aboutsummaryrefslogtreecommitdiff
path: root/fs/jfs
diff options
context:
space:
mode:
Diffstat (limited to 'fs/jfs')
-rw-r--r--fs/jfs/Makefile4
-rw-r--r--fs/jfs/acl.c136
-rw-r--r--fs/jfs/file.c43
-rw-r--r--fs/jfs/inode.c40
-rw-r--r--fs/jfs/ioctl.c51
-rw-r--r--fs/jfs/jfs_acl.h9
-rw-r--r--fs/jfs/jfs_discard.c121
-rw-r--r--fs/jfs/jfs_discard.h26
-rw-r--r--fs/jfs/jfs_dmap.c206
-rw-r--r--fs/jfs/jfs_dmap.h2
-rw-r--r--fs/jfs/jfs_dtree.c125
-rw-r--r--fs/jfs/jfs_dtree.h2
-rw-r--r--fs/jfs/jfs_extent.c8
-rw-r--r--fs/jfs/jfs_filsys.h3
-rw-r--r--fs/jfs/jfs_imap.c125
-rw-r--r--fs/jfs/jfs_incore.h12
-rw-r--r--fs/jfs/jfs_inode.c21
-rw-r--r--fs/jfs/jfs_inode.h4
-rw-r--r--fs/jfs/jfs_logmgr.c47
-rw-r--r--fs/jfs/jfs_logmgr.h2
-rw-r--r--fs/jfs/jfs_metapage.c20
-rw-r--r--fs/jfs/jfs_metapage.h2
-rw-r--r--fs/jfs/jfs_superblock.h1
-rw-r--r--fs/jfs/jfs_txnmgr.c23
-rw-r--r--fs/jfs/jfs_umount.c4
-rw-r--r--fs/jfs/jfs_xattr.h7
-rw-r--r--fs/jfs/jfs_xtree.c62
-rw-r--r--fs/jfs/namei.c167
-rw-r--r--fs/jfs/resize.c8
-rw-r--r--fs/jfs/super.c243
-rw-r--r--fs/jfs/xattr.c184
31 files changed, 1024 insertions, 684 deletions
diff --git a/fs/jfs/Makefile b/fs/jfs/Makefile
index 3adb6395e42..d20d4737b3e 100644
--- a/fs/jfs/Makefile
+++ b/fs/jfs/Makefile
@@ -6,11 +6,11 @@ obj-$(CONFIG_JFS_FS) += jfs.o
jfs-y := super.o file.o inode.o namei.o jfs_mount.o jfs_umount.o \
jfs_xtree.o jfs_imap.o jfs_debug.o jfs_dmap.o \
- jfs_unicode.o jfs_dtree.o jfs_inode.o \
+ jfs_unicode.o jfs_dtree.o jfs_inode.o jfs_discard.o \
jfs_extent.o symlink.o jfs_metapage.o \
jfs_logmgr.o jfs_txnmgr.o jfs_uniupr.o \
resize.o xattr.o ioctl.o
jfs-$(CONFIG_JFS_POSIX_ACL) += acl.o
-EXTRA_CFLAGS += -D_JFS_4K
+ccflags-y := -D_JFS_4K
diff --git a/fs/jfs/acl.c b/fs/jfs/acl.c
index 1057a4998e4..0c8ca830b11 100644
--- a/fs/jfs/acl.c
+++ b/fs/jfs/acl.c
@@ -27,7 +27,7 @@
#include "jfs_xattr.h"
#include "jfs_acl.h"
-static struct posix_acl *jfs_get_acl(struct inode *inode, int type)
+struct posix_acl *jfs_get_acl(struct inode *inode, int type)
{
struct posix_acl *acl;
char *ea_name;
@@ -64,7 +64,7 @@ static struct posix_acl *jfs_get_acl(struct inode *inode, int type)
else
acl = ERR_PTR(size);
} else {
- acl = posix_acl_from_xattr(value, size);
+ acl = posix_acl_from_xattr(&init_user_ns, value, size);
}
kfree(value);
if (!IS_ERR(acl))
@@ -72,7 +72,7 @@ static struct posix_acl *jfs_get_acl(struct inode *inode, int type)
return acl;
}
-static int jfs_set_acl(tid_t tid, struct inode *inode, int type,
+static int __jfs_set_acl(tid_t tid, struct inode *inode, int type,
struct posix_acl *acl)
{
char *ea_name;
@@ -80,27 +80,32 @@ static int jfs_set_acl(tid_t tid, struct inode *inode, int type,
int size = 0;
char *value = NULL;
- if (S_ISLNK(inode->i_mode))
- return -EOPNOTSUPP;
-
- switch(type) {
- case ACL_TYPE_ACCESS:
- ea_name = POSIX_ACL_XATTR_ACCESS;
- break;
- case ACL_TYPE_DEFAULT:
- ea_name = POSIX_ACL_XATTR_DEFAULT;
- if (!S_ISDIR(inode->i_mode))
- return acl ? -EACCES : 0;
- break;
- default:
- return -EINVAL;
+ switch (type) {
+ case ACL_TYPE_ACCESS:
+ ea_name = POSIX_ACL_XATTR_ACCESS;
+ if (acl) {
+ rc = posix_acl_equiv_mode(acl, &inode->i_mode);
+ if (rc < 0)
+ return rc;
+ inode->i_ctime = CURRENT_TIME;
+ mark_inode_dirty(inode);
+ if (rc == 0)
+ acl = NULL;
+ }
+ break;
+ case ACL_TYPE_DEFAULT:
+ ea_name = POSIX_ACL_XATTR_DEFAULT;
+ break;
+ default:
+ return -EINVAL;
}
+
if (acl) {
size = posix_acl_xattr_size(acl->a_count);
value = kmalloc(size, GFP_KERNEL);
if (!value)
return -ENOMEM;
- rc = posix_acl_to_xattr(acl, value, size);
+ rc = posix_acl_to_xattr(&init_user_ns, acl, value, size);
if (rc < 0)
goto out;
}
@@ -114,94 +119,43 @@ out:
return rc;
}
-int jfs_check_acl(struct inode *inode, int mask)
+int jfs_set_acl(struct inode *inode, struct posix_acl *acl, int type)
{
- struct posix_acl *acl = jfs_get_acl(inode, ACL_TYPE_ACCESS);
-
- if (IS_ERR(acl))
- return PTR_ERR(acl);
- if (acl) {
- int error = posix_acl_permission(inode, acl, mask);
- posix_acl_release(acl);
- return error;
- }
+ int rc;
+ tid_t tid;
- return -EAGAIN;
+ tid = txBegin(inode->i_sb, 0);
+ mutex_lock(&JFS_IP(inode)->commit_mutex);
+ rc = __jfs_set_acl(tid, inode, type, acl);
+ if (!rc)
+ rc = txCommit(tid, 1, &inode, 0);
+ txEnd(tid);
+ mutex_unlock(&JFS_IP(inode)->commit_mutex);
+ return rc;
}
int jfs_init_acl(tid_t tid, struct inode *inode, struct inode *dir)
{
- struct posix_acl *acl = NULL;
- struct posix_acl *clone;
- mode_t mode;
+ struct posix_acl *default_acl, *acl;
int rc = 0;
- if (S_ISLNK(inode->i_mode))
- return 0;
+ rc = posix_acl_create(dir, &inode->i_mode, &default_acl, &acl);
+ if (rc)
+ return rc;
- acl = jfs_get_acl(dir, ACL_TYPE_DEFAULT);
- if (IS_ERR(acl))
- return PTR_ERR(acl);
+ if (default_acl) {
+ rc = __jfs_set_acl(tid, inode, ACL_TYPE_DEFAULT, default_acl);
+ posix_acl_release(default_acl);
+ }
if (acl) {
- if (S_ISDIR(inode->i_mode)) {
- rc = jfs_set_acl(tid, inode, ACL_TYPE_DEFAULT, acl);
- if (rc)
- goto cleanup;
- }
- clone = posix_acl_clone(acl, GFP_KERNEL);
- if (!clone) {
- rc = -ENOMEM;
- goto cleanup;
- }
- mode = inode->i_mode;
- rc = posix_acl_create_masq(clone, &mode);
- if (rc >= 0) {
- inode->i_mode = mode;
- if (rc > 0)
- rc = jfs_set_acl(tid, inode, ACL_TYPE_ACCESS,
- clone);
- }
- posix_acl_release(clone);
-cleanup:
+ if (!rc)
+ rc = __jfs_set_acl(tid, inode, ACL_TYPE_ACCESS, acl);
posix_acl_release(acl);
- } else
- inode->i_mode &= ~current_umask();
+ }
JFS_IP(inode)->mode2 = (JFS_IP(inode)->mode2 & 0xffff0000) |
inode->i_mode;
return rc;
}
-
-int jfs_acl_chmod(struct inode *inode)
-{
- struct posix_acl *acl, *clone;
- int rc;
-
- if (S_ISLNK(inode->i_mode))
- return -EOPNOTSUPP;
-
- acl = jfs_get_acl(inode, ACL_TYPE_ACCESS);
- if (IS_ERR(acl) || !acl)
- return PTR_ERR(acl);
-
- clone = posix_acl_clone(acl, GFP_KERNEL);
- posix_acl_release(acl);
- if (!clone)
- return -ENOMEM;
-
- rc = posix_acl_chmod_masq(clone, inode->i_mode);
- if (!rc) {
- tid_t tid = txBegin(inode->i_sb, 0);
- mutex_lock(&JFS_IP(inode)->commit_mutex);
- rc = jfs_set_acl(tid, inode, ACL_TYPE_ACCESS, clone);
- if (!rc)
- rc = txCommit(tid, 1, &inode, 0);
- txEnd(tid);
- mutex_unlock(&JFS_IP(inode)->commit_mutex);
- }
-
- posix_acl_release(clone);
- return rc;
-}
diff --git a/fs/jfs/file.c b/fs/jfs/file.c
index c5ce6c1d1ff..33aa0cc1f8b 100644
--- a/fs/jfs/file.c
+++ b/fs/jfs/file.c
@@ -19,6 +19,7 @@
#include <linux/mm.h>
#include <linux/fs.h>
+#include <linux/posix_acl.h>
#include <linux/quotaops.h>
#include "jfs_incore.h"
#include "jfs_inode.h"
@@ -28,19 +29,26 @@
#include "jfs_acl.h"
#include "jfs_debug.h"
-int jfs_fsync(struct file *file, int datasync)
+int jfs_fsync(struct file *file, loff_t start, loff_t end, int datasync)
{
struct inode *inode = file->f_mapping->host;
int rc = 0;
+ rc = filemap_write_and_wait_range(inode->i_mapping, start, end);
+ if (rc)
+ return rc;
+
+ mutex_lock(&inode->i_mutex);
if (!(inode->i_state & I_DIRTY) ||
(datasync && !(inode->i_state & I_DIRTY_DATASYNC))) {
/* Make sure committed changes hit the disk */
jfs_flush_journal(JFS_SBI(inode->i_sb)->log, 1);
+ mutex_unlock(&inode->i_mutex);
return rc;
}
rc |= jfs_commit_inode(inode, 1);
+ mutex_unlock(&inode->i_mutex);
return rc ? -EIO : 0;
}
@@ -66,9 +74,9 @@ static int jfs_open(struct inode *inode, struct file *file)
struct jfs_inode_info *ji = JFS_IP(inode);
spin_lock_irq(&ji->ag_lock);
if (ji->active_ag == -1) {
- ji->active_ag = ji->agno;
- atomic_inc(
- &JFS_SBI(inode->i_sb)->bmap->db_active[ji->agno]);
+ struct jfs_sb_info *jfs_sb = JFS_SBI(inode->i_sb);
+ ji->active_ag = BLKTOAG(addressPXD(&ji->ixpxd), jfs_sb);
+ atomic_inc( &jfs_sb->bmap->db_active[ji->active_ag]);
}
spin_unlock_irq(&ji->ag_lock);
}
@@ -101,8 +109,8 @@ int jfs_setattr(struct dentry *dentry, struct iattr *iattr)
if (is_quota_modification(inode, iattr))
dquot_initialize(inode);
- if ((iattr->ia_valid & ATTR_UID && iattr->ia_uid != inode->i_uid) ||
- (iattr->ia_valid & ATTR_GID && iattr->ia_gid != inode->i_gid)) {
+ if ((iattr->ia_valid & ATTR_UID && !uid_eq(iattr->ia_uid, inode->i_uid)) ||
+ (iattr->ia_valid & ATTR_GID && !gid_eq(iattr->ia_gid, inode->i_gid))) {
rc = dquot_transfer(inode, iattr);
if (rc)
return rc;
@@ -110,41 +118,46 @@ int jfs_setattr(struct dentry *dentry, struct iattr *iattr)
if ((iattr->ia_valid & ATTR_SIZE) &&
iattr->ia_size != i_size_read(inode)) {
- rc = vmtruncate(inode, iattr->ia_size);
+ inode_dio_wait(inode);
+
+ rc = inode_newsize_ok(inode, iattr->ia_size);
if (rc)
return rc;
+
+ truncate_setsize(inode, iattr->ia_size);
+ jfs_truncate(inode);
}
setattr_copy(inode, iattr);
mark_inode_dirty(inode);
if (iattr->ia_valid & ATTR_MODE)
- rc = jfs_acl_chmod(inode);
+ rc = posix_acl_chmod(inode, inode->i_mode);
return rc;
}
const struct inode_operations jfs_file_inode_operations = {
- .truncate = jfs_truncate,
.setxattr = jfs_setxattr,
.getxattr = jfs_getxattr,
.listxattr = jfs_listxattr,
.removexattr = jfs_removexattr,
.setattr = jfs_setattr,
#ifdef CONFIG_JFS_POSIX_ACL
- .check_acl = jfs_check_acl,
+ .get_acl = jfs_get_acl,
+ .set_acl = jfs_set_acl,
#endif
};
const struct file_operations jfs_file_operations = {
.open = jfs_open,
.llseek = generic_file_llseek,
- .write = do_sync_write,
- .read = do_sync_read,
- .aio_read = generic_file_aio_read,
- .aio_write = generic_file_aio_write,
+ .write = new_sync_write,
+ .read = new_sync_read,
+ .read_iter = generic_file_read_iter,
+ .write_iter = generic_file_write_iter,
.mmap = generic_file_mmap,
.splice_read = generic_file_splice_read,
- .splice_write = generic_file_splice_write,
+ .splice_write = iter_file_splice_write,
.fsync = jfs_fsync,
.release = jfs_release,
.unlocked_ioctl = jfs_ioctl,
diff --git a/fs/jfs/inode.c b/fs/jfs/inode.c
index 9978803ceed..bd3df1ca3c9 100644
--- a/fs/jfs/inode.c
+++ b/fs/jfs/inode.c
@@ -23,6 +23,7 @@
#include <linux/pagemap.h>
#include <linux/quotaops.h>
#include <linux/writeback.h>
+#include <linux/aio.h>
#include "jfs_incore.h"
#include "jfs_inode.h"
#include "jfs_filsys.h"
@@ -125,7 +126,7 @@ int jfs_write_inode(struct inode *inode, struct writeback_control *wbc)
{
int wait = wbc->sync_mode == WB_SYNC_ALL;
- if (test_cflag(COMMIT_Nolink, inode))
+ if (inode->i_nlink == 0)
return 0;
/*
* If COMMIT_DIRTY is not set, the inode isn't really dirty.
