summaryrefslogtreecommitdiffstats
path: root/fs
diff options
context:
space:
mode:
Diffstat (limited to 'fs')
-rw-r--r--fs/Kconfig4
-rw-r--r--fs/afs/file.c2
-rw-r--r--fs/dcache.c29
-rw-r--r--fs/debugfs/inode.c149
-rw-r--r--fs/devpts/inode.c85
-rw-r--r--fs/exec.c14
-rw-r--r--fs/namei.c122
-rw-r--r--fs/proc/base.c3
-rw-r--r--fs/sysfs/dir.c224
-rw-r--r--fs/sysfs/inode.c11
-rw-r--r--fs/sysfs/mount.c2
-rw-r--r--fs/sysfs/sysfs.h17
12 files changed, 526 insertions, 136 deletions
diff --git a/fs/Kconfig b/fs/Kconfig
index d621f02a3f9e..aa195265362f 100644
--- a/fs/Kconfig
+++ b/fs/Kconfig
@@ -4,6 +4,10 @@
menu "File systems"
+# Use unaligned word dcache accesses
+config DCACHE_WORD_ACCESS
+ bool
+
if BLOCK
source "fs/ext2/Kconfig"
diff --git a/fs/afs/file.c b/fs/afs/file.c
index 14d89fa58fee..8f6e9234d565 100644
--- a/fs/afs/file.c
+++ b/fs/afs/file.c
@@ -251,7 +251,7 @@ static int afs_readpages(struct file *file, struct address_space *mapping,
ASSERT(key != NULL);
vnode = AFS_FS_I(mapping->host);
- if (vnode->flags & AFS_VNODE_DELETED) {
+ if (test_bit(AFS_VNODE_DELETED, &vnode->flags)) {
_leave(" = -ESTALE");
return -ESTALE;
}
diff --git a/fs/dcache.c b/fs/dcache.c
index bcbdb33fcc20..11828de68dce 100644
--- a/fs/dcache.c
+++ b/fs/dcache.c
@@ -105,10 +105,10 @@ static unsigned int d_hash_shift __read_mostly;
static struct hlist_bl_head *dentry_hashtable __read_mostly;
static inline struct hlist_bl_head *d_hash(const struct dentry *parent,
- unsigned long hash)
+ unsigned int hash)
{
- hash += ((unsigned long) parent ^ GOLDEN_RATIO_PRIME) / L1_CACHE_BYTES;
- hash = hash ^ ((hash ^ GOLDEN_RATIO_PRIME) >> D_HASHBITS);
+ hash += (unsigned long) parent / L1_CACHE_BYTES;
+ hash = hash + (hash >> D_HASHBITS);
return dentry_hashtable + (hash & D_HASHMASK);
}
@@ -144,6 +144,28 @@ int proc_nr_dentry(ctl_table *table, int write, void __user *buffer,
static inline int dentry_cmp(const unsigned char *cs, size_t scount,
const unsigned char *ct, size_t tcount)
{
+#ifdef CONFIG_DCACHE_WORD_ACCESS
+ unsigned long a,b,mask;
+
+ if (unlikely(scount != tcount))
+ return 1;
+
+ for (;;) {
+ a = *(unsigned long *)cs;
+ b = *(unsigned long *)ct;
+ if (tcount < sizeof(unsigned long))
+ break;
+ if (unlikely(a != b))
+ return 1;
+ cs += sizeof(unsigned long);
+ ct += sizeof(unsigned long);
+ tcount -= sizeof(unsigned long);
+ if (!tcount)
+ return 0;
+ }
+ mask = ~(~0ul << tcount*8);
+ return unlikely(!!((a ^ b) & mask));
+#else
if (scount != tcount)
return 1;
@@ -155,6 +177,7 @@ static inline int dentry_cmp(const unsigned char *cs, size_t scount,
tcount--;
} while (tcount);
return 0;
+#endif
}
static void __d_free(struct rcu_head *head)
diff --git a/fs/debugfs/inode.c b/fs/debugfs/inode.c
index 956d5ddddf6e..b80bc846a15a 100644
--- a/fs/debugfs/inode.c
+++ b/fs/debugfs/inode.c
@@ -23,9 +23,13 @@
#include <linux/debugfs.h>
#include <linux/fsnotify.h>
#include <linux/string.h>
+#include <linux/seq_file.h>
+#include <linux/parser.h>
#include <linux/magic.h>
#include <linux/slab.h>
+#define DEBUGFS_DEFAULT_MODE 0755
+
static struct vfsmount *debugfs_mount;
static int debugfs_mount_count;
static bool debugfs_registered;
@@ -125,11 +129,154 @@ static inline int debugfs_positive(struct dentry *dentry)
return dentry->d_inode && !d_unhashed(dentry);
}
+struct debugfs_mount_opts {
+ uid_t uid;
+ gid_t gid;
+ umode_t mode;
+};
+
+enum {
+ Opt_uid,
+ Opt_gid,
+ Opt_mode,
+ Opt_err
+};
+
