Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Date:   Thu, 27 Oct 2022 13:49:25 +0800
From:   Feng Tang <feng.tang@intel.com>
To:     "Huang, Ying" <ying.huang@intel.com>
CC:     Andrew Morton <akpm@linux-foundation.org>,
        Johannes Weiner <hannes@cmpxchg.org>,
        "Hocko, Michal" <mhocko@suse.com>, Tejun Heo <tj@kernel.org>,
        Zefan Li <lizefan.x@bytedance.com>,
        Waiman Long <longman@redhat.com>,
        "aneesh.kumar@linux.ibm.com" <aneesh.kumar@linux.ibm.com>,
        "linux-mm@kvack.org" <linux-mm@kvack.org>,
        "cgroups@vger.kernel.org" <cgroups@vger.kernel.org>,
        "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
        "Hansen, Dave" <dave.hansen@intel.com>,
        "Chen, Tim C" <tim.c.chen@intel.com>,
        "Yin, Fengwei" <fengwei.yin@intel.com>
Subject: Re: [PATCH] mm/vmscan: respect cpuset policy during page demotion
Message-ID: <Y1ob5XxqJzTDjBYy@feng-clx>
References: <20221026074343.6517-1-feng.tang@intel.com>
 <878rl1luh1.fsf@yhuang6-desk2.ccr.corp.intel.com>
Content-Type: text/plain; charset="us-ascii"
Content-Disposition: inline
In-Reply-To: <878rl1luh1.fsf@yhuang6-desk2.ccr.corp.intel.com>
MIME-Version: 1.0
X-MS-Exchange-AntiSpam-MessageData-ChunkCount: 1
X-MS-Exchange-AntiSpam-MessageData-0: =?us-ascii?Q?ySezv/6msHsee49n1FguV+bRmn7o6o54Qd7JRbqz+kJVTyPrauFQZ1iLa5pL?=
 =?us-ascii?Q?i56F8MY3v6pgju1+n9iTYbx1AUiE+Hafhm0czAs2y0YPttiXhIfs7gFHvK1y?=
 =?us-ascii?Q?UBxpSnUUv4woeY6HaDspFbEFQLcmhZ3+HVA+fIZ5QNG/S4YoZXYGPQUUZji/?=
 =?us-ascii?Q?EoGaWj4aOjBzq4BzwmwPBOcABNTPnbiAB3ZbPW9NyAQjh81OC0oX8p2BNnZ9?=
 =?us-ascii?Q?zd4gP4wqkcaPK0QDBj/V/R8VEA2J/oGdeXBFHEh+o/QAkJd+SSFbJAO7/SGJ?=
 =?us-ascii?Q?CLk8IppgMfaJ8NXylGxxtnsz2s1FUZGSDxDbKok+w8Vs67U/sOc3SjNh4DEZ?=
 =?us-ascii?Q?0I0/bwQ78ti+rGYT1YJPvgSKPXQFs0LxiXNPpfC210L4/ZwrMdnI3jho4K4r?=
 =?us-ascii?Q?+oB/8GWlxFocWOmrI0ImfN5s/vbWsH+X7xXQuS4ykiMahj/jPGUbmeZ+Plvc?=
 =?us-ascii?Q?7l8dN/p6gCMO+LMzLw3+sBXrVVEjhQwc1JVWvIf5l1FM1cZdc1XsSGfeDGty?=
 =?us-ascii?Q?mv0IMClCHKYsOhDCr5oWuc3Pd7A7Lbe07X6MS9bYuaiuU74MVfnhqgGJIv7b?=
 =?us-ascii?Q?3FbCbRIe0HZnHItQSpVZdeMb3fg/6FJvQ3+xUHGoX5oWCWF3DV5NsClXC4xO?=
 =?us-ascii?Q?qRPpihRfIqCQfYMfMfV0E31JPlqk7jM84WkXyk/VtN5Vb7kngOQkyrdyu6Sk?=
