aboutsummaryrefslogtreecommitdiff
path: root/fs
diff options
context:
space:
mode:
Diffstat (limited to 'fs')
-rw-r--r--fs/9p/v9fs_vfs.h23
-rw-r--r--fs/9p/vfs_file.c6
-rw-r--r--fs/9p/vfs_inode.c23
-rw-r--r--fs/9p/vfs_inode_dotl.c27
-rw-r--r--fs/9p/vfs_super.c4
-rw-r--r--fs/btrfs/extent_io.c4
-rw-r--r--fs/btrfs/volumes.c4
-rw-r--r--fs/cifs/file.c12
-rw-r--r--fs/cifs/smb2misc.c17
-rw-r--r--fs/cifs/smb2ops.c15
-rw-r--r--fs/devpts/inode.c1
-rw-r--r--fs/ext2/super.c39
-rw-r--r--fs/ext4/resize.c3
-rw-r--r--fs/jbd2/transaction.c33
-rw-r--r--fs/nfs/pagelist.c29
-rw-r--r--fs/nfsd/nfs3proc.c16
-rw-r--r--fs/nfsd/nfs3xdr.c1
-rw-r--r--fs/nfsd/nfsctl.c2
-rw-r--r--fs/pipe.c14
-rw-r--r--fs/splice.c4
20 files changed, 194 insertions, 83 deletions
diff --git a/fs/9p/v9fs_vfs.h b/fs/9p/v9fs_vfs.h
index 5a0db6dec8d1..aaee1e6584e6 100644
--- a/fs/9p/v9fs_vfs.h
+++ b/fs/9p/v9fs_vfs.h
@@ -40,6 +40,9 @@
*/
#define P9_LOCK_TIMEOUT (30*HZ)
+/* flags for v9fs_stat2inode() & v9fs_stat2inode_dotl() */
+#define V9FS_STAT2INODE_KEEP_ISIZE 1
+
extern struct file_system_type v9fs_fs_type;
extern const struct address_space_operations v9fs_addr_operations;
extern const struct file_operations v9fs_file_operations;
@@ -61,8 +64,10 @@ int v9fs_init_inode(struct v9fs_session_info *v9ses,
struct inode *inode, umode_t mode, dev_t);
void v9fs_evict_inode(struct inode *inode);
ino_t v9fs_qid2ino(struct p9_qid *qid);
-void v9fs_stat2inode(struct p9_wstat *, struct inode *, struct super_block *);
-void v9fs_stat2inode_dotl(struct p9_stat_dotl *, struct inode *);
+void v9fs_stat2inode(struct p9_wstat *stat, struct inode *inode,
+ struct super_block *sb, unsigned int flags);
+void v9fs_stat2inode_dotl(struct p9_stat_dotl *stat, struct inode *inode,
+ unsigned int flags);
int v9fs_dir_release(struct inode *inode, struct file *filp);
int v9fs_file_open(struct inode *inode, struct file *file);
void v9fs_inode2stat(struct inode *inode, struct p9_wstat *stat);
@@ -83,4 +88,18 @@ static inline void v9fs_invalidate_inode_attr(struct inode *inode)
}
int v9fs_open_to_dotl_flags(int flags);
+
+static inline void v9fs_i_size_write(struct inode *inode, loff_t i_size)
+{
+ /*
+ * 32-bit need the lock, concurrent updates could break the
+ * sequences and make i_size_read() loop forever.
+ * 64-bit updates are atomic and can skip the locking.
