Re: [PATCH v2 resend] mm/memory_hotplug: Make unpopulated zones PCP structures unreachable during hot remove

From: Michal Hocko <mhocko@suse.com>
To: Mel Gorman <mgorman@techsingularity.net>
Cc: Andrew Morton <akpm@linux-foundation.org>,
	Oscar Salvador <osalvador@suse.de>,
	"Michael S. Tsirkin" <mst@redhat.com>,
	Vlastimil Babka <vbabka@suse.cz>,
	Alexander Duyck <alexanderduyck@fb.com>,
	Minchan Kim <minchan@kernel.org>,
	David Hildenbrand <david@redhat.com>,
	Linux-MM <linux-mm@kvack.org>,
	LKML <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH v2 resend] mm/memory_hotplug: Make unpopulated zones PCP structures unreachable during hot remove
Date: Tue, 13 Apr 2021 08:40:30 +0200	[thread overview]
Message-ID: <YHU83jawLzKu+N2P@dhcp22.suse.cz> (raw)
In-Reply-To: <20210412120842.GY3697@techsingularity.net>

On Mon 12-04-21 13:08:42, Mel Gorman wrote:
> zone_pcp_reset allegedly protects against a race with drain_pages
> using local_irq_save but this is bogus. local_irq_save only operates
> on the local CPU. If memory hotplug is running on CPU A and drain_pages
> is running on CPU B, disabling IRQs on CPU A does not affect CPU B and
> offers no protection.
> 
> This patch deletes IRQ disable/enable on the grounds that IRQs protect
> nothing and assumes the existing hotplug paths guarantees the PCP cannot be
> used after zone_pcp_enable(). That should be the case already because all
> the pages have been freed and there is no page to put on the PCP lists.

Yes, that is the case since ec6e8c7e0314 ("mm, page_alloc: disable
pcplists during memory offline"). Prior to this commit the behavior was
undefined but full zone/node hotremove is rare enough that an existing
race was likely never observed.

Acked-by: Michal Hocko <mhocko@suse.com>

Thanks!

> Signed-off-by: Mel Gorman <mgorman@techsingularity.net>
> ---
> Resending for email address correction and adding lists
> 
> Changelog since v1
> o Minimal fix
> 
>  mm/page_alloc.c | 4 ----
>  1 file changed, 4 deletions(-)
> 
> diff --git a/mm/page_alloc.c b/mm/page_alloc.c
> index 5e8aedb64b57..9bf0db982f14 100644
> --- a/mm/page_alloc.c
> +++ b/mm/page_alloc.c
> @@ -8952,12 +8952,9 @@ void zone_pcp_enable(struct zone *zone)
>  
>  void zone_pcp_reset(struct zone *zone)
>  {
> -	unsigned long flags;
>  	int cpu;
>  	struct per_cpu_pageset *pset;
>  
> -	/* avoid races with drain_pages()  */
> -	local_irq_save(flags);
>  	if (zone->pageset != &boot_pageset) {
>  		for_each_online_cpu(cpu) {
>  			pset = per_cpu_ptr(zone->pageset, cpu);
> @@ -8966,7 +8963,6 @@ void zone_pcp_reset(struct zone *zone)
>  		free_percpu(zone->pageset);
>  		zone->pageset = &boot_pageset;
>  	}
> -	local_irq_restore(flags);
>  }
>  
>  #ifdef CONFIG_MEMORY_HOTREMOVE

-- 
Michal Hocko
SUSE Labs