FAILED: patch "[PATCH] SUNRPC: Return an error from xdr_buf_to_bvec() on overflow" failed to apply to 6.18-stable tree

FAILED: patch "[PATCH] SUNRPC: Return an error from xdr_buf_to_bvec() on overflow" failed to apply to 6.18-stable tree

[gregkh]

The patch below does not apply to the 6.18-stable tree.
If someone wants it applied there, or to any other stable or longterm
tree, then please email the backport, including the original git commit
id to <stable@vger.kernel.org>.

To reproduce the conflict and resubmit, you may use the following commands:

git fetch https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/ linux-6.18.y
git checkout FETCH_HEAD
git cherry-pick -x 18c1cc69886192e33536498289d26dba6894e3d5
# <resolve conflicts, build, test, etc.>
git commit -s
git send-email --to '<stable@vger.kernel.org>' --in-reply-to '2026072057-lapped-rebate-93bd@gregkh' --subject-prefix 'PATCH 6.18.y' 'HEAD^..'

Possible dependencies:



thanks,

greg k-h

------------------ original commit in Linus's tree ------------------

From 18c1cc69886192e33536498289d26dba6894e3d5 Mon Sep 17 00:00:00 2001
From: Chuck Lever <cel@kernel.org>
Date: Tue, 19 May 2026 09:34:22 -0400
Subject: [PATCH] SUNRPC: Return an error from xdr_buf_to_bvec() on overflow

xdr_buf_to_bvec() returns a slot count even when the caller's bvec
budget is exhausted partway through the xdr_buf. Callers feed that
count into iov_iter_bvec() and continue as if the conversion had
succeeded, silently sending or writing fewer bytes than the data
length declares. For an NFS WRITE the server reports the truncated
transfer to the client as full success.

The overflow represents an internal invariant violation: a higher
layer reserved a bvec budget too small for the xdr_buf it then
asked the encoder to convert. That is a server-side fault, not a
media I/O failure and not a malformed client argument.

Change xdr_buf_to_bvec() to return a signed int and have the
overflow label return -ESERVERFAULT. Update the three callers to
detect the negative return and fail the request: nfsd_vfs_write()
folds the error into host_err, which nfserrno() translates to
nfserr_serverfault for the WRITE reply; svc_udp_sendto() and
svc_tcp_sendmsg() propagate the error out of the send path.

Reported-by: Chris Mason <clm@meta.com>
Fixes: 2eb2b9358181 ("SUNRPC: Convert svc_tcp_sendmsg to use bio_vecs directly")
Cc: stable@vger.kernel.org
Reviewed-by: Jeff Layton <jlayton@kernel.org>
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>

diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c
index ba97e287c007..cba473969429 100644
--- a/fs/nfsd/vfs.c
+++ b/fs/nfsd/vfs.c
@@ -1440,7 +1440,7 @@ nfsd_vfs_write(struct svc_rqst *rqstp, struct svc_fh *fhp,
 	unsigned long		exp_op_flags = 0;
 	unsigned int		pflags = current->flags;
 	bool			restore_flags = false;
-	unsigned int		nvecs;
+	int			nvecs;
 
 	trace_nfsd_write_opened(rqstp, fhp, offset, *cnt);
 
@@ -1480,6 +1480,10 @@ nfsd_vfs_write(struct svc_rqst *rqstp, struct svc_fh *fhp,
 	}
 
 	nvecs = xdr_buf_to_bvec(rqstp->rq_bvec, rqstp->rq_maxpages, payload);
+	if (nvecs < 0) {
+		host_err = nvecs;
+		goto out_nfserr;
+	}
 
 	since = READ_ONCE(file->f_wb_err);
 	if (verf)
diff --git a/include/linux/sunrpc/xdr.h b/include/linux/sunrpc/xdr.h
index 31971b01d962..b102b4f21e6b 100644
--- a/include/linux/sunrpc/xdr.h
+++ b/include/linux/sunrpc/xdr.h
@@ -138,8 +138,8 @@ void	xdr_terminate_string(const struct xdr_buf *, const u32);
 size_t	xdr_buf_pagecount(const struct xdr_buf *buf);
 int	xdr_alloc_bvec(struct xdr_buf *buf, gfp_t gfp);
 void	xdr_free_bvec(struct xdr_buf *buf);
-unsigned int xdr_buf_to_bvec(struct bio_vec *bvec, unsigned int bvec_size,
-			     const struct xdr_buf *xdr);
+int xdr_buf_to_bvec(struct bio_vec *bvec, unsigned int bvec_size,
+		    const struct xdr_buf *xdr);
 int xdr_buf_to_sg(const struct xdr_buf *buf, unsigned int offset,
 		  unsigned int len, struct scatterlist *sg, unsigned int nsg);
 int xdr_buf_to_sg_alloc(const struct xdr_buf *buf, unsigned int offset,
diff --git a/net/sunrpc/svcsock.c b/net/sunrpc/svcsock.c
index 7be3de1a1aed..c434b6a6637d 100644
--- a/net/sunrpc/svcsock.c
+++ b/net/sunrpc/svcsock.c
@@ -732,7 +732,7 @@ static int svc_udp_sendto(struct svc_rqst *rqstp)
 		.msg_flags	= MSG_SPLICE_PAGES,
 		.msg_controllen	= sizeof(buffer),
 	};
-	unsigned int count;
+	int count;
 	int err;
 
 	svc_udp_release_ctxt(xprt, rqstp->rq_xprt_ctxt);
@@ -746,6 +746,10 @@ static int svc_udp_sendto(struct svc_rqst *rqstp)
 		goto out_notconn;
 
 	count = xdr_buf_to_bvec(svsk->sk_bvec, SUNRPC_MAX_UDP_SENDPAGES, xdr);
+	if (count < 0) {
+		err = count;
+		goto out_trace;
+	}
 
 	iov_iter_bvec(&msg.msg_iter, ITER_SOURCE, svsk->sk_bvec,
 		      count, rqstp->rq_res.len);
@@ -757,6 +761,7 @@ static int svc_udp_sendto(struct svc_rqst *rqstp)
 		err = sock_sendmsg(svsk->sk_sock, &msg);
 	}
 
