Merge branch 'slab/for-6.12/kmem_cache_args' into slab/for-next

Merge kmem_cache_create() refactoring by Christian Brauner.

Note this includes a merge of the vfs.file tree that contains the
prerequisity kmem_cache_create_rcu() work.

drivers/net/tun.c

@@ -3451,6 +3451,12 @@ static int tun_chr_fasync(int fd, struct file *file, int on)
 	struct tun_file *tfile = file->private_data;
 	int ret;
 
+	if (on) {
+		ret = file_f_owner_allocate(file);
+		if (ret)
+			goto out;
+	}
+
 	if ((ret = fasync_helper(fd, file, on, &tfile->fasync)) < 0)
 		goto out;
 

drivers/tty/tty_io.c

@@ -2225,6 +2225,12 @@ static int __tty_fasync(int fd, struct file *filp, int on)
 	if (tty_paranoia_check(tty, file_inode(filp), "tty_fasync"))
 		goto out;
 
+	if (on) {
+		retval = file_f_owner_allocate(filp);
+		if (retval)
+			goto out;
+	}
+
 	retval = fasync_helper(fd, filp, on, &tty->fasync);
 	if (retval <= 0)
 		goto out;

fs/fcntl.c

@@ -33,6 +33,8 @@
 #include <asm/siginfo.h>
 #include <linux/uaccess.h>
 
+#include "internal.h"
+
 #define SETFL_MASK (O_APPEND | O_NONBLOCK | O_NDELAY | O_DIRECT | O_NOATIME)
 
 static int setfl(int fd, struct file * filp, unsigned int arg)
@@ -87,22 +89,64 @@ static int setfl(int fd, struct file * filp, unsigned int arg)
 	return error;
 }
 
+/*
+ * Allocate an file->f_owner struct if it doesn't exist, handling racing
+ * allocations correctly.
+ */
+int file_f_owner_allocate(struct file *file)
+{
+	struct fown_struct *f_owner;
+
+	f_owner = file_f_owner(file);
+	if (f_owner)
+		return 0;
+
+	f_owner = kzalloc(sizeof(struct fown_struct), GFP_KERNEL);
+	if (!f_owner)
+		return -ENOMEM;
+
+	rwlock_init(&f_owner->lock);
+	f_owner->file = file;
+	/* If someone else raced us, drop our allocation. */
+	if (unlikely(cmpxchg(&file->f_owner, NULL, f_owner)))
+		kfree(f_owner);
+	return 0;
+}
+EXPORT_SYMBOL(file_f_owner_allocate);
+
+void file_f_owner_release(struct file *file)
+{
+	struct fown_struct *f_owner;
+
+	f_owner = file_f_owner(file);
+	if (f_owner) {
+		put_pid(f_owner->pid);
+		kfree(f_owner);
+	}
+}
+
 static void f_modown(struct file *filp, struct pid *pid, enum pid_type type,
                      int force)
 {
-	write_lock_irq(&filp->f_owner.lock);
-	if (force || !filp->f_owner.pid) {
-		put_pid(filp->f_owner.pid);
-		filp->f_owner.pid = get_pid(pid);
-		filp->f_owner.pid_type = type;
+	struct fown_struct *f_owner;
+
+	f_owner = file_f_owner(filp);
+	if (WARN_ON_ONCE(!f_owner))
+		return;
+
+	write_lock_irq(&f_owner->lock);
+	if (force || !f_owner->pid) {
+		put_pid(f_owner->pid);
+		f_owner->pid = get_pid(pid);
+		f_owner->pid_type = type;
 
 		if (pid) {
 			const struct cred *cred = current_cred();
-			filp->f_owner.uid = cred->uid;
-			filp->f_owner.euid = cred->euid;
+			f_owner->uid = cred->uid;
+			f_owner->euid = cred->euid;
 		}
 	}
-	write_unlock_irq(&filp->f_owner.lock);
+	write_unlock_irq(&f_owner->lock);
 }
 
 void __f_setown(struct file *filp, struct pid *pid, enum pid_type type,
@@ -119,6 +163,8 @@ int f_setown(struct file *filp, int who, int force)
 	struct pid *pid = NULL;
 	int ret = 0;
 
+	might_sleep();
+
 	type = PIDTYPE_TGID;
 	if (who < 0) {
 		/* avoid overflow below */
@@ -129,6 +175,10 @@ int f_setown(struct file *filp, int who, int force)
 		who = -who;
 	}
 
+	ret = file_f_owner_allocate(filp);
+	if (ret)
+		return ret;
+
 	rcu_read_lock();
 	if (who) {
 		pid = find_vpid(who);
@@ -152,16 +202,21 @@ void f_delown(struct file *filp)
 pid_t f_getown(struct file *filp)
 {
 	pid_t pid = 0;
+	struct fown_struct *f_owner;
+
+	f_owner = file_f_owner(filp);
+	if (!f_owner)
+		return pid;
 
-	read_lock_irq(&filp->f_owner.lock);
+	read_lock_irq(&f_owner->lock);
 	rcu_read_lock();
-	if (pid_task(filp->f_owner.pid, filp->f_owner.pid_type)) {
-		pid = pid_vnr(filp->f_owner.pid);
-		if (filp->f_owner.pid_type == PIDTYPE_PGID)
+	if (pid_task(f_owner->pid, f_owner->pid_type)) {
+		pid = pid_vnr(f_owner->pid);
+		if (f_owner->pid_type == PIDTYPE_PGID)
 			pid = -pid;
 	}
 	rcu_read_unlock();
-	read_unlock_irq(&filp->f_owner.lock);
+	read_unlock_irq(&f_owner->lock);
 	return pid;
 }
 
@@ -194,6 +249,10 @@ static int f_setown_ex(struct file *filp, unsigned long arg)
 		return -EINVAL;
 	}
 
+	ret = file_f_owner_allocate(filp);
+	if (ret)
+		return ret;
+
 	rcu_read_lock();
 	pid = find_vpid(owner.pid);
 	if (owner.pid && !pid)
@@ -210,13 +269,20 @@ static int f_getown_ex(struct file *filp, unsigned long arg)
 	struct f_owner_ex __user *owner_p = (void __user *)arg;
 	struct f_owner_ex owner = {};
 	int ret = 0;
+	struct fown_struct *f_owner;
+	enum pid_type pid_type = PIDTYPE_PID;
 
