Merge tag 'nfsd-6.17' of git://git.kernel.org/pub/scm/linux/kernel/git/cel/linux

Pull nfsd updates from Chuck Lever:
 "NFSD is finally able to offer write delegations to clients that open
  files with O_WRONLY, thanks to patches from Dai Ngo. We're expecting
  this to accelerate a few interesting corner cases.

  The cap on the number of operations per NFSv4 COMPOUND has been
  lifted. Now, clients that send COMPOUNDs containing dozens of
  operations (for example, a long stream of LOOKUP operations to walk a
  pathname in a single round trip) will no longer be rejected.

  This release re-enables the ability for NFSD to perform NFSv4.2 COPY
  operations asynchronously. This feature has been disabled to mitigate
  the risk of denial-of-service when too many such requests arrive.

  Many thanks to the contributors, reviewers, testers, and bug reporters
  who participated during the v6.17 development cycle"

* tag 'nfsd-6.17' of git://git.kernel.org/pub/scm/linux/kernel/git/cel/linux: (32 commits)
  nfsd: Drop dprintk in blocklayout xdr functions
  sunrpc: make svc_tcp_sendmsg() take a signed sentp pointer
  sunrpc: rearrange struct svc_rqst for fewer cachelines
  sunrpc: return better error in svcauth_gss_accept() on alloc failure
  sunrpc: reset rq_accept_statp when starting a new RPC
  sunrpc: remove SVC_SYSERR
  sunrpc: fix handling of unknown auth status codes
  NFSD: Simplify struct knfsd_fh
  NFSD: Access a knfsd_fh's fsid by pointer
  Revert "NFSD: Force all NFSv4.2 COPY requests to be synchronous"
  NFSD: Avoid multiple -Wflex-array-member-not-at-end warnings
  NFSD: Use vfs_iocb_iter_write()
  NFSD: Use vfs_iocb_iter_read()
  NFSD: Clean up kdoc for nfsd_open_local_fh()
  NFSD: Clean up kdoc for nfsd_file_put_local()
  NFSD: Remove definition for trace_nfsd_ctl_maxconn
  NFSD: Remove definition for trace_nfsd_file_gc_recent
  NFSD: Remove definitions for unused trace_nfsd_file_lru trace points
  NFSD: Remove definition for trace_nfsd_file_unhash_and_queue
  nfsd: Use correct error code when decoding extents
  ...

Author: torvalds

fs/nfsd/blocklayout.c

@@ -178,11 +178,13 @@ nfsd4_block_proc_layoutcommit(struct inode *inode,
 {
 	struct iomap *iomaps;
 	int nr_iomaps;
+	__be32 nfserr;
 
-	nr_iomaps = nfsd4_block_decode_layoutupdate(lcp->lc_up_layout,
-			lcp->lc_up_len, &iomaps, i_blocksize(inode));
-	if (nr_iomaps < 0)
-		return nfserrno(nr_iomaps);
+	nfserr = nfsd4_block_decode_layoutupdate(lcp->lc_up_layout,
+			lcp->lc_up_len, &iomaps, &nr_iomaps,
+			i_blocksize(inode));
+	if (nfserr != nfs_ok)
+		return nfserr;
 
 	return nfsd4_block_commit_blocks(inode, lcp, iomaps, nr_iomaps);
 }
@@ -316,11 +318,13 @@ nfsd4_scsi_proc_layoutcommit(struct inode *inode,
 {
 	struct iomap *iomaps;
 	int nr_iomaps;
+	__be32 nfserr;
 
-	nr_iomaps = nfsd4_scsi_decode_layoutupdate(lcp->lc_up_layout,
-			lcp->lc_up_len, &iomaps, i_blocksize(inode));
-	if (nr_iomaps < 0)
-		return nfserrno(nr_iomaps);
+	nfserr = nfsd4_scsi_decode_layoutupdate(lcp->lc_up_layout,
+			lcp->lc_up_len, &iomaps, &nr_iomaps,
+			i_blocksize(inode));
+	if (nfserr != nfs_ok)
+		return nfserr;
 
 	return nfsd4_block_commit_blocks(inode, lcp, iomaps, nr_iomaps);
 }

fs/nfsd/blocklayoutxdr.c

@@ -112,35 +112,46 @@ nfsd4_block_encode_getdeviceinfo(struct xdr_stream *xdr,
 	return 0;
 }
 
