Re: [PATCH] mm: Use aligned zone start for pfn_to_bitidx calculation

[Mel Gorman]

On Tue, Dec 04, 2012 at 02:10:01PM -0800, Laura Abbott wrote:
> The current calculation in pfn_to_bitidx assumes that
> (pfn - zone->zone_start_pfn) >> pageblock_order will return the
> same bit for all pfn in a pageblock. If zone_start_pfn is not
> aligned to pageblock_nr_pages, this may not always be correct.
> 
> Consider the following with pageblock order = 10, zone start 2MB:
> 
> pfn     | pfn - zone start | (pfn - zone start) >> page block order
> ----------------------------------------------------------------
> 0x26000 | 0x25e00	   |  0x97
> 0x26100 | 0x25f00	   |  0x97
> 0x26200 | 0x26000	   |  0x98
> 0x26300 | 0x26100	   |  0x98
> 
> This means that calling {get,set}_pageblock_migratetype on a single
> page will not set the migratetype for the full block. The correct
> fix is to round down zone_start_pfn for the bit index calculation.
> Rather than do this calculation everytime, store this precalcualted
> algined start in the zone structure to allow the actual start_pfn to
> be used elsewhere.
> 
> Change-Id: I13e2f53f50db294f38ec86138c17c6fe29f0ee82
> Signed-off-by: Laura Abbott <lauraa@xxxxxxxxxxxxxx>

Hi Laura,

There should be no need to add a new zone field. It's probably ok in terms
of functionality but it does mean that we have to worry about things like
hotplug (FWIW, should be fine) and the memory overhead is added even on
CONFIG_SPARSEMEM where it is not needed. Instead, mask out the lower bits
in pfn_to_bitidx() using the same round_down trick you already do. The
cost is negligible.

Thanks.

-- 
Mel Gorman
SUSE Labs
--
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/