+ */
+ if (sizeof(i_size) > sizeof(long))
+ spin_lock(&inode->i_lock);
+ i_size_write(inode, i_size);
+ if (sizeof(i_size) > sizeof(long))
+ spin_unlock(&inode->i_lock);
+}
#endif
diff --git a/fs/9p/vfs_file.c b/fs/9p/vfs_file.c
index 398a3eddb2df..2f035b15180e 100644
--- a/fs/9p/vfs_file.c
+++ b/fs/9p/vfs_file.c
@@ -442,7 +442,11 @@ v9fs_file_write_iter(struct kiocb *iocb, struct iov_iter *from)
i_size = i_size_read(inode);
if (iocb->ki_pos > i_size) {
inode_add_bytes(inode, iocb->ki_pos - i_size);
- i_size_write(inode, iocb->ki_pos);
+ /*
+ * Need to serialize against i_size_write() in
+ * v9fs_stat2inode()
+ */
+ v9fs_i_size_write(inode, iocb->ki_pos);
}
return retval;
}
diff --git a/fs/9p/vfs_inode.c b/fs/9p/vfs_inode.c
index f8ab4a66acaf..ddd1eb6aedee 100644
--- a/fs/9p/vfs_inode.c
+++ b/fs/9p/vfs_inode.c
@@ -538,7 +538,7 @@ static struct inode *v9fs_qid_iget(struct super_block *sb,
if (retval)
goto error;
- v9fs_stat2inode(st, inode, sb);
+ v9fs_stat2inode(st, inode, sb, 0);
v9fs_cache_inode_get_cookie(inode);
unlock_new_inode(inode);
return inode;
@@ -1078,7 +1078,7 @@ v9fs_vfs_getattr(struct vfsmount *mnt, struct dentry *dentry,
if (IS_ERR(st))
return PTR_ERR(st);
- v9fs_stat2inode(st, d_inode(dentry), dentry->d_sb);
+ v9fs_stat2inode(st, d_inode(dentry), dentry->d_sb, 0);
generic_fillattr(d_inode(dentry), stat);
p9stat_free(st);
@@ -1156,12 +1156,13 @@ static int v9fs_vfs_setattr(struct dentry *dentry, struct iattr *iattr)
* @stat: Plan 9 metadata (mistat) structure
* @inode: inode to populate
* @sb: superblock of filesystem
+ * @flags: control flags (e.g. V9FS_STAT2INODE_KEEP_ISIZE)
*
*/
void
v9fs_stat2inode(struct p9_wstat *stat, struct inode *inode,
- struct super_block *sb)
+ struct super_block *sb, unsigned int flags)
{
umode_t mode;
char ext[32];
@@ -1202,10 +1203,11 @@ v9fs_stat2inode(struct p9_wstat *stat, struct inode *inode,
mode = p9mode2perm(v9ses, stat);
mode |= inode->i_mode & ~S_IALLUGO;
inode->i_mode = mode;
- i_size_write(inode, stat->length);
+ if (!(flags & V9FS_STAT2INODE_KEEP_ISIZE))
+ v9fs_i_size_write(inode, stat->length);
/* not real number of blocks, but 512 byte ones ... */
- inode->i_blocks = (i_size_read(inode) + 512 - 1) >> 9;
+ inode->i_blocks = (stat->length + 512 - 1) >> 9;
v9inode->cache_validity &= ~V9FS_INO_INVALID_ATTR;
}
@@ -1402,9 +1404,9 @@ int v9fs_refresh_inode(struct p9_fid *fid, struct inode *inode)
{
int umode;
dev_t rdev;
- loff_t i_size;
struct p9_wstat *st;
struct v9fs_session_info *v9ses;
+ unsigned int flags;
v9ses = v9fs_inode2v9ses(inode);
st = p9_client_stat(fid);
@@ -1417,16 +1419,13 @@ int v9fs_refresh_inode(struct p9_fid *fid, struct inode *inode)
if ((inode->i_mode & S_IFMT) != (umode & S_IFMT))
goto out;
- spin_lock(&inode->i_lock);
/*
* We don't want to refresh inode->i_size,
* because we may have cached data
*/
- i_size = inode->i_size;
- v9fs_stat2inode(st, inode, inode->i_sb);
- if (v9ses->cache == CACHE_LOOSE || v9ses->cache == CACHE_FSCACHE)
- inode->i_size = i_size;
- spin_unlock(&inode->i_lock);
+ flags = (v9ses->cache == CACHE_LOOSE || v9ses->cache == CACHE_FSCACHE) ?
