* defined IFF we hold CEPH_CAP_FILE_SHARED (which will be revoked by
* the MDS if/when the directory is modified).
*/
- static int __dcache_readdir(struct file *filp,
- void *dirent, filldir_t filldir)
+ static int __dcache_readdir(struct file *file, struct dir_context *ctx)
{
- struct ceph_file_info *fi = filp->private_data;
- struct dentry *parent = filp->f_dentry;
+ struct ceph_file_info *fi = file->private_data;
+ struct dentry *parent = file->f_dentry;
struct inode *dir = parent->d_inode;
struct list_head *p;
struct dentry *dentry, *last;
last = fi->dentry;
fi->dentry = NULL;
- dout("__dcache_readdir %p at %llu (last %p)\n", dir, filp->f_pos,
+ dout("__dcache_readdir %p at %llu (last %p)\n", dir, ctx->pos,
last);
spin_lock(&parent->d_lock);
/* start at beginning? */
- if (filp->f_pos == 2 || last == NULL ||
- filp->f_pos < ceph_dentry(last)->offset) {
+ if (ctx->pos == 2 || last == NULL ||
+ ctx->pos < ceph_dentry(last)->offset) {
if (list_empty(&parent->d_subdirs))
goto out_unlock;
p = parent->d_subdirs.prev;
if (!d_unhashed(dentry) && dentry->d_inode &&
ceph_snap(dentry->d_inode) != CEPH_SNAPDIR &&
ceph_ino(dentry->d_inode) != CEPH_INO_CEPH &&
- filp->f_pos <= di->offset)
+ ctx->pos <= di->offset)
break;
dout(" skipping %p %.*s at %llu (%llu)%s%s\n", dentry,
dentry->d_name.len, dentry->d_name.name, di->offset,
- filp->f_pos, d_unhashed(dentry) ? " unhashed" : "",
+ ctx->pos, d_unhashed(dentry) ? " unhashed" : "",
!dentry->d_inode ? " null" : "");
spin_unlock(&dentry->d_lock);
p = p->prev;
spin_unlock(&dentry->d_lock);
spin_unlock(&parent->d_lock);
- dout(" %llu (%llu) dentry %p %.*s %p\n", di->offset, filp->f_pos,
+ dout(" %llu (%llu) dentry %p %.*s %p\n", di->offset, ctx->pos,
dentry, dentry->d_name.len, dentry->d_name.name, dentry->d_inode);
- filp->f_pos = di->offset;
- err = filldir(dirent, dentry->d_name.name,
- dentry->d_name.len, di->offset,
+ ctx->pos = di->offset;
+ if (!dir_emit(ctx, dentry->d_name.name,
+ dentry->d_name.len,
ceph_translate_ino(dentry->d_sb, dentry->d_inode->i_ino),
- dentry->d_inode->i_mode >> 12);
-
- if (last) {
- if (err < 0) {
+ dentry->d_inode->i_mode >> 12)) {
+ if (last) {
/* remember our position */
fi->dentry = last;
fi->next_offset = di->offset;
- } else {
- dput(last);
}
+ dput(dentry);
+ return 0;
}
- last = dentry;
- if (err < 0)
- goto out;
+ if (last)
+ dput(last);
+ last = dentry;
- filp->f_pos++;
+ ctx->pos++;
/* make sure a dentry wasn't dropped while we didn't have parent lock */
if (!ceph_dir_is_complete(dir)) {
return 0;
}
- static int ceph_readdir(struct file *filp, void *dirent, filldir_t filldir)
+ static int ceph_readdir(struct file *file, struct dir_context *ctx)
{
- struct ceph_file_info *fi = filp->private_data;
- struct inode *inode = file_inode(filp);
+ struct ceph_file_info *fi = file->private_data;
+ struct inode *inode = file_inode(file);
struct ceph_inode_info *ci = ceph_inode(inode);
struct ceph_fs_client *fsc = ceph_inode_to_client(inode);
struct ceph_mds_client *mdsc = fsc->mdsc;
