namei cleanups
Subsystems (7)
Al Viro
maintainerPATCH 1/1 namei cleanups
Small namei.c patch series, mostly to simplify the rules
for nameidata state. It's in #work.namei and it's actually from the previous cycle - didn't post it for review in time...
Changes visible outside of fs/namei.c: file_open_root()
calling conventions change, some freed bits in LOOKUP_... space.
Al Viro
maintainerPATCH 4/4 namei: make sure nd->depth is always valid
In reply to Al ViroZero it in set_nameidata() rather than in path_init(). That way it always matches the number of valid nd->stack[] entries. Since terminate_walk() does zero it (after having emptied the stack), we don't need to reinitialize it in subsequent path_init().
--- fs/namei.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
fs/namei.c +1 -1
@@ -581,6 +581,7 @@ static void __set_nameidata(struct nameidata *p, int dfd, struct filename *name) {struct nameidata *old = current->nameidata; p->stack = p->internal; + p->depth =0 ; p->dfd = dfd; p->name = name; p->path.mnt =NULL ;
@@ -2320,7 +2321,6 @@ static const char *path_init(struct nameidata *nd, unsigned flags) nd->flags = flags; nd->state |=ND_JUMPED ; - nd->depth =0 ; nd->m_seq =__read_seqcount_begin (&mount_lock.seqcount); nd->r_seq =__read_seqcount_begin (&rename_lock.seqcount);
Signature · 1 line
2.11.0
Al Viro
maintainerPATCH 2/4 take LOOKUP_{ROOT,ROOT_GRABBED,JUMPED} out of LOOKUP_... space
In reply to Al ViroSeparate field in nameidata (nd->state) holding the flags that should be internal-only - that way we both get some spare bits in LOOKUP_... and get simpler rules for nd->root lifetime rules, since we can set the replacement of LOOKUP_ROOT (ND_ROOT_PRESET) at the same time we set nd->root.
--- Documentation/filesystems/path-lookup.rst | 6 ++-- fs/namei.c | 54 ++++++++++++++++++------------- fs/nfs/nfstrace.h | 4 --- include/linux/namei.h | 3 -- 4 files changed, 34 insertions(+), 33 deletions(-)
Documentation/filesystems/path-lookup.rst +3 -3
@@ -1321,18 +1321,18 @@ to lookup: RCU-walk, REF-walk, and REF-walk with forced revalidation. yet. This is primarily used to tell the audit subsystem the full context of a particular access being audited. -``LOOKUP_ROOT`` indicates that the ``root`` field in the ``nameidata`` was +``ND_ROOT_PRESET`` indicates that the ``root`` field in the ``nameidata`` was provided by the caller, so it shouldn't be released when it is no longer needed. -``LOOKUP_JUMPED`` means that the current dentry was chosen not because +``ND_JUMPED`` means that the current dentry was chosen not because it had the right name but for some other reason. This happens when following "``..``", following a symlink to ``/``, crossing a mount point or accessing a "``/proc/$PID/fd/$FD``" symlink (also known as a "magic link"). In this case the filesystem has not been asked to revalidate the name (with ``d_revalidate()``). In such cases the inode may still need to be revalidated, so ``d_op->d_weak_revalidate()`` is called if -``LOOKUP_JUMPED`` is set when the look completes - which may be at the +``ND_JUMPED`` is set when the look completes - which may be at the final component or, when creating, unlinking, or renaming, at the penultimate component. Resolution-restriction flags
fs/namei.c +31 -23
186 lines not on this page · raw message
fs/nfs/nfstrace.h +0 -4
@@ -271,8 +271,6 @@ TRACE_DEFINE_ENUM(LOOKUP_OPEN);TRACE_DEFINE_ENUM (LOOKUP_CREATE );TRACE_DEFINE_ENUM (LOOKUP_EXCL );TRACE_DEFINE_ENUM (LOOKUP_RENAME_TARGET ); -TRACE_DEFINE_ENUM (LOOKUP_JUMPED ); -TRACE_DEFINE_ENUM (LOOKUP_ROOT );TRACE_DEFINE_ENUM (LOOKUP_EMPTY );TRACE_DEFINE_ENUM (LOOKUP_DOWN );
