login
Search
Hi Linus,
Here's a set of patches that remove all calls to iget() and all read_inode()
functions. They should be removed for two reasons: firstly they don't lend
themselves to good error handling, and secondly their presence is a temptation
for code outside a filesystem to call iget() to access inodes within that
filesystem.There are a few benefits to this:
(1) Error handling gets simpler as you can return an error code rather than
having to call is_bad_inode().(2) You can now tell the difference between ENOMEM and EIO occurring in the
read_inode() path.(3) The code should get smaller. iget() is an inline function that is
typically called 2-3 times per filesystem that uses it. By folding the
iget code into the read_inode code for each filesystem, it eliminates
some duplication.A tarball of the patches can be retrieved from:
http://people.redhat.com/~dhowells/iget-remove.tar.bz2
Additionally, there are a couple of patches that introduce an ERR_CAST() macro
to be used instead of ERR_PTR(PTR_ERR(p)) constructs and apply it to all such
instances in the kernel.Of the patches directly relevant to the subject:
The first patch adds a function, iget_failed() that is a canned piece of code
for killing an inode when the inode construction path fails.The second and third patches makes AFS and GFS2 use iget_failed() rather than
interpolating the sequence directly.The fourth patch marks iget() and read_inode() as being deprecated.
The final patch removes iget() and read_inode() completely.
Each of the other patches modify a filesystem that used iget() and read_inode()
to use iget_locked() instead. The standard procedure was to convert:void thingyfs_read_inode(struct inode *inode)
{
...
}into:
struct inode *thingyfs_iget(struct super_block *sp, unsigned long ino)
{
struct inode *inode;
int ret;inode = iget_locked(sb, ino);
if (!inode)
return ERR_PTR(-ENOMEM);
if (!(inode->i_state...
Is there any particular reason not to push these through -mm?
-
No.
David
-
Remove the old iget() call and the read_inode() superblock operation it uses
as these are really obsolete, and the use of read_inode() does not produce
proper error handling (no distinction between ENOMEM and EIO when marking
an inode bad).Furthermore, this removes the temptation to use iget() to find an inode by
number in a filesystem from code outside that filesystem.iget_locked() should be used instead. A new function is added in an earlier
patch (iget_failed) that is to be called to mark an inode as bad, unlock it and
release it should the get routine fail.
Mark iget() and read_inode() as being obsolete and remove references to them
from the documentation.Typically a filesystem will be modified such that the read_inode function
becomes an internal iget function, for example the following:void thingyfs_read_inode(struct inode *inode)
{
...
}would be changed into something like:
struct inode *thingyfs_iget(struct super_block *sp, unsigned long ino)
{
struct inode *inode;
int ret;inode = iget_locked(sb, ino);
if (!inode)
return ERR_PTR(-ENOMEM);
if (!(inode->i_state & I_NEW))
return inode;...
unlock_new_inode(inode);
return inode;
error:
iget_failed(inode);
return ERR_PTR(ret);
}and then thingyfs_iget() would be called rather than iget(), for example:
ret = -EINVAL;
inode = iget(sb, ino);
if (!inode || is_bad_inode(inode))
goto error;becomes:
inode = thingyfs_iget(sb, ino);
if (IS_ERR(inode)) {
ret = PTR_ERR(inode);
goto error;
}Note that is_bad_inode() does not need to be called. The error returned by
thingyfs_iget() should render it unnecessary.Signed-off-by: David Howells <dhowells@redhat.com>
---Documentation/filesystems/Exporting | 5 -----
Documentation/filesystems/Locking | 3 ---
Documentation/filesystems/porting | 12 ++++++------
Documentation/filesystems/vfs.txt | 17 +++--------------
fs/inode.c | ...
Stop the HPPFS filesystem from using iget() and read_inode(). Provide an
hppfs_iget(), and call that instead of iget(). hppfs_iget() then uses
iget_locked() directly and returns a proper error code instead of an inode in
the event of an error.hppfs_fill_sb_common() returns any error incurred when getting the root inode
instead of EINVAL.Note that the contents of hppfs_kern.c need to be examined:
(*) The HPPFS inode retains a pointer to the proc dentry it is shadowing, but
whilst it does appear to retain a reference to it, it doesn't appear to
destroy the reference if the inode goes away.(*) hppfs_iget() should perhaps subsume init_inode() and hppfs_read_inode().
(*) It would appear that all hppfs inodes are the same inode because iget()
was being called with inode number 0, which forms the lookup key.Signed-off-by: David Howells <dhowells@redhat.com>
---fs/hppfs/hppfs_kern.c | 27 ++++++++++++++++++++++-----
1 files changed, 22 insertions(+), 5 deletions(-)diff --git a/fs/hppfs/hppfs_kern.c b/fs/hppfs/hppfs_kern.c
index affb741..a1e1f0f 100644
--- a/fs/hppfs/hppfs_kern.c
+++ b/fs/hppfs/hppfs_kern.c
@@ -155,6 +155,20 @@ static void hppfs_read_inode(struct inode *ino)
ino->i_blocks = proc_ino->i_blocks;
}+static struct inode *hppfs_iget(struct super_block *sb)
+{
+ struct inode *inode;
+
+ inode = iget_locked(sb, 0);
+ if (!inode)
+ return ERR_PTR(-ENOMEM);
+ if (inode->i_state & I_NEW) {
+ hppfs_read_inode(inode);
+ unlock_new_inode(inode);
+ }
+ return inode;
+}
+
static struct dentry *hppfs_lookup(struct inode *ino, struct dentry *dentry,
struct nameidata *nd)
{
@@ -190,9 +204,11 @@ static struct dentry *hppfs_lookup(struct inode *ino, struct dentry *dentry,
if(IS_ERR(proc_dentry))
return(proc_dentry);- inode = iget(ino->i_sb, 0);
- if(inode == NULL)
+ inode = hppfs_iget(ino->i_sb);
+ if (IS_ERR(inode)) {
+ err = PTR_ERR(inode);
goto ou...
Stop the OPENPROMFS filesystem from using iget() and read_inode(). Replace
openpromfs_read_inode() with openpromfs_iget(), and call that instead of
iget(). openpromfs_iget() then uses iget_locked() directly and returns a
proper error code instead of an inode in the event of an error.openpromfs_fill_super() returns any error incurred when getting the root inode
instead of ENOMEM (not that it currently incurs any other error).Signed-off-by: David Howells <dhowells@redhat.com>
---fs/openpromfs/inode.c | 45 ++++++++++++++++++++++++++++++---------------
1 files changed, 30 insertions(+), 15 deletions(-)diff --git a/fs/openpromfs/inode.c b/fs/openpromfs/inode.c
index dd86be2..9c6017b 100644
--- a/fs/openpromfs/inode.c
+++ b/fs/openpromfs/inode.c
@@ -38,6 +38,8 @@ struct op_inode_info {
union op_inode_data u;
};+static struct inode *openprom_iget(struct super_block *sb, ino_t ino);
+
static inline struct op_inode_info *OP_I(struct inode *inode)
{
return container_of(inode, struct op_inode_info, vfs_inode);
@@ -226,10 +228,10 @@ static struct dentry *openpromfs_lookup(struct inode *dir, struct dentry *dentry
return ERR_PTR(-ENOENT);found:
- inode = iget(dir->i_sb, ino);
+ inode = openprom_iget(dir->i_sb, ino);
mutex_unlock(&op_mutex);
- if (!inode)
- return ERR_PTR(-EINVAL);
+ if (IS_ERR(inode))
+ return ERR_CAST(inode);
ent_oi = OP_I(inode);
ent_oi->type = ent_type;
ent_oi->u = ent_data;
@@ -348,14 +350,23 @@ static void openprom_destroy_inode(struct inode *inode)
kmem_cache_free(op_inode_cachep, OP_I(inode));
}-static void openprom_read_inode(struct inode * inode)
+static struct inode *openprom_iget(struct super_block *sb, ino_t ino)
{
- inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME;
- if (inode->i_ino == OPENPROM_ROOT_INO) {
- inode->i_op = &openprom_inode_operations;
- inode->i_fop = &openprom_operations;
- inode->i_mode = S_IFDIR | S_IRUGO | S_IXUG...
Stop the HOSTFS filesystem from using iget() and read_inode(). Provide
hostfs_iget(), and call that instead of iget(). hostfs_iget() then uses
iget_locked() directly and returns a proper error code instead of an inode in
the event of an error.hostfs_fill_sb_common() returns any error incurred when getting the root inode
instead of EINVAL.Note that the contents of hostfs_kern.c need to be examined:
(*) hostfs_iget() should perhaps subsume init_inode() and hostfs_read_inode().
