Re: [PATCH] mm: fix calculation accounting dirtyable highmem

[Mel Gorman]

On Wed, Jul 13, 2016 at 11:23:13AM +0900, Minchan Kim wrote:
> When I tested vmscale in mmtest in 32bit, I found the benchmark
> was slow down 0.5 times.
> 
>                 base        node
>                    1    global-1
> User           12.98       16.04
> System        147.61      166.42
> Elapsed        26.48       38.08
> 
> With vmstat, I found IO wait avg is much increased compared to
> base.
> 
> The reason was highmem_dirtyable_memory accumulates free pages
> and highmem_file_pages from HIGHMEM to MOVABLE zones which was
> wrong. With that, dirth_thresh in throtlle_vm_write is always
> 0 so that it calls congestion_wait frequently if writeback
> starts.
> 
> With this patch, it is much recovered.
> 
>                 base        node          fi
>                    1    global-1         fix
> User           12.98       16.04       13.78
> System        147.61      166.42      143.92
> Elapsed        26.48       38.08       29.64
> 
> Signed-off-by: Minchan Kim <minchan@xxxxxxxxxx>

Thanks. I'll pick this up and send a follow-on series to Andrew with
this included.

-- 
Mel Gorman
SUSE Labs