Re: [PATCH -mm] mm: more likely reclaim MADV_SEQUENTIAL mappings

[KOSAKI Motohiro]

Hi Andrew,

>> in my experience,
>>   - page_referenced_one is performance critical point.
>>     you should test some benchmark.
>>   - its patch improved mmaped-copy performance about 5%.
>>     (Of cource, you should test in current -mm. MM code was changed widely)
>>
>> So, I'm looking for your test result :)
>
> The change seems logical and I queued it for 2.6.28.

Great.

> But yes, testing for what-does-this-improve is good and useful, but so
> is testing for what-does-this-worsen.  How do we do that in this case?

In general, page_referenced_one is important for reclaim throuput.
if crap page_referenced_one changing happend,
system reclaim throuput become slow down.

Of cource, I don't think this patch cause performance regression :-)

So, any benchmark with memcgroup memory restriction is good choice.

btw:
maybe, I will able to post mamped-copy improve mesurement of Johannes's patch
after OLS.
--
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/