+out_trace:
 	trace_svcsock_udp_send(xprt, err);
 
 	mutex_unlock(&xprt->xpt_mutex);
@@ -1237,7 +1242,7 @@ static int svc_tcp_sendmsg(struct svc_sock *svsk, struct svc_rqst *rqstp,
 	struct msghdr msg = {
 		.msg_flags	= MSG_SPLICE_PAGES,
 	};
-	unsigned int count;
+	int count;
 	void *buf;
 	int ret;
 
@@ -1253,10 +1258,15 @@ static int svc_tcp_sendmsg(struct svc_sock *svsk, struct svc_rqst *rqstp,
 
 	count = xdr_buf_to_bvec(svsk->sk_bvec + 1, rqstp->rq_maxpages,
 				&rqstp->rq_res);
+	if (count < 0) {
+		ret = count;
+		goto out;
+	}
 
 	iov_iter_bvec(&msg.msg_iter, ITER_SOURCE, svsk->sk_bvec,
 		      1 + count, sizeof(marker) + rqstp->rq_res.len);
 	ret = sock_sendmsg(svsk->sk_sock, &msg);
+out:
 	page_frag_free(buf);
 	return ret;
 }
diff --git a/net/sunrpc/xdr.c b/net/sunrpc/xdr.c
index 8f52782d8a37..fa6a30b5f046 100644
--- a/net/sunrpc/xdr.c
+++ b/net/sunrpc/xdr.c
@@ -139,13 +139,14 @@ xdr_free_bvec(struct xdr_buf *buf)
 /**
  * xdr_buf_to_bvec - Copy components of an xdr_buf into a bio_vec array
  * @bvec: bio_vec array to populate
- * @bvec_size: element count of @bio_vec
+ * @bvec_size: element count of @bvec
  * @xdr: xdr_buf to be copied
  *
- * Returns the number of entries consumed in @bvec.
+ * Returns the number of entries consumed in @bvec on success, or
+ * -ESERVERFAULT when @xdr does not fit within @bvec_size entries.
  */
-unsigned int xdr_buf_to_bvec(struct bio_vec *bvec, unsigned int bvec_size,
-			     const struct xdr_buf *xdr)
+int xdr_buf_to_bvec(struct bio_vec *bvec, unsigned int bvec_size,
+		    const struct xdr_buf *xdr)
 {
 	const struct kvec *head = xdr->head;
 	const struct kvec *tail = xdr->tail;
@@ -187,7 +188,7 @@ unsigned int xdr_buf_to_bvec(struct bio_vec *bvec, unsigned int bvec_size,
 
 bvec_overflow:
 	pr_warn_once("%s: bio_vec array overflow\n", __func__);
-	return count;
+	return -ESERVERFAULT;
 }
 EXPORT_SYMBOL_GPL(xdr_buf_to_bvec);
 

[PATCH 6.18.y 1/5] NFSD: pass nfsd_file to nfsd_iter_read()

[Sasha Levin]

From: Mike Snitzer <snitzer@kernel.org>

[ Upstream commit 803bc849f0039291f546ba0e2237faebeb5c073e ]

Prepare for nfsd_iter_read() to use the DIO alignment stored in
nfsd_file by passing the nfsd_file to nfsd_iter_read() rather than
just the file which is associaed with the nfsd_file.

This means nfsd4_encode_readv() now also needs the nfsd_file rather
than the file.  Instead of changing the file arg to be the nfsd_file,
we discard the file arg as the nfsd_file (and indeed the file) is
already available via the "read" argument.

Signed-off-by: Mike Snitzer <snitzer@kernel.org>
Reviewed-by: Jeff Layton <jlayton@kernel.org>
Reviewed-by: NeilBrown <neil@brown.name>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Stable-dep-of: 18c1cc698861 ("SUNRPC: Return an error from xdr_buf_to_bvec() on overflow")
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 fs/nfsd/nfs4xdr.c | 8 ++++----
 fs/nfsd/vfs.c     | 7 ++++---
 fs/nfsd/vfs.h     | 2 +-
 3 files changed, 9 insertions(+), 8 deletions(-)

diff --git a/fs/nfsd/nfs4xdr.c b/fs/nfsd/nfs4xdr.c
index a96029c967e6bd..1e158881c1af9e 100644
--- a/fs/nfsd/nfs4xdr.c
+++ b/fs/nfsd/nfs4xdr.c
@@ -4478,7 +4478,7 @@ static __be32 nfsd4_encode_splice_read(
 
 static __be32 nfsd4_encode_readv(struct nfsd4_compoundres *resp,
 				 struct nfsd4_read *read,
-				 struct file *file, unsigned long maxcount)
+				 unsigned long maxcount)
 {
 	struct xdr_stream *xdr = resp->xdr;
 	unsigned int base = xdr->buf->page_len & ~PAGE_MASK;
@@ -4489,7 +4489,7 @@ static __be32 nfsd4_encode_readv(struct nfsd4_compoundres *resp,
 	if (xdr_reserve_space_vec(xdr, maxcount) < 0)
 		return nfserr_resource;
 
