mbox series

[v1,0/1] Report perpage metadata information.

Message ID 20230913173000.4016218-1-souravpanda@google.com
Headers
Series Report perpage metadata information. |

Message

Sourav Panda Sept. 13, 2023, 5:29 p.m. UTC 
  Hi!

This patch adds a new per-node PageMetadata field to
/sys/devices/system/node/nodeN/meminfo and a global PageMetadata field
to /proc/meminfo. This information can be used by users to see how much
memory is being used by per-page metadata, which can vary depending on
build configuration, machine architecture, and system use.

Per-page metadata is the amount of memory that Linux needs in order to
manage memory at the page granularity. The majority of such memory is
used by "struct page" and "page_ext" data structures.


Background
----------

Kernel overhead observability is missing some of the largest
allocations during runtime, including vmemmap (struct pages) and
page_ext. This patch aims to address this problem by exporting a
new metric PageMetadata.

On the contrary, the kernel does provide observibility for boot memory
allocations. For example, the metric reserved_pages depicts the pages
allocated by the bootmem allocator. This can be simply calculated as
present_pages - managed_pages, which are both exported in /proc/zoneinfo.
The metric reserved_pages is primarily composed of struct pages and
page_ext.

What about the struct pages (allocated by bootmem allocator) that are
free'd during hugetlbfs allocations and then allocated by buddy-allocator
once hugtlbfs pages are free'd?

/proc/meminfo MemTotal changes: MemTotal does not include memblock
allocations but includes buddy allocations. However, during runtime
memblock allocations can be shifted into buddy allocations, and therefore
become part of MemTotal.

Once the struct pages get allocated by buddy allocator, we lose track of
these struct page allocations overhead accounting. Therefore, we must
export a new metric that we shall refer to as PageMetadata (exported by
node). This shall also comprise the struct page and page_ext allocations
made during runtime.

Results and analysis
--------------------

Memory model: Sparsemem-vmemmap
$ echo 1 > /proc/sys/vm/hugetlb_optimize_vmemmap

$ cat /proc/meminfo | grep MemTotal
	MemTotal:       32918196 kB
$ cat /proc/meminfo | grep Meta
	PageMetadata:     589824 kB
$ cat /sys/devices/system/node/node0/meminfo | grep Meta
	Node 0 PageMetadata:     294912 kB
$ cat /sys/devices/system/node/node1/meminfo | grep Meta
	Node 1 PageMetadata:     294912 kB


AFTER HUGTLBFS RESERVATION
$ echo 512 > /proc/sys/vm/nr_hugepages

$ cat /proc/meminfo | grep MemTotal

MemTotal:       32934580 kB
$ cat /proc/meminfo | grep Meta
PageMetadata:     575488 kB
$ cat /sys/devices/system/node/node0/meminfo | grep Meta
Node 0 PageMetadata:     287744 kB
$ cat /sys/devices/system/node/node1/meminfo | grep Meta
Node 1 PageMetadata:     287744 kB

AFTER FREEING HUGTLBFS RESERVATION
$ echo 0 > /proc/sys/vm/nr_hugepages
$ cat /proc/meminfo | grep MemTotal
MemTotal:       32934580 kB
$ cat /proc/meminfo | grep Meta
PageMetadata:    589824 kB
$ cat /sys/devices/system/node/node0/meminfo | grep Meta
Node 0 PageMetadata:       294912 kB
$ cat /sys/devices/system/node/node1/meminfo | grep Meta
Node 1 PageMetadata:       294912 kB

Sourav Panda (1):
  mm: report per-page metadata information

 Documentation/filesystems/proc.rst |  3 +++
 drivers/base/node.c                |  2 ++
 fs/proc/meminfo.c                  |  7 +++++++
 include/linux/mmzone.h             |  3 +++
 include/linux/vmstat.h             |  4 ++++
 mm/hugetlb.c                       |  8 +++++++-
 mm/hugetlb_vmemmap.c               |  9 ++++++++-
 mm/mm_init.c                       |  3 +++
 mm/page_alloc.c                    |  1 +
 mm/page_ext.c                      | 17 +++++++++++++----
 mm/sparse-vmemmap.c                |  3 +++
 mm/sparse.c                        |  7 ++++++-
 mm/vmstat.c                        | 21 +++++++++++++++++++++
 13 files changed, 81 insertions(+), 7 deletions(-)

Comments

Matthew Wilcox Sept. 13, 2023, 5:56 p.m. UTC | #1 
On Wed, Sep 13, 2023 at 10:30:00AM -0700, Sourav Panda wrote:
> @@ -387,8 +390,12 @@ static int alloc_vmemmap_page_list(unsigned long start, unsigned long end,
>  
>  	while (nr_pages--) {
>  		page = alloc_pages_node(nid, gfp_mask, 0);
> -		if (!page)
> +		if (!page) {
>  			goto out;
> +		} else {
> +			__mod_node_page_state(NODE_DATA(page_to_nid(page)),
> +					      NR_PAGE_METADATA, 1);
> +		}
>  		list_add_tail(&page->lru, list);

What a strange way of writing this.  Why not simply:

		if (!page)
			goto out;
+		__mod_node_page_state(NODE_DATA(page_to_nid(page)),
+				NR_PAGE_METADATA, 1);
		list_add_tail(&page->lru, list);

> @@ -314,6 +319,10 @@ static void free_page_ext(void *addr)
>  		BUG_ON(PageReserved(page));
>  		kmemleak_free(addr);
>  		free_pages_exact(addr, table_size);
> +
> +		__mod_node_page_state(NODE_DATA(page_to_nid(page)), NR_PAGE_METADATA,
> +				      (long)-1 * (PAGE_ALIGN(table_size) >> PAGE_SHIFT));

Why not spell that as "-1L"?

And while I'm asking questions, why NODE_DATA(page_to_nid(page)) instead
of page_pgdat(page)?

> @@ -2274,4 +2275,24 @@ static int __init extfrag_debug_init(void)
>  }
>  
>  module_init(extfrag_debug_init);
> +
> +// Page metadata size (struct page and page_ext) in pages

Don't use // comments.

> +void __init writeout_early_perpage_metadata(void)

"writeout" is something swap does.  I'm sure this has a better name,
though I can't think what it might be.

> +{
> +	int nid;
> +	struct pglist_data *pgdat;
> +
> +	for_each_online_pgdat(pgdat) {
> +		nid = pgdat->node_id;
> +		__mod_node_page_state(NODE_DATA(nid), NR_PAGE_METADATA,
> +				      early_perpage_metadata[nid]);
> +	}
> +}
>  #endif
> -- 
> 2.42.0.283.g2d96d420d3-goog
>