+ V9FS_STAT2INODE_KEEP_ISIZE : 0;
+ v9fs_stat2inode(st, inode, inode->i_sb, flags);
out:
p9stat_free(st);
kfree(st);
diff --git a/fs/9p/vfs_inode_dotl.c b/fs/9p/vfs_inode_dotl.c
index c3dd0d42bb3a..425bc1a2b8c1 100644
--- a/fs/9p/vfs_inode_dotl.c
+++ b/fs/9p/vfs_inode_dotl.c
@@ -143,7 +143,7 @@ static struct inode *v9fs_qid_iget_dotl(struct super_block *sb,
if (retval)
goto error;
- v9fs_stat2inode_dotl(st, inode);
+ v9fs_stat2inode_dotl(st, inode, 0);
v9fs_cache_inode_get_cookie(inode);
retval = v9fs_get_acl(inode, fid);
if (retval)
@@ -496,7 +496,7 @@ v9fs_vfs_getattr_dotl(struct vfsmount *mnt, struct dentry *dentry,
if (IS_ERR(st))
return PTR_ERR(st);
- v9fs_stat2inode_dotl(st, d_inode(dentry));
+ v9fs_stat2inode_dotl(st, d_inode(dentry), 0);
generic_fillattr(d_inode(dentry), stat);
/* Change block size to what the server returned */
stat->blksize = st->st_blksize;
@@ -607,11 +607,13 @@ int v9fs_vfs_setattr_dotl(struct dentry *dentry, struct iattr *iattr)
* v9fs_stat2inode_dotl - populate an inode structure with stat info
* @stat: stat structure
* @inode: inode to populate
+ * @flags: ctrl flags (e.g. V9FS_STAT2INODE_KEEP_ISIZE)
*
*/
void
-v9fs_stat2inode_dotl(struct p9_stat_dotl *stat, struct inode *inode)
+v9fs_stat2inode_dotl(struct p9_stat_dotl *stat, struct inode *inode,
+ unsigned int flags)
{
umode_t mode;
struct v9fs_inode *v9inode = V9FS_I(inode);
@@ -631,7 +633,8 @@ v9fs_stat2inode_dotl(struct p9_stat_dotl *stat, struct inode *inode)
mode |= inode->i_mode & ~S_IALLUGO;
inode->i_mode = mode;
- i_size_write(inode, stat->st_size);
+ if (!(flags & V9FS_STAT2INODE_KEEP_ISIZE))
+ v9fs_i_size_write(inode, stat->st_size);
inode->i_blocks = stat->st_blocks;
} else {
if (stat->st_result_mask & P9_STATS_ATIME) {
@@ -661,8 +664,9 @@ v9fs_stat2inode_dotl(struct p9_stat_dotl *stat, struct inode *inode)
}
if (stat->st_result_mask & P9_STATS_RDEV)
inode->i_rdev = new_decode_dev(stat->st_rdev);
- if (stat->st_result_mask & P9_STATS_SIZE)
- i_size_write(inode, stat->st_size);
+ if (!(flags & V9FS_STAT2INODE_KEEP_ISIZE) &&
+ stat->st_result_mask & P9_STATS_SIZE)
+ v9fs_i_size_write(inode, stat->st_size);
if (stat->st_result_mask & P9_STATS_BLOCKS)
inode->i_blocks = stat->st_blocks;
}
@@ -928,9 +932,9 @@ v9fs_vfs_get_link_dotl(struct dentry *dentry,
int v9fs_refresh_inode_dotl(struct p9_fid *fid, struct inode *inode)
{
- loff_t i_size;
struct p9_stat_dotl *st;
struct v9fs_session_info *v9ses;
+ unsigned int flags;
v9ses = v9fs_inode2v9ses(inode);
st = p9_client_getattr_dotl(fid, P9_STATS_ALL);
@@ -942,16 +946,13 @@ int v9fs_refresh_inode_dotl(struct p9_fid *fid, struct inode *inode)
if ((inode->i_mode & S_IFMT) != (st->st_mode & S_IFMT))
goto out;
- spin_lock(&inode->i_lock);
/*
* We don't want to refresh inode->i_size,
* because we may have cached data
*/
- i_size = inode->i_size;
- v9fs_stat2inode_dotl(st, inode);
- if (v9ses->cache == CACHE_LOOSE || v9ses->cache == CACHE_FSCACHE)
- inode->i_size = i_size;
- spin_unlock(&inode->i_lock);
+ flags = (v9ses->cache == CACHE_LOOSE || v9ses->cache == CACHE_FSCACHE) ?
