From: Nicolas Schichan <nschichan@xxxxxxxxxx>
commit 3b89624ab54b9dc2d92fc08ce2670e5f19ad8ec8 upstream.
The code in txq_put_data() would use txq->tx_curr_desc to index the
tso_hdrs/tso_hdrs_dma buffers, for less than 8 bytes unaligned
fragments, which is already moved to the next descriptor at the
beginning of the function.
If that fragment was the last of the the skb, the next skb would use
that same space to place the ip headers, overwritting that small
fragment data.
Fixes: 91986fd3d335 (net: mv643xx_eth: Ensure proper data alignment in TSO TX path)
Signed-off-by: Nicolas Schichan <nschichan@xxxxxxxxxx>
Reviewed-by: Philipp Kirchhofer <philipp@xxxxxxxxxxxxxxxxxxxxx>
Signed-off-by: David S. Miller <davem@xxxxxxxxxxxxx>
Signed-off-by: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx>
---
drivers/net/ethernet/marvell/mv643xx_eth.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
--- a/drivers/net/ethernet/marvell/mv643xx_eth.c
+++ b/drivers/net/ethernet/marvell/mv643xx_eth.c
@@ -762,10 +762,10 @@ txq_put_data_tso(struct net_device *dev,
if (length <= 8 && (uintptr_t)data & 0x7) {
/* Copy unaligned small data fragment to TSO header data area */
- memcpy(txq->tso_hdrs + txq->tx_curr_desc * TSO_HEADER_SIZE,
+ memcpy(txq->tso_hdrs + tx_index * TSO_HEADER_SIZE,
data, length);
desc->buf_ptr = txq->tso_hdrs_dma
- + txq->tx_curr_desc * TSO_HEADER_SIZE;
+ + tx_index * TSO_HEADER_SIZE;
} else {
/* Alignment is okay, map buffer and hand off to hardware */
txq->tx_desc_mapping[tx_index] = DESC_DMA_MAP_SINGLE;
