[PATCH 0/8] Make NFSv4 OPEN(CREATE) less brittle

[PATCH 0/8] Make NFSv4 OPEN(CREATE) less brittle

[Chuck Lever]

fsdevel experts TL;DR: patch 8/8 is the only one that touches the
VFS. The other patches prepare NFSD for 8/8.

Attempt to address occasional reports of test failures caused by
NFSv4 OPEN(CREATE) failing internally only after the target file
object has already been created.

The basic approach is to re-organize the NFSv4 OPEN code path so
that common failure modes occur /before/ the call to vfs_create()
rather than afterwards. In addition, the local file is opened and
created atomically so that another client can't race and de-permit
that file just after it is created but before the server has opened
it.

This series was posted a few weeks ago as an RFC. Since then, Red
Hat QE has used a Lustre racer-based reproducer to confirm that
the issue is not reproducible. Therefore I'd like to include this
series in the NFSD 5.19 pull request.


---

Chuck Lever (8):
      NFSD: Clean up nfsd3_proc_create()
      NFSD: Avoid calling fh_drop_write() twice in do_nfsd_create()
      NFSD: Refactor nfsd_create_setattr()
      NFSD: Refactor NFSv3 CREATE
      NFSD: Refactor NFSv4 OPEN(CREATE)
      NFSD: Remove do_nfsd_create()
      NFSD: Clean up nfsd_open_verified()
      NFSD: Instantiate a struct file when creating a regular NFSv4 file


 fs/nfsd/filecache.c |  51 +++++++--
 fs/nfsd/filecache.h |   2 +
 fs/nfsd/nfs3proc.c  | 141 +++++++++++++++++++++----
 fs/nfsd/nfs4proc.c  | 197 +++++++++++++++++++++++++++++++++--
 fs/nfsd/nfs4state.c |  16 ++-
 fs/nfsd/vfs.c       | 245 ++++++++++----------------------------------
 fs/nfsd/vfs.h       |  14 +--
 fs/nfsd/xdr4.h      |   1 +
 fs/open.c           |  44 ++++++++
 include/linux/fs.h  |   2 +
 10 files changed, 471 insertions(+), 242 deletions(-)

--
Chuck Lever

[PATCH 1/8] NFSD: Clean up nfsd3_proc_create()

[Chuck Lever]

As near as I can tell, mode bit masking and setting S_IFREG is
already done by do_nfsd_create() and vfs_create(). The NFSv4 path
(do_open_lookup), for example, does not bother with this special
processing.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
---
 fs/nfsd/nfs3proc.c |   16 ++--------------
 1 file changed, 2 insertions(+), 14 deletions(-)

diff --git a/fs/nfsd/nfs3proc.c b/fs/nfsd/nfs3proc.c
index 936eebd4c56d..981a2a71c5af 100644
--- a/fs/nfsd/nfs3proc.c
+++ b/fs/nfsd/nfs3proc.c
@@ -229,8 +229,7 @@ nfsd3_proc_create(struct svc_rqst *rqstp)
 {
 	struct nfsd3_createargs *argp = rqstp->rq_argp;
 	struct nfsd3_diropres *resp = rqstp->rq_resp;
-	svc_fh		*dirfhp, *newfhp = NULL;
-	struct iattr	*attr;
+	svc_fh *dirfhp, *newfhp;
 
 	dprintk("nfsd: CREATE(3)   %s %.*s\n",
 				SVCFH_fmt(&argp->fh),
@@ -239,20 +238,9 @@ nfsd3_proc_create(struct svc_rqst *rqstp)
 
 	dirfhp = fh_copy(&resp->dirfh, &argp->fh);
 	newfhp = fh_init(&resp->fh, NFS3_FHSIZE);
-	attr   = &argp->attrs;
-
-	/* Unfudge the mode bits */
-	attr->ia_mode &= ~S_IFMT;
-	if (!(attr->ia_valid & ATTR_MODE)) { 
-		attr->ia_valid |= ATTR_MODE;
-		attr->ia_mode = S_IFREG;
-	} else {
-		attr->ia_mode = (attr->ia_mode & ~S_IFMT) | S_IFREG;
-	}
 
-	/* Now create the file and set attributes */
 	resp->status = do_nfsd_create(rqstp, dirfhp, argp->name, argp->len,
-				      attr, newfhp, argp->createmode,
+				      &argp->attrs, newfhp, argp->createmode,
 				      (u32 *)argp->verf, NULL, NULL);
 	return rpc_success;
 }

[PATCH 2/8] NFSD: Avoid calling fh_drop_write() twice in do_nfsd_create()

[Chuck Lever]

Clean up: The "out" label already invokes fh_drop_write().

Note that fh_drop_write() is already careful not to invoke
mnt_drop_write() if either it has already been done or there is
nothing to drop. Therefore no change in behavior is expected.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
---
 fs/nfsd/vfs.c |    5 +----
 1 file changed, 1 insertion(+), 4 deletions(-)

diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c
index 4c1984f07cdc..bbed7a986784 100644
--- a/fs/nfsd/vfs.c
+++ b/fs/nfsd/vfs.c
@@ -1479,7 +1479,6 @@ do_nfsd_create(struct svc_rqst *rqstp, struct svc_fh *fhp,
 		case NFS3_CREATE_GUARDED:
 			err = nfserr_exist;
 		}
-		fh_drop_write(fhp);
 		goto out;
 	}
 
@@ -1487,10 +1486,8 @@ do_nfsd_create(struct svc_rqst *rqstp, struct svc_fh *fhp,
 		iap->ia_mode &= ~current_umask();
 
 	host_err = vfs_create(&init_user_ns, dirp, dchild, iap->ia_mode, true);
-	if (host_err < 0) {
-		fh_drop_write(fhp);
+	if (host_err < 0)
 		goto out_nfserr;
-	}
 	if (created)
 		*created = true;
 

[PATCH 3/8] NFSD: Refactor nfsd_create_setattr()

[Chuck Lever]

I'd like to move do_nfsd_create() out of vfs.c. Therefore
nfsd_create_setattr() needs to be made publicly visible.

Note that both call sites in vfs.c commit both the new object and
its parent directory, so just combine those common metadata commits
into nfsd_create_setattr().