+ V9FS_STAT2INODE_KEEP_ISIZE : 0;
+ v9fs_stat2inode_dotl(st, inode, flags);
out:
kfree(st);
return 0;
diff --git a/fs/9p/vfs_super.c b/fs/9p/vfs_super.c
index de3ed8629196..06ff1a9089e8 100644
--- a/fs/9p/vfs_super.c
+++ b/fs/9p/vfs_super.c
@@ -165,7 +165,7 @@ static struct dentry *v9fs_mount(struct file_system_type *fs_type, int flags,
goto release_sb;
}
d_inode(root)->i_ino = v9fs_qid2ino(&st->qid);
- v9fs_stat2inode_dotl(st, d_inode(root));
+ v9fs_stat2inode_dotl(st, d_inode(root), 0);
kfree(st);
} else {
struct p9_wstat *st = NULL;
@@ -176,7 +176,7 @@ static struct dentry *v9fs_mount(struct file_system_type *fs_type, int flags,
}
d_inode(root)->i_ino = v9fs_qid2ino(&st->qid);
- v9fs_stat2inode(st, d_inode(root), sb);
+ v9fs_stat2inode(st, d_inode(root), sb, 0);
p9stat_free(st);
kfree(st);
diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c
index f605bfa96905..84051b229fd2 100644
--- a/fs/btrfs/extent_io.c
+++ b/fs/btrfs/extent_io.c
@@ -3018,11 +3018,11 @@ static int __do_readpage(struct extent_io_tree *tree,
*/
if (test_bit(EXTENT_FLAG_COMPRESSED, &em->flags) &&
prev_em_start && *prev_em_start != (u64)-1 &&
- *prev_em_start != em->orig_start)
+ *prev_em_start != em->start)
force_bio_submit = true;
if (prev_em_start)
- *prev_em_start = em->orig_start;
+ *prev_em_start = em->start;
free_extent_map(em);
em = NULL;
diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c
index 1fe447e3ed8f..dbb78c31a9fb 100644
--- a/fs/btrfs/volumes.c
+++ b/fs/btrfs/volumes.c
@@ -6439,10 +6439,10 @@ static int btrfs_check_chunk_valid(struct btrfs_root *root,
}
if ((type & BTRFS_BLOCK_GROUP_RAID10 && sub_stripes != 2) ||
- (type & BTRFS_BLOCK_GROUP_RAID1 && num_stripes < 1) ||
+ (type & BTRFS_BLOCK_GROUP_RAID1 && num_stripes != 2) ||
(type & BTRFS_BLOCK_GROUP_RAID5 && num_stripes < 2) ||
(type & BTRFS_BLOCK_GROUP_RAID6 && num_stripes < 3) ||
- (type & BTRFS_BLOCK_GROUP_DUP && num_stripes > 2) ||
+ (type & BTRFS_BLOCK_GROUP_DUP && num_stripes != 2) ||
((type & BTRFS_BLOCK_GROUP_PROFILE_MASK) == 0 &&
num_stripes != 1)) {
btrfs_err(root->fs_info,
diff --git a/fs/cifs/file.c b/fs/cifs/file.c
index 8ec296308729..1c5099fffaec 100644
--- a/fs/cifs/file.c
+++ b/fs/cifs/file.c
@@ -2797,14 +2797,16 @@ cifs_strict_writev(struct kiocb *iocb, struct iov_iter *from)
* these pages but not on the region from pos to ppos+len-1.
*/
written = cifs_user_writev(iocb, from);
- if (written > 0 && CIFS_CACHE_READ(cinode)) {
+ if (CIFS_CACHE_READ(cinode)) {
/*
- * Windows 7 server can delay breaking level2 oplock if a write
- * request comes - break it on the client to prevent reading
- * an old data.
+ * We have read level caching and we have just sent a write
+ * request to the server thus making data in the cache stale.
+ * Zap the cache and set oplock/lease level to NONE to avoid
+ * reading stale data from the cache. All subsequent read
+ * operations will read new data from the server.
*/
cifs_zap_mapping(inode);
- cifs_dbg(FYI, "Set no oplock for inode=%p after a write operation\n",
+ cifs_dbg(FYI, "Set Oplock/Lease to NONE for inode=%p after write\n",
inode);
cinode->oplock = 0;
}
diff --git a/fs/cifs/smb2misc.c b/fs/cifs/smb2misc.c
index e96a74da756f..244d27bb8fba 100644
--- a/fs/cifs/smb2misc.c
+++ b/fs/cifs/smb2misc.c
@@ -474,7 +474,6 @@ smb2_tcon_has_lease(struct cifs_tcon *tcon, struct smb2_lease_break *rsp,
__u8 lease_state;
struct list_head *tmp;
struct cifsFileInfo *cfile;
- struct TCP_Server_Info *server = tcon->ses->server;
struct cifs_pending_open *open;
struct cifsInodeInfo *cinode;
int ack_req = le32_to_cpu(rsp->Flags &
@@ -494,13 +493,25 @@ smb2_tcon_has_lease(struct cifs_tcon *tcon, struct smb2_lease_break *rsp,
cifs_dbg(FYI, "lease key match, lease break 0x%x\n",
le32_to_cpu(rsp->NewLeaseState));
- server->ops->set_oplock_level(cinode, lease_state, 0, NULL);
-
if (ack_req)
cfile->oplock_break_cancelled = false;
else
cfile->oplock_break_cancelled = true;
+ set_bit(CIFS_INODE_PENDING_OPLOCK_BREAK, &cinode->flags);
+
+ /*
+ * Set or clear flags depending on the lease state being READ.
