From 38035c04f5865c4ef9597d6beed6a7178f90f64a Mon Sep 17 00:00:00 2001 From: Amir Goldstein Date: Fri, 22 Apr 2022 15:03:13 +0300 Subject: inotify: move control flags from mask to mark flags The inotify control flags in the mark mask (e.g. FS_IN_ONE_SHOT) are not relevant to object interest mask, so move them to the mark flags. This frees up some bits in the object interest mask. Link: https://lore.kernel.org/r/20220422120327.3459282-3-amir73il@gmail.com Signed-off-by: Amir Goldstein Signed-off-by: Jan Kara --- include/linux/fsnotify_backend.h | 16 +++++++++------- 1 file changed, 9 insertions(+), 7 deletions(-) (limited to 'include/linux') diff --git a/include/linux/fsnotify_backend.h b/include/linux/fsnotify_backend.h index 0805b74cae44..b1c72edd9784 100644 --- a/include/linux/fsnotify_backend.h +++ b/include/linux/fsnotify_backend.h @@ -55,7 +55,6 @@ #define FS_ACCESS_PERM 0x00020000 /* access event in a permissions hook */ #define FS_OPEN_EXEC_PERM 0x00040000 /* open/exec event in a permission hook */ -#define FS_EXCL_UNLINK 0x04000000 /* do not send events if object is unlinked */ /* * Set on inode mark that cares about things that happen to its children. * Always set for dnotify and inotify. @@ -66,7 +65,6 @@ #define FS_RENAME 0x10000000 /* File was renamed */ #define FS_DN_MULTISHOT 0x20000000 /* dnotify multishot */ #define FS_ISDIR 0x40000000 /* event occurred against dir */ -#define FS_IN_ONESHOT 0x80000000 /* only send event once */ #define FS_MOVE (FS_MOVED_FROM | FS_MOVED_TO) @@ -106,8 +104,7 @@ FS_ERROR) /* Extra flags that may be reported with event or control handling of events */ -#define ALL_FSNOTIFY_FLAGS (FS_EXCL_UNLINK | FS_ISDIR | FS_IN_ONESHOT | \ - FS_DN_MULTISHOT | FS_EVENT_ON_CHILD) +#define ALL_FSNOTIFY_FLAGS (FS_ISDIR | FS_EVENT_ON_CHILD | FS_DN_MULTISHOT) #define ALL_FSNOTIFY_BITS (ALL_FSNOTIFY_EVENTS | ALL_FSNOTIFY_FLAGS) @@ -473,9 +470,14 @@ struct fsnotify_mark { struct fsnotify_mark_connector *connector; /* Events types to ignore [mark->lock, group->mark_mutex] */ __u32 ignored_mask; -#define FSNOTIFY_MARK_FLAG_IGNORED_SURV_MODIFY 0x01 -#define FSNOTIFY_MARK_FLAG_ALIVE 0x02 -#define FSNOTIFY_MARK_FLAG_ATTACHED 0x04 + /* General fsnotify mark flags */ +#define FSNOTIFY_MARK_FLAG_ALIVE 0x0001 +#define FSNOTIFY_MARK_FLAG_ATTACHED 0x0002 + /* inotify mark flags */ +#define FSNOTIFY_MARK_FLAG_EXCL_UNLINK 0x0010 +#define FSNOTIFY_MARK_FLAG_IN_ONESHOT 0x0020 + /* fanotify mark flags */ +#define FSNOTIFY_MARK_FLAG_IGNORED_SURV_MODIFY 0x0100 unsigned int flags; /* flags [mark->lock] */ }; -- cgit v1.2.3 From 867a448d587e7fa845bceaf4ee1c632448f2a9fa Mon Sep 17 00:00:00 2001 From: Amir Goldstein Date: Fri, 22 Apr 2022 15:03:15 +0300 Subject: fsnotify: pass flags argument to fsnotify_alloc_group() Add flags argument to fsnotify_alloc_group(), define and use the flag FSNOTIFY_GROUP_USER in inotify and fanotify instead of the helper fsnotify_alloc_user_group() to indicate user allocation. Although the flag FSNOTIFY_GROUP_USER is currently not used after group allocation, we store the flags argument in the group struct for future use of other group flags. Link: https://lore.kernel.org/r/20220422120327.3459282-5-amir73il@gmail.com Signed-off-by: Amir Goldstein