[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <a4f1212f-3903-abbc-772a-1ddee6f7f98b@huawei.com>
Date: Tue, 7 Nov 2017 19:23:52 +0800
From: Yisheng Xie <xieyisheng1@...wei.com>
To: Christopher Lameter <cl@...ux.com>,
Vlastimil Babka <vbabka@...e.cz>
CC: <akpm@...ux-foundation.org>, <mhocko@...e.com>, <mingo@...nel.org>,
<rientjes@...gle.com>, <n-horiguchi@...jp.nec.com>,
<salls@...ucsb.edu>, <linux-mm@...ck.org>,
<linux-kernel@...r.kernel.org>, <tanxiaojun@...wei.com>,
<linux-api@...r.kernel.org>, Andi Kleen <ak@...ux.intel.com>
Subject: Re: [PATCH RFC v2 4/4] mm/mempolicy: add nodes_empty check in
SYSC_migrate_pages
hi Christopher and Vlastimil,
Thanks for your comment!
On 2017/11/6 23:29, Christopher Lameter wrote:
> On Mon, 6 Nov 2017, Vlastimil Babka wrote:
>
>> I'm not sure what exactly is the EPERM intention. Should really the
>> capability of THIS process override the cpuset restriction of the TARGET
>> process? Maybe yes. Then, does "insufficient privilege (CAP_SYS_NICE) to
>
> CAP_SYS_NICE never overrides cpuset restrictions. The cap can be used to
> migrate pages that are *also* mapped by other processes (and thus move
> pages of another process which may have different cpu set restrictions!).
So you means the specified nodes should be a subset of target cpu set, right?
Thanks
Yisheng Xie
> The cap should not allow migrating pages to nodes that are not allowed by
> the cpuset of the current process.
>
>
>
> .
>
Powered by blists - more mailing lists