merge nfs bugfixes into nfsd for-3.19 branch
In addition to nfsd bugfixes, there are some fixes in -rc5 for client
bugs that can interfere with my testing.
diff --git a/drivers/staging/android/ashmem.c b/drivers/staging/android/ashmem.c
index ad4f579..27eecfe 100644
--- a/drivers/staging/android/ashmem.c
+++ b/drivers/staging/android/ashmem.c
@@ -446,7 +446,7 @@
loff_t start = range->pgstart * PAGE_SIZE;
loff_t end = (range->pgend + 1) * PAGE_SIZE;
- do_fallocate(range->asma->file,
+ vfs_fallocate(range->asma->file,
FALLOC_FL_PUNCH_HOLE | FALLOC_FL_KEEP_SIZE,
start, end - start);
range->purged = ASHMEM_WAS_PURGED;
diff --git a/fs/ioctl.c b/fs/ioctl.c
index 8ac3fad..0bd6142 100644
--- a/fs/ioctl.c
+++ b/fs/ioctl.c
@@ -443,7 +443,7 @@
return -EINVAL;
}
- return do_fallocate(filp, FALLOC_FL_KEEP_SIZE, sr.l_start, sr.l_len);
+ return vfs_fallocate(filp, FALLOC_FL_KEEP_SIZE, sr.l_start, sr.l_len);
}
static int file_ioctl(struct file *filp, unsigned int cmd,
diff --git a/fs/lockd/mon.c b/fs/lockd/mon.c
index 9106f42..1cc6ec5 100644
--- a/fs/lockd/mon.c
+++ b/fs/lockd/mon.c
@@ -214,7 +214,7 @@
if (unlikely(res.status != 0))
status = -EIO;
if (unlikely(status < 0)) {
- printk(KERN_NOTICE "lockd: cannot monitor %s\n", nsm->sm_name);
+ pr_notice_ratelimited("lockd: cannot monitor %s\n", nsm->sm_name);
return status;
}
diff --git a/fs/nfsd/nfs4proc.c b/fs/nfsd/nfs4proc.c
index 0beb023..74fb15e 100644
--- a/fs/nfsd/nfs4proc.c
+++ b/fs/nfsd/nfs4proc.c
@@ -33,6 +33,7 @@
* SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
*/
#include <linux/file.h>
+#include <linux/falloc.h>
#include <linux/slab.h>
#include "idmap.h"
@@ -1014,6 +1015,44 @@
}
static __be32
+nfsd4_fallocate(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
+ struct nfsd4_fallocate *fallocate, int flags)
+{
+ __be32 status = nfserr_notsupp;
+ struct file *file;
+
+ status = nfs4_preprocess_stateid_op(SVC_NET(rqstp), cstate,
+ &fallocate->falloc_stateid,
+ WR_STATE, &file);
+ if (status != nfs_ok) {
+ dprintk("NFSD: nfsd4_fallocate: couldn't process stateid!\n");
+ return status;
+ }
+
+ status = nfsd4_vfs_fallocate(rqstp, &cstate->current_fh, file,
+ fallocate->falloc_offset,
+ fallocate->falloc_length,
+ flags);
+ fput(file);
+ return status;
+}
+
+static __be32
+nfsd4_allocate(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
+ struct nfsd4_fallocate *fallocate)
+{
+ return nfsd4_fallocate(rqstp, cstate, fallocate, 0);
+}
+
+static __be32
+nfsd4_deallocate(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
+ struct nfsd4_fallocate *fallocate)
+{
+ return nfsd4_fallocate(rqstp, cstate, fallocate,
+ FALLOC_FL_PUNCH_HOLE | FALLOC_FL_KEEP_SIZE);
+}
+
+static __be32
nfsd4_seek(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
struct nfsd4_seek *seek)
{
@@ -1929,6 +1968,18 @@
},
/* NFSv4.2 operations */