-	nfserr = nfsd_iter_read(resp->rqstp, read->rd_fhp, file,
+	nfserr = nfsd_iter_read(resp->rqstp, read->rd_fhp, read->rd_nf,
 				read->rd_offset, &maxcount, base,
 				&read->rd_eof);
 	read->rd_length = maxcount;
@@ -4536,7 +4536,7 @@ nfsd4_encode_read(struct nfsd4_compoundres *resp, __be32 nfserr,
 	if (file->f_op->splice_read && splice_ok)
 		nfserr = nfsd4_encode_splice_read(resp, read, file, maxcount);
 	else
-		nfserr = nfsd4_encode_readv(resp, read, file, maxcount);
+		nfserr = nfsd4_encode_readv(resp, read, maxcount);
 	if (nfserr) {
 		xdr_truncate_encode(xdr, eof_offset);
 		return nfserr;
@@ -5432,7 +5432,7 @@ nfsd4_encode_read_plus_data(struct nfsd4_compoundres *resp,
 	if (file->f_op->splice_read && splice_ok)
 		nfserr = nfsd4_encode_splice_read(resp, read, file, maxcount);
 	else
-		nfserr = nfsd4_encode_readv(resp, read, file, maxcount);
+		nfserr = nfsd4_encode_readv(resp, read, maxcount);
 	if (nfserr)
 		return nfserr;
 
diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c
index 20330372387801..4874446c2468d8 100644
--- a/fs/nfsd/vfs.c
+++ b/fs/nfsd/vfs.c
@@ -1078,7 +1078,7 @@ __be32 nfsd_splice_read(struct svc_rqst *rqstp, struct svc_fh *fhp,
  * nfsd_iter_read - Perform a VFS read using an iterator
  * @rqstp: RPC transaction context
  * @fhp: file handle of file to be read
- * @file: opened struct file of file to be read
+ * @nf: opened struct nfsd_file of file to be read
  * @offset: starting byte offset
  * @count: IN: requested number of bytes; OUT: number of bytes read
  * @base: offset in first page of read buffer
@@ -1091,9 +1091,10 @@ __be32 nfsd_splice_read(struct svc_rqst *rqstp, struct svc_fh *fhp,
  * returned.
  */
 __be32 nfsd_iter_read(struct svc_rqst *rqstp, struct svc_fh *fhp,
-		      struct file *file, loff_t offset, unsigned long *count,
+		      struct nfsd_file *nf, loff_t offset, unsigned long *count,
 		      unsigned int base, u32 *eof)
 {
+	struct file *file = nf->nf_file;
 	unsigned long v, total;
 	struct iov_iter iter;
 	struct kiocb kiocb;
@@ -1346,7 +1347,7 @@ __be32 nfsd_read(struct svc_rqst *rqstp, struct svc_fh *fhp,
 	if (file->f_op->splice_read && nfsd_read_splice_ok(rqstp))
 		err = nfsd_splice_read(rqstp, fhp, file, offset, count, eof);
 	else
-		err = nfsd_iter_read(rqstp, fhp, file, offset, count, 0, eof);
+		err = nfsd_iter_read(rqstp, fhp, nf, offset, count, 0, eof);
 
 	nfsd_file_put(nf);
 	trace_nfsd_read_done(rqstp, fhp, offset, *count);
diff --git a/fs/nfsd/vfs.h b/fs/nfsd/vfs.h
index 9652c1de53ec13..1dd3ae3ceb3a8e 100644
--- a/fs/nfsd/vfs.h
+++ b/fs/nfsd/vfs.h
@@ -122,7 +122,7 @@ __be32		nfsd_splice_read(struct svc_rqst *rqstp, struct svc_fh *fhp,
 				unsigned long *count,
 				u32 *eof);
 __be32		nfsd_iter_read(struct svc_rqst *rqstp, struct svc_fh *fhp,
-				struct file *file, loff_t offset,
+				struct nfsd_file *nf, loff_t offset,
 				unsigned long *count, unsigned int base,
 				u32 *eof);
 bool		nfsd_read_splice_ok(struct svc_rqst *rqstp);
-- 
2.53.0

[PATCH 6.18.y 2/5] sunrpc: allocate a separate bvec array for socket sends

[Sasha Levin]

From: Jeff Layton <jlayton@kernel.org>

[ Upstream commit 6b3b697d65d46a0f640216a3f6c72856c159c567 ]

svc_tcp_sendmsg() calls xdr_buf_to_bvec() with the second slot of
rq_bvec as the start, but doesn't reduce the array length by one, which
could lead to an array overrun. Also, rq_bvec is always rq_maxpages in
length, which can be too short in some cases, since the TCP record
marker consumes a slot.

Fix both problems by adding a separate bvec array to the svc_sock that
is specifically for sending. For TCP, make this array one slot longer
than rq_maxpages, to account for the record marker. For UDP, only
allocate as large an array as we need since it's limited to 64k of
payload.

Signed-off-by: Jeff Layton <jlayton@kernel.org>
Reviewed-by: NeilBrown <neil@brown.name>
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Stable-dep-of: 18c1cc698861 ("SUNRPC: Return an error from xdr_buf_to_bvec() on overflow")
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 include/linux/sunrpc/svcsock.h |  3 ++
 net/sunrpc/svcsock.c           | 55 +++++++++++++++++++++++++++++-----
 2 files changed, 51 insertions(+), 7 deletions(-)

diff --git a/include/linux/sunrpc/svcsock.h b/include/linux/sunrpc/svcsock.h
index 963bbe251e5210..de37069aba9089 100644
--- a/include/linux/sunrpc/svcsock.h
+++ b/include/linux/sunrpc/svcsock.h
@@ -26,6 +26,9 @@ struct svc_sock {
 	void			(*sk_odata)(struct sock *);
 	void			(*sk_owspace)(struct sock *);
 
