This patch intend is to improve overall io performance, reduce
cpu and system usage.
Machinarium:
IO read/write logic is highly simplified and now based on conditions
instead of previously made internal readahead buffer and msg
queues. This allows to give user more control over io operations and
avoid doing undesired readahead copying.
Above that, TLS implementation reworked to use async logic instead of
separate BIO layer.
Direct pooling functions removed.
New objects machine_cond_t and machine_iov_t objects.
Odyssey:
Introduced readahead, io and relay objects.
Relay allows to connect two io handles for direct data retransmission
without double copying.
Removed configuration options no longer needed:
packet_read_size, packet_write_queue
Goal is to improve overall performance during massive peaks
of connecting clients.
This patch makes router a shared structure, instead of
being run in separate system worker thread/coroutine.
Now router and each route object is protected by
separate mutex. Worker threads directly access router object
without calls to separate thread.
This patch moves Odyssey from stream based write caching
to message driven. All write operations are delayed
and happen on a group of messages.
This removes necessity to do IO and wait for write completion during calls
from SSL_write() context. Write timings should behave more predictable now and
number of used syscalls should be also decreased.
This patch also simplifies caching management. All clients are simply
reusing cached messages without prolonged contention.
Several configuration options are no longer used:
pipeline, cache, cache_chunk
Andrey Borodin