From 2da6de30e60dd9bb14600eff1cc99df2fa2ddae3 Mon Sep 17 00:00:00 2001 From: Hugh Dickins Date: Mon, 8 Sep 2025 15:23:15 -0700 Subject: mm: folio_may_be_lru_cached() unless folio_test_large() mm/swap.c and mm/mlock.c agree to drain any per-CPU batch as soon as a large folio is added: so collect_longterm_unpinnable_folios() just wastes effort when calling lru_add_drain[_all]() on a large folio. But although there is good reason not to batch up PMD-sized folios, we might well benefit from batching a small number of low-order mTHPs (though unclear how that "small number" limitation will be implemented). So ask if folio_may_be_lru_cached() rather than !folio_test_large(), to insulate those particular checks from future change. Name preferred to "folio_is_batchable" because large folios can well be put on a batch: it's just the per-CPU LRU caches, drained much later, which need care. Marked for stable, to counter the increase in lru_add_drain_all()s from "mm/gup: check ref_count instead of lru before migration". Link: https://lkml.kernel.org/r/57d2eaf8-3607-f318-e0c5-be02dce61ad0@google.com Fixes: 9a4e9f3b2d73 ("mm: update get_user_pages_longterm to migrate pages allocated from CMA region") Signed-off-by: Hugh Dickins Suggested-by: David Hildenbrand Acked-by: David Hildenbrand Cc: "Aneesh Kumar K.V" Cc: Axel Rasmussen Cc: Chris Li Cc: Christoph Hellwig Cc: Jason Gunthorpe Cc: Johannes Weiner Cc: John Hubbard Cc: Keir Fraser Cc: Konstantin Khlebnikov Cc: Li Zhe Cc: Matthew Wilcox (Oracle) Cc: Peter Xu Cc: Rik van Riel Cc: Shivank Garg Cc: Vlastimil Babka Cc: Wei Xu Cc: Will Deacon Cc: yangge Cc: Yuanchu Xie Cc: Yu Zhao Cc: Signed-off-by: Andrew Morton --- include/linux/swap.h | 10 ++++++++++ 1 file changed, 10 insertions(+) (limited to 'include/linux') diff --git a/include/linux/swap.h b/include/linux/swap.h index 2fe6ed2cc3fd..7012a0f758d8 100644 --- a/include/linux/swap.h +++ b/include/linux/swap.h @@ -385,6 +385,16 @@ void folio_add_lru_vma(struct folio *, struct vm_area_struct *); void mark_page_accessed(struct page *); void folio_mark_accessed(struct folio *); +static inline bool folio_may_be_lru_cached(struct folio *folio) +{ + /* + * Holding PMD-sized folios in per-CPU LRU cache unbalances accounting. + * Holding small numbers of low-order mTHP folios in per-CPU LRU cache + * will be sensible, but nobody has implemented and tested that yet. + */ + return !folio_test_large(folio); +} + extern atomic_t lru_disable_count; static inline bool lru_cache_disabled(void) -- cgit v1.2.3 From e6a0deb6fa5b0fc134ee2aa127d1cfc9456d8445 Mon Sep 17 00:00:00 2001 From: SeongJae Park Date: Mon, 8 Sep 2025 13:15:12 -0700 Subject: mm/damon/core: introduce damon_call_control->dealloc_on_cancel Patch series "mm/damon/sysfs: fix refresh_ms control overwriting on multi-kdamonds usages". Automatic esssential DAMON/DAMOS status update feature of DAMON sysfs interface (refresh_ms) is broken [1] for multiple DAMON contexts (kdamonds) use case, since it uses a global single damon_call_control object for all created DAMON contexts. The fields of the object, particularly the list field is over-written for the contexts and it makes unexpected results including user-space hangup and kernel crashes [2]. Fix it by extending damon_call_control for the use case and updating the usage on DAMON sysfs interface to use per-context dynamically allocated damon_call_control object. This patch (of 2): When damon_call_control->repeat is set, damon_call() is executed asynchronously, and is eventually canceled when kdamond finishes. If the damon_call_control object is dynamically