Merge tag 'nfsd-5.17-2' of git://git.kernel.org/pub/scm/linux/kernel/…

…git/cel/linux

Pull more nfsd fixes from Chuck Lever:
 "Ensure that NFS clients cannot send file size or offset values that
  can cause the NFS server to crash or to return incorrect or surprising
  results.

  In particular, fix how the NFS server handles values larger than
  OFFSET_MAX"

* tag 'nfsd-5.17-2' of git://git.kernel.org/pub/scm/linux/kernel/git/cel/linux:
  NFSD: Deprecate NFS_OFFSET_MAX
  NFSD: Fix offset type in I/O trace points
  NFSD: COMMIT operations must not return NFS?ERR_INVAL
  NFSD: Clamp WRITE offsets
  NFSD: Fix NFSv3 SETATTR/CREATE's handling of large file sizes
  NFSD: Fix ia_size underflow
  NFSD: Fix the behavior of READ near OFFSET_MAX

fs/nfsd/nfs3proc.c

@@ -150,13 +150,17 @@ nfsd3_proc_read(struct svc_rqst *rqstp)
 	unsigned int len;
 	int v;
 
-	argp->count = min_t(u32, argp->count, max_blocksize);
-
 	dprintk("nfsd: READ(3) %s %lu bytes at %Lu\n",
 				SVCFH_fmt(&argp->fh),
 				(unsigned long) argp->count,
 				(unsigned long long) argp->offset);
 
+	argp->count = min_t(u32, argp->count, max_blocksize);
+	if (argp->offset > (u64)OFFSET_MAX)
+		argp->offset = (u64)OFFSET_MAX;
+	if (argp->offset + argp->count > (u64)OFFSET_MAX)
+		argp->count = (u64)OFFSET_MAX - argp->offset;
+
 	v = 0;
 	len = argp->count;
 	resp->pages = rqstp->rq_next_page;
@@ -199,6 +203,11 @@ nfsd3_proc_write(struct svc_rqst *rqstp)
 				(unsigned long long) argp->offset,
 				argp->stable? " stable" : "");
 
+	resp->status = nfserr_fbig;
+	if (argp->offset > (u64)OFFSET_MAX ||
+	    argp->offset + argp->len > (u64)OFFSET_MAX)
+		return rpc_success;
+
 	fh_copy(&resp->fh, &argp->fh);
 	resp->committed = argp->stable;
 	nvecs = svc_fill_write_vector(rqstp, &argp->payload);
@@ -651,15 +660,9 @@ nfsd3_proc_commit(struct svc_rqst *rqstp)
 				argp->count,
 				(unsigned long long) argp->offset);
 
-	if (argp->offset > NFS_OFFSET_MAX) {
-		resp->status = nfserr_inval;
-		goto out;
-	}
-
 	fh_copy(&resp->fh, &argp->fh);
 	resp->status = nfsd_commit(rqstp, &resp->fh, argp->offset,
 				   argp->count, resp->verf);
-out:
 	return rpc_success;
 }
 

fs/nfsd/nfs3xdr.c

@@ -254,7 +254,7 @@ svcxdr_decode_sattr3(struct svc_rqst *rqstp, struct xdr_stream *xdr,
 		if (xdr_stream_decode_u64(xdr, &newsize) < 0)
 			return false;
 		iap->ia_valid |= ATTR_SIZE;
-		iap->ia_size = min_t(u64, newsize, NFS_OFFSET_MAX);
+		iap->ia_size = newsize;
 	}
 	if (xdr_stream_decode_u32(xdr, &set_it) < 0)
 		return false;
@@ -1060,7 +1060,7 @@ svcxdr_encode_entry3_common(struct nfsd3_readdirres *resp, const char *name,
 		return false;
 	/* cookie */
 	resp->cookie_offset = dirlist->len;
-	if (xdr_stream_encode_u64(xdr, NFS_OFFSET_MAX) < 0)
+	if (xdr_stream_encode_u64(xdr, OFFSET_MAX) < 0)
 		return false;
 
 	return true;

fs/nfsd/nfs4proc.c

@@ -782,12 +782,16 @@ nfsd4_read(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 	__be32 status;
 
 	read->rd_nf = NULL;
-	if (read->rd_offset >= OFFSET_MAX)
-		return nfserr_inval;
 
 	trace_nfsd_read_start(rqstp, &cstate->current_fh,
 			      read->rd_offset, read->rd_length);
 