+ * HANDLE caching flag should be added when the client starts
+ * to defer closing remote file handles with HANDLE leases.
+ */
+ if (lease_state & SMB2_LEASE_READ_CACHING_HE)
+ set_bit(CIFS_INODE_DOWNGRADE_OPLOCK_TO_L2,
+ &cinode->flags);
+ else
+ clear_bit(CIFS_INODE_DOWNGRADE_OPLOCK_TO_L2,
+ &cinode->flags);
+
queue_work(cifsoplockd_wq, &cfile->oplock_break);
kfree(lw);
return true;
diff --git a/fs/cifs/smb2ops.c b/fs/cifs/smb2ops.c
index 2db7968febfe..97d8e2a3df9b 100644
--- a/fs/cifs/smb2ops.c
+++ b/fs/cifs/smb2ops.c
@@ -1377,6 +1377,15 @@ smb2_downgrade_oplock(struct TCP_Server_Info *server,
}
static void
+smb21_downgrade_oplock(struct TCP_Server_Info *server,
+ struct cifsInodeInfo *cinode, bool set_level2)
+{
+ server->ops->set_oplock_level(cinode,
+ set_level2 ? SMB2_LEASE_READ_CACHING_HE :
+ 0, 0, NULL);
+}
+
+static void
smb2_set_oplock_level(struct cifsInodeInfo *cinode, __u32 oplock,
unsigned int epoch, bool *purge_cache)
{
@@ -1681,7 +1690,7 @@ struct smb_version_operations smb21_operations = {
.print_stats = smb2_print_stats,
.is_oplock_break = smb2_is_valid_oplock_break,
.handle_cancelled_mid = smb2_handle_cancelled_mid,
- .downgrade_oplock = smb2_downgrade_oplock,
+ .downgrade_oplock = smb21_downgrade_oplock,
.need_neg = smb2_need_neg,
.negotiate = smb2_negotiate,
.negotiate_wsize = smb2_negotiate_wsize,
@@ -1765,7 +1774,7 @@ struct smb_version_operations smb30_operations = {
.dump_share_caps = smb2_dump_share_caps,
.is_oplock_break = smb2_is_valid_oplock_break,
.handle_cancelled_mid = smb2_handle_cancelled_mid,
- .downgrade_oplock = smb2_downgrade_oplock,
+ .downgrade_oplock = smb21_downgrade_oplock,
.need_neg = smb2_need_neg,
.negotiate = smb2_negotiate,
.negotiate_wsize = smb2_negotiate_wsize,
@@ -1855,7 +1864,7 @@ struct smb_version_operations smb311_operations = {
.dump_share_caps = smb2_dump_share_caps,
.is_oplock_break = smb2_is_valid_oplock_break,
.handle_cancelled_mid = smb2_handle_cancelled_mid,
- .downgrade_oplock = smb2_downgrade_oplock,
+ .downgrade_oplock = smb21_downgrade_oplock,
.need_neg = smb2_need_neg,
.negotiate = smb2_negotiate,
.negotiate_wsize = smb2_negotiate_wsize,
diff --git a/fs/devpts/inode.c b/fs/devpts/inode.c
index 108df2e3602c..81be3ba17cbf 100644
--- a/fs/devpts/inode.c
+++ b/fs/devpts/inode.c
@@ -396,6 +396,7 @@ devpts_fill_super(struct super_block *s, void *data, int silent)
s->s_blocksize_bits = 10;
s->s_magic = DEVPTS_SUPER_MAGIC;
s->s_op = &devpts_sops;
+ s->s_d_op = &simple_dentry_operations;
s->s_time_gran = 1;
error = -ENOMEM;
diff --git a/fs/ext2/super.c b/fs/ext2/super.c
index 4afc22a5db16..ca1289103449 100644
--- a/fs/ext2/super.c
+++ b/fs/ext2/super.c
@@ -724,7 +724,8 @@ static loff_t ext2_max_size(int bits)
{
loff_t res = EXT2_NDIR_BLOCKS;
int meta_blocks;
- loff_t upper_limit;
+ unsigned int upper_limit;