-	read_lock_irq(&filp->f_owner.lock);
-	rcu_read_lock();
-	if (pid_task(filp->f_owner.pid, filp->f_owner.pid_type))
-		owner.pid = pid_vnr(filp->f_owner.pid);
-	rcu_read_unlock();
-	switch (filp->f_owner.pid_type) {
+	f_owner = file_f_owner(filp);
+	if (f_owner) {
+		read_lock_irq(&f_owner->lock);
+		rcu_read_lock();
+		if (pid_task(f_owner->pid, f_owner->pid_type))
+			owner.pid = pid_vnr(f_owner->pid);
+		rcu_read_unlock();
+		pid_type = f_owner->pid_type;
+	}
+
+	switch (pid_type) {
 	case PIDTYPE_PID:
 		owner.type = F_OWNER_TID;
 		break;
@@ -234,7 +300,8 @@ static int f_getown_ex(struct file *filp, unsigned long arg)
 		ret = -EINVAL;
 		break;
 	}
-	read_unlock_irq(&filp->f_owner.lock);
+	if (f_owner)
+		read_unlock_irq(&f_owner->lock);
 
 	if (!ret) {
 		ret = copy_to_user(owner_p, &owner, sizeof(owner));
@@ -248,14 +315,18 @@ static int f_getown_ex(struct file *filp, unsigned long arg)
 static int f_getowner_uids(struct file *filp, unsigned long arg)
 {
 	struct user_namespace *user_ns = current_user_ns();
+	struct fown_struct *f_owner;
 	uid_t __user *dst = (void __user *)arg;
-	uid_t src[2];
+	uid_t src[2] = {0, 0};
 	int err;
 
-	read_lock_irq(&filp->f_owner.lock);
-	src[0] = from_kuid(user_ns, filp->f_owner.uid);
-	src[1] = from_kuid(user_ns, filp->f_owner.euid);
-	read_unlock_irq(&filp->f_owner.lock);
+	f_owner = file_f_owner(filp);
+	if (f_owner) {
+		read_lock_irq(&f_owner->lock);
+		src[0] = from_kuid(user_ns, f_owner->uid);
+		src[1] = from_kuid(user_ns, f_owner->euid);
+		read_unlock_irq(&f_owner->lock);
+	}
 
 	err  = put_user(src[0], &dst[0]);
 	err |= put_user(src[1], &dst[1]);
@@ -343,6 +414,30 @@ static long f_dupfd_query(int fd, struct file *filp)
 	return f.file == filp;
 }
 
+static int f_owner_sig(struct file *filp, int signum, bool setsig)
+{
+	int ret = 0;
+	struct fown_struct *f_owner;
+
+	might_sleep();
+
+	if (setsig) {
+		if (!valid_signal(signum))
+			return -EINVAL;
+
+		ret = file_f_owner_allocate(filp);
+		if (ret)
+			return ret;
+	}
+
+	f_owner = file_f_owner(filp);
+	if (setsig)
+		f_owner->signum = signum;
+	else if (f_owner)
+		ret = f_owner->signum;
+	return ret;
+}
+
 static long do_fcntl(int fd, unsigned int cmd, unsigned long arg,
 		struct file *filp)
 {
@@ -421,15 +516,10 @@ static long do_fcntl(int fd, unsigned int cmd, unsigned long arg,
 		err = f_getowner_uids(filp, arg);
 		break;
 	case F_GETSIG:
-		err = filp->f_owner.signum;
+		err = f_owner_sig(filp, 0, false);
 		break;
 	case F_SETSIG:
-		/* arg == 0 restores default behaviour. */
-		if (!valid_signal(argi)) {
-			break;
-		}
-		err = 0;
-		filp->f_owner.signum = argi;
+		err = f_owner_sig(filp, argi, true);
 		break;
 	case F_GETLEASE:
 		err = fcntl_getlease(filp);
@@ -844,14 +934,19 @@ static void send_sigurg_to_task(struct task_struct *p,
 		do_send_sig_info(SIGURG, SEND_SIG_PRIV, p, type);
 }
 
-int send_sigurg(struct fown_struct *fown)
+int send_sigurg(struct file *file)
 {
+	struct fown_struct *fown;
 	struct task_struct *p;
 	enum pid_type type;
 	struct pid *pid;
 	unsigned long flags;
 	int ret = 0;
 	
+	fown = file_f_owner(file);
+	if (!fown)
+		return 0;
+
 	read_lock_irqsave(&fown->lock, flags);
 
 	type = fown->pid_type;
@@ -1027,13 +1122,16 @@ static void kill_fasync_rcu(struct fasync_struct *fa, int sig, int band)
 		}
 		read_lock_irqsave(&fa->fa_lock, flags);
 		if (fa->fa_file) {
-			fown = &fa->fa_file->f_owner;
+			fown = file_f_owner(fa->fa_file);
+			if (!fown)
+				goto next;
 			/* Don't send SIGURG to processes which have not set a
 			   queued signum: SIGURG has its own default signalling
 			   mechanism. */
 			if (!(sig == SIGURG && fown->signum == 0))
 				send_sigio(fown, fa->fa_fd, band);
 		}
+next:
 		read_unlock_irqrestore(&fa->fa_lock, flags);
 		fa = rcu_dereference(fa->fa_next);
 	}

fs/file_table.c

@@ -155,7 +155,6 @@ static int init_file(struct file *f, int flags, const struct cred *cred)
 		return error;
 	}
 
