summaryrefslogtreecommitdiff
path: root/fs/ceph
diff options
context:
space:
mode:
Diffstat (limited to 'fs/ceph')
-rw-r--r--fs/ceph/caps.c6
-rw-r--r--fs/ceph/dir.c21
-rw-r--r--fs/ceph/file.c36
-rw-r--r--fs/ceph/inode.c6
-rw-r--r--fs/ceph/mds_client.c19
-rw-r--r--fs/ceph/super.h5
6 files changed, 68 insertions, 25 deletions
diff --git a/fs/ceph/caps.c b/fs/ceph/caps.c
index f605753c8fe9..8d74ad7ba556 100644
--- a/fs/ceph/caps.c
+++ b/fs/ceph/caps.c
@@ -1811,7 +1811,7 @@ out:
spin_unlock(&ci->i_unsafe_lock);
}
-int ceph_fsync(struct file *file, int datasync)
+int ceph_fsync(struct file *file, loff_t start, loff_t end, int datasync)
{
struct inode *inode = file->f_mapping->host;
struct ceph_inode_info *ci = ceph_inode(inode);
@@ -1822,9 +1822,10 @@ int ceph_fsync(struct file *file, int datasync)
dout("fsync %p%s\n", inode, datasync ? " datasync" : "");
sync_write_wait(inode);
- ret = filemap_write_and_wait(inode->i_mapping);
+ ret = filemap_write_and_wait_range(inode->i_mapping, start, end);
if (ret < 0)
return ret;
+ mutex_lock(&inode->i_mutex);
dirty = try_flush_caps(inode, NULL, &flush_tid);
dout("fsync dirty caps are %s\n", ceph_cap_string(dirty));
@@ -1841,6 +1842,7 @@ int ceph_fsync(struct file *file, int datasync)
}
dout("fsync %p%s done\n", inode, datasync ? " datasync" : "");
+ mutex_unlock(&inode->i_mutex);
return ret;
}
diff --git a/fs/ceph/dir.c b/fs/ceph/dir.c
index ef8f08c343e8..1065ac779840 100644
--- a/fs/ceph/dir.c
+++ b/fs/ceph/dir.c
@@ -252,7 +252,7 @@ static int ceph_readdir(struct file *filp, void *dirent, filldir_t filldir)
off = 1;
}
if (filp->f_pos == 1) {
- ino_t ino = filp->f_dentry->d_parent->d_inode->i_ino;
+ ino_t ino = parent_ino(filp->f_dentry);
dout("readdir off 1 -> '..'\n");
if (filldir(dirent, "..", 2, ceph_make_fpos(0, 1),
ceph_translate_ino(inode->i_sb, ino),
@@ -446,14 +446,19 @@ static loff_t ceph_dir_llseek(struct file *file, loff_t offset, int origin)
loff_t retval;
mutex_lock(&inode->i_mutex);
+ retval = -EINVAL;
switch (origin) {
case SEEK_END:
offset += inode->i_size + 2; /* FIXME */
break;
case SEEK_CUR:
offset += file->f_pos;
+ case SEEK_SET:
+ break;
+ default:
+ goto out;
}
- retval = -EINVAL;
+
if (offset >= 0 && offset <= inode->i_sb->s_maxbytes) {
if (offset != file->f_pos) {
file->f_pos = offset;
@@ -477,6 +482,7 @@ static loff_t ceph_dir_llseek(struct file *file, loff_t offset, int origin)
if (offset > old_offset)
fi->dir_release_count--;
}
+out:
mutex_unlock(&inode->i_mutex);
return retval;
}
@@ -566,7 +572,6 @@ static struct dentry *ceph_lookup(struct inode *dir, struct dentry *dentry,
/* open (but not create!) intent? */
if (nd &&
(nd->flags & LOOKUP_OPEN) &&
- (nd->flags & LOOKUP_CONTINUE) == 0 && /* only open last component */
!(nd->intent.open.flags & O_CREAT)) {
int mode = nd->intent.open.create_mode & ~current->fs->umask;
return ceph_lookup_open(dir, dentry, nd, mode, 1);
@@ -1113,7 +1118,8 @@ static ssize_t ceph_read_dir(struct file *file, char __user *buf, size_t size,
* an fsync() on a dir will wait for any uncommitted directory
* operations to commit.