+ unsigned int ppb = 1 << (bits-2);
/* This is calculated to be the largest file size for a
* dense, file such that the total number of
@@ -738,24 +739,34 @@ static loff_t ext2_max_size(int bits)
/* total blocks in file system block size */
upper_limit >>= (bits - 9);
+ /* Compute how many blocks we can address by block tree */
+ res += 1LL << (bits-2);
+ res += 1LL << (2*(bits-2));
+ res += 1LL << (3*(bits-2));
+ /* Does block tree limit file size? */
+ if (res < upper_limit)
+ goto check_lfs;
+ res = upper_limit;
+ /* How many metadata blocks are needed for addressing upper_limit? */
+ upper_limit -= EXT2_NDIR_BLOCKS;
/* indirect blocks */
meta_blocks = 1;
+ upper_limit -= ppb;
/* double indirect blocks */
- meta_blocks += 1 + (1LL << (bits-2));
- /* tripple indirect blocks */
- meta_blocks += 1 + (1LL << (bits-2)) + (1LL << (2*(bits-2)));
-
- upper_limit -= meta_blocks;
- upper_limit <<= bits;
-
- res += 1LL << (bits-2);
- res += 1LL << (2*(bits-2));
- res += 1LL << (3*(bits-2));
+ if (upper_limit < ppb * ppb) {
+ meta_blocks += 1 + DIV_ROUND_UP(upper_limit, ppb);
+ res -= meta_blocks;
+ goto check_lfs;
+ }
+ meta_blocks += 1 + ppb;
+ upper_limit -= ppb * ppb;
+ /* tripple indirect blocks for the rest */
+ meta_blocks += 1 + DIV_ROUND_UP(upper_limit, ppb) +
+ DIV_ROUND_UP(upper_limit, ppb*ppb);
+ res -= meta_blocks;
+check_lfs:
res <<= bits;
- if (res > upper_limit)
- res = upper_limit;
-
if (res > MAX_LFS_FILESIZE)
res = MAX_LFS_FILESIZE;
diff --git a/fs/ext4/resize.c b/fs/ext4/resize.c
index d80d93fbe4bc..4c9da0ab4dfd 100644
--- a/fs/ext4/resize.c
+++ b/fs/ext4/resize.c
@@ -1930,7 +1930,8 @@ retry:
le16_to_cpu(es->s_reserved_gdt_blocks);
n_group = n_desc_blocks * EXT4_DESC_PER_BLOCK(sb);
n_blocks_count = (ext4_fsblk_t)n_group *
- EXT4_BLOCKS_PER_GROUP(sb);
+ EXT4_BLOCKS_PER_GROUP(sb) +
+ le32_to_cpu(es->s_first_data_block);
n_group--; /* set to last group number */
}
diff --git a/fs/jbd2/transaction.c b/fs/jbd2/transaction.c
index b320c1ba7fdc..799f96c67211 100644
--- a/fs/jbd2/transaction.c
+++ b/fs/jbd2/transaction.c
@@ -1211,11 +1211,12 @@ int jbd2_journal_get_undo_access(handle_t *handle, struct buffer_head *bh)
struct journal_head *jh;
char *committed_data = NULL;
- JBUFFER_TRACE(jh, "entry");
if (jbd2_write_access_granted(handle, bh, true))
return 0;
jh = jbd2_journal_add_journal_head(bh);
+ JBUFFER_TRACE(jh, "entry");
+
/*
* Do this first --- it can drop the journal lock, so we want to
* make sure that obtaining the committed_data is done
@@ -1326,15 +1327,17 @@ int jbd2_journal_dirty_metadata(handle_t *handle, struct buffer_head *bh)
if (is_handle_aborted(handle))
return -EROFS;
- if (!buffer_jbd(bh)) {
- ret = -EUCLEAN;
- goto out;
- }
+ if (!buffer_jbd(bh))
+ return -EUCLEAN;
+
/*
* We don't grab jh reference here since the buffer must be part
* of the running transaction.