 =?us-ascii?Q?7gUYduk1hxA6ScId4mIOtOrzHyjk4mOgh9Hhci+e0NZHPWc5mhUU9gUKP7ti?=
 =?us-ascii?Q?kfzH4OeYVzQfcp3gGDeiIz0eKtjit05w1dUKni21wY6gnpZUQvzZwgHaUx3P?=
 =?us-ascii?Q?AQweqHV94pmA6P3SpEgkEUv3pD13oZrwoNLhsNOOGdIKoYRRA/fbmbBf+hXb?=
 =?us-ascii?Q?k+AsZERvBg4GLShqjm4QwriH+xH9mn8TUROMfBAkThqTafw9v8Q5yveo5qDW?=
 =?us-ascii?Q?mdJcYKW8bQ6hc1CH43RqSGz1BL8H4CJyfMV9FxKSKIJzVKCiiUsLm0sPsWNg?=
 =?us-ascii?Q?dA1mKpaGMa7sM26Z0uHbZ0Ze54aSQtptRIip1oLR5GuRfcc6e5XMDh1u9+xw?=
 =?us-ascii?Q?4FMNi1jKkj5VkHqon/8cFbmtf3eNc9rXlBwREZiavD3ytlsIBTn1amxPPWZM?=
 =?us-ascii?Q?b1G5s4uduIQjkomEoPsfw6H6IuXQi+/FPTcWyG+fC2ut/oEdNSMNUFDh/Jax?=
 =?us-ascii?Q?7oo+LLv6uxtqF6zs2ax1hgk2H+2hgINXbZ7ZvTrphciU0ajmCei2pXUqpVOy?=
 =?us-ascii?Q?cr7GOliNtaLuFkxAnZukGxLEv4cG0P88YcmN/Hdbe/9DaVTGETPbDvq2z3hS?=
 =?us-ascii?Q?GHFzfE3nj9+IQbnsZbGCNmClAKEMtQ6GYdGyB88KAQgh7csEPpk+JOZiKpWW?=
 =?us-ascii?Q?RpZjJhI6o78oQOXCR3Z13050NUw1IABJNmVedUrc1stVYPW9rYRu/SFoJQVB?=
 =?us-ascii?Q?nPhre7YrHKACace4iTBCZXfzthHehbLRuDCEQhCfI3Dz8jogiV3wHYJJtO7M?=
 =?us-ascii?Q?tIwCtF/xll3VOvD7uH36VZ2MVPV/QP8gQGJZ97lHkCnkALFCj7U08i9CGwGA?=
 =?us-ascii?Q?C1yH2LEuC1WL7ykPmW4zLgcKGeJstdA+sOZTD/98?=
X-MS-Exchange-CrossTenant-Network-Message-Id: ed2ca9e4-bb68-4277-b72b-08dab7df075d
X-MS-Exchange-CrossTenant-AuthSource: MN0PR11MB6304.namprd11.prod.outlook.com
X-MS-Exchange-CrossTenant-AuthAs: Internal
X-MS-Exchange-CrossTenant-OriginalArrivalTime: 27 Oct 2022 05:49:35.6924
 (UTC)
X-MS-Exchange-CrossTenant-FromEntityHeader: Hosted
X-MS-Exchange-CrossTenant-Id: 46c98d88-e344-4ed4-8496-4ed7712e255d
X-MS-Exchange-CrossTenant-MailboxType: HOSTED
X-MS-Exchange-CrossTenant-UserPrincipalName: pFAn4atekJe2fwLF0hj3Fy14XRE15ydEYS+awOle1KI+uFO6cLUI4vWLKv5zvOttBCqKAWMWlq2wDWob2YYhMA==
X-MS-Exchange-Transport-CrossTenantHeadersStamped: CH0PR11MB5316
X-OriginatorOrg: intel.com
X-Spam-Status: No, score=-4.9 required=5.0 tests=BAYES_00,DKIMWL_WL_HIGH,
        DKIM_SIGNED,DKIM_VALID,DKIM_VALID_AU,DKIM_VALID_EF,RCVD_IN_DNSWL_MED,
        SPF_HELO_NONE,SPF_NONE autolearn=ham autolearn_force=no version=3.4.6
X-Spam-Checker-Version: SpamAssassin 3.4.6 (2021-04-09) on
        lindbergh.monkeyblade.net
Precedence: bulk
List-ID: <linux-kernel.vger.kernel.org>
X-Mailing-List: linux-kernel@vger.kernel.org

On Thu, Oct 27, 2022 at 01:13:30PM +0800, Huang, Ying wrote:
> Feng Tang <feng.tang@intel.com> writes:
> 
> > In page reclaim path, memory could be demoted from faster memory tier
> > to slower memory tier. Currently, there is no check about cpuset's
> > memory policy, that even if the target demotion node is not allowd
> > by cpuset, the demotion will still happen, which breaks the cpuset
> > semantics.
> >
> > So add cpuset policy check in the demotion path and skip demotion
> > if the demotion targets are not allowed by cpuset.
> >
> > Signed-off-by: Feng Tang <feng.tang@intel.com>
> > ---
[...]
> > index 18f6497994ec..c205d98283bc 100644
> > --- a/mm/vmscan.c
> > +++ b/mm/vmscan.c
> > @@ -1537,9 +1537,21 @@ static struct page *alloc_demote_page(struct page *page, unsigned long private)
> >  {
> >  	struct page *target_page;
> >  	nodemask_t *allowed_mask;
> > -	struct migration_target_control *mtc;
> > +	struct migration_target_control *mtc = (void *)private;
> >  
> > -	mtc = (struct migration_target_control *)private;
> 
> I think we should avoid (void *) conversion here.

OK, will change back.

> > +#if IS_ENABLED(CONFIG_MEMCG) && IS_ENABLED(CONFIG_CPUSETS)
> > +	struct mem_cgroup *memcg;
> > +	nodemask_t cpuset_nmask;
> > +
> > +	memcg = page_memcg(page);
> > +	cpuset_get_allowed_mem_nodes(memcg->css.cgroup, &cpuset_nmask);
> > +
> > +	if (!node_isset(mtc->nid, cpuset_nmask)) {
> > +		if (mtc->nmask)
> > +			nodes_and(*mtc->nmask, *mtc->nmask, cpuset_nmask);
> > +		return alloc_migration_target(page, (unsigned long)mtc);
> > +	}
> 
> If node_isset(mtc->nid, cpuset_nmask) == true, we should keep the
> original 2 steps allocation and apply nodes_and() on node mask.

Good catch! Yes, the nodes_and() call should be taken out from this
check and done before calling node_isset().

> > +#endif
> >  
> >  	allowed_mask = mtc->nmask;
> >  	/*
> > @@ -1649,6 +1661,7 @@ static unsigned int shrink_folio_list(struct list_head *folio_list,
> >  		enum folio_references references = FOLIOREF_RECLAIM;
> >  		bool dirty, writeback;
> >  		unsigned int nr_pages;
> > +		bool skip_this_demotion = false;
> >  
> >  		cond_resched();
> >  
> > @@ -1658,6 +1671,22 @@ static unsigned int shrink_folio_list(struct list_head *folio_list,
> >  		if (!folio_trylock(folio))
> >  			goto keep;
> >  
> > +#if IS_ENABLED(CONFIG_MEMCG) && IS_ENABLED(CONFIG_CPUSETS)
> > +		if (do_demote_pass) {
> > +			struct mem_cgroup *memcg;
> > +			nodemask_t nmask, nmask1;
> > +
> > +			node_get_allowed_targets(pgdat, &nmask);
> 
> pgdat will not change in the loop, so we can move this out of the loop?
 
Yes

> > +			memcg = folio_memcg(folio);
> > +			if (memcg)
> > +				cpuset_get_allowed_mem_nodes(memcg->css.cgroup,
> > +								&nmask1);
> > +
> > +			if (!nodes_intersects(nmask, nmask1))
> > +				skip_this_demotion = true;
> > +		}
> 
> If nodes_intersects() == true, we will call
> cpuset_get_allowed_mem_nodes() twice.  Better to pass the intersecting
> mask to demote_folio_list()?
 
The pages in the loop may come from different mem control group, and
the cpuset's nodemask could be different, I don't know how to save
this per-page info to be used later in demote_folio_list.

Thanks,
Feng

> > +#endif
> > +
> >  		VM_BUG_ON_FOLIO(folio_test_active(folio), folio);
> >  
> >  		nr_pages = folio_nr_pages(folio);
> > @@ -1799,7 +1828,7 @@ static unsigned int shrink_folio_list(struct list_head *folio_list,
> >  		 * Before reclaiming the folio, try to relocate
> >  		 * its contents to another node.
> >  		 */
> > -		if (do_demote_pass &&
> > +		if (do_demote_pass && !skip_this_demotion &&
> >  		    (thp_migration_supported() || !folio_test_large(folio))) {
> >  			list_add(&folio->lru, &demote_folios);
> >  			folio_unlock(folio);
> 
> Best Regards,
> Huang, Ying