Message-ID: <CAOUHufY3s31+7yCzQD_+gdsX-Z+1ef9VTARJTcmd65aemJC8TA@mail.gmail.com>
Date: Sat, 11 Jan 2025 15:12:47 -0700
From: Yu Zhao <yuzhao@...gle.com>
To: Chen Ridong <chenridong@...weicloud.com>, Wei Xu <weixugc@...gle.com>
Cc: akpm@...ux-foundation.org, mhocko@...e.com, hannes@...xchg.org, 
	yosryahmed@...gle.com, david@...hat.com, willy@...radead.org, 
	ryan.roberts@....com, baohua@...nel.org, 21cnbao@...il.com, 
	wangkefeng.wang@...wei.com, linux-mm@...ck.org, linux-kernel@...r.kernel.org, 
	chenridong@...wei.com, wangweiyang2@...wei.com, xieym_ict@...mail.com
Subject: Re: [PATCH v7 mm-unstable] mm: vmscan: retry folios written back
 while isolated for traditional LRU

On Sat, Jan 11, 2025 at 2:25 AM Chen Ridong <chenridong@...weicloud.com> wrote:
>
> From: Chen Ridong <chenridong@...wei.com>
>
> As commit 359a5e1416ca ("mm: multi-gen LRU: retry folios written back
> while isolated") mentioned:
>
>   The page reclaim isolates a batch of folios from the tail of one of the
>   LRU lists and works on those folios one by one.  For a suitable
>   swap-backed folio, if the swap device is async, it queues that folio for
>   writeback.  After the page reclaim finishes an entire batch, it puts back
>   the folios it queued for writeback to the head of the original LRU list.
>
>   In the meantime, the page writeback flushes the queued folios also by
>   batches.  Its batching logic is independent from that of the page
>   reclaim. For each of the folios it writes back, the page writeback calls
>   folio_rotate_reclaimable() which tries to rotate a folio to the tail.
>
>   folio_rotate_reclaimable() only works for a folio after the page reclaim
>   has put it back.  If an async swap device is fast enough, the page
>   writeback can finish with that folio while the page reclaim is still
>   working on the rest of the batch containing it.  In this case, that folio
>   will remain at the head and the page reclaim will not retry it before
>   reaching there".
>
> The commit 359a5e1416ca ("mm: multi-gen LRU: retry folios written back
> while isolated") only fixed the issue for mglru. However, this issue
> also exists in the traditional active/inactive LRU and was found at [1].

The active/inactive LRU needs more careful thoughts due to its
complexity. Details below.

> It can be reproduced with below steps:
>
> 1. Compile with CONFIG_TRANSPARENT_HUGEPAGE=y
> 2. Mount memcg v1, and create memcg named test_memcg and set
>    limit_in_bytes=1G, memsw.limit_in_bytes=2G.
> 3. Create a 1G swap file, and allocate 1.05G anon memory in test_memcg.
>
> It was found that:
>
>   cat memory.limit_in_bytes
>   1073741824
>   cat memory.memsw.limit_in_bytes
>   2147483648
>   cat memory.usage_in_bytes
>   1073664000
>   cat memory.memsw.usage_in_bytes
>   1129840640
>
>   free -h
>                 total        used        free
>   Mem:           31Gi       1.2Gi        28Gi
>   Swap:         1.0Gi       1.0Gi       2.0Mi
>
> As shown above, the test_memcg used about 50M swap, but almost 1G swap
> memory was used, which means that 900M+ may be wasted because other memcgs
> can not use these swap memory.
>
> This issue should be fixed in the same way as mglru. Therefore, the common
> logic was extracted to the 'find_folios_written_back' function firstly,
> which is then reused in the 'shrink_inactive_list' function. Finally,
> retry reclaiming those folios that may have missed the rotation for
> traditional LRU.
>
> After change, the same test case. only 54M swap was used.
>
>   cat memory.usage_in_bytes
>   1073463296
>   cat memory.memsw.usage_in_bytes
>   1129828352
>
>   free -h
>                 total        used        free
>   Mem:           31Gi       1.2Gi        28Gi
>   Swap:         1.0Gi        54Mi       969Mi
>
> [1] https://lore.kernel.org/linux-kernel/20241010081802.290893-1-chenridong@huaweicloud.com/
> [2] https://lore.kernel.org/linux-kernel/CAGsJ_4zqL8ZHNRZ44o_CC69kE7DBVXvbZfvmQxMGiFqRxqHQdA@mail.gmail.com/
> Signed-off-by: Chen Ridong <chenridong@...wei.com>
> ---
>
> v6->v7:
>  - fix conflict based on mm-unstable.
>  - update the commit message(quote from YU's commit message, and add
>    improvements after change.)
>  - restore 'is_retrying' to 'skip_retry' to keep original semantics.
>
> v6: https://lore.kernel.org/linux-kernel/20241223082004.3759152-1-chenridong@huaweicloud.com/
>
>  mm/vmscan.c | 114 ++++++++++++++++++++++++++++++++++------------------
>  1 file changed, 76 insertions(+), 38 deletions(-)
>
> diff --git a/mm/vmscan.c b/mm/vmscan.c
> index 01dce6f26..6861b6937 100644
> --- a/mm/vmscan.c
> +++ b/mm/vmscan.c
> @@ -183,6 +183,9 @@ struct scan_control {
>         struct reclaim_state reclaim_state;
>  };
>
> +static inline void find_folios_written_back(struct list_head *list,
> +               struct list_head *clean, struct lruvec *lruvec, int type, bool is_retrying);
> +
>  #ifdef ARCH_HAS_PREFETCHW
>  #define prefetchw_prev_lru_folio(_folio, _base, _field)                        \
>         do {                                                            \
> @@ -1960,14 +1963,18 @@ static unsigned long shrink_inactive_list(unsigned long nr_to_scan,
>                 enum lru_list lru)
>  {
>         LIST_HEAD(folio_list);
> +       LIST_HEAD(clean_list);
>         unsigned long nr_scanned;
> -       unsigned int nr_reclaimed = 0;
> +       unsigned int nr_reclaimed, total_reclaimed = 0;
> +       unsigned int nr_pageout = 0;
> +       unsigned int nr_unqueued_dirty = 0;
>         unsigned long nr_taken;
>         struct reclaim_stat stat;
>         bool file = is_file_lru(lru);
>         enum vm_event_item item;
>         struct pglist_data *pgdat = lruvec_pgdat(lruvec);
>         bool stalled = false;
> +       bool skip_retry = false;
>
>         while (unlikely(too_many_isolated(pgdat, file, sc))) {
>                 if (stalled)
> @@ -2001,22 +2008,47 @@ static unsigned long shrink_inactive_list(unsigned long nr_to_scan,
>         if (nr_taken == 0)
>                 return 0;
>
> +retry:
>         nr_reclaimed = shrink_folio_list(&folio_list, pgdat, sc, &stat, false);
>
> +       sc->nr.dirty += stat.nr_dirty;
> +       sc->nr.congested += stat.nr_congested;
> +       sc->nr.unqueued_dirty += stat.nr_unqueued_dirty;
> +       sc->nr.writeback += stat.nr_writeback;

