Re: [PATCHv7 10/19] x86/mm: Make __PHYSICAL_MASK_SHIFT and __VIRTUAL_MASK_SHIFT dynamic

From	"Kirill A. Shutemov" <kirill@shutemov.name>
Newsgroups	linux.kernel
Subject	Re: [PATCHv7 10/19] x86/mm: Make __PHYSICAL_MASK_SHIFT and __VIRTUAL_MASK_SHIFT dynamic
Date	2017-09-28 12:30 +0200
Message-ID	<uuEeT-7MF-19@gated-at.bofh.it> (permalink)
References	<ur1Wp-5gk-5@gated-at.bofh.it> <ur1Wq-5gk-29@gated-at.bofh.it> <uuCmK-6C9-31@gated-at.bofh.it>
Organization	linux.* mail to news gateway

Show all headers | View raw

On Thu, Sep 28, 2017 at 10:28:13AM +0200, Ingo Molnar wrote:
> 
> * Kirill A. Shutemov <kirill.shutemov@linux.intel.com> wrote:
> 
> > --- a/arch/x86/mm/dump_pagetables.c
> > +++ b/arch/x86/mm/dump_pagetables.c
> > @@ -82,8 +82,8 @@ static struct addr_marker address_markers[] = {
> >  	{ 0/* VMALLOC_START */, "vmalloc() Area" },
> >  	{ 0/* VMEMMAP_START */, "Vmemmap" },
> >  #ifdef CONFIG_KASAN
> > -	{ KASAN_SHADOW_START,	"KASAN shadow" },
> > -	{ KASAN_SHADOW_END,	"KASAN shadow end" },
> > +	{ 0/* KASAN_SHADOW_START */,	"KASAN shadow" },
> > +	{ 0/* KASAN_SHADOW_END */,	"KASAN shadow end" },
> 
> What's this? Looks hacky.

KASAN_SHADOW_START and KASAN_SHADOW_END depend on __VIRTUAL_MASK_SHIFT,
which is dynamic for boot-time switching case. It means we cannot
initialize the corresponding address_markers fields compile-time, so we do
it boot-time.

I used the same approach we use to deal with dynamic VMALLOC_START,
VMEMMAP_START and PAGE_OFFSET.

-- 
 Kirill A. Shutemov

Thread

Re: [PATCHv7 10/19] x86/mm: Make __PHYSICAL_MASK_SHIFT and  __VIRTUAL_MASK_SHIFT dynamic Ingo Molnar <mingo@kernel.org> - 2017-09-28 10:30 +0200
  Re: [PATCHv7 10/19] x86/mm: Make __PHYSICAL_MASK_SHIFT and  __VIRTUAL_MASK_SHIFT dynamic "Kirill A. Shutemov" <kirill@shutemov.name> - 2017-09-28 12:30 +0200
    Re: [PATCHv7 10/19] x86/mm: Make __PHYSICAL_MASK_SHIFT and  __VIRTUAL_MASK_SHIFT dynamic Ingo Molnar <mingo@kernel.org> - 2017-09-28 12:50 +0200

csiph-web