- unsigned frag = fpos_frag(filp->f_pos);
- int off = fpos_off(filp->f_pos);
+ unsigned frag = fpos_frag(ctx->pos);
+ int off = fpos_off(ctx->pos);
int err;
u32 ftype;
struct ceph_mds_reply_info_parsed *rinfo;
const int max_entries = fsc->mount_options->max_readdir;
const int max_bytes = fsc->mount_options->max_readdir_bytes;
- dout("readdir %p filp %p frag %u off %u\n", inode, filp, frag, off);
+ dout("readdir %p file %p frag %u off %u\n", inode, file, frag, off);
if (fi->flags & CEPH_F_ATEND)
return 0;
/* always start with . and .. */
- if (filp->f_pos == 0) {
+ if (ctx->pos == 0) {
/* note dir version at start of readdir so we can tell
* if any dentries get dropped */
fi->dir_release_count = atomic_read(&ci->i_release_count);
dout("readdir off 0 -> '.'\n");
- if (filldir(dirent, ".", 1, ceph_make_fpos(0, 0),
+ if (!dir_emit(ctx, ".", 1,
ceph_translate_ino(inode->i_sb, inode->i_ino),
- inode->i_mode >> 12) < 0)
+ inode->i_mode >> 12))
return 0;
- filp->f_pos = 1;
+ ctx->pos = 1;
off = 1;
}
- if (filp->f_pos == 1) {
- ino_t ino = parent_ino(filp->f_dentry);
+ if (ctx->pos == 1) {
+ ino_t ino = parent_ino(file->f_dentry);
dout("readdir off 1 -> '..'\n");
- if (filldir(dirent, "..", 2, ceph_make_fpos(0, 1),
+ if (!dir_emit(ctx, "..", 2,
ceph_translate_ino(inode->i_sb, ino),
- inode->i_mode >> 12) < 0)
+ inode->i_mode >> 12))
return 0;
- filp->f_pos = 2;
+ ctx->pos = 2;
off = 2;
}
/* can we use the dcache? */
spin_lock(&ci->i_ceph_lock);
- if ((filp->f_pos == 2 || fi->dentry) &&
+ if ((ctx->pos == 2 || fi->dentry) &&
!ceph_test_mount_opt(fsc, NOASYNCREADDIR) &&
ceph_snap(inode) != CEPH_SNAPDIR &&
__ceph_dir_is_complete(ci) &&
__ceph_caps_issued_mask(ci, CEPH_CAP_FILE_SHARED, 1)) {
spin_unlock(&ci->i_ceph_lock);
- err = __dcache_readdir(filp, dirent, filldir);
+ err = __dcache_readdir(file, ctx);
if (err != -EAGAIN)
return err;
} else {
return PTR_ERR(req);
req->r_inode = inode;
ihold(inode);
- req->r_dentry = dget(filp->f_dentry);
+ req->r_dentry = dget(file->f_dentry);
/* hints to request -> mds selection code */
req->r_direct_mode = USE_AUTH_MDS;
req->r_direct_hash = ceph_frag_value(frag);
rinfo = &fi->last_readdir->r_reply_info;
dout("readdir frag %x num %d off %d chunkoff %d\n", frag,
rinfo->dir_nr, off, fi->offset);
+
+ ctx->pos = ceph_make_fpos(frag, off);
while (off >= fi->offset && off - fi->offset < rinfo->dir_nr) {
- u64 pos = ceph_make_fpos(frag, off);
struct ceph_mds_reply_inode *in =
rinfo->dir_in[off - fi->offset].in;
struct ceph_vino vino;
ino_t ino;
dout("readdir off %d (%d/%d) -> %lld '%.*s' %p\n",
- off, off - fi->offset, rinfo->dir_nr, pos,
+ off, off - fi->offset, rinfo->dir_nr, ctx->pos,
rinfo->dir_dname_len[off - fi->offset],
rinfo->dir_dname[off - fi->offset], in);
BUG_ON(!in);
vino.ino = le64_to_cpu(in->ino);
vino.snap = le64_to_cpu(in->snapid);
ino = ceph_vino_to_ino(vino);
- if (filldir(dirent,
+ if (!dir_emit(ctx,
rinfo->dir_dname[off - fi->offset],
rinfo->dir_dname_len[off - fi->offset],
- pos,
- ceph_translate_ino(inode->i_sb, ino), ftype) < 0) {