(*) It would appear that all hostfs inodes are the same inode because iget()
was being called with inode number 0 - which forms the lookup key.Signed-off-by: David Howells <dhowells@redhat.com>
---fs/hostfs/hostfs_kern.c | 58 ++++++++++++++++++++++++++++++++---------------
1 files changed, 39 insertions(+), 19 deletions(-)diff --git a/fs/hostfs/hostfs_kern.c b/fs/hostfs/hostfs_kern.c
index c778620..c6a456a 100644
--- a/fs/hostfs/hostfs_kern.c
+++ b/fs/hostfs/hostfs_kern.c
@@ -208,7 +208,7 @@ static char *follow_link(char *link)
return ERR_PTR(n);
}-static int read_inode(struct inode *ino)
+static int hostfs_read_inode(struct inode *ino)
{
char *name;
int err = 0;
@@ -238,6 +238,25 @@ static int read_inode(struct inode *ino)
return err;
}+static struct inode *hostfs_iget(struct super_block *sb)
+{
+ struct inode *inode;
+ long ret;
+
+ inode = iget_locked(sb, 0);
+ if (!inode)
+ return ERR_PTR(-ENOMEM);
+ if (inode->i_state & I_NEW) {
+ ret = hostfs_read_inode(inode);
+ if (ret < 0) {
+ iget_failed(inode);
+ return ERR_PTR(ret);
+ }
+ unlock_new_inode(inode);
+ }
+ return inode;
+}
+
int hostfs_statfs(struct dentry *dentry, struct kstatfs *sf)
{
/* do_statfs uses struct statfs64 internally, but the linux kernel
@@ -305,17 +324,11 @@ static void hostfs_destroy_inode(struct inode *inode)
kfree(HOSTFS_I(inode));
}-static void hostfs_read_inode(struct inode *inode)
-{
- read_inode(inode);
-}
-
static co...
Stop the UFS filesystem from using iget() and read_inode(). Replace
ufs_read_inode() with ufs_iget(), and call that instead of iget().
ufs_iget() then uses iget_locked() directly and returns a proper error code
instead of an inode in the event of an error.ufs_fill_super() returns any error incurred when getting the root inode
instead of EINVAL.Signed-off-by: David Howells <dhowells@redhat.com>
---fs/ufs/inode.c | 34 ++++++++++++++++++++--------------
fs/ufs/namei.c | 6 +++---
fs/ufs/super.c | 14 +++++++++-----
include/linux/ufs_fs.h | 2 +-
4 files changed, 33 insertions(+), 23 deletions(-)diff --git a/fs/ufs/inode.c b/fs/ufs/inode.c
index f18b791..d7c2d3c 100644
--- a/fs/ufs/inode.c
+++ b/fs/ufs/inode.c
@@ -698,26 +698,30 @@ static int ufs2_read_inode(struct inode *inode, struct ufs2_inode *ufs2_inode)
return 0;
}-void ufs_read_inode(struct inode * inode)
+struct inode *ufs_iget (struct super_block *sb, unsigned long ino)
{
- struct ufs_inode_info *ufsi = UFS_I(inode);
- struct super_block * sb;
- struct ufs_sb_private_info * uspi;
+ struct ufs_inode_info *ufsi;
+ struct ufs_sb_private_info * uspi = UFS_SB(sb)->s_uspi;
struct buffer_head * bh;
+ struct inode *inode;
int err;- UFSD("ENTER, ino %lu\n", inode->i_ino);
-
- sb = inode->i_sb;
- uspi = UFS_SB(sb)->s_uspi;
+ UFSD("ENTER, ino %lu\n", ino);- if (inode->i_ino < UFS_ROOTINO ||
- inode->i_ino > (uspi->s_ncg * uspi->s_ipg)) {
+ if (ino < UFS_ROOTINO || ino > (uspi->s_ncg * uspi->s_ipg)) {
ufs_warning(sb, "ufs_read_inode", "bad inode number (%lu)\n",
- inode->i_ino);
- goto bad_inode;
+ ino);
+ return ERR_PTR(-EIO);
}+ inode = iget_locked(sb, ino);
+ if (!inode)
+ return ERR_PTR(-ENOMEM);
+ if (!(inode->i_state & I_NEW))
+ return inode;
+
+ ufsi = UFS_I(inode);
+
bh = sb_bread(sb, uspi->s_sbbase + ufs_inotofsba(inode->i_ino));
if (!bh) {
...
Stop the JFS filesystem from using iget() and read_inode(). Replace
jfs_read_inode() with jfs_iget(), and call that instead of iget().
jfs_iget() then uses iget_locked() directly and returns a proper error code
instead of an inode in the event of an error.jfs_fill_super() returns any error incurred when getting the root inode
instead of EINVAL.Signed-off-by: David Howells <dhowells@redhat.com>
Acked-by: Dave Kleikamp <shaggy@linux.vnet.ibm.com>
---fs/jfs/inode.c | 20 ++++++++++++++++----
fs/jfs/jfs_inode.h | 2 +-
fs/jfs/namei.c | 34 ++++++++++++++--------------------
fs/jfs/super.c | 15 +++++++++------
4 files changed, 40 insertions(+), 31 deletions(-)diff --git a/fs/jfs/inode.c b/fs/jfs/inode.c
index 3467dde..5081e76 100644
--- a/fs/jfs/inode.c
+++ b/fs/jfs/inode.c
@@ -31,11 +31,21 @@
#include "jfs_debug.h"-void jfs_read_inode(struct inode *inode)
+struct inode *jfs_iget(struct super_block *sb, unsigned long ino)
{
- if (diRead(inode)) {
- make_bad_inode(inode);
- return;
+ struct inode *inode;
+ int ret;
+
+ inode = iget_locked(sb, ino);
+ if (!inode)
+ return ERR_PTR(-ENOMEM);
+ if (!(inode->i_state & I_NEW))
+ return inode;
+
+ ret = diRead(inode);
+ if (ret < 0) {
+ iget_failed(inode);
+ return ERR_PTR(ret);
}if (S_ISREG(inode->i_mode)) {
@@ -55,6 +65,8 @@ void jfs_read_inode(struct inode *inode)
inode->i_op = &jfs_file_inode_operations;
init_special_inode(inode, inode->i_mode, inode->i_rdev);
}
+ unlock_new_inode(inode);
+ return inode;
}/*
diff --git a/fs/jfs/jfs_inode.h b/fs/jfs/jfs_inode.h
index f0ec72b..71ea106 100644
--- a/fs/jfs/jfs_inode.h
+++ b/fs/jfs/jfs_inode.h
@@ -22,7 +22,7 @@ extern struct inode *ialloc(struct inode *, umode_t);
extern int jfs_fsync(struct file *, struct dentry *, int);
extern int jfs_ioctl(struct inode *, struct file *,
unsigned int, unsigned long);
-extern void jfs_read_inode(struct inode *);
+extern ...
Stop the SYSV filesystem from using iget() and read_inode(). Replace
sysv_read_inode() with sysv_iget(), and call that instead of iget().
sysv_iget() then uses iget_locked() directly and returns a proper error code
instead of an inode in the event of an error.Signed-off-by: David Howells <dhowells@redhat.com>
---fs/sysv/inode.c | 25 ++++++++++++++++---------
fs/sysv/namei.c | 6 +++---
fs/sysv/super.c | 4 ++--
fs/sysv/sysv.h | 1 +
4 files changed, 22 insertions(+), 14 deletions(-)diff --git a/fs/sysv/inode.c b/fs/sysv/inode.c
index 7c4e5d3..9d512a8 100644
--- a/fs/sysv/inode.c
+++ b/fs/sysv/inode.c
@@ -169,20 +169,27 @@ void sysv_set_inode(struct inode *inode, dev_t rdev)
init_special_inode(inode, inode->i_mode, rdev);
}-static void sysv_read_inode(struct inode *inode)
+struct inode *sysv_iget(struct super_block *sb, unsigned int ino)
{
- struct super_block * sb = inode->i_sb;
struct sysv_sb_info * sbi = SYSV_SB(sb);
struct buffer_head * bh;
struct sysv_inode * raw_inode;
struct sysv_inode_info * si;
- unsigned int block, ino = inode->i_ino;
+ struct inode *inode;
+ unsigned int block;if (!ino || ino > sbi->s_ninodes) {
printk("Bad inode number on dev %s: %d is out of range\n",
- inode->i_sb->s_id, ino);
- goto bad_inode;
+ sb->s_id, ino);
+ return ERR_PTR(-EIO);
}
+
+ inode = iget_locked(sb, ino);
+ if (!inode)
+ return ERR_PTR(-ENOMEM);
+ if (!(inode->i_state & I_NEW))
+ return inode;
+
raw_inode = sysv_raw_inode(sb, ino, &bh);
if (!raw_inode) {
printk("Major problem: unable to read inode from dev %s\n",
@@ -214,11 +221,12 @@ static void sysv_read_inode(struct inode *inode)
old_decode_dev(fs32_to_cpu(sbi, si->i_data[0])));
else
sysv_set_inode(inode, 0);
- return;
+ unlock_new_inode(inode);
+ return inode;bad_inode:
- make_bad_inode(inode);
- return;
+ iget_failed(inode);
+ return ERR_PTR(-EIO);
}static s...
