Re: [PATCH 13/40] autonuma: CPU follow memory algorithm

On 2012a1'06ae??30ae?JPY 21:04, Andrea Arcangeli wrote:

On Sat, Jun 30, 2012 at 02:58:29PM +0800, Nai Xia wrote:

quoted

OK, I think I'd stop discussing this topic now. Without strict and comprehensive
research on this topic, further arguments seems to me to be purely based on
imagination.

I suggest to consider how ptep_clear_and_test_young works on the
pte_young bit on the VM swapping code. Then apply your "concern" to
the pte_young bit scan. If you can't NAK the swapping code in the
kernel, well I guess you can't nack AutoNUMA as well because of that
specific concern.

And no I'm not saying this is trivial or obvious, I appreciate your
thoughts a lot, just I'm quite convinced this is a subtle detail but
an irrelevant one that gets lost in the noise.

quoted

If you insist on ignoring any constructive suggestions from others,
it's pretty much ok to do so.  But I (and possibly many others who are
watching)
am pretty much  possible to do a LOL to your development style.

Well if you think answering your emails means ignoring your
suggestions, be my guest.

Thanks for you kindness, I was just feeling a little bit uncomfortable
about Dor's tune of "Anyone can beat anything". It's OK now, since
anyone with eyes can easily find out I am not a spamming person on
this list.


Thanks,
Nai

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>