I think this change breaks the tests on the stats above, e.g.,
wakeup_flusher_threads(), because the same dirty/writeback folio can
be counted twice. The reason for that is that
folio_test_dirty/writeback() can't account for dirty/writeback buffer
heads, which can only be done by folio_check_dirty_writeback().

For MGLRU, it has been broken since day 1 and commit 1bc542c6a0d1
("mm/vmscan: wake up flushers conditionally to avoid cgroup OOM")
doesn't account for this either. I'll get around to that.

> +       sc->nr.immediate += stat.nr_immediate;
> +       total_reclaimed += nr_reclaimed;
> +       nr_pageout += stat.nr_pageout;
> +       nr_unqueued_dirty += stat.nr_unqueued_dirty;
> +
> +       trace_mm_vmscan_lru_shrink_inactive(pgdat->node_id,
> +                       nr_scanned, nr_reclaimed, &stat, sc->priority, file);
> +
> +       find_folios_written_back(&folio_list, &clean_list, lruvec, 0, skip_retry);
> +
>         spin_lock_irq(&lruvec->lru_lock);
>         move_folios_to_lru(lruvec, &folio_list);
>
>         __mod_lruvec_state(lruvec, PGDEMOTE_KSWAPD + reclaimer_offset(),
>                                         stat.nr_demoted);
> -       __mod_node_page_state(pgdat, NR_ISOLATED_ANON + file, -nr_taken);
>         item = PGSTEAL_KSWAPD + reclaimer_offset();
>         if (!cgroup_reclaim(sc))
>                 __count_vm_events(item, nr_reclaimed);
>         __count_memcg_events(lruvec_memcg(lruvec), item, nr_reclaimed);
>         __count_vm_events(PGSTEAL_ANON + file, nr_reclaimed);
> +
> +       if (!list_empty(&clean_list)) {
> +               list_splice_init(&clean_list, &folio_list);
> +               skip_retry = true;
> +               spin_unlock_irq(&lruvec->lru_lock);
> +               goto retry;
> +       }
> +       __mod_node_page_state(pgdat, NR_ISOLATED_ANON + file, -nr_taken);
>         spin_unlock_irq(&lruvec->lru_lock);
> +       sc->nr.taken += nr_taken;
> +       if (file)
> +               sc->nr.file_taken += nr_taken;
>
> -       lru_note_cost(lruvec, file, stat.nr_pageout, nr_scanned - nr_reclaimed);
> +       lru_note_cost(lruvec, file, nr_pageout, nr_scanned - total_reclaimed);
>
>         /*
>          * If dirty folios are scanned that are not queued for IO, it
> @@ -2029,7 +2061,7 @@ static unsigned long shrink_inactive_list(unsigned long nr_to_scan,
>          * the flushers simply cannot keep up with the allocation
>          * rate. Nudge the flusher threads in case they are asleep.
>          */
> -       if (stat.nr_unqueued_dirty == nr_taken) {
> +       if (nr_unqueued_dirty == nr_taken) {
>                 wakeup_flusher_threads(WB_REASON_VMSCAN);
>                 /*
>                  * For cgroupv1 dirty throttling is achieved by waking up
> @@ -2044,18 +2076,7 @@ static unsigned long shrink_inactive_list(unsigned long nr_to_scan,
>                         reclaim_throttle(pgdat, VMSCAN_THROTTLE_WRITEBACK);
>         }
>
> -       sc->nr.dirty += stat.nr_dirty;
> -       sc->nr.congested += stat.nr_congested;
> -       sc->nr.unqueued_dirty += stat.nr_unqueued_dirty;
> -       sc->nr.writeback += stat.nr_writeback;
> -       sc->nr.immediate += stat.nr_immediate;
> -       sc->nr.taken += nr_taken;
> -       if (file)
> -               sc->nr.file_taken += nr_taken;
> -
> -       trace_mm_vmscan_lru_shrink_inactive(pgdat->node_id,