Stop the QNX4 filesystem from using iget() and read_inode(). Replace
qnx4_read_inode() with qnx4_iget(), and call that instead of iget().
qnx4_iget() then uses iget_locked() directly and returns a proper error code
instead of an inode in the event of an error.qnx4_fill_super() returns any error incurred when getting the root inode
instead of EINVAL.Signed-off-by: David Howells <dhowells@redhat.com>
---fs/qnx4/inode.c | 45 ++++++++++++++++++++++++++++++---------------
fs/qnx4/namei.c | 8 +++++---
include/linux/qnx4_fs.h | 1 +
3 files changed, 36 insertions(+), 18 deletions(-)diff --git a/fs/qnx4/inode.c b/fs/qnx4/inode.c
index 1bc8d87..3eba8fe 100644
--- a/fs/qnx4/inode.c
+++ b/fs/qnx4/inode.c
@@ -125,7 +125,6 @@ static int qnx4_write_inode(struct inode *inode, int unused)
static void qnx4_put_super(struct super_block *sb);
static struct inode *qnx4_alloc_inode(struct super_block *sb);
static void qnx4_destroy_inode(struct inode *inode);
-static void qnx4_read_inode(struct inode *);
static int qnx4_remount(struct super_block *sb, int *flags, char *data);
static int qnx4_statfs(struct dentry *, struct kstatfs *);@@ -133,7 +132,6 @@ static const struct super_operations qnx4_sops =
{
.alloc_inode = qnx4_alloc_inode,
.destroy_inode = qnx4_destroy_inode,
- .read_inode = qnx4_read_inode,
.put_super = qnx4_put_super,
.statfs = qnx4_statfs,
.remount_fs = qnx4_remount,
@@ -357,6 +355,7 @@ static int qnx4_fill_super(struct super_block *s, void *data, int silent)
struct inode *root;
const char *errmsg;
struct qnx4_sb_info *qs;
+ int ret = -EINVAL;qs = kzalloc(sizeof(struct qnx4_sb_info), GFP_KERNEL);
if (!qs)
@@ -396,12 +395,14 @@ static int qnx4_fill_super(struct super_block *s, void *data, int silent)
}/* does root not have inode number QNX4_ROOT_INO ?? */
- root = iget(s, QNX4_ROOT_INO * QNX4_INODES_PER_BLOCK);
- if (!root) {
+ root = qnx4_iget(s, QNX4_ROOT_INO * QNX4_INODES_PER_BLOCK);
...
Stop the PROCFS filesystem from using iget() and read_inode(). Merge
procfs_read_inode() into procfs_get_inode(), and have that call iget_locked()
instead of iget().Signed-off-by: David Howells <dhowells@redhat.com>
---fs/proc/inode.c | 60 ++++++++++++++++++++++++++-----------------------------
1 files changed, 28 insertions(+), 32 deletions(-)diff --git a/fs/proc/inode.c b/fs/proc/inode.c
index 0e4d37c..7a563c5 100644
--- a/fs/proc/inode.c
+++ b/fs/proc/inode.c
@@ -78,11 +78,6 @@ static void proc_delete_inode(struct inode *inode)struct vfsmount *proc_mnt;
-static void proc_read_inode(struct inode * inode)
-{
- inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME;
-}
-
static struct kmem_cache * proc_inode_cachep;static struct inode *proc_alloc_inode(struct super_block *sb)
@@ -135,7 +130,6 @@ static int proc_remount(struct super_block *sb, int *flags, char *data)
static const struct super_operations proc_sops = {
.alloc_inode = proc_alloc_inode,
.destroy_inode = proc_destroy_inode,
- .read_inode = proc_read_inode,
.drop_inode = generic_delete_inode,
.delete_inode = proc_delete_inode,
.statfs = simple_statfs,
@@ -408,39 +402,41 @@ struct inode *proc_get_inode(struct super_block *sb, unsigned int ino,
if (de != NULL && !try_module_get(de->owner))
goto out_mod;- inode = iget(sb, ino);
+ inode = iget_locked(sb, ino);
if (!inode)
goto out_ino;
-
- PROC_I(inode)->fd = 0;
- PROC_I(inode)->pde = de;
- if (de) {
- if (de->mode) {
- inode->i_mode = de->mode;
- inode->i_uid = de->uid;
- inode->i_gid = de->gid;
- }
- if (de->size)
- inode->i_size = de->size;
- if (de->nlink)
- inode->i_nlink = de->nlink;
- if (de->proc_iops)
- inode->i_op = de->proc_iops;
- if (de->proc_fops) {
- if (S_ISREG(inode->i_mode)) {
+ if (inode->i_state & I_NEW) {
+ inode->i_mtime = inode->i_atime = inode->i_ct...
Stop the ROMFS filesystem from using iget() and read_inode(). Replace
romfs_read_inode() with romfs_iget(), and call that instead of iget().
romfs_iget() then uses iget_locked() directly and returns a proper error code
instead of an inode in the event of an error.romfs_fill_super() returns any error incurred when getting the root inode
instead of EINVAL.Signed-off-by: David Howells <dhowells@redhat.com>
---fs/romfs/inode.c | 45 +++++++++++++++++++++++++++++++--------------
1 files changed, 31 insertions(+), 14 deletions(-)diff --git a/fs/romfs/inode.c b/fs/romfs/inode.c
index dae7945..5071c9a 100644
--- a/fs/romfs/inode.c
+++ b/fs/romfs/inode.c
@@ -84,6 +84,8 @@ struct romfs_inode_info {
struct inode vfs_inode;
};+static struct inode *romfs_iget(struct super_block *, unsigned long);
+
/* instead of private superblock data */
static inline unsigned long romfs_maxsize(struct super_block *sb)
{
@@ -117,7 +119,7 @@ static int romfs_fill_super(struct super_block *s, void *data, int silent)
struct buffer_head *bh;
struct romfs_super_block *rsb;
struct inode *root;
- int sz;
+ int sz, ret = -EINVAL;/* I would parse the options here, but there are none.. :) */
@@ -157,10 +159,13 @@ static int romfs_fill_super(struct super_block *s, void *data, int silent)
& ROMFH_MASK;s->s_op = &romfs_ops;
- root = iget(s, sz);
- if (!root)
+ root = romfs_iget(s, sz);
+ if (IS_ERR(root)) {
+ ret = PTR_ERR(root);
goto out;
+ }+ ret = -ENOMEM;
s->s_root = d_alloc_root(root);
if (!s->s_root)
goto outiput;
@@ -173,7 +178,7 @@ outiput:
out:
brelse(bh);
outnobh:
- return -EINVAL;
+ return ret;
}/* That's simple too. */
@@ -389,8 +394,11 @@ romfs_lookup(struct inode *dir, struct dentry *dentry, struct nameidata *nd)
if ((be32_to_cpu(ri.next) & ROMFH_TYPE) == ROMFH_HRD)
offset = be32_to_cpu(ri.spec) & ROMFH_MASK;- if ((inode = iget(dir->i_sb, offset)))
- goto outi;
+ i...