+	read->rd_length = min_t(u32, read->rd_length, svc_max_payload(rqstp));
+	if (read->rd_offset > (u64)OFFSET_MAX)
+		read->rd_offset = (u64)OFFSET_MAX;
+	if (read->rd_offset + read->rd_length > (u64)OFFSET_MAX)
+		read->rd_length = (u64)OFFSET_MAX - read->rd_offset;
+
 	/*
 	 * If we do a zero copy read, then a client will see read data
 	 * that reflects the state of the file *after* performing the
@@ -1018,8 +1022,9 @@ nfsd4_write(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 	unsigned long cnt;
 	int nvecs;
 
-	if (write->wr_offset >= OFFSET_MAX)
-		return nfserr_inval;
+	if (write->wr_offset > (u64)OFFSET_MAX ||
+	    write->wr_offset + write->wr_buflen > (u64)OFFSET_MAX)
+		return nfserr_fbig;
 
 	cnt = write->wr_buflen;
 	trace_nfsd_write_start(rqstp, &cstate->current_fh,

fs/nfsd/nfs4xdr.c

@@ -3495,7 +3495,7 @@ nfsd4_encode_dirent(void *ccdv, const char *name, int namlen,
 	p = xdr_reserve_space(xdr, 3*4 + namlen);
 	if (!p)
 		goto fail;
-	p = xdr_encode_hyper(p, NFS_OFFSET_MAX);    /* offset of next entry */
+	p = xdr_encode_hyper(p, OFFSET_MAX);        /* offset of next entry */
 	p = xdr_encode_array(p, name, namlen);      /* name length & name */
 
 	nfserr = nfsd4_encode_dirent_fattr(xdr, cd, name, namlen);
@@ -3986,10 +3986,8 @@ nfsd4_encode_read(struct nfsd4_compoundres *resp, __be32 nfserr,
 	}
 	xdr_commit_encode(xdr);
 
-	maxcount = svc_max_payload(resp->rqstp);
-	maxcount = min_t(unsigned long, maxcount,
+	maxcount = min_t(unsigned long, read->rd_length,
 			 (xdr->buf->buflen - xdr->buf->len));
-	maxcount = min_t(unsigned long, maxcount, read->rd_length);
 
 	if (file->f_op->splice_read &&
 	    test_bit(RQ_SPLICE_OK, &resp->rqstp->rq_flags))
@@ -4826,10 +4824,8 @@ nfsd4_encode_read_plus(struct nfsd4_compoundres *resp, __be32 nfserr,
 		return nfserr_resource;
 	xdr_commit_encode(xdr);
 
-	maxcount = svc_max_payload(resp->rqstp);
-	maxcount = min_t(unsigned long, maxcount,
+	maxcount = min_t(unsigned long, read->rd_length,
 			 (xdr->buf->buflen - xdr->buf->len));
-	maxcount = min_t(unsigned long, maxcount, read->rd_length);
 	count    = maxcount;
 
 	eof = read->rd_offset >= i_size_read(file_inode(file));

fs/nfsd/trace.h

@@ -306,22 +306,22 @@ TRACE_EVENT(nfsd_export_update,
 DECLARE_EVENT_CLASS(nfsd_io_class,
 	TP_PROTO(struct svc_rqst *rqstp,
 		 struct svc_fh	*fhp,
-		 loff_t		offset,
-		 unsigned long	len),
+		 u64		offset,
+		 u32		len),
 	TP_ARGS(rqstp, fhp, offset, len),
 	TP_STRUCT__entry(
 		__field(u32, xid)
 		__field(u32, fh_hash)
-		__field(loff_t, offset)
-		__field(unsigned long, len)
+		__field(u64, offset)
+		__field(u32, len)
 	),
 	TP_fast_assign(
 		__entry->xid = be32_to_cpu(rqstp->rq_xid);
 		__entry->fh_hash = knfsd_fh_hash(&fhp->fh_handle);
 		__entry->offset = offset;
 		__entry->len = len;
 	),
-	TP_printk("xid=0x%08x fh_hash=0x%08x offset=%lld len=%lu",
+	TP_printk("xid=0x%08x fh_hash=0x%08x offset=%llu len=%u",
 		  __entry->xid, __entry->fh_hash,
 		  __entry->offset, __entry->len)
 )
@@ -330,8 +330,8 @@ DECLARE_EVENT_CLASS(nfsd_io_class,
 DEFINE_EVENT(nfsd_io_class, nfsd_##name,	\
 	TP_PROTO(struct svc_rqst *rqstp,	\
 		 struct svc_fh	*fhp,		\
-		 loff_t		offset,		\
-		 unsigned long	len),		\
+		 u64		offset,		\
+		 u32		len),		\
 	TP_ARGS(rqstp, fhp, offset, len))
 