@@ -153,7 +154,7 @@ void jfs_evict_inode(struct inode *inode)
dquot_initialize(inode);
if (JFS_IP(inode)->fileset == FILESYSTEM_I) {
- truncate_inode_pages(&inode->i_data, 0);
+ truncate_inode_pages_final(&inode->i_data);
if (test_cflag(COMMIT_Freewmap, inode))
jfs_free_zero_link(inode);
@@ -167,13 +168,13 @@ void jfs_evict_inode(struct inode *inode)
dquot_free_inode(inode);
}
} else {
- truncate_inode_pages(&inode->i_data, 0);
+ truncate_inode_pages_final(&inode->i_data);
}
- end_writeback(inode);
+ clear_inode(inode);
dquot_drop(inode);
}
-void jfs_dirty_inode(struct inode *inode)
+void jfs_dirty_inode(struct inode *inode, int flags)
{
static int noisy = 5;
@@ -300,6 +301,16 @@ static int jfs_readpages(struct file *file, struct address_space *mapping,
return mpage_readpages(mapping, pages, nr_pages, jfs_get_block);
}
+static void jfs_write_failed(struct address_space *mapping, loff_t to)
+{
+ struct inode *inode = mapping->host;
+
+ if (to > inode->i_size) {
+ truncate_pagecache(inode, inode->i_size);
+ jfs_truncate(inode);
+ }
+}
+
static int jfs_write_begin(struct file *file, struct address_space *mapping,
loff_t pos, unsigned len, unsigned flags,
struct page **pagep, void **fsdata)
@@ -308,11 +319,8 @@ static int jfs_write_begin(struct file *file, struct address_space *mapping,
ret = nobh_write_begin(mapping, pos, len, flags, pagep, fsdata,
jfs_get_block);
- if (unlikely(ret)) {
- loff_t isize = mapping->host->i_size;
- if (pos + len > isize)
- vmtruncate(mapping->host, isize);
- }
+ if (unlikely(ret))
+ jfs_write_failed(mapping, pos + len);
return ret;
}
@@ -323,14 +331,15 @@ static sector_t jfs_bmap(struct address_space *mapping, sector_t block)
}
static ssize_t jfs_direct_IO(int rw, struct kiocb *iocb,
- const struct iovec *iov, loff_t offset, unsigned long nr_segs)
+ struct iov_iter *iter, loff_t offset)
{
struct file *file = iocb->ki_filp;
+ struct address_space *mapping = file->f_mapping;
struct inode *inode = file->f_mapping->host;
+ size_t count = iov_iter_count(iter);
ssize_t ret;
- ret = blockdev_direct_IO(rw, iocb, inode, inode->i_sb->s_bdev, iov,
- offset, nr_segs, jfs_get_block, NULL);
+ ret = blockdev_direct_IO(rw, iocb, inode, iter, offset, jfs_get_block);
/*
* In case of error extending write may have instantiated a few
@@ -338,10 +347,10 @@ static ssize_t jfs_direct_IO(int rw, struct kiocb *iocb,
*/
if (unlikely((rw & WRITE) && ret < 0)) {
loff_t isize = i_size_read(inode);
- loff_t end = offset + iov_length(iov, nr_segs);
+ loff_t end = offset + count;
if (end > isize)
- vmtruncate(inode, isize);
+ jfs_write_failed(mapping, end);
}
return ret;
@@ -352,7 +361,6 @@ const struct address_space_operations jfs_aops = {
.readpages = jfs_readpages,
.writepage = jfs_writepage,
.writepages = jfs_writepages,
- .sync_page = block_sync_page,
.write_begin = jfs_write_begin,
.write_end = nobh_write_end,
.bmap = jfs_bmap,
diff --git a/fs/jfs/ioctl.c b/fs/jfs/ioctl.c
index afe222bf300..93a1232894f 100644
--- a/fs/jfs/ioctl.c
+++ b/fs/jfs/ioctl.c
@@ -11,13 +11,17 @@
#include <linux/mount.h>
#include <linux/time.h>
#include <linux/sched.h>
+#include <linux/blkdev.h>
#include <asm/current.h>
#include <asm/uaccess.h>
+#include "jfs_filsys.h"
+#include "jfs_debug.h"
#include "jfs_incore.h"
#include "jfs_dinode.h"
#include "jfs_inode.h"
-
+#include "jfs_dmap.h"
+#include "jfs_discard.h"
static struct {
long jfs_flag;
@@ -54,7 +58,7 @@ static long jfs_map_ext2(unsigned long flags, int from)
long jfs_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
{
- struct inode *inode = filp->f_dentry->d_inode;
+ struct inode *inode = file_inode(filp);
struct jfs_inode_info *jfs_inode = JFS_IP(inode);
unsigned int flags;
@@ -68,11 +72,11 @@ long jfs_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
unsigned int oldflags;
int err;
- err = mnt_want_write(filp->f_path.mnt);
+ err = mnt_want_write_file(filp);
if (err)
return err;
- if (!is_owner_or_cap(inode)) {
+ if (!inode_owner_or_capable(inode)) {
err = -EACCES;
goto setflags_out;
}
@@ -120,9 +124,43 @@ long jfs_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
inode->i_ctime = CURRENT_TIME_SEC;
mark_inode_dirty(inode);
setflags_out:
- mnt_drop_write(filp->f_path.mnt);
+ mnt_drop_write_file(filp);
return err;
}
+
+ case FITRIM:
+ {
+ struct super_block *sb = inode->i_sb;
+ struct request_queue *q = bdev_get_queue(sb->s_bdev);
+ struct fstrim_range range;
+ s64 ret = 0;
+
+ if (!capable(CAP_SYS_ADMIN))
+ return -EPERM;
+
+ if (!blk_queue_discard(q)) {
+ jfs_warn("FITRIM not supported on device");
+ return -EOPNOTSUPP;
+ }
+
+ if (copy_from_user(&range, (struct fstrim_range __user *)arg,
+ sizeof(range)))
+ return -EFAULT;
+
+ range.minlen = max_t(unsigned int, range.minlen,
+ q->limits.discard_granularity);
+
+ ret = jfs_ioc_trim(inode, &range);
+ if (ret < 0)
+ return ret;
+
+ if (copy_to_user((struct fstrim_range __user *)arg, &range,
+ sizeof(range)))
+ return -EFAULT;
+
+ return 0;
+ }
+
default:
return -ENOTTY;
}
@@ -142,6 +180,9 @@ long jfs_compat_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
case JFS_IOC_SETFLAGS32:
cmd = JFS_IOC_SETFLAGS;
break;
+ case FITRIM:
+ cmd = FITRIM;
+ break;
}
return jfs_ioctl(filp, cmd, arg);
}
diff --git a/fs/jfs/jfs_acl.h b/fs/jfs/jfs_acl.h
index 54e07559878..489f993b7b1 100644
--- a/fs/jfs/jfs_acl.h
+++ b/fs/jfs/jfs_acl.h
@@ -20,9 +20,9 @@
#ifdef CONFIG_JFS_POSIX_ACL
-int jfs_check_acl(struct inode *, int);
+struct posix_acl *jfs_get_acl(struct inode *inode, int type);
+int jfs_set_acl(struct inode *inode, struct posix_acl *acl, int type);
int jfs_init_acl(tid_t, struct inode *, struct inode *);
-int jfs_acl_chmod(struct inode *inode);
#else
@@ -32,10 +32,5 @@ static inline int jfs_init_acl(tid_t tid, struct inode *inode,
return 0;
}
-static inline int jfs_acl_chmod(struct inode *inode)
-{
- return 0;
-}
-
#endif
#endif /* _H_JFS_ACL */
diff --git a/fs/jfs/jfs_discard.c b/fs/jfs/jfs_discard.c
new file mode 100644
index 00000000000..dfcd5030455
--- /dev/null
+++ b/fs/jfs/jfs_discard.c
@@ -0,0 +1,121 @@
+/*
+ * Copyright (C) Tino Reichardt, 2012
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See
+ * the GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ */
+
+#include <linux/fs.h>
+#include <linux/slab.h>
+#include <linux/blkdev.h>
+
+#include "jfs_incore.h"
+#include "jfs_superblock.h"
+#include "jfs_discard.h"
+#include "jfs_dmap.h"
+#include "jfs_debug.h"
+
+
+/*
+ * NAME: jfs_issue_discard()
+ *
+ * FUNCTION: TRIM the specified block range on device, if supported
+ *
+ * PARAMETERS:
+ * ip - pointer to in-core inode
+ * blkno - starting block number to be trimmed (0..N)
+ * nblocks - number of blocks to be trimmed
+ *
+ * RETURN VALUES:
+ * none
+ *
+ * serialization: IREAD_LOCK(ipbmap) held on entry/exit;
+ */
+void jfs_issue_discard(struct inode *ip, u64 blkno, u64 nblocks)
+{
+ struct super_block *sb = ip->i_sb;
+ int r = 0;
+
+ r = sb_issue_discard(sb, blkno, nblocks, GFP_NOFS, 0);
+ if (unlikely(r != 0)) {
+ jfs_err("JFS: sb_issue_discard" \
+ "(%p, %llu, %llu, GFP_NOFS, 0) = %d => failed!\n",
+ sb, (unsigned long long)blkno,
+ (unsigned long long)nblocks, r);
+ }
+
+ jfs_info("JFS: sb_issue_discard" \
+ "(%p, %llu, %llu, GFP_NOFS, 0) = %d\n",
+ sb, (unsigned long long)blkno,
+ (unsigned long long)nblocks, r);
+
+ return;
+}
+
+/*
+ * NAME: jfs_ioc_trim()
+ *
+ * FUNCTION: attempt to discard (TRIM) all free blocks from the
+ * filesystem.
+ *
+ * PARAMETERS:
+ * ip - pointer to in-core inode;
+ * range - the range, given by user space
+ *
+ * RETURN VALUES:
+ * 0 - success
+ * -EIO - i/o error
+ */
+int jfs_ioc_trim(struct inode *ip, struct fstrim_range *range)
+{
+ struct inode *ipbmap = JFS_SBI(ip->i_sb)->ipbmap;
+ struct bmap *bmp = JFS_SBI(ip->i_sb)->bmap;
+ struct super_block *sb = ipbmap->i_sb;
+ int agno, agno_end;
+ u64 start, end, minlen;
+ u64 trimmed = 0;
+
+ /**
+ * convert byte values to block size of filesystem:
+ * start: First Byte to trim
+ * len: number of Bytes to trim from start
+ * minlen: minimum extent length in Bytes
+ */
+ start = range->start >> sb->s_blocksize_bits;
+ end = start + (range->len >> sb->s_blocksize_bits) - 1;
+ minlen = range->minlen >> sb->s_blocksize_bits;
+ if (minlen == 0)
+ minlen = 1;
+
+ if (minlen > bmp->db_agsize ||
+ start >= bmp->db_mapsize ||
+ range->len < sb->s_blocksize)
+ return -EINVAL;
+
+ if (end >= bmp->db_mapsize)
+ end = bmp->db_mapsize - 1;
+
+ /**
+ * we trim all ag's within the range
+ */
+ agno = BLKTOAG(start, JFS_SBI(ip->i_sb));
+ agno_end = BLKTOAG(end, JFS_SBI(ip->i_sb));
+ while (agno <= agno_end) {
+ trimmed += dbDiscardAG(ip, agno, minlen);
+ agno++;
+ }
+ range->len = trimmed << sb->s_blocksize_bits;
+
+ return 0;
+}
diff --git a/fs/jfs/jfs_discard.h b/fs/jfs/jfs_discard.h
new file mode 100644
index 00000000000..40d1ee6081a
--- /dev/null
+++ b/fs/jfs/jfs_discard.h
@@ -0,0 +1,26 @@
+/*
+ * Copyright (C) Tino Reichardt, 2012
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See
+ * the GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ */
+#ifndef _H_JFS_DISCARD
+#define _H_JFS_DISCARD
+
+struct fstrim_range;
+
+extern void jfs_issue_discard(struct inode *ip, u64 blkno, u64 nblocks);
+extern int jfs_ioc_trim(struct inode *ip, struct fstrim_range *range);
+
+#endif /* _H_JFS_DISCARD */
diff --git a/fs/jfs/jfs_dmap.c b/fs/jfs/jfs_dmap.c
index c92ea3b3ea5..2d514c7affc 100644
--- a/fs/jfs/jfs_dmap.c
+++ b/fs/jfs/jfs_dmap.c
@@ -1,5 +1,6 @@
/*
* Copyright (C) International Business Machines Corp., 2000-2004
+ * Portions Copyright (C) Tino Reichardt, 2012
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
@@ -25,6 +26,7 @@
#include "jfs_lock.h"
#include "jfs_metapage.h"
#include "jfs_debug.h"
+#include "jfs_discard.h"
/*
* SERIALIZATION of the Block Allocation Map.
@@ -104,7 +106,6 @@ static int dbFreeBits(struct bmap * bmp, struct dmap * dp, s64 blkno,
static int dbFreeDmap(struct bmap * bmp, struct dmap * dp, s64 blkno,
int nblocks);
static int dbMaxBud(u8 * cp);
-s64 dbMapFileSizeToMapSize(struct inode *ipbmap);
static int blkstol2(s64 nb);
static int cntlz(u32 value);
@@ -145,7 +146,6 @@ static const s8 budtab[256] = {
2, 1, 1, 1, 1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, -1
};
-
/*
* NAME: dbMount()
*
@@ -310,7 +310,6 @@ int dbSync(struct inode *ipbmap)
return (0);
}
-
/*
* NAME: dbFree()
*
@@ -337,6 +336,7 @@ int dbFree(struct inode *ip, s64 blkno, s64 nblocks)
s64 lblkno, rem;
struct inode *ipbmap = JFS_SBI(ip->i_sb)->ipbmap;
struct bmap *bmp = JFS_SBI(ip->i_sb)->bmap;
+ struct super_block *sb = ipbmap->i_sb;
IREAD_LOCK(ipbmap, RDWRLOCK_DMAP);
@@ -346,11 +346,17 @@ int dbFree(struct inode *ip, s64 blkno, s64 nblocks)
printk(KERN_ERR "blkno = %Lx, nblocks = %Lx\n",
(unsigned long long) blkno,
(unsigned long long) nblocks);
- jfs_error(ip->i_sb,
- "dbFree: block to be freed is outside the map");
+ jfs_error(ip->i_sb, "block to be freed is outside the map\n");
return -EIO;
}
+ /**
+ * TRIM the blocks, when mounted with discard option
+ */
+ if (JFS_SBI(sb)->flag & JFS_DISCARD)
+ if (JFS_SBI(sb)->minblks_trim <= nblocks)
+ jfs_issue_discard(ipbmap, blkno, nblocks);
+
/*
* free the blocks a dmap at a time.
*/
@@ -377,7 +383,7 @@ int dbFree(struct inode *ip, s64 blkno, s64 nblocks)
/* free the blocks. */
if ((rc = dbFreeDmap(bmp, dp, blkno, nb))) {
- jfs_error(ip->i_sb, "dbFree: error in block map\n");
+ jfs_error(ip->i_sb, "error in block map\n");
release_metapage(mp);
IREAD_UNLOCK(ipbmap);
return (rc);
@@ -434,8 +440,7 @@ dbUpdatePMap(struct inode *ipbmap,
printk(KERN_ERR "blkno = %Lx, nblocks = %Lx\n",
(unsigned long long) blkno,
(unsigned long long) nblocks);
- jfs_error(ipbmap->i_sb,
- "dbUpdatePMap: blocks are outside the map");
+ jfs_error(ipbmap->i_sb, "blocks are outside the map\n");
return -EIO;
}
@@ -719,7 +724,7 @@ int dbAlloc(struct inode *ip, s64 hint, s64 nblocks, s64 * results)
/* the hint should be within the map */
if (hint >= mapSize) {
- jfs_error(ip->i_sb, "dbAlloc: the hint is outside the map");
+ jfs_error(ip->i_sb, "the hint is outside the map\n");
return -EIO;
}
@@ -1050,8 +1055,7 @@ static int dbExtend(struct inode *ip, s64 blkno, s64 nblocks, s64 addnblocks)
bmp = sbi->bmap;
if (lastblkno < 0 || lastblkno >= bmp->db_mapsize) {
IREAD_UNLOCK(ipbmap);
- jfs_error(ip->i_sb,
- "dbExtend: the block is outside the filesystem");
+ jfs_error(ip->i_sb, "the block is outside the filesystem\n");
return -EIO;
}
@@ -1095,7 +1099,6 @@ static int dbExtend(struct inode *ip, s64 blkno, s64 nblocks, s64 addnblocks)
/* we were not successful */
release_metapage(mp);
-
return (rc);
}
@@ -1128,8 +1131,7 @@ static int dbAllocNext(struct bmap * bmp, struct dmap * dp, s64 blkno,
u32 mask;
if (dp->tree.leafidx != cpu_to_le32(LEAFIND)) {
- jfs_error(bmp->db_ipbmap->i_sb,
- "dbAllocNext: Corrupt dmap page");
+ jfs_error(bmp->db_ipbmap->i_sb, "Corrupt dmap page\n");
return -EIO;
}
@@ -1206,7 +1208,7 @@ static int dbAllocNext(struct bmap * bmp, struct dmap * dp, s64 blkno,
* by this leaf.
*/
l2size =
- min((int)leaf[word], NLSTOL2BSZ(nwords));
+ min_t(int, leaf[word], NLSTOL2BSZ(nwords));
/* determine how many words were handled.
*/
@@ -1259,8 +1261,7 @@ dbAllocNear(struct bmap * bmp,
s8 *leaf;
if (dp->tree.leafidx != cpu_to_le32(LEAFIND)) {
- jfs_error(bmp->db_ipbmap->i_sb,
- "dbAllocNear: Corrupt dmap page");
+ jfs_error(bmp->db_ipbmap->i_sb, "Corrupt dmap page\n");
return -EIO;
}
@@ -1375,8 +1376,7 @@ dbAllocAG(struct bmap * bmp, int agno, s64 nblocks, int l2nb, s64 * results)
*/
if (l2nb > bmp->db_agl2size) {
jfs_error(bmp->db_ipbmap->i_sb,
- "dbAllocAG: allocation request is larger than the "
- "allocation group size");
+ "allocation request is larger than the allocation group size\n");
return -EIO;
}
@@ -1411,7 +1411,7 @@ dbAllocAG(struct bmap * bmp, int agno, s64 nblocks, int l2nb, s64 * results)
(unsigned long long) blkno,
(unsigned long long) nblocks);
jfs_error(bmp->db_ipbmap->i_sb,
- "dbAllocAG: dbAllocCtl failed in free AG");
+ "dbAllocCtl failed in free AG\n");
}
return (rc);
}
@@ -1427,8 +1427,7 @@ dbAllocAG(struct bmap * bmp, int agno, s64 nblocks, int l2nb, s64 * results)
budmin = dcp->budmin;
if (dcp->leafidx != cpu_to_le32(CTLLEAFIND)) {
- jfs_error(bmp->db_ipbmap->i_sb,
- "dbAllocAG: Corrupt dmapctl page");
+ jfs_error(bmp->db_ipbmap->i_sb, "Corrupt dmapctl page\n");
release_metapage(mp);
return -EIO;
}
@@ -1469,7 +1468,7 @@ dbAllocAG(struct bmap * bmp, int agno, s64 nblocks, int l2nb, s64 * results)
}
if (n == 4) {
jfs_error(bmp->db_ipbmap->i_sb,
- "dbAllocAG: failed descending stree");
+ "failed descending stree\n");
release_metapage(mp);
return -EIO;
}
@@ -1509,8 +1508,7 @@ dbAllocAG(struct bmap * bmp, int agno, s64 nblocks, int l2nb, s64 * results)
&blkno))) {
if (rc == -ENOSPC) {
jfs_error(bmp->db_ipbmap->i_sb,
- "dbAllocAG: control page "
- "inconsistent");
+ "control page inconsistent\n");
return -EIO;
}
return (rc);
@@ -1522,7 +1520,7 @@ dbAllocAG(struct bmap * bmp, int agno, s64 nblocks, int l2nb, s64 * results)
rc = dbAllocCtl(bmp, nblocks, l2nb, blkno, results);
if (rc == -ENOSPC) {
jfs_error(bmp->db_ipbmap->i_sb,
- "dbAllocAG: unable to allocate blocks");
+ "unable to allocate blocks\n");
rc = -EIO;
}
return (rc);
@@ -1581,8 +1579,7 @@ static int dbAllocAny(struct bmap * bmp, s64 nblocks, int l2nb, s64 * results)
*/
rc = dbAllocCtl(bmp, nblocks, l2nb, blkno, results);
if (rc == -ENOSPC) {
- jfs_error(bmp->db_ipbmap->i_sb,
- "dbAllocAny: unable to allocate blocks");
+ jfs_error(bmp->db_ipbmap->i_sb, "unable to allocate blocks\n");
return -EIO;
}
return (rc);
@@ -1590,6 +1587,116 @@ static int dbAllocAny(struct bmap * bmp, s64 nblocks, int l2nb, s64 * results)
/*
+ * NAME: dbDiscardAG()
+ *
+ * FUNCTION: attempt to discard (TRIM) all free blocks of specific AG
+ *
+ * algorithm:
+ * 1) allocate blocks, as large as possible and save them
+ * while holding IWRITE_LOCK on ipbmap
+ * 2) trim all these saved block/length values
+ * 3) mark the blocks free again
+ *
+ * benefit:
+ * - we work only on one ag at some time, minimizing how long we
+ * need to lock ipbmap
+ * - reading / writing the fs is possible most time, even on
+ * trimming
+ *
+ * downside:
+ * - we write two times to the dmapctl and dmap pages
+ * - but for me, this seems the best way, better ideas?
+ * /TR 2012
+ *
+ * PARAMETERS:
+ * ip - pointer to in-core inode
+ * agno - ag to trim
+ * minlen - minimum value of contiguous blocks
+ *
+ * RETURN VALUES:
+ * s64 - actual number of blocks trimmed
+ */
+s64 dbDiscardAG(struct inode *ip, int agno, s64 minlen)
+{
+ struct inode *ipbmap = JFS_SBI(ip->i_sb)->ipbmap;
+ struct bmap *bmp = JFS_SBI(ip->i_sb)->bmap;
+ s64 nblocks, blkno;
+ u64 trimmed = 0;
+ int rc, l2nb;
+ struct super_block *sb = ipbmap->i_sb;
+
+ struct range2trim {
+ u64 blkno;
+ u64 nblocks;
+ } *totrim, *tt;
+
+ /* max blkno / nblocks pairs to trim */
+ int count = 0, range_cnt;
+ u64 max_ranges;
+
+ /* prevent others from writing new stuff here, while trimming */
+ IWRITE_LOCK(ipbmap, RDWRLOCK_DMAP);
+
+ nblocks = bmp->db_agfree[agno];
+ max_ranges = nblocks;
+ do_div(max_ranges, minlen);
+ range_cnt = min_t(u64, max_ranges + 1, 32 * 1024);
+ totrim = kmalloc(sizeof(struct range2trim) * range_cnt, GFP_NOFS);
+ if (totrim == NULL) {
+ jfs_error(bmp->db_ipbmap->i_sb, "no memory for trim array\n");
+ IWRITE_UNLOCK(ipbmap);
+ return 0;
+ }
+
+ tt = totrim;
+ while (nblocks >= minlen) {
+ l2nb = BLKSTOL2(nblocks);
+
+ /* 0 = okay, -EIO = fatal, -ENOSPC -> try smaller block */
+ rc = dbAllocAG(bmp, agno, nblocks, l2nb, &blkno);
+ if (rc == 0) {
+ tt->blkno = blkno;
+ tt->nblocks = nblocks;
+ tt++; count++;
+
+ /* the whole ag is free, trim now */
+ if (bmp->db_agfree[agno] == 0)
+ break;
+
+ /* give a hint for the next while */
+ nblocks = bmp->db_agfree[agno];
+ continue;
+ } else if (rc == -ENOSPC) {
+ /* search for next smaller log2 block */
+ l2nb = BLKSTOL2(nblocks) - 1;
+ nblocks = 1 << l2nb;
+ } else {
+ /* Trim any already allocated blocks */
+ jfs_error(bmp->db_ipbmap->i_sb, "-EIO\n");
+ break;
+ }
+
+ /* check, if our trim array is full */
+ if (unlikely(count >= range_cnt - 1))
+ break;
+ }
+ IWRITE_UNLOCK(ipbmap);
+
+ tt->nblocks = 0; /* mark the current end */
+ for (tt = totrim; tt->nblocks != 0; tt++) {
+ /* when mounted with online discard, dbFree() will
+ * call jfs_issue_discard() itself */
+ if (!(JFS_SBI(sb)->flag & JFS_DISCARD))
+ jfs_issue_discard(ip, tt->blkno, tt->nblocks);
+ dbFree(ip, tt->blkno, tt->nblocks);
+ trimmed += tt->nblocks;
+ }
+ kfree(totrim);
+
+ return trimmed;
+}
+
+/*
* NAME: dbFindCtl()
*
* FUNCTION: starting at a specified dmap control page level and block
@@ -1643,13 +1750,13 @@ static int dbFindCtl(struct bmap * bmp, int l2nb, int level, s64 * blkno)
if (dcp->leafidx != cpu_to_le32(CTLLEAFIND)) {
jfs_error(bmp->db_ipbmap->i_sb,
- "dbFindCtl: Corrupt dmapctl page");
+ "Corrupt dmapctl page\n");
release_metapage(mp);
return -EIO;
}
/* search the tree within the dmap control page for
- * sufficent free space. if sufficient free space is found,
+ * sufficient free space. if sufficient free space is found,
* dbFindLeaf() returns the index of the leaf at which
* free space was found.