Stop the JFFS2 filesystem from using iget() and read_inode(). Replace
jffs2_read_inode() with jffs2_iget(), and call that instead of iget().
jffs2_iget() then uses iget_locked() directly and returns a proper error code
instead of an inode in the event of an error.jffs2_do_fill_super() returns any error incurred when getting the root inode
instead of EINVAL.Signed-off-by: David Howells <dhowells@redhat.com>
---fs/jffs2/dir.c | 6 +++--
fs/jffs2/fs.c | 56 ++++++++++++++++++++++++++++++++-------------------
fs/jffs2/os-linux.h | 2 +-
fs/jffs2/super.c | 1 -
4 files changed, 39 insertions(+), 26 deletions(-)diff --git a/fs/jffs2/dir.c b/fs/jffs2/dir.c
index c1dfca3..ca0add3 100644
--- a/fs/jffs2/dir.c
+++ b/fs/jffs2/dir.c
@@ -101,10 +101,10 @@ static struct dentry *jffs2_lookup(struct inode *dir_i, struct dentry *target,
ino = fd->ino;
up(&dir_f->sem);
if (ino) {
- inode = iget(dir_i->i_sb, ino);
- if (!inode) {
+ inode = jffs2_iget(dir_i->i_sb, ino);
+ if (IS_ERR(inode)) {
printk(KERN_WARNING "iget() failed for ino #%u\n", ino);
- return (ERR_PTR(-EIO));
+ return ERR_CAST(inode);
}
}diff --git a/fs/jffs2/fs.c b/fs/jffs2/fs.c
index 8bc727b..900be8b 100644
--- a/fs/jffs2/fs.c
+++ b/fs/jffs2/fs.c
@@ -227,16 +227,23 @@ void jffs2_clear_inode (struct inode *inode)
jffs2_do_clear_inode(c, f);
}-void jffs2_read_inode (struct inode *inode)
+struct inode *jffs2_iget(struct super_block *sb, unsigned long ino)
{
struct jffs2_inode_info *f;
struct jffs2_sb_info *c;
struct jffs2_raw_inode latest_node;
union jffs2_device_node jdev;
+ struct inode *inode;
dev_t rdev = 0;
int ret;- D1(printk(KERN_DEBUG "jffs2_read_inode(): inode->i_ino == %lu\n", inode->i_ino));
+ D1(printk(KERN_DEBUG "jffs2_iget(): ino == %lu\n", ino));
+
+ inode = iget_locked(sb, ino);
+ if (!inode)
+ return ERR_PTR(-ENOMEM);
+ if (!(inode->i_state & I_NEW))
+ return inode;f ...
Stop the MINIX filesystem from using iget() and read_inode(). Replace
minix_read_inode() with minix_iget(), and call that instead of iget().
minix_iget() then uses iget_locked() directly and returns a proper error code
instead of an inode in the event of an error.minix_fill_super() returns any error incurred when getting the root inode
instead of EINVAL.Signed-off-by: David Howells <dhowells@redhat.com>
---fs/minix/inode.c | 43 +++++++++++++++++++++++++++++--------------
fs/minix/minix.h | 1 +
fs/minix/namei.c | 7 +++----
3 files changed, 33 insertions(+), 18 deletions(-)diff --git a/fs/minix/inode.c b/fs/minix/inode.c
index 43668d7..ee329f8 100644
--- a/fs/minix/inode.c
+++ b/fs/minix/inode.c
@@ -18,7 +18,6 @@
#include <linux/highuid.h>
#include <linux/vfs.h>-static void minix_read_inode(struct inode * inode);
static int minix_write_inode(struct inode * inode, int wait);
static int minix_statfs(struct dentry *dentry, struct kstatfs *buf);
static int minix_remount (struct super_block * sb, int * flags, char * data);
@@ -96,7 +95,6 @@ static void destroy_inodecache(void)
static const struct super_operations minix_sops = {
.alloc_inode = minix_alloc_inode,
.destroy_inode = minix_destroy_inode,
- .read_inode = minix_read_inode,
.write_inode = minix_write_inode,
.delete_inode = minix_delete_inode,
.put_super = minix_put_super,
@@ -149,6 +147,7 @@ static int minix_fill_super(struct super_block *s, void *data, int silent)
unsigned long i, block;
struct inode *root_inode;
struct minix_sb_info *sbi;
+ int ret = -EINVAL;sbi = kzalloc(sizeof(struct minix_sb_info), GFP_KERNEL);
if (!sbi)
@@ -246,10 +245,13 @@ static int minix_fill_super(struct super_block *s, void *data, int silent)/* set up enough so that it can read an inode */
s->s_op = &minix_sops;
- root_inode = iget(s, MINIX_ROOT_INO);
- if (!root_inode || is_bad_inode(root_inode))
+ root_inode = minix_iget(s, MINIX_ROOT_INO);
+ if (I...
Stop the ISOFS filesystem from using read_inode(). Make isofs_read_inode()
return an error code, and make isofs_iget() pass it on.Signed-off-by: David Howells <dhowells@redhat.com>
---fs/isofs/inode.c | 25 +++++++++++++++++--------
1 files changed, 17 insertions(+), 8 deletions(-)diff --git a/fs/isofs/inode.c b/fs/isofs/inode.c
index 043b470..28d990b 100644
--- a/fs/isofs/inode.c
+++ b/fs/isofs/inode.c
@@ -54,7 +54,7 @@ static void isofs_put_super(struct super_block *sb)
return;
}-static void isofs_read_inode(struct inode *);
+static int isofs_read_inode(struct inode *);
static int isofs_statfs (struct dentry *, struct kstatfs *);static struct kmem_cache *isofs_inode_cachep;
@@ -107,7 +107,6 @@ static int isofs_remount(struct super_block *sb, int *flags, char *data)
static const struct super_operations isofs_sops = {
.alloc_inode = isofs_alloc_inode,
.destroy_inode = isofs_destroy_inode,
- .read_inode = isofs_read_inode,
.put_super = isofs_put_super,
.statfs = isofs_statfs,
.remount_fs = isofs_remount,
@@ -1186,7 +1185,7 @@ out_toomany:
goto out;
}-static void isofs_read_inode(struct inode *inode)
+static int isofs_read_inode(struct inode *inode)
{
struct super_block *sb = inode->i_sb;
struct isofs_sb_info *sbi = ISOFS_SB(sb);
@@ -1199,6 +1198,7 @@ static void isofs_read_inode(struct inode *inode)
unsigned int de_len;
unsigned long offset;
struct iso_inode_info *ei = ISOFS_I(inode);
+ int ret = -EIO;block = ei->i_iget5_block;
bh = sb_bread(inode->i_sb, block);
@@ -1216,6 +1216,7 @@ static void isofs_read_inode(struct inode *inode)
tmpde = kmalloc(de_len, GFP_KERNEL);
if (tmpde == NULL) {
printk(KERN_INFO "%s: out of memory\n", __func__);
+ ret = -ENOMEM;
goto fail;
}
memcpy(tmpde, bh->b_data + offset, frag1);
@@ -1259,8 +1260,10 @@ static void isofs_read_inode(struct inode *inode)ei->i_section_size = isonum_733(de->size);
if (de->flags[-high_si...
Stop the HFSPLUS filesystem from using iget() and read_inode(). Replace
hfsplus_read_inode() with hfsplus_iget(), and call that instead of iget().
hfsplus_iget() then uses iget_locked() directly and returns a proper error code
instead of an inode in the event of an error.hfsplus_fill_super() returns any error incurred when getting the root inode.
Signed-off-by: David Howells <dhowells@redhat.com>
---fs/hfsplus/btree.c | 6 ++++--
fs/hfsplus/dir.c | 6 +++---
fs/hfsplus/hfsplus_fs.h | 3 +++
fs/hfsplus/super.c | 47 +++++++++++++++++++++++++++++++++--------------
4 files changed, 43 insertions(+), 19 deletions(-)diff --git a/fs/hfsplus/btree.c b/fs/hfsplus/btree.c
index 050d29c..bb54336 100644
--- a/fs/hfsplus/btree.c
+++ b/fs/hfsplus/btree.c
@@ -22,6 +22,7 @@ struct hfs_btree *hfs_btree_open(struct super_block *sb, u32 id)
struct hfs_btree *tree;
struct hfs_btree_header_rec *head;
struct address_space *mapping;
+ struct inode *inode;
struct page *page;
unsigned int size;@@ -33,9 +34,10 @@ struct hfs_btree *hfs_btree_open(struct super_block *sb, u32 id)
spin_lock_init(&tree->hash_lock);
tree->sb = sb;
tree->cnid = id;
- tree->inode = iget(sb, id);
- if (!tree->inode)
+ inode = hfsplus_iget(sb, id);
+ if (IS_ERR(inode))
goto free_tree;
+ tree->inode = inode;mapping = tree->inode->i_mapping;
page = read_mapping_page(mapping, 0, NULL);
diff --git a/fs/hfsplus/dir.c b/fs/hfsplus/dir.c
index 1955ee6..2968364 100644
--- a/fs/hfsplus/dir.c
+++ b/fs/hfsplus/dir.c
@@ -97,9 +97,9 @@ again:
goto fail;
}
hfs_find_exit(&fd);
- inode = iget(dir->i_sb, cnid);
- if (!inode)
- return ERR_PTR(-EACCES);
+ inode = hfsplus_iget(dir->i_sb, cnid);
+ if (IS_ERR(inode))
+ return ERR_CAST(inode);
if (S_ISREG(inode->i_mode))
HFSPLUS_I(inode).dev = linkid;
out:
diff --git a/fs/hfsplus/hfsplus_fs.h b/fs/hfsplus/hfsplus_fs.h
index d9f5eda..d72d0a8 100644
--- ...
