[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <0f4d092d-1421-7163-d937-f8aa681db594@linux.alibaba.com>
Date: Mon, 15 Apr 2019 15:10:00 -0700
From: Yang Shi <yang.shi@...ux.alibaba.com>
To: Dave Hansen <dave.hansen@...el.com>, mhocko@...e.com,
mgorman@...hsingularity.net, riel@...riel.com, hannes@...xchg.org,
akpm@...ux-foundation.org, keith.busch@...el.com,
dan.j.williams@...el.com, fengguang.wu@...el.com, fan.du@...el.com,
ying.huang@...el.com, ziy@...dia.com
Cc: linux-mm@...ck.org, linux-kernel@...r.kernel.org
Subject: Re: [v2 PATCH 5/9] mm: vmscan: demote anon DRAM pages to PMEM node
On 4/11/19 7:31 AM, Dave Hansen wrote:
> On 4/10/19 8:56 PM, Yang Shi wrote:
>> include/linux/gfp.h | 12 ++++
>> include/linux/migrate.h | 1 +
>> include/trace/events/migrate.h | 3 +-
>> mm/debug.c | 1 +
>> mm/internal.h | 13 +++++
>> mm/migrate.c | 15 ++++-
>> mm/vmscan.c | 127 +++++++++++++++++++++++++++++++++++------
>> 7 files changed, 149 insertions(+), 23 deletions(-)
> Yikes, that's a lot of code.
>
> And it only handles anonymous pages?
Yes, for the time being. But, it is easy to extend to all kind of pages.
>
> Also, I don't see anything in the code tying this to strictly demote
> from DRAM to PMEM. Is that the end effect, or is it really implemented
> that way and I missed it?
No, not restrict to PMEM. It just tries to demote from "preferred node"
(or called compute node) to a memory-only node. In the hardware with
PMEM, PMEM would be the memory-only node.
Thanks,
Yang
Powered by blists - more mailing lists