-	rwlock_init(&f->f_owner.lock);
 	spin_lock_init(&f->f_lock);
 	mutex_init(&f->f_pos_lock);
 	f->f_flags = flags;
@@ -425,7 +424,7 @@ static void __fput(struct file *file)
 		cdev_put(inode->i_cdev);
 	}
 	fops_put(file->f_op);
-	put_pid(file->f_owner.pid);
+	file_f_owner_release(file);
 	put_file_access(file);
 	dput(dentry);
 	if (unlikely(mode & FMODE_NEED_UNMOUNT))
@@ -512,9 +511,14 @@ EXPORT_SYMBOL(__fput_sync);
 
 void __init files_init(void)
 {
-	filp_cachep = kmem_cache_create("filp", sizeof(struct file), 0,
-				SLAB_TYPESAFE_BY_RCU | SLAB_HWCACHE_ALIGN |
-				SLAB_PANIC | SLAB_ACCOUNT, NULL);
+	struct kmem_cache_args args = {
+		.use_freeptr_offset = true,
+		.freeptr_offset = offsetof(struct file, f_freeptr),
+	};
+
+	filp_cachep = kmem_cache_create("filp", sizeof(struct file), &args,
+				SLAB_HWCACHE_ALIGN | SLAB_PANIC |
+				SLAB_ACCOUNT | SLAB_TYPESAFE_BY_RCU);
 	percpu_counter_init(&nr_files, 0, GFP_KERNEL);
 }
 

fs/internal.h

@@ -337,3 +337,4 @@ static inline bool path_mounted(const struct path *path)
 {
 	return path->mnt->mnt_root == path->dentry;
 }
+void file_f_owner_release(struct file *file);

fs/locks.c

@@ -1451,7 +1451,7 @@ int lease_modify(struct file_lease *fl, int arg, struct list_head *dispose)
 		struct file *filp = fl->c.flc_file;
 
 		f_delown(filp);
-		filp->f_owner.signum = 0;
+		file_f_owner(filp)->signum = 0;
 		fasync_helper(0, fl->c.flc_file, 0, &fl->fl_fasync);
 		if (fl->fl_fasync != NULL) {
 			printk(KERN_ERR "locks_delete_lock: fasync == %p\n", fl->fl_fasync);
@@ -1783,6 +1783,10 @@ generic_add_lease(struct file *filp, int arg, struct file_lease **flp, void **pr
 	lease = *flp;
 	trace_generic_add_lease(inode, lease);
 
+	error = file_f_owner_allocate(filp);
+	if (error)
+		return error;
+
 	/* Note that arg is never F_UNLCK here */
 	ctx = locks_get_lock_context(inode, arg);
 	if (!ctx)

fs/notify/dnotify/dnotify.c

@@ -110,7 +110,7 @@ static int dnotify_handle_event(struct fsnotify_mark *inode_mark, u32 mask,
 			prev = &dn->dn_next;
 			continue;
 		}
-		fown = &dn->dn_filp->f_owner;
+		fown = file_f_owner(dn->dn_filp);
 		send_sigio(fown, dn->dn_fd, POLL_MSG);
 		if (dn->dn_mask & FS_DN_MULTISHOT)
 			prev = &dn->dn_next;
@@ -316,6 +316,10 @@ int fcntl_dirnotify(int fd, struct file *filp, unsigned int arg)
 		goto out_err;
 	}
 
+	error = file_f_owner_allocate(filp);
+	if (error)
+		goto out_err;
+
 	/* set up the new_fsn_mark and new_dn_mark */
 	new_fsn_mark = &new_dn_mark->fsn_mark;
 	fsnotify_init_mark(new_fsn_mark, dnotify_group);

include/linux/fs.h

@@ -947,6 +947,7 @@ static inline unsigned imajor(const struct inode *inode)
 }
 
 struct fown_struct {
+	struct file *file;	/* backpointer for security modules */
 	rwlock_t lock;          /* protects pid, uid, euid fields */
 	struct pid *pid;	/* pid or -pgrp where SIGIO should be sent */
 	enum pid_type pid_type;	/* Kind of process group SIGIO should be sent to */
@@ -986,52 +987,65 @@ static inline int ra_has_index(struct file_ra_state *ra, pgoff_t index)
 		index <  ra->start + ra->size);
 }
 
-/*
- * f_{lock,count,pos_lock} members can be highly contended and share
- * the same cacheline. f_{lock,mode} are very frequently used together
- * and so share the same cacheline as well. The read-mostly
- * f_{path,inode,op} are kept on a separate cacheline.
+/**
+ * struct file - Represents a file
+ * @f_count: reference count
+ * @f_lock: Protects f_ep, f_flags. Must not be taken from IRQ context.
+ * @f_mode: FMODE_* flags often used in hotpaths
+ * @f_op: file operations
+ * @f_mapping: Contents of a cacheable, mappable object.
+ * @private_data: filesystem or driver specific data
+ * @f_inode: cached inode
+ * @f_flags: file flags
+ * @f_iocb_flags: iocb flags
+ * @f_cred: stashed credentials of creator/opener
+ * @f_path: path of the file
+ * @f_pos_lock: lock protecting file position
+ * @f_pos: file position
+ * @f_version: file version
+ * @f_security: LSM security context of this file
+ * @f_owner: file owner
+ * @f_wb_err: writeback error
+ * @f_sb_err: per sb writeback errors
+ * @f_ep: link of all epoll hooks for this file
+ * @f_task_work: task work entry point
+ * @f_llist: work queue entrypoint
+ * @f_ra: file's readahead state
+ * @f_freeptr: Pointer used by SLAB_TYPESAFE_BY_RCU file cache (don't touch.)
  */
 struct file {
-	union {
-		/* fput() uses task work when closing and freeing file (default). */
-		struct callback_head 	f_task_work;
-		/* fput() must use workqueue (most kernel threads). */
-		struct llist_node	f_llist;
-		unsigned int 		f_iocb_flags;
-	};
-
-	/*
-	 * Protects f_ep, f_flags.
-	 * Must not be taken from IRQ context.
-	 */
-	spinlock_t		f_lock;
-	fmode_t			f_mode;
-	atomic_long_t		f_count;
-	struct mutex		f_pos_lock;
-	loff_t			f_pos;
-	unsigned int		f_flags;
-	struct fown_struct	f_owner;
-	const struct cred	*f_cred;
-	struct file_ra_state	f_ra;
-	struct path		f_path;
-	struct inode		*f_inode;	/* cached value */
+	atomic_long_t			f_count;
+	spinlock_t			f_lock;
+	fmode_t				f_mode;
 	const struct file_operations	*f_op;
-
-	u64			f_version;
+	struct address_space		*f_mapping;
+	void				*private_data;
+	struct inode			*f_inode;
+	unsigned int			f_flags;
+	unsigned int			f_iocb_flags;
+	const struct cred		*f_cred;
+	/* --- cacheline 1 boundary (64 bytes) --- */
+	struct path			f_path;
+	struct mutex			f_pos_lock;
+	loff_t				f_pos;
+	u64				f_version;
+	/* --- cacheline 2 boundary (128 bytes) --- */
 #ifdef CONFIG_SECURITY
-	void			*f_security;
+	void				*f_security;
 #endif
-	/* needed for tty driver, and maybe others */
-	void			*private_data;
-
+	struct fown_struct		*f_owner;
+	errseq_t			f_wb_err;
+	errseq_t			f_sb_err;
 #ifdef CONFIG_EPOLL
-	/* Used by fs/eventpoll.c to link all the hooks to this file */
-	struct hlist_head	*f_ep;
-#endif /* #ifdef CONFIG_EPOLL */
-	struct address_space	*f_mapping;
-	errseq_t		f_wb_err;
-	errseq_t		f_sb_err; /* for syncfs */
+	struct hlist_head		*f_ep;
+#endif
+	union {
+		struct callback_head	f_task_work;
+		struct llist_node	f_llist;
+		struct file_ra_state	f_ra;
+		freeptr_t		f_freeptr;
+	};
+	/* --- cacheline 3 boundary (192 bytes) --- */
 } __randomize_layout
   __attribute__((aligned(4)));	/* lest something weird decides that 2 is OK */
 
