On Tue, Mar 18, 2025 at 10:10 PM Yosry Ahmed <yosry.ahmed@xxxxxxxxx> wrote:
>
> On Tue, Mar 18, 2025 at 09:53:30PM +0800, Zhongkun He wrote:
> > With this patch 'commit <68cd9050d871> ("mm: add swappiness= arg to
> > memory.reclaim")', we can submit an additional swappiness=<val> argument
> > to memory.reclaim. It is very useful because we can dynamically adjust
> > the reclamation ratio based on the anonymous folios and file folios of
> > each cgroup. For example,when swappiness is set to 0, we only reclaim
> > from file folios.
> >
> > However,we have also encountered a new issue: when swappiness is set to
> > the MAX_SWAPPINESS, it may still only reclaim file folios.
> >
> > So, we hope to add a new arg 'swappiness=max' in memory.reclaim where
> > proactive memory reclaim only reclaims from anonymous folios when
> > swappiness is set to max. The swappiness semantics from a user
> > perspective remain unchanged.
> >
> > For example, something like this:
> >
> > echo "2M swappiness=max" > /sys/fs/cgroup/memory.reclaim
> >
> > will perform reclaim on the rootcg with a swappiness setting of 'max' (a
> > new mode) regardless of the file folios. Users have a more comprehensive
> > view of the application's memory distribution because there are many
> > metrics available. For example, if we find that a certain cgroup has a
> > large number of inactive anon folios, we can reclaim only those and skip
> > file folios, because with the zram/zswap, the IO tradeoff that
> > cache_trim_mode or other file first logic is making doesn't hold -
> > file refaults will cause IO, whereas anon decompression will not.
> >
> > With this patch, the swappiness argument of memory.reclaim has a new
> > mode 'max', means reclaiming just from anonymous folios both in traditional
> > LRU and MGLRU.
>
> Is MGLRU handled in this patch?
Yes, The value of ONLY_ANON_RECLAIM_MODE is 201, and the MGLRU select the
evictable type like this:
#define evictable_min_seq(min_seq, swappiness) \
min((min_seq)[!(swappiness)], (min_seq)[(swappiness) <= MAX_SWAPPINESS])
#define for_each_evictable_type(type, swappiness) \
for ((type) = !(swappiness); (type) <= ((swappiness) <=
MAX_SWAPPINESS); (type)++)
if the swappiness=0, the type is LRU_GEN_FILE(1);
if the swappiness=201 (>MAX_SWAPPINESS),
for ((type) = 0; (type) <= 0); (type)++)
The type is always LRU_GEN_ANON(0).
>
> >
> > Here is the previous discussion:
> > https://lore.kernel.org/all/20250314033350.1156370-1-hezhongkun.hzk@xxxxxxxxxxxxx/
> > https://lore.kernel.org/all/20250312094337.2296278-1-hezhongkun.hzk@xxxxxxxxxxxxx/
> >
> > Suggested-by: Yosry Ahmed <yosry.ahmed@xxxxxxxxx>
> > Signed-off-by: Zhongkun He <hezhongkun.hzk@xxxxxxxxxxxxx>
> > ---
> > Documentation/admin-guide/cgroup-v2.rst | 4 ++++
> > include/linux/swap.h | 4 ++++
> > mm/memcontrol.c | 5 +++++
> > mm/vmscan.c | 10 ++++++++++
> > 4 files changed, 23 insertions(+)
> >
> > diff --git a/Documentation/admin-guide/cgroup-v2.rst b/Documentation/admin-guide/cgroup-v2.rst
> > index cb1b4e759b7e..c39ef4314499 100644
> > --- a/Documentation/admin-guide/cgroup-v2.rst
> > +++ b/Documentation/admin-guide/cgroup-v2.rst
> > @@ -1343,6 +1343,10 @@ The following nested keys are defined.
> > same semantics as vm.swappiness applied to memcg reclaim with
> > all the existing limitations and potential future extensions.
> >
> > + If set swappiness=max, memory reclamation will exclusively
> > + target the anonymous folio list for both traditional LRU and
> > + MGLRU reclamation algorithms.
