Re: [PATCH, V2] i386: instead of poisoning .init zone, changeprotection bits to force a fault

[Andrew Morton]

Eric Dumazet <dada1@xxxxxxxxxxxxx> wrote:
>
> Andrew Morton a écrit :
> 
> > But I'm inclined to drop the whole patch - I don't see how it can detect
> > any bugs which CONFIG_DEBUG_PAGEALLOC won't find.
> > 
> 
> If CONFIG_DEBUG_PAGEALLOC is selected, does a page freed by free_page(addr); 
> guaranted not being reused later ?

No.

So with your patch, any access to the freed page will oops.  With
CONFIG_DEBUG_PAGEALLOC it'll only oops if that page is presently free.

So if enough people run with CONFIG_DEBUG_PAGEALLOC for enough time, we'll
find the same bugs.

> Anyway, the CONFIG_DEBUG_INITDATA was a temporary patch in order to track the 
> accesses to non possible cpus percpu data. So if you feel all such accesses 
> were cleaned, we can drop the patch...

I think so..
-
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/