* Re: [PATCH mm-new] mm/migrate: move NR_KERNEL_FILE_PAGES on cross-node migration
2026-10-27 4:32 [PATCH mm-new] mm/migrate: move NR_KERNEL_FILE_PAGES on cross-node migration Ridong Chen
@ 2026-10-27 4:59 ` Garg, Shivank
2026-10-27 14:49 ` Zi Yan
2026-10-28 8:36 ` David Hildenbrand (Arm)
2 siblings, 0 replies; 4+ messages in thread
From: Garg, Shivank @ 2026-10-27 4:59 UTC (permalink / raw)
To: david@kernel.org, ridong.chen@linux.dev,
akpm@linux-foundation.org
Cc: chenridong@xiaomi.com, shakeel.butt@linux.dev, apopple@nvidia.com,
ziy@nvidia.com, gourry@gourry.net, linux-mm@kvack.org,
matthew.brost@intel.com, joshua.hahnjy@gmail.com,
ying.huang@linux.alibaba.com, boris@bur.io, byungchul@sk.com,
linux-kernel@vger.kernel.org
On Tue, 2026-10-27 at 12:32 +0800, Ridong Chen wrote:
> From: Ridong Chen <chenridong@xiaomi.com>
>
> NR_KERNEL_FILE_PAGES counts page cache folios of mappings flagged
> AS_KERNEL_FILE (currently the btrfs btree inode). It is a per-node
> counter: filemap_add_folio() adds to folio_pgdat() of the new folio and
> filemap_unaccount_folio() subtracts from folio_pgdat() of the folio
> being removed.
>
> __folio_migrate_mapping() does not move it when such a folio is
> migrated to another node, so the source node keeps the count forever
> and the destination node underflows once the folio is freed. btree
> folios are migratable through btree_migrate_folio(), so demotion,
> DAMON-based migration (DAMOS_MIGRATE_{HOT,COLD}) or memory offlining
> all hit this.
>
> Reproducer, on a QEMU guest with two memory tiers (node0 is DRAM with
> all CPUs, node1 is a CPU-less slower node described via HMAT) and
> /sys/kernel/mm/numa/demotion_enabled set:
>
> 1. mount a btrfs and create 20000 empty files, so that btree folios
> are cached on node0;
> 2. echo 2G > /sys/devices/system/node/node0/reclaim, which demotes
> them to node1;
> 3. umount and drop caches.
>
> nr_kernel_file_pages in /sys/devices/system/node/nodeN/vmstat after
> each step:
>
> without this patch with this patch
> node0 node1 node0 node1
> 1. cached 1868 0 1868 0
> 2. demoted 1868 0 36 1832
> 3. dropped 1832 0 (*) 0 0
>
> (*) the counter has underflowed to -1832; negative values read as 0.
>
> The global value in /proc/vmstat is 0 after step 3 in both cases.
>
> Move it together with the other node-keyed page cache counters. It is
> a node-only counter, so use the node accessor, matching how it is
> accounted elsewhere. Interrupts are disabled here, as for the zone
> counters below.
>
> This was pointed out by Shivank while reviewing the NR_FILE_THPS fix
> for the same function.