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
---
 fs/nfsd/vfs.c |   79 ++++++++++++++++++++++++++++++---------------------------
 fs/nfsd/vfs.h |    2 +
 2 files changed, 44 insertions(+), 37 deletions(-)

diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c
index bbed7a986784..83c989a5d6f3 100644
--- a/fs/nfsd/vfs.c
+++ b/fs/nfsd/vfs.c
@@ -1181,14 +1181,26 @@ nfsd_commit(struct svc_rqst *rqstp, struct svc_fh *fhp, u64 offset,
 	return err;
 }
 
-static __be32
-nfsd_create_setattr(struct svc_rqst *rqstp, struct svc_fh *resfhp,
-			struct iattr *iap)
+/**
+ * nfsd_create_setattr - Set a created file's attributes
+ * @rqstp: RPC transaction being executed
+ * @fhp: NFS filehandle of parent directory
+ * @resfhp: NFS filehandle of new object
+ * @iap: requested attributes of new object
+ *
+ * Returns nfs_ok on success, or an nfsstat in network byte order.
+ */
+__be32
+nfsd_create_setattr(struct svc_rqst *rqstp, struct svc_fh *fhp,
+		    struct svc_fh *resfhp, struct iattr *iap)
 {
+	__be32 status;
+
 	/*
-	 * Mode has already been set earlier in create:
+	 * Mode has already been set by file creation.
 	 */
 	iap->ia_valid &= ~ATTR_MODE;
+
 	/*
 	 * Setting uid/gid works only for root.  Irix appears to
 	 * send along the gid on create when it tries to implement
@@ -1196,10 +1208,31 @@ nfsd_create_setattr(struct svc_rqst *rqstp, struct svc_fh *resfhp,
 	 */
 	if (!uid_eq(current_fsuid(), GLOBAL_ROOT_UID))
 		iap->ia_valid &= ~(ATTR_UID|ATTR_GID);
+
+	/*
+	 * Callers expect new file metadata to be committed even
+	 * if the attributes have not changed.
+	 */
 	if (iap->ia_valid)
-		return nfsd_setattr(rqstp, resfhp, iap, 0, (time64_t)0);
-	/* Callers expect file metadata to be committed here */
-	return nfserrno(commit_metadata(resfhp));
+		status = nfsd_setattr(rqstp, resfhp, iap, 0, (time64_t)0);
+	else
+		status = nfserrno(commit_metadata(resfhp));
+
+	/*
+	 * Transactional filesystems had a chance to commit changes
+	 * for both parent and child simultaneously making the
+	 * following commit_metadata a noop in many cases.
+	 */
+	if (!status)
+		status = nfserrno(commit_metadata(fhp));
+
+	/*
+	 * Update the new filehandle to pick up the new attributes.
+	 */
+	if (!status)
+		status = fh_update(resfhp);
+
+	return status;
 }
 
 /* HPUX client sometimes creates a file in mode 000, and sets size to 0.
@@ -1226,7 +1259,6 @@ nfsd_create_locked(struct svc_rqst *rqstp, struct svc_fh *fhp,
 	struct dentry	*dentry, *dchild;
 	struct inode	*dirp;
 	__be32		err;
-	__be32		err2;
 	int		host_err;
 
 	dentry = fhp->fh_dentry;
@@ -1299,22 +1331,8 @@ nfsd_create_locked(struct svc_rqst *rqstp, struct svc_fh *fhp,
 	if (host_err < 0)
 		goto out_nfserr;
 
-	err = nfsd_create_setattr(rqstp, resfhp, iap);
+	err = nfsd_create_setattr(rqstp, fhp, resfhp, iap);
 
-	/*
-	 * nfsd_create_setattr already committed the child.  Transactional
-	 * filesystems had a chance to commit changes for both parent and
-	 * child simultaneously making the following commit_metadata a
-	 * noop.
-	 */
-	err2 = nfserrno(commit_metadata(fhp));
-	if (err2)
-		err = err2;
-	/*
-	 * Update the file handle to get the new inode info.
-	 */
-	if (!err)
-		err = fh_update(resfhp);
 out:
 	dput(dchild);
 	return err;
@@ -1505,20 +1523,7 @@ do_nfsd_create(struct svc_rqst *rqstp, struct svc_fh *fhp,
 	}
 
  set_attr:
-	err = nfsd_create_setattr(rqstp, resfhp, iap);
-
-	/*
-	 * nfsd_create_setattr already committed the child
-	 * (and possibly also the parent).
-	 */
-	if (!err)
-		err = nfserrno(commit_metadata(fhp));
-
-	/*
-	 * Update the filehandle to get the new inode info.
-	 */
-	if (!err)
-		err = fh_update(resfhp);
+	err = nfsd_create_setattr(rqstp, fhp, resfhp, iap);
 
  out:
 	fh_unlock(fhp);
diff --git a/fs/nfsd/vfs.h b/fs/nfsd/vfs.h
index ccb87b2864f6..1f32a83456b0 100644
--- a/fs/nfsd/vfs.h
+++ b/fs/nfsd/vfs.h
@@ -69,6 +69,8 @@ __be32		nfsd_create(struct svc_rqst *, struct svc_fh *,
 				char *name, int len, struct iattr *attrs,
 				int type, dev_t rdev, struct svc_fh *res);
 __be32		nfsd_access(struct svc_rqst *, struct svc_fh *, u32 *, u32 *);
+__be32		nfsd_create_setattr(struct svc_rqst *rqstp, struct svc_fh *fhp,
+				struct svc_fh *resfhp, struct iattr *iap);
 __be32		do_nfsd_create(struct svc_rqst *, struct svc_fh *,
 				char *name, int len, struct iattr *attrs,
 				struct svc_fh *res, int createmode,

[PATCH 4/8] NFSD: Refactor NFSv3 CREATE

[Chuck Lever]

The NFSv3 CREATE and NFSv4 OPEN(CREATE) use cases are about to
diverge such that it makes sense to split do_nfsd_create() into one
version for NFSv3 and one for NFSv4.

As a first step, copy do_nfsd_create() to nfs3proc.c and remove
NFSv4-specific logic.

One immediate legibility benefit is that the logic for handling
NFSv3 createhow is now quite straightforward. NFSv4 createhow
has some subtleties that IMO do not belong in generic code.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
---
 fs/nfsd/nfs3proc.c |  127 ++++++++++++++++++++++++++++++++++++++++++++++++++--
 1 file changed, 121 insertions(+), 6 deletions(-)

diff --git a/fs/nfsd/nfs3proc.c b/fs/nfsd/nfs3proc.c
index 981a2a71c5af..981a3a7a6e16 100644
--- a/fs/nfsd/nfs3proc.c
+++ b/fs/nfsd/nfs3proc.c
@@ -8,6 +8,7 @@
 #include <linux/fs.h>
 #include <linux/ext2_fs.h>
 #include <linux/magic.h>
+#include <linux/namei.h>
 
 #include "cache.h"
 #include "xdr3.h"
@@ -220,10 +221,126 @@ nfsd3_proc_write(struct svc_rqst *rqstp)
 }
 
