On 7/11/19 8:26 PM, Michal Kubecek wrote: > > I'm aware it's not a realistic test. It was written as quick and simple > check of the pre-4.19 patch, but it shows that even TLP may not get > through. Most of TLP probes send new data, not rtx. But yes, I get your point. SO_SNDBUF=15000 in your case is seriously wrong. Lets code a safety feature over SO_SNDBUF to not allow pathological small values, because I do not want to support a constrained TCP stack in 2019.
- RE: [PATCH net 2/4] tcp: tcp_fragment() shoul... Prout, Andrew - LLSC - MITLL
- Re: [PATCH net 2/4] tcp: tcp_fragment() ... Eric Dumazet
- RE: [PATCH net 2/4] tcp: tcp_fragmen... Prout, Andrew - LLSC - MITLL
- Re: [PATCH net 2/4] tcp: tcp_fra... Eric Dumazet
- Re: [PATCH net 2/4] tcp: tcp... Christoph Paasch
- Re: [PATCH net 2/4] tcp... Eric Dumazet
- Re: [PATCH net 2/4]... Michal Kubecek
- Re: [PATCH net ... Eric Dumazet
- Re: [PATCH net 2/4] tcp... Eric Dumazet
- RE: [PATCH net 2/4] tcp: tcp... Prout, Andrew - LLSC - MITLL
- Re: [PATCH net 2/4] tcp... Eric Dumazet
- Re: [PATCH net 2/4]... Jonathan Lemon
- Re: [PATCH net ... Eric Dumazet
- RE: [PATCH net 2/4] tcp... Prout, Andrew - LLSC - MITLL
- Re: [PATCH net 2/4] tcp: tcp_fragment() ... Bernd
- Re: [PATCH net 2/4] tcp: tcp_fragmen... Neal Cardwell