Stop the FAT filesystem from using iget() and read_inode(). Replace
the call to iget() with a call to ilookup().Signed-off-by: David Howells <dhowells@redhat.com>
---fs/fat/inode.c | 6 ++----
1 files changed, 2 insertions(+), 4 deletions(-)diff --git a/fs/fat/inode.c b/fs/fat/inode.c
index 960ed3d..9ae2e9f 100644
--- a/fs/fat/inode.c
+++ b/fs/fat/inode.c
@@ -629,8 +629,6 @@ static const struct super_operations fat_sops = {
.clear_inode = fat_clear_inode,
.remount_fs = fat_remount,- .read_inode = make_bad_inode,
-
.show_options = fat_show_options,
};@@ -667,8 +665,8 @@ static struct dentry *fat_get_dentry(struct super_block *sb, void *inump)
struct dentry *result;
__u32 *fh = inump;- inode = iget(sb, fh[0]);
- if (!inode || is_bad_inode(inode) || inode->i_generation != fh[1]) {
+ inode = ilookup(sb, fh[0]);
+ if (!inode || inode->i_generation != fh[1]) {
if (inode)
iput(inode);
inode = NULL;-
Stop the FreeVXFS filesystem from using iget() and read_inode(). Replace
vxfs_read_inode() with vxfs_iget(), and call that instead of iget().
vxfs_iget() then uses iget_locked() directly and returns a proper error code
instead of an inode in the event of an error.vxfs_fill_super() returns any error incurred when getting the root inode
instead of EINVAL.Signed-off-by: David Howells <dhowells@redhat.com>
---fs/freevxfs/vxfs_extern.h | 2 +-
fs/freevxfs/vxfs_inode.c | 45 +++++++++++++++++++++++++++++----------------
fs/freevxfs/vxfs_lookup.c | 6 +++---
fs/freevxfs/vxfs_super.c | 10 +++++++---
4 files changed, 40 insertions(+), 23 deletions(-)diff --git a/fs/freevxfs/vxfs_extern.h b/fs/freevxfs/vxfs_extern.h
index 91ccee8..f307694 100644
--- a/fs/freevxfs/vxfs_extern.h
+++ b/fs/freevxfs/vxfs_extern.h
@@ -58,7 +58,7 @@ extern struct inode * vxfs_get_fake_inode(struct super_block *,
extern void vxfs_put_fake_inode(struct inode *);
extern struct vxfs_inode_info * vxfs_blkiget(struct super_block *, u_long, ino_t);
extern struct vxfs_inode_info * vxfs_stiget(struct super_block *, ino_t);
-extern void vxfs_read_inode(struct inode *);
+extern struct inode * vxfs_iget(struct super_block *, unsigned long);
extern void vxfs_clear_inode(struct inode *);/* vxfs_lookup.c */
diff --git a/fs/freevxfs/vxfs_inode.c b/fs/freevxfs/vxfs_inode.c
index d1f7c5b..95a0af9 100644
--- a/fs/freevxfs/vxfs_inode.c
+++ b/fs/freevxfs/vxfs_inode.c
@@ -129,7 +129,7 @@ fail:
* Description:
* Search the for inode number @ino in the filesystem
* described by @sbp. Use the specified inode table (@ilistp).
- * Returns the matching VxFS inode on success, else a NULL pointer.
+ * Returns the matching VxFS inode on success, else an error code.
*/
static struct vxfs_inode_info *
__vxfs_iget(ino_t ino, struct inode *ilistp)
@@ -157,12 +157,12 @@ __vxfs_iget(ino_t ino, struct inode *ilistp)
}printk(KERN_WARNING "vxfs: error on page %p\n", pp);
- r...
Stop the EXT4 filesystem from using iget() and read_inode(). Replace
ext4_read_inode() with ext4_iget(), and call that instead of iget().
ext4_iget() then uses iget_locked() directly and returns a proper error code
instead of an inode in the event of an error.ext4_fill_super() returns any error incurred when getting the root inode
instead of EINVAL.Signed-off-by: David Howells <dhowells@redhat.com>
Acked-by: "Theodore Ts'o" <tytso@mit.edu>
Acked-by: Jan Kara <jack@suse.cz>
---fs/ext4/ialloc.c | 58 ++++++++++++++++++++++++++++-------------------
fs/ext4/inode.c | 25 +++++++++++++++-----
fs/ext4/namei.c | 29 +++++++-----------------
fs/ext4/resize.c | 7 ++----
fs/ext4/super.c | 36 ++++++++++++++++-------------
include/linux/ext4_fs.h | 2 +-
6 files changed, 87 insertions(+), 70 deletions(-)diff --git a/fs/ext4/ialloc.c b/fs/ext4/ialloc.c
index 427f830..4fba007 100644
--- a/fs/ext4/ialloc.c
+++ b/fs/ext4/ialloc.c
@@ -659,14 +659,15 @@ struct inode *ext4_orphan_get(struct super_block *sb, unsigned long ino)
unsigned long max_ino = le32_to_cpu(EXT4_SB(sb)->s_es->s_inodes_count);
unsigned long block_group;
int bit;
- struct buffer_head *bitmap_bh = NULL;
+ struct buffer_head *bitmap_bh;
struct inode *inode = NULL;
+ long err = -EIO;/* Error cases - e2fsck has already cleaned up for us */
if (ino > max_ino) {
ext4_warning(sb, __FUNCTION__,
"bad orphan ino %lu! e2fsck was run?", ino);
- goto out;
+ goto error;
}block_group = (ino - 1) / EXT4_INODES_PER_GROUP(sb);
@@ -675,38 +676,49 @@ struct inode *ext4_orphan_get(struct super_block *sb, unsigned long ino)
if (!bitmap_bh) {
ext4_warning(sb, __FUNCTION__,
"inode bitmap error for orphan %lu", ino);
- goto out;
+ goto error;
}/* Having the inode bit set should be a 100% indicator that this
* is a valid orphan (no e2fsck run on fs). Orphans also include
* inodes t...
Stop the EXT3 filesystem from using iget() and read_inode(). Replace
ext3_read_inode() with ext3_iget(), and call that instead of iget().
ext3_iget() then uses iget_locked() directly and returns a proper error code
instead of an inode in the event of an error.ext3_fill_super() returns any error incurred when getting the root inode
instead of EINVAL.Signed-off-by: David Howells <dhowells@redhat.com>
Acked-by: "Theodore Ts'o" <tytso@mit.edu>
Acked-by: Jan Kara <jack@suse.cz>
---fs/ext3/ialloc.c | 58 ++++++++++++++++++++++++++++-------------------
fs/ext3/inode.c | 25 +++++++++++++++-----
fs/ext3/namei.c | 29 +++++++-----------------
fs/ext3/resize.c | 7 ++----
fs/ext3/super.c | 40 ++++++++++++++++++--------------
include/linux/ext3_fs.h | 2 +-
6 files changed, 89 insertions(+), 72 deletions(-)diff --git a/fs/ext3/ialloc.c b/fs/ext3/ialloc.c
index e45dbd6..5724adb 100644
--- a/fs/ext3/ialloc.c
+++ b/fs/ext3/ialloc.c
@@ -645,14 +645,15 @@ struct inode *ext3_orphan_get(struct super_block *sb, unsigned long ino)
unsigned long max_ino = le32_to_cpu(EXT3_SB(sb)->s_es->s_inodes_count);
unsigned long block_group;
int bit;
- struct buffer_head *bitmap_bh = NULL;
+ struct buffer_head *bitmap_bh;
struct inode *inode = NULL;
+ long err = -EIO;/* Error cases - e2fsck has already cleaned up for us */
if (ino > max_ino) {
ext3_warning(sb, __FUNCTION__,
"bad orphan ino %lu! e2fsck was run?", ino);
- goto out;
+ goto error;
}block_group = (ino - 1) / EXT3_INODES_PER_GROUP(sb);
@@ -661,38 +662,49 @@ struct inode *ext3_orphan_get(struct super_block *sb, unsigned long ino)
if (!bitmap_bh) {
ext3_warning(sb, __FUNCTION__,
"inode bitmap error for orphan %lu", ino);
- goto out;
+ goto error;
}/* Having the inode bit set should be a 100% indicator that this
* is a valid orphan (no e2fsck run on fs). Orphans also include
* inode...
