[PATCH 00/25] mm/fbatch: drain lru_add_drain() and _all() · Hugh Dickins <hughd@google.com> · 2026-08-24
[PATCH 01/25] mm/fbatch: remove !CONFIG_SMP special case of folio_activate() · Hugh Dickins <hughd@google.com> · 2026-08-24
Re: [PATCH 01/25] mm/fbatch: remove !CONFIG_SMP special case of folio_activate() · "David Hildenbrand (Arm)" <david@kernel.org> · 2026-08-27
Re: [PATCH 01/25] mm/fbatch: remove !CONFIG_SMP special case of folio_activate() · "Vlastimil Babka (SUSE)" <vbabka@kernel.org> · 2026-09-08
[PATCH 02/25] mm/fbatch: allow folios_put_refs() to skip xa_is_value() entries · Hugh Dickins <hughd@google.com> · 2026-08-24
Re: [PATCH 02/25] mm/fbatch: allow folios_put_refs() to skip xa_is_value() entries · "David Hildenbrand (Arm)" <david@kernel.org> · 2026-08-27
Re: [PATCH 02/25] mm/fbatch: allow folios_put_refs() to skip xa_is_value() entries · "Vlastimil Babka (SUSE)" <vbabka@kernel.org> · 2026-09-08
[PATCH 03/25] mm/fbatch: temporarily disable lazyfree and mlock+munlock batching · Hugh Dickins <hughd@google.com> · 2026-08-24
Re: [PATCH 03/25] mm/fbatch: temporarily disable lazyfree and mlock+munlock batching · "Vlastimil Babka (SUSE)" <vbabka@kernel.org> · 2026-09-08
[PATCH 04/25] mm/fbatch: lru bit set, no extra ref, while folio on per-cpu fbatch · Hugh Dickins <hughd@google.com> · 2026-08-24
Re: [PATCH 04/25] mm/fbatch: lru bit set, no extra ref, while folio on per-cpu fbatch · Kiryl Shutsemau <kas@kernel.org> · 2026-08-27
Re: [PATCH 04/25] mm/fbatch: lru bit set, no extra ref, while folio on per-cpu fbatch · Hugh Dickins <hughd@google.com> · 2026-08-28
Re: [PATCH 04/25] mm/fbatch: lru bit set, no extra ref, while folio on per-cpu fbatch · Matthew Wilcox <willy@infradead.org> · 2026-08-28
Re: [PATCH 04/25] mm/fbatch: lru bit set, no extra ref, while folio on per-cpu fbatch · Hugh Dickins <hughd@google.com> · 2026-08-28
Re: [PATCH 04/25] mm/fbatch: lru bit set, no extra ref, while folio on per-cpu fbatch · Hugh Dickins <hughd@google.com> · 2026-09-02
[PATCH 05/25] mm/fbatch: lru_add_del_folio()+folio_add_lru() after clear_lru() · Hugh Dickins <hughd@google.com> · 2026-08-24
[PATCH 06/25] mm/fbatch: fbatch_drain_lazyfree(onstack fbatch) before ptl unlock · Hugh Dickins <hughd@google.com> · 2026-08-24
[PATCH 07/25] mm/fbatch: LRU_NEXT_ACTIVATE bit to optimize folio_activate() · Hugh Dickins <hughd@google.com> · 2026-08-24
Re: [PATCH 07/25] mm/fbatch: LRU_NEXT_ACTIVATE bit to optimize folio_activate() · Kiryl Shutsemau <kas@kernel.org> · 2026-08-27
Re: [PATCH 07/25] mm/fbatch: LRU_NEXT_ACTIVATE bit to optimize folio_activate() · Hugh Dickins <hughd@google.com> · 2026-08-28
Re: [PATCH 07/25] mm/fbatch: LRU_NEXT_ACTIVATE bit to optimize folio_activate() · Kiryl Shutsemau <kas@kernel.org> · 2026-08-31
Re: [PATCH 07/25] mm/fbatch: LRU_NEXT_ACTIVATE bit to optimize folio_activate() · Hugh Dickins <hughd@google.com> · 2026-09-03
Re: [PATCH 07/25] mm/fbatch: LRU_NEXT_ACTIVATE bit to optimize folio_activate() · Kiryl Shutsemau <kas@kernel.org> · 2026-09-03
Re: [PATCH 07/25] mm/fbatch: LRU_NEXT_ACTIVATE bit to optimize folio_activate() · Hugh Dickins <hughd@google.com> · 2026-09-03
[PATCH 08/25] mm/fbatch: replace mlock_new_folio() by __folio_add_lru(,mlockit) · Hugh Dickins <hughd@google.com> · 2026-08-24
[PATCH 09/25] mm/fbatch: restore mlock+munlock batching, without extra ref · Hugh Dickins <hughd@google.com> · 2026-08-24
[PATCH 10/25] mm/fbatch: remove several uses of mlock_drain_local() · Hugh Dickins <hughd@google.com> · 2026-08-24
[PATCH 11/25] mm/fbatch: remove migration's PAGE_WAS_MLOCKED lru_add_drain() · Hugh Dickins <hughd@google.com> · 2026-08-24
[PATCH 12/25] mm/fbatch: remove percpu_pvec_drained and folios_put() · Hugh Dickins <hughd@google.com> · 2026-08-24
[PATCH 13/25] mm/fbatch: no lru_add_drain() to collect_longterm_unpinnable_folios() · Hugh Dickins <hughd@google.com> · 2026-08-24
[PATCH alt 13/25] mm/fbatch: no lru_add_drain() to collect_longterm_unpinnable_folios() · Hugh Dickins <hughd@google.com> · 2026-08-24
Re: [PATCH alt 13/25] mm/fbatch: no lru_add_drain() to collect_longterm_unpinnable_folios() · "David Hildenbrand (Arm)" <david@kernel.org> · 2026-08-24