+	/* For sends (protected by xpt_mutex) */
+	struct bio_vec		*sk_bvec;
+
 	/* private TCP part */
 	/* On-the-wire fragment header: */
 	__be32			sk_marker;
diff --git a/net/sunrpc/svcsock.c b/net/sunrpc/svcsock.c
index 7b90abc5cf0ee1..48700b99909fac 100644
--- a/net/sunrpc/svcsock.c
+++ b/net/sunrpc/svcsock.c
@@ -68,6 +68,17 @@
 
 #define RPCDBG_FACILITY	RPCDBG_SVCXPRT
 
+/*
+ * For UDP:
+ * 1 for header page
+ * enough pages for RPCSVC_MAXPAYLOAD_UDP
+ * 1 in case payload is not aligned
+ * 1 for tail page
+ */
+enum {
+	SUNRPC_MAX_UDP_SENDPAGES = 1 + RPCSVC_MAXPAYLOAD_UDP / PAGE_SIZE + 1 + 1
+};
+
 /* To-do: to avoid tying up an nfsd thread while waiting for a
  * handshake request, the request could instead be deferred.
  */
@@ -740,14 +751,14 @@ static int svc_udp_sendto(struct svc_rqst *rqstp)
 	if (svc_xprt_is_dead(xprt))
 		goto out_notconn;
 
-	count = xdr_buf_to_bvec(rqstp->rq_bvec, rqstp->rq_maxpages, xdr);
+	count = xdr_buf_to_bvec(svsk->sk_bvec, SUNRPC_MAX_UDP_SENDPAGES, xdr);
 
-	iov_iter_bvec(&msg.msg_iter, ITER_SOURCE, rqstp->rq_bvec,
+	iov_iter_bvec(&msg.msg_iter, ITER_SOURCE, svsk->sk_bvec,
 		      count, rqstp->rq_res.len);
 	err = sock_sendmsg(svsk->sk_sock, &msg);
 	if (err == -ECONNREFUSED) {
 		/* ICMP error on earlier request. */
-		iov_iter_bvec(&msg.msg_iter, ITER_SOURCE, rqstp->rq_bvec,
+		iov_iter_bvec(&msg.msg_iter, ITER_SOURCE, svsk->sk_bvec,
 			      count, rqstp->rq_res.len);
 		err = sock_sendmsg(svsk->sk_sock, &msg);
 	}
@@ -1235,19 +1246,19 @@ static int svc_tcp_sendmsg(struct svc_sock *svsk, struct svc_rqst *rqstp,
 	int ret;
 
 	/* The stream record marker is copied into a temporary page
-	 * fragment buffer so that it can be included in rq_bvec.
+	 * fragment buffer so that it can be included in sk_bvec.
 	 */
 	buf = page_frag_alloc(&svsk->sk_frag_cache, sizeof(marker),
 			      GFP_KERNEL);
 	if (!buf)
 		return -ENOMEM;
 	memcpy(buf, &marker, sizeof(marker));
-	bvec_set_virt(rqstp->rq_bvec, buf, sizeof(marker));
+	bvec_set_virt(svsk->sk_bvec, buf, sizeof(marker));
 
-	count = xdr_buf_to_bvec(rqstp->rq_bvec + 1, rqstp->rq_maxpages,
+	count = xdr_buf_to_bvec(svsk->sk_bvec + 1, rqstp->rq_maxpages,
 				&rqstp->rq_res);
 
-	iov_iter_bvec(&msg.msg_iter, ITER_SOURCE, rqstp->rq_bvec,
+	iov_iter_bvec(&msg.msg_iter, ITER_SOURCE, svsk->sk_bvec,
 		      1 + count, sizeof(marker) + rqstp->rq_res.len);
 	ret = sock_sendmsg(svsk->sk_sock, &msg);
 	page_frag_free(buf);
@@ -1392,6 +1403,20 @@ void svc_sock_update_bufs(struct svc_serv *serv)
 	spin_unlock_bh(&serv->sv_lock);
 }
 
+static int svc_sock_sendpages(struct svc_serv *serv, struct socket *sock, int flags)
+{
+	switch (sock->type) {
+	case SOCK_STREAM:
+		/* +1 for TCP record marker */
+		if (flags & SVC_SOCK_TEMPORARY)
+			return svc_serv_maxpages(serv) + 1;
+		return 0;
+	case SOCK_DGRAM:
+		return SUNRPC_MAX_UDP_SENDPAGES;
+	}
+	return -EINVAL;
+}
+
 /*
  * Initialize socket for RPC use and create svc_sock struct
  */
@@ -1402,12 +1427,26 @@ static struct svc_sock *svc_setup_socket(struct svc_serv *serv,
 	struct svc_sock	*svsk;
 	struct sock	*inet;
 	int		pmap_register = !(flags & SVC_SOCK_ANONYMOUS);
+	int		sendpages;
 	unsigned long	pages;
 
+	sendpages = svc_sock_sendpages(serv, sock, flags);
+	if (sendpages < 0)
+		return ERR_PTR(sendpages);
+
 	pages = svc_serv_maxpages(serv);
 	svsk = kzalloc(struct_size(svsk, sk_pages, pages), GFP_KERNEL);
 	if (!svsk)
 		return ERR_PTR(-ENOMEM);
+
+	if (sendpages) {
+		svsk->sk_bvec = kcalloc(sendpages, sizeof(*svsk->sk_bvec), GFP_KERNEL);
+		if (!svsk->sk_bvec) {
+			kfree(svsk);
+			return ERR_PTR(-ENOMEM);
+		}
+	}
+
 	svsk->sk_maxpages = pages;
 
 	inet = sock->sk;
@@ -1419,6 +1458,7 @@ static struct svc_sock *svc_setup_socket(struct svc_serv *serv,
 				     inet->sk_protocol,
 				     ntohs(inet_sk(inet)->inet_sport));
 		if (err < 0) {
+			kfree(svsk->sk_bvec);
 			kfree(svsk);
 			return ERR_PTR(err);
 		}
@@ -1636,5 +1676,6 @@ static void svc_sock_free(struct svc_xprt *xprt)
 		sock_release(sock);
 
 	page_frag_cache_drain(&svsk->sk_frag_cache);
+	kfree(svsk->sk_bvec);
 	kfree(svsk);
 }
-- 
2.53.0