*/
-static int ceph_dir_fsync(struct file *file, int datasync)
+static int ceph_dir_fsync(struct file *file, loff_t start, loff_t end,
+ int datasync)
{
struct inode *inode = file->f_path.dentry->d_inode;
struct ceph_inode_info *ci = ceph_inode(inode);
@@ -1123,6 +1129,11 @@ static int ceph_dir_fsync(struct file *file, int datasync)
int ret = 0;
dout("dir_fsync %p\n", inode);
+ ret = filemap_write_and_wait_range(inode->i_mapping, start, end);
+ if (ret)
+ return ret;
+ mutex_lock(&inode->i_mutex);
+
spin_lock(&ci->i_unsafe_lock);
if (list_empty(head))
goto out;
@@ -1156,6 +1167,8 @@ static int ceph_dir_fsync(struct file *file, int datasync)
} while (req->r_tid < last_tid);
out:
spin_unlock(&ci->i_unsafe_lock);
+ mutex_unlock(&inode->i_mutex);
+
return ret;
}
diff --git a/fs/ceph/file.c b/fs/ceph/file.c
index 9542f07d0b93..0d0eae05598f 100644
--- a/fs/ceph/file.c
+++ b/fs/ceph/file.c
@@ -226,7 +226,7 @@ struct dentry *ceph_lookup_open(struct inode *dir, struct dentry *dentry,
struct inode *parent_inode = get_dentry_parent_inode(file->f_dentry);
struct ceph_mds_request *req;
int err;
- int flags = nd->intent.open.flags - 1; /* silly vfs! */
+ int flags = nd->intent.open.flags;
dout("ceph_lookup_open dentry %p '%.*s' flags %d mode 0%o\n",
dentry, dentry->d_name.len, dentry->d_name.name, flags, mode);
@@ -290,7 +290,6 @@ static int striped_read(struct inode *inode,
struct ceph_inode_info *ci = ceph_inode(inode);
u64 pos, this_len;
int io_align, page_align;
- int page_off = off & ~PAGE_CACHE_MASK; /* first byte's offset in page */
int left, pages_left;
int read;
struct page **page_pos;
@@ -326,12 +325,11 @@ more:
ret, hit_stripe ? " HITSTRIPE" : "", was_short ? " SHORT" : "");
if (ret > 0) {
- int didpages =
- ((pos & ~PAGE_CACHE_MASK) + ret) >> PAGE_CACHE_SHIFT;
+ int didpages = (page_align + ret) >> PAGE_CACHE_SHIFT;
if (read < pos - off) {
dout(" zero gap %llu to %llu\n", off + read, pos);
- ceph_zero_page_vector_range(page_off + read,
+ ceph_zero_page_vector_range(page_align + read,
pos - off - read, pages);
}
pos += ret;
@@ -356,7 +354,7 @@ more:
left = inode->i_size - pos;
dout("zero tail %d\n", left);
- ceph_zero_page_vector_range(page_off + read, left,
+ ceph_zero_page_vector_range(page_align + read, left,
pages);
read += left;
}
@@ -478,9 +476,6 @@ static ssize_t ceph_sync_write(struct file *file, const char __user *data,
else
pos = *offset;
- io_align = pos & ~PAGE_MASK;
- buf_align = (unsigned long)data & ~PAGE_MASK;
-
ret = filemap_write_and_wait_range(inode->i_mapping, pos, pos + left);
if (ret < 0)
return ret;
@@ -504,6 +499,8 @@ static ssize_t ceph_sync_write(struct file *file, const char __user *data,
* boundary. this isn't atomic, unfortunately. :(
*/
more:
+ io_align = pos & ~PAGE_MASK;
+ buf_align = (unsigned long)data & ~PAGE_MASK;
len = left;
if (file->f_flags & O_DIRECT) {
/* write from beginning of first page, regardless of
@@ -593,6 +590,7 @@ out:
pos += len;
written += len;
left -= len;
+ data += written;
if (left)
goto more;
@@ -770,13 +768,16 @@ static loff_t ceph_llseek(struct file *file, loff_t offset, int origin)
mutex_lock(&inode->i_mutex);
__ceph_do_pending_vmtruncate(inode);
- switch (origin) {
- case SEEK_END:
+ if (origin != SEEK_CUR || origin != SEEK_SET) {
ret = ceph_do_getattr(inode, CEPH_STAT_CAP_SIZE);
if (ret < 0) {
offset = ret;
goto out;
}
+ }
+
+ switch (origin) {
+ case SEEK_END:
offset += inode->i_size;
break;
case SEEK_CUR:
@@ -792,6 +793,19 @@ static loff_t ceph_llseek(struct file *file, loff_t offset, int origin)
}
offset += file->f_pos;
break;
+ case SEEK_DATA:
+ if (offset >= inode->i_size) {
+ ret = -ENXIO;
+ goto out;
+ }
+ break;
+ case SEEK_HOLE:
+ if (offset >= inode->i_size) {
+ ret = -ENXIO;
+ goto out;
+ }
+ offset = inode->i_size;
+ break;
}
if (offset < 0 || offset > inode->i_sb->s_maxbytes) {
diff --git a/fs/ceph/inode.c b/fs/ceph/inode.c
index d8858e96ab18..dfb2831d8d85 100644
--- a/fs/ceph/inode.c
+++ b/fs/ceph/inode.c
@@ -1795,17 +1795,17 @@ int ceph_do_getattr(struct inode *inode, int mask)
* Check inode permissions. We verify we have a valid value for
* the AUTH cap, then call the generic handler.