*/
@@ -1664,7 +1771,7 @@ static int dbFindCtl(struct bmap * bmp, int l2nb, int level, s64 * blkno)
if (rc) {
if (lev != level) {
jfs_error(bmp->db_ipbmap->i_sb,
- "dbFindCtl: dmap inconsistent");
+ "dmap inconsistent\n");
return -EIO;
}
return -ENOSPC;
@@ -1788,14 +1895,14 @@ dbAllocCtl(struct bmap * bmp, s64 nblocks, int l2nb, s64 blkno, s64 * results)
if (dp->tree.stree[ROOT] != L2BPERDMAP) {
release_metapage(mp);
jfs_error(bmp->db_ipbmap->i_sb,
- "dbAllocCtl: the dmap is not all free");
+ "the dmap is not all free\n");
rc = -EIO;
goto backout;
}
/* determine how many blocks to allocate from this dmap.
*/
- nb = min(n, (s64)BPERDMAP);
+ nb = min_t(s64, n, BPERDMAP);
/* allocate the blocks from the dmap.
*/
@@ -1835,7 +1942,7 @@ dbAllocCtl(struct bmap * bmp, s64 nblocks, int l2nb, s64 blkno, s64 * results)
* to indicate that we have leaked blocks.
*/
jfs_error(bmp->db_ipbmap->i_sb,
- "dbAllocCtl: I/O Error: Block Leakage.");
+ "I/O Error: Block Leakage\n");
continue;
}
dp = (struct dmap *) mp->data;
@@ -1847,8 +1954,7 @@ dbAllocCtl(struct bmap * bmp, s64 nblocks, int l2nb, s64 blkno, s64 * results)
* to indicate that we have leaked blocks.
*/
release_metapage(mp);
- jfs_error(bmp->db_ipbmap->i_sb,
- "dbAllocCtl: Block Leakage.");
+ jfs_error(bmp->db_ipbmap->i_sb, "Block Leakage\n");
continue;
}
@@ -2145,8 +2251,7 @@ static void dbAllocBits(struct bmap * bmp, struct dmap * dp, s64 blkno,
for (; nwords > 0; nwords -= nw) {
if (leaf[word] < BUDMIN) {
jfs_error(bmp->db_ipbmap->i_sb,
- "dbAllocBits: leaf page "
- "corrupt");
+ "leaf page corrupt\n");
break;
}
@@ -2155,7 +2260,8 @@ static void dbAllocBits(struct bmap * bmp, struct dmap * dp, s64 blkno,
* of bits being allocated and the l2 number
* of bits currently described by this leaf.
*/
- size = min((int)leaf[word], NLSTOL2BSZ(nwords));
+ size = min_t(int, leaf[word],
+ NLSTOL2BSZ(nwords));
/* update the leaf to reflect the allocation.
* in addition to setting the leaf value to
@@ -2418,8 +2524,7 @@ dbAdjCtl(struct bmap * bmp, s64 blkno, int newval, int alloc, int level)
dcp = (struct dmapctl *) mp->data;
if (dcp->leafidx != cpu_to_le32(CTLLEAFIND)) {
- jfs_error(bmp->db_ipbmap->i_sb,
- "dbAdjCtl: Corrupt dmapctl page");
+ jfs_error(bmp->db_ipbmap->i_sb, "Corrupt dmapctl page\n");
release_metapage(mp);
return -EIO;
}
@@ -2520,8 +2625,7 @@ dbAdjCtl(struct bmap * bmp, s64 blkno, int newval, int alloc, int level)
assert(level == bmp->db_maxlevel);
if (bmp->db_maxfreebud != oldroot) {
jfs_error(bmp->db_ipbmap->i_sb,
- "dbAdjCtl: the maximum free buddy is "
- "not the old root");
+ "the maximum free buddy is not the old root\n");
}
bmp->db_maxfreebud = dcp->stree[ROOT];
}
@@ -2744,7 +2848,7 @@ static int dbJoin(dmtree_t * tp, int leafno, int newval)
/* check which (leafno or buddy) is the left buddy.
* the left buddy gets to claim the blocks resulting
* from the join while the right gets to claim none.
- * the left buddy is also eligable to participate in
+ * the left buddy is also eligible to participate in
* a join at the next higher level while the right
* is not.
*
@@ -3161,7 +3265,7 @@ static int dbAllocDmapBU(struct bmap * bmp, struct dmap * dp, s64 blkno,
{
int rc;
int dbitno, word, rembits, nb, nwords, wbitno, agno;
- s8 oldroot, *leaf;
+ s8 oldroot;
struct dmaptree *tp = (struct dmaptree *) & dp->tree;
/* save the current value of the root (i.e. maximum free string)
@@ -3169,9 +3273,6 @@ static int dbAllocDmapBU(struct bmap * bmp, struct dmap * dp, s64 blkno,
*/
oldroot = tp->stree[ROOT];
- /* pick up a pointer to the leaves of the dmap tree */
- leaf = tp->stree + LEAFIND;
-
/* determine the bit number and word within the dmap of the
* starting block.
*/
@@ -3366,7 +3467,7 @@ int dbExtendFS(struct inode *ipbmap, s64 blkno, s64 nblocks)
p = BMAPBLKNO + nbperpage; /* L2 page */
l2mp = read_metapage(ipbmap, p, PSIZE, 0);
if (!l2mp) {
- jfs_error(ipbmap->i_sb, "dbExtendFS: L2 page could not be read");
+ jfs_error(ipbmap->i_sb, "L2 page could not be read\n");
return -EIO;
}
l2dcp = (struct dmapctl *) l2mp->data;
@@ -3463,7 +3564,7 @@ int dbExtendFS(struct inode *ipbmap, s64 blkno, s64 nblocks)
if (mp == NULL)
goto errout;
- n = min(nblocks, (s64)BPERDMAP);
+ n = min_t(s64, nblocks, BPERDMAP);
}
dp = (struct dmap *) mp->data;
@@ -3531,8 +3632,7 @@ int dbExtendFS(struct inode *ipbmap, s64 blkno, s64 nblocks)
}
} /* for each L1 in a L2 */
- jfs_error(ipbmap->i_sb,
- "dbExtendFS: function has not returned as expected");
+ jfs_error(ipbmap->i_sb, "function has not returned as expected\n");
errout:
if (l0mp)
release_metapage(l0mp);
@@ -3602,7 +3702,7 @@ void dbFinalizeBmap(struct inode *ipbmap)
}
if (bmp->db_agpref >= bmp->db_numag) {
jfs_error(ipbmap->i_sb,
- "cannot find ag with average freespace");
+ "cannot find ag with average freespace\n");
}
}
diff --git a/fs/jfs/jfs_dmap.h b/fs/jfs/jfs_dmap.h
index 6dcb906c55d..562b9a7e431 100644
--- a/fs/jfs/jfs_dmap.h
+++ b/fs/jfs/jfs_dmap.h
@@ -311,4 +311,6 @@ extern int dbAllocBottomUp(struct inode *ip, s64 blkno, s64 nblocks);
extern int dbExtendFS(struct inode *ipbmap, s64 blkno, s64 nblocks);
extern void dbFinalizeBmap(struct inode *ipbmap);
extern s64 dbMapFileSizeToMapSize(struct inode *ipbmap);
+extern s64 dbDiscardAG(struct inode *ip, int agno, s64 minlen);
+
#endif /* _H_JFS_DMAP */
diff --git a/fs/jfs/jfs_dtree.c b/fs/jfs/jfs_dtree.c
index 9197a1b0d02..984c2bbf4f6 100644
--- a/fs/jfs/jfs_dtree.c
+++ b/fs/jfs/jfs_dtree.c
@@ -124,21 +124,21 @@ struct dtsplit {
#define DT_PAGE(IP, MP) BT_PAGE(IP, MP, dtpage_t, i_dtroot)
/* get page buffer for specified block address */
-#define DT_GETPAGE(IP, BN, MP, SIZE, P, RC)\
-{\
- BT_GETPAGE(IP, BN, MP, dtpage_t, SIZE, P, RC, i_dtroot)\
- if (!(RC))\
- {\
- if (((P)->header.nextindex > (((BN)==0)?DTROOTMAXSLOT:(P)->header.maxslot)) ||\
- ((BN) && ((P)->header.maxslot > DTPAGEMAXSLOT)))\
- {\
- BT_PUTPAGE(MP);\
- jfs_error((IP)->i_sb, "DT_GETPAGE: dtree page corrupt");\
- MP = NULL;\
- RC = -EIO;\
- }\
- }\
-}
+#define DT_GETPAGE(IP, BN, MP, SIZE, P, RC) \
+do { \
+ BT_GETPAGE(IP, BN, MP, dtpage_t, SIZE, P, RC, i_dtroot); \
+ if (!(RC)) { \
+ if (((P)->header.nextindex > \
+ (((BN) == 0) ? DTROOTMAXSLOT : (P)->header.maxslot)) || \
+ ((BN) && ((P)->header.maxslot > DTPAGEMAXSLOT))) { \
+ BT_PUTPAGE(MP); \
+ jfs_error((IP)->i_sb, \
+ "DT_GETPAGE: dtree page corrupt\n"); \
+ MP = NULL; \
+ RC = -EIO; \
+ } \
+ } \
+} while (0)
/* for consistency */
#define DT_PUTPAGE(MP) BT_PUTPAGE(MP)
@@ -776,7 +776,7 @@ int dtSearch(struct inode *ip, struct component_name * key, ino_t * data,
/* Something's corrupted, mark filesystem dirty so
* chkdsk will fix it.
*/
- jfs_error(sb, "stack overrun in dtSearch!");
+ jfs_error(sb, "stack overrun!\n");
BT_STACK_DUMP(btstack);
rc = -EIO;
goto out;
@@ -3002,9 +3002,9 @@ static inline struct jfs_dirent *next_jfs_dirent(struct jfs_dirent *dirent)
* return: offset = (pn, index) of start entry
* of next jfs_readdir()/dtRead()
*/
-int jfs_readdir(struct file *filp, void *dirent, filldir_t filldir)
+int jfs_readdir(struct file *file, struct dir_context *ctx)
{
- struct inode *ip = filp->f_path.dentry->d_inode;
+ struct inode *ip = file_inode(file);
struct nls_table *codepage = JFS_SBI(ip->i_sb)->nls_tab;
int rc = 0;
loff_t dtpos; /* legacy OS/2 style position */
@@ -3033,7 +3033,7 @@ int jfs_readdir(struct file *filp, void *dirent, filldir_t filldir)
int overflow, fix_page, page_fixed = 0;
static int unique_pos = 2; /* If we can't fix broken index */
- if (filp->f_pos == DIREND)
+ if (ctx->pos == DIREND)
return 0;
if (DO_INDEX(ip)) {
@@ -3045,7 +3045,15 @@ int jfs_readdir(struct file *filp, void *dirent, filldir_t filldir)
*/
do_index = 1;
- dir_index = (u32) filp->f_pos;
+ dir_index = (u32) ctx->pos;
+
+ /*
+ * NFSv4 reserves cookies 1 and 2 for . and .. so the value
+ * we return to the vfs is one greater than the one we use
+ * internally.
+ */
+ if (dir_index)
+ dir_index--;
if (dir_index > 1) {
struct dir_table_slot dirtab_slot;
@@ -3053,25 +3061,25 @@ int jfs_readdir(struct file *filp, void *dirent, filldir_t filldir)
if (dtEmpty(ip) ||
(dir_index >= JFS_IP(ip)->next_index)) {
/* Stale position. Directory has shrunk */
- filp->f_pos = DIREND;
+ ctx->pos = DIREND;
return 0;
}
repeat:
rc = read_index(ip, dir_index, &dirtab_slot);
if (rc) {
- filp->f_pos = DIREND;
+ ctx->pos = DIREND;
return rc;
}
if (dirtab_slot.flag == DIR_INDEX_FREE) {
if (loop_count++ > JFS_IP(ip)->next_index) {
jfs_err("jfs_readdir detected "
"infinite loop!");
- filp->f_pos = DIREND;
+ ctx->pos = DIREND;
return 0;
}
dir_index = le32_to_cpu(dirtab_slot.addr2);
if (dir_index == -1) {
- filp->f_pos = DIREND;
+ ctx->pos = DIREND;
return 0;
}
goto repeat;
@@ -3080,13 +3088,13 @@ int jfs_readdir(struct file *filp, void *dirent, filldir_t filldir)
index = dirtab_slot.slot;
DT_GETPAGE(ip, bn, mp, PSIZE, p, rc);
if (rc) {
- filp->f_pos = DIREND;
+ ctx->pos = DIREND;
return 0;
}
if (p->header.flag & BT_INTERNAL) {
jfs_err("jfs_readdir: bad index table");
DT_PUTPAGE(mp);
- filp->f_pos = -1;
+ ctx->pos = DIREND;
return 0;
}
} else {
@@ -3094,23 +3102,22 @@ int jfs_readdir(struct file *filp, void *dirent, filldir_t filldir)
/*
* self "."
*/
- filp->f_pos = 0;
- if (filldir(dirent, ".", 1, 0, ip->i_ino,
- DT_DIR))
+ ctx->pos = 1;
+ if (!dir_emit(ctx, ".", 1, ip->i_ino, DT_DIR))
return 0;
}
/*
* parent ".."
*/
- filp->f_pos = 1;
- if (filldir(dirent, "..", 2, 1, PARENT(ip), DT_DIR))
+ ctx->pos = 2;
+ if (!dir_emit(ctx, "..", 2, PARENT(ip), DT_DIR))
return 0;
/*
* Find first entry of left-most leaf
*/
if (dtEmpty(ip)) {
- filp->f_pos = DIREND;
+ ctx->pos = DIREND;
return 0;
}
@@ -3123,28 +3130,25 @@ int jfs_readdir(struct file *filp, void *dirent, filldir_t filldir)
/*
* Legacy filesystem - OS/2 & Linux JFS < 0.3.6
*
- * pn = index = 0: First entry "."
- * pn = 0; index = 1: Second entry ".."
+ * pn = 0; index = 1: First entry "."
+ * pn = 0; index = 2: Second entry ".."
* pn > 0: Real entries, pn=1 -> leftmost page
* pn = index = -1: No more entries
*/
- dtpos = filp->f_pos;
- if (dtpos == 0) {
+ dtpos = ctx->pos;
+ if (dtpos < 2) {
/* build "." entry */
-
- if (filldir(dirent, ".", 1, filp->f_pos, ip->i_ino,
- DT_DIR))
+ ctx->pos = 1;
+ if (!dir_emit(ctx, ".", 1, ip->i_ino, DT_DIR))
return 0;
- dtoffset->index = 1;
- filp->f_pos = dtpos;
+ dtoffset->index = 2;
+ ctx->pos = dtpos;
}
if (dtoffset->pn == 0) {
- if (dtoffset->index == 1) {
+ if (dtoffset->index == 2) {
/* build ".." entry */
-
- if (filldir(dirent, "..", 2, filp->f_pos,
- PARENT(ip), DT_DIR))
+ if (!dir_emit(ctx, "..", 2, PARENT(ip), DT_DIR))
return 0;
} else {
jfs_err("jfs_readdir called with "
@@ -3152,18 +3156,18 @@ int jfs_readdir(struct file *filp, void *dirent, filldir_t filldir)
}
dtoffset->pn = 1;
dtoffset->index = 0;
- filp->f_pos = dtpos;
+ ctx->pos = dtpos;
}
if (dtEmpty(ip)) {
- filp->f_pos = DIREND;
+ ctx->pos = DIREND;
return 0;
}
- if ((rc = dtReadNext(ip, &filp->f_pos, &btstack))) {
+ if ((rc = dtReadNext(ip, &ctx->pos, &btstack))) {
jfs_err("jfs_readdir: unexpected rc = %d "
"from dtReadNext", rc);
- filp->f_pos = DIREND;
+ ctx->pos = DIREND;
return 0;
}
/* get start leaf page and index */
@@ -3171,7 +3175,7 @@ int jfs_readdir(struct file *filp, void *dirent, filldir_t filldir)
/* offset beyond directory eof ? */
if (bn < 0) {
- filp->f_pos = DIREND;
+ ctx->pos = DIREND;
return 0;
}
}
@@ -3180,7 +3184,7 @@ int jfs_readdir(struct file *filp, void *dirent, filldir_t filldir)
if (dirent_buf == 0) {
DT_PUTPAGE(mp);
jfs_warn("jfs_readdir: __get_free_page failed!");
- filp->f_pos = DIREND;
+ ctx->pos = DIREND;
return -ENOMEM;
}
@@ -3233,6 +3237,12 @@ int jfs_readdir(struct file *filp, void *dirent, filldir_t filldir)
}
jfs_dirent->position = unique_pos++;
}
+ /*
+ * We add 1 to the index because we may
+ * use a value of 2 internally, and NFSv4
+ * doesn't like that.