> -                       nr_scanned, nr_reclaimed, &stat, sc->priority, file);
> -       return nr_reclaimed;
> +       return total_reclaimed;
>  }
>
>  /*
> @@ -4637,8 +4658,6 @@ static int evict_folios(struct lruvec *lruvec, struct scan_control *sc, int swap
>         int reclaimed;
>         LIST_HEAD(list);
>         LIST_HEAD(clean);
> -       struct folio *folio;
> -       struct folio *next;
>         enum vm_event_item item;
>         struct reclaim_stat stat;
>         struct lru_gen_mm_walk *walk;
> @@ -4668,26 +4687,7 @@ static int evict_folios(struct lruvec *lruvec, struct scan_control *sc, int swap
>                         scanned, reclaimed, &stat, sc->priority,
>                         type ? LRU_INACTIVE_FILE : LRU_INACTIVE_ANON);
>
> -       list_for_each_entry_safe_reverse(folio, next, &list, lru) {
> -               DEFINE_MIN_SEQ(lruvec);
> -
> -               if (!folio_evictable(folio)) {
> -                       list_del(&folio->lru);
> -                       folio_putback_lru(folio);
> -                       continue;
> -               }
> -
> -               /* retry folios that may have missed folio_rotate_reclaimable() */
> -               if (!skip_retry && !folio_test_active(folio) && !folio_mapped(folio) &&
> -                   !folio_test_dirty(folio) && !folio_test_writeback(folio)) {
> -                       list_move(&folio->lru, &clean);
> -                       continue;
> -               }
> -
> -               /* don't add rejected folios to the oldest generation */
> -               if (lru_gen_folio_seq(lruvec, folio, false) == min_seq[type])
> -                       set_mask_bits(&folio->flags, LRU_REFS_FLAGS, BIT(PG_active));
> -       }
> +       find_folios_written_back(&list, &clean, lruvec, type, skip_retry);
>
>         spin_lock_irq(&lruvec->lru_lock);
>
> @@ -5706,6 +5706,44 @@ static void lru_gen_shrink_node(struct pglist_data *pgdat, struct scan_control *
>
>  #endif /* CONFIG_LRU_GEN */
>
> +/**
> + * find_folios_written_back - Find and move the written back folios to a new list.
> + * @list: filios list
> + * @clean: the written back folios list
> + * @lruvec: the lruvec
> + * @type: LRU_GEN_ANON/LRU_GEN_FILE, only for multi-gen LRU
> + * @skip_retry: whether skip retry.
> + */
> +static inline void find_folios_written_back(struct list_head *list,
> +               struct list_head *clean, struct lruvec *lruvec, int type, bool skip_retry)
> +{
> +       struct folio *folio;
> +       struct folio *next;
> +
> +       list_for_each_entry_safe_reverse(folio, next, list, lru) {
> +#ifdef CONFIG_LRU_GEN
> +               DEFINE_MIN_SEQ(lruvec);
> +#endif
> +               if (!folio_evictable(folio)) {
> +                       list_del(&folio->lru);
> +                       folio_putback_lru(folio);
> +                       continue;
> +               }
> +
> +               /* retry folios that may have missed folio_rotate_reclaimable() */
> +               if (!skip_retry && !folio_test_active(folio) && !folio_mapped(folio) &&
> +                   !folio_test_dirty(folio) && !folio_test_writeback(folio)) {

Have you verified that this condition also holds for the
active/inactive LRU or did you just assume it? IOW, how do we know the
active/inactive LRU doesn't think this folio should be kept (and put
back to the head of the inactive LRU list).

Powered by blists - more mailing lists