>
> Fixes: e3a9ac4e866e ("mm: add vmstat for kernel_file pages")
> Suggested-by: Shivank Garg <shivankg@amd.com>
> Assisted-by: LLM
> Signed-off-by: Ridong Chen <chenridong@xiaomi.com>
> ---
> mm/migrate.c | 6 ++++++
> 1 file changed, 6 insertions(+)
>
> diff --git a/mm/migrate.c b/mm/migrate.c
> index 99e3026e3126..f829c5017c96 100644
> --- a/mm/migrate.c
> +++ b/mm/migrate.c
> @@ -705,6 +705,12 @@ static int __folio_migrate_mapping(struct address_space *mapping,
> mod_lruvec_state(old_lruvec, NR_FILE_THPS, -nr);
> mod_lruvec_state(new_lruvec, NR_FILE_THPS, nr);
> }
> + if (test_bit(AS_KERNEL_FILE, &mapping->flags)) {
> + __mod_node_page_state(oldzone->zone_pgdat,
> + NR_KERNEL_FILE_PAGES, -nr);
> + __mod_node_page_state(newzone->zone_pgdat,
> + NR_KERNEL_FILE_PAGES, nr);
> + }
> #ifdef CONFIG_SWAP
> if (folio_test_swapcache(folio)) {
> mod_lruvec_state(old_lruvec, NR_SWAPCACHE, -nr);
LGTM
Reviewed-by: Shivank Garg <shivankg@amd.com>
Thanks,
Shivank
^ permalink raw reply [flat|nested] 4+ messages in thread* Re: [PATCH mm-new] mm/migrate: move NR_KERNEL_FILE_PAGES on cross-node migration
2026-10-27 4:32 [PATCH mm-new] mm/migrate: move NR_KERNEL_FILE_PAGES on cross-node migration Ridong Chen
2026-10-27 4:59 ` Garg, Shivank
@ 2026-10-27 14:49 ` Zi Yan
2026-10-28 8:36 ` David Hildenbrand (Arm)
2 siblings, 0 replies; 4+ messages in thread
From: Zi Yan @ 2026-10-27 14:49 UTC (permalink / raw)
To: Ridong Chen
Cc: Andrew Morton, David Hildenbrand, Matthew Brost, Joshua Hahn,
Byungchul Park, Gregory Price, Ying Huang, Alistair Popple,
Boris Burkov, Shakeel Butt, linux-mm, linux-kernel, Shivank Garg,
Ridong Chen
On 27 Oct 2026, at 0:32, Ridong Chen wrote:
> From: Ridong Chen <chenridong@xiaomi.com>
>
> NR_KERNEL_FILE_PAGES counts page cache folios of mappings flagged
> AS_KERNEL_FILE (currently the btrfs btree inode). It is a per-node
> counter: filemap_add_folio() adds to folio_pgdat() of the new folio and
> filemap_unaccount_folio() subtracts from folio_pgdat() of the folio
> being removed.
>
> __folio_migrate_mapping() does not move it when such a folio is
> migrated to another node, so the source node keeps the count forever
> and the destination node underflows once the folio is freed. btree
> folios are migratable through btree_migrate_folio(), so demotion,
> DAMON-based migration (DAMOS_MIGRATE_{HOT,COLD}) or memory offlining
> all hit this.
>
> Reproducer, on a QEMU guest with two memory tiers (node0 is DRAM with
> all CPUs, node1 is a CPU-less slower node described via HMAT) and
> /sys/kernel/mm/numa/demotion_enabled set:
>
> 1. mount a btrfs and create 20000 empty files, so that btree folios
> are cached on node0;
> 2. echo 2G > /sys/devices/system/node/node0/reclaim, which demotes
> them to node1;
> 3. umount and drop caches.
>
> nr_kernel_file_pages in /sys/devices/system/node/nodeN/vmstat after
> each step:
>
> without this patch with this patch
> node0 node1 node0 node1
> 1. cached 1868 0 1868 0
> 2. demoted 1868 0 36 1832
> 3. dropped 1832 0 (*) 0 0
>
> (*) the counter has underflowed to -1832; negative values read as 0.
>
> The global value in /proc/vmstat is 0 after step 3 in both cases.
>
> Move it together with the other node-keyed page cache counters. It is
> a node-only counter, so use the node accessor, matching how it is
> accounted elsewhere. Interrupts are disabled here, as for the zone
> counters below.
>
> This was pointed out by Shivank while reviewing the NR_FILE_THPS fix
> for the same function.
>
> Fixes: e3a9ac4e866e ("mm: add vmstat for kernel_file pages")
> Suggested-by: Shivank Garg <shivankg@amd.com>
> Assisted-by: LLM
> Signed-off-by: Ridong Chen <chenridong@xiaomi.com>
> ---
> mm/migrate.c | 6 ++++++
> 1 file changed, 6 insertions(+)
>
LGTM.