@@ -288,8 +286,6 @@ TRACE_DEFINE_ENUM(LOOKUP_DOWN); { LOOKUP_CREATE, "CREATE" }, \ { LOOKUP_EXCL, "EXCL" }, \ { LOOKUP_RENAME_TARGET, "RENAME_TARGET" }, \ - { LOOKUP_JUMPED, "JUMPED" }, \ - { LOOKUP_ROOT, "ROOT" }, \ { LOOKUP_EMPTY, "EMPTY" }, \ { LOOKUP_DOWN, "DOWN" })
include/linux/namei.h +0 -3
@@ -36,9 +36,6 @@ enum {LAST_NORM, LAST_ROOT, LAST_DOT, LAST_DOTDOT};/* internal use only */ #define LOOKUP_PARENT 0x0010 -#define LOOKUP_JUMPED 0x1000 -#define LOOKUP_ROOT 0x2000 -#define LOOKUP_ROOT_GRABBED 0x0008/* Scoping flags for lookup. */ #define LOOKUP_NO_SYMLINKS 0x010000/* No symlink crossing. */
Signature · 1 line
2.11.0
Al Viro
maintainerPATCH 3/4 teach set_nameidata() to handle setting the root as well
In reply to Al ViroThat way we don't need the callers to mess with manually setting any fields of nameidata instances. Old set_nameidata() gets renamed (__set_nameidata()), new becomes an inlined helper that takes a struct path pointer and deals with setting nd->root and putting ND_ROOT_PRESET in nd->state when new argument is non-NULL.
--- fs/namei.c | 28 ++++++++++++++++------------ 1 file changed, 16 insertions(+), 12 deletions(-)
fs/namei.c +16 -12
@@ -577,7 +577,7 @@ struct nameidata {#define ND_ROOT_GRABBED 2#define ND_JUMPED 4 -static void set_nameidata (struct nameidata *p,int dfd,struct filename *name) +static void __set_nameidata (struct nameidata *p,int dfd,struct filename *name) {struct nameidata *old = current->nameidata; p->stack = p->internal;
@@ -587,10 +587,20 @@ static void set_nameidata(struct nameidata *p, int dfd, struct filename *name) p->path.dentry =NULL ; p->total_link_count = old ? old->total_link_count :0 ; p->saved = old; - p->state =0 ; current->nameidata = p; } +static inline void set_nameidata (struct nameidata *p,int dfd,struct filename *name, +const struct path *root) +{ +__set_nameidata (p, dfd, name); + p->state =0 ; +if (unlikely (root)) { + p->state =ND_ROOT_PRESET ; + p->root = *root; + } +} +static void restore_nameidata (void ) {struct nameidata *now = current->nameidata, *old = now->saved;
@@ -2454,11 +2464,7 @@ int filename_lookup(int dfd, struct filename *name, unsigned flags,struct nameidata nd;if (IS_ERR (name))return PTR_ERR (name); -set_nameidata (&nd, dfd, name); -if (unlikely (root)) { - nd.root = *root; - nd.state =ND_ROOT_PRESET ; - } +set_nameidata (&nd, dfd, name, root); retval =path_lookupat (&nd, flags |LOOKUP_RCU , path);if (unlikely (retval == -ECHILD )) retval =path_lookupat (&nd, flags, path);
@@ -2499,7 +2505,7 @@ static struct filename *filename_parentat(int dfd, struct filename *name,if (IS_ERR (name))return name; -set_nameidata (&nd, dfd, name); +set_nameidata (&nd, dfd, name,NULL ); retval =path_parentat (&nd, flags |LOOKUP_RCU , parent);if (unlikely (retval == -ECHILD )) retval =path_parentat (&nd, flags, parent);
@@ -3530,7 +3536,7 @@ struct file *do_filp_open(int dfd, struct filename *pathname,int flags = op->lookup_flags;struct file *filp; -set_nameidata (&nd, dfd, pathname); +set_nameidata (&nd, dfd, pathname,NULL ); filp =path_openat (&nd, op, flags |LOOKUP_RCU );if (unlikely (filp ==ERR_PTR (-ECHILD ))) filp =path_openat (&nd, op, flags);
@@ -3555,9 +3561,7 @@ struct file *do_file_open_root(const struct path *root,if (IS_ERR (filename))return ERR_CAST (filename); -set_nameidata (&nd,-1 , filename); - nd.root = *root; - nd.state =ND_ROOT_PRESET ; +set_nameidata (&nd,-1 , filename, root); file =path_openat (&nd, op, flags |LOOKUP_RCU );if (unlikely (file ==ERR_PTR (-ECHILD ))) file =path_openat (&nd, op, flags);
Signature · 1 line
2.11.0
Al Viro
maintainerPATCH 1/4 switch file_open_root() to struct path
In reply to Al Viro... and provide file_open_root_mnt(), using the root of given mount.