 /*
- * With NFSv3, CREATE processing is a lot easier than with NFSv2.
- * At least in theory; we'll see how it fares in practice when the
- * first reports about SunOS compatibility problems start to pour in...
+ * Implement NFSv3's unchecked, guarded, and exclusive CREATE
+ * semantics for regular files. Except for the created file,
+ * this operation is stateless on the server.
+ *
+ * Upon return, caller must release @fhp and @resfhp.
  */
+static __be32
+nfsd3_create_file(struct svc_rqst *rqstp, struct svc_fh *fhp,
+		  struct svc_fh *resfhp, struct nfsd3_createargs *argp)
+{
+	struct iattr *iap = &argp->attrs;
+	struct dentry *parent, *child;
+	__u32 v_mtime, v_atime;
+	struct inode *inode;
+	__be32 status;
+	int host_err;
+
+	if (isdotent(argp->name, argp->len))
+		return nfserr_exist;
+	if (!(iap->ia_valid & ATTR_MODE))
+		iap->ia_mode = 0;
+
+	status = fh_verify(rqstp, fhp, S_IFDIR, NFSD_MAY_EXEC);
+	if (status != nfs_ok)
+		return status;
+
+	parent = fhp->fh_dentry;
+	inode = d_inode(parent);
+
+	host_err = fh_want_write(fhp);
+	if (host_err)
+		return nfserrno(host_err);
+
+	fh_lock_nested(fhp, I_MUTEX_PARENT);
+
+	child = lookup_one_len(argp->name, parent, argp->len);
+	if (IS_ERR(child)) {
+		status = nfserrno(PTR_ERR(child));
+		goto out;
+	}
+
+	if (d_really_is_negative(child)) {
+		status = fh_verify(rqstp, fhp, S_IFDIR, NFSD_MAY_CREATE);
+		if (status != nfs_ok)
+			goto out;
+	}
+
+	status = fh_compose(resfhp, fhp->fh_export, child, fhp);
+	if (status != nfs_ok)
+		goto out;
+
+	v_mtime = 0;
+	v_atime = 0;
+	if (argp->createmode == NFS3_CREATE_EXCLUSIVE) {
+		u32 *verifier = (u32 *)argp->verf;
+
+		/*
+		 * Solaris 7 gets confused (bugid 4218508) if these have
+		 * the high bit set, as do xfs filesystems without the
+		 * "bigtime" feature. So just clear the high bits.
+		 */
+		v_mtime = verifier[0] & 0x7fffffff;
+		v_atime = verifier[1] & 0x7fffffff;
+	}
+
+	if (d_really_is_positive(child)) {
+		status = nfs_ok;
+
+		switch (argp->createmode) {
+		case NFS3_CREATE_UNCHECKED:
+			if (!d_is_reg(child))
+				break;
+			iap->ia_valid &= ATTR_SIZE;
+			goto set_attr;
+		case NFS3_CREATE_GUARDED:
+			status = nfserr_exist;
+			break;
+		case NFS3_CREATE_EXCLUSIVE:
+			if (d_inode(child)->i_mtime.tv_sec == v_mtime &&
+			    d_inode(child)->i_atime.tv_sec == v_atime &&
+			    d_inode(child)->i_size == 0) {
+				break;
+			}
+			status = nfserr_exist;
+		}
+		goto out;
+	}
+
+	if (!IS_POSIXACL(inode))
+		iap->ia_mode &= ~current_umask();
+
+	host_err = vfs_create(&init_user_ns, inode, child, iap->ia_mode, true);
+	if (host_err < 0) {
+		status = nfserrno(host_err);
+		goto out;
+	}
+
+	/* A newly created file already has a file size of zero. */
+	if ((iap->ia_valid & ATTR_SIZE) && (iap->ia_size == 0))
+		iap->ia_valid &= ~ATTR_SIZE;
+	if (argp->createmode == NFS3_CREATE_EXCLUSIVE) {
+		iap->ia_valid = ATTR_MTIME | ATTR_ATIME |
+				ATTR_MTIME_SET | ATTR_ATIME_SET;
+		iap->ia_mtime.tv_sec = v_mtime;
+		iap->ia_atime.tv_sec = v_atime;
+		iap->ia_mtime.tv_nsec = 0;
+		iap->ia_atime.tv_nsec = 0;
+	}
+
+set_attr:
+	status = nfsd_create_setattr(rqstp, fhp, resfhp, iap);
+
+out:
+	fh_unlock(fhp);
+	if (child && !IS_ERR(child))
+		dput(child);
+	fh_drop_write(fhp);
+	return status;
+}
+
 static __be32
 nfsd3_proc_create(struct svc_rqst *rqstp)
 {
@@ -239,9 +356,7 @@ nfsd3_proc_create(struct svc_rqst *rqstp)
 	dirfhp = fh_copy(&resp->dirfh, &argp->fh);
 	newfhp = fh_init(&resp->fh, NFS3_FHSIZE);
 
-	resp->status = do_nfsd_create(rqstp, dirfhp, argp->name, argp->len,
-				      &argp->attrs, newfhp, argp->createmode,
-				      (u32 *)argp->verf, NULL, NULL);
+	resp->status = nfsd3_create_file(rqstp, dirfhp, newfhp, argp);
 	return rpc_success;
 }
 

[PATCH 5/8] NFSD: Refactor NFSv4 OPEN(CREATE)

[Chuck Lever]

Copy do_nfsd_create() to nfs4proc.c and remove NFSv3-specific logic.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
---
 fs/nfsd/nfs4proc.c |  162 +++++++++++++++++++++++++++++++++++++++++++++++++---
 1 file changed, 152 insertions(+), 10 deletions(-)

diff --git a/fs/nfsd/nfs4proc.c b/fs/nfsd/nfs4proc.c
index b207c76a873f..99c0485a29e9 100644
--- a/fs/nfsd/nfs4proc.c
+++ b/fs/nfsd/nfs4proc.c
@@ -37,6 +37,8 @@
 #include <linux/falloc.h>
 #include <linux/slab.h>
 #include <linux/kthread.h>
+#include <linux/namei.h>
+
 #include <linux/sunrpc/addr.h>
 #include <linux/nfs_ssc.h>
 
@@ -235,6 +237,154 @@ static void nfsd4_set_open_owner_reply_cache(struct nfsd4_compound_state *cstate
 			&resfh->fh_handle);
 }
 