@@ -1076,6 +1090,12 @@ struct file_lease;
 #define OFFT_OFFSET_MAX	type_max(off_t)
 #endif
 
+int file_f_owner_allocate(struct file *file);
+static inline struct fown_struct *file_f_owner(const struct file *file)
+{
+	return READ_ONCE(file->f_owner);
+}
+
 extern void send_sigio(struct fown_struct *fown, int fd, int band);
 
 static inline struct inode *file_inode(const struct file *f)
@@ -1124,7 +1144,7 @@ extern void __f_setown(struct file *filp, struct pid *, enum pid_type, int force
 extern int f_setown(struct file *filp, int who, int force);
 extern void f_delown(struct file *filp);
 extern pid_t f_getown(struct file *filp);
-extern int send_sigurg(struct fown_struct *fown);
+extern int send_sigurg(struct file *file);
 
 /*
  * sb->s_flags.  Note that these mirror the equivalent MS_* flags where

include/linux/slab.h

@@ -212,6 +212,12 @@ enum _slab_flag_bits {
 #define SLAB_NO_OBJ_EXT		__SLAB_FLAG_UNUSED
 #endif
 
+/*
+ * freeptr_t represents a SLUB freelist pointer, which might be encoded
+ * and not dereferenceable if CONFIG_SLAB_FREELIST_HARDENED is enabled.
+ */
+typedef struct { unsigned long v; } freeptr_t;
+
 /*
  * ZERO_SIZE_PTR will be returned for zero sized kmalloc requests.
  *
@@ -234,14 +240,173 @@ struct mem_cgroup;
  */
 bool slab_is_available(void);
 
-struct kmem_cache *kmem_cache_create(const char *name, unsigned int size,
-			unsigned int align, slab_flags_t flags,
-			void (*ctor)(void *));
-struct kmem_cache *kmem_cache_create_usercopy(const char *name,
-			unsigned int size, unsigned int align,
-			slab_flags_t flags,
-			unsigned int useroffset, unsigned int usersize,
-			void (*ctor)(void *));
+/**
+ * struct kmem_cache_args - Less common arguments for kmem_cache_create()
+ *
+ * Any uninitialized fields of the structure are interpreted as unused. The
+ * exception is @freeptr_offset where %0 is a valid value, so
+ * @use_freeptr_offset must be also set to %true in order to interpret the field
+ * as used. For @useroffset %0 is also valid, but only with non-%0
+ * @usersize.
+ *
+ * When %NULL args is passed to kmem_cache_create(), it is equivalent to all
+ * fields unused.
+ */
+struct kmem_cache_args {
+	/**
+	 * @align: The required alignment for the objects.
+	 *
+	 * %0 means no specific alignment is requested.
+	 */
+	unsigned int align;
+	/**
+	 * @useroffset: Usercopy region offset.
+	 *
+	 * %0 is a valid offset, when @usersize is non-%0
+	 */
+	unsigned int useroffset;
+	/**
+	 * @usersize: Usercopy region size.
+	 *
+	 * %0 means no usercopy region is specified.
+	 */
+	unsigned int usersize;
+	/**
+	 * @freeptr_offset: Custom offset for the free pointer
+	 * in &SLAB_TYPESAFE_BY_RCU caches
+	 *
+	 * By default &SLAB_TYPESAFE_BY_RCU caches place the free pointer
+	 * outside of the object. This might cause the object to grow in size.
+	 * Cache creators that have a reason to avoid this can specify a custom
+	 * free pointer offset in their struct where the free pointer will be
+	 * placed.
+	 *
+	 * Note that placing the free pointer inside the object requires the
+	 * caller to ensure that no fields are invalidated that are required to
+	 * guard against object recycling (See &SLAB_TYPESAFE_BY_RCU for
+	 * details).
+	 *
+	 * Using %0 as a value for @freeptr_offset is valid. If @freeptr_offset
+	 * is specified, %use_freeptr_offset must be set %true.
+	 *
+	 * Note that @ctor currently isn't supported with custom free pointers
+	 * as a @ctor requires an external free pointer.
+	 */
+	unsigned int freeptr_offset;
+	/**
+	 * @use_freeptr_offset: Whether a @freeptr_offset is used.
+	 */
+	bool use_freeptr_offset;
+	/**
+	 * @ctor: A constructor for the objects.
+	 *
+	 * The constructor is invoked for each object in a newly allocated slab
+	 * page. It is the cache user's responsibility to free object in the
+	 * same state as after calling the constructor, or deal appropriately
+	 * with any differences between a freshly constructed and a reallocated
+	 * object.
+	 *
+	 * %NULL means no constructor.
+	 */
+	void (*ctor)(void *);
+};
+
+struct kmem_cache *__kmem_cache_create_args(const char *name,
+					    unsigned int object_size,
+					    struct kmem_cache_args *args,
+					    slab_flags_t flags);
+static inline struct kmem_cache *
+__kmem_cache_create(const char *name, unsigned int size, unsigned int align,
+		    slab_flags_t flags, void (*ctor)(void *))
+{
+	struct kmem_cache_args kmem_args = {
+		.align	= align,
+		.ctor	= ctor,
+	};
+
+	return __kmem_cache_create_args(name, size, &kmem_args, flags);
+}
+
+/**
+ * kmem_cache_create_usercopy - Create a kmem cache with a region suitable
+ * for copying to userspace.
+ * @name: A string which is used in /proc/slabinfo to identify this cache.
+ * @size: The size of objects to be created in this cache.
+ * @align: The required alignment for the objects.
+ * @flags: SLAB flags
+ * @useroffset: Usercopy region offset
+ * @usersize: Usercopy region size
+ * @ctor: A constructor for the objects, or %NULL.
+ *
+ * This is a legacy wrapper, new code should use either KMEM_CACHE_USERCOPY()
+ * if whitelisting a single field is sufficient, or kmem_cache_create() with
+ * the necessary parameters passed via the args parameter (see
+ * &struct kmem_cache_args)
+ *
+ * Return: a pointer to the cache on success, NULL on failure.
+ */
+static inline struct kmem_cache *
+kmem_cache_create_usercopy(const char *name, unsigned int size,
+			   unsigned int align, slab_flags_t flags,
+			   unsigned int useroffset, unsigned int usersize,
+			   void (*ctor)(void *))
+{
+	struct kmem_cache_args kmem_args = {
+		.align		= align,
+		.ctor		= ctor,
+		.useroffset	= useroffset,
+		.usersize	= usersize,
+	};
+
+	return __kmem_cache_create_args(name, size, &kmem_args, flags);
+}
+
+/* If NULL is passed for @args, use this variant with default arguments. */
+static inline struct kmem_cache *
+__kmem_cache_default_args(const char *name, unsigned int size,
+			  struct kmem_cache_args *args,
+			  slab_flags_t flags)
+{
+	struct kmem_cache_args kmem_default_args = {};
+
+	/* Make sure we don't get passed garbage. */
+	if (WARN_ON_ONCE(args))
+		return ERR_PTR(-EINVAL);
+
+	return __kmem_cache_create_args(name, size, &kmem_default_args, flags);
+}
+
+/**
+ * kmem_cache_create - Create a kmem cache.
+ * @__name: A string which is used in /proc/slabinfo to identify this cache.
+ * @__object_size: The size of objects to be created in this cache.
+ * @__args: Optional arguments, see &struct kmem_cache_args. Passing %NULL
+ *	    means defaults will be used for all the arguments.
+ *
+ * This is currently implemented as a macro using ``_Generic()`` to call
+ * either the new variant of the function, or a legacy one.
+ *
+ * The new variant has 4 parameters:
+ * ``kmem_cache_create(name, object_size, args, flags)``
+ *
+ * See __kmem_cache_create_args() which implements this.
+ *
+ * The legacy variant has 5 parameters:
+ * ``kmem_cache_create(name, object_size, align, flags, ctor)``
+ *
+ * The align and ctor parameters map to the respective fields of
+ * &struct kmem_cache_args
+ *
+ * Context: Cannot be called within a interrupt, but can be interrupted.
+ *
+ * Return: a pointer to the cache on success, NULL on failure.
+ */
+#define kmem_cache_create(__name, __object_size, __args, ...)           \
+	_Generic((__args),                                              \
+		struct kmem_cache_args *: __kmem_cache_create_args,	\
+		void *: __kmem_cache_default_args,			\
+		default: __kmem_cache_create)(__name, __object_size, __args, __VA_ARGS__)
+
 void kmem_cache_destroy(struct kmem_cache *s);
 int kmem_cache_shrink(struct kmem_cache *s);
 
