Received-SPF: pass (google.com: domain of linux-kernel+bounces-113571-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:45e3:2400::1 as permitted sender) client-ip=2604:1380:45e3:2400::1;
From: Sasha Levin <sashal@kernel.org>
To: linux-kernel@vger.kernel.org,
	stable@vger.kernel.org
Cc: Xuan Zhuo <xuanzhuo@linux.alibaba.com>,
	"Michael S . Tsirkin" <mst@redhat.com>,
	Sasha Levin <sashal@kernel.org>
Subject: [PATCH 6.8 680/715] virtio: packed: fix unmap leak for indirect desc table
Date: Sun, 24 Mar 2024 18:34:19 -0400
Message-ID: <20240324223455.1342824-681-sashal@kernel.org>
In-Reply-To: <20240324223455.1342824-1-sashal@kernel.org>
References: <20240324223455.1342824-1-sashal@kernel.org>
Precedence: bulk
MIME-Version: 1.0
Content-Transfer-Encoding: 8bit

From: Xuan Zhuo <xuanzhuo@linux.alibaba.com>

[ Upstream commit d5c0ed17fea60cca9bc3bf1278b49ba79242bbcd ]

When use_dma_api and premapped are true, then the do_unmap is false.

Because the do_unmap is false, vring_unmap_extra_packed is not called by
detach_buf_packed.

  if (unlikely(vq->do_unmap)) {
                curr = id;
                for (i = 0; i < state->num; i++) {
                        vring_unmap_extra_packed(vq,
                                                 &vq->packed.desc_extra[curr]);
                        curr = vq->packed.desc_extra[curr].next;
                }
  }

So the indirect desc table is not unmapped. This causes the unmap leak.

So here, we check vq->use_dma_api instead. Synchronously, dma info is
updated based on use_dma_api judgment

This bug does not occur, because no driver use the premapped with
indirect.

Fixes: b319940f83c2 ("virtio_ring: skip unmap for premapped")
Signed-off-by: Xuan Zhuo <xuanzhuo@linux.alibaba.com>
Message-Id: <20240223071833.26095-1-xuanzhuo@linux.alibaba.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 drivers/virtio/virtio_ring.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/virtio/virtio_ring.c b/drivers/virtio/virtio_ring.c
index 49299b1f9ec74..6f7e5010a6735 100644
--- a/drivers/virtio/virtio_ring.c
+++ b/drivers/virtio/virtio_ring.c
@@ -1340,7 +1340,7 @@ static int virtqueue_add_indirect_packed(struct vring_virtqueue *vq,
 				sizeof(struct vring_packed_desc));
 	vq->packed.vring.desc[head].id = cpu_to_le16(id);
 
-	if (vq->do_unmap) {
+	if (vq->use_dma_api) {
 		vq->packed.desc_extra[id].addr = addr;
 		vq->packed.desc_extra[id].len = total_sg *
 				sizeof(struct vring_packed_desc);
@@ -1481,7 +1481,7 @@ static inline int virtqueue_add_packed(struct virtqueue *_vq,
 			desc[i].len = cpu_to_le32(sg->length);
 			desc[i].id = cpu_to_le16(id);
 
-			if (unlikely(vq->do_unmap)) {
+			if (unlikely(vq->use_dma_api)) {
 				vq->packed.desc_extra[curr].addr = addr;
 				vq->packed.desc_extra[curr].len = sg->length;
 				vq->packed.desc_extra[curr].flags =
@@ -1615,7 +1615,7 @@ static void detach_buf_packed(struct vring_virtqueue *vq,
 	vq->free_head = id;
 	vq->vq.num_free += state->num;
 
-	if (unlikely(vq->do_unmap)) {
+	if (unlikely(vq->use_dma_api)) {
 		curr = id;
 		for (i = 0; i < state->num; i++) {
 			vring_unmap_extra_packed(vq,
-- 
2.43.0