--- Documentation/filesystems/porting.rst | 9 +++++++++ arch/um/drivers/mconsole_kern.c | 2 +- fs/coredump.c | 4 ++-- fs/fhandle.c | 2 +- fs/internal.h | 2 +- fs/kernel_read_file.c | 2 +- fs/namei.c | 8 +++----- fs/open.c | 4 ++-- fs/proc/proc_sysctl.c | 2 +- include/linux/fs.h | 8 +++++++- kernel/usermode_driver.c | 2 +- 11 files changed, 29 insertions(+), 16 deletions(-)
Documentation/filesystems/porting.rst +9 -0
@@ -890,3 +890,12 @@ been called or returned with non -EIOCBQUEUED code. mnt_want_write_file() can now only be paired with mnt_drop_write_file(), whereas previously it could be paired with mnt_drop_write() as well. + +--- + +**mandatory** + +Calling conventions for file_open_root() changed; now it takes struct path * +instead of passing mount and dentry separately. For callers that used to +pass <mnt, mnt->mnt_root> pair (i.e. the root of given mount), a new helper +is provided - file_open_root_mnt(). In-tree users adjusted.
arch/um/drivers/mconsole_kern.c +1 -1
@@ -140,7 +140,7 @@ void mconsole_proc(struct mc_request *req)mconsole_reply (req,"Proc not available" ,1 ,0 ); goto out; } - file =file_open_root (mnt->mnt_root, mnt, ptr,O_RDONLY ,0 ); + file =file_open_root_mnt (mnt, ptr,O_RDONLY ,0 );if (IS_ERR (file)) {mconsole_reply (req,"Failed to open file" ,1 ,0 );printk (KERN_ERR "open /proc/%s: %ld\n" , ptr,PTR_ERR (file));
fs/coredump.c +2 -2
@@ -755,8 +755,8 @@ void do_coredump(const kernel_siginfo_t *siginfo)task_lock (&init_task);get_fs_root (init_task.fs, &root);task_unlock (&init_task); - cprm.file =file_open_root (root.dentry, root.mnt, - cn.corename, open_flags,0600 ); + cprm.file =file_open_root (&root, cn.corename, + open_flags,0600 );path_put (&root); }else { cprm.file =filp_open (cn.corename, open_flags,0600 );
fs/fhandle.c +1 -1
@@ -229,7 +229,7 @@ static long do_handle_open(int mountdirfd, struct file_handle __user *ufh,path_put (&path);return fd; } - file =file_open_root (path.dentry, path.mnt,"" , open_flag,0 ); + file =file_open_root (&path,"" , open_flag,0 );if (IS_ERR (file)) {put_unused_fd (fd); retval =PTR_ERR (file);
fs/internal.h +1 -1
@@ -129,7 +129,7 @@ struct open_flags { };extern struct file *do_filp_open (int dfd,struct filename *pathname,const struct open_flags *op); -extern struct file *do_file_open_root (struct dentry *,struct vfsmount *, +extern struct file *do_file_open_root (const struct path *,const char *,const struct open_flags *);extern struct open_how build_open_how (int flags,umode_t mode);extern int build_open_flags (const struct open_how *how,struct open_flags *op);
fs/kernel_read_file.c +1 -1
@@ -160,7 +160,7 @@ int kernel_read_file_from_path_initns(const char *path, loff_t offset,get_fs_root (init_task.fs, &root);task_unlock (&init_task); - file =file_open_root (root.dentry, root.mnt, path,O_RDONLY ,0 ); + file =file_open_root (&root, path,O_RDONLY ,0 );path_put (&root);if (IS_ERR (file))return PTR_ERR (file);
fs/namei.c +3 -5
@@ -3533,7 +3533,7 @@ struct file *do_filp_open(int dfd, struct filename *pathname,return filp; } -struct file *do_file_open_root (struct dentry *dentry,struct vfsmount *mnt, +struct file *do_file_open_root (const struct path *root,const char *name,const struct open_flags *op) {struct nameidata nd;