@@ -253,20 +418,23 @@ int kmem_cache_shrink(struct kmem_cache *s);
  * f.e. add ____cacheline_aligned_in_smp to the struct declaration
  * then the objects will be properly aligned in SMP configurations.
  */
-#define KMEM_CACHE(__struct, __flags)					\
-		kmem_cache_create(#__struct, sizeof(struct __struct),	\
-			__alignof__(struct __struct), (__flags), NULL)
+#define KMEM_CACHE(__struct, __flags)                                   \
+	__kmem_cache_create_args(#__struct, sizeof(struct __struct),    \
+			&(struct kmem_cache_args) {			\
+				.align	= __alignof__(struct __struct), \
+			}, (__flags))
 
 /*
  * To whitelist a single field for copying to/from usercopy, use this
  * macro instead for KMEM_CACHE() above.
  */
-#define KMEM_CACHE_USERCOPY(__struct, __flags, __field)			\
-		kmem_cache_create_usercopy(#__struct,			\
-			sizeof(struct __struct),			\
-			__alignof__(struct __struct), (__flags),	\
-			offsetof(struct __struct, __field),		\
-			sizeof_field(struct __struct, __field), NULL)
+#define KMEM_CACHE_USERCOPY(__struct, __flags, __field)						\
+	__kmem_cache_create_args(#__struct, sizeof(struct __struct),				\
+			&(struct kmem_cache_args) {						\
+				.align		= __alignof__(struct __struct),			\
+				.useroffset	= offsetof(struct __struct, __field),		\
+				.usersize	= sizeof_field(struct __struct, __field),	\
+			}, (__flags))
 
 /*
  * Common kmalloc functions provided by all allocators

io_uring/io_uring.c

@@ -3638,6 +3638,11 @@ SYSCALL_DEFINE2(io_uring_setup, u32, entries,
 
 static int __init io_uring_init(void)
 {
+	struct kmem_cache_args kmem_args = {
+		.useroffset = offsetof(struct io_kiocb, cmd.data),
+		.usersize = sizeof_field(struct io_kiocb, cmd.data),
+	};
+
 #define __BUILD_BUG_VERIFY_OFFSET_SIZE(stype, eoffset, esize, ename) do { \
 	BUILD_BUG_ON(offsetof(stype, ename) != eoffset); \
 	BUILD_BUG_ON(sizeof_field(stype, ename) != esize); \
@@ -3722,12 +3727,9 @@ static int __init io_uring_init(void)
 	 * range, and HARDENED_USERCOPY will complain if we haven't
 	 * correctly annotated this range.
 	 */
-	req_cachep = kmem_cache_create_usercopy("io_kiocb",
-				sizeof(struct io_kiocb), 0,
-				SLAB_HWCACHE_ALIGN | SLAB_PANIC |
-				SLAB_ACCOUNT | SLAB_TYPESAFE_BY_RCU,
-				offsetof(struct io_kiocb, cmd.data),
-				sizeof_field(struct io_kiocb, cmd.data), NULL);
+	req_cachep = kmem_cache_create("io_kiocb", sizeof(struct io_kiocb), &kmem_args,
+				SLAB_HWCACHE_ALIGN | SLAB_PANIC | SLAB_ACCOUNT |
+				SLAB_TYPESAFE_BY_RCU);
 	io_buf_cachep = KMEM_CACHE(io_buffer,
 					  SLAB_HWCACHE_ALIGN | SLAB_PANIC | SLAB_ACCOUNT);
 

mm/slab.h

@@ -422,7 +422,9 @@ kmalloc_slab(size_t size, kmem_buckets *b, gfp_t flags, unsigned long caller)
 gfp_t kmalloc_fix_flags(gfp_t flags);
 