[PATCH 6.18.y 3/5] SUNRPC: Add helpers to convert xdr_buf byte ranges to scatterlists

[Sasha Levin]

From: Chuck Lever <chuck.lever@oracle.com>

[ Upstream commit e9be933959b581effd426f93b86654f5fbf0c574 ]

The crypto/krb5 library accepts data in scatterlist form, but
the GSS-API layer presents RPC payloads as struct xdr_buf.
Bridge that gap with a pair of helper functions:

  xdr_buf_to_sg()        - populate a caller-supplied scatterlist
                           array from a byte range
  xdr_buf_to_sg_alloc()  - populate a caller-supplied inline
                           scatterlist, chaining to a heap-
                           allocated overflow for large payloads

The inline array (typically stack-allocated at eight entries)
covers the common case of small RPCs with no heap allocation
on the encrypt/decrypt path. Only buffers spanning many pages
incur a kmalloc for the chained extension.

The segment-walking logic follows the same head, page array,
tail traversal as xdr_process_buf(), but populates a
scatterlist directly rather than invoking a per-segment
callback. sg_next() traversal makes the walker safe for
chained scatterlists. Once subsequent patches reroute all
per-message crypto operations through crypto/krb5,
xdr_process_buf() loses its last callers and is removed.

Assisted-by: Claude:claude-opus-4-6
Reviewed-by: Jeff Layton <jlayton@kernel.org>
Acked-by: Anna Schumaker <anna.schumaker@hammerspace.com>
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Stable-dep-of: 18c1cc698861 ("SUNRPC: Return an error from xdr_buf_to_bvec() on overflow")
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 include/linux/sunrpc/xdr.h |  15 +++
 net/sunrpc/xdr.c           | 199 +++++++++++++++++++++++++++++++++++++
 2 files changed, 214 insertions(+)

diff --git a/include/linux/sunrpc/xdr.h b/include/linux/sunrpc/xdr.h
index 152597750f55b3..4b4e767e2ab5bd 100644
--- a/include/linux/sunrpc/xdr.h
+++ b/include/linux/sunrpc/xdr.h
@@ -140,6 +140,21 @@ int	xdr_alloc_bvec(struct xdr_buf *buf, gfp_t gfp);
 void	xdr_free_bvec(struct xdr_buf *buf);
 unsigned int xdr_buf_to_bvec(struct bio_vec *bvec, unsigned int bvec_size,
 			     const struct xdr_buf *xdr);
+int xdr_buf_to_sg(const struct xdr_buf *buf, unsigned int offset,
+		  unsigned int len, struct scatterlist *sg, unsigned int nsg);
+int xdr_buf_to_sg_alloc(const struct xdr_buf *buf, unsigned int offset,
+			unsigned int len, struct scatterlist *sg_head,
+			unsigned int sg_head_nents,
+			struct scatterlist **sg_overflow, gfp_t gfp);
+
+/*
+ * Inline scatterlist entries for xdr_buf_to_sg_alloc().  Sized to cover the
+ * head kvec, tail kvec, and a few page fragments without any heap allocation.
+ */
+enum {
+	XDR_BUF_TO_SG_NENTS	= 8,
+};
+
 
 static inline __be32 *xdr_encode_array(__be32 *p, const void *s, unsigned int len)
 {
diff --git a/net/sunrpc/xdr.c b/net/sunrpc/xdr.c
index 70efc727a9cd82..f372be09dcce41 100644
--- a/net/sunrpc/xdr.c
+++ b/net/sunrpc/xdr.c
@@ -187,6 +187,205 @@ unsigned int xdr_buf_to_bvec(struct bio_vec *bvec, unsigned int bvec_size,
 }
 EXPORT_SYMBOL_GPL(xdr_buf_to_bvec);
 