+static const match_table_t tokens = {
+ {Opt_uid, "uid=%u"},
+ {Opt_gid, "gid=%u"},
+ {Opt_mode, "mode=%o"},
+ {Opt_err, NULL}
+};
+
+struct debugfs_fs_info {
+ struct debugfs_mount_opts mount_opts;
+};
+
+static int debugfs_parse_options(char *data, struct debugfs_mount_opts *opts)
+{
+ substring_t args[MAX_OPT_ARGS];
+ int option;
+ int token;
+ char *p;
+
+ opts->mode = DEBUGFS_DEFAULT_MODE;
+
+ while ((p = strsep(&data, ",")) != NULL) {
+ if (!*p)
+ continue;
+
+ token = match_token(p, tokens, args);
+ switch (token) {
+ case Opt_uid:
+ if (match_int(&args[0], &option))
+ return -EINVAL;
+ opts->uid = option;
+ break;
+ case Opt_gid:
+ if (match_octal(&args[0], &option))
+ return -EINVAL;
+ opts->gid = option;
+ break;
+ case Opt_mode:
+ if (match_octal(&args[0], &option))
+ return -EINVAL;
+ opts->mode = option & S_IALLUGO;
+ break;
+ /*
+ * We might like to report bad mount options here;
+ * but traditionally debugfs has ignored all mount options
+ */
+ }
+ }
+
+ return 0;
+}
+
+static int debugfs_apply_options(struct super_block *sb)
+{
+ struct debugfs_fs_info *fsi = sb->s_fs_info;
+ struct inode *inode = sb->s_root->d_inode;
+ struct debugfs_mount_opts *opts = &fsi->mount_opts;
+
+ inode->i_mode &= ~S_IALLUGO;
+ inode->i_mode |= opts->mode;
+
+ inode->i_uid = opts->uid;
+ inode->i_gid = opts->gid;
+
+ return 0;
+}
+
+static int debugfs_remount(struct super_block *sb, int *flags, char *data)
+{
+ int err;
+ struct debugfs_fs_info *fsi = sb->s_fs_info;
+
+ err = debugfs_parse_options(data, &fsi->mount_opts);
+ if (err)
+ goto fail;
+
+ debugfs_apply_options(sb);
+
+fail:
+ return err;
+}
+
+static int debugfs_show_options(struct seq_file *m, struct dentry *root)
+{
+ struct debugfs_fs_info *fsi = root->d_sb->s_fs_info;
+ struct debugfs_mount_opts *opts = &fsi->mount_opts;
+
+ if (opts->uid != 0)
+ seq_printf(m, ",uid=%u", opts->uid);
+ if (opts->gid != 0)
+ seq_printf(m, ",gid=%u", opts->gid);
+ if (opts->mode != DEBUGFS_DEFAULT_MODE)
+ seq_printf(m, ",mode=%o", opts->mode);
+
+ return 0;
+}
+
+static const struct super_operations debugfs_super_operations = {
+ .statfs = simple_statfs,
+ .remount_fs = debugfs_remount,
+ .show_options = debugfs_show_options,
+};
+
static int debug_fill_super(struct super_block *sb, void *data, int silent)
{
static struct tree_descr debug_files[] = {{""}};
+ struct debugfs_fs_info *fsi;
+ int err;
+
+ save_mount_options(sb, data);
+
+ fsi = kzalloc(sizeof(struct debugfs_fs_info), GFP_KERNEL);
+ sb->s_fs_info = fsi;
+ if (!fsi) {
+ err = -ENOMEM;
+ goto fail;
+ }
+
+ err = debugfs_parse_options(data, &fsi->mount_opts);
+ if (err)
+ goto fail;
+
+ err = simple_fill_super(sb, DEBUGFS_MAGIC, debug_files);
+ if (err)
+ goto fail;
+
+ sb->s_op = &debugfs_super_operations;
+
+ debugfs_apply_options(sb);
+
+ return 0;
- return simple_fill_super(sb, DEBUGFS_MAGIC, debug_files);
+fail:
+ kfree(fsi);
+ sb->s_fs_info = NULL;
+ return err;
}
static struct dentry *debug_mount(struct file_system_type *fs_type,
diff --git a/fs/devpts/inode.c b/fs/devpts/inode.c
index c4e2a58a2e82..1c6f908e38ca 100644
--- a/fs/devpts/inode.c
+++ b/fs/devpts/inode.c
@@ -36,7 +36,61 @@
#define DEVPTS_DEFAULT_PTMX_MODE 0000
#define PTMX_MINOR 2
-extern int pty_limit; /* Config limit on Unix98 ptys */
+/*
+ * sysctl support for setting limits on the number of Unix98 ptys allocated.
+ * Otherwise one can eat up all kernel memory by opening /dev/ptmx repeatedly.