Re: [PATCH alt 13/25] mm/fbatch: no lru_add_drain() to collect_longterm_unpinnable_folios() · Hugh Dickins <hughd@google.com> · 2026-08-27
Re: [PATCH alt 13/25] mm/fbatch: no lru_add_drain() to collect_longterm_unpinnable_folios() · "David Hildenbrand (Arm)" <david@kernel.org> · 2026-08-27
[PATCH 14/25] mm/fbatch: no lru_add_drain() nor _all() for memfd_wait_for_pins() · Hugh Dickins <hughd@google.com> · 2026-08-24
[PATCH 15/25] mm/fbatch: remove shake_folio() shake_page() from memory-failure · Hugh Dickins <hughd@google.com> · 2026-08-24
Re: [PATCH 15/25] mm/fbatch: remove shake_folio() shake_page() from memory-failure · Miaohe Lin <linmiaohe@huawei.com> · 2026-09-08
[PATCH 16/25] mm/fbatch: remove lru_cache_disable(() from NUMA folio migration · Hugh Dickins <hughd@google.com> · 2026-08-24
[PATCH 17/25] mm/fbatch: no lru_cache_disable() in __alloc_contig_migrate_range() · Hugh Dickins <hughd@google.com> · 2026-08-24
[PATCH 18/25] mm/fbatch: remove lru_add_drain() and _all() calls from various · Hugh Dickins <hughd@google.com> · 2026-08-24
[PATCH 19/25] mm/fbatch: vm/stat_refresh include lru_add_drain() on each cpu · Hugh Dickins <hughd@google.com> · 2026-08-24
[PATCH 20/25] s390/fbatch: no lru_add_drain_all() in s390_wiggle_split_folio() · Hugh Dickins <hughd@google.com> · 2026-08-24
Re: [PATCH 20/25] s390/fbatch: no lru_add_drain_all() in s390_wiggle_split_folio() · Claudio Imbrenda <imbrenda@linux.ibm.com> · 2026-08-26
Re: [PATCH 20/25] s390/fbatch: no lru_add_drain_all() in s390_wiggle_split_folio() · Hugh Dickins <hughd@google.com> · 2026-08-27
Re: [PATCH 20/25] s390/fbatch: no lru_add_drain_all() in s390_wiggle_split_folio() · Claudio Imbrenda <imbrenda@linux.ibm.com> · 2026-08-27
Re: [PATCH 20/25] s390/fbatch: no lru_add_drain_all() in s390_wiggle_split_folio() · "David Hildenbrand (Arm)" <david@kernel.org> · 2026-08-27
Re: [PATCH 20/25] s390/fbatch: no lru_add_drain_all() in s390_wiggle_split_folio() · Hugh Dickins <hughd@google.com> · 2026-08-28
Re: [PATCH 20/25] s390/fbatch: no lru_add_drain_all() in s390_wiggle_split_folio() · Claudio Imbrenda <imbrenda@linux.ibm.com> · 2026-08-28
Re: [PATCH 20/25] s390/fbatch: no lru_add_drain_all() in s390_wiggle_split_folio() · Hugh Dickins <hughd@google.com> · 2026-08-28
[PATCH 21/25] block/fbatch: no lru_add_drain_all() in invalidate_bdev() · Hugh Dickins <hughd@google.com> · 2026-08-24
[PATCH 22/25] fs/fbatch: drop_caches invalidate_bh_lrus() not lru_add_drain_all() · Hugh Dickins <hughd@google.com> · 2026-08-24
[PATCH 23/25] fs,mm/fbatch: use invalidate_bh_lrus() not invalidate_bh_lrus_cpu() · Hugh Dickins <hughd@google.com> · 2026-08-24
[PATCH 24/25] fs,mm/fbatch: lru_cache_disable() keep off buffer_head lrus only · Hugh Dickins <hughd@google.com> · 2026-08-24
[PATCH 25/25] mm/fbatch: move lru_add_drain_all() declaration to mm/internal.h · Hugh Dickins <hughd@google.com> · 2026-08-24
Re: [PATCH 00/25] mm/fbatch: drain lru_add_drain() and _all() · "David Hildenbrand (Arm)" <david@kernel.org> · 2026-08-27
[PATCH 26/25] mm/fbatch: drop reference inside the loop when draining · Hugh Dickins <hughd@google.com> · 2026-09-02

Re: [PATCH 04/25] mm/fbatch: lru bit set, no extra ref, while folio on per-cpu fbatch

From: Kiryl Shutsemau <kas@kernel.org>
Date: 2026-08-27 11:46:20
Also in: linux-fsdevel, linux-mm, lkml

On Mon, Aug 24, 2026 at 07:01:20AM -0700, Hugh Dickins wrote:

Treat folios on a per-cpu fbatch as if they were already on the lruvec:
with PG_lru set, without holding an extra reference. This will enable
the removal of most lru_add_drain() and lru_add_drain_all() calls soon.

Recognize such a folio by 0x02 set in the folio->lru.next pointer by
folio_add_lru().

Hm. pfmemalloc (__GFP_MEMALLOC) thingy already claims the bit. Is it
safe because such memory is never on LRU?

Are pfmemalloc and PG_lru mutually exclusive?

Do we want to be explicit about this? Like, folio/page_is_pfmemalloc()
shouldn't return true for PG_lru folios/pages or something.

-- 
  Kiryl Shutsemau / Kirill A. Shutemov

`h`	back out one level
`j`	next message in thread
`k`	previous message in thread
`l`	drill in
`Esc`	close help / fold thread tree
`?`	toggle this help