+ */
+ jfs_dirent->position++;
} else {
jfs_dirent->position = dtpos;
len = min(d_namleft, DTLHDRDATALEN_LEGACY);
@@ -3252,8 +3262,7 @@ int jfs_readdir(struct file *filp, void *dirent, filldir_t filldir)
/* Sanity Check */
if (d_namleft == 0) {
jfs_error(ip->i_sb,
- "JFS:Dtree error: ino = "
- "%ld, bn=%Ld, index = %d",
+ "JFS:Dtree error: ino = %ld, bn=%lld, index = %d\n",
(long)ip->i_ino,
(long long)bn,
i);
@@ -3295,9 +3304,9 @@ skip_one:
jfs_dirent = (struct jfs_dirent *) dirent_buf;
while (jfs_dirents--) {
- filp->f_pos = jfs_dirent->position;
- if (filldir(dirent, jfs_dirent->name,
- jfs_dirent->name_len, filp->f_pos,
+ ctx->pos = jfs_dirent->position;
+ if (!dir_emit(ctx, jfs_dirent->name,
+ jfs_dirent->name_len,
jfs_dirent->ino, DT_UNKNOWN))
goto out;
jfs_dirent = next_jfs_dirent(jfs_dirent);
@@ -3309,7 +3318,7 @@ skip_one:
}
if (!overflow && (bn == 0)) {
- filp->f_pos = DIREND;
+ ctx->pos = DIREND;
break;
}
@@ -3373,7 +3382,7 @@ static int dtReadFirst(struct inode *ip, struct btstack * btstack)
*/
if (BT_STACK_FULL(btstack)) {
DT_PUTPAGE(mp);
- jfs_error(ip->i_sb, "dtReadFirst: btstack overrun");
+ jfs_error(ip->i_sb, "btstack overrun\n");
BT_STACK_DUMP(btstack);
return -EIO;
}
diff --git a/fs/jfs/jfs_dtree.h b/fs/jfs/jfs_dtree.h
index 2545bb31723..fd4169e6e69 100644
--- a/fs/jfs/jfs_dtree.h
+++ b/fs/jfs/jfs_dtree.h
@@ -265,5 +265,5 @@ extern int dtDelete(tid_t tid, struct inode *ip, struct component_name * key,
extern int dtModify(tid_t tid, struct inode *ip, struct component_name * key,
ino_t * orig_ino, ino_t new_ino, int flag);
-extern int jfs_readdir(struct file *filp, void *dirent, filldir_t filldir);
+extern int jfs_readdir(struct file *file, struct dir_context *ctx);
#endif /* !_H_JFS_DTREE */
diff --git a/fs/jfs/jfs_extent.c b/fs/jfs/jfs_extent.c
index 5d3bbd10f8d..2ae7d59ab10 100644
--- a/fs/jfs/jfs_extent.c
+++ b/fs/jfs/jfs_extent.c
@@ -126,7 +126,7 @@ extAlloc(struct inode *ip, s64 xlen, s64 pno, xad_t * xp, bool abnr)
/* allocate the disk blocks for the extent. initially, extBalloc()
* will try to allocate disk blocks for the requested size (xlen).
- * if this fails (xlen contiguous free blocks not avaliable), it'll
+ * if this fails (xlen contiguous free blocks not available), it'll
* try to allocate a smaller number of blocks (producing a smaller
* extent), with this smaller number of blocks consisting of the
* requested number of blocks rounded down to the next smaller
@@ -388,7 +388,7 @@ int extHint(struct inode *ip, s64 offset, xad_t * xp)
if ((rc == 0) && xlen) {
if (xlen != nbperpage) {
- jfs_error(ip->i_sb, "extHint: corrupt xtree");
+ jfs_error(ip->i_sb, "corrupt xtree\n");
rc = -EIO;
}
XADaddress(xp, xaddr);
@@ -481,7 +481,7 @@ int extFill(struct inode *ip, xad_t * xp)
*
* initially, we will try to allocate disk blocks for the
* requested size (nblocks). if this fails (nblocks
- * contiguous free blocks not avaliable), we'll try to allocate
+ * contiguous free blocks not available), we'll try to allocate
* a smaller number of blocks (producing a smaller extent), with
* this smaller number of blocks consisting of the requested
* number of blocks rounded down to the next smaller power of 2
@@ -575,7 +575,7 @@ extBalloc(struct inode *ip, s64 hint, s64 * nblocks, s64 * blkno)
* to a new set of blocks. If moving the extent, we initially
* will try to allocate disk blocks for the requested size
* (newnblks). if this fails (new contiguous free blocks not
- * avaliable), we'll try to allocate a smaller number of
+ * available), we'll try to allocate a smaller number of
* blocks (producing a smaller extent), with this smaller
* number of blocks consisting of the requested number of
* blocks rounded down to the next smaller power of 2
diff --git a/fs/jfs/jfs_filsys.h b/fs/jfs/jfs_filsys.h
index b3f5463fbe5..b67d64671bb 100644
--- a/fs/jfs/jfs_filsys.h
+++ b/fs/jfs/jfs_filsys.h
@@ -45,6 +45,9 @@
/* mount time flag to disable journaling to disk */
#define JFS_NOINTEGRITY 0x00000040
+/* mount time flag to enable TRIM to ssd disks */
+#define JFS_DISCARD 0x00000080
+
/* commit option */
#define JFS_COMMIT 0x00000f00 /* commit option mask */
#define JFS_GROUPCOMMIT 0x00000100 /* group (of 1) commit */
diff --git a/fs/jfs/jfs_imap.c b/fs/jfs/jfs_imap.c
index 3a09423b6c2..f321986e73d 100644
--- a/fs/jfs/jfs_imap.c
+++ b/fs/jfs/jfs_imap.c
@@ -386,7 +386,7 @@ int diRead(struct inode *ip)
dp += rel_inode;
if (ip->i_ino != le32_to_cpu(dp->di_number)) {
- jfs_error(ip->i_sb, "diRead: i_ino != di_number");
+ jfs_error(ip->i_sb, "i_ino != di_number\n");
rc = -EIO;
} else if (le32_to_cpu(dp->di_nlink) == 0)
rc = -ESTALE;
@@ -397,7 +397,7 @@ int diRead(struct inode *ip)
release_metapage(mp);
/* set the ag for the inode */
- JFS_IP(ip)->agno = BLKTOAG(agstart, sbi);
+ JFS_IP(ip)->agstart = agstart;
JFS_IP(ip)->active_ag = -1;
return (rc);
@@ -457,7 +457,7 @@ struct inode *diReadSpecial(struct super_block *sb, ino_t inum, int secondary)
/* read the page of fixed disk inode (AIT) in raw mode */
mp = read_metapage(ip, address << sbi->l2nbperpage, PSIZE, 1);
if (mp == NULL) {
- ip->i_nlink = 1; /* Don't want iput() deleting it */
+ set_nlink(ip, 1); /* Don't want iput() deleting it */
iput(ip);
return (NULL);
}
@@ -469,7 +469,7 @@ struct inode *diReadSpecial(struct super_block *sb, ino_t inum, int secondary)
/* copy on-disk inode to in-memory inode */
if ((copy_from_dinode(dp, ip)) != 0) {
/* handle bad return by returning NULL for ip */
- ip->i_nlink = 1; /* Don't want iput() deleting it */
+ set_nlink(ip, 1); /* Don't want iput() deleting it */
iput(ip);
/* release the page */
release_metapage(mp);
@@ -625,7 +625,7 @@ int diWrite(tid_t tid, struct inode *ip)
if (!addressPXD(&(jfs_ip->ixpxd)) ||
(lengthPXD(&(jfs_ip->ixpxd)) !=
JFS_IP(ipimap)->i_imap->im_nbperiext)) {
- jfs_error(ip->i_sb, "diWrite: ixpxd invalid");
+ jfs_error(ip->i_sb, "ixpxd invalid\n");
return -EIO;
}
@@ -893,15 +893,14 @@ int diFree(struct inode *ip)
if (iagno >= imap->im_nextiag) {
print_hex_dump(KERN_ERR, "imap: ", DUMP_PREFIX_ADDRESS, 16, 4,
imap, 32, 0);
- jfs_error(ip->i_sb,
- "diFree: inum = %d, iagno = %d, nextiag = %d",
+ jfs_error(ip->i_sb, "inum = %d, iagno = %d, nextiag = %d\n",
(uint) inum, iagno, imap->im_nextiag);
return -EIO;
}
/* get the allocation group for this ino.
*/
- agno = JFS_IP(ip)->agno;
+ agno = BLKTOAG(JFS_IP(ip)->agstart, JFS_SBI(ip->i_sb));
/* Lock the AG specific inode map information
*/
@@ -930,15 +929,14 @@ int diFree(struct inode *ip)
mask = HIGHORDER >> bitno;
if (!(le32_to_cpu(iagp->wmap[extno]) & mask)) {
- jfs_error(ip->i_sb,
- "diFree: wmap shows inode already free");
+ jfs_error(ip->i_sb, "wmap shows inode already free\n");
}
if (!addressPXD(&iagp->inoext[extno])) {
release_metapage(mp);
IREAD_UNLOCK(ipimap);
AG_UNLOCK(imap, agno);
- jfs_error(ip->i_sb, "diFree: invalid inoext");
+ jfs_error(ip->i_sb, "invalid inoext\n");
return -EIO;
}
@@ -950,7 +948,7 @@ int diFree(struct inode *ip)
release_metapage(mp);
IREAD_UNLOCK(ipimap);
AG_UNLOCK(imap, agno);
- jfs_error(ip->i_sb, "diFree: numfree > numinos");
+ jfs_error(ip->i_sb, "numfree > numinos\n");
return -EIO;
}
/*
@@ -1069,7 +1067,7 @@ int diFree(struct inode *ip)
*/
if (iagp->nfreeexts == cpu_to_le32(EXTSPERIAG - 1)) {
/* in preparation for removing the iag from the
- * ag extent free list, read the iags preceeding
+ * ag extent free list, read the iags preceding
* and following the iag on the ag extent free
* list.
*/
@@ -1095,7 +1093,7 @@ int diFree(struct inode *ip)
int inofreefwd = le32_to_cpu(iagp->inofreefwd);
/* in preparation for removing the iag from the
- * ag inode free list, read the iags preceeding
+ * ag inode free list, read the iags preceding
* and following the iag on the ag inode free
* list. before reading these iags, we must make
* sure that we already don't have them in hand
@@ -1199,7 +1197,7 @@ int diFree(struct inode *ip)
* for the inode being freed.
*/
if (iagp->pmap[extno] != 0) {
- jfs_error(ip->i_sb, "diFree: the pmap does not show inode free");
+ jfs_error(ip->i_sb, "the pmap does not show inode free\n");
}
iagp->wmap[extno] = 0;
PXDlength(&iagp->inoext[extno], 0);
@@ -1315,12 +1313,11 @@ int diFree(struct inode *ip)
static inline void
diInitInode(struct inode *ip, int iagno, int ino, int extno, struct iag * iagp)
{
- struct jfs_sb_info *sbi = JFS_SBI(ip->i_sb);
struct jfs_inode_info *jfs_ip = JFS_IP(ip);
ip->i_ino = (iagno << L2INOSPERIAG) + ino;
jfs_ip->ixpxd = iagp->inoext[extno];
- jfs_ip->agno = BLKTOAG(le64_to_cpu(iagp->agstart), sbi);
+ jfs_ip->agstart = le64_to_cpu(iagp->agstart);
jfs_ip->active_ag = -1;
}
@@ -1379,7 +1376,7 @@ int diAlloc(struct inode *pip, bool dir, struct inode *ip)
*/
/* get the ag number of this iag */
- agno = JFS_IP(pip)->agno;
+ agno = BLKTOAG(JFS_IP(pip)->agstart, JFS_SBI(pip->i_sb));
if (atomic_read(&JFS_SBI(pip->i_sb)->bmap->db_active[agno])) {
/*
@@ -1494,7 +1491,7 @@ int diAlloc(struct inode *pip, bool dir, struct inode *ip)
/* mask any prior bits for the starting words of the
* summary map.
*/
- mask = ONES << (EXTSPERSUM - bitno);
+ mask = (bitno == 0) ? 0 : (ONES << (EXTSPERSUM - bitno));
inosmap = le32_to_cpu(iagp->inosmap[sword]) | mask;
extsmap = le32_to_cpu(iagp->extsmap[sword]) | mask;
@@ -1519,8 +1516,7 @@ int diAlloc(struct inode *pip, bool dir, struct inode *ip)
release_metapage(mp);
AG_UNLOCK(imap, agno);
jfs_error(ip->i_sb,
- "diAlloc: can't find free bit "
- "in wmap");
+ "can't find free bit in wmap\n");
return -EIO;
}
@@ -1661,7 +1657,7 @@ diAllocAG(struct inomap * imap, int agno, bool dir, struct inode *ip)
numinos = imap->im_agctl[agno].numinos;
if (numfree > numinos) {
- jfs_error(ip->i_sb, "diAllocAG: numfree > numinos");
+ jfs_error(ip->i_sb, "numfree > numinos\n");
return -EIO;
}
@@ -1681,7 +1677,7 @@ diAllocAG(struct inomap * imap, int agno, bool dir, struct inode *ip)
* try to allocate a new extent of free inodes.
*/
if (addext) {
- /* if free space is not avaliable for this new extent, try
+ /* if free space is not available for this new extent, try
* below to allocate a free and existing (already backed)
* inode from the ag.
*/
@@ -1812,8 +1808,7 @@ static int diAllocIno(struct inomap * imap, int agno, struct inode *ip)
if (!iagp->nfreeinos) {
IREAD_UNLOCK(imap->im_ipimap);
release_metapage(mp);
- jfs_error(ip->i_sb,
- "diAllocIno: nfreeinos = 0, but iag on freelist");
+ jfs_error(ip->i_sb, "nfreeinos = 0, but iag on freelist\n");
return -EIO;
}
@@ -1825,7 +1820,7 @@ static int diAllocIno(struct inomap * imap, int agno, struct inode *ip)
IREAD_UNLOCK(imap->im_ipimap);
release_metapage(mp);
jfs_error(ip->i_sb,
- "diAllocIno: free inode not found in summary map");
+ "free inode not found in summary map\n");
return -EIO;
}
@@ -1840,7 +1835,7 @@ static int diAllocIno(struct inomap * imap, int agno, struct inode *ip)
if (rem >= EXTSPERSUM) {
IREAD_UNLOCK(imap->im_ipimap);
release_metapage(mp);
- jfs_error(ip->i_sb, "diAllocIno: no free extent found");
+ jfs_error(ip->i_sb, "no free extent found\n");
return -EIO;
}
extno = (sword << L2EXTSPERSUM) + rem;
@@ -1851,7 +1846,7 @@ static int diAllocIno(struct inomap * imap, int agno, struct inode *ip)
if (rem >= INOSPEREXT) {
IREAD_UNLOCK(imap->im_ipimap);
release_metapage(mp);
- jfs_error(ip->i_sb, "diAllocIno: free inode not found");
+ jfs_error(ip->i_sb, "free inode not found\n");
return -EIO;
}
@@ -1937,7 +1932,7 @@ static int diAllocExt(struct inomap * imap, int agno, struct inode *ip)
IREAD_LOCK(imap->im_ipimap, RDWRLOCK_IMAP);
if ((rc = diIAGRead(imap, iagno, &mp))) {
IREAD_UNLOCK(imap->im_ipimap);
- jfs_error(ip->i_sb, "diAllocExt: error reading iag");
+ jfs_error(ip->i_sb, "error reading iag\n");
return rc;
}
iagp = (struct iag *) mp->data;
@@ -1949,8 +1944,7 @@ static int diAllocExt(struct inomap * imap, int agno, struct inode *ip)
if (sword >= SMAPSZ) {
release_metapage(mp);
IREAD_UNLOCK(imap->im_ipimap);
- jfs_error(ip->i_sb,
- "diAllocExt: free ext summary map not found");
+ jfs_error(ip->i_sb, "free ext summary map not found\n");
return -EIO;
}
if (~iagp->extsmap[sword])
@@ -1963,7 +1957,7 @@ static int diAllocExt(struct inomap * imap, int agno, struct inode *ip)
if (rem >= EXTSPERSUM) {
release_metapage(mp);
IREAD_UNLOCK(imap->im_ipimap);
- jfs_error(ip->i_sb, "diAllocExt: free extent not found");
+ jfs_error(ip->i_sb, "free extent not found\n");
return -EIO;
}
extno = (sword << L2EXTSPERSUM) + rem;
@@ -2036,7 +2030,7 @@ static int diAllocBit(struct inomap * imap, struct iag * iagp, int ino)
/* check if this is the last free inode within the iag.
* if so, it will have to be removed from the ag free
- * inode list, so get the iags preceeding and following
+ * inode list, so get the iags preceding and following
* it on the list.
*/
if (iagp->nfreeinos == cpu_to_le32(1)) {
@@ -2082,8 +2076,7 @@ static int diAllocBit(struct inomap * imap, struct iag * iagp, int ino)
if (bmp)
release_metapage(bmp);
- jfs_error(imap->im_ipimap->i_sb,
- "diAllocBit: iag inconsistent");
+ jfs_error(imap->im_ipimap->i_sb, "iag inconsistent\n");
return -EIO;
}
@@ -2190,7 +2183,7 @@ static int diNewExt(struct inomap * imap, struct iag * iagp, int extno)
/* better have free extents.
*/
if (!iagp->nfreeexts) {
- jfs_error(imap->im_ipimap->i_sb, "diNewExt: no free extents");
+ jfs_error(imap->im_ipimap->i_sb, "no free extents\n");
return -EIO;
}
@@ -2208,7 +2201,7 @@ static int diNewExt(struct inomap * imap, struct iag * iagp, int extno)
/* check if this is the last free extent within the
* iag. if so, the iag must be removed from the ag
- * free extent list, so get the iags preceeding and
+ * free extent list, so get the iags preceding and
* following the iag on this list.