+/**
+ * xdr_buf_to_sg - Populate a scatterlist from an xdr_buf range
+ * @buf: xdr_buf to map
+ * @offset: starting byte offset within @buf
+ * @len: number of bytes to cover
+ * @sg: scatterlist array initialized with sg_init_table()
+ * @nsg: number of entries available in @sg
+ *
+ * @sg is traversed with sg_next(), so callers may pass a list
+ * assembled with sg_chain().
+ *
+ * Return: on success, the number of scatterlist entries used; the
+ * last used entry is marked with sg_mark_end().  On failure, a
+ * negative errno.
+ */
+int xdr_buf_to_sg(const struct xdr_buf *buf, unsigned int offset,
+		  unsigned int len, struct scatterlist *sg, unsigned int nsg)
+{
+	unsigned int page_len, thislen, page_offset;
+	struct scatterlist *cur = sg, *prev = NULL;
+	int nents = 0;
+	int i;
+
+	if (len == 0)
+		return 0;
+
+	if (offset >= buf->head[0].iov_len) {
+		offset -= buf->head[0].iov_len;
+	} else {
+		thislen = min_t(unsigned int,
+				buf->head[0].iov_len - offset, len);
+		if (nents >= nsg)
+			return -ENOSPC;
+		sg_set_buf(cur, buf->head[0].iov_base + offset,
+			   thislen);
+		prev = cur;
+		cur = sg_next(cur);
+		nents++;
+		len -= thislen;
+		offset = 0;
+	}
+	if (len == 0)
+		goto done;
+
+	if (offset >= buf->page_len) {
+		offset -= buf->page_len;
+	} else {
+		page_len = min(buf->page_len - offset, len);
+		len -= page_len;
+		page_offset = (offset + buf->page_base) & (PAGE_SIZE - 1);
+		i = (offset + buf->page_base) >> PAGE_SHIFT;
+		thislen = PAGE_SIZE - page_offset;
+		do {
+			if (thislen > page_len)
+				thislen = page_len;
+			if (nents >= nsg)
+				return -ENOSPC;
+			sg_set_page(cur, buf->pages[i],
+				    thislen, page_offset);
+			prev = cur;
+			cur = sg_next(cur);
+			nents++;
+			page_len -= thislen;
+			i++;
+			page_offset = 0;
+			thislen = PAGE_SIZE;
+		} while (page_len != 0);
+		offset = 0;
+	}
+	if (len == 0)
+		goto done;
+
+	if (offset < buf->tail[0].iov_len) {
+		thislen = min_t(unsigned int,
+				buf->tail[0].iov_len - offset, len);
+		if (nents >= nsg)
+			return -ENOSPC;
+		sg_set_buf(cur, buf->tail[0].iov_base + offset,
+			   thislen);
+		prev = cur;
+		nents++;
+		len -= thislen;
+	}
+	if (len != 0)
+		return -EINVAL;
+
+done:
+	if (prev)
+		sg_mark_end(prev);
+	return nents;
+}
+EXPORT_SYMBOL_GPL(xdr_buf_to_sg);
+
+/*
+ * Count the scatterlist entries needed to cover [offset, offset + len)
+ * within @buf.  Mirrors the walk in xdr_buf_to_sg() so the caller can
+ * size an allocation that matches the requested sub-range rather than
+ * the full xdr_buf.
+ */
+static unsigned int xdr_buf_sg_nents(const struct xdr_buf *buf,
+				     unsigned int offset, unsigned int len)
+{
+	unsigned int nsg = 0, thislen, page_offset;
+
+	if (len == 0)
+		return 0;
+
+	if (offset < buf->head[0].iov_len) {
+		thislen = min_t(unsigned int,
+				buf->head[0].iov_len - offset, len);
+		nsg++;
+		len -= thislen;
+		offset = 0;
+	} else {
+		offset -= buf->head[0].iov_len;
+	}
+	if (len == 0)
+		return nsg;
+
+	if (offset < buf->page_len) {
+		thislen = min(buf->page_len - offset, len);
+		page_offset = (offset + buf->page_base) & (PAGE_SIZE - 1);
+		nsg += DIV_ROUND_UP(page_offset + thislen, PAGE_SIZE);
+		len -= thislen;
+		offset = 0;
+	} else {
+		offset -= buf->page_len;
+	}
+	if (len == 0)
+		return nsg;
+
+	if (offset < buf->tail[0].iov_len)
+		nsg++;
+	return nsg;
+}
+
+/**
+ * xdr_buf_to_sg_alloc - Populate a scatterlist for an xdr_buf range
+ * @buf: xdr_buf to map
+ * @offset: starting byte offset within @buf
+ * @len: number of bytes to cover
+ * @sg_head: caller-provided scatterlist array (typically stack-allocated)
+ * @sg_head_nents: number of entries in @sg_head
+ * @sg_overflow: OUT: chained extension, or NULL when @sg_head sufficed
+ * @gfp: memory allocation flags for overflow
+ *
+ * Populates @sg_head directly when the xdr_buf fits.  When more
+ * entries are needed, an overflow scatterlist is allocated and
+ * chained from @sg_head so that the result is traversable with
+ * sg_next().
+ *
+ * Return: on success, the number of populated scatterlist entries
+ * (counting only data entries, not chain entries).  @sg_head is
+ * the head of the resulting list.  Caller must kfree @sg_overflow
+ * when done.  On failure, a negative errno.
+ */
+int xdr_buf_to_sg_alloc(const struct xdr_buf *buf, unsigned int offset,
+			unsigned int len, struct scatterlist *sg_head,
+			unsigned int sg_head_nents,
+			struct scatterlist **sg_overflow, gfp_t gfp)
+{
+	unsigned int nsg;
+	int ret;
+
+	*sg_overflow = NULL;
+	if (len == 0)
+		return 0;
+
+	nsg = xdr_buf_sg_nents(buf, offset, len);
+	if (nsg == 0)
+		return -EINVAL;
+
+	if (nsg <= sg_head_nents) {
+		sg_init_table(sg_head, nsg);
+	} else {
+		/* +1 replaces the slot sg_chain() consumes as the link. */
+		unsigned int overflow_nents = nsg - sg_head_nents + 1;
+		struct scatterlist *overflow;
+
+		overflow = kmalloc_array(overflow_nents, sizeof(*overflow),
+					 gfp);
+		if (!overflow)
+			return -ENOMEM;
+
+		sg_init_table(sg_head, sg_head_nents);
+		sg_init_table(overflow, overflow_nents);
+		sg_chain(sg_head, sg_head_nents, overflow);
+		*sg_overflow = overflow;
+	}
+
+	ret = xdr_buf_to_sg(buf, offset, len, sg_head, nsg);
+	if (ret < 0) {
+		kfree(*sg_overflow);
+		*sg_overflow = NULL;
+	}
+	return ret;
+}
+EXPORT_SYMBOL_GPL(xdr_buf_to_sg_alloc);
+
 /**
  * xdr_inline_pages - Prepare receive buffer for a large reply
  * @xdr: xdr_buf into which reply will be placed
-- 
2.53.0

[PATCH 6.18.y 4/5] SUNRPC: Bound-check xdr_buf_to_bvec() stores before writing

[Sasha Levin]

From: Chuck Lever <chuck.lever@oracle.com>

[ Upstream commit 42f5b80dda6b86e424054baf1475df686c403d5c ]

xdr_buf_to_bvec() writes a bio_vec into the caller's array before
testing whether that slot is in range, and the head branch performs
the store with no check at all. When the caller's budget is exactly
used up, the next store lands one element past the end of the array.
The overflow label returns count - 1, which masks the surplus store
but cannot undo it.

rq_bvec, the array passed by nfsd_vfs_write(), is allocated to
exactly rq_maxpages entries with no slack. The OOB store can land in
adjacent slab memory; the bv_len and bv_offset fields written there
are derived from client-supplied RPC payload sizes.