*/
-int ceph_permission(struct inode *inode, int mask, unsigned int flags)
+int ceph_permission(struct inode *inode, int mask)
{
int err;
- if (flags & IPERM_FLAG_RCU)
+ if (mask & MAY_NOT_BLOCK)
return -ECHILD;
err = ceph_do_getattr(inode, CEPH_CAP_AUTH_SHARED);
if (!err)
- err = generic_permission(inode, mask, flags, NULL);
+ err = generic_permission(inode, mask);
return err;
}
diff --git a/fs/ceph/mds_client.c b/fs/ceph/mds_client.c
index 79743d146be6..0c1d91756528 100644
--- a/fs/ceph/mds_client.c
+++ b/fs/ceph/mds_client.c
@@ -1438,12 +1438,15 @@ char *ceph_mdsc_build_path(struct dentry *dentry, int *plen, u64 *base,
struct dentry *temp;
char *path;
int len, pos;
+ unsigned seq;
if (dentry == NULL)
return ERR_PTR(-EINVAL);
retry:
len = 0;
+ seq = read_seqbegin(&rename_lock);
+ rcu_read_lock();
for (temp = dentry; !IS_ROOT(temp);) {
struct inode *inode = temp->d_inode;
if (inode && ceph_snap(inode) == CEPH_SNAPDIR)
@@ -1455,10 +1458,12 @@ retry:
len += 1 + temp->d_name.len;
temp = temp->d_parent;
if (temp == NULL) {
+ rcu_read_unlock();
pr_err("build_path corrupt dentry %p\n", dentry);
return ERR_PTR(-EINVAL);
}
}
+ rcu_read_unlock();
if (len)
len--; /* no leading '/' */
@@ -1467,9 +1472,12 @@ retry:
return ERR_PTR(-ENOMEM);
pos = len;
path[pos] = 0; /* trailing null */
+ rcu_read_lock();
for (temp = dentry; !IS_ROOT(temp) && pos != 0; ) {
- struct inode *inode = temp->d_inode;
+ struct inode *inode;
+ spin_lock(&temp->d_lock);
+ inode = temp->d_inode;
if (inode && ceph_snap(inode) == CEPH_SNAPDIR) {
dout("build_path path+%d: %p SNAPDIR\n",
pos, temp);
@@ -1478,21 +1486,26 @@ retry:
break;
} else {
pos -= temp->d_name.len;
- if (pos < 0)
+ if (pos < 0) {
+ spin_unlock(&temp->d_lock);
break;
+ }
strncpy(path + pos, temp->d_name.name,
temp->d_name.len);
}
+ spin_unlock(&temp->d_lock);
if (pos)
path[--pos] = '/';
temp = temp->d_parent;
if (temp == NULL) {
+ rcu_read_unlock();
pr_err("build_path corrupt dentry\n");
kfree(path);
return ERR_PTR(-EINVAL);
}
}
- if (pos != 0) {
+ rcu_read_unlock();
+ if (pos != 0 || read_seqretry(&rename_lock, seq)) {
pr_err("build_path did not end path lookup where "
"expected, namelen is %d, pos is %d\n", len, pos);
/* presumably this is only possible if racing with a
diff --git a/fs/ceph/super.h b/fs/ceph/super.h
index f5cabefa98dc..30446b144e3d 100644
--- a/fs/ceph/super.h
+++ b/fs/ceph/super.h
@@ -692,7 +692,7 @@ extern void ceph_queue_invalidate(struct inode *inode);
extern void ceph_queue_writeback(struct inode *inode);
extern int ceph_do_getattr(struct inode *inode, int mask);
-extern int ceph_permission(struct inode *inode, int mask, unsigned int flags);
+extern int ceph_permission(struct inode *inode, int mask);
extern int ceph_setattr(struct dentry *dentry, struct iattr *attr);
extern int ceph_getattr(struct vfsmount *mnt, struct dentry *dentry,
struct kstat *stat);
@@ -728,7 +728,8 @@ extern void ceph_put_cap(struct ceph_mds_client *mdsc,
extern void ceph_queue_caps_release(struct inode *inode);
extern int ceph_write_inode(struct inode *inode, struct writeback_control *wbc);
-extern int ceph_fsync(struct file *file, int datasync);
+extern int ceph_fsync(struct file *file, loff_t start, loff_t end,
+ int datasync);
extern void ceph_kick_flushing_caps(struct ceph_mds_client *mdsc,
struct ceph_mds_session *session);
extern struct ceph_cap *ceph_get_cap_for_mds(struct ceph_inode_info *ci,
generated by cgit v1.2.3 (git 2.25.1) at 2024-12-07 20:12:02 +0000