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;
From:   Punit Agrawal <punit.agrawal@bytedance.com>
To:     William Lam <william.lam@bytedance.com>
Cc:     Andrew Morton <akpm@linux-foundation.org>, linux-mm@kvack.org,
        linux-kernel@vger.kernel.org
Subject: Re: [PATCH] mm: compaction: include compound page count for
 scanning in pageblock isolation
In-Reply-To: <20220711202806.22296-1-william.lam@bytedance.com> (William Lam's
        message of "Mon, 11 Jul 2022 21:28:06 +0100")
References: <20220711202806.22296-1-william.lam@bytedance.com>
User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/27.1 (gnu/linux)
Date:   Fri, 15 Jul 2022 10:23:07 +0100
Message-ID: <874jzin30k.fsf@stealth>
MIME-Version: 1.0
Content-Type: text/plain
Precedence: bulk

Hi William,

William Lam <william.lam@bytedance.com> writes:

> The number of scanned pages can be lower than the number of isolated
> pages when isolating mirgratable or free pageblock. The metric is being
> reported in trace event and also used in vmstat.
>
> This behaviour is confusing since currently the count for isolated pages
> takes account of compound page but not for the case of scanned pages.
> And given that the number of isolated pages(nr_taken) reported in
> mm_compaction_isolate_template trace event is on a single-page basis,
> the ambiguity when reporting the number of scanned pages can be removed
> by also including compound page count.

A minor suggestion - It maybe useful to include an example trace output
to highlight the issue.

>
> Signed-off-by: William Lam <william.lam@bytedance.com>
> ---
>  mm/compaction.c | 3 +++
>  1 file changed, 3 insertions(+)
>
> diff --git a/mm/compaction.c b/mm/compaction.c
> index 1f89b969c12b..1b51cf2d32b6 100644
> --- a/mm/compaction.c
> +++ b/mm/compaction.c
> @@ -616,6 +616,7 @@ static unsigned long isolate_freepages_block(struct compact_control *cc,
>  			break;
>  		set_page_private(page, order);
>  
> +		nr_scanned += isolated - 1;
>  		total_isolated += isolated;
>  		cc->nr_freepages += isolated;
>  		list_add_tail(&page->lru, freelist);
> @@ -1101,6 +1102,7 @@ isolate_migratepages_block(struct compact_control *cc, unsigned long low_pfn,
>  isolate_success_no_list:
>  		cc->nr_migratepages += compound_nr(page);
>  		nr_isolated += compound_nr(page);
> +		nr_scanned += compound_nr(page) - 1;
>  
>  		/*
>  		 * Avoid isolating too much unless this block is being
> @@ -1504,6 +1506,7 @@ fast_isolate_freepages(struct compact_control *cc)
>  			if (__isolate_free_page(page, order)) {
>  				set_page_private(page, order);
>  				nr_isolated = 1 << order;
> +				nr_scanned += nr_isolated - 1;
>  				cc->nr_freepages += nr_isolated;
>  				list_add_tail(&page->lru, &cc->freepages);
>  				count_compact_events(COMPACTISOLATED, nr_isolated);

Regardless of the comment above -

Reviewed-by: Punit Agrawal <punit.agrawal@bytedance.com>