On 11/28/2018 04:05 PM, Cong Wang wrote: > While we are on this page, mlx5e_lro_update_hdr() incorrectly assumes > TCP header is located right after struct iphdr, which is wrong if we could > have IP options on this path. > > It could the hardware which already verified this corner case though. > GRO makes sure IPv4 header has no options, so I would not be surprised that LRO on mlx5 has the same logic.
- [Patch net v2] mlx5: fixup checksum for short ethernet fram... Cong Wang
- Re: [Patch net v2] mlx5: fixup checksum for short ethe... Eric Dumazet
- Re: [Patch net v2] mlx5: fixup checksum for short ... Cong Wang
- Re: [Patch net v2] mlx5: fixup checksum for sh... Eric Dumazet
- Re: [Patch net v2] mlx5: fixup checksum fo... Cong Wang
- Re: [Patch net v2] mlx5: fixup checks... Cong Wang
- Re: [Patch net v2] mlx5: fixup ch... Eric Dumazet
- Re: [Patch net v2] mlx5: fixup ch... Saeed Mahameed
- Re: [Patch net v2] mlx5: fixup checks... Eric Dumazet
- Re: [Patch net v2] mlx5: fixup ch... Cong Wang
- Re: [Patch net v2] mlx5: fixu... Eric Dumazet
- Re: [Patch net v2] mlx5: fixu... Cong Wang
- Re: [Patch net v2] mlx5: fixu... Eric Dumazet
- Re: [Patch net v2] mlx5: fixu... Saeed Mahameed
- Re: [Patch net v2] mlx5: fixup checksum for short ethe... David Miller