> > +
>
> I don't think we need to specify LRU and MGLRU here. What about:
>
> Setting swappiness=max exclusively reclaims anonymous memory.
>
Agree, thanks.
> > memory.peak
> > A read-write single value file which exists on non-root cgroups.
> >
> > diff --git a/include/linux/swap.h b/include/linux/swap.h
> > index b13b72645db3..a94efac10fe5 100644
> > --- a/include/linux/swap.h
> > +++ b/include/linux/swap.h
> > @@ -419,6 +419,10 @@ extern unsigned long try_to_free_pages(struct zonelist *zonelist, int order,
> > #define MEMCG_RECLAIM_PROACTIVE (1 << 2)
> > #define MIN_SWAPPINESS 0
> > #define MAX_SWAPPINESS 200
> > +
> > +/* Just recliam from anon folios in proactive memory reclaim */
> > +#define ONLY_ANON_RECLAIM_MODE (MAX_SWAPPINESS + 1)
> > +
>
> This is a swappiness value so let's keep that clear, e.g.
> SWAPPINESS_ANON_ONLY or similar.
>
OK.
> > extern unsigned long try_to_free_mem_cgroup_pages(struct mem_cgroup *memcg,
> > unsigned long nr_pages,
> > gfp_t gfp_mask,
> > diff --git a/mm/memcontrol.c b/mm/memcontrol.c
> > index 4de6acb9b8ec..0d0400f141d1 100644
> > --- a/mm/memcontrol.c
> > +++ b/mm/memcontrol.c
> > @@ -4291,11 +4291,13 @@ static ssize_t memory_oom_group_write(struct kernfs_open_file *of,
> >
> > enum {
> > MEMORY_RECLAIM_SWAPPINESS = 0,
> > + MEMORY_RECLAIM_ONLY_ANON_MODE,
> > MEMORY_RECLAIM_NULL,
> > };
> >
> > static const match_table_t tokens = {
> > { MEMORY_RECLAIM_SWAPPINESS, "swappiness=%d"},
> > + { MEMORY_RECLAIM_ONLY_ANON_MODE, "swappiness=max"},
>
> MEMORY_RECLAIM_SWAPPINESS_MAX?
>
OK.
> > { MEMORY_RECLAIM_NULL, NULL },
> > };
> >
> > @@ -4329,6 +4331,9 @@ static ssize_t memory_reclaim(struct kernfs_open_file *of, char *buf,
> > if (swappiness < MIN_SWAPPINESS || swappiness > MAX_SWAPPINESS)
> > return -EINVAL;
> > break;
> > + case MEMORY_RECLAIM_ONLY_ANON_MODE:
> > + swappiness = ONLY_ANON_RECLAIM_MODE;
> > + break;
> > default:
> > return -EINVAL;
> > }
> > diff --git a/mm/vmscan.c b/mm/vmscan.c
> > index c767d71c43d7..779a9a3cf715 100644
> > --- a/mm/vmscan.c
> > +++ b/mm/vmscan.c
> > @@ -2438,6 +2438,16 @@ static void get_scan_count(struct lruvec *lruvec, struct scan_control *sc,
> > goto out;
> > }
> >
> > + /*
> > + * Do not bother scanning file folios if the memory reclaim
> > + * invoked by userspace through memory.reclaim and set
> > + * 'swappiness=max'.
> > + */
>
> /* Proactive reclaim initiated by userspace for anonymous memory only */
>
Looks clearer.
> > + if (sc->proactive && (swappiness == ONLY_ANON_RECLAIM_MODE)) {
>
> Do we need to check sc->proactive here? Supposedly this swappiness value
> can only be passed in from proactive reclaim. Instead of silently
> ignoring the value from other paths, I wonder if we should WARN on
> !sc->proactive instead.
>
I'm also hesitating on how to handle this judgment. WARN looks good.
> > + scan_balance = SCAN_ANON;
> > + goto out;
> > + }
> > +
> > /*
> > * Do not apply any pressure balancing cleverness when the
> > * system is close to OOM, scan both anon and file equally
> > --
> > 2.39.5
> >