*/
if (iagp->nfreeexts == cpu_to_le32(1)) {
@@ -2262,7 +2255,7 @@ static int diNewExt(struct inomap * imap, struct iag * iagp, int extno)
}
if (ciagp == NULL) {
jfs_error(imap->im_ipimap->i_sb,
- "diNewExt: ciagp == NULL");
+ "ciagp == NULL\n");
rc = -EIO;
goto error_out;
}
@@ -2499,12 +2492,12 @@ diNewIAG(struct inomap * imap, int *iagnop, int agno, struct metapage ** mpp)
IWRITE_UNLOCK(ipimap);
IAGFREE_UNLOCK(imap);
jfs_error(imap->im_ipimap->i_sb,
- "diNewIAG: ipimap->i_size is wrong");
+ "ipimap->i_size is wrong\n");
return -EIO;
}
- /* get the next avaliable iag number */
+ /* get the next available iag number */
iagno = imap->im_nextiag;
/* make sure that we have not exceeded the maximum inode
@@ -2615,7 +2608,7 @@ diNewIAG(struct inomap * imap, int *iagnop, int agno, struct metapage ** mpp)
duplicateIXtree(sb, blkno, xlen, &xaddr);
- /* update the next avaliable iag number */
+ /* update the next available iag number */
imap->im_nextiag += 1;
/* Add the iag to the iag free list so we don't lose the iag
@@ -2759,8 +2752,7 @@ diUpdatePMap(struct inode *ipimap,
iagno = INOTOIAG(inum);
/* make sure that the iag is contained within the map */
if (iagno >= imap->im_nextiag) {
- jfs_error(ipimap->i_sb,
- "diUpdatePMap: the iag is outside the map");
+ jfs_error(ipimap->i_sb, "the iag is outside the map\n");
return -EIO;
}
/* read the iag */
@@ -2789,13 +2781,13 @@ diUpdatePMap(struct inode *ipimap,
*/
if (!(le32_to_cpu(iagp->wmap[extno]) & mask)) {
jfs_error(ipimap->i_sb,
- "diUpdatePMap: inode %ld not marked as "
- "allocated in wmap!", inum);
+ "inode %ld not marked as allocated in wmap!\n",
+ inum);
}
if (!(le32_to_cpu(iagp->pmap[extno]) & mask)) {
jfs_error(ipimap->i_sb,
- "diUpdatePMap: inode %ld not marked as "
- "allocated in pmap!", inum);
+ "inode %ld not marked as allocated in pmap!\n",
+ inum);
}
/* update the bitmap for the extent of the freed inode */
iagp->pmap[extno] &= cpu_to_le32(~mask);
@@ -2810,15 +2802,13 @@ diUpdatePMap(struct inode *ipimap,
if (!(le32_to_cpu(iagp->wmap[extno]) & mask)) {
release_metapage(mp);
jfs_error(ipimap->i_sb,
- "diUpdatePMap: the inode is not allocated in "
- "the working map");
+ "the inode is not allocated in the working map\n");
return -EIO;
}
if ((le32_to_cpu(iagp->pmap[extno]) & mask) != 0) {
release_metapage(mp);
jfs_error(ipimap->i_sb,
- "diUpdatePMap: the inode is not free in the "
- "persistent map");
+ "the inode is not free in the persistent map\n");
return -EIO;
}
/* update the bitmap for the extent of the allocated inode */
@@ -2910,8 +2900,7 @@ int diExtendFS(struct inode *ipimap, struct inode *ipbmap)
iagp = (struct iag *) bp->data;
if (le32_to_cpu(iagp->iagnum) != i) {
release_metapage(bp);
- jfs_error(ipimap->i_sb,
- "diExtendFs: unexpected value of iagnum");
+ jfs_error(ipimap->i_sb, "unexpected value of iagnum\n");
return -EIO;
}
@@ -2921,10 +2910,9 @@ int diExtendFS(struct inode *ipimap, struct inode *ipbmap)
continue;
}
- /* agstart that computes to the same ag is treated as same; */
agstart = le64_to_cpu(iagp->agstart);
- /* iagp->agstart = agstart & ~(mp->db_agsize - 1); */
n = agstart >> mp->db_agl2size;
+ iagp->agstart = cpu_to_le64((s64)n << mp->db_agl2size);
/* compute backed inodes */
numinos = (EXTSPERIAG - le32_to_cpu(iagp->nfreeexts))
@@ -2988,8 +2976,7 @@ int diExtendFS(struct inode *ipimap, struct inode *ipbmap)
if (xnuminos != atomic_read(&imap->im_numinos) ||
xnumfree != atomic_read(&imap->im_numfree)) {
- jfs_error(ipimap->i_sb,
- "diExtendFs: numinos or numfree incorrect");
+ jfs_error(ipimap->i_sb, "numinos or numfree incorrect\n");
return -EIO;
}
@@ -3078,17 +3065,17 @@ static int copy_from_dinode(struct dinode * dip, struct inode *ip)
ip->i_mode |= 0001;
}
}
- ip->i_nlink = le32_to_cpu(dip->di_nlink);
+ set_nlink(ip, le32_to_cpu(dip->di_nlink));
- jfs_ip->saved_uid = le32_to_cpu(dip->di_uid);
- if (sbi->uid == -1)
+ jfs_ip->saved_uid = make_kuid(&init_user_ns, le32_to_cpu(dip->di_uid));
+ if (!uid_valid(sbi->uid))
ip->i_uid = jfs_ip->saved_uid;
else {
ip->i_uid = sbi->uid;
}
- jfs_ip->saved_gid = le32_to_cpu(dip->di_gid);
- if (sbi->gid == -1)
+ jfs_ip->saved_gid = make_kgid(&init_user_ns, le32_to_cpu(dip->di_gid));
+ if (!gid_valid(sbi->gid))
ip->i_gid = jfs_ip->saved_gid;
else {
ip->i_gid = sbi->gid;
@@ -3152,14 +3139,16 @@ static void copy_to_dinode(struct dinode * dip, struct inode *ip)
dip->di_size = cpu_to_le64(ip->i_size);
dip->di_nblocks = cpu_to_le64(PBLK2LBLK(ip->i_sb, ip->i_blocks));
dip->di_nlink = cpu_to_le32(ip->i_nlink);
- if (sbi->uid == -1)
- dip->di_uid = cpu_to_le32(ip->i_uid);
+ if (!uid_valid(sbi->uid))
+ dip->di_uid = cpu_to_le32(i_uid_read(ip));
else
- dip->di_uid = cpu_to_le32(jfs_ip->saved_uid);
- if (sbi->gid == -1)
- dip->di_gid = cpu_to_le32(ip->i_gid);
+ dip->di_uid =cpu_to_le32(from_kuid(&init_user_ns,
+ jfs_ip->saved_uid));
+ if (!gid_valid(sbi->gid))
+ dip->di_gid = cpu_to_le32(i_gid_read(ip));
else
- dip->di_gid = cpu_to_le32(jfs_ip->saved_gid);
+ dip->di_gid = cpu_to_le32(from_kgid(&init_user_ns,
+ jfs_ip->saved_gid));
jfs_get_inode_flags(jfs_ip);
/*
* mode2 is only needed for storing the higher order bits.
diff --git a/fs/jfs/jfs_incore.h b/fs/jfs/jfs_incore.h
index 1439f119ec8..cf47f09e8ac 100644
--- a/fs/jfs/jfs_incore.h
+++ b/fs/jfs/jfs_incore.h
@@ -38,8 +38,8 @@
struct jfs_inode_info {
int fileset; /* fileset number (always 16)*/
uint mode2; /* jfs-specific mode */
- uint saved_uid; /* saved for uid mount option */
- uint saved_gid; /* saved for gid mount option */
+ kuid_t saved_uid; /* saved for uid mount option */
+ kgid_t saved_gid; /* saved for gid mount option */
pxd_t ixpxd; /* inode extent descriptor */
dxd_t acl; /* dxd describing acl */
dxd_t ea; /* dxd describing ea */
@@ -50,8 +50,9 @@ struct jfs_inode_info {
short btindex; /* btpage entry index*/
struct inode *ipimap; /* inode map */
unsigned long cflag; /* commit flags */
+ u64 agstart; /* agstart of the containing IAG */
u16 bxflag; /* xflag of pseudo buffer? */
- unchar agno; /* ag number */
+ unchar pad;
signed char active_ag; /* ag currently allocating from */
lid_t blid; /* lid of pseudo buffer? */
lid_t atlhead; /* anonymous tlock list head */
@@ -191,9 +192,10 @@ struct jfs_sb_info {
uint state; /* mount/recovery state */
unsigned long flag; /* mount time flags */
uint p_state; /* state prior to going no integrity */
- uint uid; /* uid to override on-disk uid */
- uint gid; /* gid to override on-disk gid */
+ kuid_t uid; /* uid to override on-disk uid */
+ kgid_t gid; /* gid to override on-disk gid */
uint umask; /* umask to override on-disk umask */
+ uint minblks_trim; /* minimum blocks, for online trim */
};
/* jfs_sb_info commit_state */
diff --git a/fs/jfs/jfs_inode.c b/fs/jfs/jfs_inode.c
index 2686531e235..6b0f816201a 100644
--- a/fs/jfs/jfs_inode.c
+++ b/fs/jfs/jfs_inode.c
@@ -29,20 +29,20 @@
void jfs_set_inode_flags(struct inode *inode)
{
unsigned int flags = JFS_IP(inode)->mode2;
-
- inode->i_flags &= ~(S_IMMUTABLE | S_APPEND |
- S_NOATIME | S_DIRSYNC | S_SYNC);
+ unsigned int new_fl = 0;
if (flags & JFS_IMMUTABLE_FL)
- inode->i_flags |= S_IMMUTABLE;
+ new_fl |= S_IMMUTABLE;
if (flags & JFS_APPEND_FL)
- inode->i_flags |= S_APPEND;
+ new_fl |= S_APPEND;
if (flags & JFS_NOATIME_FL)
- inode->i_flags |= S_NOATIME;
+ new_fl |= S_NOATIME;
if (flags & JFS_DIRSYNC_FL)
- inode->i_flags |= S_DIRSYNC;
+ new_fl |= S_DIRSYNC;
if (flags & JFS_SYNC_FL)
- inode->i_flags |= S_SYNC;
+ new_fl |= S_SYNC;
+ inode_set_flags(inode, new_fl, S_IMMUTABLE | S_APPEND | S_NOATIME |
+ S_DIRSYNC | S_SYNC);
}
void jfs_get_inode_flags(struct jfs_inode_info *jfs_ip)
@@ -95,7 +95,7 @@ struct inode *ialloc(struct inode *parent, umode_t mode)
if (insert_inode_locked(inode) < 0) {
rc = -EINVAL;
- goto fail_unlock;
+ goto fail_put;
}
inode_init_owner(inode, parent, mode);
@@ -156,8 +156,7 @@ struct inode *ialloc(struct inode *parent, umode_t mode)
fail_drop:
dquot_drop(inode);
inode->i_flags |= S_NOQUOTA;
-fail_unlock:
- inode->i_nlink = 0;
+ clear_nlink(inode);
unlock_new_inode(inode);
fail_put:
iput(inode);
diff --git a/fs/jfs/jfs_inode.h b/fs/jfs/jfs_inode.h
index 155e91eff07..9271cfe4a14 100644
--- a/fs/jfs/jfs_inode.h
+++ b/fs/jfs/jfs_inode.h
@@ -21,14 +21,14 @@
struct fid;
extern struct inode *ialloc(struct inode *, umode_t);
-extern int jfs_fsync(struct file *, int);
+extern int jfs_fsync(struct file *, loff_t, loff_t, int);
extern long jfs_ioctl(struct file *, unsigned int, unsigned long);
extern long jfs_compat_ioctl(struct file *, unsigned int, unsigned long);
extern struct inode *jfs_iget(struct super_block *, unsigned long);
extern int jfs_commit_inode(struct inode *, int);
extern int jfs_write_inode(struct inode *, struct writeback_control *);
extern void jfs_evict_inode(struct inode *);
-extern void jfs_dirty_inode(struct inode *);
+extern void jfs_dirty_inode(struct inode *, int);
extern void jfs_truncate(struct inode *);
extern void jfs_truncate_nolock(struct inode *, loff_t);
extern void jfs_free_zero_link(struct inode *);
diff --git a/fs/jfs/jfs_logmgr.c b/fs/jfs/jfs_logmgr.c
index e1b8493b9aa..0acddf60af5 100644
--- a/fs/jfs/jfs_logmgr.c
+++ b/fs/jfs/jfs_logmgr.c
@@ -67,6 +67,7 @@
#include <linux/buffer_head.h> /* for sync_blockdev() */
#include <linux/bio.h>
#include <linux/freezer.h>
+#include <linux/export.h>
#include <linux/delay.h>
#include <linux/mutex.h>
#include <linux/seq_file.h>
@@ -166,7 +167,7 @@ do { \
* Global list of active external journals
*/
static LIST_HEAD(jfs_external_logs);
-static struct jfs_log *dummy_log = NULL;
+static struct jfs_log *dummy_log;
static DEFINE_MUTEX(jfs_log_mutex);
/*
@@ -1057,7 +1058,8 @@ static int lmLogSync(struct jfs_log * log, int hard_sync)
*/
void jfs_syncpt(struct jfs_log *log, int hard_sync)
{ LOG_LOCK(log);
- lmLogSync(log, hard_sync);
+ if (!test_bit(log_QUIESCE, &log->flag))
+ lmLogSync(log, hard_sync);
LOG_UNLOCK(log);
}
@@ -1120,16 +1122,13 @@ int lmLogOpen(struct super_block *sb)
* file systems to log may have n-to-1 relationship;
*/
- bdev = open_by_devnum(sbi->logdev, FMODE_READ|FMODE_WRITE);
+ bdev = blkdev_get_by_dev(sbi->logdev, FMODE_READ|FMODE_WRITE|FMODE_EXCL,
+ log);
if (IS_ERR(bdev)) {
- rc = -PTR_ERR(bdev);
+ rc = PTR_ERR(bdev);
goto free;
}
- if ((rc = bd_claim(bdev, log))) {
- goto close;
- }
-
log->bdev = bdev;
memcpy(log->uuid, sbi->loguuid, sizeof(log->uuid));
@@ -1137,7 +1136,7 @@ int lmLogOpen(struct super_block *sb)
* initialize log:
*/
if ((rc = lmLogInit(log)))
- goto unclaim;
+ goto close;
list_add(&log->journal_list, &jfs_external_logs);
@@ -1163,11 +1162,8 @@ journal_found:
list_del(&log->journal_list);
lbmLogShutdown(log);
- unclaim:
- bd_release(bdev);
-
close: /* close external log device */
- blkdev_put(bdev, FMODE_READ|FMODE_WRITE);
+ blkdev_put(bdev, FMODE_READ|FMODE_WRITE|FMODE_EXCL);
free: /* free log descriptor */
mutex_unlock(&jfs_log_mutex);
@@ -1512,8 +1508,7 @@ int lmLogClose(struct super_block *sb)
bdev = log->bdev;
rc = lmLogShutdown(log);
- bd_release(bdev);
- blkdev_put(bdev, FMODE_READ|FMODE_WRITE);
+ blkdev_put(bdev, FMODE_READ|FMODE_WRITE|FMODE_EXCL);
kfree(log);
@@ -2003,19 +1998,24 @@ static int lbmRead(struct jfs_log * log, int pn, struct lbuf ** bpp)
bio = bio_alloc(GFP_NOFS, 1);
- bio->bi_sector = bp->l_blkno << (log->l2bsize - 9);
+ bio->bi_iter.bi_sector = bp->l_blkno << (log->l2bsize - 9);
bio->bi_bdev = log->bdev;
bio->bi_io_vec[0].bv_page = bp->l_page;
bio->bi_io_vec[0].bv_len = LOGPSIZE;
bio->bi_io_vec[0].bv_offset = bp->l_offset;
bio->bi_vcnt = 1;
- bio->bi_idx = 0;
- bio->bi_size = LOGPSIZE;
+ bio->bi_iter.bi_size = LOGPSIZE;
bio->bi_end_io = lbmIODone;
bio->bi_private = bp;
- submit_bio(READ_SYNC, bio);
+ /*check if journaling to disk has been disabled*/
+ if (log->no_integrity) {
+ bio->bi_iter.bi_size = 0;
+ lbmIODone(bio, 0);
+ } else {
+ submit_bio(READ_SYNC, bio);
+ }
wait_event(bp->l_ioevent, (bp->l_flag != lbmREAD));
@@ -2144,22 +2144,21 @@ static void lbmStartIO(struct lbuf * bp)
jfs_info("lbmStartIO\n");
bio = bio_alloc(GFP_NOFS, 1);
- bio->bi_sector = bp->l_blkno << (log->l2bsize - 9);
+ bio->bi_iter.bi_sector = bp->l_blkno << (log->l2bsize - 9);
bio->bi_bdev = log->bdev;
bio->bi_io_vec[0].bv_page = bp->l_page;
bio->bi_io_vec[0].bv_len = LOGPSIZE;
bio->bi_io_vec[0].bv_offset = bp->l_offset;
bio->bi_vcnt = 1;
- bio->bi_idx = 0;
- bio->bi_size = LOGPSIZE;
+ bio->bi_iter.bi_size = LOGPSIZE;
bio->bi_end_io = lbmIODone;
bio->bi_private = bp;
/* check if journaling to disk has been disabled */
if (log->no_integrity) {
- bio->bi_size = 0;
+ bio->bi_iter.bi_size = 0;
lbmIODone(bio, 0);
} else {
submit_bio(WRITE_SYNC, bio);
@@ -2355,7 +2354,7 @@ int jfsIOWait(void *arg)
if (freezing(current)) {
spin_unlock_irq(&log_redrive_lock);
- refrigerator();
+ try_to_freeze();
} else {
set_current_state(TASK_INTERRUPTIBLE);
spin_unlock_irq(&log_redrive_lock);
diff --git a/fs/jfs/jfs_logmgr.h b/fs/jfs/jfs_logmgr.h
index 9236bc49ae7..e38c2159885 100644
--- a/fs/jfs/jfs_logmgr.h
+++ b/fs/jfs/jfs_logmgr.h
@@ -288,7 +288,7 @@ struct lrd {
/*
* SYNCPT: log sync point
*
- * replay log upto syncpt address specified;
+ * replay log up to syncpt address specified;
*/
struct {
__le32 sync; /* 4: syncpt address (0 = here) */
diff --git a/fs/jfs/jfs_metapage.c b/fs/jfs/jfs_metapage.c
index 48b44bd8267..49ba7ff1bbb 100644
--- a/fs/jfs/jfs_metapage.c
+++ b/fs/jfs/jfs_metapage.c
@@ -416,7 +416,7 @@ static int metapage_writepage(struct page *page, struct writeback_control *wbc)
* count from hitting zero before we're through
*/
inc_io(page);
- if (!bio->bi_size)
+ if (!bio->bi_iter.bi_size)
goto dump_bio;
submit_bio(WRITE, bio);
nr_underway++;
@@ -438,7 +438,7 @@ static int metapage_writepage(struct page *page, struct writeback_control *wbc)
bio = bio_alloc(GFP_NOFS, 1);
bio->bi_bdev = inode->i_sb->s_bdev;
- bio->bi_sector = pblock << (inode->i_blkbits - 9);
+ bio->bi_iter.bi_sector = pblock << (inode->i_blkbits - 9);
bio->bi_end_io = metapage_write_end_io;
bio->bi_private = page;
@@ -452,7 +452,7 @@ static int metapage_writepage(struct page *page, struct writeback_control *wbc)
if (bio) {
if (bio_add_page(bio, page, bio_bytes, bio_offset) < bio_bytes)
goto add_failed;
- if (!bio->bi_size)
+ if (!bio->bi_iter.bi_size)
goto dump_bio;
submit_bio(WRITE, bio);
@@ -517,7 +517,8 @@ static int metapage_readpage(struct file *fp, struct page *page)
bio = bio_alloc(GFP_NOFS, 1);
bio->bi_bdev = inode->i_sb->s_bdev;
- bio->bi_sector = pblock << (inode->i_blkbits - 9);
+ bio->bi_iter.bi_sector =
+ pblock << (inode->i_blkbits - 9);
bio->bi_end_io = metapage_read_end_io;
bio->bi_private = page;
len = xlen << inode->i_blkbits;
@@ -571,9 +572,10 @@ static int metapage_releasepage(struct page *page, gfp_t gfp_mask)
return ret;
}
-static void metapage_invalidatepage(struct page *page, unsigned long offset)
+static void metapage_invalidatepage(struct page *page, unsigned int offset,
+ unsigned int length)
{
- BUG_ON(offset);
+ BUG_ON(offset || length < PAGE_CACHE_SIZE);
BUG_ON(PageWriteback(page));
@@ -583,7 +585,6 @@ static void metapage_invalidatepage(struct page *page, unsigned long offset)
const struct address_space_operations jfs_metapage_aops = {
.readpage = metapage_readpage,
.writepage = metapage_writepage,
- .sync_page = block_sync_page,
.releasepage = metapage_releasepage,
.invalidatepage = metapage_invalidatepage,
.set_page_dirty = __set_page_dirty_nobuffers,
@@ -647,7 +648,7 @@ struct metapage *__get_metapage(struct inode *inode, unsigned long lblock,
if (mp) {
if (mp->logical_size != size) {
jfs_error(inode->i_sb,
- "__get_metapage: mp->logical_size != size");
+ "get_mp->logical_size != size\n");
jfs_err("logical_size = %d, size = %d",
mp->logical_size, size);
dump_stack();
@@ -658,8 +659,7 @@ struct metapage *__get_metapage(struct inode *inode, unsigned long lblock,
if (test_bit(META_discard, &mp->flag)) {
if (!new) {
jfs_error(inode->i_sb,
- "__get_metapage: using a "
- "discarded metapage");
+ "using a discarded metapage\n");
discard_metapage(mp);
goto unlock;
}
diff --git a/fs/jfs/jfs_metapage.h b/fs/jfs/jfs_metapage.h
index d94f8d9e87d..a78beda85f6 100644
--- a/fs/jfs/jfs_metapage.h
+++ b/fs/jfs/jfs_metapage.h
@@ -75,7 +75,7 @@ extern void grab_metapage(struct metapage *);
extern void force_metapage(struct metapage *);
/*
- * hold_metapage and put_metapage are used in conjuction. The page lock
+ * hold_metapage and put_metapage are used in conjunction. The page lock
* is not dropped between the two, so no other threads can get or release
* the metapage
*/
diff --git a/fs/jfs/jfs_superblock.h b/fs/jfs/jfs_superblock.h
index 884fc21ab8e..04847b8d307 100644
--- a/fs/jfs/jfs_superblock.h
+++ b/fs/jfs/jfs_superblock.h
@@ -108,6 +108,7 @@ struct jfs_superblock {
extern int readSuper(struct super_block *, struct buffer_head **);
extern int updateSuper(struct super_block *, uint);
+__printf(2, 3)
extern void jfs_error(struct super_block *, const char *, ...);
extern int jfs_mount(struct super_block *);
extern int jfs_mount_rw(struct super_block *, int);
diff --git a/fs/jfs/jfs_txnmgr.c b/fs/jfs/jfs_txnmgr.c
index 9466957ec84..564c4f279ac 100644
--- a/fs/jfs/jfs_txnmgr.c
+++ b/fs/jfs/jfs_txnmgr.c
@@ -636,7 +636,7 @@ struct tlock *txLock(tid_t tid, struct inode *ip, struct metapage * mp,
* the inode of the page and available to all anonymous
* transactions until txCommit() time at which point
* they are transferred to the transaction tlock list of
- * the commiting transaction of the inode)
+ * the committing transaction of the inode)
*/
if (xtid == 0) {
tlck->tid = tid;
@@ -1143,7 +1143,6 @@ int txCommit(tid_t tid, /* transaction identifier */
struct jfs_log *log;
struct tblock *tblk;
struct lrd *lrd;
- int lsn;
struct inode *ip;
struct jfs_inode_info *jfs_ip;
int k, n;
@@ -1310,7 +1309,7 @@ int txCommit(tid_t tid, /* transaction identifier */
*/
lrd->type = cpu_to_le16(LOG_COMMIT);
lrd->length = 0;
- lsn = lmLog(log, tblk, lrd, NULL);
+ lmLog(log, tblk, lrd, NULL);
lmGroupCommit(log, tblk);
@@ -2685,7 +2684,7 @@ void txAbort(tid_t tid, int dirty)
* mark filesystem dirty
*/
if (dirty)
- jfs_error(tblk->sb, "txAbort");
+ jfs_error(tblk->sb, "\n");
return;
}
@@ -2801,7 +2800,7 @@ int jfs_lazycommit(void *arg)
if (freezing(current)) {
LAZY_UNLOCK(flags);
- refrigerator();
+ try_to_freeze();
} else {
DECLARE_WAITQUEUE(wq, current);
@@ -2935,7 +2934,6 @@ int jfs_sync(void *arg)
{
struct inode *ip;
struct jfs_inode_info *jfs_ip;
- int rc;
tid_t tid;
do {
@@ -2961,7 +2959,7 @@ int jfs_sync(void *arg)
*/
TXN_UNLOCK();
tid = txBegin(ip->i_sb, COMMIT_INODE);
- rc = txCommit(tid, 1, &ip, 0);
+ txCommit(tid, 1, &ip, 0);
txEnd(tid);
mutex_unlock(&jfs_ip->commit_mutex);
@@ -2979,12 +2977,9 @@ int jfs_sync(void *arg)
* put back on the anon_list.
