- Name
-
ul_select_fast_usec
- Default
-
32
- Scope
- per-process
When spinning in a select()
call, causes accelerated sockets to be polled for
N μsecs before unaccelerated sockets are polled. This reduces latency for accelerated
sockets, possibly at the expense of latency on unaccelerated sockets. Because
accelerated sockets are typically the parts of the application which are most
performance-sensitive this is typically a good tradeoff.