 DEFINE_NFSD_IO_EVENT(read_start);

fs/nfsd/vfs.c

@@ -435,6 +435,10 @@ nfsd_setattr(struct svc_rqst *rqstp, struct svc_fh *fhp, struct iattr *iap,
 			.ia_size	= iap->ia_size,
 		};
 
+		host_err = -EFBIG;
+		if (iap->ia_size < 0)
+			goto out_unlock;
+
 		host_err = notify_change(&init_user_ns, dentry, &size_attr, NULL);
 		if (host_err)
 			goto out_unlock;
@@ -1110,42 +1114,61 @@ nfsd_write(struct svc_rqst *rqstp, struct svc_fh *fhp, loff_t offset,
 }
 
 #ifdef CONFIG_NFSD_V3
-/*
- * Commit all pending writes to stable storage.
+/**
+ * nfsd_commit - Commit pending writes to stable storage
+ * @rqstp: RPC request being processed
+ * @fhp: NFS filehandle
+ * @offset: raw offset from beginning of file
+ * @count: raw count of bytes to sync
+ * @verf: filled in with the server's current write verifier
  *
- * Note: we only guarantee that data that lies within the range specified
- * by the 'offset' and 'count' parameters will be synced.
+ * Note: we guarantee that data that lies within the range specified
+ * by the 'offset' and 'count' parameters will be synced. The server
+ * is permitted to sync data that lies outside this range at the
+ * same time.
  *
  * Unfortunately we cannot lock the file to make sure we return full WCC
  * data to the client, as locking happens lower down in the filesystem.
+ *
+ * Return values:
+ *   An nfsstat value in network byte order.
  */
 __be32
-nfsd_commit(struct svc_rqst *rqstp, struct svc_fh *fhp,
-               loff_t offset, unsigned long count, __be32 *verf)
+nfsd_commit(struct svc_rqst *rqstp, struct svc_fh *fhp, u64 offset,
+	    u32 count, __be32 *verf)
 {
+	u64			maxbytes;
+	loff_t			start, end;
 	struct nfsd_net		*nn;
 	struct nfsd_file	*nf;
-	loff_t			end = LLONG_MAX;
-	__be32			err = nfserr_inval;
-
-	if (offset < 0)
-		goto out;
-	if (count != 0) {
-		end = offset + (loff_t)count - 1;
-		if (end < offset)
-			goto out;
-	}
+	__be32			err;
 
 	err = nfsd_file_acquire(rqstp, fhp,
 			NFSD_MAY_WRITE|NFSD_MAY_NOT_BREAK_LEASE, &nf);
 	if (err)
 		goto out;
+
+	/*
+	 * Convert the client-provided (offset, count) range to a
+	 * (start, end) range. If the client-provided range falls
+	 * outside the maximum file size of the underlying FS,
+	 * clamp the sync range appropriately.
+	 */
+	start = 0;
+	end = LLONG_MAX;
+	maxbytes = (u64)fhp->fh_dentry->d_sb->s_maxbytes;
+	if (offset < maxbytes) {
+		start = offset;
+		if (count && (offset + count - 1 < maxbytes))
+			end = offset + count - 1;
+	}
+
 	nn = net_generic(nf->nf_net, nfsd_net_id);
 	if (EX_ISSYNC(fhp->fh_export)) {
 		errseq_t since = READ_ONCE(nf->nf_file->f_wb_err);
 		int err2;
 
-		err2 = vfs_fsync_range(nf->nf_file, offset, end, 0);
+		err2 = vfs_fsync_range(nf->nf_file, start, end, 0);
 		switch (err2) {
 		case 0:
 			nfsd_copy_write_verifier(verf, nn);

fs/nfsd/vfs.h

@@ -74,8 +74,8 @@ __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 *, struct svc_fh *,
-				loff_t, unsigned long, __be32 *verf);
+__be32		nfsd_commit(struct svc_rqst *rqst, struct svc_fh *fhp,
+				u64 offset, u32 count, __be32 *verf);
 #endif /* CONFIG_NFSD_V3 */
 #ifdef CONFIG_NFSD_V4
 __be32		nfsd_getxattr(struct svc_rqst *rqstp, struct svc_fh *fhp,

include/linux/nfs.h

@@ -36,14 +36,6 @@ static inline void nfs_copy_fh(struct nfs_fh *target, const struct nfs_fh *sourc
 	memcpy(target->data, source->data, source->size);
 }
 
-
-/*
- * This is really a general kernel constant, but since nothing like
- * this is defined in the kernel headers, I have to do it here.
- */
-#define NFS_OFFSET_MAX		((__s64)((~(__u64)0) >> 1))
-
-
 enum nfs3_stable_how {
 	NFS_UNSTABLE = 0,
 	NFS_DATA_SYNC = 1,