Re: [PATCH v10 13/17] mm: vmalloc: add flag preventing guard hole allocation

[Andrey Ryabinin]

On 01/30/2015 02:12 AM, Andrew Morton wrote:
> On Thu, 29 Jan 2015 18:11:57 +0300 Andrey Ryabinin <a.ryabinin@xxxxxxxxxxx> wrote:
> 
>> For instrumenting global variables KASan will shadow memory
>> backing memory for modules. So on module loading we will need
>> to allocate shadow memory and map it at exact virtual address.
> 
> I don't understand.  What does "map it at exact virtual address" mean?
> 

I mean that if module_alloc() returned address x, than
shadow memory should be mapped exactly at address kasan_mem_to_shadow(x).

>> __vmalloc_node_range() seems like the best fit for that purpose,
>> except it puts a guard hole after allocated area.
> 
> Why is the guard hole a problem?
> 

Because of guard hole in shadow some future allocations of shadow memory
will fail. Requested address ( kasan_mem_to_shadow(x) ) will be already occupied
by guard hole of previous allocation.


--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/