+static inline bool nfsd4_create_is_exclusive(int createmode)
+{
+	return createmode == NFS4_CREATE_EXCLUSIVE ||
+		createmode == NFS4_CREATE_EXCLUSIVE4_1;
+}
+
+/*
+ * Implement NFSv4's unchecked, guarded, and exclusive create
+ * semantics for regular files. Open state for this new file is
+ * subsequently fabricated in nfsd4_process_open2().
+ *
+ * Upon return, caller must release @fhp and @resfhp.
+ */
+static __be32
+nfsd4_create_file(struct svc_rqst *rqstp, struct svc_fh *fhp,
+		  struct svc_fh *resfhp, struct nfsd4_open *open)
+{
+	struct iattr *iap = &open->op_iattr;
+	struct dentry *parent, *child;
+	__u32 v_mtime, v_atime;
+	struct inode *inode;
+	__be32 status;
+	int host_err;
+
+	if (isdotent(open->op_fname, open->op_fnamelen))
+		return nfserr_exist;
+	if (!(iap->ia_valid & ATTR_MODE))
+		iap->ia_mode = 0;
+
+	status = fh_verify(rqstp, fhp, S_IFDIR, NFSD_MAY_EXEC);
+	if (status != nfs_ok)
+		return status;
+	parent = fhp->fh_dentry;
+	inode = d_inode(parent);
+
+	host_err = fh_want_write(fhp);
+	if (host_err)
+		return nfserrno(host_err);
+
+	fh_lock_nested(fhp, I_MUTEX_PARENT);
+
+	child = lookup_one_len(open->op_fname, parent, open->op_fnamelen);
+	if (IS_ERR(child)) {
+		status = nfserrno(PTR_ERR(child));
+		goto out;
+	}
+
+	if (d_really_is_negative(child)) {
+		status = fh_verify(rqstp, fhp, S_IFDIR, NFSD_MAY_CREATE);
+		if (status != nfs_ok)
+			goto out;
+	}
+
+	status = fh_compose(resfhp, fhp->fh_export, child, fhp);
+	if (status != nfs_ok)
+		goto out;
+
+	v_mtime = 0;
+	v_atime = 0;
+	if (nfsd4_create_is_exclusive(open->op_createmode)) {
+		u32 *verifier = (u32 *)open->op_verf.data;
+
+		/*
+		 * Solaris 7 gets confused (bugid 4218508) if these have
+		 * the high bit set, as do xfs filesystems without the
+		 * "bigtime" feature. So just clear the high bits. If this
+		 * is ever changed to use different attrs for storing the
+		 * verifier, then do_open_lookup() will also need to be
+		 * fixed accordingly.
+		 */
+		v_mtime = verifier[0] & 0x7fffffff;
+		v_atime = verifier[1] & 0x7fffffff;
+	}
+
+	if (d_really_is_positive(child)) {
+		status = nfs_ok;
+
+		switch (open->op_createmode) {
+		case NFS4_CREATE_UNCHECKED:
+			if (!d_is_reg(child))
+				break;
+
+			/*
+			 * In NFSv4, we don't want to truncate the file
+			 * now. This would be wrong if the OPEN fails for
+			 * some other reason. Furthermore, if the size is
+			 * nonzero, we should ignore it according to spec!
+			 */
+			open->op_truncate = (iap->ia_valid & ATTR_SIZE) &&
+						!iap->ia_size;
+			break;
+		case NFS4_CREATE_GUARDED:
+			status = nfserr_exist;
+			break;
+		case NFS4_CREATE_EXCLUSIVE:
+			if (d_inode(child)->i_mtime.tv_sec == v_mtime &&
+			    d_inode(child)->i_atime.tv_sec == v_atime &&
+			    d_inode(child)->i_size == 0) {
+				open->op_created = true;
+				break;		/* subtle */
+			}
+			status = nfserr_exist;
+			break;
+		case NFS4_CREATE_EXCLUSIVE4_1:
+			if (d_inode(child)->i_mtime.tv_sec == v_mtime &&
+			    d_inode(child)->i_atime.tv_sec == v_atime &&
+			    d_inode(child)->i_size == 0) {
+				open->op_created = true;
+				goto set_attr;	/* subtle */
+			}
+			status = nfserr_exist;
+		}
+		goto out;
+	}
+
+	if (!IS_POSIXACL(inode))
+		iap->ia_mode &= ~current_umask();
+
+	host_err = vfs_create(&init_user_ns, inode, child, iap->ia_mode, true);
+	if (host_err < 0) {
+		status = nfserrno(host_err);
+		goto out;
+	}
+	open->op_created = true;
+
+	/* A newly created file already has a file size of zero. */
+	if ((iap->ia_valid & ATTR_SIZE) && (iap->ia_size == 0))
+		iap->ia_valid &= ~ATTR_SIZE;
+	if (nfsd4_create_is_exclusive(open->op_createmode)) {
+		iap->ia_valid = ATTR_MTIME | ATTR_ATIME |
+				ATTR_MTIME_SET|ATTR_ATIME_SET;
+		iap->ia_mtime.tv_sec = v_mtime;
+		iap->ia_atime.tv_sec = v_atime;
+		iap->ia_mtime.tv_nsec = 0;
+		iap->ia_atime.tv_nsec = 0;
+	}
+
+set_attr:
+	status = nfsd_create_setattr(rqstp, fhp, resfhp, iap);
+
+out:
+	fh_unlock(fhp);
+	if (child && !IS_ERR(child))
+		dput(child);
+	fh_drop_write(fhp);
+	return status;
+}
+
 static __be32
 do_open_lookup(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, struct nfsd4_open *open, struct svc_fh **resfh)
 {
@@ -264,16 +414,8 @@ do_open_lookup(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, stru
 		 * yes          | yes    | GUARDED4        | GUARDED4
 		 */
 
-		/*
-		 * Note: create modes (UNCHECKED,GUARDED...) are the same
-		 * in NFSv4 as in v3 except EXCLUSIVE4_1.
-		 */
 		current->fs->umask = open->op_umask;
-		status = do_nfsd_create(rqstp, current_fh, open->op_fname,
-					open->op_fnamelen, &open->op_iattr,
-					*resfh, open->op_createmode,
-					(u32 *)open->op_verf.data,
-					&open->op_truncate, &open->op_created);
+		status = nfsd4_create_file(rqstp, current_fh, *resfh, open);
 		current->fs->umask = 0;
 
 		if (!status && open->op_label.len)
@@ -284,7 +426,7 @@ do_open_lookup(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, stru
 		 * use the returned bitmask to indicate which attributes
 		 * we used to store the verifier:
 		 */
-		if (nfsd_create_is_exclusive(open->op_createmode) && status == 0)
+		if (nfsd4_create_is_exclusive(open->op_createmode) && status == 0)
 			open->op_bmval[1] |= (FATTR4_WORD1_TIME_ACCESS |
 						FATTR4_WORD1_TIME_MODIFY);
 	} else

[PATCH 6/8] NFSD: Remove do_nfsd_create()

[Chuck Lever]

Now that its two callers have their own version-specific instance of
this function, do_nfsd_create() is no longer used.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
---
 fs/nfsd/vfs.c |  150 ---------------------------------------------------------
 fs/nfsd/vfs.h |   10 ----
 2 files changed, 160 deletions(-)

diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c
index 83c989a5d6f3..0b0dbb8e0894 100644
--- a/fs/nfsd/vfs.c
+++ b/fs/nfsd/vfs.c
@@ -1387,156 +1387,6 @@ nfsd_create(struct svc_rqst *rqstp, struct svc_fh *fhp,
 					rdev, resfhp);
 }
 
