4.9-stable review patch. If anyone has any objections, please let me know.
------------------
From: Sarah Newman srn@prgmr.com
[ Not relevant upstream, therefore no upstream commit. ]
To fix, unmap the page as soon as possible.
When swiotlb is in use, calling dma_unmap_page means that the original page mapped with dma_map_page must still be valid, as swiotlb will copy data from its internal cache back to the originally requested DMA location.
When GRO is enabled, before this patch all references to the original frag may be put and the page freed before dma_unmap_page in mlx4_en_free_frag is called.
It is possible there is a path where the use-after-free occurs even with GRO disabled, but this has not been observed so far.
The bug can be trivially detected by doing the following:
* Compile the kernel with DEBUG_PAGEALLOC * Run the kernel as a Xen Dom0 * Leave GRO enabled on the interface * Run a 10 second or more test with iperf over the interface.
This bug was likely introduced in commit 4cce66cdd14a ("mlx4_en: map entire pages to increase throughput"), first part of u3.6.
It was incidentally fixed in commit 34db548bfb95 ("mlx4: add page recycling in receive path"), first part of v4.12.
This version applies to the v4.9 series.
Signed-off-by: Sarah Newman srn@prgmr.com Tested-by: Sarah Newman srn@prgmr.com Cc: Tariq Toukan tariqt@mellanox.com Cc: Yishai Hadas yishaih@mellanox.com Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org --- drivers/net/ethernet/mellanox/mlx4/en_rx.c | 32 ++++++++++++++++++----------- 1 file changed, 20 insertions(+), 12 deletions(-)
--- a/drivers/net/ethernet/mellanox/mlx4/en_rx.c +++ b/drivers/net/ethernet/mellanox/mlx4/en_rx.c @@ -142,16 +142,17 @@ static void mlx4_en_free_frag(struct mlx struct mlx4_en_rx_alloc *frags, int i) { - const struct mlx4_en_frag_info *frag_info = &priv->frag_info[i]; - u32 next_frag_end = frags[i].page_offset + 2 * frag_info->frag_stride; - - - if (next_frag_end > frags[i].page_size) - dma_unmap_page(priv->ddev, frags[i].dma, frags[i].page_size, - frag_info->dma_dir); + if (frags[i].page) { + const struct mlx4_en_frag_info *frag_info = &priv->frag_info[i]; + u32 next_frag_end = frags[i].page_offset + + 2 * frag_info->frag_stride;
- if (frags[i].page) + if (next_frag_end > frags[i].page_size) { + dma_unmap_page(priv->ddev, frags[i].dma, + frags[i].page_size, frag_info->dma_dir); + } put_page(frags[i].page); + } }
static int mlx4_en_init_allocator(struct mlx4_en_priv *priv, @@ -586,21 +587,28 @@ static int mlx4_en_complete_rx_desc(stru int length) { struct skb_frag_struct *skb_frags_rx = skb_shinfo(skb)->frags; - struct mlx4_en_frag_info *frag_info; int nr; dma_addr_t dma;
/* Collect used fragments while replacing them in the HW descriptors */ for (nr = 0; nr < priv->num_frags; nr++) { - frag_info = &priv->frag_info[nr]; + struct mlx4_en_frag_info *frag_info = &priv->frag_info[nr]; + u32 next_frag_end = frags[nr].page_offset + + 2 * frag_info->frag_stride; + if (length <= frag_info->frag_prefix_size) break; if (unlikely(!frags[nr].page)) goto fail;
dma = be64_to_cpu(rx_desc->data[nr].addr); - dma_sync_single_for_cpu(priv->ddev, dma, frag_info->frag_size, - DMA_FROM_DEVICE); + if (next_frag_end > frags[nr].page_size) + dma_unmap_page(priv->ddev, frags[nr].dma, + frags[nr].page_size, frag_info->dma_dir); + else + dma_sync_single_for_cpu(priv->ddev, dma, + frag_info->frag_size, + DMA_FROM_DEVICE);
/* Save page reference in skb */ __skb_frag_set_page(&skb_frags_rx[nr], frags[nr].page);