*/
jh = bh2jh(bh);
+ jbd_debug(5, "journal_head %p\n", jh);
+ JBUFFER_TRACE(jh, "entry");
+
/*
* This and the following assertions are unreliable since we may see jh
* in inconsistent state unless we grab bh_state lock. But this is
@@ -1368,9 +1371,6 @@ int jbd2_journal_dirty_metadata(handle_t *handle, struct buffer_head *bh)
}
journal = transaction->t_journal;
- jbd_debug(5, "journal_head %p\n", jh);
- JBUFFER_TRACE(jh, "entry");
-
jbd_lock_bh_state(bh);
if (jh->b_modified == 0) {
@@ -1568,14 +1568,21 @@ int jbd2_journal_forget (handle_t *handle, struct buffer_head *bh)
/* However, if the buffer is still owned by a prior
* (committing) transaction, we can't drop it yet... */
JBUFFER_TRACE(jh, "belongs to older transaction");
- /* ... but we CAN drop it from the new transaction if we
- * have also modified it since the original commit. */
+ /* ... but we CAN drop it from the new transaction through
+ * marking the buffer as freed and set j_next_transaction to
+ * the new transaction, so that not only the commit code
+ * knows it should clear dirty bits when it is done with the
+ * buffer, but also the buffer can be checkpointed only
+ * after the new transaction commits. */
- if (jh->b_next_transaction) {
- J_ASSERT(jh->b_next_transaction == transaction);
+ set_buffer_freed(bh);
+
+ if (!jh->b_next_transaction) {
spin_lock(&journal->j_list_lock);
- jh->b_next_transaction = NULL;
+ jh->b_next_transaction = transaction;
spin_unlock(&journal->j_list_lock);
+ } else {
+ J_ASSERT(jh->b_next_transaction == transaction);
/*
* only drop a reference if this transaction modified
diff --git a/fs/nfs/pagelist.c b/fs/nfs/pagelist.c
index 892c88542ebd..fad4d5188aaf 100644
--- a/fs/nfs/pagelist.c
+++ b/fs/nfs/pagelist.c
@@ -975,6 +975,17 @@ static void nfs_pageio_doio(struct nfs_pageio_descriptor *desc)
}
}
+static void
+nfs_pageio_cleanup_request(struct nfs_pageio_descriptor *desc,
+ struct nfs_page *req)
+{
+ LIST_HEAD(head);
+
+ nfs_list_remove_request(req);
+ nfs_list_add_request(req, &head);
+ desc->pg_completion_ops->error_cleanup(&head);
+}
+
/**
* nfs_pageio_add_request - Attempt to coalesce a request into a page list.
* @desc: destination io descriptor
@@ -1012,10 +1023,8 @@ static int __nfs_pageio_add_request(struct nfs_pageio_descriptor *desc,
nfs_page_group_unlock(req);
desc->pg_moreio = 1;
nfs_pageio_doio(desc);
- if (desc->pg_error < 0)
- return 0;
- if (mirror->pg_recoalesce)
- return 0;
+ if (desc->pg_error < 0 || mirror->pg_recoalesce)
+ goto out_cleanup_subreq;
/* retry add_request for this subreq */
nfs_page_group_lock(req, false);
continue;
@@ -1048,6 +1057,10 @@ err_ptr:
desc->pg_error = PTR_ERR(subreq);
nfs_page_group_unlock(req);
return 0;
+out_cleanup_subreq:
+ if (req != subreq)
+ nfs_pageio_cleanup_request(desc, subreq);
+ return 0;
}
static int nfs_do_recoalesce(struct nfs_pageio_descriptor *desc)
@@ -1066,7 +1079,6 @@ static int nfs_do_recoalesce(struct nfs_pageio_descriptor *desc)
struct nfs_page *req;
req = list_first_entry(&head, struct nfs_page, wb_list);
- nfs_list_remove_request(req);
if (__nfs_pageio_add_request(desc, req))
continue;
if (desc->pg_error < 0) {
@@ -1141,11 +1153,14 @@ int nfs_pageio_add_request(struct nfs_pageio_descriptor *desc,
if (nfs_pgio_has_mirroring(desc))
desc->pg_mirror_idx = midx;
if (!nfs_pageio_add_request_mirror(desc, dupreq))
- goto out_failed;
+ goto out_cleanup_subreq;
}
return 1;
+out_cleanup_subreq:
+ if (req != dupreq)
+ nfs_pageio_cleanup_request(desc, dupreq);
out_failed:
/*
* We might have failed before sending any reqs over wire.