Signed-off-by: Jan Kara --- include/linux/fsnotify_backend.h | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/fsnotify_backend.h b/include/linux/fsnotify_backend.h index b1c72edd9784..f0bf557af009 100644 --- a/include/linux/fsnotify_backend.h +++ b/include/linux/fsnotify_backend.h @@ -210,6 +210,9 @@ struct fsnotify_group { unsigned int priority; bool shutdown; /* group is being shut down, don't queue more events */ +#define FSNOTIFY_GROUP_USER 0x01 /* user allocated group */ + int flags; + /* stores all fastpath marks assoc with this group so they can be cleaned on unregister */ struct mutex mark_mutex; /* protect marks_list */ atomic_t user_waits; /* Number of tasks waiting for user @@ -543,8 +546,9 @@ static inline void fsnotify_update_flags(struct dentry *dentry) /* called from fsnotify listeners, such as fanotify or dnotify */ /* create a new group */ -extern struct fsnotify_group *fsnotify_alloc_group(const struct fsnotify_ops *ops); -extern struct fsnotify_group *fsnotify_alloc_user_group(const struct fsnotify_ops *ops); +extern struct fsnotify_group *fsnotify_alloc_group( + const struct fsnotify_ops *ops, + int flags); /* get reference to a group */ extern void fsnotify_get_group(struct fsnotify_group *group); /* drop reference on a group from fsnotify_alloc_group */ -- cgit v1.2.3 From f3010343d9e119da35ee864b3a28993bb5c78ed7 Mon Sep 17 00:00:00 2001 From: Amir Goldstein Date: Fri, 22 Apr 2022 15:03:16 +0300 Subject: fsnotify: make allow_dups a property of the group Instead of passing the allow_dups argument to fsnotify_add_mark() as an argument, define the group flag FSNOTIFY_GROUP_DUPS to express the allow_dups behavior and set this behavior at group creation time for all calls of fsnotify_add_mark(). Rename the allow_dups argument to generic add_flags argument for future use. Link: https://lore.kernel.org/r/20220422120327.3459282-6-amir73il@gmail.com Suggested-by: Jan Kara Signed-off-by: Amir Goldstein Signed-off-by: Jan Kara --- include/linux/fsnotify_backend.h | 13 +++++++------ 1 file changed, 7 insertions(+), 6 deletions(-) (limited to 'include/linux') diff --git a/include/linux/fsnotify_backend.h b/include/linux/fsnotify_backend.h index f0bf557af009..dd440e6ff528 100644 --- a/include/linux/fsnotify_backend.h +++ b/include/linux/fsnotify_backend.h @@ -211,6 +211,7 @@ struct fsnotify_group { bool shutdown; /* group is being shut down, don't queue more events */ #define FSNOTIFY_GROUP_USER 0x01 /* user allocated group */ +#define FSNOTIFY_GROUP_DUPS 0x02 /* allow multiple marks per object */ int flags; /* stores all fastpath marks assoc with this group so they can be cleaned on unregister */ @@ -641,26 +642,26 @@ extern int fsnotify_get_conn_fsid(const struct fsnotify_mark_connector *conn, /* attach the mark to the object */ extern int fsnotify_add_mark(struct fsnotify_mark *mark, fsnotify_connp_t *connp, unsigned int obj_type, - int allow_dups, __kernel_fsid_t *fsid); + int add_flags, __kernel_fsid_t *fsid); extern int fsnotify_add_mark_locked(struct fsnotify_mark *mark, fsnotify_connp_t *connp, - unsigned int obj_type, int allow_dups, + unsigned int obj_type, int add_flags, __kernel_fsid_t *fsid); /* attach the mark to the inode */ static inline int fsnotify_add_inode_mark(struct