+ */
+static int pty_limit = NR_UNIX98_PTY_DEFAULT;
+static int pty_reserve = NR_UNIX98_PTY_RESERVE;
+static int pty_limit_min;
+static int pty_limit_max = INT_MAX;
+static int pty_count;
+
+static struct ctl_table pty_table[] = {
+ {
+ .procname = "max",
+ .maxlen = sizeof(int),
+ .mode = 0644,
+ .data = &pty_limit,
+ .proc_handler = proc_dointvec_minmax,
+ .extra1 = &pty_limit_min,
+ .extra2 = &pty_limit_max,
+ }, {
+ .procname = "reserve",
+ .maxlen = sizeof(int),
+ .mode = 0644,
+ .data = &pty_reserve,
+ .proc_handler = proc_dointvec_minmax,
+ .extra1 = &pty_limit_min,
+ .extra2 = &pty_limit_max,
+ }, {
+ .procname = "nr",
+ .maxlen = sizeof(int),
+ .mode = 0444,
+ .data = &pty_count,
+ .proc_handler = proc_dointvec,
+ },
+ {}
+};
+
+static struct ctl_table pty_kern_table[] = {
+ {
+ .procname = "pty",
+ .mode = 0555,
+ .child = pty_table,
+ },
+ {}
+};
+
+static struct ctl_table pty_root_table[] = {
+ {
+ .procname = "kernel",
+ .mode = 0555,
+ .child = pty_kern_table,
+ },
+ {}
+};
+
static DEFINE_MUTEX(allocated_ptys_lock);
static struct vfsmount *devpts_mnt;
@@ -49,10 +103,11 @@ struct pts_mount_opts {
umode_t mode;
umode_t ptmxmode;
int newinstance;
+ int max;
};
enum {
- Opt_uid, Opt_gid, Opt_mode, Opt_ptmxmode, Opt_newinstance,
+ Opt_uid, Opt_gid, Opt_mode, Opt_ptmxmode, Opt_newinstance, Opt_max,
Opt_err
};
@@ -63,6 +118,7 @@ static const match_table_t tokens = {
#ifdef CONFIG_DEVPTS_MULTIPLE_INSTANCES
{Opt_ptmxmode, "ptmxmode=%o"},
{Opt_newinstance, "newinstance"},
+ {Opt_max, "max=%d"},
#endif
{Opt_err, NULL}
};
@@ -109,6 +165,7 @@ static int parse_mount_options(char *data, int op, struct pts_mount_opts *opts)
opts->gid = 0;
opts->mode = DEVPTS_DEFAULT_MODE;
opts->ptmxmode = DEVPTS_DEFAULT_PTMX_MODE;
+ opts->max = NR_UNIX98_PTY_MAX;
/* newinstance makes sense only on initial mount */
if (op == PARSE_MOUNT)
@@ -152,6 +209,12 @@ static int parse_mount_options(char *data, int op, struct pts_mount_opts *opts)
if (op == PARSE_MOUNT)
opts->newinstance = 1;
break;
+ case Opt_max:
+ if (match_int(&args[0], &option) ||
+ option < 0 || option > NR_UNIX98_PTY_MAX)
+ return -EINVAL;
+ opts->max = option;
+ break;
#endif
default:
printk(KERN_ERR "devpts: called with bogus options\n");
@@ -258,6 +321,8 @@ static int devpts_show_options(struct seq_file *seq, struct dentry *root)
seq_printf(seq, ",mode=%03o", opts->mode);
#ifdef CONFIG_DEVPTS_MULTIPLE_INSTANCES
seq_printf(seq, ",ptmxmode=%03o", opts->ptmxmode);
+ if (opts->max < NR_UNIX98_PTY_MAX)
+ seq_printf(seq, ",max=%d", opts->max);
#endif
return 0;
@@ -438,6 +503,12 @@ retry:
return -ENOMEM;
mutex_lock(&allocated_ptys_lock);
+ if (pty_count >= pty_limit -
+ (fsi->mount_opts.newinstance ? pty_reserve : 0)) {
+ mutex_unlock(&allocated_ptys_lock);
+ return -ENOSPC;
+ }
+
ida_ret = ida_get_new(&fsi->allocated_ptys, &index);
if (ida_ret < 0) {
mutex_unlock(&allocated_ptys_lock);
@@ -446,11 +517,12 @@ retry:
return -EIO;
}
- if (index >= pty_limit) {
+ if (index >= fsi->mount_opts.max) {
ida_remove(&fsi->allocated_ptys, index);
mutex_unlock(&allocated_ptys_lock);
- return -EIO;
+ return -ENOSPC;
}
+ pty_count++;
mutex_unlock(&allocated_ptys_lock);
return index;
}
