[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <07970c9c-9816-650b-26ab-ed9aa46d0653@linux.alibaba.com>
Date: Mon, 7 Feb 2022 09:38:44 +0800
From: Baolin Wang <baolin.wang@...ux.alibaba.com>
To: Huang Ying <ying.huang@...el.com>,
Andrew Morton <akpm@...ux-foundation.org>
Cc: linux-mm@...ck.org, linux-kernel@...r.kernel.org,
Dave Hansen <dave.hansen@...ux.intel.com>,
Zi Yan <ziy@...dia.com>, Oscar Salvador <osalvador@...e.de>,
Yang Shi <shy828301@...il.com>,
zhongjiang-ali <zhongjiang-ali@...ux.alibaba.com>,
Xunlei Pang <xlpang@...ux.alibaba.com>
Subject: Re: [PATCH] mm,migrate: fix establishing demotion target
On 1/28/2022 1:59 PM, Huang Ying wrote:
> In commit ac16ec835314 ("mm: migrate: support multiple target nodes
> demotion"), after the first demotion target node is found, we will
> continue to check the next candidate obtained via
> find_next_best_node(). This is to find all demotion target nodes with
> same NUMA distance. But one side effect of find_next_best_node() is
> that the candidate node returned will be set in "used" parameter, even
> if the candidate node isn't passed in the following NUMA distance
> checking, the candidate node will not be used as demotion target node
> for the following nodes. For example, for system as follows,
>
> node distances:
> node 0 1 2 3
> 0: 10 21 17 28
> 1: 21 10 28 17
> 2: 17 28 10 28
> 3: 28 17 28 10
>
> when we establish demotion target node for node 0, in the first round
> node 2 is added to the demotion target node set. Then in the second
> round, node 3 is checked and failed because distance(0, 3) >
> distance(0, 2). But node 3 is set in "used" nodemask too. When we
> establish demotion target node for node 1, there is no available node.
> This is wrong, node 3 should be set as the demotion target of node 1.
>
> To fix this, if the candidate node is failed to pass the distance
> checking, it will be cleared in "used" nodemask. So that it can be
> used for the following node.
>
> The bug can be reproduced and fixed with this patch on a 2 socket
> server machine with DRAM and PMEM.
>
> Fixes: ac16ec835314 ("mm: migrate: support multiple target nodes demotion")
> Signed-off-by: "Huang, Ying" <ying.huang@...el.com>
> Cc: Baolin Wang <baolin.wang@...ux.alibaba.com>
> Cc: Dave Hansen <dave.hansen@...ux.intel.com>
> Cc: Zi Yan <ziy@...dia.com>
> Cc: Oscar Salvador <osalvador@...e.de>
> Cc: Yang Shi <shy828301@...il.com>
> Cc: Baolin Wang <baolin.wang@...ux.alibaba.com>
> Cc: zhongjiang-ali <zhongjiang-ali@...ux.alibaba.com>
> Cc: Xunlei Pang <xlpang@...ux.alibaba.com>
> ---
Good catch. Thanks.
Reviewed-by: Baolin Wang <baolin.wang@...ux.alibaba.com>
> mm/migrate.c | 7 +++++--
> 1 file changed, 5 insertions(+), 2 deletions(-)
>
> diff --git a/mm/migrate.c b/mm/migrate.c
> index c7da064b4781..e8a6933af68d 100644
> --- a/mm/migrate.c
> +++ b/mm/migrate.c
> @@ -3082,18 +3082,21 @@ static int establish_migrate_target(int node, nodemask_t *used,
> if (best_distance != -1) {
> val = node_distance(node, migration_target);
> if (val > best_distance)
> - return NUMA_NO_NODE;
> + goto out_clear;
> }
>
> index = nd->nr;
> if (WARN_ONCE(index >= DEMOTION_TARGET_NODES,
> "Exceeds maximum demotion target nodes\n"))
> - return NUMA_NO_NODE;
> + goto out_clear;
>
> nd->nodes[index] = migration_target;
> nd->nr++;
>
> return migration_target;
> +out_clear:
> + node_clear(migration_target, *used);
> + return NUMA_NO_NODE;
> }
>
> /*
Powered by blists - more mailing lists