@@ -1185,7 +1200,7 @@ static void nfs_pageio_complete_mirror(struct nfs_pageio_descriptor *desc,
desc->pg_mirror_idx = mirror_idx;
for (;;) {
nfs_pageio_doio(desc);
- if (!mirror->pg_recoalesce)
+ if (desc->pg_error < 0 || !mirror->pg_recoalesce)
break;
if (!nfs_do_recoalesce(desc))
break;
diff --git a/fs/nfsd/nfs3proc.c b/fs/nfsd/nfs3proc.c
index d818e4ffd79f..00b472fe77c1 100644
--- a/fs/nfsd/nfs3proc.c
+++ b/fs/nfsd/nfs3proc.c
@@ -431,8 +431,19 @@ nfsd3_proc_readdir(struct svc_rqst *rqstp, struct nfsd3_readdirargs *argp,
&resp->common, nfs3svc_encode_entry);
memcpy(resp->verf, argp->verf, 8);
resp->count = resp->buffer - argp->buffer;
- if (resp->offset)
- xdr_encode_hyper(resp->offset, argp->cookie);
+ if (resp->offset) {
+ loff_t offset = argp->cookie;
+
+ if (unlikely(resp->offset1)) {
+ /* we ended up with offset on a page boundary */
+ *resp->offset = htonl(offset >> 32);
+ *resp->offset1 = htonl(offset & 0xffffffff);
+ resp->offset1 = NULL;
+ } else {
+ xdr_encode_hyper(resp->offset, offset);
+ }
+ resp->offset = NULL;
+ }
RETURN_STATUS(nfserr);
}
@@ -500,6 +511,7 @@ nfsd3_proc_readdirplus(struct svc_rqst *rqstp, struct nfsd3_readdirargs *argp,
} else {
xdr_encode_hyper(resp->offset, offset);
}
+ resp->offset = NULL;
}
RETURN_STATUS(nfserr);
diff --git a/fs/nfsd/nfs3xdr.c b/fs/nfsd/nfs3xdr.c
index 452334694a5d..7e50248ca432 100644
--- a/fs/nfsd/nfs3xdr.c
+++ b/fs/nfsd/nfs3xdr.c
@@ -899,6 +899,7 @@ encode_entry(struct readdir_cd *ccd, const char *name, int namlen,
} else {
xdr_encode_hyper(cd->offset, offset64);
}
+ cd->offset = NULL;
}
/*
diff --git a/fs/nfsd/nfsctl.c b/fs/nfsd/nfsctl.c
index 797a155c9a67..f704f90db36c 100644
--- a/fs/nfsd/nfsctl.c
+++ b/fs/nfsd/nfsctl.c
@@ -1103,7 +1103,7 @@ static ssize_t write_v4_end_grace(struct file *file, char *buf, size_t size)
case 'Y':
case 'y':
case '1':
- if (nn->nfsd_serv)
+ if (!nn->nfsd_serv)
return -EBUSY;
nfsd4_end_grace(nn);
break;
diff --git a/fs/pipe.c b/fs/pipe.c
index 34345535f63d..388e09a689de 100644
--- a/fs/pipe.c
+++ b/fs/pipe.c
@@ -238,6 +238,14 @@ static const struct pipe_buf_operations anon_pipe_buf_ops = {
.get = generic_pipe_buf_get,
};
+static const struct pipe_buf_operations anon_pipe_buf_nomerge_ops = {
+ .can_merge = 0,
+ .confirm = generic_pipe_buf_confirm,
+ .release = anon_pipe_buf_release,
+ .steal = anon_pipe_buf_steal,
+ .get = generic_pipe_buf_get,
+};
+
static const struct pipe_buf_operations packet_pipe_buf_ops = {
.can_merge = 0,
.confirm = generic_pipe_buf_confirm,
@@ -246,6 +254,12 @@ static const struct pipe_buf_operations packet_pipe_buf_ops = {
.get = generic_pipe_buf_get,
};
+void pipe_buf_mark_unmergeable(struct pipe_buffer *buf)
+{
+ if (buf->ops == &anon_pipe_buf_ops)
+ buf->ops = &anon_pipe_buf_nomerge_ops;
+}
+
static ssize_t
pipe_read(struct kiocb *iocb, struct iov_iter *to)
{
diff --git a/fs/splice.c b/fs/splice.c
index f16b5868bade..015ffc2b9d37 100644
--- a/fs/splice.c
+++ b/fs/splice.c
@@ -1595,6 +1595,8 @@ retry:
*/
obuf->flags &= ~PIPE_BUF_FLAG_GIFT;
+ pipe_buf_mark_unmergeable(obuf);
+
obuf->len = len;
opipe->nrbufs++;
ibuf->offset += obuf->len;
@@ -1669,6 +1671,8 @@ static int link_pipe(struct pipe_inode_info *ipipe,
*/
obuf->flags &= ~PIPE_BUF_FLAG_GIFT;
+ pipe_buf_mark_unmergeable(obuf);
+
if (obuf->len > len)
obuf->len = len;
generated by cgit v1.2.3 (git 2.25.1) at 2025-12-20 17:42:46 +0000