[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <4734132d-bae3-d3ab-33b3-a4c3282ad5cd@huawei.com>
Date: Tue, 19 Apr 2022 22:07:46 +0800
From: Kefeng Wang <wangkefeng.wang@...wei.com>
To: Andrew Morton <akpm@...ux-foundation.org>,
Peng Liu <liupeng256@...wei.com>
CC: <mike.kravetz@...cle.com>, <david@...hat.com>,
<yaozhenguo1@...il.com>, <baolin.wang@...ux.alibaba.com>,
<songmuchun@...edance.com>, <liuyuntao10@...wei.com>,
<linux-mm@...ck.org>, <linux-kernel@...r.kernel.org>,
<dave@...olabs.net>
Subject: Re: [PATCH v4] hugetlb: Fix wrong use of nr_online_nodes
On 2022/4/19 12:03, Andrew Morton wrote:
> On Sat, 16 Apr 2022 10:35:26 +0000 Peng Liu <liupeng256@...wei.com> wrote:
>
>> Certain systems are designed to have sparse/discontiguous nodes. In
>> this case, nr_online_nodes can not be used to walk through numa node.
>> Also, a valid node may be greater than nr_online_nodes.
>>
>> However, in hugetlb, it is assumed that nodes are contiguous. Recheck
>> all the places that use nr_online_nodes, and repair them one by one.
> oops.
>
> What are the user-visible runtime effects of this flaw?
For example, there are four node=0,1,2,3, and nid = 1 is offline
node,nr_online_nodes = 3
1) per-node alloc (hugepages=1:2) fails,
2) per-node alloc (hugepages=3:2) fails, but it could succeed.
I assume that there is no user-visible runtime effects.
> .
Powered by blists - more mailing lists