Re: [PATCH 01/15] mm: poison struct page for ptlock

[Andrew Morton]

Hugh Dickins <hugh@xxxxxxxxxxx> wrote:
>
>  On Thu, 10 Nov 2005, Andrew Morton wrote:
>  > Hugh Dickins <hugh@xxxxxxxxxxx> wrote:
>  > > On Thu, 10 Nov 2005, Ingo Molnar wrote:
>  > > > 
>  > > > yuck. What is the real problem btw? AFAICS there's enough space for a 
>  > > > 2-word spinlock in struct page for pagetables.
>  > > 
>  > > Yes.  There is no real problem.  But my patch offends good taste.
>  > 
>  > Isn't it going to overrun page.lru with CONFIG_DEBUG_SPINLOCK?
> 
>  No.

!no, methinks.

On 32-bit architectures with CONFIG_PREEMPT, CONFIG_DEBUG_SPINLOCK,
CONFIG_NR_CPUS >= CONFIG_SPLIT_PTLOCK_CPUS we have a 32-byte page, a
20-byte spinlock and offsetof(page, private) == 12.

IOW we're assuming that no 32-bit architectures will obtain pagetables from
slab?
-
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/