Stop the EXT2 filesystem from using iget() and read_inode(). Replace
ext2_read_inode() with ext2_iget(), and call that instead of iget().
ext2_iget() then uses iget_locked() directly and returns a proper error code
instead of an inode in the event of an error.ext2_fill_super() returns any error incurred when getting the root inode
instead of EINVAL.Signed-off-by: David Howells <dhowells@redhat.com>
Acked-by: "Theodore Ts'o" <tytso@mit.edu>
---fs/ext2/ext2.h | 2 +-
fs/ext2/inode.c | 30 ++++++++++++++++++++++--------
fs/ext2/namei.c | 12 ++++++------
fs/ext2/super.c | 32 ++++++++++++++++++--------------
4 files changed, 47 insertions(+), 29 deletions(-)diff --git a/fs/ext2/ext2.h b/fs/ext2/ext2.h
index 9fd0ec5..85ca3de 100644
--- a/fs/ext2/ext2.h
+++ b/fs/ext2/ext2.h
@@ -123,7 +123,7 @@ extern void ext2_check_inodes_bitmap (struct super_block *);
extern unsigned long ext2_count_free (struct buffer_head *, unsigned);/* inode.c */
-extern void ext2_read_inode (struct inode *);
+extern struct inode *ext2_iget (struct super_block *, unsigned long);
extern int ext2_write_inode (struct inode *, int);
extern void ext2_put_inode (struct inode *);
extern void ext2_delete_inode (struct inode *);
diff --git a/fs/ext2/inode.c b/fs/ext2/inode.c
index 0079b2c..fa5e115 100644
--- a/fs/ext2/inode.c
+++ b/fs/ext2/inode.c
@@ -1074,20 +1074,32 @@ void ext2_get_inode_flags(struct ext2_inode_info *ei)
ei->i_flags |= EXT2_DIRSYNC_FL;
}-void ext2_read_inode (struct inode * inode)
+struct inode *ext2_iget (struct super_block *sb, unsigned long ino)
{
- struct ext2_inode_info *ei = EXT2_I(inode);
- ino_t ino = inode->i_ino;
+ struct ext2_inode_info *ei;
struct buffer_head * bh;
- struct ext2_inode * raw_inode = ext2_get_inode(inode->i_sb, ino, &bh);
+ struct ext2_inode * raw_inode;
+ struct inode *inode;
+ long ret = -EIO;
int n;+ inode = iget_locked(sb, ino);
+ if (!inode)
+ return ERR_PTR(-ENOMEM);
+ if (!(inode...
Stop the FUSE filesystem from using read_inode(), which it doesn't use anyway.
Signed-off-by: David Howells <dhowells@redhat.com>
---fs/fuse/inode.c | 6 ------
1 files changed, 0 insertions(+), 6 deletions(-)diff --git a/fs/fuse/inode.c b/fs/fuse/inode.c
index 5448f62..2986654 100644
--- a/fs/fuse/inode.c
+++ b/fs/fuse/inode.c
@@ -73,11 +73,6 @@ static void fuse_destroy_inode(struct inode *inode)
kmem_cache_free(fuse_inode_cachep, inode);
}-static void fuse_read_inode(struct inode *inode)
-{
- /* No op */
-}
-
void fuse_send_forget(struct fuse_conn *fc, struct fuse_req *req,
unsigned long nodeid, u64 nlookup)
{
@@ -452,7 +447,6 @@ static struct inode *get_root_inode(struct super_block *sb, unsigned mode)
static const struct super_operations fuse_super_operations = {
.alloc_inode = fuse_alloc_inode,
.destroy_inode = fuse_destroy_inode,
- .read_inode = fuse_read_inode,
.clear_inode = fuse_clear_inode,
.drop_inode = generic_delete_inode,
.remount_fs = fuse_remount_fs,-
Stop the EFS filesystem from using iget() and read_inode(). Replace
efs_read_inode() with efs_iget(), and call that instead of iget().
efs_iget() then uses iget_locked() directly and returns a proper error code
instead of an inode in the event of an error.efs_fill_super() returns any error incurred when getting the root inode
instead of EACCES.Signed-off-by: David Howells <dhowells@redhat.com>
---fs/efs/inode.c | 25 +++++++++++++++++--------
fs/efs/namei.c | 23 ++++++++++++-----------
fs/efs/super.c | 18 ++++++++++++------
include/linux/efs_fs.h | 2 +-
4 files changed, 42 insertions(+), 26 deletions(-)diff --git a/fs/efs/inode.c b/fs/efs/inode.c
index 174696f..627c302 100644
--- a/fs/efs/inode.c
+++ b/fs/efs/inode.c
@@ -45,17 +45,26 @@ static inline void extent_copy(efs_extent *src, efs_extent *dst) {
return;
}-void efs_read_inode(struct inode *inode)
+struct inode *efs_iget(struct super_block *super, unsigned long ino)
{
int i, inode_index;
dev_t device;
u32 rdev;
struct buffer_head *bh;
- struct efs_sb_info *sb = SUPER_INFO(inode->i_sb);
- struct efs_inode_info *in = INODE_INFO(inode);
+ struct efs_sb_info *sb = SUPER_INFO(super);
+ struct efs_inode_info *in;
efs_block_t block, offset;
struct efs_dinode *efs_inode;
-
+ struct inode *inode;
+
+ inode = iget_locked(super, ino);
+ if (IS_ERR(inode))
+ return ERR_PTR(-ENOMEM);
+ if (!(inode->i_state & I_NEW))
+ return inode;
+
+ in = INODE_INFO(inode);
+
/*
** EFS layout:
**
@@ -159,13 +168,13 @@ void efs_read_inode(struct inode *inode)
break;
}- return;
+ unlock_new_inode(inode);
+ return inode;read_inode_error:
printk(KERN_WARNING "EFS: failed to read inode %lu\n", inode->i_ino);
- make_bad_inode(inode);
-
- return;
+ iget_failed(inode);
+ return ERR_PTR(-EIO);
}static inline efs_block_t
diff --git a/fs/efs/namei.c b/fs/efs/namei.c
index 5276b19..77d7106 100644
--- ...
Stop the CIFS filesystem from using iget() and read_inode(). Replace
cifs_read_inode() with cifs_iget(), and call that instead of iget().
cifs_iget() then uses iget_locked() directly and returns a proper error code
instead of an inode in the event of an error.cifs_read_super() now returns any error incurred when getting the root inode
instead of ENOMEM.Signed-off-by: David Howells <dhowells@redhat.com>
---fs/cifs/cifsfs.c | 8 ++++----
fs/cifs/cifsfs.h | 1 +
fs/cifs/inode.c | 35 ++++++++++++++++++++++++++---------
3 files changed, 31 insertions(+), 13 deletions(-)diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c
index cabb6a5..1cb13f5 100644
--- a/fs/cifs/cifsfs.c
+++ b/fs/cifs/cifsfs.c
@@ -126,10 +126,11 @@ cifs_read_super(struct super_block *sb, void *data,
#endif
sb->s_blocksize = CIFS_MAX_MSGSIZE;
sb->s_blocksize_bits = 14; /* default 2**14 = CIFS_MAX_MSGSIZE */
- inode = iget(sb, ROOT_I);
+ inode = cifs_iget(sb, ROOT_I);- if (!inode) {
- rc = -ENOMEM;
+ if (IS_ERR(inode)) {
+ rc = PTR_ERR(inode);
+ inode = NULL;
goto out_no_root;
}@@ -481,7 +482,6 @@ static int cifs_remount(struct super_block *sb, int *flags, char *data)
}static const struct super_operations cifs_super_ops = {
- .read_inode = cifs_read_inode,
.put_super = cifs_put_super,
.statfs = cifs_statfs,
.alloc_inode = cifs_alloc_inode,
diff --git a/fs/cifs/cifsfs.h b/fs/cifs/cifsfs.h
index a20de77..74a3190 100644
--- a/fs/cifs/cifsfs.h
+++ b/fs/cifs/cifsfs.h
@@ -43,6 +43,7 @@ extern void cifs_read_inode(struct inode *);/* Functions related to inodes */
extern const struct inode_operations cifs_dir_inode_ops;
+extern struct inode *cifs_iget(struct super_block *, unsigned long);
extern int cifs_create(struct inode *, struct dentry *, int,
struct nameidata *);
extern struct dentry *cifs_lookup(struct inode *, struct dentry *,
diff --git a/fs/cifs/inode.c b/fs/cifs/inode.c
index dd41677..48966b9 100644
--- a/fs/cifs/inode...
