On 22/07/21(Thu) 15:03, Jan Klemkow wrote:
> Hi,
>
> This calculation of the receive window has a logic error:
>
> If win is 0 it will be overwritten by (rcv_adv - rcv_nxt). Thus, win
> will be (rcv_adv - rcv_nxt) even if its below (sb_hiwat / 4).
Why is this a problem?
> We could just remove the dead (sb_hiwat / 4) code, or reorder the
> conditions to keep the original feature.
>
> OK?
>
> bye,
> Jan
>
> Index: netinet/tcp_output.c
> ===
> RCS file: /cvs/src/sys/netinet/tcp_output.c,v
> retrieving revision 1.130
> diff -u -p -r1.130 tcp_output.c
> --- netinet/tcp_output.c 8 Feb 2021 19:37:15 - 1.130
> +++ netinet/tcp_output.c 22 Jul 2021 12:33:13 -
> @@ -812,12 +812,12 @@ send:
>* Calculate receive window. Don't shrink window,
>* but avoid silly window syndrome.
>*/
> - if (win < (long)(so->so_rcv.sb_hiwat / 4) && win < (long)tp->t_maxseg)
> - win = 0;
> if (win > (long)TCP_MAXWIN << tp->rcv_scale)
> win = (long)TCP_MAXWIN << tp->rcv_scale;
> if (win < (long)(int32_t)(tp->rcv_adv - tp->rcv_nxt))
> win = (long)(int32_t)(tp->rcv_adv - tp->rcv_nxt);
> + if (win < (long)(so->so_rcv.sb_hiwat / 4) && win < (long)tp->t_maxseg)
> + win = 0;
> if (flags & TH_RST)
> win = 0;
> th->th_win = htons((u_int16_t) (win>>tp->rcv_scale));
>