Move the in-range check ahead of the store in the head, page-loop,
and tail branches. With the check at the top of each sequence, count
is incremented only after a successful store, so the overflow label
can return count directly.

Reported-by: Chris Mason <clm@meta.com>
Fixes: 2eb2b9358181 ("SUNRPC: Convert svc_tcp_sendmsg to use bio_vecs directly")
Cc: stable@vger.kernel.org
Reviewed-by: Jeff Layton <jlayton@kernel.org>
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Stable-dep-of: 18c1cc698861 ("SUNRPC: Return an error from xdr_buf_to_bvec() on overflow")
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 net/sunrpc/xdr.c | 14 +++++++++-----
 1 file changed, 9 insertions(+), 5 deletions(-)

diff --git a/net/sunrpc/xdr.c b/net/sunrpc/xdr.c
index f372be09dcce41..82d3e8f65e486a 100644
--- a/net/sunrpc/xdr.c
+++ b/net/sunrpc/xdr.c
@@ -152,6 +152,8 @@ unsigned int xdr_buf_to_bvec(struct bio_vec *bvec, unsigned int bvec_size,
 	unsigned int count = 0;
 
 	if (head->iov_len) {
+		if (unlikely(count >= bvec_size))
+			goto bvec_overflow;
 		bvec_set_virt(bvec++, head->iov_base, head->iov_len);
 		++count;
 	}
@@ -165,25 +167,27 @@ unsigned int xdr_buf_to_bvec(struct bio_vec *bvec, unsigned int bvec_size,
 		while (remaining > 0) {
 			len = min_t(unsigned int, remaining,
 				    PAGE_SIZE - offset);
+			if (unlikely(count >= bvec_size))
+				goto bvec_overflow;
 			bvec_set_page(bvec++, *pages++, len, offset);
 			remaining -= len;
 			offset = 0;
-			if (unlikely(++count > bvec_size))
-				goto bvec_overflow;
+			++count;
 		}
 	}
 
 	if (tail->iov_len) {
-		bvec_set_virt(bvec, tail->iov_base, tail->iov_len);
-		if (unlikely(++count > bvec_size))
+		if (unlikely(count >= bvec_size))
 			goto bvec_overflow;
+		bvec_set_virt(bvec, tail->iov_base, tail->iov_len);
+		++count;
 	}
 
 	return count;
 
 bvec_overflow:
 	pr_warn_once("%s: bio_vec array overflow\n", __func__);
-	return count - 1;
+	return count;
 }
 EXPORT_SYMBOL_GPL(xdr_buf_to_bvec);
 
-- 
2.53.0

[PATCH 6.18.y 5/5] SUNRPC: Return an error from xdr_buf_to_bvec() on overflow

[Sasha Levin]

From: Chuck Lever <chuck.lever@oracle.com>

[ Upstream commit 18c1cc69886192e33536498289d26dba6894e3d5 ]

xdr_buf_to_bvec() returns a slot count even when the caller's bvec
budget is exhausted partway through the xdr_buf. Callers feed that
count into iov_iter_bvec() and continue as if the conversion had
succeeded, silently sending or writing fewer bytes than the data
length declares. For an NFS WRITE the server reports the truncated
transfer to the client as full success.

The overflow represents an internal invariant violation: a higher
layer reserved a bvec budget too small for the xdr_buf it then
asked the encoder to convert. That is a server-side fault, not a
media I/O failure and not a malformed client argument.

Change xdr_buf_to_bvec() to return a signed int and have the
overflow label return -ESERVERFAULT. Update the three callers to
detect the negative return and fail the request: nfsd_vfs_write()
folds the error into host_err, which nfserrno() translates to
nfserr_serverfault for the WRITE reply; svc_udp_sendto() and
svc_tcp_sendmsg() propagate the error out of the send path.

Reported-by: Chris Mason <clm@meta.com>
Fixes: 2eb2b9358181 ("SUNRPC: Convert svc_tcp_sendmsg to use bio_vecs directly")
Cc: stable@vger.kernel.org
Reviewed-by: Jeff Layton <jlayton@kernel.org>
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 fs/nfsd/vfs.c              |  8 +++++++-
 include/linux/sunrpc/xdr.h |  4 ++--
 net/sunrpc/svcsock.c       | 14 ++++++++++++--
 net/sunrpc/xdr.c           | 11 ++++++-----
 4 files changed, 27 insertions(+), 10 deletions(-)

diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c
index 4874446c2468d8..ba882a4ea9cb9d 100644
--- a/fs/nfsd/vfs.c
+++ b/fs/nfsd/vfs.c
@@ -1203,7 +1203,7 @@ nfsd_vfs_write(struct svc_rqst *rqstp, struct svc_fh *fhp,
 	unsigned long		exp_op_flags = 0;
 	unsigned int		pflags = current->flags;
 	bool			restore_flags = false;
-	unsigned int		nvecs;
+	int			nvecs;
 
 	trace_nfsd_write_opened(rqstp, fhp, offset, *cnt);
 