-int
+/**
+ * nfsd4_block_decode_layoutupdate - decode the block layout extent array
+ * @p: pointer to the xdr data
+ * @len: number of bytes to decode
+ * @iomapp: pointer to store the decoded extent array
+ * @nr_iomapsp: pointer to store the number of extents
+ * @block_size: alignment of extent offset and length
+ *
+ * This function decodes the opaque field of the layoutupdate4 structure
+ * in a layoutcommit request for the block layout driver. The field is
+ * actually an array of extents sent by the client. It also checks that
+ * the file offset, storage offset and length of each extent are aligned
+ * by @block_size.
+ *
+ * Return values:
+ *   %nfs_ok: Successful decoding, @iomapp and @nr_iomapsp are valid
+ *   %nfserr_bad_xdr: The encoded array in @p is invalid
+ *   %nfserr_inval: An unaligned extent found
+ *   %nfserr_delay: Failed to allocate memory for @iomapp
+ */
+__be32
 nfsd4_block_decode_layoutupdate(__be32 *p, u32 len, struct iomap **iomapp,
-		u32 block_size)
+		int *nr_iomapsp, u32 block_size)
 {
 	struct iomap *iomaps;
 	u32 nr_iomaps, i;
 
-	if (len < sizeof(u32)) {
-		dprintk("%s: extent array too small: %u\n", __func__, len);
-		return -EINVAL;
-	}
+	if (len < sizeof(u32))
+		return nfserr_bad_xdr;
 	len -= sizeof(u32);
-	if (len % PNFS_BLOCK_EXTENT_SIZE) {
-		dprintk("%s: extent array invalid: %u\n", __func__, len);
-		return -EINVAL;
-	}
+	if (len % PNFS_BLOCK_EXTENT_SIZE)
+		return nfserr_bad_xdr;
 
 	nr_iomaps = be32_to_cpup(p++);
-	if (nr_iomaps != len / PNFS_BLOCK_EXTENT_SIZE) {
-		dprintk("%s: extent array size mismatch: %u/%u\n",
-			__func__, len, nr_iomaps);
-		return -EINVAL;
-	}
+	if (nr_iomaps != len / PNFS_BLOCK_EXTENT_SIZE)
+		return nfserr_bad_xdr;
 
 	iomaps = kcalloc(nr_iomaps, sizeof(*iomaps), GFP_KERNEL);
-	if (!iomaps) {
-		dprintk("%s: failed to allocate extent array\n", __func__);
-		return -ENOMEM;
-	}
+	if (!iomaps)
+		return nfserr_delay;
 
 	for (i = 0; i < nr_iomaps; i++) {
 		struct pnfs_block_extent bex;
@@ -150,26 +161,18 @@ nfsd4_block_decode_layoutupdate(__be32 *p, u32 len, struct iomap **iomapp,
 
 		p = xdr_decode_hyper(p, &bex.foff);
 		if (bex.foff & (block_size - 1)) {
-			dprintk("%s: unaligned offset 0x%llx\n",
-				__func__, bex.foff);
 			goto fail;
 		}
 		p = xdr_decode_hyper(p, &bex.len);
 		if (bex.len & (block_size - 1)) {
-			dprintk("%s: unaligned length 0x%llx\n",
-				__func__, bex.foff);
 			goto fail;
 		}
 		p = xdr_decode_hyper(p, &bex.soff);
 		if (bex.soff & (block_size - 1)) {
-			dprintk("%s: unaligned disk offset 0x%llx\n",
-				__func__, bex.soff);
 			goto fail;
 		}
 		bex.es = be32_to_cpup(p++);
 		if (bex.es != PNFS_BLOCK_READWRITE_DATA) {
-			dprintk("%s: incorrect extent state %d\n",
-				__func__, bex.es);
 			goto fail;
 		}
 
@@ -178,59 +181,71 @@ nfsd4_block_decode_layoutupdate(__be32 *p, u32 len, struct iomap **iomapp,
 	}
 
 	*iomapp = iomaps;
-	return nr_iomaps;
+	*nr_iomapsp = nr_iomaps;
+	return nfs_ok;
 fail:
 	kfree(iomaps);
-	return -EINVAL;
+	return nfserr_inval;
 }
 
-int
+/**
+ * nfsd4_scsi_decode_layoutupdate - decode the scsi layout extent array
+ * @p: pointer to the xdr data
+ * @len: number of bytes to decode
+ * @iomapp: pointer to store the decoded extent array
+ * @nr_iomapsp: pointer to store the number of extents
+ * @block_size: alignment of extent offset and length
+ *
+ * This function decodes the opaque field of the layoutupdate4 structure
+ * in a layoutcommit request for the scsi layout driver. The field is
+ * actually an array of extents sent by the client. It also checks that
+ * the offset and length of each extent are aligned by @block_size.
+ *
+ * Return values:
+ *   %nfs_ok: Successful decoding, @iomapp and @nr_iomapsp are valid
+ *   %nfserr_bad_xdr: The encoded array in @p is invalid
+ *   %nfserr_inval: An unaligned extent found
+ *   %nfserr_delay: Failed to allocate memory for @iomapp
+ */
+__be32
 nfsd4_scsi_decode_layoutupdate(__be32 *p, u32 len, struct iomap **iomapp,
-		u32 block_size)
+		int *nr_iomapsp, u32 block_size)
 {
 	struct iomap *iomaps;
 	u32 nr_iomaps, expected, i;
 