Acked-by: Zi Yan <ziy@nvidia.com>
Best Regards,
Yan, Zi
^ permalink raw reply [flat|nested] 4+ messages in thread* Re: [PATCH mm-new] mm/migrate: move NR_KERNEL_FILE_PAGES on cross-node migration
2026-10-27 4:32 [PATCH mm-new] mm/migrate: move NR_KERNEL_FILE_PAGES on cross-node migration Ridong Chen
2026-10-27 4:59 ` Garg, Shivank
2026-10-27 14:49 ` Zi Yan
@ 2026-10-28 8:36 ` David Hildenbrand (Arm)
2 siblings, 0 replies; 4+ messages in thread
From: David Hildenbrand (Arm) @ 2026-10-28 8:36 UTC (permalink / raw)
To: Ridong Chen, Andrew Morton
Cc: Zi Yan, Matthew Brost, Joshua Hahn, Byungchul Park, Gregory Price,
Ying Huang, Alistair Popple, Boris Burkov, Shakeel Butt, linux-mm,
linux-kernel, Shivank Garg, Ridong Chen
On 10/27/26 05:32, Ridong Chen wrote:
> From: Ridong Chen <chenridong@xiaomi.com>
>
> NR_KERNEL_FILE_PAGES counts page cache folios of mappings flagged
> AS_KERNEL_FILE (currently the btrfs btree inode). It is a per-node
> counter: filemap_add_folio() adds to folio_pgdat() of the new folio and
> filemap_unaccount_folio() subtracts from folio_pgdat() of the folio
> being removed.
>
> __folio_migrate_mapping() does not move it when such a folio is
> migrated to another node, so the source node keeps the count forever
> and the destination node underflows once the folio is freed. btree
> folios are migratable through btree_migrate_folio(), so demotion,
> DAMON-based migration (DAMOS_MIGRATE_{HOT,COLD}) or memory offlining
> all hit this.
>
> Reproducer, on a QEMU guest with two memory tiers (node0 is DRAM with
> all CPUs, node1 is a CPU-less slower node described via HMAT) and
> /sys/kernel/mm/numa/demotion_enabled set:
>
> 1. mount a btrfs and create 20000 empty files, so that btree folios
> are cached on node0;
> 2. echo 2G > /sys/devices/system/node/node0/reclaim, which demotes
> them to node1;
> 3. umount and drop caches.
>
> nr_kernel_file_pages in /sys/devices/system/node/nodeN/vmstat after
> each step:
>
> without this patch with this patch
> node0 node1 node0 node1
> 1. cached 1868 0 1868 0
> 2. demoted 1868 0 36 1832
> 3. dropped 1832 0 (*) 0 0
>
> (*) the counter has underflowed to -1832; negative values read as 0.
>
> The global value in /proc/vmstat is 0 after step 3 in both cases.
>
> Move it together with the other node-keyed page cache counters. It is
> a node-only counter, so use the node accessor, matching how it is
> accounted elsewhere. Interrupts are disabled here, as for the zone
> counters below.
>
> This was pointed out by Shivank while reviewing the NR_FILE_THPS fix
> for the same function.
>
> Fixes: e3a9ac4e866e ("mm: add vmstat for kernel_file pages")
> Suggested-by: Shivank Garg <shivankg@amd.com>
> Assisted-by: LLM
> Signed-off-by: Ridong Chen <chenridong@xiaomi.com>
> ---
> mm/migrate.c | 6 ++++++
> 1 file changed, 6 insertions(+)
>
> diff --git a/mm/migrate.c b/mm/migrate.c
> index 99e3026e3126..f829c5017c96 100644
> --- a/mm/migrate.c
> +++ b/mm/migrate.c
> @@ -705,6 +705,12 @@ static int __folio_migrate_mapping(struct address_space *mapping,
> mod_lruvec_state(old_lruvec, NR_FILE_THPS, -nr);
> mod_lruvec_state(new_lruvec, NR_FILE_THPS, nr);
> }
> + if (test_bit(AS_KERNEL_FILE, &mapping->flags)) {
> + __mod_node_page_state(oldzone->zone_pgdat,
> + NR_KERNEL_FILE_PAGES, -nr);
> + __mod_node_page_state(newzone->zone_pgdat,
> + NR_KERNEL_FILE_PAGES, nr);
> + }
> #ifdef CONFIG_SWAP
> if (folio_test_swapcache(folio)) {
> mod_lruvec_state(old_lruvec, NR_SWAPCACHE, -nr);
Today I learned about NR_KERNEL_FILE_PAGES :D
Acked-by: David Hildenbrand (Arm) <david@kernel.org>
--
Cheers,
David
^ permalink raw reply [flat|nested] 4+ messages in thread