Add page migration support via swap to the NUMA policy layer
[original mail had a mistyped linux-mm address]
This is the Postfix program at host mx2.suse.de.
I'm sorry to have to inform you that your message could not be be delivered to one or more recipients. It's attached below.
For further assistance, please send mail to <postmaster>
If you do so, please include this problem report. You can delete your own text from the attached returned message.
The Postfix program
<[email protected]>: host vger.kernel.org[209.132.176.167] said: 554
5.0.0 Hi [195.135.220.15], unresolvable address:
<[email protected]>; nosuchuser; [email protected] (in reply to RCPT TO command)
Quoted 4 lines · Christoph Lameter
This patch adds page migration support to the NUMA policy layer. An additional flag
MPOL_MF_MOVEis introduced for mbind. IfMPOL_MF_MOVEis specified then pages that do not conform to the memory policy will be evicted from memory. When they get pages back in new pages will be allocated following the numa policy.
That part looks ok.
Quoted 5 lines
In addition this also adds a
move_pagesfunction that may be used from outside of the policy layer to move pages between nodes (needed by the cpuset support and the /proc interface). The design is intended to support future direct page migration without going through swap space.
Please split that out and resubmit if there are really other users
(what /proc support?)
Review
Quoted 1 line
(quoted diff) +1
-0+ WARN_ON(isolate_lru_page(page, pagelist) == 0);
WARN_ONs are not supposed to have side effects.
-Andi