+ ceph_translate_ino(inode->i_sb, ino), ftype)) {
dout("filldir stopping us...\n");
return 0;
}
off++;
- filp->f_pos = pos + 1;
+ ctx->pos++;
}
if (fi->last_name) {
if (!ceph_frag_is_rightmost(frag)) {
frag = ceph_frag_next(frag);
off = 0;
- filp->f_pos = ceph_make_fpos(frag, off);
+ ctx->pos = ceph_make_fpos(frag, off);
dout("readdir next frag is %x\n", frag);
goto more;
}
if (atomic_read(&ci->i_release_count) == fi->dir_release_count) {
dout(" marking %p complete\n", inode);
__ceph_dir_set_complete(ci, fi->dir_release_count);
- ci->i_max_offset = filp->f_pos;
+ ci->i_max_offset = ctx->pos;
}
spin_unlock(&ci->i_ceph_lock);
- dout("readdir %p filp %p done.\n", inode, filp);
+ dout("readdir %p file %p done.\n", inode, file);
return 0;
}
req->r_locked_dir = dir;
req->r_dentry_drop = CEPH_CAP_FILE_SHARED;
req->r_dentry_unless = CEPH_CAP_FILE_EXCL;
+ /* release LINK_SHARED on source inode (mds will lock it) */
+ req->r_old_inode_drop = CEPH_CAP_LINK_SHARED;
err = ceph_mdsc_do_request(mdsc, dir, req);
if (err) {
d_drop(dentry);
const struct file_operations ceph_dir_fops = {
.read = ceph_read_dir,
- .readdir = ceph_readdir,
+ .iterate = ceph_readdir,
.llseek = ceph_dir_llseek,
.open = ceph_open,
.release = ceph_release,
{
struct ceph_fs_client *fsc = ceph_inode_to_client(inode);
struct ceph_inode_info *ci = ceph_inode(inode);
- u64 pos, this_len;
+ u64 pos, this_len, left;
int io_align, page_align;
- int left, pages_left;
+ int pages_left;
int read;
struct page **page_pos;
int ret;
ret = 0;
hit_stripe = this_len < left;
was_short = ret >= 0 && ret < this_len;
- dout("striped_read %llu~%u (read %u) got %d%s%s\n", pos, left, read,
+ dout("striped_read %llu~%llu (read %u) got %d%s%s\n", pos, left, read,
ret, hit_stripe ? " HITSTRIPE" : "", was_short ? " SHORT" : "");
if (ret > 0) {
if (pos + left > inode->i_size)
left = inode->i_size - pos;
- dout("zero tail %d\n", left);
+ dout("zero tail %llu\n", left);
ceph_zero_page_vector_range(page_align + read, left,
pages);
read += left;
if ((got & (CEPH_CAP_FILE_CACHE|CEPH_CAP_FILE_LAZYIO)) == 0 ||
(iocb->ki_filp->f_flags & O_DIRECT) ||
- (inode->i_sb->s_flags & MS_SYNCHRONOUS) ||
(fi->flags & CEPH_F_SYNC))
/* hmm, this isn't really async... */
ret = ceph_sync_read(filp, base, len, ppos, &checkeof);
&ceph_sb_to_client(inode->i_sb)->client->osdc;
ssize_t count, written = 0;
int err, want, got;
- bool hold_mutex;
if (ceph_snap(inode) != CEPH_NOSNAP)
return -EROFS;
mutex_lock(&inode->i_mutex);
- hold_mutex = true;
err = generic_segment_checks(iov, &nr_segs, &count, VERIFY_READ);
if (err)
if ((got & (CEPH_CAP_FILE_BUFFER|CEPH_CAP_FILE_LAZYIO)) == 0 ||
(iocb->ki_filp->f_flags & O_DIRECT) ||
- (inode->i_sb->s_flags & MS_SYNCHRONOUS) ||
(fi->flags & CEPH_F_SYNC)) {
mutex_unlock(&inode->i_mutex);
written = ceph_sync_write(file, iov->iov_base, count,
pos, &iocb->ki_pos);
+ if (written == -EOLDSNAPC) {
+ dout("aio_write %p %llx.%llx %llu~%u"
+ "got EOLDSNAPC, retrying\n",
+ inode, ceph_vinop(inode),
+ pos, (unsigned)iov->iov_len);