@@ -462,6 +534,7 @@ void devpts_kill_index(struct inode *ptmx_inode, int idx)
mutex_lock(&allocated_ptys_lock);
ida_remove(&fsi->allocated_ptys, idx);
+ pty_count--;
mutex_unlock(&allocated_ptys_lock);
}
@@ -558,11 +631,15 @@ void devpts_pty_kill(struct tty_struct *tty)
static int __init init_devpts_fs(void)
{
int err = register_filesystem(&devpts_fs_type);
+ struct ctl_table_header *table;
+
if (!err) {
+ table = register_sysctl_table(pty_root_table);
devpts_mnt = kern_mount(&devpts_fs_type);
if (IS_ERR(devpts_mnt)) {
err = PTR_ERR(devpts_mnt);
unregister_filesystem(&devpts_fs_type);
+ unregister_sysctl_table(table);
}
}
return err;
diff --git a/fs/exec.c b/fs/exec.c
index 153dee14fe55..95551c6da090 100644
--- a/fs/exec.c
+++ b/fs/exec.c
@@ -63,6 +63,8 @@
#include <trace/events/task.h>
#include "internal.h"
+#include <trace/events/sched.h>
+
int core_uses_pid;
char core_pattern[CORENAME_MAX_SIZE] = "core";
unsigned int core_pipe_limit;
@@ -848,6 +850,7 @@ static int exec_mmap(struct mm_struct *mm)
if (old_mm) {
up_read(&old_mm->mmap_sem);
BUG_ON(active_mm != old_mm);
+ setmax_mm_hiwater_rss(&tsk->signal->maxrss, old_mm);
mm_update_next_owner(old_mm);
mmput(old_mm);
return 0;
@@ -975,8 +978,8 @@ static int de_thread(struct task_struct *tsk)
sig->notify_count = 0;
no_thread_group:
- if (current->mm)
- setmax_mm_hiwater_rss(&sig->maxrss, current->mm);
+ /* we have changed execution domain */
+ tsk->exit_signal = SIGCHLD;
exit_itimers(sig);
flush_itimer_signals();
@@ -1402,9 +1405,10 @@ int search_binary_handler(struct linux_binprm *bprm,struct pt_regs *regs)
*/
bprm->recursion_depth = depth;
if (retval >= 0) {
- if (depth == 0)
- ptrace_event(PTRACE_EVENT_EXEC,
- old_pid);
+ if (depth == 0) {
+ trace_sched_process_exec(current, old_pid, bprm);
+ ptrace_event(PTRACE_EVENT_EXEC, old_pid);
+ }
put_binfmt(fmt);
allow_write_access(bprm->file);
if (bprm->file)
diff --git a/fs/namei.c b/fs/namei.c
index 46ea9cc16647..fa96a26d3291 100644
--- a/fs/namei.c
+++ b/fs/namei.c
@@ -1374,6 +1374,126 @@ static inline int can_lookup(struct inode *inode)
return 1;
}
+/*
+ * We can do the critical dentry name comparison and hashing
+ * operations one word at a time, but we are limited to:
+ *
+ * - Architectures with fast unaligned word accesses. We could
+ * do a "get_unaligned()" if this helps and is sufficiently
+ * fast.
+ *
+ * - Little-endian machines (so that we can generate the mask
+ * of low bytes efficiently). Again, we *could* do a byte
+ * swapping load on big-endian architectures if that is not
+ * expensive enough to make the optimization worthless.
+ *
+ * - non-CONFIG_DEBUG_PAGEALLOC configurations (so that we
+ * do not trap on the (extremely unlikely) case of a page
+ * crossing operation.
+ *
+ * - Furthermore, we need an efficient 64-bit compile for the
+ * 64-bit case in order to generate the "number of bytes in
+ * the final mask". Again, that could be replaced with a
+ * efficient population count instruction or similar.
+ */
+#ifdef CONFIG_DCACHE_WORD_ACCESS
+
+#ifdef CONFIG_64BIT
+
+/*
+ * Jan Achrenius on G+: microoptimized version of
+ * the simpler "(mask & ONEBYTES) * ONEBYTES >> 56"
+ * that works for the bytemasks without having to
+ * mask them first.