@@ -1243,7 +1243,13 @@ nfsd_vfs_write(struct svc_rqst *rqstp, struct svc_fh *fhp,
 	}
 
 	nvecs = xdr_buf_to_bvec(rqstp->rq_bvec, rqstp->rq_maxpages, payload);
+	if (nvecs < 0) {
+		host_err = nvecs;
+		goto out_nfserr;
+	}
+
 	iov_iter_bvec(&iter, ITER_SOURCE, rqstp->rq_bvec, nvecs, *cnt);
+
 	since = READ_ONCE(file->f_wb_err);
 	if (verf)
 		nfsd_copy_write_verifier(verf, nn);
diff --git a/include/linux/sunrpc/xdr.h b/include/linux/sunrpc/xdr.h
index 4b4e767e2ab5bd..656e0ad323af1b 100644
--- a/include/linux/sunrpc/xdr.h
+++ b/include/linux/sunrpc/xdr.h
@@ -138,8 +138,8 @@ void	xdr_terminate_string(const struct xdr_buf *, const u32);
 size_t	xdr_buf_pagecount(const struct xdr_buf *buf);
 int	xdr_alloc_bvec(struct xdr_buf *buf, gfp_t gfp);
 void	xdr_free_bvec(struct xdr_buf *buf);
-unsigned int xdr_buf_to_bvec(struct bio_vec *bvec, unsigned int bvec_size,
-			     const struct xdr_buf *xdr);
+int xdr_buf_to_bvec(struct bio_vec *bvec, unsigned int bvec_size,
+		    const struct xdr_buf *xdr);
 int xdr_buf_to_sg(const struct xdr_buf *buf, unsigned int offset,
 		  unsigned int len, struct scatterlist *sg, unsigned int nsg);
 int xdr_buf_to_sg_alloc(const struct xdr_buf *buf, unsigned int offset,
diff --git a/net/sunrpc/svcsock.c b/net/sunrpc/svcsock.c
index 48700b99909fac..038294fa3942a5 100644
--- a/net/sunrpc/svcsock.c
+++ b/net/sunrpc/svcsock.c
@@ -738,7 +738,7 @@ static int svc_udp_sendto(struct svc_rqst *rqstp)
 		.msg_flags	= MSG_SPLICE_PAGES,
 		.msg_controllen	= sizeof(buffer),
 	};
-	unsigned int count;
+	int count;
 	int err;
 
 	svc_udp_release_ctxt(xprt, rqstp->rq_xprt_ctxt);
@@ -752,6 +752,10 @@ static int svc_udp_sendto(struct svc_rqst *rqstp)
 		goto out_notconn;
 
 	count = xdr_buf_to_bvec(svsk->sk_bvec, SUNRPC_MAX_UDP_SENDPAGES, xdr);
+	if (count < 0) {
+		err = count;
+		goto out_trace;
+	}
 
 	iov_iter_bvec(&msg.msg_iter, ITER_SOURCE, svsk->sk_bvec,
 		      count, rqstp->rq_res.len);
@@ -763,6 +767,7 @@ static int svc_udp_sendto(struct svc_rqst *rqstp)
 		err = sock_sendmsg(svsk->sk_sock, &msg);
 	}
 
+out_trace:
 	trace_svcsock_udp_send(xprt, err);
 
 	mutex_unlock(&xprt->xpt_mutex);
@@ -1241,7 +1246,7 @@ static int svc_tcp_sendmsg(struct svc_sock *svsk, struct svc_rqst *rqstp,
 	struct msghdr msg = {
 		.msg_flags	= MSG_SPLICE_PAGES,
 	};
-	unsigned int count;
+	int count;
 	void *buf;
 	int ret;
 
@@ -1257,10 +1262,15 @@ static int svc_tcp_sendmsg(struct svc_sock *svsk, struct svc_rqst *rqstp,
 
 	count = xdr_buf_to_bvec(svsk->sk_bvec + 1, rqstp->rq_maxpages,
 				&rqstp->rq_res);
+	if (count < 0) {
+		ret = count;
+		goto out;
+	}
 
 	iov_iter_bvec(&msg.msg_iter, ITER_SOURCE, svsk->sk_bvec,
 		      1 + count, sizeof(marker) + rqstp->rq_res.len);
 	ret = sock_sendmsg(svsk->sk_sock, &msg);
+out:
 	page_frag_free(buf);
 	return ret;
 }
diff --git a/net/sunrpc/xdr.c b/net/sunrpc/xdr.c
index 82d3e8f65e486a..fb4dab2070007f 100644
--- a/net/sunrpc/xdr.c
+++ b/net/sunrpc/xdr.c
@@ -139,13 +139,14 @@ xdr_free_bvec(struct xdr_buf *buf)
 /**
  * xdr_buf_to_bvec - Copy components of an xdr_buf into a bio_vec array
  * @bvec: bio_vec array to populate
- * @bvec_size: element count of @bio_vec
+ * @bvec_size: element count of @bvec
  * @xdr: xdr_buf to be copied
  *
- * Returns the number of entries consumed in @bvec.
+ * Returns the number of entries consumed in @bvec on success, or
+ * -ESERVERFAULT when @xdr does not fit within @bvec_size entries.
  */
-unsigned int xdr_buf_to_bvec(struct bio_vec *bvec, unsigned int bvec_size,
-			     const struct xdr_buf *xdr)
+int xdr_buf_to_bvec(struct bio_vec *bvec, unsigned int bvec_size,
+		    const struct xdr_buf *xdr)
 {
 	const struct kvec *head = xdr->head;
 	const struct kvec *tail = xdr->tail;
@@ -187,7 +188,7 @@ unsigned int xdr_buf_to_bvec(struct bio_vec *bvec, unsigned int bvec_size,
 
 bvec_overflow:
 	pr_warn_once("%s: bio_vec array overflow\n", __func__);
-	return count;
+	return -ESERVERFAULT;
 }
 EXPORT_SYMBOL_GPL(xdr_buf_to_bvec);
 
-- 
2.53.0