fsnotify_mark *mark, struct inode *inode, - int allow_dups) + int add_flags) { return fsnotify_add_mark(mark, &inode->i_fsnotify_marks, - FSNOTIFY_OBJ_TYPE_INODE, allow_dups, NULL); + FSNOTIFY_OBJ_TYPE_INODE, add_flags, NULL); } static inline int fsnotify_add_inode_mark_locked(struct fsnotify_mark *mark, struct inode *inode, - int allow_dups) + int add_flags) { return fsnotify_add_mark_locked(mark, &inode->i_fsnotify_marks, - FSNOTIFY_OBJ_TYPE_INODE, allow_dups, + FSNOTIFY_OBJ_TYPE_INODE, add_flags, NULL); } -- cgit v1.2.3 From 43b245a788e2d8f1bb742668a9bdace02fcb3e96 Mon Sep 17 00:00:00 2001 From: Amir Goldstein Date: Fri, 22 Apr 2022 15:03:17 +0300 Subject: fsnotify: create helpers for group mark_mutex lock Create helpers to take and release the group mark_mutex lock. Define a flag FSNOTIFY_GROUP_NOFS in fsnotify_group that determines if the mark_mutex lock is fs reclaim safe or not. If not safe, the lock helpers take the lock and disable direct fs reclaim. In that case we annotate the mutex with a different lockdep class to express to lockdep that an allocation of mark of an fs reclaim safe group may take the group lock of another "NOFS" group to evict inodes. For now, converted only the callers in common code and no backend defines the NOFS flag. It is intended to be set by fanotify for evictable marks support. Link: https://lore.kernel.org/r/20220422120327.3459282-7-amir73il@gmail.com Suggested-by: Jan Kara Link: https://lore.kernel.org/r/20220321112310.vpr7oxro2xkz5llh@quack3.lan/ Signed-off-by: Amir Goldstein Signed-off-by: Jan Kara --- include/linux/fsnotify_backend.h | 28 ++++++++++++++++++++++++++++ 1 file changed, 28 insertions(+) (limited to 'include/linux') diff --git a/include/linux/fsnotify_backend.h b/include/linux/fsnotify_backend.h index dd440e6ff528..d62111e83244 100644 --- a/include/linux/fsnotify_backend.h +++ b/include/linux/fsnotify_backend.h @@ -20,6 +20,7 @@ #include #include #include +#include /* * IN_* from inotfy.h lines up EXACTLY with FS_*, this is so we can easily @@ -212,7 +213,9 @@ struct fsnotify_group { #define FSNOTIFY_GROUP_USER 0x01 /* user allocated group */ #define FSNOTIFY_GROUP_DUPS 0x02 /* allow multiple marks per object */ +#define FSNOTIFY_GROUP_NOFS 0x04 /* group lock is not direct reclaim safe */ int flags; + unsigned int owner_flags; /* stored flags of mark_mutex owner */ /* stores all fastpath marks assoc with this group so they can be cleaned on unregister */ struct mutex mark_mutex; /* protect marks_list */ @@ -254,6 +257,31 @@ struct fsnotify_group { }; }; +/* + * These helpers are used to prevent deadlock when reclaiming inodes with + * evictable marks of the same group that is allocating a new mark. + */ +static inline void fsnotify_group_lock(struct fsnotify_group *group) +{ + mutex_lock(&group->mark_mutex); + if (group->flags & FSNOTIFY_GROUP_NOFS) + group->owner_flags = memalloc_nofs_save(); +} + +static inline void fsnotify_group_unlock(struct fsnotify_group *group) +{ + if (group->flags & FSNOTIFY_GROUP_NOFS) + memalloc_nofs_restore(group->owner_flags); + mutex_unlock(&group->mark_mutex); +} + +static inline void fsnotify_group_assert_locked(struct fsnotify_group *group) +{ + WARN_ON_ONCE(!mutex_is_locked(&group->mark_mutex)); + if (group->flags & FSNOTIFY_GROUP_NOFS) + WARN_ON_ONCE(!