-/*
- * NFSv3 and NFSv4 version of nfsd_create
- */
-__be32
-do_nfsd_create(struct svc_rqst *rqstp, struct svc_fh *fhp,
-		char *fname, int flen, struct iattr *iap,
-		struct svc_fh *resfhp, int createmode, u32 *verifier,
-	        bool *truncp, bool *created)
-{
-	struct dentry	*dentry, *dchild = NULL;
-	struct inode	*dirp;
-	__be32		err;
-	int		host_err;
-	__u32		v_mtime=0, v_atime=0;
-
-	err = nfserr_perm;
-	if (!flen)
-		goto out;
-	err = nfserr_exist;
-	if (isdotent(fname, flen))
-		goto out;
-	if (!(iap->ia_valid & ATTR_MODE))
-		iap->ia_mode = 0;
-	err = fh_verify(rqstp, fhp, S_IFDIR, NFSD_MAY_EXEC);
-	if (err)
-		goto out;
-
-	dentry = fhp->fh_dentry;
-	dirp = d_inode(dentry);
-
-	host_err = fh_want_write(fhp);
-	if (host_err)
-		goto out_nfserr;
-
-	fh_lock_nested(fhp, I_MUTEX_PARENT);
-
-	/*
-	 * Compose the response file handle.
-	 */
-	dchild = lookup_one_len(fname, dentry, flen);
-	host_err = PTR_ERR(dchild);
-	if (IS_ERR(dchild))
-		goto out_nfserr;
-
-	/* If file doesn't exist, check for permissions to create one */
-	if (d_really_is_negative(dchild)) {
-		err = fh_verify(rqstp, fhp, S_IFDIR, NFSD_MAY_CREATE);
-		if (err)
-			goto out;
-	}
-
-	err = fh_compose(resfhp, fhp->fh_export, dchild, fhp);
-	if (err)
-		goto out;
-
-	if (nfsd_create_is_exclusive(createmode)) {
-		/* solaris7 gets confused (bugid 4218508) if these have
-		 * the high bit set, as do xfs filesystems without the
-		 * "bigtime" feature.  So just clear the high bits. If this is
-		 * ever changed to use different attrs for storing the
-		 * verifier, then do_open_lookup() will also need to be fixed
-		 * accordingly.
-		 */
-		v_mtime = verifier[0]&0x7fffffff;
-		v_atime = verifier[1]&0x7fffffff;
-	}
-	
-	if (d_really_is_positive(dchild)) {
-		err = 0;
-
-		switch (createmode) {
-		case NFS3_CREATE_UNCHECKED:
-			if (! d_is_reg(dchild))
-				goto out;
-			else if (truncp) {
-				/* in nfsv4, we need to treat this case a little
-				 * differently.  we don't want to truncate the
-				 * file now; this would be wrong if the OPEN
-				 * fails for some other reason.  furthermore,
-				 * if the size is nonzero, we should ignore it
-				 * according to spec!
-				 */
-				*truncp = (iap->ia_valid & ATTR_SIZE) && !iap->ia_size;
-			}
-			else {
-				iap->ia_valid &= ATTR_SIZE;
-				goto set_attr;
-			}
-			break;
-		case NFS3_CREATE_EXCLUSIVE:
-			if (   d_inode(dchild)->i_mtime.tv_sec == v_mtime
-			    && d_inode(dchild)->i_atime.tv_sec == v_atime
-			    && d_inode(dchild)->i_size  == 0 ) {
-				if (created)
-					*created = true;
-				break;
-			}
-			fallthrough;
-		case NFS4_CREATE_EXCLUSIVE4_1:
-			if (   d_inode(dchild)->i_mtime.tv_sec == v_mtime
-			    && d_inode(dchild)->i_atime.tv_sec == v_atime
-			    && d_inode(dchild)->i_size  == 0 ) {
-				if (created)
-					*created = true;
-				goto set_attr;
-			}
-			fallthrough;
-		case NFS3_CREATE_GUARDED:
-			err = nfserr_exist;
-		}
-		goto out;
-	}
-
-	if (!IS_POSIXACL(dirp))
-		iap->ia_mode &= ~current_umask();
-
-	host_err = vfs_create(&init_user_ns, dirp, dchild, iap->ia_mode, true);
-	if (host_err < 0)
-		goto out_nfserr;
-	if (created)
-		*created = true;
-
-	nfsd_check_ignore_resizing(iap);
-
-	if (nfsd_create_is_exclusive(createmode)) {
-		/* Cram the verifier into atime/mtime */
-		iap->ia_valid = ATTR_MTIME|ATTR_ATIME
-			| ATTR_MTIME_SET|ATTR_ATIME_SET;
-		/* XXX someone who knows this better please fix it for nsec */ 
-		iap->ia_mtime.tv_sec = v_mtime;
-		iap->ia_atime.tv_sec = v_atime;
-		iap->ia_mtime.tv_nsec = 0;
-		iap->ia_atime.tv_nsec = 0;
-	}
-
- set_attr:
-	err = nfsd_create_setattr(rqstp, fhp, resfhp, iap);
-
- out:
-	fh_unlock(fhp);
-	if (dchild && !IS_ERR(dchild))
-		dput(dchild);
-	fh_drop_write(fhp);
- 	return err;
- 
- out_nfserr:
-	err = nfserrno(host_err);
-	goto out;
-}
-
 /*
  * Read a symlink. On entry, *lenp must contain the maximum path length that
  * fits into the buffer. On return, it contains the true length.
diff --git a/fs/nfsd/vfs.h b/fs/nfsd/vfs.h
index 1f32a83456b0..f99794b033a5 100644
--- a/fs/nfsd/vfs.h
+++ b/fs/nfsd/vfs.h
@@ -71,10 +71,6 @@ __be32		nfsd_create(struct svc_rqst *, struct svc_fh *,
 __be32		nfsd_access(struct svc_rqst *, struct svc_fh *, u32 *, u32 *);
 __be32		nfsd_create_setattr(struct svc_rqst *rqstp, struct svc_fh *fhp,
 				struct svc_fh *resfhp, struct iattr *iap);
-__be32		do_nfsd_create(struct svc_rqst *, struct svc_fh *,
-				char *name, int len, struct iattr *attrs,
-				struct svc_fh *res, int createmode,
-				u32 *verifier, bool *truncp, bool *created);
 __be32		nfsd_commit(struct svc_rqst *rqst, struct svc_fh *fhp,
 				u64 offset, u32 count, __be32 *verf);
 #ifdef CONFIG_NFSD_V4
@@ -161,10 +157,4 @@ static inline __be32 fh_getattr(const struct svc_fh *fh, struct kstat *stat)
 				    AT_STATX_SYNC_AS_STAT));
 }
 
-static inline int nfsd_create_is_exclusive(int createmode)
-{
-	return createmode == NFS3_CREATE_EXCLUSIVE
-	       || createmode == NFS4_CREATE_EXCLUSIVE4_1;
-}
-
 #endif /* LINUX_NFSD_VFS_H */

[PATCH 7/8] NFSD: Clean up nfsd_open_verified()

[Chuck Lever]

Its only caller always passes S_IFREG as the @type parameter. As an
additional clean-up, add a kerneldoc comment.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
---
 fs/nfsd/filecache.c |    4 ++--
 fs/nfsd/vfs.c       |   15 ++++++++++++---
 fs/nfsd/vfs.h       |    2 +-
 3 files changed, 15 insertions(+), 6 deletions(-)

diff --git a/fs/nfsd/filecache.c b/fs/nfsd/filecache.c
index 2c1b027774d4..781bef7e42d9 100644
--- a/fs/nfsd/filecache.c
+++ b/fs/nfsd/filecache.c
@@ -995,8 +995,8 @@ nfsd_file_acquire(struct svc_rqst *rqstp, struct svc_fh *fhp,
 
 	nf->nf_mark = nfsd_file_mark_find_or_create(nf);
 	if (nf->nf_mark)
-		status = nfsd_open_verified(rqstp, fhp, S_IFREG,
-				may_flags, &nf->nf_file);
+		status = nfsd_open_verified(rqstp, fhp, may_flags,
+					    &nf->nf_file);
 	else
 		status = nfserr_jukebox;
 	/*
diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c
index 0b0dbb8e0894..840e3af63a6f 100644
--- a/fs/nfsd/vfs.c
+++ b/fs/nfsd/vfs.c
@@ -827,14 +827,23 @@ nfsd_open(struct svc_rqst *rqstp, struct svc_fh *fhp, umode_t type,
 	return err;
 }
 