allocated, finding the place to deallocate the object is difficult. Introduce a new damon_call_control field, namely dealloc_on_cancel, to ask the kdamond deallocates those dynamically allocated objects when those are canceled. Link: https://lkml.kernel.org/r/20250908201513.60802-3-sj@kernel.org Link: https://lkml.kernel.org/r/20250908201513.60802-2-sj@kernel.org Fixes: d809a7c64ba8 ("mm/damon/sysfs: implement refresh_ms file internal work") Signed-off-by: SeongJae Park Cc: Yunjeong Mun Signed-off-by: Andrew Morton --- include/linux/damon.h | 2 ++ mm/damon/core.c | 8 ++++++-- 2 files changed, 8 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/damon.h b/include/linux/damon.h index f13664c62ddd..9e62b2a85538 100644 --- a/include/linux/damon.h +++ b/include/linux/damon.h @@ -636,6 +636,7 @@ struct damon_operations { * @data: Data that will be passed to @fn. * @repeat: Repeat invocations. * @return_code: Return code from @fn invocation. + * @dealloc_on_cancel: De-allocate when canceled. * * Control damon_call(), which requests specific kdamond to invoke a given * function. Refer to damon_call() for more details. @@ -645,6 +646,7 @@ struct damon_call_control { void *data; bool repeat; int return_code; + bool dealloc_on_cancel; /* private: internal use only */ /* informs if the kdamond finished handling of the request */ struct completion completion; diff --git a/mm/damon/core.c b/mm/damon/core.c index c2e0b469fd43..08065b363972 100644 --- a/mm/damon/core.c +++ b/mm/damon/core.c @@ -2479,10 +2479,14 @@ static void kdamond_call(struct damon_ctx *ctx, bool cancel) mutex_lock(&ctx->call_controls_lock); list_del(&control->list); mutex_unlock(&ctx->call_controls_lock); - if (!control->repeat) + if (!control->repeat) { complete(&control->completion); - else + } else if (control->canceled && control->dealloc_on_cancel) { + kfree(control); + continue; + } else { list_add(&control->list, &repeat_controls); + } } control = list_first_entry_or_null(&repeat_controls, struct damon_call_control, list); -- cgit v1.2.3 From 6b4be64fd9fec16418f365c2d8e47a7566e9eba5 Mon Sep 17 00:00:00 2001 From: Jianbo Liu Date: Mon, 15 Sep 2025 15:24:32 +0300 Subject: net/mlx5e: Harden uplink netdev access against device unbind The function mlx5_uplink_netdev_get() gets the uplink netdevice pointer from mdev->mlx5e_res.uplink_netdev. However, the netdevice can be removed and its pointer cleared when unbound from the mlx5_core.eth driver. This results in a NULL pointer, causing a kernel panic. BUG: unable to handle page fault for address: 0000000000001300 at RIP: 0010:mlx5e_vport_rep_load+0x22a/0x270 [mlx5_core] Call Trace: mlx5_esw_offloads_rep_load+0x68/0xe0 [mlx5_core] esw_offloads_enable+0x593/0x910 [mlx5_core] mlx5_eswitch_enable_locked+0x341/0x420 [mlx5_core] mlx5_devlink_eswitch_mode_set+0x17e/0x3a0 [mlx5_core] devlink_nl_eswitch_set_doit+0x60/0xd0 genl_family_rcv_msg_doit+0xe0/0x130 genl_rcv_msg+0x183/0x290 netlink_rcv_skb+0x4b/0xf0 genl_rcv+0x24/0x40 netlink_unicast+0x255/0x380 netlink_sendmsg+0x1f3/0x420 __sock_sendmsg+0x38/0x60 __sys_sendto+0x119/0x180 do_syscall_64+0x53/0x1d0 entry_SYSCALL_64_after_hwframe+0x4b/0x53 Ensure the pointer is valid before use by checking it for NULL. If it is valid, immediately call netdev_hold() to take a reference, and preventing the netdevice from being freed while it is in use. Fixes: 7a9fb35e8c3a ("net/mlx5e: Do not reload ethernet ports when changing eswitch mode") Signed-off-by: Jianbo Liu Reviewed-by: Cosmin Ratiu Reviewed-by: Jiri Pirko Reviewed-by: Dragos Tatulea Signed-off-by: Tariq Toukan Link: https://patch.msgid.link/1757939074-617281-2-git-send-email-tariqt@nvidia.