+ */
+static inline long count_masked_bytes(unsigned long mask)
+{
+ return mask*0x0001020304050608 >> 56;
+}
+
+static inline unsigned int fold_hash(unsigned long hash)
+{
+ hash += hash >> (8*sizeof(int));
+ return hash;
+}
+
+#else /* 32-bit case */
+
+/* Carl Chatfield / Jan Achrenius G+ version for 32-bit */
+static inline long count_masked_bytes(long mask)
+{
+ /* (000000 0000ff 00ffff ffffff) -> ( 1 1 2 3 ) */
+ long a = (0x0ff0001+mask) >> 23;
+ /* Fix the 1 for 00 case */
+ return a & mask;
+}
+
+#define fold_hash(x) (x)
+
+#endif
+
+unsigned int full_name_hash(const unsigned char *name, unsigned int len)
+{
+ unsigned long a, mask;
+ unsigned long hash = 0;
+
+ for (;;) {
+ a = *(unsigned long *)name;
+ hash *= 9;
+ if (len < sizeof(unsigned long))
+ break;
+ hash += a;
+ name += sizeof(unsigned long);
+ len -= sizeof(unsigned long);
+ if (!len)
+ goto done;
+ }
+ mask = ~(~0ul << len*8);
+ hash += mask & a;
+done:
+ return fold_hash(hash);
+}
+EXPORT_SYMBOL(full_name_hash);
+
+#define ONEBYTES 0x0101010101010101ul
+#define SLASHBYTES 0x2f2f2f2f2f2f2f2ful
+#define HIGHBITS 0x8080808080808080ul
+
+/* Return the high bit set in the first byte that is a zero */
+static inline unsigned long has_zero(unsigned long a)
+{
+ return ((a - ONEBYTES) & ~a) & HIGHBITS;
+}
+
+/*
+ * Calculate the length and hash of the path component, and
+ * return the length of the component;
+ */
+static inline unsigned long hash_name(const char *name, unsigned int *hashp)
+{
+ unsigned long a, mask, hash, len;
+
+ hash = a = 0;
+ len = -sizeof(unsigned long);
+ do {
+ hash = (hash + a) * 9;
+ len += sizeof(unsigned long);
+ a = *(unsigned long *)(name+len);
+ /* Do we have any NUL or '/' bytes in this word? */
+ mask = has_zero(a) | has_zero(a ^ SLASHBYTES);
+ } while (!mask);
+
+ /* The mask *below* the first high bit set */
+ mask = (mask - 1) & ~mask;
+ mask >>= 7;
+ hash += a & mask;
+ *hashp = fold_hash(hash);
+
+ return len + count_masked_bytes(mask);
+}
+
+#else
+
unsigned int full_name_hash(const unsigned char *name, unsigned int len)
{
unsigned long hash = init_name_hash();
@@ -1402,6 +1522,8 @@ static inline unsigned long hash_name(const char *name, unsigned int *hashp)
return len;
}
+#endif
+
/*
* Name resolution.
* This is the basic name resolution function, turning a pathname into
diff --git a/fs/proc/base.c b/fs/proc/base.c
index d4548dd49b02..965d4bde3a3b 100644
--- a/fs/proc/base.c
+++ b/fs/proc/base.c
@@ -1310,8 +1310,7 @@ sched_autogroup_write(struct file *file, const char __user *buf,
if (!p)
return -ESRCH;
- err = nice;
- err = proc_sched_autogroup_set_nice(p, &err);
+ err = proc_sched_autogroup_set_nice(p, nice);
if (err)
count = err;
diff --git a/fs/sysfs/dir.c b/fs/sysfs/dir.c
index 7fdf6a7b7436..2a7a3f5d1ca6 100644
--- a/fs/sysfs/dir.c
+++ b/fs/sysfs/dir.c
@@ -22,76 +22,103 @@
#include <linux/mutex.h>
#include <linux/slab.h>
#include <linux/security.h>
+#include <linux/hash.h>
#include "sysfs.h"
DEFINE_MUTEX(sysfs_mutex);
DEFINE_SPINLOCK(sysfs_assoc_lock);
+#define to_sysfs_dirent(X) rb_entry((X), struct sysfs_dirent, s_rb);
+
static DEFINE_SPINLOCK(sysfs_ino_lock);
static DEFINE_IDA(sysfs_ino_ida);
/**
- * sysfs_link_sibling - link sysfs_dirent into sibling list
+ * sysfs_name_hash
+ * @ns: Namespace tag to hash
+ * @name: Null terminated string to hash
+ *
+ * Returns 31 bit hash of ns + name (so it fits in an off_t )
+ */
+static unsigned int sysfs_name_hash(const void *ns, const char *name)
+{
+ unsigned long hash = init_name_hash();
+ unsigned int len = strlen(name);
+ while (len--)
+ hash = partial_name_hash(*name++, hash);
+ hash = ( end_name_hash(hash) ^ hash_ptr( (void *)ns, 31 ) );
+ hash &= 0x7fffffffU;
+ /* Reserve hash numbers 0, 1 and INT_MAX for magic directory entries */
+ if (hash < 1)
+ hash += 2;
+ if (hash >= INT_MAX)
+ hash = INT_MAX - 1;
+ return hash;
+}
+
+static int sysfs_name_compare(unsigned int hash, const void *ns,
+ const char *name, const struct sysfs_dirent *sd)
+{
+ if (hash != sd->s_hash)
+ return hash - sd->s_hash;
+ if (ns != sd->s_ns)
+ return ns - sd->s_ns;
+ return strcmp(name, sd->s_name);
+}
+
+static int sysfs_sd_compare(const struct sysfs_dirent *left,
+ const struct sysfs_dirent *right)
+{
+ return sysfs_name_compare(left->s_hash, left->s_ns, left->s_name,
+ right);
+}
+
+/**
+ * sysfs_link_subling - link sysfs_dirent into sibling rbtree
* @sd: sysfs_dirent of interest
*
- * Link @sd into its sibling list which starts from
+ * Link @sd into its sibling rbtree which starts from
* sd->s_parent->s_dir.children.
*
* Locking:
* mutex_lock(sysfs_mutex)
+ *
+ * RETURNS:
+ * 0 on susccess -EEXIST on failure.