-	if (len < sizeof(u32)) {
-		dprintk("%s: extent array too small: %u\n", __func__, len);
-		return -EINVAL;
-	}
+	if (len < sizeof(u32))
+		return nfserr_bad_xdr;
 
 	nr_iomaps = be32_to_cpup(p++);
 	expected = sizeof(__be32) + nr_iomaps * PNFS_SCSI_RANGE_SIZE;
-	if (len != expected) {
-		dprintk("%s: extent array size mismatch: %u/%u\n",
-			__func__, len, expected);
-		return -EINVAL;
-	}
+	if (len != expected)
+		return nfserr_bad_xdr;
 
 	iomaps = kcalloc(nr_iomaps, sizeof(*iomaps), GFP_KERNEL);
-	if (!iomaps) {
-		dprintk("%s: failed to allocate extent array\n", __func__);
-		return -ENOMEM;
-	}
+	if (!iomaps)
+		return nfserr_delay;
 
 	for (i = 0; i < nr_iomaps; i++) {
 		u64 val;
 
 		p = xdr_decode_hyper(p, &val);
 		if (val & (block_size - 1)) {
-			dprintk("%s: unaligned offset 0x%llx\n", __func__, val);
 			goto fail;
 		}
 		iomaps[i].offset = val;
 
 		p = xdr_decode_hyper(p, &val);
 		if (val & (block_size - 1)) {
-			dprintk("%s: unaligned length 0x%llx\n", __func__, val);
 			goto fail;
 		}
 		iomaps[i].length = val;
 	}
 
 	*iomapp = iomaps;
-	return nr_iomaps;
+	*nr_iomapsp = nr_iomaps;
+	return nfs_ok;
 fail:
 	kfree(iomaps);
-	return -EINVAL;
+	return nfserr_inval;
 }

fs/nfsd/blocklayoutxdr.h

@@ -54,9 +54,9 @@ __be32 nfsd4_block_encode_getdeviceinfo(struct xdr_stream *xdr,
 		const struct nfsd4_getdeviceinfo *gdp);
 __be32 nfsd4_block_encode_layoutget(struct xdr_stream *xdr,
 		const struct nfsd4_layoutget *lgp);
-int nfsd4_block_decode_layoutupdate(__be32 *p, u32 len, struct iomap **iomapp,
-		u32 block_size);
-int nfsd4_scsi_decode_layoutupdate(__be32 *p, u32 len, struct iomap **iomapp,
-		u32 block_size);
+__be32 nfsd4_block_decode_layoutupdate(__be32 *p, u32 len,
+		struct iomap **iomapp, int *nr_iomapsp, u32 block_size);
+__be32 nfsd4_scsi_decode_layoutupdate(__be32 *p, u32 len,
+		struct iomap **iomapp, int *nr_iomapsp, u32 block_size);
 
 #endif /* _NFSD_BLOCKLAYOUTXDR_H */

fs/nfsd/export.c

@@ -82,8 +82,7 @@ static int expkey_parse(struct cache_detail *cd, char *mesg, int mlen)
 	int len;
 	struct auth_domain *dom = NULL;
 	int err;
-	int fsidtype;
-	char *ep;
+	u8 fsidtype;
 	struct svc_expkey key;
 	struct svc_expkey *ek = NULL;
 
@@ -109,10 +108,9 @@ static int expkey_parse(struct cache_detail *cd, char *mesg, int mlen)
 	err = -EINVAL;
 	if (qword_get(&mesg, buf, PAGE_SIZE) <= 0)
 		goto out;
-	fsidtype = simple_strtoul(buf, &ep, 10);
-	if (*ep)
+	if (kstrtou8(buf, 10, &fsidtype))
 		goto out;
-	dprintk("found fsidtype %d\n", fsidtype);
+	dprintk("found fsidtype %u\n", fsidtype);
 	if (key_len(fsidtype)==0) /* invalid type */
 		goto out;
 	if ((len=qword_get(&mesg, buf, PAGE_SIZE)) <= 0)

fs/nfsd/export.h

@@ -88,7 +88,7 @@ struct svc_expkey {
 	struct cache_head	h;
 
 	struct auth_domain *	ek_client;
-	int			ek_fsidtype;
+	u8			ek_fsidtype;
 	u32			ek_fsid[6];
 
 	struct path		ek_path;

fs/nfsd/filecache.c

@@ -372,7 +372,7 @@ nfsd_file_put(struct nfsd_file *nf)
 
