About SECTION_SIZE_BITS for Sparsemem

[Russell King - ARM Linux]

On Mon, Jul 12, 2010 at 09:28:16PM +0900, Minchan Kim wrote:
> It just checks that there is a section and section_mem_map has SECTION_HAS_MEM_MAP.
> The first section in above case has just 80M memory but section has 256M. 
> So, 0x25000000 - 28000000 is the hole. If you pass pfn whihc is 0x2500000,
> let's see pfn_valid.

That's not a hole as such.  It's an incompletely filled section, which is
precisely what sparsemem is all about.

> 1. We pass pfn_to_section_nr check 
> 2. Both __nr_to_section and valid_section is vaild.
> 
> static inline int pfn_valid(unsigned long pfn) 
> {
>         if (pfn_to_section_nr(pfn) >= NR_MEM_SECTIONS)
>                 return 0;
>         return valid_section(__nr_to_section(pfn_to_section_nr(pfn)));
> }
> 
> 
> What prevent above hole's case?

That looks too loose to be useful - that means invalid page table entries
can pass the pfn_valid() test when they should actually fail, and I'd
imagine there will be many more such problems like that.

> I think at least pfn_valid in sparsemem need bank range check like pfn_valid of ARM 
> in FLATMEM.

I agree - pfn_valid() on sparsemem needs to be tightened.