*/
-static void sysfs_link_sibling(struct sysfs_dirent *sd)
+static int sysfs_link_sibling(struct sysfs_dirent *sd)
{
- struct sysfs_dirent *parent_sd = sd->s_parent;
-
- struct rb_node **p;
- struct rb_node *parent;
+ struct rb_node **node = &sd->s_parent->s_dir.children.rb_node;
+ struct rb_node *parent = NULL;
if (sysfs_type(sd) == SYSFS_DIR)
- parent_sd->s_dir.subdirs++;
-
- p = &parent_sd->s_dir.inode_tree.rb_node;
- parent = NULL;
- while (*p) {
- parent = *p;
-#define node rb_entry(parent, struct sysfs_dirent, inode_node)
- if (sd->s_ino < node->s_ino) {
- p = &node->inode_node.rb_left;
- } else if (sd->s_ino > node->s_ino) {
- p = &node->inode_node.rb_right;
- } else {
- printk(KERN_CRIT "sysfs: inserting duplicate inode '%lx'\n",
- (unsigned long) sd->s_ino);
- BUG();
- }
-#undef node
- }
- rb_link_node(&sd->inode_node, parent, p);
- rb_insert_color(&sd->inode_node, &parent_sd->s_dir.inode_tree);
-
- p = &parent_sd->s_dir.name_tree.rb_node;
- parent = NULL;
- while (*p) {
- int c;
- parent = *p;
-#define node rb_entry(parent, struct sysfs_dirent, name_node)
- c = strcmp(sd->s_name, node->s_name);
- if (c < 0) {
- p = &node->name_node.rb_left;
- } else {
- p = &node->name_node.rb_right;
- }
-#undef node
+ sd->s_parent->s_dir.subdirs++;
+
+ while (*node) {
+ struct sysfs_dirent *pos;
+ int result;
+
+ pos = to_sysfs_dirent(*node);
+ parent = *node;
+ result = sysfs_sd_compare(sd, pos);
+ if (result < 0)
+ node = &pos->s_rb.rb_left;
+ else if (result > 0)
+ node = &pos->s_rb.rb_right;
+ else
+ return -EEXIST;
}
- rb_link_node(&sd->name_node, parent, p);
- rb_insert_color(&sd->name_node, &parent_sd->s_dir.name_tree);
+ /* add new node and rebalance the tree */
+ rb_link_node(&sd->s_rb, parent, node);
+ rb_insert_color(&sd->s_rb, &sd->s_parent->s_dir.children);
+ return 0;
}
/**
- * sysfs_unlink_sibling - unlink sysfs_dirent from sibling list
+ * sysfs_unlink_sibling - unlink sysfs_dirent from sibling rbtree
* @sd: sysfs_dirent of interest
*
- * Unlink @sd from its sibling list which starts from
+ * Unlink @sd from its sibling rbtree which starts from
* sd->s_parent->s_dir.children.
*
* Locking:
@@ -102,8 +129,7 @@ static void sysfs_unlink_sibling(struct sysfs_dirent *sd)
if (sysfs_type(sd) == SYSFS_DIR)
sd->s_parent->s_dir.subdirs--;
- rb_erase(&sd->inode_node, &sd->s_parent->s_dir.inode_tree);
- rb_erase(&sd->name_node, &sd->s_parent->s_dir.name_tree);
+ rb_erase(&sd->s_rb, &sd->s_parent->s_dir.children);
}
/**
@@ -198,7 +224,7 @@ static void sysfs_deactivate(struct sysfs_dirent *sd)
rwsem_release(&sd->dep_map, 1, _RET_IP_);
}
-static int sysfs_alloc_ino(ino_t *pino)
+static int sysfs_alloc_ino(unsigned int *pino)
{
int ino, rc;
@@ -217,7 +243,7 @@ static int sysfs_alloc_ino(ino_t *pino)
return rc;
}
-static void sysfs_free_ino(ino_t ino)
+static void sysfs_free_ino(unsigned int ino)
{
spin_lock(&sysfs_ino_lock);
ida_remove(&sysfs_ino_ida, ino);
@@ -402,6 +428,7 @@ void sysfs_addrm_start(struct sysfs_addrm_cxt *acxt,
int __sysfs_add_one(struct sysfs_addrm_cxt *acxt, struct sysfs_dirent *sd)
{
struct sysfs_inode_attrs *ps_iattr;
+ int ret;
if (!!sysfs_ns_type(acxt->parent_sd) != !!sd->s_ns) {
WARN(1, KERN_WARNING "sysfs: ns %s in '%s' for '%s'\n",
@@ -410,12 +437,12 @@ int __sysfs_add_one(struct sysfs_addrm_cxt *acxt, struct sysfs_dirent *sd)
return -EINVAL;
}
- if (sysfs_find_dirent(acxt->parent_sd, sd->s_ns, sd->s_name))
- return -EEXIST;
-
+ sd->s_hash = sysfs_name_hash(sd->s_ns, sd->s_name);