 /* Functions provided by the slab allocators */
-int __kmem_cache_create(struct kmem_cache *, slab_flags_t flags);
+int do_kmem_cache_create(struct kmem_cache *s, const char *name,
+			 unsigned int size, struct kmem_cache_args *args,
+			 slab_flags_t flags);
 
 void __init kmem_cache_init(void);
 extern void create_boot_cache(struct kmem_cache *, const char *name,

mm/slab_common.c

@@ -215,32 +215,29 @@ struct kmem_cache *find_mergeable(unsigned int size, unsigned int align,
 }
 
 static struct kmem_cache *create_cache(const char *name,
-		unsigned int object_size, unsigned int align,
-		slab_flags_t flags, unsigned int useroffset,
-		unsigned int usersize, void (*ctor)(void *),
-		struct kmem_cache *root_cache)
+				       unsigned int object_size,
+				       struct kmem_cache_args *args,
+				       slab_flags_t flags)
 {
 	struct kmem_cache *s;
 	int err;
 
-	if (WARN_ON(useroffset + usersize > object_size))
-		useroffset = usersize = 0;
+	if (WARN_ON(args->useroffset + args->usersize > object_size))
+		args->useroffset = args->usersize = 0;
+
+	/* If a custom freelist pointer is requested make sure it's sane. */
+	err = -EINVAL;
+	if (args->use_freeptr_offset &&
+	    (args->freeptr_offset >= object_size ||
+	     !(flags & SLAB_TYPESAFE_BY_RCU) ||
+	     !IS_ALIGNED(args->freeptr_offset, sizeof(freeptr_t))))
+		goto out;
 
 	err = -ENOMEM;
 	s = kmem_cache_zalloc(kmem_cache, GFP_KERNEL);
 	if (!s)
 		goto out;
-
-	s->name = name;
-	s->size = s->object_size = object_size;
-	s->align = align;
-	s->ctor = ctor;
-#ifdef CONFIG_HARDENED_USERCOPY
-	s->useroffset = useroffset;
-	s->usersize = usersize;
-#endif
-
-	err = __kmem_cache_create(s, flags);
+	err = do_kmem_cache_create(s, name, object_size, args, flags);
 	if (err)
 		goto out_free_cache;
 
@@ -255,39 +252,24 @@ static struct kmem_cache *create_cache(const char *name,
 }
 
 /**
- * kmem_cache_create_usercopy - Create a cache with a region suitable
- * for copying to userspace
+ * __kmem_cache_create_args - Create a kmem cache.
  * @name: A string which is used in /proc/slabinfo to identify this cache.
- * @size: The size of objects to be created in this cache.
- * @align: The required alignment for the objects.
- * @flags: SLAB flags
- * @useroffset: Usercopy region offset
- * @usersize: Usercopy region size
- * @ctor: A constructor for the objects.
+ * @object_size: The size of objects to be created in this cache.
+ * @args: Additional arguments for the cache creation (see
+ *        &struct kmem_cache_args).
+ * @flags: See %SLAB_* flags for an explanation of individual @flags.
  *
- * Cannot be called within a interrupt, but can be interrupted.
- * The @ctor is run when new pages are allocated by the cache.
+ * Not to be called directly, use the kmem_cache_create() wrapper with the same
+ * parameters.
  *
- * The flags are
- *
- * %SLAB_POISON - Poison the slab with a known test pattern (a5a5a5a5)
- * to catch references to uninitialised memory.
- *
- * %SLAB_RED_ZONE - Insert `Red` zones around the allocated memory to check
- * for buffer overruns.
- *
- * %SLAB_HWCACHE_ALIGN - Align the objects in this cache to a hardware
- * cacheline.  This can be beneficial if you're counting cycles as closely
- * as davem.
+ * Context: Cannot be called within a interrupt, but can be interrupted.
  *
  * Return: a pointer to the cache on success, NULL on failure.
  */
-struct kmem_cache *
-kmem_cache_create_usercopy(const char *name,
-		  unsigned int size, unsigned int align,
-		  slab_flags_t flags,
-		  unsigned int useroffset, unsigned int usersize,
-		  void (*ctor)(void *))
+struct kmem_cache *__kmem_cache_create_args(const char *name,
+					    unsigned int object_size,
+					    struct kmem_cache_args *args,
+					    slab_flags_t flags)
 {
 	struct kmem_cache *s = NULL;
 	const char *cache_name;
@@ -309,7 +291,7 @@ kmem_cache_create_usercopy(const char *name,
 
 	mutex_lock(&slab_mutex);
 
-	err = kmem_cache_sanity_check(name, size);
+	err = kmem_cache_sanity_check(name, object_size);
 	if (err) {
 		goto out_unlock;
 	}
@@ -330,12 +312,14 @@ kmem_cache_create_usercopy(const char *name,
 
 	/* Fail closed on bad usersize of useroffset values. */
 	if (!IS_ENABLED(CONFIG_HARDENED_USERCOPY) ||
-	    WARN_ON(!usersize && useroffset) ||
-	    WARN_ON(size < usersize || size - usersize < useroffset))
-		usersize = useroffset = 0;
-
-	if (!usersize)
-		s = __kmem_cache_alias(name, size, align, flags, ctor);
+	    WARN_ON(!args->usersize && args->useroffset) ||
+	    WARN_ON(object_size < args->usersize ||
+		    object_size - args->usersize < args->useroffset))
+		args->usersize = args->useroffset = 0;
+
+	if (!args->usersize)
+		s = __kmem_cache_alias(name, object_size, args->align, flags,
+				       args->ctor);
 	if (s)
 		goto out_unlock;
 
