> Ported from 7ac9ca128885c5dd561e6fbd6bbeaddb86d6264c to the latest upstream > fibers branch. Adapted to the current API which renamed > implRegister/implDeregister to implStartPoll/implStopPoll and added > Mode/EventFD/Cleaner/PollerGroup architecture.
Francesco Nigro has updated the pull request incrementally with one additional commit since the last revision: Add JMH benchmark for VT read poller registration overhead RPC-style benchmark where a platform NIO echo server (selectNow spin loop) handles requests from virtual threads doing blocking socket I/O. Each JMH invocation spawns a VT that writes a request then blocking- reads the response, exercising the poller registration path. Per-thread connections with pre-allocated buffers for zero GC in steady state. Use with -Djdk.pollerMode=2 to compare edge-triggered vs one-shot epoll registration. ------------- Changes: - all: https://git.openjdk.org/loom/pull/223/files - new: https://git.openjdk.org/loom/pull/223/files/1ac6dc35..28755d93 Webrevs: - full: https://webrevs.openjdk.org/?repo=loom&pr=223&range=02 - incr: https://webrevs.openjdk.org/?repo=loom&pr=223&range=01-02 Stats: 215 lines in 1 file changed: 215 ins; 0 del; 0 mod Patch: https://git.openjdk.org/loom/pull/223.diff Fetch: git fetch https://git.openjdk.org/loom.git pull/223/head:pull/223 PR: https://git.openjdk.org/loom/pull/223
