Re: Deadlock in combination with pthread, lwip and grpc

Sid,

On Fri, May 22, 2020 at 14:24:10 CEST, Sid Hussmann wrote:

It is very reasonable indeed, and we should hunt this issue as well. I ran into this assertion at some point as too. However, I wanted to focus first on the not so obvious "deadlock" problem, and takle the latter after. But I'm open to ideas on how to make lwIP as stable as possible.

That's only fair! I'd also love to stabilize one of the network stacks more (whether it be lwip or lxip).

Should we move this discussion to github and create two issues?

I'd much appreciate one new issue for the lwip-stress stability referring to the original test case (gRPC) and the current much reduced one (your test and netcat).

Regards