Merge tag 'nfs-for-4.4-2' of git://git.linux-nfs.org/projects/trondmy/linux-nfs

Pull NFS client bugfixes from Trond Myklebust:
 "Highlights include:

  Stable patches:
   - Fix a NFSv4 callback identifier leak that was also causing client
     crashes
   - Fix NFSv4 callback decoding issues when incoming requests are
     truncated
   - Don't declare the attribute cache valid when we call
     nfs_update_inode with an empty attribute structure.
   - Resend LAYOUTGET when there is a race that changes the seqid

  Bugfixes:
   - Fix a number of issues with the NFSv4.2 CLONE ioctl()
   - Properly set NFS v4.2 NFSDBG_FACILITY
   - NFSv4 referrals are broken; Cleanup FATTR4_WORD0_FS_LOCATIONS after
     decoding success
   - Use sliding delay when LAYOUTGET gets NFS4ERR_DELAY
   - Ensure that attrcache is revalidated after a SETATTR"

* tag 'nfs-for-4.4-2' of git://git.linux-nfs.org/projects/trondmy/linux-nfs:
  nfs4: resend LAYOUTGET when there is a race that changes the seqid
  nfs: if we have no valid attrs, then don't declare the attribute cache valid
  nfs: ensure that attrcache is revalidated after a SETATTR
  nfs4: limit callback decoding to received bytes
  nfs4: start callback_ident at idr 1
  nfs: use sliding delay when LAYOUTGET gets NFS4ERR_DELAY
  NFS4: Cleanup FATTR4_WORD0_FS_LOCATIONS after decoding success
  NFS: Properly set NFS v4.2 NFSDBG_FACILITY
  nfs: reduce the amount of ifdefs for v4.2 in nfs4file.c
  nfs: use btrfs ioctl defintions for clone
  nfs: allow intra-file CLONE
  nfs: offer native ioctls even if CONFIG_COMPAT is set
  nfs: pass on count for CLONE operations

Author: torvalds

fs/nfs/callback_xdr.c

@@ -78,7 +78,8 @@ static __be32 *read_buf(struct xdr_stream *xdr, int nbytes)
 
 	p = xdr_inline_decode(xdr, nbytes);
 	if (unlikely(p == NULL))
-		printk(KERN_WARNING "NFS: NFSv4 callback reply buffer overflowed!\n");
+		printk(KERN_WARNING "NFS: NFSv4 callback reply buffer overflowed "
+							"or truncated request.\n");
 	return p;
 }
 
@@ -889,6 +890,7 @@ static __be32 nfs4_callback_compound(struct svc_rqst *rqstp, void *argp, void *r
 	struct cb_compound_hdr_arg hdr_arg = { 0 };
 	struct cb_compound_hdr_res hdr_res = { NULL };
 	struct xdr_stream xdr_in, xdr_out;
+	struct xdr_buf *rq_arg = &rqstp->rq_arg;
 	__be32 *p, status;
 	struct cb_process_state cps = {
 		.drc_status = 0,
@@ -900,7 +902,8 @@ static __be32 nfs4_callback_compound(struct svc_rqst *rqstp, void *argp, void *r
 
 	dprintk("%s: start\n", __func__);
 
-	xdr_init_decode(&xdr_in, &rqstp->rq_arg, rqstp->rq_arg.head[0].iov_base);
+	rq_arg->len = rq_arg->head[0].iov_len + rq_arg->page_len;
+	xdr_init_decode(&xdr_in, rq_arg, rq_arg->head[0].iov_base);
 
 	p = (__be32*)((char *)rqstp->rq_res.head[0].iov_base + rqstp->rq_res.head[0].iov_len);
 	xdr_init_encode(&xdr_out, &rqstp->rq_res, p);

