net: bcmgenet: Fix FCS generation for fragmented skbuffs

[ Upstream commit e584f2ff1e6cc9b1d99e8a6b0f3415940d1b3eb3 ]

The flag DMA_TX_APPEND_CRC was only written to the first DMA descriptor
in the TX path, where each descriptor corresponds to a single skbuff
fragment (or the skbuff head). This led to packets with no FCS appearing
on the wire if the kernel allocated the packet in fragments, which would
always happen when using PACKET_MMAP/TPACKET (cf. tpacket_fill_skb() in
net/af_packet.c).

Fixes: 1c1008c793fa ("net: bcmgenet: add main driver file")
Signed-off-by: Adrian Cinal <adriancinal1@gmail.com>
Acked-by: Doug Berger <opendmb@gmail.com>
Acked-by: Florian Fainelli <florian.fainelli@broadcom.com>
Link: https://lore.kernel.org/r/20231228135638.1339245-1-adriancinal1@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

diff --git a/drivers/net/ethernet/broadcom/genet/bcmgenet.c b/drivers/net/ethernet/broadcom/genet/bcmgenet.c
index e5e52c0c39a55fb970c3f31bc802ad394a142a88..445e892022eb589a0c7f41a6e9fd3c89185e4ced 100644 (file)
--- a/drivers/net/ethernet/broadcom/genet/bcmgenet.c
+++ b/drivers/net/ethernet/broadcom/genet/bcmgenet.c
@@ -1625,8 +1625,10 @@ static netdev_tx_t bcmgenet_xmit(struct sk_buff *skb, struct net_device *dev)
                /* Note: if we ever change from DMA_TX_APPEND_CRC below we
                 * will need to restore software padding of "runt" packets
                 */
+               len_stat |= DMA_TX_APPEND_CRC;
+
                if (!i) {
-                       len_stat |= DMA_TX_APPEND_CRC | DMA_SOP;
+                       len_stat |= DMA_SOP;
                        if (skb->ip_summed == CHECKSUM_PARTIAL)
                                len_stat |= DMA_TX_DO_CSUM;
                }