@@ -345,9 +329,8 @@ kmem_cache_create_usercopy(const char *name,
 		goto out_unlock;
 	}
 
-	s = create_cache(cache_name, size,
-			 calculate_alignment(flags, align, size),
-			 flags, useroffset, usersize, ctor, NULL);
+	args->align = calculate_alignment(flags, args->align, object_size);
+	s = create_cache(cache_name, object_size, args, flags);
 	if (IS_ERR(s)) {
 		err = PTR_ERR(s);
 		kfree_const(cache_name);
@@ -369,41 +352,7 @@ kmem_cache_create_usercopy(const char *name,
 	}
 	return s;
 }
-EXPORT_SYMBOL(kmem_cache_create_usercopy);
-
-/**
- * kmem_cache_create - Create a cache.
- * @name: A string which is used in /proc/slabinfo to identify this cache.
- * @size: The size of objects to be created in this cache.
- * @align: The required alignment for the objects.
- * @flags: SLAB flags
- * @ctor: A constructor for the objects.
- *
- * Cannot be called within a interrupt, but can be interrupted.
- * The @ctor is run when new pages are allocated by the cache.
- *
- * The flags are
- *
- * %SLAB_POISON - Poison the slab with a known test pattern (a5a5a5a5)
- * to catch references to uninitialised memory.
- *
- * %SLAB_RED_ZONE - Insert `Red` zones around the allocated memory to check
- * for buffer overruns.
- *
- * %SLAB_HWCACHE_ALIGN - Align the objects in this cache to a hardware
- * cacheline.  This can be beneficial if you're counting cycles as closely
- * as davem.
- *
- * Return: a pointer to the cache on success, NULL on failure.
- */
-struct kmem_cache *
-kmem_cache_create(const char *name, unsigned int size, unsigned int align,
-		slab_flags_t flags, void (*ctor)(void *))
-{
-	return kmem_cache_create_usercopy(name, size, align, flags, 0, 0,
-					  ctor);
-}
-EXPORT_SYMBOL(kmem_cache_create);
+EXPORT_SYMBOL(__kmem_cache_create_args);
 
 static struct kmem_cache *kmem_buckets_cache __ro_after_init;
 
@@ -689,9 +638,7 @@ void __init create_boot_cache(struct kmem_cache *s, const char *name,
 {
 	int err;
 	unsigned int align = ARCH_KMALLOC_MINALIGN;
-
-	s->name = name;
-	s->size = s->object_size = size;
+	struct kmem_cache_args kmem_args = {};
 
 	/*
 	 * kmalloc caches guarantee alignment of at least the largest
@@ -700,14 +647,14 @@ void __init create_boot_cache(struct kmem_cache *s, const char *name,
 	 */
 	if (flags & SLAB_KMALLOC)
 		align = max(align, 1U << (ffs(size) - 1));
-	s->align = calculate_alignment(flags, align, size);
+	kmem_args.align = calculate_alignment(flags, align, size);
 
 #ifdef CONFIG_HARDENED_USERCOPY
-	s->useroffset = useroffset;
-	s->usersize = usersize;
+	kmem_args.useroffset = useroffset;
+	kmem_args.usersize = usersize;
 #endif
 
-	err = __kmem_cache_create(s, flags);
+	err = do_kmem_cache_create(s, name, size, &kmem_args, flags);
 
 	if (err)
 		panic("Creation of kmalloc slab %s size=%u failed. Reason %d\n",

mm/slub.c

@@ -465,12 +465,6 @@ static struct workqueue_struct *flushwq;
  * 			Core slab cache functions
  *******************************************************************/
 
-/*
- * freeptr_t represents a SLUB freelist pointer, which might be encoded
- * and not dereferenceable if CONFIG_SLAB_FREELIST_HARDENED is enabled.
- */
-typedef struct { unsigned long v; } freeptr_t;
-
 /*
  * Returns freelist pointer (ptr). With hardening, this is obfuscated
  * with an XOR of the address where the pointer is held and a per-cache
@@ -4019,6 +4013,8 @@ static void *__slab_alloc_node(struct kmem_cache *s,
 /*
  * If the object has been wiped upon free, make sure it's fully initialized by
  * zeroing out freelist pointer.
+ *
+ * Note that we also wipe custom freelist pointers.
  */
 static __always_inline void maybe_wipe_obj_freeptr(struct kmem_cache *s,
 						   void *obj)
@@ -5282,7 +5278,7 @@ static void set_cpu_partial(struct kmem_cache *s)
  * calculate_sizes() determines the order and the distribution of data within
  * a slab object.
  */
-static int calculate_sizes(struct kmem_cache *s)
+static int calculate_sizes(struct kmem_cache_args *args, struct kmem_cache *s)
 {
 	slab_flags_t flags = s->flags;
 	unsigned int size = s->object_size;
@@ -5323,7 +5319,8 @@ static int calculate_sizes(struct kmem_cache *s)
 	 */
 	s->inuse = size;
 
-	if ((flags & (SLAB_TYPESAFE_BY_RCU | SLAB_POISON)) || s->ctor ||
+	if (((flags & SLAB_TYPESAFE_BY_RCU) && !args->use_freeptr_offset) ||
+	    (flags & SLAB_POISON) || s->ctor ||
 	    ((flags & SLAB_RED_ZONE) &&
 	     (s->object_size < sizeof(void *) || slub_debug_orig_size(s)))) {
 		/*
@@ -5344,6 +5341,8 @@ static int calculate_sizes(struct kmem_cache *s)
 		 */
 		s->offset = size;
 		size += sizeof(void *);
+	} else if ((flags & SLAB_TYPESAFE_BY_RCU) && args->use_freeptr_offset) {
+		s->offset = args->freeptr_offset;
 	} else {
 		/*
 		 * Store freelist pointer near middle of object to keep
@@ -5418,65 +5417,6 @@ static int calculate_sizes(struct kmem_cache *s)
 	return !!oo_objects(s->oo);
 }
 
