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;
Message-ID: <c9384e52-3f3b-cb1b-a607-955cd7066422@redhat.com>
Date:   Tue, 13 Jun 2023 11:56:02 +0200
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:102.0) Gecko/20100101
 Thunderbird/102.12.0
Content-Language: en-US
To:     Baolin Wang <baolin.wang@linux.alibaba.com>,
        akpm@linux-foundation.org
Cc:     mgorman@techsingularity.net, vbabka@suse.cz, ying.huang@intel.com,
        linux-mm@kvack.org, linux-kernel@vger.kernel.org
References: <770f9f61472b24b6bc89adbd71a77d9cf62bb54f.1686646361.git.baolin.wang@linux.alibaba.com>
From:   David Hildenbrand <david@redhat.com>
Organization: Red Hat
Subject: Re: [PATCH v2] mm: compaction: skip memory hole rapidly when
 isolating migratable pages
In-Reply-To: <770f9f61472b24b6bc89adbd71a77d9cf62bb54f.1686646361.git.baolin.wang@linux.alibaba.com>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 7bit
Precedence: bulk

On 13.06.23 10:55, Baolin Wang wrote:
> On some machines, the normal zone can have a large memory hole like
> below memory layout, and we can see the range from 0x100000000 to
> 0x1800000000 is a hole. So when isolating some migratable pages, the
> scanner can meet the hole and it will take more time to skip the large
> hole. From my measurement, I can see the isolation scanner will take
> 80us ~ 100us to skip the large hole [0x100000000 - 0x1800000000].
> 
> So adding a new helper to fast search next online memory section
> to skip the large hole can help to find next suitable pageblock
> efficiently. With this patch, I can see the large hole scanning only
> takes < 1us.
> 
> [    0.000000] Zone ranges:
> [    0.000000]   DMA      [mem 0x0000000040000000-0x00000000ffffffff]
> [    0.000000]   DMA32    empty
> [    0.000000]   Normal   [mem 0x0000000100000000-0x0000001fa7ffffff]
> [    0.000000] Movable zone start for each node
> [    0.000000] Early memory node ranges
> [    0.000000]   node   0: [mem 0x0000000040000000-0x0000000fffffffff]
> [    0.000000]   node   0: [mem 0x0000001800000000-0x0000001fa3c7ffff]
> [    0.000000]   node   0: [mem 0x0000001fa3c80000-0x0000001fa3ffffff]
> [    0.000000]   node   0: [mem 0x0000001fa4000000-0x0000001fa402ffff]
> [    0.000000]   node   0: [mem 0x0000001fa4030000-0x0000001fa40effff]
> [    0.000000]   node   0: [mem 0x0000001fa40f0000-0x0000001fa73cffff]
> [    0.000000]   node   0: [mem 0x0000001fa73d0000-0x0000001fa745ffff]
> [    0.000000]   node   0: [mem 0x0000001fa7460000-0x0000001fa746ffff]
> [    0.000000]   node   0: [mem 0x0000001fa7470000-0x0000001fa758ffff]
> [    0.000000]   node   0: [mem 0x0000001fa7590000-0x0000001fa7ffffff]
> 
> Signed-off-by: Baolin Wang <baolin.wang@linux.alibaba.com>
> ---
> Changes from v1:
>   - Fix building errors if CONFIG_SPARSEMEM is not selected.
>   - Use NR_MEM_SECTIONS instead of '-1' per Huang Ying.
> ---
>   include/linux/mmzone.h | 10 ++++++++++
>   mm/compaction.c        | 30 +++++++++++++++++++++++++++++-
>   2 files changed, 39 insertions(+), 1 deletion(-)
> 
> diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
> index 5a7ada0413da..5ff1fa2efe28 100644
> --- a/include/linux/mmzone.h
> +++ b/include/linux/mmzone.h
> @@ -2000,6 +2000,16 @@ static inline unsigned long next_present_section_nr(unsigned long section_nr)
>   	return -1;
>   }
>   
> +static inline unsigned long next_online_section_nr(unsigned long section_nr)
> +{
> +	while (++section_nr <= __highest_present_section_nr) {
> +		if (online_section_nr(section_nr))
> +			return section_nr;
> +	}
> +
> +	return NR_MEM_SECTIONS;
> +}
> +
>   /*
>    * These are _only_ used during initialisation, therefore they
>    * can use __initdata ...  They could have names to indicate
> diff --git a/mm/compaction.c b/mm/compaction.c
> index 3398ef3a55fe..c31ff6123891 100644
> --- a/mm/compaction.c
> +++ b/mm/compaction.c
> @@ -229,6 +229,28 @@ static void reset_cached_positions(struct zone *zone)
>   				pageblock_start_pfn(zone_end_pfn(zone) - 1);
>   }
>   
> +#ifdef CONFIG_SPARSEMEM
> +static unsigned long skip_hole_pageblock(unsigned long start_pfn)
> +{
> +	unsigned long next_online_nr;
> +	unsigned long start_nr = pfn_to_section_nr(start_pfn);
> +
> +	if (online_section_nr(start_nr))
> +		return 0;
> +
> +	next_online_nr = next_online_section_nr(start_nr);
> +	if (next_online_nr < NR_MEM_SECTIONS)
> +		return section_nr_to_pfn(next_online_nr);
> +

I would simply inline next_online_section_nr and simplify (and add a 
comment):

/*
  * If the PFN falls into an offline section, return the start PFN of the
  * next online section. If the PFN falls into an online section or if
  * there is no next online section, return 0.
  */
static unsigned long skip_hole_pageblock(unsigned long start_pfn)
{
	unsigned long nr = pfn_to_section_nr(start_pfn);

	if (online_section_nr(nr))
		return 0;

	while (++nr <= __highest_present_section_nr) {
		if (online_section_nr(nr))
			return section_nr_to_pfn(nr);
	}
	return 0
}

Easier, no?

And maybe just call that function "skip_offline_sections()" then? 
Because we're not operating on pageblocks.

> +	return 0;
> +}
> +#else
> +static unsigned long skip_hole_pageblock(unsigned long start_pfn)
> +{
> +	return 0;
> +}
> +#endif
> +
>   /*
>    * Compound pages of >= pageblock_order should consistently be skipped until
>    * released. It is always pointless to compact pages of such order (if they are
> @@ -1991,8 +2013,14 @@ static isolate_migrate_t isolate_migratepages(struct compact_control *cc)
>   
>   		page = pageblock_pfn_to_page(block_start_pfn,
>   						block_end_pfn, cc->zone);
> -		if (!page)
> +		if (!page) {
> +			unsigned long next_pfn;
> +
> +			next_pfn = skip_hole_pageblock(block_start_pfn);
> +			if (next_pfn != 0)

if (next_pfn)

> +				block_end_pfn = next_pfn;
>   			continue;
> +		}
>   
>   		/*
>   		 * If isolation recently failed, do not retry. Only check the

-- 
Cheers,

David / dhildenb