*/
- /* Take off anon_list */
- list_del(&jfs_ip->anon_inode_list);
-
- /* Put on anon_list2 */
- list_add(&jfs_ip->anon_inode_list,
- &TxAnchor.anon_list2);
+ /* Move from anon_list to anon_list2 */
+ list_move(&jfs_ip->anon_inode_list,
+ &TxAnchor.anon_list2);
TXN_UNLOCK();
iput(ip);
@@ -2996,7 +2991,7 @@ int jfs_sync(void *arg)
if (freezing(current)) {
TXN_UNLOCK();
- refrigerator();
+ try_to_freeze();
} else {
set_current_state(TASK_INTERRUPTIBLE);
TXN_UNLOCK();
diff --git a/fs/jfs/jfs_umount.c b/fs/jfs/jfs_umount.c
index adcf92d3b60..7971f37534a 100644
--- a/fs/jfs/jfs_umount.c
+++ b/fs/jfs/jfs_umount.c
@@ -68,7 +68,7 @@ int jfs_umount(struct super_block *sb)
/*
* Wait for outstanding transactions to be written to log:
*/
- jfs_flush_journal(log, 1);
+ jfs_flush_journal(log, 2);
/*
* close fileset inode allocation map (aka fileset inode)
@@ -146,7 +146,7 @@ int jfs_umount_rw(struct super_block *sb)
*
* remove file system from log active file system list.
*/
- jfs_flush_journal(log, 1);
+ jfs_flush_journal(log, 2);
/*
* Make sure all metadata makes it to disk
diff --git a/fs/jfs/jfs_xattr.h b/fs/jfs/jfs_xattr.h
index 88b6cc535bf..e8d717dabca 100644
--- a/fs/jfs/jfs_xattr.h
+++ b/fs/jfs/jfs_xattr.h
@@ -61,11 +61,14 @@ extern ssize_t jfs_getxattr(struct dentry *, const char *, void *, size_t);
extern ssize_t jfs_listxattr(struct dentry *, char *, size_t);
extern int jfs_removexattr(struct dentry *, const char *);
+extern const struct xattr_handler *jfs_xattr_handlers[];
+
#ifdef CONFIG_JFS_SECURITY
-extern int jfs_init_security(tid_t, struct inode *, struct inode *);
+extern int jfs_init_security(tid_t, struct inode *, struct inode *,
+ const struct qstr *);
#else
static inline int jfs_init_security(tid_t tid, struct inode *inode,
- struct inode *dir)
+ struct inode *dir, const struct qstr *qstr)
{
return 0;
}
diff --git a/fs/jfs/jfs_xtree.c b/fs/jfs/jfs_xtree.c
index 6c50871e622..5ad7748860c 100644
--- a/fs/jfs/jfs_xtree.c
+++ b/fs/jfs/jfs_xtree.c
@@ -64,22 +64,23 @@
/* get page buffer for specified block address */
/* ToDo: Replace this ugly macro with a function */
-#define XT_GETPAGE(IP, BN, MP, SIZE, P, RC)\
-{\
- BT_GETPAGE(IP, BN, MP, xtpage_t, SIZE, P, RC, i_xtroot)\
- if (!(RC))\
- {\
- if ((le16_to_cpu((P)->header.nextindex) < XTENTRYSTART) ||\
- (le16_to_cpu((P)->header.nextindex) > le16_to_cpu((P)->header.maxentry)) ||\
- (le16_to_cpu((P)->header.maxentry) > (((BN)==0)?XTROOTMAXSLOT:PSIZE>>L2XTSLOTSIZE)))\
- {\
- jfs_error((IP)->i_sb, "XT_GETPAGE: xtree page corrupt");\
- BT_PUTPAGE(MP);\
- MP = NULL;\
- RC = -EIO;\
- }\
- }\
-}
+#define XT_GETPAGE(IP, BN, MP, SIZE, P, RC) \
+do { \
+ BT_GETPAGE(IP, BN, MP, xtpage_t, SIZE, P, RC, i_xtroot); \
+ if (!(RC)) { \
+ if ((le16_to_cpu((P)->header.nextindex) < XTENTRYSTART) || \
+ (le16_to_cpu((P)->header.nextindex) > \
+ le16_to_cpu((P)->header.maxentry)) || \
+ (le16_to_cpu((P)->header.maxentry) > \
+ (((BN) == 0) ? XTROOTMAXSLOT : PSIZE >> L2XTSLOTSIZE))) { \
+ jfs_error((IP)->i_sb, \
+ "XT_GETPAGE: xtree page corrupt\n"); \
+ BT_PUTPAGE(MP); \
+ MP = NULL; \
+ RC = -EIO; \
+ } \
+ } \
+} while (0)
/* for consistency */
#define XT_PUTPAGE(MP) BT_PUTPAGE(MP)
@@ -499,7 +500,7 @@ static int xtSearch(struct inode *ip, s64 xoff, s64 *nextp,
/* push (bn, index) of the parent page/entry */
if (BT_STACK_FULL(btstack)) {
- jfs_error(ip->i_sb, "stack overrun in xtSearch!");
+ jfs_error(ip->i_sb, "stack overrun!\n");
XT_PUTPAGE(mp);
return -EIO;
}
@@ -1385,7 +1386,7 @@ int xtExtend(tid_t tid, /* transaction id */
if (cmp != 0) {
XT_PUTPAGE(mp);
- jfs_error(ip->i_sb, "xtExtend: xtSearch did not find extent");
+ jfs_error(ip->i_sb, "xtSearch did not find extent\n");
return -EIO;
}
@@ -1393,7 +1394,7 @@ int xtExtend(tid_t tid, /* transaction id */
xad = &p->xad[index];
if ((offsetXAD(xad) + lengthXAD(xad)) != xoff) {
XT_PUTPAGE(mp);
- jfs_error(ip->i_sb, "xtExtend: extension is not contiguous");
+ jfs_error(ip->i_sb, "extension is not contiguous\n");
return -EIO;
}
@@ -1552,7 +1553,7 @@ printf("xtTailgate: nxoff:0x%lx nxlen:0x%x nxaddr:0x%lx\n",
if (cmp != 0) {
XT_PUTPAGE(mp);
- jfs_error(ip->i_sb, "xtTailgate: couldn't find extent");
+ jfs_error(ip->i_sb, "couldn't find extent\n");
return -EIO;
}
@@ -1560,8 +1561,7 @@ printf("xtTailgate: nxoff:0x%lx nxlen:0x%x nxaddr:0x%lx\n",
nextindex = le16_to_cpu(p->header.nextindex);
if (index != nextindex - 1) {
XT_PUTPAGE(mp);
- jfs_error(ip->i_sb,
- "xtTailgate: the entry found is not the last entry");
+ jfs_error(ip->i_sb, "the entry found is not the last entry\n");
return -EIO;
}
@@ -1734,7 +1734,7 @@ int xtUpdate(tid_t tid, struct inode *ip, xad_t * nxad)
if (cmp != 0) {
XT_PUTPAGE(mp);
- jfs_error(ip->i_sb, "xtUpdate: Could not find extent");
+ jfs_error(ip->i_sb, "Could not find extent\n");
return -EIO;
}
@@ -1758,7 +1758,7 @@ int xtUpdate(tid_t tid, struct inode *ip, xad_t * nxad)
(nxoff + nxlen > xoff + xlen)) {
XT_PUTPAGE(mp);
jfs_error(ip->i_sb,
- "xtUpdate: nXAD in not completely contained within XAD");
+ "nXAD in not completely contained within XAD\n");
return -EIO;
}
@@ -1907,7 +1907,7 @@ int xtUpdate(tid_t tid, struct inode *ip, xad_t * nxad)
if (xoff >= nxoff) {
XT_PUTPAGE(mp);
- jfs_error(ip->i_sb, "xtUpdate: xoff >= nxoff");
+ jfs_error(ip->i_sb, "xoff >= nxoff\n");
return -EIO;
}
/* #endif _JFS_WIP_COALESCE */
@@ -2048,14 +2048,13 @@ int xtUpdate(tid_t tid, struct inode *ip, xad_t * nxad)
if (cmp != 0) {
XT_PUTPAGE(mp);
- jfs_error(ip->i_sb, "xtUpdate: xtSearch failed");
+ jfs_error(ip->i_sb, "xtSearch failed\n");
return -EIO;
}
if (index0 != index) {
XT_PUTPAGE(mp);
- jfs_error(ip->i_sb,
- "xtUpdate: unexpected value of index");
+ jfs_error(ip->i_sb, "unexpected value of index\n");
return -EIO;
}
}
@@ -3650,7 +3649,7 @@ s64 xtTruncate(tid_t tid, struct inode *ip, s64 newsize, int flag)
getChild:
/* save current parent entry for the child page */
if (BT_STACK_FULL(&btstack)) {
- jfs_error(ip->i_sb, "stack overrun in xtTruncate!");
+ jfs_error(ip->i_sb, "stack overrun!\n");
XT_PUTPAGE(mp);
return -EIO;
}
@@ -3751,8 +3750,7 @@ s64 xtTruncate_pmap(tid_t tid, struct inode *ip, s64 committed_size)
if (cmp != 0) {
XT_PUTPAGE(mp);
- jfs_error(ip->i_sb,
- "xtTruncate_pmap: did not find extent");
+ jfs_error(ip->i_sb, "did not find extent\n");
return -EIO;
}
} else {
@@ -3851,7 +3849,7 @@ s64 xtTruncate_pmap(tid_t tid, struct inode *ip, s64 committed_size)
getChild:
/* save current parent entry for the child page */
if (BT_STACK_FULL(&btstack)) {
- jfs_error(ip->i_sb, "stack overrun in xtTruncate_pmap!");
+ jfs_error(ip->i_sb, "stack overrun!\n");
XT_PUTPAGE(mp);
return -EIO;
}
diff --git a/fs/jfs/namei.c b/fs/jfs/namei.c
index 231ca4af9bc..d59c7defb1e 100644
--- a/fs/jfs/namei.c
+++ b/fs/jfs/namei.c
@@ -18,6 +18,7 @@
*/
#include <linux/fs.h>
+#include <linux/namei.h>
#include <linux/ctype.h>
#include <linux/quotaops.h>
#include <linux/exportfs.h>
@@ -71,8 +72,8 @@ static inline void free_ea_wmap(struct inode *inode)
* RETURN: Errors from subroutines
*
*/
-static int jfs_create(struct inode *dip, struct dentry *dentry, int mode,
- struct nameidata *nd)
+static int jfs_create(struct inode *dip, struct dentry *dentry, umode_t mode,
+ bool excl)
{
int rc = 0;
tid_t tid; /* transaction id */
@@ -114,7 +115,7 @@ static int jfs_create(struct inode *dip, struct dentry *dentry, int mode,
if (rc)
goto out3;
- rc = jfs_init_security(tid, ip, dip);
+ rc = jfs_init_security(tid, ip, dip, &dentry->d_name);
if (rc) {
txAbort(tid, 0);
goto out3;
@@ -171,12 +172,12 @@ static int jfs_create(struct inode *dip, struct dentry *dentry, int mode,
mutex_unlock(&JFS_IP(dip)->commit_mutex);
if (rc) {
free_ea_wmap(ip);
- ip->i_nlink = 0;
+ clear_nlink(ip);
unlock_new_inode(ip);
iput(ip);
} else {
- d_instantiate(dentry, ip);
unlock_new_inode(ip);
+ d_instantiate(dentry, ip);
}
out2:
@@ -204,7 +205,7 @@ static int jfs_create(struct inode *dip, struct dentry *dentry, int mode,
* note:
* EACCESS: user needs search+write permission on the parent directory
*/
-static int jfs_mkdir(struct inode *dip, struct dentry *dentry, int mode)
+static int jfs_mkdir(struct inode *dip, struct dentry *dentry, umode_t mode)
{
int rc = 0;
tid_t tid; /* transaction id */
@@ -219,12 +220,6 @@ static int jfs_mkdir(struct inode *dip, struct dentry *dentry, int mode)
dquot_initialize(dip);
- /* link count overflow on parent directory ? */
- if (dip->i_nlink == JFS_LINK_MAX) {
- rc = -EMLINK;
- goto out1;
- }
-
/*
* search parent directory for entry/freespace
* (dtSearch() returns parent directory page pinned)
@@ -252,7 +247,7 @@ static int jfs_mkdir(struct inode *dip, struct dentry *dentry, int mode)
if (rc)
goto out3;
- rc = jfs_init_security(tid, ip, dip);
+ rc = jfs_init_security(tid, ip, dip, &dentry->d_name);
if (rc) {
txAbort(tid, 0);
goto out3;
@@ -291,7 +286,7 @@ static int jfs_mkdir(struct inode *dip, struct dentry *dentry, int mode)
goto out3;
}
- ip->i_nlink = 2; /* for '.' */
+ set_nlink(ip, 2); /* for '.' */
ip->i_op = &jfs_dir_inode_operations;
ip->i_fop = &jfs_dir_operations;
@@ -310,12 +305,12 @@ static int jfs_mkdir(struct inode *dip, struct dentry *dentry, int mode)
mutex_unlock(&JFS_IP(dip)->commit_mutex);
if (rc) {
free_ea_wmap(ip);
- ip->i_nlink = 0;
+ clear_nlink(ip);
unlock_new_inode(ip);
iput(ip);
} else {
- d_instantiate(dentry, ip);
unlock_new_inode(ip);
+ d_instantiate(dentry, ip);
}
out2:
@@ -805,12 +800,6 @@ static int jfs_link(struct dentry *old_dentry,
jfs_info("jfs_link: %s %s", old_dentry->d_name.name,
dentry->d_name.name);
- if (ip->i_nlink == JFS_LINK_MAX)
- return -EMLINK;
-
- if (ip->i_nlink == 0)
- return -ENOENT;
-
dquot_initialize(dir);
tid = txBegin(ip->i_sb, 0);
@@ -846,7 +835,7 @@ static int jfs_link(struct dentry *old_dentry,
rc = txCommit(tid, 2, &iplist[0], 0);
if (rc) {
- ip->i_nlink--; /* never instantiated */
+ drop_nlink(ip); /* never instantiated */
iput(ip);
} else
d_instantiate(dentry, ip);
@@ -895,7 +884,7 @@ static int jfs_symlink(struct inode *dip, struct dentry *dentry,
unchar *i_fastsymlink;
s64 xlen = 0;
int bmask = 0, xsize;
- s64 extent = 0, xaddr;
+ s64 xaddr;
struct metapage *mp;
struct super_block *sb;
struct tblock *tblk;
@@ -931,7 +920,7 @@ static int jfs_symlink(struct inode *dip, struct dentry *dentry,
mutex_lock_nested(&JFS_IP(dip)->commit_mutex, COMMIT_MUTEX_PARENT);
mutex_lock_nested(&JFS_IP(ip)->commit_mutex, COMMIT_MUTEX_CHILD);
- rc = jfs_init_security(tid, ip, dip);
+ rc = jfs_init_security(tid, ip, dip, &dentry->d_name);
if (rc)
goto out3;
@@ -995,7 +984,6 @@ static int jfs_symlink(struct inode *dip, struct dentry *dentry,
txAbort(tid, 0);
goto out3;
}
- extent = xaddr;
ip->i_size = ssize - 1;
while (ssize) {
/* This is kind of silly since PATH_MAX == 4K */
@@ -1051,12 +1039,12 @@ static int jfs_symlink(struct inode *dip, struct dentry *dentry,
mutex_unlock(&JFS_IP(dip)->commit_mutex);
if (rc) {
free_ea_wmap(ip);
- ip->i_nlink = 0;
+ clear_nlink(ip);
unlock_new_inode(ip);
iput(ip);
} else {
- d_instantiate(dentry, ip);
unlock_new_inode(ip);
+ d_instantiate(dentry, ip);
}
out2:
@@ -1141,10 +1129,6 @@ static int jfs_rename(struct inode *old_dir, struct dentry *old_dentry,
rc = -ENOTEMPTY;
goto out3;
}
- } else if ((new_dir != old_dir) &&
- (new_dir->i_nlink == JFS_LINK_MAX)) {
- rc = -EMLINK;
- goto out3;
}
} else if (new_ip) {
IWRITE_LOCK(new_ip, RDWRLOCK_NORMAL);
@@ -1192,7 +1176,7 @@ static int jfs_rename(struct inode *old_dir, struct dentry *old_dentry,
if (!S_ISDIR(old_ip->i_mode) && new_ip)
IWRITE_UNLOCK(new_ip);
jfs_error(new_ip->i_sb,
- "jfs_rename: new_ip->i_nlink != 0");
+ "new_ip->i_nlink != 0\n");
return -EIO;
}
tblk = tid_to_tblock(tid);
@@ -1356,7 +1340,7 @@ static int jfs_rename(struct inode *old_dir, struct dentry *old_dentry,
* FUNCTION: Create a special file (device)
*/
static int jfs_mknod(struct inode *dir, struct dentry *dentry,
- int mode, dev_t rdev)
+ umode_t mode, dev_t rdev)
{
struct jfs_inode_info *jfs_ip;
struct btstack btstack;
@@ -1394,7 +1378,7 @@ static int jfs_mknod(struct inode *dir, struct dentry *dentry,
if (rc)
goto out3;
- rc = jfs_init_security(tid, ip, dir);
+ rc = jfs_init_security(tid, ip, dir, &dentry->d_name);
if (rc) {
txAbort(tid, 0);
goto out3;
@@ -1436,12 +1420,12 @@ static int jfs_mknod(struct inode *dir, struct dentry *dentry,
mutex_unlock(&JFS_IP(dir)->commit_mutex);
if (rc) {
free_ea_wmap(ip);
- ip->i_nlink = 0;
+ clear_nlink(ip);
unlock_new_inode(ip);
iput(ip);
} else {
- d_instantiate(dentry, ip);
unlock_new_inode(ip);
+ d_instantiate(dentry, ip);
}
out1:
@@ -1452,51 +1436,32 @@ static int jfs_mknod(struct inode *dir, struct dentry *dentry,
return rc;
}
-static struct dentry *jfs_lookup(struct inode *dip, struct dentry *dentry, struct nameidata *nd)
+static struct dentry *jfs_lookup(struct inode *dip, struct dentry *dentry, unsigned int flags)
{
struct btstack btstack;
ino_t inum;
struct inode *ip;
struct component_name key;
- const char *name = dentry->d_name.name;
- int len = dentry->d_name.len;
int rc;
- jfs_info("jfs_lookup: name = %s", name);
-
- if (JFS_SBI(dip->i_sb)->mntflag & JFS_OS2)
- dentry->d_op = &jfs_ci_dentry_operations;
-
- if ((name[0] == '.') && (len == 1))
- inum = dip->i_ino;
- else if (strcmp(name, "..") == 0)
- inum = PARENT(dip);
- else {
- if ((rc = get_UCSname(&key, dentry)))
- return ERR_PTR(rc);
- rc = dtSearch(dip, &key, &inum, &btstack, JFS_LOOKUP);
- free_UCSname(&key);
- if (rc == -ENOENT) {
- d_add(dentry, NULL);
- return NULL;
- } else if (rc) {
- jfs_err("jfs_lookup: dtSearch returned %d", rc);
- return ERR_PTR(rc);
- }
- }
-
- ip = jfs_iget(dip->i_sb, inum);
- if (IS_ERR(ip)) {
- jfs_err("jfs_lookup: iget failed on inum %d", (uint) inum);
- return ERR_CAST(ip);
+ jfs_info("jfs_lookup: name = %s", dentry->d_name.name);
+
+ if ((rc = get_UCSname(&key, dentry)))
+ return ERR_PTR(rc);
+ rc = dtSearch(dip, &key, &inum, &btstack, JFS_LOOKUP);
+ free_UCSname(&key);
+ if (rc == -ENOENT) {
+ ip = NULL;
+ } else if (rc) {
+ jfs_err("jfs_lookup: dtSearch returned %d", rc);
+ ip = ERR_PTR(rc);
+ } else {
+ ip = jfs_iget(dip->i_sb, inum);
+ if (IS_ERR(ip))
+ jfs_err("jfs_lookup: iget failed on inum %d", (uint)inum);
}
- dentry = d_splice_alias(ip, dentry);
-
- if (dentry && (JFS_SBI(dip->i_sb)->mntflag & JFS_OS2))
- dentry->d_op = &jfs_ci_dentry_operations;
-
- return dentry;
+ return d_splice_alias(ip, dentry);
}