fs/nfs/inode.c

@@ -618,7 +618,10 @@ void nfs_setattr_update_inode(struct inode *inode, struct iattr *attr,
 		nfs_inc_stats(inode, NFSIOS_SETATTRTRUNC);
 		nfs_vmtruncate(inode, attr->ia_size);
 	}
-	nfs_update_inode(inode, fattr);
+	if (fattr->valid)
+		nfs_update_inode(inode, fattr);
+	else
+		NFS_I(inode)->cache_validity |= NFS_INO_INVALID_ATTR;
 	spin_unlock(&inode->i_lock);
 }
 EXPORT_SYMBOL_GPL(nfs_setattr_update_inode);
@@ -1824,7 +1827,11 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr)
 		if ((long)fattr->gencount - (long)nfsi->attr_gencount > 0)
 			nfsi->attr_gencount = fattr->gencount;
 	}
-	invalid &= ~NFS_INO_INVALID_ATTR;
+
+	/* Don't declare attrcache up to date if there were no attrs! */
+	if (fattr->valid != 0)
+		invalid &= ~NFS_INO_INVALID_ATTR;
+
 	/* Don't invalidate the data if we were to blame */
 	if (!(S_ISREG(inode->i_mode) || S_ISDIR(inode->i_mode)
 				|| S_ISLNK(inode->i_mode)))

fs/nfs/nfs42proc.c

@@ -14,7 +14,7 @@
 #include "pnfs.h"
 #include "internal.h"
 
-#define NFSDBG_FACILITY NFSDBG_PNFS
+#define NFSDBG_FACILITY NFSDBG_PROC
 
 static int nfs42_set_rw_stateid(nfs4_stateid *dst, struct file *file,
 				fmode_t fmode)
@@ -284,6 +284,7 @@ static int _nfs42_proc_clone(struct rpc_message *msg, struct file *src_f,
 		.dst_fh = NFS_FH(dst_inode),
 		.src_offset = src_offset,
 		.dst_offset = dst_offset,
+		.count = count,
 		.dst_bitmask = server->cache_consistency_bitmask,
 	};
 	struct nfs42_clone_res res = {

fs/nfs/nfs4client.c

@@ -33,7 +33,7 @@ static int nfs_get_cb_ident_idr(struct nfs_client *clp, int minorversion)
 		return ret;
 	idr_preload(GFP_KERNEL);
 	spin_lock(&nn->nfs_client_lock);
-	ret = idr_alloc(&nn->cb_ident_idr, clp, 0, 0, GFP_NOWAIT);
+	ret = idr_alloc(&nn->cb_ident_idr, clp, 1, 0, GFP_NOWAIT);
 	if (ret >= 0)
 		clp->cl_cb_ident = ret;
 	spin_unlock(&nn->nfs_client_lock);

fs/nfs/nfs4file.c

@@ -7,6 +7,7 @@
 #include <linux/file.h>
 #include <linux/falloc.h>
 #include <linux/nfs_fs.h>
+#include <uapi/linux/btrfs.h>	/* BTRFS_IOC_CLONE/BTRFS_IOC_CLONE_RANGE */
 #include "delegation.h"
 #include "internal.h"
 #include "iostat.h"
@@ -203,6 +204,7 @@ nfs42_ioctl_clone(struct file *dst_file, unsigned long srcfd,
 	struct fd src_file;
 	struct inode *src_inode;
 	unsigned int bs = server->clone_blksize;
+	bool same_inode = false;
 	int ret;
 
 	/* dst file must be opened for writing */
@@ -221,10 +223,8 @@ nfs42_ioctl_clone(struct file *dst_file, unsigned long srcfd,
 
 	src_inode = file_inode(src_file.file);
 
-	/* src and dst must be different files */
-	ret = -EINVAL;
 	if (src_inode == dst_inode)
-		goto out_fput;
+		same_inode = true;
 
 	/* src file must be opened for reading */
 	if (!(src_file.file->f_mode & FMODE_READ))
@@ -249,8 +249,16 @@ nfs42_ioctl_clone(struct file *dst_file, unsigned long srcfd,
 			goto out_fput;
 	}
 
+	/* verify if ranges are overlapped within the same file */
+	if (same_inode) {
+		if (dst_off + count > src_off && dst_off < src_off + count)
+			goto out_fput;
+	}
+
 	/* XXX: do we lock at all? what if server needs CB_RECALL_LAYOUT? */
-	if (dst_inode < src_inode) {
+	if (same_inode) {
+		mutex_lock(&src_inode->i_mutex);
+	} else if (dst_inode < src_inode) {
 		mutex_lock_nested(&dst_inode->i_mutex, I_MUTEX_PARENT);
 		mutex_lock_nested(&src_inode->i_mutex, I_MUTEX_CHILD);
 	} else {
@@ -275,7 +283,9 @@ nfs42_ioctl_clone(struct file *dst_file, unsigned long srcfd,
 		truncate_inode_pages_range(&dst_inode->i_data, dst_off, dst_off + count - 1);
 