+ [OP_ALLOCATE] = {
+ .op_func = (nfsd4op_func)nfsd4_allocate,
+ .op_flags = OP_MODIFIES_SOMETHING | OP_CACHEME,
+ .op_name = "OP_ALLOCATE",
+ .op_rsize_bop = (nfsd4op_rsize)nfsd4_write_rsize,
+ },
+ [OP_DEALLOCATE] = {
+ .op_func = (nfsd4op_func)nfsd4_deallocate,
+ .op_flags = OP_MODIFIES_SOMETHING | OP_CACHEME,
+ .op_name = "OP_DEALLOCATE",
+ .op_rsize_bop = (nfsd4op_rsize)nfsd4_write_rsize,
+ },
[OP_SEEK] = {
.op_func = (nfsd4op_func)nfsd4_seek,
.op_name = "OP_SEEK",
diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c
index e9c3afe..1379d86 100644
--- a/fs/nfsd/nfs4state.c
+++ b/fs/nfsd/nfs4state.c
@@ -275,9 +275,11 @@
return x;
}
-static void nfsd4_free_file(struct nfs4_file *f)
+static void nfsd4_free_file_rcu(struct rcu_head *rcu)
{
- kmem_cache_free(file_slab, f);
+ struct nfs4_file *fp = container_of(rcu, struct nfs4_file, fi_rcu);
+
+ kmem_cache_free(file_slab, fp);
}
static inline void
@@ -286,9 +288,10 @@
might_lock(&state_lock);
if (atomic_dec_and_lock(&fi->fi_ref, &state_lock)) {
- hlist_del(&fi->fi_hash);
+ hlist_del_rcu(&fi->fi_hash);
spin_unlock(&state_lock);
- nfsd4_free_file(fi);
+ WARN_ON_ONCE(!list_empty(&fi->fi_delegations));
+ call_rcu(&fi->fi_rcu, nfsd4_free_file_rcu);
}
}
@@ -1440,7 +1443,7 @@
list_add(&new->se_perclnt, &clp->cl_sessions);
spin_unlock(&clp->cl_lock);
- if (cses->flags & SESSION4_BACK_CHAN) {
+ {
struct sockaddr *sa = svc_addr(rqstp);
/*
* This is a little silly; with sessions there's no real
@@ -3057,10 +3060,9 @@
}
/* OPEN Share state helper functions */
-static void nfsd4_init_file(struct nfs4_file *fp, struct knfsd_fh *fh)
+static void nfsd4_init_file(struct knfsd_fh *fh, unsigned int hashval,
+ struct nfs4_file *fp)
{
- unsigned int hashval = file_hashval(fh);
-
lockdep_assert_held(&state_lock);
atomic_set(&fp->fi_ref, 1);
@@ -3073,7 +3075,7 @@
fp->fi_share_deny = 0;
memset(fp->fi_fds, 0, sizeof(fp->fi_fds));
memset(fp->fi_access, 0, sizeof(fp->fi_access));
- hlist_add_head(&fp->fi_hash, &file_hashtbl[hashval]);
+ hlist_add_head_rcu(&fp->fi_hash, &file_hashtbl[hashval]);
}
void
@@ -3294,17 +3296,14 @@
/* search file_hashtbl[] for file */
static struct nfs4_file *
-find_file_locked(struct knfsd_fh *fh)
+find_file_locked(struct knfsd_fh *fh, unsigned int hashval)
{
- unsigned int hashval = file_hashval(fh);
struct nfs4_file *fp;
- lockdep_assert_held(&state_lock);
-
- hlist_for_each_entry(fp, &file_hashtbl[hashval], fi_hash) {
+ hlist_for_each_entry_rcu(fp, &file_hashtbl[hashval], fi_hash) {
if (nfsd_fh_match(&fp->fi_fhandle, fh)) {
- get_nfs4_file(fp);
- return fp;
+ if (atomic_inc_not_zero(&fp->fi_ref))
+ return fp;
}
}
return NULL;
@@ -3314,10 +3313,11 @@
find_file(struct knfsd_fh *fh)
{
struct nfs4_file *fp;
+ unsigned int hashval = file_hashval(fh);
- spin_lock(&state_lock);
- fp = find_file_locked(fh);