sd->s_parent = sysfs_get(acxt->parent_sd);
- sysfs_link_sibling(sd);
+ ret = sysfs_link_sibling(sd);
+ if (ret)
+ return ret;
/* Update timestamps on the parent */
ps_iattr = acxt->parent_sd->s_iattr;
@@ -565,8 +592,8 @@ struct sysfs_dirent *sysfs_find_dirent(struct sysfs_dirent *parent_sd,
const void *ns,
const unsigned char *name)
{
- struct rb_node *p = parent_sd->s_dir.name_tree.rb_node;
- struct sysfs_dirent *found = NULL;
+ struct rb_node *node = parent_sd->s_dir.children.rb_node;
+ unsigned int hash;
if (!!sysfs_ns_type(parent_sd) != !!ns) {
WARN(1, KERN_WARNING "sysfs: ns %s in '%s' for '%s'\n",
@@ -575,33 +602,21 @@ struct sysfs_dirent *sysfs_find_dirent(struct sysfs_dirent *parent_sd,
return NULL;
}
- while (p) {
- int c;
-#define node rb_entry(p, struct sysfs_dirent, name_node)
- c = strcmp(name, node->s_name);
- if (c < 0) {
- p = node->name_node.rb_left;
- } else if (c > 0) {
- p = node->name_node.rb_right;
- } else {
- found = node;
- p = node->name_node.rb_left;
- }
-#undef node
- }
-
- if (found) {
- while (found->s_ns != ns) {
- p = rb_next(&found->name_node);
- if (!p)
- return NULL;
- found = rb_entry(p, struct sysfs_dirent, name_node);
- if (strcmp(name, found->s_name))
- return NULL;
- }
+ hash = sysfs_name_hash(ns, name);
+ while (node) {
+ struct sysfs_dirent *sd;
+ int result;
+
+ sd = to_sysfs_dirent(node);
+ result = sysfs_name_compare(hash, ns, name, sd);
+ if (result < 0)
+ node = node->rb_left;
+ else if (result > 0)
+ node = node->rb_right;
+ else
+ return sd;
}
-
- return found;
+ return NULL;
}
/**
@@ -804,9 +819,9 @@ static void __sysfs_remove_dir(struct sysfs_dirent *dir_sd)
pr_debug("sysfs %s: removing dir\n", dir_sd->s_name);
sysfs_addrm_start(&acxt, dir_sd);
- pos = rb_first(&dir_sd->s_dir.inode_tree);
+ pos = rb_first(&dir_sd->s_dir.children);
while (pos) {
- struct sysfs_dirent *sd = rb_entry(pos, struct sysfs_dirent, inode_node);
+ struct sysfs_dirent *sd = to_sysfs_dirent(pos);
pos = rb_next(pos);
if (sysfs_type(sd) != SYSFS_DIR)
sysfs_remove_one(&acxt, sd);
@@ -863,6 +878,7 @@ int sysfs_rename(struct sysfs_dirent *sd,
dup_name = sd->s_name;
sd->s_name = new_name;
+ sd->s_hash = sysfs_name_hash(sd->s_ns, sd->s_name);
}
/* Move to the appropriate place in the appropriate directories rbtree. */
@@ -919,38 +935,36 @@ static int sysfs_dir_release(struct inode *inode, struct file *filp)
}
static struct sysfs_dirent *sysfs_dir_pos(const void *ns,
- struct sysfs_dirent *parent_sd, ino_t ino, struct sysfs_dirent *pos)
+ struct sysfs_dirent *parent_sd, loff_t hash, struct sysfs_dirent *pos)
{
if (pos) {
int valid = !(pos->s_flags & SYSFS_FLAG_REMOVED) &&
pos->s_parent == parent_sd &&
- ino == pos->s_ino;
+ hash == pos->s_hash;
sysfs_put(pos);
if (!valid)
pos = NULL;
}
- if (!pos && (ino > 1) && (ino < INT_MAX)) {
- struct rb_node *p = parent_sd->s_dir.inode_tree.rb_node;
- while (p) {
-#define node rb_entry(p, struct sysfs_dirent, inode_node)
- if (ino < node->s_ino) {
- pos = node;
- p = node->inode_node.rb_left;
- } else if (ino > node->s_ino) {
- p = node->inode_node.rb_right;
- } else {
- pos = node;
+ if (!pos && (hash > 1) && (hash < INT_MAX)) {
+ struct rb_node *node = parent_sd->s_dir.children.rb_node;
+ while (node) {
+ pos = to_sysfs_dirent(node);
+
+ if (hash < pos->s_hash)
+ node = node->rb_left;
+ else if (hash > pos->s_hash)
+ node = node->rb_right;
+ else
break;
- }
-#undef node
}
}
+ /* Skip over entries in the wrong namespace */
while (pos && pos->s_ns != ns) {