 out_unlock:
-	if (dst_inode < src_inode) {
+	if (same_inode) {
+		mutex_unlock(&src_inode->i_mutex);
+	} else if (dst_inode < src_inode) {
 		mutex_unlock(&src_inode->i_mutex);
 		mutex_unlock(&dst_inode->i_mutex);
 	} else {
@@ -291,46 +301,31 @@ nfs42_ioctl_clone(struct file *dst_file, unsigned long srcfd,
 
 static long nfs42_ioctl_clone_range(struct file *dst_file, void __user *argp)
 {
-	struct nfs_ioctl_clone_range_args args;
+	struct btrfs_ioctl_clone_range_args args;
 
 	if (copy_from_user(&args, argp, sizeof(args)))
 		return -EFAULT;
 
-	return nfs42_ioctl_clone(dst_file, args.src_fd, args.src_off, args.dst_off, args.count);
-}
-#else
-static long nfs42_ioctl_clone(struct file *dst_file, unsigned long srcfd,
-		u64 src_off, u64 dst_off, u64 count)
-{
-	return -ENOTTY;
-}
-
-static long nfs42_ioctl_clone_range(struct file *dst_file, void __user *argp)
-{
-	return -ENOTTY;
+	return nfs42_ioctl_clone(dst_file, args.src_fd, args.src_offset,
+				 args.dest_offset, args.src_length);
 }
-#endif /* CONFIG_NFS_V4_2 */
 
 long nfs4_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
 {
 	void __user *argp = (void __user *)arg;
 
 	switch (cmd) {
-	case NFS_IOC_CLONE:
+	case BTRFS_IOC_CLONE:
 		return nfs42_ioctl_clone(file, arg, 0, 0, 0);
-	case NFS_IOC_CLONE_RANGE:
+	case BTRFS_IOC_CLONE_RANGE:
 		return nfs42_ioctl_clone_range(file, argp);
 	}
 
 	return -ENOTTY;
 }
+#endif /* CONFIG_NFS_V4_2 */
 
 const struct file_operations nfs4_file_operations = {
-#ifdef CONFIG_NFS_V4_2
-	.llseek		= nfs4_file_llseek,
-#else
-	.llseek		= nfs_file_llseek,
-#endif
 	.read_iter	= nfs_file_read,
 	.write_iter	= nfs_file_write,
 	.mmap		= nfs_file_mmap,
@@ -342,14 +337,14 @@ const struct file_operations nfs4_file_operations = {
 	.flock		= nfs_flock,
 	.splice_read	= nfs_file_splice_read,
 	.splice_write	= iter_file_splice_write,
-#ifdef CONFIG_NFS_V4_2
-	.fallocate	= nfs42_fallocate,
-#endif /* CONFIG_NFS_V4_2 */
 	.check_flags	= nfs_check_flags,
 	.setlease	= simple_nosetlease,
-#ifdef CONFIG_COMPAT
+#ifdef CONFIG_NFS_V4_2
+	.llseek		= nfs4_file_llseek,
+	.fallocate	= nfs42_fallocate,
 	.unlocked_ioctl = nfs4_ioctl,
-#else
 	.compat_ioctl	= nfs4_ioctl,
-#endif /* CONFIG_COMPAT */
+#else
+	.llseek		= nfs_file_llseek,
+#endif
 };

fs/nfs/nfs4proc.c

@@ -7866,7 +7866,7 @@ static void nfs4_layoutget_done(struct rpc_task *task, void *calldata)
 			spin_unlock(&inode->i_lock);
 		goto out_restart;
 	}
-	if (nfs4_async_handle_error(task, server, state, NULL) == -EAGAIN)
+	if (nfs4_async_handle_error(task, server, state, &lgp->timeout) == -EAGAIN)
 		goto out_restart;
 out:
 	dprintk("<-- %s\n", __func__);

fs/nfs/nfs4xdr.c

@@ -3615,6 +3615,7 @@ static int decode_attr_fs_locations(struct xdr_stream *xdr, uint32_t *bitmap, st
 	status = 0;
 	if (unlikely(!(bitmap[0] & FATTR4_WORD0_FS_LOCATIONS)))
 		goto out;
+	bitmap[0] &= ~FATTR4_WORD0_FS_LOCATIONS;
 	status = -EIO;
 	/* Ignore borken servers that return unrequested attrs */
 	if (unlikely(res == NULL))

fs/nfs/pnfs.c

@@ -872,33 +872,38 @@ send_layoutget(struct pnfs_layout_hdr *lo,
 
 	dprintk("--> %s\n", __func__);
 