-static int kmem_cache_open(struct kmem_cache *s, slab_flags_t flags)
-{
-	s->flags = kmem_cache_flags(flags, s->name);
-#ifdef CONFIG_SLAB_FREELIST_HARDENED
-	s->random = get_random_long();
-#endif
-
-	if (!calculate_sizes(s))
-		goto error;
-	if (disable_higher_order_debug) {
-		/*
-		 * Disable debugging flags that store metadata if the min slab
-		 * order increased.
-		 */
-		if (get_order(s->size) > get_order(s->object_size)) {
-			s->flags &= ~DEBUG_METADATA_FLAGS;
-			s->offset = 0;
-			if (!calculate_sizes(s))
-				goto error;
-		}
-	}
-
-#ifdef system_has_freelist_aba
-	if (system_has_freelist_aba() && !(s->flags & SLAB_NO_CMPXCHG)) {
-		/* Enable fast mode */
-		s->flags |= __CMPXCHG_DOUBLE;
-	}
-#endif
-
-	/*
-	 * The larger the object size is, the more slabs we want on the partial
-	 * list to avoid pounding the page allocator excessively.
-	 */
-	s->min_partial = min_t(unsigned long, MAX_PARTIAL, ilog2(s->size) / 2);
-	s->min_partial = max_t(unsigned long, MIN_PARTIAL, s->min_partial);
-
-	set_cpu_partial(s);
-
-#ifdef CONFIG_NUMA
-	s->remote_node_defrag_ratio = 1000;
-#endif
-
-	/* Initialize the pre-computed randomized freelist if slab is up */
-	if (slab_state >= UP) {
-		if (init_cache_random_seq(s))
-			goto error;
-	}
-
-	if (!init_kmem_cache_nodes(s))
-		goto error;
-
-	if (alloc_kmem_cache_cpus(s))
-		return 0;
-
-error:
-	__kmem_cache_release(s);
-	return -EINVAL;
-}
-
 static void list_slab_objects(struct kmem_cache *s, struct slab *slab,
 			      const char *text)
 {
@@ -6030,28 +5970,90 @@ __kmem_cache_alias(const char *name, unsigned int size, unsigned int align,
 	return s;
 }
 
-int __kmem_cache_create(struct kmem_cache *s, slab_flags_t flags)
+int do_kmem_cache_create(struct kmem_cache *s, const char *name,
+			 unsigned int size, struct kmem_cache_args *args,
+			 slab_flags_t flags)
 {
-	int err;
+	int err = -EINVAL;
 
-	err = kmem_cache_open(s, flags);
-	if (err)
-		return err;
+	s->name = name;
+	s->size = s->object_size = size;
+
+	s->flags = kmem_cache_flags(flags, s->name);
+#ifdef CONFIG_SLAB_FREELIST_HARDENED
+	s->random = get_random_long();
+#endif
+	s->align = args->align;
+	s->ctor = args->ctor;
+#ifdef CONFIG_HARDENED_USERCOPY
+	s->useroffset = args->useroffset;
+	s->usersize = args->usersize;
+#endif
+
+	if (!calculate_sizes(args, s))
+		goto out;
+	if (disable_higher_order_debug) {
+		/*
+		 * Disable debugging flags that store metadata if the min slab
+		 * order increased.
+		 */
+		if (get_order(s->size) > get_order(s->object_size)) {
+			s->flags &= ~DEBUG_METADATA_FLAGS;
+			s->offset = 0;
+			if (!calculate_sizes(args, s))
+				goto out;
+		}
+	}
+
+#ifdef system_has_freelist_aba
+	if (system_has_freelist_aba() && !(s->flags & SLAB_NO_CMPXCHG)) {
+		/* Enable fast mode */
+		s->flags |= __CMPXCHG_DOUBLE;
+	}
+#endif
+
+	/*
+	 * The larger the object size is, the more slabs we want on the partial
+	 * list to avoid pounding the page allocator excessively.
+	 */
+	s->min_partial = min_t(unsigned long, MAX_PARTIAL, ilog2(s->size) / 2);
+	s->min_partial = max_t(unsigned long, MIN_PARTIAL, s->min_partial);
+
+	set_cpu_partial(s);
+
+#ifdef CONFIG_NUMA
+	s->remote_node_defrag_ratio = 1000;
+#endif
+
+	/* Initialize the pre-computed randomized freelist if slab is up */
+	if (slab_state >= UP) {
+		if (init_cache_random_seq(s))
+			goto out;
+	}
+
+	if (!init_kmem_cache_nodes(s))
+		goto out;
+
+	if (!alloc_kmem_cache_cpus(s))
+		goto out;
 
 	/* Mutex is not taken during early boot */
-	if (slab_state <= UP)
-		return 0;
+	if (slab_state <= UP) {
+		err = 0;
+		goto out;
+	}
 
 	err = sysfs_slab_add(s);
-	if (err) {
-		__kmem_cache_release(s);
-		return err;
-	}
+	if (err)
+		goto out;
 
 	if (s->flags & SLAB_STORE_USER)
 		debugfs_slab_add(s);
 
-	return 0;
+out:
+	if (err)
+		__kmem_cache_release(s);
+	return err;
 }
 
 #ifdef SLAB_SUPPORTS_SYSFS

net/core/sock.c

@@ -3429,7 +3429,7 @@ static void sock_def_destruct(struct sock *sk)
 void sk_send_sigurg(struct sock *sk)
 {
 	if (sk->sk_socket && sk->sk_socket->file)
-		if (send_sigurg(&sk->sk_socket->file->f_owner))
+		if (send_sigurg(sk->sk_socket->file))
 			sk_wake_async(sk, SOCK_WAKE_URG, POLL_PRI);
 }
 EXPORT_SYMBOL(sk_send_sigurg);

security/selinux/hooks.c

@@ -3950,7 +3950,7 @@ static int selinux_file_send_sigiotask(struct task_struct *tsk,
 	struct file_security_struct *fsec;
 
 	/* struct fown_struct is never outside the context of a struct file */
-	file = container_of(fown, struct file, f_owner);
+	file = fown->file;
 
 	fsec = selinux_file(file);
 

security/smack/smack_lsm.c

@@ -1950,7 +1950,7 @@ static int smack_file_send_sigiotask(struct task_struct *tsk,
 	/*
 	 * struct fown_struct is never outside the context of a struct file
 	 */
-	file = container_of(fown, struct file, f_owner);
+	file = fown->file;
 
 	/* we don't log here as rc can be overriden */
 	blob = smack_file(file);