Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Date:   Tue, 25 Feb 2020 17:32:24 -0800 (PST)
From:   David Rientjes <rientjes@google.com>
To:     Mel Gorman <mgorman@techsingularity.net>
cc:     Andrew Morton <akpm@linux-foundation.org>,
        Michal Hocko <mhocko@suse.com>,
        Vlastimil Babka <vbabka@suse.cz>,
        Ivan Babrou <ivan@cloudflare.com>,
        Rik van Riel <riel@surriel.com>,
        Linux-MM <linux-mm@kvack.org>,
        Linux Kernel Mailing List <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH 2/3] mm, page_alloc: Disable watermark boosting if THP
 is disabled at boot
In-Reply-To: <20200225141534.5044-3-mgorman@techsingularity.net>
Message-ID: <alpine.DEB.2.21.2002251728520.14021@chino.kir.corp.google.com>
References: <20200225141534.5044-1-mgorman@techsingularity.net> <20200225141534.5044-3-mgorman@techsingularity.net>
User-Agent: Alpine 2.21 (DEB 202 2017-01-01)
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Tue, 25 Feb 2020, Mel Gorman wrote:

> Watermark boosting is intended to increase the success rate and reduce
> latency of high-order allocations, particularly THP. If THP is disabled
> at boot, then it makes sense to disable watermark boosting as well. While
> there are other high-order allocations that potentially benefit, they
> are relatively rare.
> 
> Signed-off-by: Mel Gorman <mgorman@techsingularity.net>
> ---
>  mm/huge_memory.c | 1 +
>  1 file changed, 1 insertion(+)
> 
> diff --git a/mm/huge_memory.c b/mm/huge_memory.c
> index b08b199f9a11..565bb9973ff8 100644
> --- a/mm/huge_memory.c
> +++ b/mm/huge_memory.c
> @@ -472,6 +472,7 @@ static int __init setup_transparent_hugepage(char *str)
>  			  &transparent_hugepage_flags);
>  		clear_bit(TRANSPARENT_HUGEPAGE_REQ_MADV_FLAG,
>  			  &transparent_hugepage_flags);
> +		disable_watermark_boosting();
>  		ret = 1;
>  	}
>  out:

Seems like watermark boosting can help prevent fragmentation so it 
benefits all hugepage sized allocations for the long term and that would 
include dynamic provisioning of hugetlb memory or hugetlb overcommit?