-	lgp = kzalloc(sizeof(*lgp), gfp_flags);
-	if (lgp == NULL)
-		return NULL;
+	/*
+	 * Synchronously retrieve layout information from server and
+	 * store in lseg. If we race with a concurrent seqid morphing
+	 * op, then re-send the LAYOUTGET.
+	 */
+	do {
+		lgp = kzalloc(sizeof(*lgp), gfp_flags);
+		if (lgp == NULL)
+			return NULL;
+
+		i_size = i_size_read(ino);
+
+		lgp->args.minlength = PAGE_CACHE_SIZE;
+		if (lgp->args.minlength > range->length)
+			lgp->args.minlength = range->length;
+		if (range->iomode == IOMODE_READ) {
+			if (range->offset >= i_size)
+				lgp->args.minlength = 0;
+			else if (i_size - range->offset < lgp->args.minlength)
+				lgp->args.minlength = i_size - range->offset;
+		}
+		lgp->args.maxcount = PNFS_LAYOUT_MAXSIZE;
+		lgp->args.range = *range;
+		lgp->args.type = server->pnfs_curr_ld->id;
+		lgp->args.inode = ino;
+		lgp->args.ctx = get_nfs_open_context(ctx);
+		lgp->gfp_flags = gfp_flags;
+		lgp->cred = lo->plh_lc_cred;
 
-	i_size = i_size_read(ino);
+		lseg = nfs4_proc_layoutget(lgp, gfp_flags);
+	} while (lseg == ERR_PTR(-EAGAIN));
 
-	lgp->args.minlength = PAGE_CACHE_SIZE;
-	if (lgp->args.minlength > range->length)
-		lgp->args.minlength = range->length;
-	if (range->iomode == IOMODE_READ) {
-		if (range->offset >= i_size)
-			lgp->args.minlength = 0;
-		else if (i_size - range->offset < lgp->args.minlength)
-			lgp->args.minlength = i_size - range->offset;
-	}
-	lgp->args.maxcount = PNFS_LAYOUT_MAXSIZE;
-	lgp->args.range = *range;
-	lgp->args.type = server->pnfs_curr_ld->id;
-	lgp->args.inode = ino;
-	lgp->args.ctx = get_nfs_open_context(ctx);
-	lgp->gfp_flags = gfp_flags;
-	lgp->cred = lo->plh_lc_cred;
-
-	/* Synchronously retrieve layout information from server and
-	 * store in lseg.
-	 */
-	lseg = nfs4_proc_layoutget(lgp, gfp_flags);
 	if (IS_ERR(lseg)) {
 		switch (PTR_ERR(lseg)) {
 		case -ENOMEM:
@@ -1687,6 +1692,7 @@ pnfs_layout_process(struct nfs4_layoutget *lgp)
 		/* existing state ID, make sure the sequence number matches. */
 		if (pnfs_layout_stateid_blocked(lo, &res->stateid)) {
 			dprintk("%s forget reply due to sequence\n", __func__);
+			status = -EAGAIN;
 			goto out_forget_reply;
 		}
 		pnfs_set_layout_stateid(lo, &res->stateid, false);

include/linux/nfs_xdr.h

@@ -251,6 +251,7 @@ struct nfs4_layoutget {
 	struct nfs4_layoutget_res res;
 	struct rpc_cred *cred;
 	gfp_t gfp_flags;
+	long timeout;
 };
 
 struct nfs4_getdeviceinfo_args {

include/uapi/linux/nfs.h

@@ -33,17 +33,6 @@
 
 #define NFS_PIPE_DIRNAME "nfs"
 
-/* NFS ioctls */
-/* Let's follow btrfs lead on CLONE to avoid messing userspace */
-#define NFS_IOC_CLONE		_IOW(0x94, 9, int)
-#define NFS_IOC_CLONE_RANGE	_IOW(0x94, 13, int)
-
-struct nfs_ioctl_clone_range_args {
-	__s64 src_fd;
-	__u64 src_off, count;
-	__u64 dst_off;
-};
-
 /*
  * NFS stats. The good thing with these values is that NFSv3 errors are
  * a superset of NFSv2 errors (with the exception of NFSERR_WFLUSH which