Christoph Paasch

At least, when I test-compiled it with mobile-IPv6-enabled I don't think I got an error. You can maybe just try out whith allyesmod. If it compiles, then 48 is fine.

Increasing the size of the skb slows down the networking stack because accessing fields of the skb will incur additional cacheline misses. The upstream community is trying to minimize the...

Isn’t this the same problem as #347?

Writing down some notes on this issue here as I am looking into it: There is packet-loss happening and from that moment on the connection is going to stall. What...

This sounds like we don't fill the WiFi channel enough to get the desired throughput. It's rather a low-level issue. Just as a shot in the dark, can you increase...

Yeah, looking at the trace it really seems like tcp_limit_output_bytes is the culprit. Please let us know if it helped. If yes, we have to fix something in that area.

Hello @jaehyun-hwang , your analysis seems to me to confirm that tcp_limit_output_bytes is the "problem". Can someone who can reproduce this issue try out increasing tcp_limit_output_bytes ? (I don't have...

Thanks!

I will take a closer look at your packet-traces.

@jaehyun-hwang Thanks for the explanation (and sorry for the delay). When the "another ack 3-4" comes in at TCP, it should schedule a call to ```mptcp_write_xmit()```, which should trigger the...