- spin_unlock(&state_lock);
+ rcu_read_lock();
+ fp = find_file_locked(fh, hashval);
+ rcu_read_unlock();
return fp;
}
@@ -3325,11 +3325,18 @@
find_or_add_file(struct nfs4_file *new, struct knfsd_fh *fh)
{
struct nfs4_file *fp;
+ unsigned int hashval = file_hashval(fh);
+
+ rcu_read_lock();
+ fp = find_file_locked(fh, hashval);
+ rcu_read_unlock();
+ if (fp)
+ return fp;
spin_lock(&state_lock);
- fp = find_file_locked(fh);
- if (fp == NULL) {
- nfsd4_init_file(new, fh);
+ fp = find_file_locked(fh, hashval);
+ if (likely(fp == NULL)) {
+ nfsd4_init_file(fh, hashval, new);
fp = new;
}
spin_unlock(&state_lock);
@@ -4127,7 +4134,7 @@
nfs4_put_stateowner(so);
}
if (open->op_file)
- nfsd4_free_file(open->op_file);
+ kmem_cache_free(file_slab, open->op_file);
if (open->op_stp)
nfs4_put_stid(&open->op_stp->st_stid);
}
diff --git a/fs/nfsd/nfs4xdr.c b/fs/nfsd/nfs4xdr.c
index eeea7a9..0622d4f 100644
--- a/fs/nfsd/nfs4xdr.c
+++ b/fs/nfsd/nfs4xdr.c
@@ -1514,6 +1514,23 @@
}
static __be32
+nfsd4_decode_fallocate(struct nfsd4_compoundargs *argp,
+ struct nfsd4_fallocate *fallocate)
+{
+ DECODE_HEAD;
+
+ status = nfsd4_decode_stateid(argp, &fallocate->falloc_stateid);
+ if (status)
+ return status;
+
+ READ_BUF(16);
+ p = xdr_decode_hyper(p, &fallocate->falloc_offset);
+ xdr_decode_hyper(p, &fallocate->falloc_length);
+
+ DECODE_TAIL;
+}
+
+static __be32
nfsd4_decode_seek(struct nfsd4_compoundargs *argp, struct nfsd4_seek *seek)
{
DECODE_HEAD;
@@ -1604,10 +1621,10 @@
[OP_RECLAIM_COMPLETE] = (nfsd4_dec)nfsd4_decode_reclaim_complete,
/* new operations for NFSv4.2 */
- [OP_ALLOCATE] = (nfsd4_dec)nfsd4_decode_notsupp,
+ [OP_ALLOCATE] = (nfsd4_dec)nfsd4_decode_fallocate,
[OP_COPY] = (nfsd4_dec)nfsd4_decode_notsupp,
[OP_COPY_NOTIFY] = (nfsd4_dec)nfsd4_decode_notsupp,
- [OP_DEALLOCATE] = (nfsd4_dec)nfsd4_decode_notsupp,
+ [OP_DEALLOCATE] = (nfsd4_dec)nfsd4_decode_fallocate,
[OP_IO_ADVISE] = (nfsd4_dec)nfsd4_decode_notsupp,
[OP_LAYOUTERROR] = (nfsd4_dec)nfsd4_decode_notsupp,
[OP_LAYOUTSTATS] = (nfsd4_dec)nfsd4_decode_notsupp,
diff --git a/fs/nfsd/state.h b/fs/nfsd/state.h
index 2712042..9d3be37 100644
--- a/fs/nfsd/state.h
+++ b/fs/nfsd/state.h
@@ -463,17 +463,24 @@
/*
* nfs4_file: a file opened by some number of (open) nfs4_stateowners.
*
- * These objects are global. nfsd only keeps one instance of a nfs4_file per
- * inode (though it may keep multiple file descriptors open per inode). These
- * are tracked in the file_hashtbl which is protected by the state_lock
- * spinlock.
+ * These objects are global. nfsd keeps one instance of a nfs4_file per
+ * filehandle (though it may keep multiple file descriptors for each). Each
+ * inode can have multiple filehandles associated with it, so there is
+ * (potentially) a many to one relationship between this struct and struct
+ * inode.
