On 9/2/20 12:56 PM, Mihail Milev wrote:
> What?
>
> Create a new sysctl parameter called tcp_disable_rst_unkn_socket,
> which by default is set to 0 - "disabled". When this parameter is
> set to 1 - "enabled", it suppresses sending a TCP RST packet as a
> response to received TCP packets destined for a socket, which is
> unknown to the kernel.
>
Well, I am not thrilled by this patch...
1) This seems hacky to only focus on RST packets, while it seems clear you
want to be able to use a user-land TCP stack.
2) No ipv6 support in your patch, this is unfortunate.
3) I do not see how you prevent another program using kernel stack conflicting
with TCP listeners/flows of your user space TCP.