static struct inode *jfs_nfs_get_inode(struct super_block *sb,
@@ -1558,13 +1523,14 @@ const struct inode_operations jfs_dir_inode_operations = {
.removexattr = jfs_removexattr,
.setattr = jfs_setattr,
#ifdef CONFIG_JFS_POSIX_ACL
- .check_acl = jfs_check_acl,
+ .get_acl = jfs_get_acl,
+ .set_acl = jfs_set_acl,
#endif
};
const struct file_operations jfs_dir_operations = {
.read = generic_read_dir,
- .readdir = jfs_readdir,
+ .iterate = jfs_readdir,
.fsync = jfs_fsync,
.unlocked_ioctl = jfs_ioctl,
#ifdef CONFIG_COMPAT
@@ -1573,7 +1539,7 @@ const struct file_operations jfs_dir_operations = {
.llseek = generic_file_llseek,
};
-static int jfs_ci_hash(struct dentry *dir, struct qstr *this)
+static int jfs_ci_hash(const struct dentry *dir, struct qstr *this)
{
unsigned long hash;
int i;
@@ -1586,32 +1552,57 @@ static int jfs_ci_hash(struct dentry *dir, struct qstr *this)
return 0;
}
-static int jfs_ci_compare(struct dentry *dir, struct qstr *a, struct qstr *b)
+static int jfs_ci_compare(const struct dentry *parent, const struct dentry *dentry,
+ unsigned int len, const char *str, const struct qstr *name)
{
int i, result = 1;
- if (a->len != b->len)
+ if (len != name->len)
goto out;
- for (i=0; i < a->len; i++) {
- if (tolower(a->name[i]) != tolower(b->name[i]))
+ for (i=0; i < len; i++) {
+ if (tolower(str[i]) != tolower(name->name[i]))
goto out;
}
result = 0;
+out:
+ return result;
+}
+static int jfs_ci_revalidate(struct dentry *dentry, unsigned int flags)
+{
/*
- * We want creates to preserve case. A negative dentry, a, that
- * has a different case than b may cause a new entry to be created
- * with the wrong case. Since we can't tell if a comes from a negative
- * dentry, we blindly replace it with b. This should be harmless if
- * a is not a negative dentry.
+ * This is not negative dentry. Always valid.
+ *
+ * Note, rename() to existing directory entry will have ->d_inode,
+ * and will use existing name which isn't specified name by user.
+ *
+ * We may be able to drop this positive dentry here. But dropping
+ * positive dentry isn't good idea. So it's unsupported like
+ * rename("filename", "FILENAME") for now.
*/
- memcpy((unsigned char *)a->name, b->name, a->len);
-out:
- return result;
+ if (dentry->d_inode)
+ return 1;
+
+ /*
+ * This may be nfsd (or something), anyway, we can't see the
+ * intent of this. So, since this can be for creation, drop it.
+ */
+ if (!flags)
+ return 0;
+
+ /*
+ * Drop the negative dentry, in order to make sure to use the
+ * case sensitive name which is specified by user if this is
+ * for creation.
+ */
+ if (flags & (LOOKUP_CREATE | LOOKUP_RENAME_TARGET))
+ return 0;
+ return 1;
}
const struct dentry_operations jfs_ci_dentry_operations =
{
.d_hash = jfs_ci_hash,
.d_compare = jfs_ci_compare,
+ .d_revalidate = jfs_ci_revalidate,
};
diff --git a/fs/jfs/resize.c b/fs/jfs/resize.c
index 1aba0039f1c..90b3bc21e9b 100644
--- a/fs/jfs/resize.c
+++ b/fs/jfs/resize.c
@@ -57,7 +57,7 @@
* 2. compute new FSCKSize from new LVSize;
* 3. set new FSSize as MIN(FSSize, LVSize-(LogSize+FSCKSize)) where
* assert(new FSSize >= old FSSize),
- * i.e., file system must not be shrinked;
+ * i.e., file system must not be shrunk;
*/
int jfs_extendfs(struct super_block *sb, s64 newLVSize, int newLogSize)
{
@@ -80,7 +80,7 @@ int jfs_extendfs(struct super_block *sb, s64 newLVSize, int newLogSize)
int log_formatted = 0;
struct inode *iplist[1];
struct jfs_superblock *j_sb, *j_sb2;
- uint old_agsize;
+ s64 old_agsize;
int agsizechanged = 0;
struct buffer_head *bh, *bh2;
@@ -182,7 +182,7 @@ int jfs_extendfs(struct super_block *sb, s64 newLVSize, int newLogSize)
*/
newFSSize = newLVSize - newLogSize - newFSCKSize;
- /* file system cannot be shrinked */
+ /* file system cannot be shrunk */
if (newFSSize < bmp->db_mapsize) {
rc = -EINVAL;
goto out;
@@ -530,7 +530,7 @@ int jfs_extendfs(struct super_block *sb, s64 newLVSize, int newLogSize)
goto resume;
error_out:
- jfs_error(sb, "jfs_extendfs");
+ jfs_error(sb, "\n");
resume:
/*
diff --git a/fs/jfs/super.c b/fs/jfs/super.c
index 0669fc1cc3b..adf8cb045b9 100644
--- a/fs/jfs/super.c
+++ b/fs/jfs/super.c
@@ -33,6 +33,7 @@
#include <linux/slab.h>
#include <asm/uaccess.h>
#include <linux/seq_file.h>
+#include <linux/blkdev.h>
#include "jfs_incore.h"
#include "jfs_filsys.h"
@@ -43,19 +44,20 @@
#include "jfs_imap.h"
#include "jfs_acl.h"
#include "jfs_debug.h"
+#include "jfs_xattr.h"
MODULE_DESCRIPTION("The Journaled Filesystem (JFS)");
MODULE_AUTHOR("Steve Best/Dave Kleikamp/Barry Arndt, IBM");
MODULE_LICENSE("GPL");
-static struct kmem_cache * jfs_inode_cachep;
+static struct kmem_cache *jfs_inode_cachep;
static const struct super_operations jfs_super_operations;
static const struct export_operations jfs_export_operations;
static struct file_system_type jfs_fs_type;
#define MAX_COMMIT_THREADS 64
-static int commit_threads = 0;
+static int commit_threads;
module_param(commit_threads, int, 0);
MODULE_PARM_DESC(commit_threads, "Number of commit threads");
@@ -82,8 +84,7 @@ static void jfs_handle_error(struct super_block *sb)
panic("JFS (device %s): panic forced after error\n",
sb->s_id);
else if (sbi->flag & JFS_ERR_REMOUNT_RO) {
- jfs_err("ERROR: (device %s): remounting filesystem "
- "as read-only\n",
+ jfs_err("ERROR: (device %s): remounting filesystem as read-only\n",
sb->s_id);
sb->s_flags |= MS_RDONLY;
}
@@ -91,16 +92,20 @@ static void jfs_handle_error(struct super_block *sb)
/* nothing is done for continue beyond marking the superblock dirty */
}
-void jfs_error(struct super_block *sb, const char * function, ...)
+void jfs_error(struct super_block *sb, const char *fmt, ...)
{
- static char error_buf[256];
+ struct va_format vaf;
va_list args;
- va_start(args, function);
- vsnprintf(error_buf, sizeof(error_buf), function, args);
- va_end(args);
+ va_start(args, fmt);
+
+ vaf.fmt = fmt;
+ vaf.va = &args;
- printk(KERN_ERR "ERROR: (device %s): %s\n", sb->s_id, error_buf);
+ pr_err("ERROR: (device %s): %pf: %pV\n",
+ sb->s_id, __builtin_return_address(0), &vaf);
+
+ va_end(args);
jfs_handle_error(sb);
}
@@ -115,6 +120,13 @@ static struct inode *jfs_alloc_inode(struct super_block *sb)
return &jfs_inode->vfs_inode;
}
+static void jfs_i_callback(struct rcu_head *head)
+{
+ struct inode *inode = container_of(head, struct inode, i_rcu);
+ struct jfs_inode_info *ji = JFS_IP(inode);
+ kmem_cache_free(jfs_inode_cachep, ji);
+}
+
static void jfs_destroy_inode(struct inode *inode)
{
struct jfs_inode_info *ji = JFS_IP(inode);
@@ -128,7 +140,7 @@ static void jfs_destroy_inode(struct inode *inode)
ji->active_ag = -1;
}
spin_unlock_irq(&ji->ag_lock);
- kmem_cache_free(jfs_inode_cachep, ji);
+ call_rcu(&inode->i_rcu, jfs_i_callback);
}
static int jfs_statfs(struct dentry *dentry, struct kstatfs *buf)
@@ -146,7 +158,7 @@ static int jfs_statfs(struct dentry *dentry, struct kstatfs *buf)
/*
* If we really return the number of allocated & free inodes, some
* applications will fail because they won't see enough free inodes.
- * We'll try to calculate some guess as to how may inodes we can
+ * We'll try to calculate some guess as to how many inodes we can
* really allocate
*
* buf->f_files = atomic_read(&imap->im_numinos);
@@ -190,7 +202,8 @@ static void jfs_put_super(struct super_block *sb)
enum {
Opt_integrity, Opt_nointegrity, Opt_iocharset, Opt_resize,
Opt_resize_nosize, Opt_errors, Opt_ignore, Opt_err, Opt_quota,
- Opt_usrquota, Opt_grpquota, Opt_uid, Opt_gid, Opt_umask
+ Opt_usrquota, Opt_grpquota, Opt_uid, Opt_gid, Opt_umask,
+ Opt_discard, Opt_nodiscard, Opt_discard_minblk
};
static const match_table_t tokens = {
@@ -207,6 +220,9 @@ static const match_table_t tokens = {
{Opt_uid, "uid=%u"},
{Opt_gid, "gid=%u"},
{Opt_umask, "umask=%u"},
+ {Opt_discard, "discard"},
+ {Opt_nodiscard, "nodiscard"},
+ {Opt_discard_minblk, "discard=%u"},
{Opt_err, NULL}
};
@@ -248,8 +264,7 @@ static int parse_options(char *options, struct super_block *sb, s64 *newLVSize,
else {
nls_map = load_nls(args[0].from);
if (!nls_map) {
- printk(KERN_ERR
- "JFS: charset not found\n");
+ pr_err("JFS: charset not found\n");
goto cleanup;
}
}
@@ -257,7 +272,10 @@ static int parse_options(char *options, struct super_block *sb, s64 *newLVSize,
case Opt_resize:
{
char *resize = args[0].from;
- *newLVSize = simple_strtoull(resize, &resize, 0);
+ int rc = kstrtoll(resize, 0, newLVSize);
+
+ if (rc)
+ goto cleanup;
break;
}
case Opt_resize_nosize:
@@ -265,8 +283,7 @@ static int parse_options(char *options, struct super_block *sb, s64 *newLVSize,
*newLVSize = sb->s_bdev->bd_inode->i_size >>
sb->s_blocksize_bits;
if (*newLVSize == 0)
- printk(KERN_ERR
- "JFS: Cannot determine volume size\n");
+ pr_err("JFS: Cannot determine volume size\n");
break;
}
case Opt_errors:
@@ -287,8 +304,7 @@ static int parse_options(char *options, struct super_block *sb, s64 *newLVSize,
*flag &= ~JFS_ERR_REMOUNT_RO;
*flag |= JFS_ERR_PANIC;
} else {
- printk(KERN_ERR
- "JFS: %s is an invalid error handler\n",
+ pr_err("JFS: %s is an invalid error handler\n",
errors);
goto cleanup;
}
@@ -307,36 +323,89 @@ static int parse_options(char *options, struct super_block *sb, s64 *newLVSize,
case Opt_usrquota:
case Opt_grpquota:
case Opt_quota:
- printk(KERN_ERR
- "JFS: quota operations not supported\n");
+ pr_err("JFS: quota operations not supported\n");
break;
#endif
case Opt_uid:
{
char *uid = args[0].from;
- sbi->uid = simple_strtoul(uid, &uid, 0);
+ uid_t val;
+ int rc = kstrtouint(uid, 0, &val);
+
+ if (rc)
+ goto cleanup;
+ sbi->uid = make_kuid(current_user_ns(), val);
+ if (!uid_valid(sbi->uid))
+ goto cleanup;
break;
}
+
case Opt_gid:
{
char *gid = args[0].from;
- sbi->gid = simple_strtoul(gid, &gid, 0);
+ gid_t val;
+ int rc = kstrtouint(gid, 0, &val);
+
+ if (rc)
+ goto cleanup;
+ sbi->gid = make_kgid(current_user_ns(), val);
+ if (!gid_valid(sbi->gid))
+ goto cleanup;
break;
}
+
case Opt_umask:
{
char *umask = args[0].from;
- sbi->umask = simple_strtoul(umask, &umask, 8);
+ int rc = kstrtouint(umask, 8, &sbi->umask);
+
+ if (rc)
+ goto cleanup;
if (sbi->umask & ~0777) {
- printk(KERN_ERR
- "JFS: Invalid value of umask\n");
+ pr_err("JFS: Invalid value of umask\n");
goto cleanup;
}
break;
}
+
+ case Opt_discard:
+ {
+ struct request_queue *q = bdev_get_queue(sb->s_bdev);
+ /* if set to 1, even copying files will cause
+ * trimming :O
+ * -> user has more control over the online trimming
+ */
+ sbi->minblks_trim = 64;
+ if (blk_queue_discard(q))
+ *flag |= JFS_DISCARD;
+ else
+ pr_err("JFS: discard option not supported on device\n");
+ break;
+ }
+
+ case Opt_nodiscard:
+ *flag &= ~JFS_DISCARD;
+ break;
+
+ case Opt_discard_minblk:
+ {
+ struct request_queue *q = bdev_get_queue(sb->s_bdev);
+ char *minblks_trim = args[0].from;
+ int rc;
+ if (blk_queue_discard(q)) {
+ *flag |= JFS_DISCARD;
+ rc = kstrtouint(minblks_trim, 0,
+ &sbi->minblks_trim);
+ if (rc)
+ goto cleanup;
+ } else
+ pr_err("JFS: discard option not supported on device\n");
+ break;
+ }
+
default:
- printk("jfs: Unrecognized mount option \"%s\" "
- " or missing value\n", p);
+ printk("jfs: Unrecognized mount option \"%s\" or missing value\n",
+ p);
goto cleanup;
}
}
@@ -361,14 +430,13 @@ static int jfs_remount(struct super_block *sb, int *flags, char *data)
int flag = JFS_SBI(sb)->flag;
int ret;
- if (!parse_options(data, sb, &newLVSize, &flag)) {
+ sync_filesystem(sb);
+ if (!parse_options(data, sb, &newLVSize, &flag))
return -EINVAL;
- }
if (newLVSize) {
if (sb->s_flags & MS_RDONLY) {
- printk(KERN_ERR
- "JFS: resize requires volume to be mounted read-write\n");
+ pr_err("JFS: resize requires volume to be mounted read-write\n");
return -EROFS;
}
rc = jfs_extendfs(sb, newLVSize, 0);
@@ -394,9 +462,8 @@ static int jfs_remount(struct super_block *sb, int *flags, char *data)
}
if ((!(sb->s_flags & MS_RDONLY)) && (*flags & MS_RDONLY)) {
rc = dquot_suspend(sb, -1);
- if (rc < 0) {
+ if (rc < 0)
return rc;
- }
rc = jfs_umount_rw(sb);
JFS_SBI(sb)->flag = flag;
return rc;
@@ -429,13 +496,16 @@ static int jfs_fill_super(struct super_block *sb, void *data, int silent)
if (!new_valid_dev(sb->s_bdev->bd_dev))
return -EOVERFLOW;
- sbi = kzalloc(sizeof (struct jfs_sb_info), GFP_KERNEL);
+ sbi = kzalloc(sizeof(struct jfs_sb_info), GFP_KERNEL);
if (!sbi)
return -ENOMEM;
sb->s_fs_info = sbi;
+ sb->s_max_links = JFS_LINK_MAX;
sbi->sb = sb;
- sbi->uid = sbi->gid = sbi->umask = -1;
+ sbi->uid = INVALID_UID;
+ sbi->gid = INVALID_GID;
+ sbi->umask = -1;
/* initialize the mount flag and determine the default error handler */
flag = JFS_ERR_REMOUNT_RO;
@@ -449,7 +519,7 @@ static int jfs_fill_super(struct super_block *sb, void *data, int silent)
#endif
if (newLVSize) {
- printk(KERN_ERR "resize option for remount only\n");
+ pr_err("resize option for remount only\n");
goto out_kfree;
}
@@ -463,6 +533,7 @@ static int jfs_fill_super(struct super_block *sb, void *data, int silent)
*/
sb->s_op = &jfs_super_operations;
sb->s_export_op = &jfs_export_operations;
+ sb->s_xattr = jfs_xattr_handlers;
#ifdef CONFIG_QUOTA
sb->dq_op = &dquot_operations;
sb->s_qcop = &dquot_quotactl_ops;
@@ -477,7 +548,6 @@ static int jfs_fill_super(struct super_block *sb, void *data, int silent)
goto out_unload;
}
inode->i_ino = 0;
- inode->i_nlink = 1;
inode->i_size = sb->s_bdev->bd_inode->i_size;
inode->i_mapping->a_ops = &jfs_metapage_aops;
insert_inode_hash(inode);
@@ -487,9 +557,8 @@ static int jfs_fill_super(struct super_block *sb, void *data, int silent)
rc = jfs_mount(sb);
if (rc) {
- if (!silent) {
+ if (!silent)
jfs_err("jfs_mount failed w/return code = %d", rc);
- }
goto out_mount_failed;
}
if (sb->s_flags & MS_RDONLY)
@@ -507,18 +576,18 @@ static int jfs_fill_super(struct super_block *sb, void *data, int silent)
sb->s_magic = JFS_SUPER_MAGIC;
+ if (sbi->mntflag & JFS_OS2)
+ sb->s_d_op = &jfs_ci_dentry_operations;
+
inode = jfs_iget(sb, ROOT_I);
if (IS_ERR(inode)) {
ret = PTR_ERR(inode);
goto out_no_rw;
}
- sb->s_root = d_alloc_root(inode);
+ sb->s_root = d_make_root(inode);
if (!sb->s_root)
goto out_no_root;
- if (sbi->mntflag & JFS_OS2)
- sb->s_root->d_op = &jfs_ci_dentry_operations;
-
/* logical blocks are represented by 40 bits in pxd_t, etc. */
sb->s_maxbytes = ((u64) sb->s_blocksize) << 40;
#if BITS_PER_LONG == 32
@@ -526,20 +595,19 @@ static int jfs_fill_super(struct super_block *sb, void *data, int silent)
* Page cache is indexed by long.