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/mellanox/mlx5/core/en_rep.c | 27 ++++++++++++++++++---- drivers/net/ethernet/mellanox/mlx5/core/esw/qos.c | 1 + drivers/net/ethernet/mellanox/mlx5/core/lib/mlx5.h | 15 +++++++++++- include/linux/mlx5/driver.h | 1 + 4 files changed, 38 insertions(+), 6 deletions(-) (limited to 'include/linux') diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c index 63a7a788fb0d..cd0242eb008c 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c @@ -1506,12 +1506,21 @@ static const struct mlx5e_profile mlx5e_uplink_rep_profile = { static int mlx5e_vport_uplink_rep_load(struct mlx5_core_dev *dev, struct mlx5_eswitch_rep *rep) { - struct mlx5e_priv *priv = netdev_priv(mlx5_uplink_netdev_get(dev)); struct mlx5e_rep_priv *rpriv = mlx5e_rep_to_rep_priv(rep); + struct net_device *netdev; + struct mlx5e_priv *priv; + int err; + + netdev = mlx5_uplink_netdev_get(dev); + if (!netdev) + return 0; + priv = netdev_priv(netdev); rpriv->netdev = priv->netdev; - return mlx5e_netdev_change_profile(priv, &mlx5e_uplink_rep_profile, - rpriv); + err = mlx5e_netdev_change_profile(priv, &mlx5e_uplink_rep_profile, + rpriv); + mlx5_uplink_netdev_put(dev, netdev); + return err; } static void @@ -1638,8 +1647,16 @@ mlx5e_vport_rep_unload(struct mlx5_eswitch_rep *rep) { struct mlx5e_rep_priv *rpriv = mlx5e_rep_to_rep_priv(rep); struct net_device *netdev = rpriv->netdev; - struct mlx5e_priv *priv = netdev_priv(netdev); - void *ppriv = priv->ppriv; + struct mlx5e_priv *priv; + void *ppriv; + + if (!netdev) { + ppriv = rpriv; + goto free_ppriv; + } + + priv = netdev_priv(netdev); + ppriv = priv->ppriv; if (rep->vport == MLX5_VPORT_UPLINK) { mlx5e_vport_uplink_rep_unload(rpriv); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/esw/qos.c b/drivers/net/ethernet/mellanox/mlx5/core/esw/qos.c index 8b4977650183..5f2d6c35f1ad 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/esw/qos.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/esw/qos.c @@ -1515,6 +1515,7 @@ static u32 mlx5_esw_qos_lag_link_speed_get_locked(struct mlx5_core_dev *mdev) speed = lksettings.base.speed; out: + mlx5_uplink_netdev_put(mdev, slave); return speed; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lib/mlx5.h b/drivers/net/ethernet/mellanox/mlx5/core/lib/mlx5.h index b111ccd03b02..74ea5da58b7e 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/lib/mlx5.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/lib/mlx5.h @@ -47,7 +47,20 @@ int mlx5_crdump_collect(struct mlx5_core_dev *dev, u32 *cr_data); static inline struct net_device *mlx5_uplink_netdev_get(struct mlx5_core_dev *mdev) { - return mdev->mlx5e_res.uplink_netdev; + struct mlx5e_resources *mlx5e_res = &mdev->mlx5e_res; + struct net_device *netdev; + + mutex_lock(&mlx5e_res->uplink_netdev_lock); + netdev = mlx5e_res->uplink_netdev; + netdev_hold(netdev, &mlx5e_res->tracker, GFP_KERNEL); + mutex_unlock(&mlx5e_res->uplink_netdev_lock); + return netdev; +} + +static inline void mlx5_uplink_netdev_put(struct mlx5_core_dev *mdev, + struct net_device *netdev) +{ + netdev_put(netdev, &mdev->mlx5e_res.tracker); } struct mlx5_sd; diff --git a/include/linux/mlx5/driver.h b/include/linux/mlx5/driver.h index 8c5fbfb85749..10fe492e1fed 100644 --- a/include/linux/mlx5/driver.h +++ b/include/linux/mlx5/driver.h @@ -663,6 +663,7 @@ struct mlx5e_resources { bool tisn_valid; } hw_objs; struct net_device *uplink_netdev; + netdevice_tracker tracker; struct mutex uplink_netdev_lock; struct mlx5_crypto_dek_priv *dek_priv; }; -- cgit v1.2.3