Commit 5ef50c3b authored by Sage Weil's avatar Sage Weil

ceph: simplify+fix atomic_open

The initial ->atomic_open op was carried over from the old intent code,
which was incomplete and didn't really work.  Replace it with a fresh
method.  In particular:

 * always attempt to do an atomic open+lookup, both for the create case
   and for lookups of existing files.
 * fix symlink handling by returning 1 to the VFS so that we can follow
   the link to its destination. This fixes a longstanding ceph bug (#2392).
Signed-off-by: default avatarSage Weil <sage@inktank.com>
parent 1a9b4993
...@@ -633,44 +633,6 @@ static struct dentry *ceph_lookup(struct inode *dir, struct dentry *dentry, ...@@ -633,44 +633,6 @@ static struct dentry *ceph_lookup(struct inode *dir, struct dentry *dentry,
return dentry; return dentry;
} }
int ceph_atomic_open(struct inode *dir, struct dentry *dentry,
struct file *file, unsigned flags, umode_t mode,
int *opened)
{
int err;
struct dentry *res = NULL;
if (!(flags & O_CREAT)) {
if (dentry->d_name.len > NAME_MAX)
return -ENAMETOOLONG;
err = ceph_init_dentry(dentry);
if (err < 0)
return err;
return ceph_lookup_open(dir, dentry, file, flags, mode, opened);
}
if (d_unhashed(dentry)) {
res = ceph_lookup(dir, dentry, 0);
if (IS_ERR(res))
return PTR_ERR(res);
if (res)
dentry = res;
}
/* We don't deal with positive dentries here */
if (dentry->d_inode)
return finish_no_open(file, res);
*opened |= FILE_CREATED;
err = ceph_lookup_open(dir, dentry, file, flags, mode, opened);
dput(res);
return err;
}
/* /*
* If we do a create but get no trace back from the MDS, follow up with * If we do a create but get no trace back from the MDS, follow up with
* a lookup (the VFS expects us to link up the provided dentry). * a lookup (the VFS expects us to link up the provided dentry).
......
...@@ -4,6 +4,7 @@ ...@@ -4,6 +4,7 @@
#include <linux/sched.h> #include <linux/sched.h>
#include <linux/slab.h> #include <linux/slab.h>
#include <linux/file.h> #include <linux/file.h>
#include <linux/mount.h>
#include <linux/namei.h> #include <linux/namei.h>
#include <linux/writeback.h> #include <linux/writeback.h>
...@@ -106,9 +107,6 @@ static int ceph_init_file(struct inode *inode, struct file *file, int fmode) ...@@ -106,9 +107,6 @@ static int ceph_init_file(struct inode *inode, struct file *file, int fmode)
} }
/* /*
* If the filp already has private_data, that means the file was
* already opened by intent during lookup, and we do nothing.
*
* If we already have the requisite capabilities, we can satisfy * If we already have the requisite capabilities, we can satisfy
* the open request locally (no need to request new caps from the * the open request locally (no need to request new caps from the
* MDS). We do, however, need to inform the MDS (asynchronously) * MDS). We do, however, need to inform the MDS (asynchronously)
...@@ -207,24 +205,29 @@ int ceph_open(struct inode *inode, struct file *file) ...@@ -207,24 +205,29 @@ int ceph_open(struct inode *inode, struct file *file)
/* /*
* Do a lookup + open with a single request. * Do a lookup + open with a single request. If we get a non-existent
* * file or symlink, return 1 so the VFS can retry.
* If this succeeds, but some subsequent check in the vfs
* may_open() fails, the struct *file gets cleaned up (i.e.
* ceph_release gets called). So fear not!
*/ */
int ceph_lookup_open(struct inode *dir, struct dentry *dentry, int ceph_atomic_open(struct inode *dir, struct dentry *dentry,
