5.4-stable review patch. If anyone has any objections, please let me know.
------------------
From: Linus Walleij linus.walleij@linaro.org
[ Upstream commit ac631873c9e7a50d2a8de457cfc4b9f86666403e ]
The recent change to allow large frames without hardware checksumming slotted in software checksumming in the driver if hardware could not do it.
This will however upset TSO (TCP Segment Offloading). Typical error dumps includes this:
skb len=2961 headroom=222 headlen=66 tailroom=0 (...) WARNING: CPU: 0 PID: 956 at net/core/dev.c:3259 skb_warn_bad_offload+0x7c/0x108 gemini-ethernet-port: caps=(0x0000010000154813, 0x00002007ffdd7889)
And the packets do not go through.
The TSO implementation is bogus: a TSO enabled driver must propagate the skb_shinfo(skb)->gso_size value to the TSO engine on the NIC.
Drop the size check and TSO offloading features for now: this needs to be fixed up properly.
After this ethernet works fine on Gemini devices with a direct connected PHY such as D-Link DNS-313.
Also tested to still be working with a DSA switch using the Gemini ethernet as conduit interface.
Link: https://lore.kernel.org/netdev/CANn89iJLfxng1sYL5Zk0mknXpyYQPCp83m3KgD2KJ2_h... Suggested-by: Eric Dumazet edumazet@google.com Fixes: d4d0c5b4d279 ("net: ethernet: cortina: Handle large frames") Signed-off-by: Linus Walleij linus.walleij@linaro.org Reviewed-by: Eric Dumazet edumazet@google.com Signed-off-by: David S. Miller davem@davemloft.net Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/net/ethernet/cortina/gemini.c | 15 ++------------- 1 file changed, 2 insertions(+), 13 deletions(-)
diff --git a/drivers/net/ethernet/cortina/gemini.c b/drivers/net/ethernet/cortina/gemini.c index 91952d086f226..8b5ed26cab03e 100644 --- a/drivers/net/ethernet/cortina/gemini.c +++ b/drivers/net/ethernet/cortina/gemini.c @@ -81,8 +81,7 @@ MODULE_PARM_DESC(debug, "Debug level (0=none,...,16=all)"); #define GMAC0_IRQ4_8 (GMAC0_MIB_INT_BIT | GMAC0_RX_OVERRUN_INT_BIT)
#define GMAC_OFFLOAD_FEATURES (NETIF_F_SG | NETIF_F_IP_CSUM | \ - NETIF_F_IPV6_CSUM | NETIF_F_RXCSUM | \ - NETIF_F_TSO | NETIF_F_TSO_ECN | NETIF_F_TSO6) + NETIF_F_IPV6_CSUM | NETIF_F_RXCSUM)
/** * struct gmac_queue_page - page buffer per-page info @@ -1155,23 +1154,13 @@ static int gmac_map_tx_bufs(struct net_device *netdev, struct sk_buff *skb, struct gmac_txdesc *txd; skb_frag_t *skb_frag; dma_addr_t mapping; - unsigned short mtu; void *buffer; int ret;
- mtu = ETH_HLEN; - mtu += netdev->mtu; - if (skb->protocol == htons(ETH_P_8021Q)) - mtu += VLAN_HLEN; - + /* TODO: implement proper TSO using MTU in word3 */ word1 = skb->len; word3 = SOF_BIT;
- if (word1 > mtu) { - word1 |= TSS_MTU_ENABLE_BIT; - word3 |= mtu; - } - if (skb->len >= ETH_FRAME_LEN) { /* Hardware offloaded checksumming isn't working on frames * bigger than 1514 bytes. A hypothesis about this is that the