Stop the autofs filesystem from using iget() and read_inode(). Replace
autofs_read_inode() with autofs_iget(), and call that instead of iget().
autofs_iget() then uses iget_locked() directly and returns a proper error code
instead of an inode in the event of an error.Signed-off-by: David Howells <dhowells@redhat.com>
---fs/autofs/autofs_i.h | 1 +
fs/autofs/inode.c | 27 ++++++++++++++++++---------
fs/autofs/root.c | 22 ++++++++++++++++++----
3 files changed, 37 insertions(+), 13 deletions(-)diff --git a/fs/autofs/autofs_i.h b/fs/autofs/autofs_i.h
index 8b4cca3..901a3e6 100644
--- a/fs/autofs/autofs_i.h
+++ b/fs/autofs/autofs_i.h
@@ -150,6 +150,7 @@ extern const struct file_operations autofs_root_operations;int autofs_fill_super(struct super_block *, void *, int);
void autofs_kill_sb(struct super_block *sb);
+struct inode *autofs_iget(struct super_block *, unsigned long);/* Queue management functions */
diff --git a/fs/autofs/inode.c b/fs/autofs/inode.c
index e7204d7..fc6a15e 100644
--- a/fs/autofs/inode.c
+++ b/fs/autofs/inode.c
@@ -52,10 +52,7 @@ out_kill_sb:
kill_anon_super(sb);
}-static void autofs_read_inode(struct inode *inode);
-
static const struct super_operations autofs_sops = {
- .read_inode = autofs_read_inode,
.statfs = simple_statfs,
};@@ -164,7 +161,9 @@ int autofs_fill_super(struct super_block *s, void *data, int silent)
s->s_time_gran = 1;
sbi->sb = s;- root_inode = iget(s, AUTOFS_ROOT_INO);
+ root_inode = autofs_iget(s, AUTOFS_ROOT_INO);
+ if (IS_ERR(root_inode))
+ goto fail_free;
root = d_alloc_root(root_inode);
pipe = NULL;@@ -230,11 +229,17 @@ fail_unlock:
return -EINVAL;
}-static void autofs_read_inode(struct inode *inode)
+struct inode *autofs_iget(struct super_block *sb, unsigned long ino)
{
- ino_t ino = inode->i_ino;
unsigned int n;
- struct autofs_sb_info *sbi = autofs_sbi(inode->i_sb);
+ struct autofs_sb_info *sbi = autofs_sbi(sb);
+ st...
Stop the AFFS filesystem from using iget() and read_inode(). Replace
affs_read_inode() with affs_iget(), and call that instead of iget().
affs_iget() then uses iget_locked() directly and returns a proper error code
instead of an inode in the event of an error.affs_fill_super() returns any error incurred when getting the root inode
instead of EINVAL.Signed-off-by: David Howells <dhowells@redhat.com>
---fs/affs/affs.h | 2 +-
fs/affs/amigaffs.c | 6 ++++--
fs/affs/inode.c | 20 +++++++++++++-------
fs/affs/namei.c | 10 ++++------
fs/affs/super.c | 12 +++++++++---
5 files changed, 31 insertions(+), 19 deletions(-)diff --git a/fs/affs/affs.h b/fs/affs/affs.h
index 232c694..0bce4ff 100644
--- a/fs/affs/affs.h
+++ b/fs/affs/affs.h
@@ -174,7 +174,7 @@ extern void affs_put_inode(struct inode *inode);
extern void affs_drop_inode(struct inode *inode);
extern void affs_delete_inode(struct inode *inode);
extern void affs_clear_inode(struct inode *inode);
-extern void affs_read_inode(struct inode *inode);
+extern struct inode *affs_iget(struct super_block *sb, unsigned long ino);
extern int affs_write_inode(struct inode *inode, int);
extern int affs_add_entry(struct inode *dir, struct inode *inode, struct dentry *dentry, s32 type);diff --git a/fs/affs/amigaffs.c b/fs/affs/amigaffs.c
index f4de4b9..8055730 100644
--- a/fs/affs/amigaffs.c
+++ b/fs/affs/amigaffs.c
@@ -170,9 +170,11 @@ affs_remove_link(struct dentry *dentry)
if (!link_bh)
goto done;- dir = iget(sb, be32_to_cpu(AFFS_TAIL(sb, link_bh)->parent));
- if (!dir)
+ dir = affs_iget(sb, be32_to_cpu(AFFS_TAIL(sb, link_bh)->parent));
+ if (IS_ERR(dir)) {
+ retval = PTR_ERR(dir);
goto done;
+ }affs_lock_dir(dir);
affs_fix_dcache(dentry, link_ino);
diff --git a/fs/affs/inode.c b/fs/affs/inode.c
index 4609a6c..edea76c 100644
--- a/fs/affs/inode.c
+++ b/fs/affs/inode.c
@@ -15,20 +15,25 @@
extern const struct inode_ope...
Stop the BEFS filesystem from using iget() and read_inode(). Replace
befs_read_inode() with befs_iget(), and call that instead of iget().
befs_iget() then uses iget_locked() directly and returns a proper error code
instead of an inode in the event of an error.befs_fill_super() returns any error incurred when getting the root inode
instead of EINVAL.Signed-off-by: David Howells <dhowells@redhat.com>
Acked-by: Will Dyson <will_dyson@pobox.com>
---fs/befs/linuxvfs.c | 39 +++++++++++++++++++++++++--------------
1 files changed, 25 insertions(+), 14 deletions(-)diff --git a/fs/befs/linuxvfs.c b/fs/befs/linuxvfs.c
index a451418..fd9866e 100644
--- a/fs/befs/linuxvfs.c
+++ b/fs/befs/linuxvfs.c
@@ -35,7 +35,7 @@ static int befs_get_block(struct inode *, sector_t, struct buffer_head *, int);
static int befs_readpage(struct file *file, struct page *page);
static sector_t befs_bmap(struct address_space *mapping, sector_t block);
static struct dentry *befs_lookup(struct inode *, struct dentry *, struct nameidata *);
-static void befs_read_inode(struct inode *ino);
+static struct inode *befs_iget(struct super_block *, unsigned long);
static struct inode *befs_alloc_inode(struct super_block *sb);
static void befs_destroy_inode(struct inode *inode);
static int befs_init_inodecache(void);
@@ -52,7 +52,6 @@ static int befs_statfs(struct dentry *, struct kstatfs *);
static int parse_options(char *, befs_mount_options *);static const struct super_operations befs_sops = {
- .read_inode = befs_read_inode, /* initialize & read inode */
.alloc_inode = befs_alloc_inode, /* allocate a new inode */
.destroy_inode = befs_destroy_inode, /* deallocate an inode */
.put_super = befs_put_super, /* uninit super */
@@ -198,9 +197,9 @@ befs_lookup(struct inode *dir, struct dentry *dentry, struct nameidata *nd)
return ERR_PTR(-ENODATA);
}- inode = iget(dir->i_sb, (ino_t) offset);
- if (!inode)
- return ERR_PTR(-EACCES);
+ inode = befs_iget(dir->i...
Stop the BFS filesystem from using iget() and read_inode(). Replace
bfs_read_inode() with bfs_iget(), and call that instead of iget().
bfs_iget() then uses iget_locked() directly and returns a proper error code
instead of an inode in the event of an error.bfs_fill_super() returns any error incurred when getting the root inode
instead of EINVAL.Signed-off-by: David Howells <dhowells@redhat.com>
---fs/bfs/bfs.h | 2 ++
fs/bfs/dir.c | 6 +++---
fs/bfs/inode.c | 32 ++++++++++++++++++++++----------
3 files changed, 27 insertions(+), 13 deletions(-)diff --git a/fs/bfs/bfs.h b/fs/bfs/bfs.h
index 130f6c6..5cf50d4 100644
--- a/fs/bfs/bfs.h
+++ b/fs/bfs/bfs.h
@@ -46,6 +46,8 @@ static inline struct bfs_inode_info *BFS_I(struct inode *inode)
#define printf(format, args...) \
printk(KERN_ERR "BFS-fs: %s(): " format, __FUNCTION__, ## args)+/* inode.c */
+extern struct inode *bfs_iget(struct super_block *sb, unsigned long ino);/* file.c */
extern const struct inode_operations bfs_file_inops;
diff --git a/fs/bfs/dir.c b/fs/bfs/dir.c
index 097f149..03c8bdb 100644
--- a/fs/bfs/dir.c
+++ b/fs/bfs/dir.c
@@ -141,10 +141,10 @@ static struct dentry * bfs_lookup(struct inode * dir, struct dentry * dentry, st
if (bh) {
unsigned long ino = (unsigned long)le16_to_cpu(de->ino);
brelse(bh);
- inode = iget(dir->i_sb, ino);
- if (!inode) {
+ inode = bfs_iget(dir->i_sb, ino);
+ if (IS_ERR(inode)) {
unlock_kernel();
- return ERR_PTR(-EACCES);
+ return ERR_CAST(inode);
}
}
unlock_kernel();
diff --git a/fs/bfs/inode.c b/fs/bfs/inode.c
index f346eb1..76798c9 100644
--- a/fs/bfs/inode.c
+++ b/fs/bfs/inode.c
@@ -32,25 +32,29 @@ MODULE_LICENSE("GPL");void dump_imap(const char *prefix, struct super_block * s);
-static void bfs_read_inode(struct inode * inode)
+struct inode *bfs_iget(struct super_block *sb, unsigned long ino)
{
- unsigned long ino = inode->i_ino;
struct bfs_inode * di;
struct buffer_head *...