+ *
+ * These are hashed by filehandle in the file_hashtbl, which is protected by
+ * the global state_lock spinlock.
*/
struct nfs4_file {
atomic_t fi_ref;
spinlock_t fi_lock;
- struct hlist_node fi_hash; /* hash by "struct inode *" */
+ struct hlist_node fi_hash; /* hash on fi_fhandle */
struct list_head fi_stateids;
- struct list_head fi_delegations;
+ union {
+ struct list_head fi_delegations;
+ struct rcu_head fi_rcu;
+ };
/* One each for O_RDONLY, O_WRONLY, O_RDWR: */
struct file * fi_fds[3];
/*
diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c
index 989129e..f199961 100644
--- a/fs/nfsd/vfs.c
+++ b/fs/nfsd/vfs.c
@@ -16,6 +16,7 @@
#include <linux/fs.h>
#include <linux/file.h>
#include <linux/splice.h>
+#include <linux/falloc.h>
#include <linux/fcntl.h>
#include <linux/namei.h>
#include <linux/delay.h>
@@ -533,6 +534,26 @@
}
#endif
+__be32 nfsd4_vfs_fallocate(struct svc_rqst *rqstp, struct svc_fh *fhp,
+ struct file *file, loff_t offset, loff_t len,
+ int flags)
+{
+ __be32 err;
+ int error;
+
+ if (!S_ISREG(file_inode(file)->i_mode))
+ return nfserr_inval;
+
+ err = nfsd_permission(rqstp, fhp->fh_export, fhp->fh_dentry, NFSD_MAY_WRITE);
+ if (err)
+ return err;
+
+ error = vfs_fallocate(file, flags, offset, len);
+ if (!error)
+ error = commit_metadata(fhp);
+
+ return nfserrno(error);
+}
#endif /* defined(CONFIG_NFSD_V4) */
#ifdef CONFIG_NFSD_V3
@@ -938,6 +959,7 @@
int stable = *stablep;
int use_wgather;
loff_t pos = offset;
+ loff_t end = LLONG_MAX;
unsigned int pflags = current->flags;
if (rqstp->rq_local)
@@ -969,10 +991,13 @@
fsnotify_modify(file);
if (stable) {
- if (use_wgather)
+ if (use_wgather) {
host_err = wait_for_concurrent_writes(file);
- else
- host_err = vfs_fsync_range(file, offset, offset+*cnt, 0);
+ } else {
+ if (*cnt)
+ end = offset + *cnt - 1;
+ host_err = vfs_fsync_range(file, offset, end, 0);
+ }
}
out_nfserr:
diff --git a/fs/nfsd/vfs.h b/fs/nfsd/vfs.h
index c2ff3f1..7ffdb14 100644
--- a/fs/nfsd/vfs.h
+++ b/fs/nfsd/vfs.h
@@ -54,6 +54,8 @@
#ifdef CONFIG_NFSD_V4
__be32 nfsd4_set_nfs4_label(struct svc_rqst *, struct svc_fh *,
struct xdr_netobj *);
+__be32 nfsd4_vfs_fallocate(struct svc_rqst *, struct svc_fh *,
+ struct file *, loff_t, loff_t, int);
#endif /* CONFIG_NFSD_V4 */
__be32 nfsd_create(struct svc_rqst *, struct svc_fh *,
char *name, int len, struct iattr *attrs,
diff --git a/fs/nfsd/xdr4.h b/fs/nfsd/xdr4.h
index 5720e94..90a5925 100644
--- a/fs/nfsd/xdr4.h
+++ b/fs/nfsd/xdr4.h
@@ -428,6 +428,13 @@
u32 rca_one_fs;
};
+struct nfsd4_fallocate {
+ /* request */
+ stateid_t falloc_stateid;
+ loff_t falloc_offset;
+ u64 falloc_length;
+};
+
struct nfsd4_seek {
/* request */
stateid_t seek_stateid;
@@ -486,6 +493,8 @@
struct nfsd4_free_stateid free_stateid;
/* NFSv4.2 */
+ struct nfsd4_fallocate allocate;
