From: Roman Gushchin <guro@fb.com>
To: Muchun Song <songmuchun@bytedance.com>
Cc: <gregkh@linuxfoundation.org>, <rafael@kernel.org>,
<adobriyan@gmail.com>, <akpm@linux-foundation.org>,
<hannes@cmpxchg.org>, <mhocko@kernel.org>,
<vdavydov.dev@gmail.com>, <hughd@google.com>, <will@kernel.org>,
<rppt@kernel.org>, <tglx@linutronix.de>, <esyr@redhat.com>,
<peterx@redhat.com>, <krisman@collabora.com>, <surenb@google.com>,
<avagin@openvz.org>, <elver@google.com>, <rdunlap@infradead.org>,
<iamjoonsoo.kim@lge.com>, <linux-kernel@vger.kernel.org>,
<linux-fsdevel@vger.kernel.org>, <linux-mm@kvack.org>,
<cgroups@vger.kernel.org>
Subject: Re: [RESEND PATCH v2 09/12] mm: memcontrol: convert vmstat slab counters to bytes
Date: Mon, 7 Dec 2020 11:46:22 -0800 [thread overview]
Message-ID: <20201207194622.GA2238414@carbon.dhcp.thefacebook.com> (raw)
In-Reply-To: <20201206101451.14706-10-songmuchun@bytedance.com>
On Sun, Dec 06, 2020 at 06:14:48PM +0800, Muchun Song wrote:
> the global and per-node counters are stored in pages, however memcg
> and lruvec counters are stored in bytes. This scheme looks weird.
> So convert all vmstat slab counters to bytes.
There is a reason for this weird scheme:
percpu caches (see struct per_cpu_nodestat) are s8, so counting in bytes
will lead to overfills. Switching to s32 can lead to an increase in
the cache thrashing, especially on small machines.
>
> Signed-off-by: Muchun Song <songmuchun@bytedance.com>
> ---
> include/linux/vmstat.h | 17 ++++++++++-------
> mm/vmstat.c | 21 ++++++++++-----------
> 2 files changed, 20 insertions(+), 18 deletions(-)
>
> diff --git a/include/linux/vmstat.h b/include/linux/vmstat.h
> index 322dcbfcc933..fd1a3d5d4926 100644
> --- a/include/linux/vmstat.h
> +++ b/include/linux/vmstat.h
> @@ -197,18 +197,26 @@ static inline
> unsigned long global_node_page_state_pages(enum node_stat_item item)
> {
> long x = atomic_long_read(&vm_node_stat[item]);
> +
> #ifdef CONFIG_SMP
> if (x < 0)
> x = 0;
> #endif
> + if (vmstat_item_in_bytes(item))
> + x >>= PAGE_SHIFT;
> return x;
> }
>
> static inline unsigned long global_node_page_state(enum node_stat_item item)
> {
> - VM_WARN_ON_ONCE(vmstat_item_in_bytes(item));
> + long x = atomic_long_read(&vm_node_stat[item]);
>
> - return global_node_page_state_pages(item);
> + VM_WARN_ON_ONCE(vmstat_item_in_bytes(item));
> +#ifdef CONFIG_SMP
> + if (x < 0)
> + x = 0;
> +#endif
> + return x;
> }
>
> static inline unsigned long zone_page_state(struct zone *zone,
> @@ -312,11 +320,6 @@ static inline void __mod_zone_page_state(struct zone *zone,
> static inline void __mod_node_page_state(struct pglist_data *pgdat,
> enum node_stat_item item, int delta)
> {
> - if (vmstat_item_in_bytes(item)) {
> - VM_WARN_ON_ONCE(delta & (PAGE_SIZE - 1));
> - delta >>= PAGE_SHIFT;
> - }
> -
> node_page_state_add(delta, pgdat, item);
> }
>
> diff --git a/mm/vmstat.c b/mm/vmstat.c
> index 8d77ee426e22..7fb0c7cb9516 100644
> --- a/mm/vmstat.c
> +++ b/mm/vmstat.c
> @@ -345,11 +345,6 @@ void __mod_node_page_state(struct pglist_data *pgdat, enum node_stat_item item,
> long x;
> long t;
>
> - if (vmstat_item_in_bytes(item)) {
> - VM_WARN_ON_ONCE(delta & (PAGE_SIZE - 1));
> - delta >>= PAGE_SHIFT;
> - }
> -
> x = delta + __this_cpu_read(*p);
>
> t = __this_cpu_read(pcp->stat_threshold);
> @@ -554,11 +549,6 @@ static inline void mod_node_state(struct pglist_data *pgdat,
> s8 __percpu *p = pcp->vm_node_stat_diff + item;
> long o, n, t, z;
>
> - if (vmstat_item_in_bytes(item)) {
> - VM_WARN_ON_ONCE(delta & (PAGE_SIZE - 1));
> - delta >>= PAGE_SHIFT;
> - }
> -
> do {
> z = 0; /* overflow to node counters */
>