+/**
+ * nfsd_open_verified - Open a regular file for the filecache
+ * @rqstp: RPC request
+ * @fhp: NFS filehandle of the file to open
+ * @may_flags: internal permission flags
+ * @filp: OUT: open "struct file *"
+ *
+ * Returns an nfsstat value in network byte order.
+ */
 __be32
-nfsd_open_verified(struct svc_rqst *rqstp, struct svc_fh *fhp, umode_t type,
-		int may_flags, struct file **filp)
+nfsd_open_verified(struct svc_rqst *rqstp, struct svc_fh *fhp, int may_flags,
+		   struct file **filp)
 {
 	__be32 err;
 
 	validate_process_creds();
-	err = __nfsd_open(rqstp, fhp, type, may_flags, filp);
+	err = __nfsd_open(rqstp, fhp, S_IFREG, may_flags, filp);
 	validate_process_creds();
 	return err;
 }
diff --git a/fs/nfsd/vfs.h b/fs/nfsd/vfs.h
index f99794b033a5..26347d76f44a 100644
--- a/fs/nfsd/vfs.h
+++ b/fs/nfsd/vfs.h
@@ -86,7 +86,7 @@ __be32		nfsd_setxattr(struct svc_rqst *rqstp, struct svc_fh *fhp,
 int 		nfsd_open_break_lease(struct inode *, int);
 __be32		nfsd_open(struct svc_rqst *, struct svc_fh *, umode_t,
 				int, struct file **);
-__be32		nfsd_open_verified(struct svc_rqst *, struct svc_fh *, umode_t,
+__be32		nfsd_open_verified(struct svc_rqst *, struct svc_fh *,
 				int, struct file **);
 __be32		nfsd_splice_read(struct svc_rqst *rqstp, struct svc_fh *fhp,
 				struct file *file, loff_t offset,

[PATCH 8/8] NFSD: Instantiate a struct file when creating a regular NFSv4 file

[Chuck Lever]

There have been reports of races that cause NFSv4 OPEN(CREATE) to
return an error even though the requested file was created. NFSv4
does not provide a status code for this case.

To mitigate some of these problems, reorganize the NFSv4
OPEN(CREATE) logic to allocate resources before the file is actually
created, and open the new file while the parent directory is still
locked.

Two new APIs are added:

+ Add an API that works like nfsd_file_acquire() but does not open
the underlying file. The OPEN(CREATE) path can use this API when it
already has an open file.

+ Add an API that is kin to dentry_open(). NFSD needs to create a
file and grab an open "struct file *" atomically. The
alloc_empty_file() has to be done before the inode create. If it
fails (for example, because the NFS server has exceeded its
max_files limit), we avoid creating the file and can still return
an error to the NFS client.

BugLink: https://bugzilla.linux-nfs.org/show_bug.cgi?id=382
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Tested-by: JianHong Yin <jiyin@redhat.com>
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
---
 fs/nfsd/filecache.c |   51 ++++++++++++++++++++++++++++++++++++++++++++-------
 fs/nfsd/filecache.h |    2 ++
 fs/nfsd/nfs4proc.c  |   43 +++++++++++++++++++++++++++++++++++++++----
 fs/nfsd/nfs4state.c |   16 +++++++++++++---
 fs/nfsd/xdr4.h      |    1 +
 fs/open.c           |   44 ++++++++++++++++++++++++++++++++++++++++++++
 include/linux/fs.h  |    2 ++
 7 files changed, 145 insertions(+), 14 deletions(-)

diff --git a/fs/nfsd/filecache.c b/fs/nfsd/filecache.c
index 781bef7e42d9..24772f246461 100644
--- a/fs/nfsd/filecache.c
+++ b/fs/nfsd/filecache.c
@@ -897,9 +897,9 @@ nfsd_file_is_cached(struct inode *inode)
 	return ret;
 }
 
-__be32
-nfsd_file_acquire(struct svc_rqst *rqstp, struct svc_fh *fhp,
-		  unsigned int may_flags, struct nfsd_file **pnf)
+static __be32
+nfsd_do_file_acquire(struct svc_rqst *rqstp, struct svc_fh *fhp,
+		     unsigned int may_flags, struct nfsd_file **pnf, bool open)
 {
 	__be32	status;
 	struct net *net = SVC_NET(rqstp);
@@ -994,10 +994,13 @@ nfsd_file_acquire(struct svc_rqst *rqstp, struct svc_fh *fhp,
 		nfsd_file_gc();
 
 	nf->nf_mark = nfsd_file_mark_find_or_create(nf);
-	if (nf->nf_mark)
-		status = nfsd_open_verified(rqstp, fhp, may_flags,
-					    &nf->nf_file);
-	else
+	if (nf->nf_mark) {
+		if (open)
+			status = nfsd_open_verified(rqstp, fhp, may_flags,
+						    &nf->nf_file);
+		else
+			status = nfs_ok;
+	} else
 		status = nfserr_jukebox;
 	/*
 	 * If construction failed, or we raced with a call to unlink()
@@ -1017,6 +1020,40 @@ nfsd_file_acquire(struct svc_rqst *rqstp, struct svc_fh *fhp,
 	goto out;
 }
 
+/**
+ * nfsd_file_acquire - Get a struct nfsd_file with an open file
+ * @rqstp: the RPC transaction being executed
+ * @fhp: the NFS filehandle of the file to be opened
+ * @may_flags: NFSD_MAY_ settings for the file
+ * @pnf: OUT: new or found "struct nfsd_file" object
+ *
+ * Returns nfs_ok and sets @pnf on success; otherwise an nfsstat in
+ * network byte order is returned.
+ */
+__be32
+nfsd_file_acquire(struct svc_rqst *rqstp, struct svc_fh *fhp,
+		  unsigned int may_flags, struct nfsd_file **pnf)
+{
+	return nfsd_do_file_acquire(rqstp, fhp, may_flags, pnf, true);
+}
+
+/**
+ * nfsd_file_create - Get a struct nfsd_file, do not open
+ * @rqstp: the RPC transaction being executed
+ * @fhp: the NFS filehandle of the file just created
+ * @may_flags: NFSD_MAY_ settings for the file
+ * @pnf: OUT: new or found "struct nfsd_file" object
+ *
+ * Returns nfs_ok and sets @pnf on success; otherwise an nfsstat in
+ * network byte order is returned.
+ */
+__be32
+nfsd_file_create(struct svc_rqst *rqstp, struct svc_fh *fhp,
+		 unsigned int may_flags, struct nfsd_file **pnf)
+{
+	return nfsd_do_file_acquire(rqstp, fhp, may_flags, pnf, false);
+}
+
 /*
  * Note that fields may be added, removed or reordered in the future. Programs
  * scraping this file for info should test the labels to ensure they're
diff --git a/fs/nfsd/filecache.h b/fs/nfsd/filecache.h
index 435ceab27897..1da0c79a5580 100644
--- a/fs/nfsd/filecache.h
+++ b/fs/nfsd/filecache.h
@@ -59,5 +59,7 @@ void nfsd_file_close_inode_sync(struct inode *inode);
 bool nfsd_file_is_cached(struct inode *inode);
 __be32 nfsd_file_acquire(struct svc_rqst *rqstp, struct svc_fh *fhp,
 		  unsigned int may_flags, struct nfsd_file **nfp);
+__be32 nfsd_file_create(struct svc_rqst *rqstp, struct svc_fh *fhp,
+		  unsigned int may_flags, struct nfsd_file **nfp);
 int	nfsd_file_cache_stats_open(struct inode *, struct file *);
 #endif /* _FS_NFSD_FILECACHE_H */
diff --git a/fs/nfsd/nfs4proc.c b/fs/nfsd/nfs4proc.c
index 99c0485a29e9..28bae84d7809 100644
--- a/fs/nfsd/nfs4proc.c
+++ b/fs/nfsd/nfs4proc.c
@@ -243,6 +243,37 @@ static inline bool nfsd4_create_is_exclusive(int createmode)
 		createmode == NFS4_CREATE_EXCLUSIVE4_1;
 }
 