 /**
  * nfsd_file_put_local - put nfsd_file reference and arm nfsd_net_put in caller
- * @nf: nfsd_file of which to put the reference
+ * @pnf: nfsd_file of which to put the reference
  *
  * First save the associated net to return to caller, then put
  * the reference of the nfsd_file.

fs/nfsd/localio.c

@@ -32,7 +32,7 @@
  * @rpc_clnt: rpc_clnt that the client established
  * @cred: cred that the client established
  * @nfs_fh: filehandle to lookup
- * @nfp: place to find the nfsd_file, or store it if it was non-NULL
+ * @pnf: place to find the nfsd_file, or store it if it was non-NULL
  * @fmode: fmode_t to use for open
  *
  * This function maps a local fh to a path on a local filesystem.

fs/nfsd/nfs3proc.c

@@ -561,7 +561,7 @@ static void nfsd3_init_dirlist_pages(struct svc_rqst *rqstp,
 	buf->pages = rqstp->rq_next_page;
 	rqstp->rq_next_page += (buf->buflen + PAGE_SIZE - 1) >> PAGE_SHIFT;
 
-	xdr_init_encode_pages(xdr, buf, buf->pages,  NULL);
+	xdr_init_encode_pages(xdr, buf);
 }
 
 /*

fs/nfsd/nfs4layouts.c

@@ -65,7 +65,7 @@ nfsd4_alloc_devid_map(const struct svc_fh *fhp)
 		return;
 
 	map->fsid_type = fh->fh_fsid_type;
-	memcpy(&map->fsid, fh->fh_fsid, fsid_len);
+	memcpy(&map->fsid, fh_fsid(fh), fsid_len);
 
 	spin_lock(&nfsd_devid_lock);
 	if (fhp->fh_export->ex_devid_map)
@@ -75,7 +75,7 @@ nfsd4_alloc_devid_map(const struct svc_fh *fhp)
 		list_for_each_entry(old, &nfsd_devid_hash[i], hash) {
 			if (old->fsid_type != fh->fh_fsid_type)
 				continue;
-			if (memcmp(old->fsid, fh->fh_fsid,
+			if (memcmp(old->fsid, fh_fsid(fh),
 					key_len(old->fsid_type)))
 				continue;
 

fs/nfsd/nfs4proc.c

@@ -1917,13 +1917,6 @@ nfsd4_copy(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 	struct nfsd42_write_res *result;
 	__be32 status;
 
-	/*
-	 * Currently, async COPY is not reliable. Force all COPY
-	 * requests to be synchronous to avoid client application
-	 * hangs waiting for COPY completion.
-	 */
-	nfsd4_copy_set_sync(copy, true);
-
 	result = &copy->cp_res;
 	nfsd_copy_write_verifier((__be32 *)&result->wr_verifier.data, nn);
 
@@ -2842,20 +2835,10 @@ nfsd4_proc_compound(struct svc_rqst *rqstp)
 
 	rqstp->rq_lease_breaker = (void **)&cstate->clp;
 
-	trace_nfsd_compound(rqstp, args->tag, args->taglen, args->client_opcnt);
+	trace_nfsd_compound(rqstp, args->tag, args->taglen, args->opcnt);
 	while (!status && resp->opcnt < args->opcnt) {
 		op = &args->ops[resp->opcnt++];
 
-		if (unlikely(resp->opcnt == NFSD_MAX_OPS_PER_COMPOUND)) {
-			/* If there are still more operations to process,
-			 * stop here and report NFS4ERR_RESOURCE. */
-			if (cstate->minorversion == 0 &&
-			    args->client_opcnt > resp->opcnt) {
-				op->status = nfserr_resource;
-				goto encode_op;
-			}
-		}
-
 		/*
 		 * The XDR decode routines may have pre-set op->status;
 		 * for example, if there is a miscellaneous XDR error
@@ -2932,7 +2915,7 @@ nfsd4_proc_compound(struct svc_rqst *rqstp)
 			status = op->status;
 		}
 
-		trace_nfsd_compound_status(args->client_opcnt, resp->opcnt,
+		trace_nfsd_compound_status(args->opcnt, resp->opcnt,
 					   status, nfsd4_op_name(op->opnum));
 
 		nfsd4_cstate_clear_replay(cstate);