> @@ -1012,19 +1002,28 @@ unsigned long node_page_state_pages(struct pglist_data *pgdat,
> enum node_stat_item item)
> {
> long x = atomic_long_read(&pgdat->vm_stat[item]);
> +
> #ifdef CONFIG_SMP
> if (x < 0)
> x = 0;
> #endif
> + if (vmstat_item_in_bytes(item))
> + x >>= PAGE_SHIFT;
> return x;
> }
>
> unsigned long node_page_state(struct pglist_data *pgdat,
> enum node_stat_item item)
> {
> + long x = atomic_long_read(&pgdat->vm_stat[item]);
> +
> VM_WARN_ON_ONCE(vmstat_item_in_bytes(item));
>
> - return node_page_state_pages(pgdat, item);
> +#ifdef CONFIG_SMP
> + if (x < 0)
> + x = 0;
> +#endif
> + return x;
> }
> #endif
>
> --
> 2.11.0
>
next prev parent reply other threads:[~2020-12-07 19:46 UTC|newest]
Thread overview: 32+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-12-06 10:14 [RESEND PATCH v2 00/12] Convert all vmstat counters to pages or bytes Muchun Song
2020-12-06 10:14 ` [RESEND PATCH v2 01/12] mm: memcontrol: fix NR_ANON_THPS account Muchun Song
2020-12-07 12:52 ` Michal Hocko
2020-12-10 16:00 ` Johannes Weiner
2020-12-10 16:02 ` Johannes Weiner
2020-12-10 16:04 ` Johannes Weiner
2020-12-10 16:56 ` [External] " Muchun Song
2020-12-06 10:14 ` [RESEND PATCH v2 02/12] mm: memcontrol: convert NR_ANON_THPS account to pages Muchun Song
2020-12-10 16:10 ` Johannes Weiner
2020-12-10 17:05 ` [External] " Muchun Song
2020-12-06 10:14 ` [RESEND PATCH v2 03/12] mm: memcontrol: convert NR_FILE_THPS " Muchun Song
2020-12-06 10:14 ` [RESEND PATCH v2 04/12] mm: memcontrol: convert NR_SHMEM_THPS " Muchun Song
2020-12-06 10:14 ` [RESEND PATCH v2 05/12] mm: memcontrol: convert NR_SHMEM_PMDMAPPED " Muchun Song
2020-12-06 10:14 ` [RESEND PATCH v2 06/12] mm: memcontrol: convert NR_FILE_PMDMAPPED " Muchun Song
2020-12-06 10:14 ` [RESEND PATCH v2 07/12] mm: memcontrol: convert kernel stack account to bytes Muchun Song
2020-12-06 10:14 ` [RESEND PATCH v2 08/12] mm: memcontrol: convert NR_KERNEL_SCS_KB " Muchun Song
2020-12-06 10:14 ` [RESEND PATCH v2 09/12] mm: memcontrol: convert vmstat slab counters " Muchun Song
2020-12-07 19:46 ` Roman Gushchin [this message]
2020-12-07 20:01 ` Roman Gushchin
2020-12-08 2:44 ` [External] " Muchun Song
2020-12-06 10:14 ` [RESEND PATCH v2 10/12] mm: memcontrol: scale stat_threshold for byted-sized vmstat Muchun Song
2020-12-06 10:14 ` [RESEND PATCH v2 11/12] mm: memcontrol: make the slab calculation consistent Muchun Song
2020-12-06 10:14 ` [RESEND PATCH v2 12/12] mm: memcontrol: remove {global_}node_page_state_pages Muchun Song
2020-12-07 13:00 ` [RESEND PATCH v2 00/12] Convert all vmstat counters to pages or bytes Michal Hocko
2020-12-07 14:52 ` [External] " Muchun Song
2020-12-07 15:02 ` Michal Hocko
2020-12-07 18:51 ` Randy Dunlap
2020-12-08 2:29 ` Muchun Song
2020-12-07 19:51 ` Roman Gushchin
2020-12-07 20:33 ` Hugh Dickins
2020-12-08 2:42 ` Muchun Song
2020-12-08 2:40 ` Muchun Song
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20201207194622.GA2238414@carbon.dhcp.thefacebook.com \
--to=guro@fb.com \
--cc=adobriyan@gmail.com \
--cc=akpm@linux-foundation.org \
--cc=avagin@openvz.org \
--cc=cgroups@vger.kernel.org \
--cc=elver@google.com \
--cc=esyr@redhat.com \
--cc=gregkh@linuxfoundation.org \
--cc=hannes@cmpxchg.org \
--cc=hughd@google.com \
--cc=iamjoonsoo.kim@lge.com \
--cc=krisman@collabora.com \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mhocko@kernel.org \
--cc=peterx@redhat.com \
--cc=rafael@kernel.org \
--cc=rdunlap@infradead.org \
--cc=rppt@kernel.org \
--cc=songmuchun@bytedance.com \
--cc=surenb@google.com \
--cc=tglx@linutronix.de \
--cc=vdavydov.dev@gmail.com \
--cc=will@kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).