Re: [patch] mm: reduce pagetable-freeing latencies

From: Benjamin Herrenschmidt
Date: Wed Jul 25 2007 - 06:10:23 EST


On Wed, 2007-07-25 at 11:46 +0200, Andi Kleen wrote:
> > You could perhaps use C99 variable length arrays to avoid the stack
> > waste when not needed, however Andi once told me that generates rather
> > dubious code.
>
> It generates frame pointers, but that's not that bad. I'm not
> aware of any other bad side effects. Ok the compiler will limit
> your goto usage, but that's more a good thing.
>
> But since you always have to strictly limit the array in kernel code anyways
> you could as well just allocate the fixed limit.

Plan is fixed array or 4 or maybe 8 entries (pointers), that shouldn't
be -too- bad. The code path I'm a bit worried about is
unmap_mapping_ranges() which goes into zapping page tables from deep
within filesystems.

At worst, I can reduce the fixed array to 1 entry. That means that if
the batch can't manage to get a page to use for the page list, it will
end up doing the flush for each page :-) But that should rarely happen,
in fact, I would expect it to be able to get a page the next time around
because it just freed one...

Ben.


-
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/