYEXEC |
> |                                   VM_ALWAYSDUMP;
> |         return 0;
> | }
> | why bother with initcall?
> | it's all constant, AFAICS
> | IOW, why not initialize it statically and be done with that?

PAGE_READONLY_EXEC isn't constant - it's constructed from pgprot_user which
is initialised in mmu.c during boot.

Will