From 84ad6e44915a797cbab81c3042c600955ca75408 Mon Sep 17 00:00:00 2001 From: Yuanhan Liu Date: Wed, 19 Apr 2017 13:26:01 +0800 Subject: [PATCH] vhost: fix dequeue zero copy For zero copy mode, we need pin the mbuf to not let the underlaying PMD driver (or the app) free the mbuf. Currently, only the heading mbuf is pinned. However, the mbuf free function would try to free all mbufs in the mbuf chain (-1 to the refcnt). This may lead the head mbuf being still pinned, while the other subsequent mbufs are actually freed. Which is wrong. It becomes more fatal after the mbuf refactor, more specificly, after the commit 8f094a9ac5d7 ("mbuf: set mbuf fields while in pool"). The refcnt resets to 1 after the last real reference. OTOH, it leads to a situtation that we never know one mbuf is actually freed or not. This would result the mbuf __just__ after the heading mbuf being freed twice: it's firstly freed (and put back to mempool) when the underlaying PMD finishes the DMA. Later, it will then be freed again when vhost unpins it. Meaning, one mbuf may be returned to the mempool twice, while in turn, being allocated twice later. Something uncertain may happen then. For example, the VM2VM case becomes broken. Fixes: b0a985d1f340 ("vhost: add dequeue zero copy") Cc: stable@dpdk.org Signed-off-by: Yuanhan Liu --- lib/librte_vhost/virtio_net.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/lib/librte_vhost/virtio_net.c b/lib/librte_vhost/virtio_net.c index 78327a6ec2..48219e0509 100644 --- a/lib/librte_vhost/virtio_net.c +++ b/lib/librte_vhost/virtio_net.c @@ -876,6 +876,8 @@ copy_desc_to_mbuf(struct virtio_net *dev, struct vring_desc *descs, "allocate memory for mbuf.\n"); return -1; } + if (unlikely(dev->dequeue_zero_copy)) + rte_mbuf_refcnt_update(cur, 1); prev->next = cur; prev->data_len = mbuf_offset; -- 2.20.1