+static __be32
+nfsd4_vfs_create(struct svc_fh *fhp, struct dentry *child,
+		 struct nfsd4_open *open)
+{
+	struct file *filp;
+	struct path path;
+	int oflags;
+
+	oflags = O_CREAT | O_LARGEFILE;
+	switch (open->op_share_access & NFS4_SHARE_ACCESS_BOTH) {
+	case NFS4_SHARE_ACCESS_WRITE:
+		oflags |= O_WRONLY;
+		break;
+	case NFS4_SHARE_ACCESS_BOTH:
+		oflags |= O_RDWR;
+		break;
+	default:
+		oflags |= O_RDONLY;
+	}
+
+	path.mnt = fhp->fh_export->ex_path.mnt;
+	path.dentry = child;
+	filp = dentry_create(&path, oflags, open->op_iattr.ia_mode,
+			     current_cred());
+	if (IS_ERR(filp))
+		return nfserrno(PTR_ERR(filp));
+
+	open->op_filp = filp;
+	return nfs_ok;
+}
+
 /*
  * Implement NFSv4's unchecked, guarded, and exclusive create
  * semantics for regular files. Open state for this new file is
@@ -355,11 +386,9 @@ nfsd4_create_file(struct svc_rqst *rqstp, struct svc_fh *fhp,
 	if (!IS_POSIXACL(inode))
 		iap->ia_mode &= ~current_umask();
 
-	host_err = vfs_create(&init_user_ns, inode, child, iap->ia_mode, true);
-	if (host_err < 0) {
-		status = nfserrno(host_err);
+	status = nfsd4_vfs_create(fhp, child, open);
+	if (status != nfs_ok)
 		goto out;
-	}
 	open->op_created = true;
 
 	/* A newly created file already has a file size of zero. */
@@ -517,6 +546,8 @@ nfsd4_open(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 		(int)open->op_fnamelen, open->op_fname,
 		open->op_openowner);
 
+	open->op_filp = NULL;
+
 	/* This check required by spec. */
 	if (open->op_create && open->op_claim_type != NFS4_OPEN_CLAIM_NULL)
 		return nfserr_inval;
@@ -613,6 +644,10 @@ nfsd4_open(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 	if (reclaim && !status)
 		nn->somebody_reclaimed = true;
 out:
+	if (open->op_filp) {
+		fput(open->op_filp);
+		open->op_filp = NULL;
+	}
 	if (resfh && resfh != &cstate->current_fh) {
 		fh_dup2(&cstate->current_fh, resfh);
 		fh_put(resfh);
diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c
index 231e5c19cdb7..131102cba06b 100644
--- a/fs/nfsd/nfs4state.c
+++ b/fs/nfsd/nfs4state.c
@@ -5093,9 +5093,19 @@ static __be32 nfs4_get_vfs_file(struct svc_rqst *rqstp, struct nfs4_file *fp,
 
 	if (!fp->fi_fds[oflag]) {
 		spin_unlock(&fp->fi_lock);
-		status = nfsd_file_acquire(rqstp, cur_fh, access, &nf);
-		if (status)
-			goto out_put_access;
+
+		if (!open->op_filp) {
+			status = nfsd_file_acquire(rqstp, cur_fh, access, &nf);
+			if (status != nfs_ok)
+				goto out_put_access;
+		} else {
+			status = nfsd_file_create(rqstp, cur_fh, access, &nf);
+			if (status != nfs_ok)
+				goto out_put_access;
+			nf->nf_file = open->op_filp;
+			open->op_filp = NULL;
+		}
+
 		spin_lock(&fp->fi_lock);
 		if (!fp->fi_fds[oflag]) {
 			fp->fi_fds[oflag] = nf;
diff --git a/fs/nfsd/xdr4.h b/fs/nfsd/xdr4.h
index 846ab6df9d48..7b744011f2d3 100644
--- a/fs/nfsd/xdr4.h
+++ b/fs/nfsd/xdr4.h
@@ -273,6 +273,7 @@ struct nfsd4_open {
 	bool		op_truncate;        /* used during processing */
 	bool		op_created;         /* used during processing */
 	struct nfs4_openowner *op_openowner; /* used during processing */
+	struct file	*op_filp;           /* used during processing */
 	struct nfs4_file *op_file;          /* used during processing */
 	struct nfs4_ol_stateid *op_stp;	    /* used during processing */
 	struct nfs4_clnt_odstate *op_odstate; /* used during processing */
diff --git a/fs/open.c b/fs/open.c
index 1315253e0247..e3cf72c269e1 100644
--- a/fs/open.c
+++ b/fs/open.c
@@ -981,6 +981,50 @@ struct file *dentry_open(const struct path *path, int flags,
 }
 EXPORT_SYMBOL(dentry_open);
 
+/**
+ * dentry_create - Create and open a file
+ * @path: path to create
+ * @flags: O_ flags
+ * @mode: mode bits for new file
+ * @cred: credentials to use
+ *
+ * Caller must hold the parent directory's lock, and have prepared
+ * a negative dentry, placed in @path->dentry, for the new file.
+ *
+ * On success, returns a "struct file *". Otherwise a ERR_PTR
+ * is returned.
+ */
+struct file *dentry_create(const struct path *path, int flags, umode_t mode,
+			   const struct cred *cred)
+{
+	struct dentry *parent;
+	struct file *f;
+	int error;
+
+	validate_creds(cred);
+	f = alloc_empty_file(flags, cred);
+	if (IS_ERR(f))
+		return f;
+
+	parent = dget_parent(path->dentry);
+	error = vfs_create(mnt_user_ns(path->mnt), d_inode(parent),
+			   path->dentry, mode, true);
+	dput(parent);
+	if (error) {
+		fput(f);
+		return ERR_PTR(error);
+	}
+
+	error = vfs_open(path, f);
+	if (error) {
+		fput(f);
+		return ERR_PTR(error);
+	}
+
+	return f;
+}
+EXPORT_SYMBOL(dentry_create);
+
 struct file *open_with_fake_path(const struct path *path, int flags,
 				struct inode *inode, const struct cred *cred)
 {
diff --git a/include/linux/fs.h b/include/linux/fs.h
index aa6c1bbdb8c4..b848355b5e6c 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -2640,6 +2640,8 @@ static inline struct file *file_open_root_mnt(struct vfsmount *mnt,
 			      name, flags, mode);
 }
 extern struct file * dentry_open(const struct path *, int, const struct cred *);
+extern struct file *dentry_create(const struct path *path, int flags,
+				  umode_t mode, const struct cred *cred);
 extern struct file * open_with_fake_path(const struct path *, int,
 					 struct inode*, const struct cred *);
 static inline struct file *file_clone_open(struct file *file)

Re: [PATCH 8/8] NFSD: Instantiate a struct file when creating a regular NFSv4 file

[Al Viro]

On Fri, May 13, 2022 at 03:40:13PM -0400, Chuck Lever wrote:

> +/**
> + * dentry_create - Create and open a file
> + * @path: path to create
> + * @flags: O_ flags
> + * @mode: mode bits for new file
> + * @cred: credentials to use
> + *
> + * Caller must hold the parent directory's lock, and have prepared
> + * a negative dentry, placed in @path->dentry, for the new file.
> + *
> + * On success, returns a "struct file *". Otherwise a ERR_PTR
> + * is returned.
> + */
> +struct file *dentry_create(const struct path *path, int flags, umode_t mode,
> +			   const struct cred *cred)
> +{
> +	struct dentry *parent;
> +	struct file *f;
> +	int error;
> +
> +	validate_creds(cred);
> +	f = alloc_empty_file(flags, cred);
> +	if (IS_ERR(f))
> +		return f;
> +
> +	parent = dget_parent(path->dentry);
> +	error = vfs_create(mnt_user_ns(path->mnt), d_inode(parent),
> +			   path->dentry, mode, true);
> +	dput(parent);

Yuck.  dget_parent() is not entirely without valid uses, but this isn't
one.  It's for the cases when parent is *not* stable and you need to grab
what had been the parent at some point (even though it might not be the
parent anymore by the time dget_parent() returns).  Here you seriously
depend upon it remaining the parent of that sucker all the way through -
otherwise vfs_create() would break.  And you really, really depend upon
its survival - the caller is holding it locked, so they would better
have it pinned.

So this dget_parent() (and dput()) is pointless and confusing;
path->dentry->d_parent would suffice.  And document that path->dentry
must not be root and that its parent should match path->mnt.

> +	if (error) {
> +		fput(f);
> +		return ERR_PTR(error);
> +	}
> +
> +	error = vfs_open(path, f);
> +	if (error) {
> +		fput(f);
> +		return ERR_PTR(error);
> +	}
> +
> +	return f;

FWIW, I'd rather have it done as
	error = vfs_create(...);
	if (!error)
		error = vfs_open(path, f);

	if (unlikely(error)) {
		fput(f);
		return ERR_PTR(error);
	}
	return f;

Re: [PATCH 8/8] NFSD: Instantiate a struct file when creating a regular NFSv4 file

[Al Viro]

On Fri, May 13, 2022 at 10:20:18PM +0000, Al Viro wrote:

> Yuck.  dget_parent() is not entirely without valid uses, but this isn't
> one.  It's for the cases when parent is *not* stable and you need to grab
> what had been the parent at some point (even though it might not be the
> parent anymore by the time dget_parent() returns).  Here you seriously
> depend upon it remaining the parent of that sucker all the way through -
> otherwise vfs_create() would break.  And you really, really depend upon
> its survival - the caller is holding it locked, so they would better
> have it pinned.

As an aside, the reason why vfs_create() takes inode of parent directory
and dentry of child is basically that it's easier to describe the locking
rules that way: vfs_create(..., dir, child, ...) must be called with
1) dir being held by caller (exclusive) and
2) child->d_parent->d_inode == dir, which is stabilized by (1)

inode of parent directory is a redundant argument - it can be easily
derived from the child dentry, for all that family.  The only real
objection against dropping it from vfs_create() and friends is that
having rules described as "inode of parent dentry of child must be held
exclusive by the caller" invites breakage along the lines of

	parent = dget_parent(child);
	inode_lock(d_inode(parent));
	vfs_create(..., child, ...);	// WRONG
	inode_unlock(d_inode(parent));
	dput(parent);

which *seems* to match the rules, but actually breaks them badly -
'parent' in the snippet above might be no longer related to child by the
time dget_parent() returns it, so we end up calling vfs_create() with
wrong directory locked, child->d_parent being completely unstable, etc.
Note that the difference from your code (which is correct, if redundant) is
rather subtle.

If you have any suggestions how to describe these locking rules without
an explicit inode-of-parent argument, I would really like to hear those.
The best I'd been able to come up with had been "there's an inode
locked exclusive by the caller that had been observed to be equal to
child->d_parent->d_inode at some point after it had been locked", which
is both cumbersome and confusing...

Re: [PATCH 8/8] NFSD: Instantiate a struct file when creating a regular NFSv4 file

[Chuck Lever III]

> On May 14, 2022, at 12:44 AM, Al Viro <viro@zeniv.linux.org.uk> wrote:
> 
> On Fri, May 13, 2022 at 10:20:18PM +0000, Al Viro wrote:
> 
>> Yuck.  dget_parent() is not entirely without valid uses, but this isn't
>> one.  It's for the cases when parent is *not* stable and you need to grab
>> what had been the parent at some point (even though it might not be the
>> parent anymore by the time dget_parent() returns).  Here you seriously
>> depend upon it remaining the parent of that sucker all the way through -
>> otherwise vfs_create() would break.  And you really, really depend upon
>> its survival - the caller is holding it locked, so they would better
>> have it pinned.
> 
> As an aside, the reason why vfs_create() takes inode of parent directory
> and dentry of child is basically that it's easier to describe the locking
> rules that way: vfs_create(..., dir, child, ...) must be called with
> 1) dir being held by caller (exclusive) and
> 2) child->d_parent->d_inode == dir, which is stabilized by (1)
> 
> inode of parent directory is a redundant argument - it can be easily
> derived from the child dentry, for all that family.  The only real
> objection against dropping it from vfs_create() and friends is that
> having rules described as "inode of parent dentry of child must be held
> exclusive by the caller" invites breakage along the lines of
> 
> 	parent = dget_parent(child);
> 	inode_lock(d_inode(parent));
> 	vfs_create(..., child, ...);	// WRONG
> 	inode_unlock(d_inode(parent));
> 	dput(parent);
> 
> which *seems* to match the rules, but actually breaks them badly -
> 'parent' in the snippet above might be no longer related to child by the
> time dget_parent() returns it, so we end up calling vfs_create() with
> wrong directory locked, child->d_parent being completely unstable, etc.
> Note that the difference from your code (which is correct, if redundant) is
> rather subtle.

I'm not sure I have anything informed to say, but here are some
random thoughts:

vfs_create() has to work for both exclusive and non-exclusive
create requests, but its caller is responsible for those semantics.
So perhaps the current vfs_create() synopsis is correct (though as
you point out above, the locking requirements could be documented
a little better?).

NFSD also has the interesting problem of re-exporting NFS mounts.
We'd like the exclusiveness of the creation request to be passed
through to the originating NFS server properly (maybe via
->atomic_open?). Perhaps that's for another day.

But locking the parent means that file creation is serialized,
which for large directories or workloads like "tar", impacts
throughput. Again, perhaps that's for another day.


> If you have any suggestions how to describe these locking rules without
> an explicit inode-of-parent argument, I would really like to hear those.
> The best I'd been able to come up with had been "there's an inode
> locked exclusive by the caller that had been observed to be equal to
> child->d_parent->d_inode at some point after it had been locked", which
> is both cumbersome and confusing...

I have a v2 of my patch which hopefully addresses your comments
(many thanks for those). I'm not sure I captured the gist of your
documentation request though. Looking at some other callers of
vfs_create() such as do_mknodat(), dentry_create's use of the
@path argument might be, well, unexpected. If it looks insane to
you, it can be adjusted.

I'll post v2 shortly and we can iterate on that.


--
Chuck Lever