Re: BUG: sk_backlog.len can overestimate

[Eric Dumazet]

On 10/1/19 1:45 PM, John Ousterhout wrote:

> 
> But this isn't really about socket resource limits (though that is
> conflated in the implementation); it's about limiting the time spent
> in a single call to __release_sock, no?

The proxy used is memory usage, not time usage.

cond_resched() or a preemptible kernel makes anything based on time flaky,
you probably do not want to play with a time limit...