@@ -3541,16 +3541,14 @@ struct file *do_file_open_root(struct dentry *dentry, struct vfsmount *mnt,struct filename *filename;int flags = op->lookup_flags |LOOKUP_ROOT ; - nd.root.mnt = mnt; - nd.root.dentry = dentry; - -if (d_is_symlink (dentry) && op->intent &LOOKUP_OPEN ) +if (d_is_symlink (root->dentry) && op->intent &LOOKUP_OPEN )return ERR_PTR (-ELOOP ); filename =getname_kernel (name);if (IS_ERR (filename))return ERR_CAST (filename); + nd.root = *root;set_nameidata (&nd,-1 , filename); file =path_openat (&nd, op, flags |LOOKUP_RCU );if (unlikely (file ==ERR_PTR (-ECHILD )))
fs/open.c +2 -2
@@ -1156,7 +1156,7 @@ struct file *filp_open(const char *filename, int flags, umode_t mode) }EXPORT_SYMBOL (filp_open); -struct file *file_open_root (struct dentry *dentry,struct vfsmount *mnt, +struct file *file_open_root (const struct path *root,const char *filename,int flags,umode_t mode) {struct open_flags op;
@@ -1164,7 +1164,7 @@ struct file *file_open_root(struct dentry *dentry, struct vfsmount *mnt,int err =build_open_flags (&how, &op);if (err)return ERR_PTR (err); -return do_file_open_root (dentry, mnt, filename, &op); +return do_file_open_root (root, filename, &op); }EXPORT_SYMBOL (file_open_root);
fs/proc/proc_sysctl.c +1 -1
@@ -1806,7 +1806,7 @@ static int process_sysctl_arg(char *param, char *val,panic ("%s: Failed to allocate path for %s\n" , __func__, param);strreplace (path,'.' ,'/' ); - file =file_open_root ((*proc_mnt)->mnt_root, *proc_mnt, path,O_WRONLY ,0 ); + file =file_open_root_mnt (*proc_mnt, path,O_WRONLY ,0 );if (IS_ERR (file)) { err =PTR_ERR (file);if (err == -ENOENT )
include/linux/fs.h +7 -1
@@ -2632,8 +2632,14 @@ extern long do_sys_open(int dfd, const char __user *filename, int flags, umode_t mode);extern struct file *file_open_name (struct filename *,int ,umode_t );extern struct file *filp_open (const char *,int ,umode_t ); -extern struct file *file_open_root (struct dentry *,struct vfsmount *, +extern struct file *file_open_root (const struct path *,const char *,int ,umode_t ); +static inline struct file *file_open_root_mnt (struct vfsmount *mnt, +const char *name,int flags,umode_t mode) +{ +return file_open_root (&(struct path ){.mnt = mnt, .dentry = mnt->mnt_root}, + name, flags, mode); +}extern struct file *dentry_open (const struct path *,int ,const struct cred *);extern struct file *open_with_fake_path (const struct path *,int ,struct inode *,const struct cred *);
kernel/usermode_driver.c +1 -1
@@ -26,7 +26,7 @@ static struct vfsmount *blob_to_mnt(const void *data, size_t len, const char *naif (IS_ERR (mnt))return mnt; - file =file_open_root (mnt->mnt_root, mnt, name,O_CREAT |O_WRONLY ,0700 ); + file =file_open_root_mnt (mnt, name,O_CREAT |O_WRONLY ,0700 );if (IS_ERR (file)) {mntput (mnt);return ERR_CAST (file);
Signature · 1 line
2.11.0
Quoted 6 lines · Al Viro
Small
namei.cpatch series, mostly to simplify the rules
for nameidata state. It's in #work.namei and it's actually from the previous cycle - didn't post it for review in time...Changes visible outside of
fs/namei.c:file_open_root()
calling conventions change, some freed bits inLOOKUP_... space.
Ah that's the follow-up to the io_uring fix from last cycle. Nice cleanup.