struct file *file, unsigned flags, umode_t mode, struct file *file, unsigned flags, umode_t mode,
int *opened) int *opened)
{ {
struct ceph_fs_client *fsc = ceph_sb_to_client(dir->i_sb); struct ceph_fs_client *fsc = ceph_sb_to_client(dir->i_sb);
struct ceph_mds_client *mdsc = fsc->mdsc; struct ceph_mds_client *mdsc = fsc->mdsc;
struct ceph_mds_request *req; struct ceph_mds_request *req;
struct dentry *ret; struct dentry *dn;
int err; int err;
dout("ceph_lookup_open dentry %p '%.*s' flags %d mode 0%o\n", dout("atomic_open %p dentry %p '%.*s' %s flags %d mode 0%o\n",
dentry, dentry->d_name.len, dentry->d_name.name, flags, mode); dir, dentry, dentry->d_name.len, dentry->d_name.name,
d_unhashed(dentry) ? "unhashed" : "hashed", flags, mode);
if (dentry->d_name.len > NAME_MAX)
return -ENAMETOOLONG;
err = ceph_init_dentry(dentry);
if (err < 0)
return err;
/* do the open */ /* do the open */
req = prepare_open_request(dir->i_sb, flags, mode); req = prepare_open_request(dir->i_sb, flags, mode);
...@@ -241,22 +244,31 @@ int ceph_lookup_open(struct inode *dir, struct dentry *dentry, ...@@ -241,22 +244,31 @@ int ceph_lookup_open(struct inode *dir, struct dentry *dentry,
(flags & (O_CREAT|O_TRUNC)) ? dir : NULL, (flags & (O_CREAT|O_TRUNC)) ? dir : NULL,
req); req);
err = ceph_handle_snapdir(req, dentry, err); err = ceph_handle_snapdir(req, dentry, err);
if (err) if (err == 0 && (flags & O_CREAT) && !req->r_reply_info.head->is_dentry)
goto out;
if ((flags & O_CREAT) && !req->r_reply_info.head->is_dentry)
err = ceph_handle_notrace_create(dir, dentry); err = ceph_handle_notrace_create(dir, dentry);
if (err)
goto out;
err = finish_open(file, req->r_dentry, ceph_open, opened);
out:
ret = ceph_finish_lookup(req, dentry, err);
ceph_mdsc_put_request(req);
dout("ceph_lookup_open result=%p\n", ret);
if (IS_ERR(ret)) if (d_unhashed(dentry)) {
return PTR_ERR(ret); dn = ceph_finish_lookup(req, dentry, err);
if (IS_ERR(dn))
err = PTR_ERR(dn);
} else {
/* we were given a hashed negative dentry */
dn = NULL;
}
if (err)
goto out_err;
if (dn || dentry->d_inode == NULL || S_ISLNK(dentry->d_inode->i_mode)) {
/* make vfs retry on splice, ENOENT, or symlink */
dout("atomic_open finish_no_open on dn %p\n", dn);
err = finish_no_open(file, dn);
} else {
dout("atomic_open finish_open on dn %p\n", dn);
err = finish_open(file, dentry, ceph_open, opened);
}
dput(ret); out_err:
ceph_mdsc_put_request(req);
dout("atomic_open result=%d\n", err);
return err; return err;
} }
......
...@@ -806,9 +806,9 @@ extern int ceph_copy_from_page_vector(struct page **pages, ...@@ -806,9 +806,9 @@ extern int ceph_copy_from_page_vector(struct page **pages,
loff_t off, size_t len); loff_t off, size_t len);
extern struct page **ceph_alloc_page_vector(int num_pages, gfp_t flags); extern struct page **ceph_alloc_page_vector(int num_pages, gfp_t flags);
extern int ceph_open(struct inode *inode, struct file *file); extern int ceph_open(struct inode *inode, struct file *file);
extern int ceph_lookup_open(struct inode *dir, struct dentry *dentry, extern int ceph_atomic_open(struct inode *dir, struct dentry *dentry,
struct file *od, unsigned flags, struct file *file, unsigned flags, umode_t mode,
umode_t mode, int *opened); int *opened);
extern int ceph_release(struct inode *inode, struct file *filp); extern int ceph_release(struct inode *inode, struct file *filp);
/* dir.c */ /* dir.c */
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment