mirror of
https://github.com/ElementsProject/lightning.git
synced 2025-02-24 23:18:25 +01:00
A fairly simple change: ccan/io will now call the underlying I/O routines repeatedly until they indicate they are unfinished, *or* fail with EAGAIN. This should make a significant difference to large nodes, which currently spend far too much time calling poll() to discover a single fd is still writable (mainly, for streaming gossip). Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> Changelog-Changed: connectd: now should use far less CPU on large nodes. |
||
---|---|---|
.. | ||
run-01-start-finish.c | ||
run-02-read.c | ||
run-03-readpartial.c | ||
run-04-writepartial.c | ||
run-05-write.c | ||
run-06-idle.c | ||
run-07-break.c | ||
run-08-hangup-on-idle.c | ||
run-08-read-after-hangup.c | ||
run-09-connect.c | ||
run-10-many.c | ||
run-12-bidir.c | ||
run-13-all-idle.c | ||
run-14-duplex-both-read.c | ||
run-15-timeout.c | ||
run-16-duplex-test.c | ||
run-17-homemade-io.c | ||
run-18-errno.c | ||
run-19-always.c | ||
run-20-io_time_override.c | ||
run-21-io_close_taken_fd.c | ||
run-22-POLLHUP-on-listening-socket.c | ||
run-30-io_flush_sync.c | ||
run-40-wakeup-mutual.c | ||
run-41-io_poll_override.c | ||
run-43-io_plan_in_started.c | ||
run-44-io_plan_out_started.c | ||
run-45-eof.c | ||
run-46-exclusive.c | ||
run-47-exclusive-duplex.c | ||
run-48-exclusive-duplex-write.c |