(current->flags & PF_MEMALLOC_NOFS)); +} + /* When calling fsnotify tell it if the data is a path or inode */ enum fsnotify_data_type { FSNOTIFY_EVENT_NONE, -- cgit v1.2.3 From c3638b5b13740fa31762d414bbce8b7a694e582a Mon Sep 17 00:00:00 2001 From: Amir Goldstein Date: Fri, 22 Apr 2022 15:03:22 +0300 Subject: fsnotify: allow adding an inode mark without pinning inode fsnotify_add_mark() and variants implicitly take a reference on inode when attaching a mark to an inode. Make that behavior opt-out with the mark flag FSNOTIFY_MARK_FLAG_NO_IREF. Instead of taking the inode reference when attaching connector to inode and dropping the inode reference when detaching connector from inode, take the inode reference on attach of the first mark that wants to hold an inode reference and drop the inode reference on detach of the last mark that wants to hold an inode reference. Backends can "upgrade" an existing mark to take an inode reference, but cannot "downgrade" a mark with inode reference to release the refernce. This leaves the choice to the backend whether or not to pin the inode when adding an inode mark. This is intended to be used when adding a mark with ignored mask that is used for optimization in cases where group can afford getting unneeded events and reinstate the mark with ignored mask when inode is accessed again after being evicted. Link: https://lore.kernel.org/r/20220422120327.3459282-12-amir73il@gmail.com Signed-off-by: Amir Goldstein Signed-off-by: Jan Kara --- include/linux/fsnotify_backend.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include/linux') diff --git a/include/linux/fsnotify_backend.h b/include/linux/fsnotify_backend.h index d62111e83244..9a1a9e78f69f 100644 --- a/include/linux/fsnotify_backend.h +++ b/include/linux/fsnotify_backend.h @@ -456,6 +456,7 @@ struct fsnotify_mark_connector { spinlock_t lock; unsigned short type; /* Type of object [lock] */ #define FSNOTIFY_CONN_FLAG_HAS_FSID 0x01 +#define FSNOTIFY_CONN_FLAG_HAS_IREF 0x02 unsigned short flags; /* flags [lock] */ __kernel_fsid_t fsid; /* fsid of filesystem containing object */ union { @@ -510,6 +511,7 @@ struct fsnotify_mark { #define FSNOTIFY_MARK_FLAG_IN_ONESHOT 0x0020 /* fanotify mark flags */ #define FSNOTIFY_MARK_FLAG_IGNORED_SURV_MODIFY 0x0100 +#define FSNOTIFY_MARK_FLAG_NO_IREF 0x0200 unsigned int flags; /* flags [mark->lock] */ }; -- cgit v1.2.3 From 5f9d3bd520261fd7a850818c71809fd580e0f30c Mon Sep 17 00:00:00 2001 From: Amir Goldstein Date: Fri, 22 Apr 2022 15:03:27 +0300 Subject: fanotify: enable "evictable" inode marks Now that the direct reclaim path is handled we can enable evictable inode marks. Link: https://lore.kernel.org/r/20220422120327.3459282-17-amir73il@gmail.com Signed-off-by: Amir Goldstein Signed-off-by: Jan Kara --- include/linux/fanotify.