- struct rb_node *p = rb_next(&pos->inode_node);
- if (!p)
+ struct rb_node *node = rb_next(&pos->s_rb);
+ if (!node)
pos = NULL;
else
- pos = rb_entry(p, struct sysfs_dirent, inode_node);
+ pos = to_sysfs_dirent(node);
}
return pos;
}
@@ -960,11 +974,11 @@ static struct sysfs_dirent *sysfs_dir_next_pos(const void *ns,
{
pos = sysfs_dir_pos(ns, parent_sd, ino, pos);
if (pos) do {
- struct rb_node *p = rb_next(&pos->inode_node);
- if (!p)
+ struct rb_node *node = rb_next(&pos->s_rb);
+ if (!node)
pos = NULL;
else
- pos = rb_entry(p, struct sysfs_dirent, inode_node);
+ pos = to_sysfs_dirent(node);
} while (pos && pos->s_ns != ns);
return pos;
}
@@ -1006,7 +1020,7 @@ static int sysfs_readdir(struct file * filp, void * dirent, filldir_t filldir)
len = strlen(name);
ino = pos->s_ino;
type = dt_type(pos);
- filp->f_pos = ino;
+ filp->f_pos = pos->s_hash;
filp->private_data = sysfs_get(pos);
mutex_unlock(&sysfs_mutex);
diff --git a/fs/sysfs/inode.c b/fs/sysfs/inode.c
index 85eb81683a29..feb2d69396cf 100644
--- a/fs/sysfs/inode.c
+++ b/fs/sysfs/inode.c
@@ -136,12 +136,13 @@ static int sysfs_sd_setsecdata(struct sysfs_dirent *sd, void **secdata, u32 *sec
void *old_secdata;
size_t old_secdata_len;
- iattrs = sd->s_iattr;
- if (!iattrs)
- iattrs = sysfs_init_inode_attrs(sd);
- if (!iattrs)
- return -ENOMEM;
+ if (!sd->s_iattr) {
+ sd->s_iattr = sysfs_init_inode_attrs(sd);
+ if (!sd->s_iattr)
+ return -ENOMEM;
+ }
+ iattrs = sd->s_iattr;
old_secdata = iattrs->ia_secdata;
old_secdata_len = iattrs->ia_secdata_len;
diff --git a/fs/sysfs/mount.c b/fs/sysfs/mount.c
index e34f0d99ea4e..140f26a34288 100644
--- a/fs/sysfs/mount.c
+++ b/fs/sysfs/mount.c
@@ -36,7 +36,7 @@ struct sysfs_dirent sysfs_root = {
.s_name = "",
.s_count = ATOMIC_INIT(1),
.s_flags = SYSFS_DIR | (KOBJ_NS_TYPE_NONE << SYSFS_NS_TYPE_SHIFT),
- .s_mode = S_IFDIR | S_IRWXU | S_IRUGO | S_IXUGO,
+ .s_mode = S_IFDIR | S_IRUGO | S_IXUGO,
.s_ino = 1,
};
diff --git a/fs/sysfs/sysfs.h b/fs/sysfs/sysfs.h
index 7484a36ee678..661a9639570b 100644
--- a/fs/sysfs/sysfs.h
+++ b/fs/sysfs/sysfs.h
@@ -20,9 +20,8 @@ struct sysfs_elem_dir {
struct kobject *kobj;
unsigned long subdirs;
-
- struct rb_root inode_tree;
- struct rb_root name_tree;
+ /* children rbtree starts here and goes through sd->s_rb */
+ struct rb_root children;
};
struct sysfs_elem_symlink {
@@ -62,8 +61,7 @@ struct sysfs_dirent {
struct sysfs_dirent *s_parent;
const char *s_name;
- struct rb_node inode_node;
- struct rb_node name_node;
+ struct rb_node s_rb;
union {
struct completion *completion;
@@ -71,6 +69,7 @@ struct sysfs_dirent {
} u;
const void *s_ns; /* namespace tag */
+ unsigned int s_hash; /* ns + name hash */
union {
struct sysfs_elem_dir s_dir;
struct sysfs_elem_symlink s_symlink;
@@ -78,9 +77,9 @@ struct sysfs_dirent {
struct sysfs_elem_bin_attr s_bin_attr;
};
- unsigned int s_flags;
+ unsigned short s_flags;
umode_t s_mode;
- ino_t s_ino;
+ unsigned int s_ino;
struct sysfs_inode_attrs *s_iattr;
};
@@ -95,11 +94,11 @@ struct sysfs_dirent {
#define SYSFS_ACTIVE_REF (SYSFS_KOBJ_ATTR | SYSFS_KOBJ_BIN_ATTR)
/* identify any namespace tag on sysfs_dirents */
-#define SYSFS_NS_TYPE_MASK 0xff00
+#define SYSFS_NS_TYPE_MASK 0xf00
#define SYSFS_NS_TYPE_SHIFT 8
#define SYSFS_FLAG_MASK ~(SYSFS_NS_TYPE_MASK|SYSFS_TYPE_MASK)
-#define SYSFS_FLAG_REMOVED 0x020000
+#define SYSFS_FLAG_REMOVED 0x02000
static inline unsigned int sysfs_type(struct sysfs_dirent *sd)
{
generated by cgit v1.2.3 (git 2.25.1) at 2024-11-25 03:43:07 +0000