From b835996f628eadb55c5fb222ba46fe9395bf73c7 Mon Sep 17 00:00:00 2001 From: Dipankar Sarma Date: Fri, 9 Sep 2005 13:04:14 -0700 Subject: [PATCH] files: lock-free fd look-up With the use of RCU in files structure, the look-up of files using fds can now be lock-free. The lookup is protected by rcu_read_lock()/rcu_read_unlock(). This patch changes the readers to use lock-free lookup. Signed-off-by: Maneesh Soni Signed-off-by: Ravikiran Thirumalai Signed-off-by: Dipankar Sarma Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/fcntl.c | 4 ++-- fs/proc/base.c | 29 +++++++++++++++-------------- fs/select.c | 13 ++++++++++--- 3 files changed, 27 insertions(+), 19 deletions(-) (limited to 'fs') diff --git a/fs/fcntl.c b/fs/fcntl.c index d2f3ed8acd932e..863b46e0d78a65 100644 --- a/fs/fcntl.c +++ b/fs/fcntl.c @@ -40,10 +40,10 @@ static inline int get_close_on_exec(unsigned int fd) struct files_struct *files = current->files; struct fdtable *fdt; int res; - spin_lock(&files->file_lock); + rcu_read_lock(); fdt = files_fdtable(files); res = FD_ISSET(fd, fdt->close_on_exec); - spin_unlock(&files->file_lock); + rcu_read_unlock(); return res; } diff --git a/fs/proc/base.c b/fs/proc/base.c index d0087a0b024be4..23db452ab428c9 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -62,6 +62,7 @@ #include #include #include +#include #include #include #include @@ -283,16 +284,16 @@ static int proc_fd_link(struct inode *inode, struct dentry **dentry, struct vfsm files = get_files_struct(task); if (files) { - spin_lock(&files->file_lock); + rcu_read_lock(); file = fcheck_files(files, fd); if (file) { *mnt = mntget(file->f_vfsmnt); *dentry = dget(file->f_dentry); - spin_unlock(&files->file_lock); + rcu_read_unlock(); put_files_struct(files); return 0; } - spin_unlock(&files->file_lock); + rcu_read_unlock(); put_files_struct(files); } return -ENOENT; @@ -1062,7 +1063,7 @@ static int proc_readfd(struct file * filp, void * dirent, filldir_t filldir) files = get_files_struct(p); if (!files) goto out; - spin_lock(&files->file_lock); + rcu_read_lock(); fdt = files_fdtable(files); for (fd = filp->f_pos-2; fd < fdt->max_fds; @@ -1071,7 +1072,7 @@ static int proc_readfd(struct file * filp, void * dirent, filldir_t filldir) if (!fcheck_files(files, fd)) continue; - spin_unlock(&files->file_lock); + rcu_read_unlock(); j = NUMBUF; i = fd; @@ -1083,12 +1084,12 @@ static int proc_readfd(struct file * filp, void * dirent, filldir_t filldir) ino = fake_ino(tid, PROC_TID_FD_DIR + fd); if (filldir(dirent, buf+j, NUMBUF-j, fd+2, ino, DT_LNK) < 0) { - spin_lock(&files->file_lock); + rcu_read_lock(); break; } - spin_lock(&files->file_lock); + rcu_read_lock(); } - spin_unlock(&files->file_lock); + rcu_read_unlock(); put_files_struct(files); } out: @@ -1263,9 +1264,9 @@ static int tid_fd_revalidate(struct dentry *dentry, struct nameidata *nd) files = get_files_struct(task); if (files) { - spin_lock(&files->file_lock); + rcu_read_lock(); if (fcheck_files(files, fd)) { - spin_unlock(&files->file_lock); + rcu_read_unlock(); put_files_struct(files); if (task_dumpable(task)) { inode->i_uid = task->euid; @@ -1277,7 +1278,7 @@ static int tid_fd_revalidate(struct dentry *dentry, struct nameidata *nd) security_task_to_inode(task, inode); return 1; } - spin_unlock(&files->file_lock); + rcu_read_unlock(); put_files_struct(files); } d_drop(dentry); @@ -1369,7 +1370,7 @@ static struct dentry *proc_lookupfd(struct inode * dir, struct dentry * dentry, if (!files) goto out_unlock; inode->i_mode = S_IFLNK; - spin_lock(&files->file_lock); + rcu_read_lock(); file = fcheck_files(files, fd); if (!file) goto out_unlock2; @@ -1377,7 +1378,7 @@ static struct dentry *proc_lookupfd(struct inode * dir, struct dentry * dentry, inode->i_mode |= S_IRUSR | S_IXUSR; if (file->f_mode & 2) inode->i_mode |= S_IWUSR | S_IXUSR; - spin_unlock(&files->file_lock); + rcu_read_unlock(); put_files_struct(files); inode->i_op = &proc_pid_link_inode_operations; inode->i_size = 64; @@ -1387,7 +1388,7 @@ static struct dentry *proc_lookupfd(struct inode * dir, struct dentry * dentry, return NULL; out_unlock2: - spin_unlock(&files->file_lock); + rcu_read_unlock(); put_files_struct(files); out_unlock: iput(inode); diff --git a/fs/select.c b/fs/select.c index 2e56325c73c479..f10a10317d5494 100644 --- a/fs/select.c +++ b/fs/select.c @@ -22,6 +22,7 @@ #include /* for STICKY_TIMEOUTS */ #include #include +#include #include @@ -185,9 +186,9 @@ int do_select(int n, fd_set_bits *fds, long *timeout) int retval, i; long __timeout = *timeout; - spin_lock(¤t->files->file_lock); + rcu_read_lock(); retval = max_select_fd(n, fds); - spin_unlock(¤t->files->file_lock); + rcu_read_unlock(); if (retval < 0) return retval; @@ -329,8 +330,10 @@ sys_select(int n, fd_set __user *inp, fd_set __user *outp, fd_set __user *exp, s goto out_nofds; /* max_fdset can increase, so grab it once to avoid race */ + rcu_read_lock(); fdt = files_fdtable(current->files); max_fdset = fdt->max_fdset; + rcu_read_unlock(); if (n > max_fdset) n = max_fdset; @@ -469,10 +472,14 @@ asmlinkage long sys_poll(struct pollfd __user * ufds, unsigned int nfds, long ti struct poll_list *head; struct poll_list *walk; struct fdtable *fdt; + int max_fdset; /* Do a sanity check on nfds ... */ + rcu_read_lock(); fdt = files_fdtable(current->files); - if (nfds > fdt->max_fdset && nfds > OPEN_MAX) + max_fdset = fdt->max_fdset; + rcu_read_unlock(); + if (nfds > max_fdset && nfds > OPEN_MAX) return -EINVAL; if (timeout) { -- cgit 1.2.3-korg