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/fanotify.h b/include/linux/fanotify.h index 419cadcd7ff5..edc28555814c 100644 --- a/include/linux/fanotify.h +++ b/include/linux/fanotify.h @@ -66,6 +66,7 @@ FAN_MARK_ONLYDIR | \ FAN_MARK_IGNORED_MASK | \ FAN_MARK_IGNORED_SURV_MODIFY | \ + FAN_MARK_EVICTABLE | \ FAN_MARK_FLUSH) /* -- cgit v1.2.3 From 14362a2541797cf9df0e86fb12dcd7950baf566e Mon Sep 17 00:00:00 2001 From: Amir Goldstein Date: Wed, 11 May 2022 22:02:12 +0300 Subject: fsnotify: introduce mark type iterator fsnotify_foreach_iter_mark_type() is used to reduce boilerplate code of iterating all marks of a specific group interested in an event by consulting the iterator report_mask. Use an open coded version of that iterator in fsnotify_iter_next() that collects all marks of the current iteration group without consulting the iterator report_mask. At the moment, the two iterator variants are the same, but this decoupling will allow us to exclude some of the group's marks from reporting the event, for example for event on child and inode marks on parent did not request to watch events on children. Fixes: 2f02fd3fa13e ("fanotify: fix ignore mask logic for events on child and on dir") Reported-by: Jan Kara Signed-off-by: Amir Goldstein Signed-off-by: Jan Kara Link: https://lore.kernel.org/r/20220511190213.831646-2-amir73il@gmail.com --- include/linux/fsnotify_backend.h | 31 ++++++++++++++++++++++++------- 1 file changed, 24 insertions(+), 7 deletions(-) (limited to 'include/linux') diff --git a/include/linux/fsnotify_backend.h b/include/linux/fsnotify_backend.h index 9a1a9e78f69f..9560734759fa 100644 --- a/include/linux/fsnotify_backend.h +++ b/include/linux/fsnotify_backend.h @@ -399,6 +399,7 @@ static inline bool fsnotify_valid_obj_type(unsigned int obj_type) struct fsnotify_iter_info { struct fsnotify_mark *marks[FSNOTIFY_ITER_TYPE_COUNT]; + struct fsnotify_group *current_group; unsigned int report_mask; int srcu_idx; }; @@ -415,20 +416,31 @@ static inline void fsnotify_iter_set_report_type( iter_info->report_mask |= (1U << iter_type); } -static inline void fsnotify_iter_set_report_type_mark( - struct fsnotify_iter_info *iter_info, int iter_type, - struct fsnotify_mark *mark) +static inline struct fsnotify_mark *fsnotify_iter_mark( + struct fsnotify_iter_info *iter_info, int iter_type) { - iter_info->marks[iter_type] = mark; - iter_info->report_mask |= (1U << iter_type); + if (fsnotify_iter_should_report_type(iter_info, iter_type)) + return iter_info->marks[iter_type]; + return NULL; +} + +static inline int fsnotify_iter_step(struct fsnotify_iter_info *iter, int type, + struct fsnotify_mark **markp) +{ + while (type < FSNOTIFY_ITER_TYPE_COUNT) { + *markp = fsnotify_iter_mark(iter, type); + if (*markp) + break; + type++; + } + return type; } #define FSNOTIFY_ITER_FUNCS(name, NAME) \ static inline struct fsnotify_mark *fsnotify_iter_##name##_mark( \ struct fsnotify_iter_info *iter_info) \ { \ - return (iter_info->report_mask & (1U << FSNOTIFY_ITER_TYPE_##NAME)) ? \ - iter_info->marks[FSNOTIFY_ITER_TYPE_##NAME] : NULL; \ + return fsnotify_iter_mark(iter_info, FSNOTIFY_ITER_TYPE_##NAME); \ } FSNOTIFY_ITER_FUNCS(inode, INODE) @@ -438,6 +450,11 @@ FSNOTIFY_ITER_FUNCS(sb, SB) #define fsnotify_foreach_iter_type(type) \ for (type = 0; type < FSNOTIFY_ITER_TYPE_COUNT; type++) +#define fsnotify_foreach_iter_mark_type(iter, mark, type) \ + for (type = 0; \ + type = fsnotify_iter_step(iter, type, &mark), \ + type < FSNOTIFY_ITER_TYPE_COUNT; \ + type++) /* * fsnotify_connp_t is what we embed in objects which connector can be attached -- cgit v1.2.3