+ struct nfsd4_fallocate deallocate;
struct nfsd4_seek seek;
} u;
struct nfs4_replay * replay;
diff --git a/fs/open.c b/fs/open.c
index de92c13..192c429 100644
--- a/fs/open.c
+++ b/fs/open.c
@@ -222,7 +222,7 @@
#endif /* BITS_PER_LONG == 32 */
-int do_fallocate(struct file *file, int mode, loff_t offset, loff_t len)
+int vfs_fallocate(struct file *file, int mode, loff_t offset, loff_t len)
{
struct inode *inode = file_inode(file);
long ret;
@@ -298,6 +298,7 @@
sb_end_write(inode->i_sb);
return ret;
}
+EXPORT_SYMBOL_GPL(vfs_fallocate);
SYSCALL_DEFINE4(fallocate, int, fd, int, mode, loff_t, offset, loff_t, len)
{
@@ -305,7 +306,7 @@
int error = -EBADF;
if (f.file) {
- error = do_fallocate(f.file, mode, offset, len);
+ error = vfs_fallocate(f.file, mode, offset, len);
fdput(f);
}
return error;
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 9ab779e..4439c99 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -2060,7 +2060,7 @@
extern long vfs_truncate(struct path *, loff_t);
extern int do_truncate(struct dentry *, loff_t start, unsigned int time_attrs,
struct file *filp);
-extern int do_fallocate(struct file *file, int mode, loff_t offset,
+extern int vfs_fallocate(struct file *file, int mode, loff_t offset,
loff_t len);
extern long do_sys_open(int dfd, const char __user *filename, int flags,
umode_t mode);
diff --git a/mm/madvise.c b/mm/madvise.c
index 0938b30..a271adc 100644
--- a/mm/madvise.c
+++ b/mm/madvise.c
@@ -326,7 +326,7 @@
*/
get_file(f);
up_read(¤t->mm->mmap_sem);
- error = do_fallocate(f,
+ error = vfs_fallocate(f,
FALLOC_FL_PUNCH_HOLE | FALLOC_FL_KEEP_SIZE,
offset, end - start);
fput(f);
diff --git a/net/sunrpc/svc.c b/net/sunrpc/svc.c
index ca8a795..349c98f 100644
--- a/net/sunrpc/svc.c
+++ b/net/sunrpc/svc.c
@@ -189,7 +189,7 @@
return err;
for_each_online_cpu(cpu) {
- BUG_ON(pidx > maxpools);
+ BUG_ON(pidx >= maxpools);
m->to_pool[cpu] = pidx;
m->pool_to[pidx] = cpu;
pidx++;
diff --git a/net/sunrpc/xdr.c b/net/sunrpc/xdr.c
index 290af97..1cb6124 100644
--- a/net/sunrpc/xdr.c
+++ b/net/sunrpc/xdr.c
@@ -617,9 +617,10 @@
fraglen = min_t(int, buf->len - len, tail->iov_len);
tail->iov_len -= fraglen;
buf->len -= fraglen;
- if (tail->iov_len && buf->len == len) {
+ if (tail->iov_len) {
xdr->p = tail->iov_base + tail->iov_len;
- /* xdr->end, xdr->iov should be set already */
+ WARN_ON_ONCE(!xdr->end);
+ WARN_ON_ONCE(!xdr->iov);
return;
}
WARN_ON_ONCE(fraglen);
@@ -631,11 +632,11 @@
old = new + fraglen;
xdr->page_ptr -= (old >> PAGE_SHIFT) - (new >> PAGE_SHIFT);
- if (buf->page_len && buf->len == len) {
+ if (buf->page_len) {
xdr->p = page_address(*xdr->page_ptr);
xdr->end = (void *)xdr->p + PAGE_SIZE;
xdr->p = (void *)xdr->p + (new % PAGE_SIZE);
- /* xdr->iov should already be NULL */
+ WARN_ON_ONCE(xdr->iov);
return;
}
if (fraglen) {