* I would use MAX_LFS_FILESIZE, but it's only half as big
*/
- sb->s_maxbytes = min(((u64) PAGE_CACHE_SIZE << 32) - 1, (u64)sb->s_maxbytes);
+ sb->s_maxbytes = min(((u64) PAGE_CACHE_SIZE << 32) - 1,
+ (u64)sb->s_maxbytes);
#endif
sb->s_time_gran = 1;
return 0;
out_no_root:
jfs_err("jfs_read_super: get root dentry failed");
- iput(inode);
out_no_rw:
rc = jfs_umount(sb);
- if (rc) {
+ if (rc)
jfs_err("jfs_umount failed with return code %d", rc);
- }
out_mount_failed:
filemap_write_and_wait(sbi->direct_inode->i_mapping);
truncate_inode_pages(sbi->direct_inode->i_mapping, 0);
@@ -558,11 +626,28 @@ static int jfs_freeze(struct super_block *sb)
{
struct jfs_sb_info *sbi = JFS_SBI(sb);
struct jfs_log *log = sbi->log;
+ int rc = 0;
if (!(sb->s_flags & MS_RDONLY)) {
txQuiesce(sb);
- lmLogShutdown(log);
- updateSuper(sb, FM_CLEAN);
+ rc = lmLogShutdown(log);
+ if (rc) {
+ jfs_error(sb, "lmLogShutdown failed\n");
+
+ /* let operations fail rather than hang */
+ txResume(sb);
+
+ return rc;
+ }
+ rc = updateSuper(sb, FM_CLEAN);
+ if (rc) {
+ jfs_err("jfs_freeze: updateSuper failed\n");
+ /*
+ * Don't fail here. Everything succeeded except
+ * marking the superblock clean, so there's really
+ * no harm in leaving it frozen for now.
+ */
+ }
}
return 0;
}
@@ -574,13 +659,18 @@ static int jfs_unfreeze(struct super_block *sb)
int rc = 0;
if (!(sb->s_flags & MS_RDONLY)) {
- updateSuper(sb, FM_MOUNT);
- if ((rc = lmLogInit(log)))
- jfs_err("jfs_unlock failed with return code %d", rc);
- else
- txResume(sb);
+ rc = updateSuper(sb, FM_MOUNT);
+ if (rc) {
+ jfs_error(sb, "updateSuper failed\n");
+ goto out;
+ }
+ rc = lmLogInit(log);
+ if (rc)
+ jfs_error(sb, "lmLogInit failed\n");
+out:
+ txResume(sb);
}
- return 0;
+ return rc;
}
static struct dentry *jfs_do_mount(struct file_system_type *fs_type,
@@ -595,6 +685,11 @@ static int jfs_sync_fs(struct super_block *sb, int wait)
/* log == NULL indicates read-only mount */
if (log) {
+ /*
+ * Write quota structures to quota file, sync_blockdev() will
+ * write them to disk later
+ */
+ dquot_writeback_dquots(sb, -1);
jfs_flush_journal(log, wait);
jfs_syncpt(log, 0);
}
@@ -602,18 +697,20 @@ static int jfs_sync_fs(struct super_block *sb, int wait)
return 0;
}
-static int jfs_show_options(struct seq_file *seq, struct vfsmount *vfs)
+static int jfs_show_options(struct seq_file *seq, struct dentry *root)
{
- struct jfs_sb_info *sbi = JFS_SBI(vfs->mnt_sb);
+ struct jfs_sb_info *sbi = JFS_SBI(root->d_sb);
- if (sbi->uid != -1)
- seq_printf(seq, ",uid=%d", sbi->uid);
- if (sbi->gid != -1)
- seq_printf(seq, ",gid=%d", sbi->gid);
+ if (uid_valid(sbi->uid))
+ seq_printf(seq, ",uid=%d", from_kuid(&init_user_ns, sbi->uid));
+ if (gid_valid(sbi->gid))
+ seq_printf(seq, ",gid=%d", from_kgid(&init_user_ns, sbi->gid));
if (sbi->umask != -1)
seq_printf(seq, ",umask=%03o", sbi->umask);
if (sbi->flag & JFS_NOINTEGRITY)
seq_puts(seq, ",nointegrity");
+ if (sbi->flag & JFS_DISCARD)
+ seq_printf(seq, ",discard=%u", sbi->minblks_trim);
if (sbi->nls_tab)
seq_printf(seq, ",iocharset=%s", sbi->nls_tab->charset);
if (sbi->flag & JFS_ERR_CONTINUE)
@@ -636,7 +733,7 @@ static int jfs_show_options(struct seq_file *seq, struct vfsmount *vfs)
/* Read data from quotafile - avoid pagecache and such because we cannot afford
* acquiring the locks... As quota files are never truncated and quota code
- * itself serializes the operations (and noone else should touch the files)
+ * itself serializes the operations (and no one else should touch the files)
* we don't have to be afraid of races */
static ssize_t jfs_quota_read(struct super_block *sb, int type, char *data,
size_t len, loff_t off)
@@ -773,6 +870,7 @@ static struct file_system_type jfs_fs_type = {
.kill_sb = kill_block_super,
.fs_flags = FS_REQUIRES_DEV,
};
+MODULE_ALIAS_FS("jfs");
static void init_once(void *foo)
{
@@ -834,7 +932,8 @@ static int __init init_jfs_fs(void)
commit_threads = MAX_COMMIT_THREADS;
for (i = 0; i < commit_threads; i++) {
- jfsCommitThread[i] = kthread_run(jfs_lazycommit, NULL, "jfsCommit");
+ jfsCommitThread[i] = kthread_run(jfs_lazycommit, NULL,
+ "jfsCommit");
if (IS_ERR(jfsCommitThread[i])) {
rc = PTR_ERR(jfsCommitThread[i]);
jfs_err("init_jfs_fs: fork failed w/rc = %d", rc);
@@ -854,8 +953,14 @@ static int __init init_jfs_fs(void)
jfs_proc_init();
#endif
- return register_filesystem(&jfs_fs_type);
+ rc = register_filesystem(&jfs_fs_type);
+ if (!rc)
+ return 0;
+#ifdef PROC_FS_JFS
+ jfs_proc_clean();
+#endif
+ kthread_stop(jfsSyncThread);
kill_committask:
for (i = 0; i < commit_threads; i++)
kthread_stop(jfsCommitThread[i]);
@@ -886,6 +991,12 @@ static void __exit exit_jfs_fs(void)
jfs_proc_clean();
#endif
unregister_filesystem(&jfs_fs_type);
+
+ /*
+ * Make sure all delayed rcu free inodes are flushed before we
+ * destroy cache.
+ */
+ rcu_barrier();
kmem_cache_destroy(jfs_inode_cachep);
}
diff --git a/fs/jfs/xattr.c b/fs/jfs/xattr.c
index 2d7f165d0f1..46325d5c34f 100644
--- a/fs/jfs/xattr.c
+++ b/fs/jfs/xattr.c
@@ -382,7 +382,7 @@ static int ea_read(struct inode *ip, struct jfs_ea_list *ealist)
nbytes = sizeDXD(&ji->ea);
if (!nbytes) {
- jfs_error(sb, "ea_read: nbytes is 0");
+ jfs_error(sb, "nbytes is 0\n");
return -EIO;
}
@@ -482,7 +482,7 @@ static int ea_get(struct inode *inode, struct ea_buffer *ea_buf, int min_size)
current_blocks = 0;
} else {
if (!(ji->ea.flag & DXD_EXTENT)) {
- jfs_error(sb, "ea_get: invalid ea.flag)");
+ jfs_error(sb, "invalid ea.flag\n");
return -EIO;
}
current_blocks = (ea_size + sb->s_blocksize - 1) >>
@@ -666,83 +666,12 @@ static int ea_put(tid_t tid, struct inode *inode, struct ea_buffer *ea_buf,
}
/*
- * can_set_system_xattr
- *
- * This code is specific to the system.* namespace. It contains policy
- * which doesn't belong in the main xattr codepath.
- */
-static int can_set_system_xattr(struct inode *inode, const char *name,
- const void *value, size_t value_len)
-{
-#ifdef CONFIG_JFS_POSIX_ACL
- struct posix_acl *acl;
- int rc;
-
- if (!is_owner_or_cap(inode))
- return -EPERM;
-
- /*
- * POSIX_ACL_XATTR_ACCESS is tied to i_mode
- */
- if (strcmp(name, POSIX_ACL_XATTR_ACCESS) == 0) {
- acl = posix_acl_from_xattr(value, value_len);
- if (IS_ERR(acl)) {
- rc = PTR_ERR(acl);
- printk(KERN_ERR "posix_acl_from_xattr returned %d\n",
- rc);
- return rc;
- }
- if (acl) {
- mode_t mode = inode->i_mode;
- rc = posix_acl_equiv_mode(acl, &mode);
- posix_acl_release(acl);
- if (rc < 0) {
- printk(KERN_ERR
- "posix_acl_equiv_mode returned %d\n",
- rc);
- return rc;
- }
- inode->i_mode = mode;
- mark_inode_dirty(inode);
- }
- /*
- * We're changing the ACL. Get rid of the cached one
- */
- forget_cached_acl(inode, ACL_TYPE_ACCESS);
-
- return 0;
- } else if (strcmp(name, POSIX_ACL_XATTR_DEFAULT) == 0) {
- acl = posix_acl_from_xattr(value, value_len);
- if (IS_ERR(acl)) {
- rc = PTR_ERR(acl);
- printk(KERN_ERR "posix_acl_from_xattr returned %d\n",
- rc);
- return rc;
- }
- posix_acl_release(acl);
-
- /*
- * We're changing the default ACL. Get rid of the cached one
- */
- forget_cached_acl(inode, ACL_TYPE_DEFAULT);
-
- return 0;
- }
-#endif /* CONFIG_JFS_POSIX_ACL */
- return -EOPNOTSUPP;
-}
-
-/*
* Most of the permission checking is done by xattr_permission in the vfs.
- * The local file system is responsible for handling the system.* namespace.
* We also need to verify that this is a namespace that we recognize.
*/
static int can_set_xattr(struct inode *inode, const char *name,
const void *value, size_t value_len)
{
- if (!strncmp(name, XATTR_SYSTEM_PREFIX, XATTR_SYSTEM_PREFIX_LEN))
- return can_set_system_xattr(inode, name, value, value_len);
-
if (!strncmp(name, XATTR_OS2_PREFIX, XATTR_OS2_PREFIX_LEN)) {
/*
* This makes sure that we aren't trying to set an
@@ -750,7 +679,7 @@ static int can_set_xattr(struct inode *inode, const char *name,
* with "os2."
*/
if (is_known_namespace(name + XATTR_OS2_PREFIX_LEN))
- return -EOPNOTSUPP;
+ return -EOPNOTSUPP;
return 0;
}
@@ -862,6 +791,19 @@ int __jfs_setxattr(tid_t tid, struct inode *inode, const char *name,
/* Completely new ea list */
xattr_size = sizeof (struct jfs_ea_list);
+ /*
+ * The size of EA value is limitted by on-disk format up to
+ * __le16, there would be an overflow if the size is equal
+ * to XATTR_SIZE_MAX (65536). In order to avoid this issue,
+ * we can pre-checkup the value size against USHRT_MAX, and
+ * return -E2BIG in this case, which is consistent with the
+ * VFS setxattr interface.
+ */
+ if (value_len >= USHRT_MAX) {
+ rc = -E2BIG;
+ goto release;
+ }
+
ea = (struct jfs_ea *) ((char *) ealist + xattr_size);
ea->flag = 0;
ea->namelen = namelen;
@@ -876,7 +818,7 @@ int __jfs_setxattr(tid_t tid, struct inode *inode, const char *name,
/* DEBUG - If we did this right, these number match */
if (xattr_size != new_size) {
printk(KERN_ERR
- "jfs_xsetattr: xattr_size = %d, new_size = %d\n",
+ "__jfs_setxattr: xattr_size = %d, new_size = %d\n",
xattr_size, new_size);
rc = -EINVAL;
@@ -912,6 +854,14 @@ int jfs_setxattr(struct dentry *dentry, const char *name, const void *value,
int rc;
tid_t tid;
+ /*
+ * If this is a request for a synthetic attribute in the system.*
+ * namespace use the generic infrastructure to resolve a handler
+ * for it via sb->s_xattr.
+ */
+ if (!strncmp(name, XATTR_SYSTEM_PREFIX, XATTR_SYSTEM_PREFIX_LEN))
+ return generic_setxattr(dentry, name, value, value_len, flags);
+
if ((rc = can_set_xattr(inode, name, value, value_len)))
return rc;
@@ -988,6 +938,14 @@ ssize_t jfs_getxattr(struct dentry *dentry, const char *name, void *data,
{
int err;
+ /*
+ * If this is a request for a synthetic attribute in the system.*
+ * namespace use the generic infrastructure to resolve a handler
+ * for it via sb->s_xattr.
+ */
+ if (!strncmp(name, XATTR_SYSTEM_PREFIX, XATTR_SYSTEM_PREFIX_LEN))
+ return generic_getxattr(dentry, name, data, buf_size);
+
if (strncmp(name, XATTR_OS2_PREFIX, XATTR_OS2_PREFIX_LEN) == 0) {
/*
* skip past "os2." prefix
@@ -1076,6 +1034,14 @@ int jfs_removexattr(struct dentry *dentry, const char *name)
int rc;
tid_t tid;
+ /*
+ * If this is a request for a synthetic attribute in the system.*
+ * namespace use the generic infrastructure to resolve a handler
+ * for it via sb->s_xattr.
+ */
+ if (!strncmp(name, XATTR_SYSTEM_PREFIX, XATTR_SYSTEM_PREFIX_LEN))
+ return generic_removexattr(dentry, name);
+
if ((rc = can_set_xattr(inode, name, NULL, 0)))
return rc;
@@ -1090,37 +1056,51 @@ int jfs_removexattr(struct dentry *dentry, const char *name)
return rc;
}
+/*
+ * List of handlers for synthetic system.* attributes. All real ondisk
+ * attributes are handled directly.
+ */
+const struct xattr_handler *jfs_xattr_handlers[] = {
+#ifdef CONFIG_JFS_POSIX_ACL
+ &posix_acl_access_xattr_handler,
+ &posix_acl_default_xattr_handler,
+#endif
+ NULL,
+};
+
+
#ifdef CONFIG_JFS_SECURITY
-int jfs_init_security(tid_t tid, struct inode *inode, struct inode *dir)
+static int jfs_initxattrs(struct inode *inode, const struct xattr *xattr_array,
+ void *fs_info)
{
- int rc;
- size_t len;
- void *value;
- char *suffix;
+ const struct xattr *xattr;
+ tid_t *tid = fs_info;
char *name;
-
- rc = security_inode_init_security(inode, dir, &suffix, &value, &len);
- if (rc) {
- if (rc == -EOPNOTSUPP)
- return 0;
- return rc;
- }
- name = kmalloc(XATTR_SECURITY_PREFIX_LEN + 1 + strlen(suffix),
- GFP_NOFS);
- if (!name) {
- rc = -ENOMEM;
- goto kmalloc_failed;
+ int err = 0;
+
+ for (xattr = xattr_array; xattr->name != NULL; xattr++) {
+ name = kmalloc(XATTR_SECURITY_PREFIX_LEN +
+ strlen(xattr->name) + 1, GFP_NOFS);
+ if (!name) {
+ err = -ENOMEM;
+ break;
+ }
+ strcpy(name, XATTR_SECURITY_PREFIX);
+ strcpy(name + XATTR_SECURITY_PREFIX_LEN, xattr->name);
+
+ err = __jfs_setxattr(*tid, inode, name,
+ xattr->value, xattr->value_len, 0);
+ kfree(name);
+ if (err < 0)
+ break;
}
- strcpy(name, XATTR_SECURITY_PREFIX);
- strcpy(name + XATTR_SECURITY_PREFIX_LEN, suffix);
-
- rc = __jfs_setxattr(tid, inode, name, value, len, 0);
-
- kfree(name);
-kmalloc_failed:
- kfree(suffix);
- kfree(value);
+ return err;
+}
- return rc;
+int jfs_init_security(tid_t tid, struct inode *inode, struct inode *dir,
+ const struct qstr *qstr)
+{
+ return security_inode_init_security(inode, dir, qstr,
+ &jfs_initxattrs, &tid);
}
#endif
generated by cgit v1.2.3-70-g09d2 (git 2.46.0) at 2025-11-04 08:27:54 +0000