On Tue, 31 Oct 2017 07:58:21 -0700 Davidlohr Bueso <[email protected]> wrote:
> On Tue, 31 Oct 2017, Jason Baron wrote: > > >The use of ep_call_nested() in ep_eventpoll_poll(), which is the .poll > >routine for an epoll fd, is used to prevent excessively deep epoll > >nesting, and to prevent circular paths. However, we are already preventing > >these conditions during EPOLL_CTL_ADD. In terms of too deep epoll chains, > >we do in fact allow deep nesting of the epoll fds themselves (deeper > >than EP_MAX_NESTS), however we don't allow more than EP_MAX_NESTS when > >an epoll file descriptor is actually connected to a wakeup source. Thus, > >we do not require the use of ep_call_nested(), since ep_eventpoll_poll(), > >which is called via ep_scan_ready_list() only continues nesting if there > >are events available. Since ep_call_nested() is implemented using a global > >lock, applications that make use of nested epoll can see large performance > >improvements with this change. > > Improvements are quite obscene actually, such as for the following > epoll_wait() > benchmark with 2 level nesting on a 80 core IvyBridge: > > ncpus vanilla dirty delta > 1 2447092 3028315 +23.75% > 4 231265 2986954 +1191.57% > 8 121631 2898796 +2283.27% > 16 59749 2902056 +4757.07% > 32 26837 2326314 +8568.30% > 64 12926 1341281 +10276.61% > > (http://linux-scalability.org/epoll/epoll-test.c) This is tempting, but boy it is late in the -rc cycle. How important are these workloads? Would the world end if we held off on this for 4.15?