+ mutex_lock(&inode->i_mutex);
+ goto retry_snap;
+ }
} else {
written = generic_file_buffered_write(iocb, iov, nr_segs,
pos, &iocb->ki_pos,
count, 0);
mutex_unlock(&inode->i_mutex);
}
- hold_mutex = false;
if (written >= 0) {
int dirty;
written = err;
}
- if (written == -EOLDSNAPC) {
- dout("aio_write %p %llx.%llx %llu~%u got EOLDSNAPC, retrying\n",
- inode, ceph_vinop(inode), pos, (unsigned)iov->iov_len);
- mutex_lock(&inode->i_mutex);
- hold_mutex = true;
- goto retry_snap;
- }
+ goto out_unlocked;
+
out:
- if (hold_mutex)
- mutex_unlock(&inode->i_mutex);
+ mutex_unlock(&inode->i_mutex);
+out_unlocked:
current->backing_dev_info = NULL;
-
return written ? written : err;
}
break;
}
- if (offset < 0 || offset > inode->i_sb->s_maxbytes) {
- offset = -EINVAL;
- goto out;
- }
-
- /* Special lock needed here? */
- if (offset != file->f_pos) {
- file->f_pos = offset;
- file->f_version = 0;
- }
+ offset = vfs_setpos(file, offset, inode->i_sb->s_maxbytes);
out:
mutex_unlock(&inode->i_mutex);
{
struct ceph_mds_session *s;
+ if (mds >= mdsc->mdsmap->m_max_mds)
+ return ERR_PTR(-EINVAL);
+
s = kzalloc(sizeof(*s), GFP_NOFS);
if (!s)
return ERR_PTR(-ENOMEM);
{
dout("remove_session_caps on %p\n", session);
iterate_session_caps(session, remove_session_caps_cb, NULL);
+
+ spin_lock(&session->s_cap_lock);
+ if (session->s_nr_caps > 0) {
+ struct super_block *sb = session->s_mdsc->fsc->sb;
+ struct inode *inode;
+ struct ceph_cap *cap, *prev = NULL;
+ struct ceph_vino vino;
+ /*
+ * iterate_session_caps() skips inodes that are being
+ * deleted, we need to wait until deletions are complete.
+ * __wait_on_freeing_inode() is designed for the job,
+ * but it is not exported, so use lookup inode function
+ * to access it.
+ */
+ while (!list_empty(&session->s_caps)) {
+ cap = list_entry(session->s_caps.next,
+ struct ceph_cap, session_caps);
+ if (cap == prev)
+ break;
+ prev = cap;
+ vino = cap->ci->i_vino;
+ spin_unlock(&session->s_cap_lock);
+
+ inode = ceph_lookup_inode(sb, vino);
+ iput(inode);
+
+ spin_lock(&session->s_cap_lock);
+ }
+ }
+ spin_unlock(&session->s_cap_lock);
+
BUG_ON(session->s_nr_caps > 0);
BUG_ON(!list_empty(&session->s_cap_flushing));
cleanup_cap_releases(session);
*base = ceph_ino(temp->d_inode);
*plen = len;
dout("build_path on %p %d built %llx '%.*s'\n",
- dentry, dentry->d_count, *base, len, path);
+ dentry, d_count(dentry), *base, len, path);
return path;
}
struct ceph_filelock *flocks;
encode_again:
- lock_flocks();
+ spin_lock(&inode->i_lock);
ceph_count_locks(inode, &num_fcntl_locks, &num_flock_locks);
- unlock_flocks();
+ spin_unlock(&inode->i_lock);
flocks = kmalloc((num_fcntl_locks+num_flock_locks) *
sizeof(struct ceph_filelock), GFP_NOFS);
if (!flocks) {
err = -ENOMEM;
goto out_free;
}
- lock_flocks();
+ spin_lock(&inode->i_lock);
err = ceph_encode_locks_to_buffer(inode, flocks,
num_fcntl_locks,
num_flock_locks);
- unlock_flocks();
+ spin_unlock(&inode->i_lock);
if (err) {
kfree(flocks);
if (err == -ENOSPC)
projects / cascardo / linux.git / commitdiff