Use iget_failed() in GFS2 to kill a failed inode.
Signed-off-by: David Howells <dhowells@redhat.com>
---fs/gfs2/inode.c | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)diff --git a/fs/gfs2/inode.c b/fs/gfs2/inode.c
index 34f7bcd..498844f 100644
--- a/fs/gfs2/inode.c
+++ b/fs/gfs2/inode.c
@@ -185,7 +185,7 @@ fail_put:
ip->i_gl->gl_object = NULL;
gfs2_glock_put(ip->i_gl);
fail:
- iput(inode);
+ iget_failed(inode);
return ERR_PTR(error);
}-
Use iget_failed() in AFS to kill a failed inode.
Signed-off-by: David Howells <dhowells@redhat.com>
---fs/afs/inode.c | 5 +----
1 files changed, 1 insertions(+), 4 deletions(-)diff --git a/fs/afs/inode.c b/fs/afs/inode.c
index d196840..ca9b02f 100644
--- a/fs/afs/inode.c
+++ b/fs/afs/inode.c
@@ -196,10 +196,7 @@ struct inode *afs_iget(struct super_block *sb, struct key *key,/* failure */
bad_inode:
- make_bad_inode(inode);
- unlock_new_inode(inode);
- iput(inode);
-
+ iget_failed(inode);
_leave(" = %d [bad]", ret);
return ERR_PTR(ret);
}-
Introduce a function to register failure in an inode construction path. This
includes marking the inode under construction as bad, unlocking it and
releasing it.Signed-off-by: David Howells <dhowells@redhat.com>
---Documentation/filesystems/porting | 18 +++++++++++++-----
fs/bad_inode.c | 14 ++++++++++++++
include/linux/fs.h | 1 +
3 files changed, 28 insertions(+), 5 deletions(-)diff --git a/Documentation/filesystems/porting b/Documentation/filesystems/porting
index dac45c9..3b0fb22 100644
--- a/Documentation/filesystems/porting
+++ b/Documentation/filesystems/porting
@@ -184,11 +184,19 @@ just takes the superblock and inode number as arguments and does the
test and set for you.e.g.
- inode = iget_locked(sb, ino);
- if (inode->i_state & I_NEW) {
- read_inode_from_disk(inode);
- unlock_new_inode(inode);
- }
+ inode = iget_locked(sb, ino);
+ if (inode->i_state & I_NEW) {
+ err = read_inode_from_disk(inode);
+ if (err < 0) {
+ iget_failed(inode);
+ return err;
+ }
+ unlock_new_inode(inode);
+ }
+
+Note that if the process of setting up a new inode fails, then iget_failed()
+should be called on the inode to render it dead, and an appropriate error
+should be passed back to the caller.---
[recommended]
diff --git a/fs/bad_inode.c b/fs/bad_inode.c
index 521ff7c..f1c2ea8 100644
--- a/fs/bad_inode.c
+++ b/fs/bad_inode.c
@@ -359,3 +359,17 @@ int is_bad_inode(struct inode *inode)
}EXPORT_SYMBOL(is_bad_inode);
+
+/**
+ * iget_failed - Mark an under-construction inode as dead and release it
+ * @inode: The inode to discard
+ *
+ * Mark an under-construction inode as dead and release it.
+ */
+void iget_failed(struct inode *inode)
+{
+ make_bad_inode(inode);
+ unlock_new_inode(inode);
+ iput(inode);
+}
+EXPORT_SYMBOL(iget_failed);
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 16421f6..c24d433 100644
--- a/in...
Convert instances of ERR_PTR(PTR_ERR(p)) to ERR_CAST(p) using:
perl -spi -e 's/ERR_PTR[(]PTR_ERR[(](.*)[)][)]/ERR_CAST(\1)/' `grep -rl 'ERR_PTR[(]*PTR_ERR' fs crypto net security`
Signed-off-by: David Howells <dhowells@redhat.com>
---crypto/cbc.c | 2 +-
crypto/cryptd.c | 4 ++--
crypto/ecb.c | 2 +-
crypto/hmac.c | 2 +-
crypto/lrw.c | 2 +-
crypto/pcbc.c | 2 +-
crypto/xcbc.c | 4 ++--
fs/9p/vfs_inode.c | 2 +-
fs/affs/namei.c | 2 +-
fs/afs/dir.c | 4 ++--
fs/afs/security.c | 2 +-
fs/fat/inode.c | 2 +-
fs/fuse/dir.c | 6 +++---
fs/gfs2/dir.c | 2 +-
fs/gfs2/ops_export.c | 2 +-
fs/gfs2/ops_inode.c | 2 +-
fs/jffs2/write.c | 4 ++--
fs/nfs/getroot.c | 8 ++++----
fs/nfsd/export.c | 4 ++--
fs/quota.c | 4 ++--
fs/reiserfs/inode.c | 2 +-
fs/reiserfs/xattr.c | 6 +++---
fs/vfat/namei.c | 2 +-
net/rxrpc/af_rxrpc.c | 6 +++---
security/keys/key.c | 2 +-
security/keys/process_keys.c | 2 +-
security/keys/request_key.c | 4 ++--
security/keys/request_key_auth.c | 2 +-
28 files changed, 44 insertions(+), 44 deletions(-)diff --git a/crypto/cbc.c b/crypto/cbc.c
index 1f2649e..ea39ba2 100644
--- a/crypto/cbc.c
+++ b/crypto/cbc.c
@@ -288,7 +288,7 @@ static struct crypto_instance *crypto_cbc_alloc(struct rtattr **tb)
alg = crypto_get_attr_alg(tb, CRYPTO_ALG_TYPE_CIPHER,
CRYPTO_ALG_TYPE_MASK);
if (IS_ERR(alg))
- return ERR_PTR(PTR_ERR(alg));
+ return ERR_CAST(alg);inst = crypto_alloc_instance("cbc", alg);
if (IS_ERR(inst))
diff --git...
Add an ERR_CAST() macro to complement ERR_PTR and co. for the purposes of
casting an error entyped as one pointer type to an error of another pointer
type whilst making it explicit as to what is going on.This provides a replacement for the ERR_PTR(PTR_ERR(p)) construct.
Signed-off-by: David Howells <dhowells@redhat.com>
---include/linux/err.h | 12 ++++++++++++
1 files changed, 12 insertions(+), 0 deletions(-)diff --git a/include/linux/err.h b/include/linux/err.h
index 1ab1d44..08409cd 100644
--- a/include/linux/err.h
+++ b/include/linux/err.h
@@ -34,6 +34,18 @@ static inline long IS_ERR(const void *ptr)
return IS_ERR_VALUE((unsigned long)ptr);
}+/**
+ * ERR_CAST - Explicitly cast an error-valued pointer to another pointer type
+ * @ptr: The pointer to cast.
+ *
+ * Explicitly cast an error-valued pointer to another pointer type in such a
+ * way as to make it clear that's what's going on.
+ */
+static inline void *ERR_CAST(const void *ptr)
+{
+ return (void *) ptr;
+}
+
#endif#endif /* _LINUX_ERR_H */
-
| Natalie Protasevich | [BUG] New Kernel Bugs |
| david | Re: Dual-Licensing Linux Kernel with GPL V2 and GPL V3 |
| Bart Van Assche | Integration of SCST in the mainstream Linux kernel |
| Greg KH | [GIT PATCH] driver core patches against 2.6.24 |
git: | |
| Jarek Poplawski | Re: [PATCH] pkt_sched: Destroy gen estimators under rtnl_lock(). |
| David Miller | [GIT]: Networking |
| Patrick McHardy | [NET_SCHED 00/04]: External SFQ classifiers/flow classifier |
| Gerrit Renker | [PATCH 27/37] dccp: Integration of dynamic feature activation - part 2 (server side) |
