summaryrefslogtreecommitdiff
path: root/src/pipe.cpp
AgeCommit message (Collapse)Author
2012-05-25Implement SP wire protocolMartin Lucina
Implements the SP wire protocol, and infrastructure for legacy wire protocol support. Also added an XS_SERVICE_ID socket option to set the service id and renamed the XS_PROTOCOL option to XS_PATTERN_VERSION. The following pattern versions are supported: PAIR: v3 PUBSUB: v1 (legacy), v4 REQREP: v2 PIPELINE: v3 SURVEY: v2 Note that all existing pattern versions have been bumped by 1 to allow for use of legacy protocols (otherwise there would be no way to distinguish between e.g. PUBSUB v3 and PUBSUB v3 using SP). Signed-off-by: Martin Lucina <martin@lucina.net>
2012-05-01Revert "Implement protocol versioning (except PGM)"Martin Sustrik
This reverts commit 512f3a604924fec9d89e2b4bfd6f73aa66309fa7.
2012-04-29Implement protocol versioning (except PGM)Martin Lucina
Implements SP protocol versioning, legacy protocol support, and the following pattern protocol versions: PAIR: v2 PUBSUB: v1 (legacy), v3 REQREP: v1 PIPELINE: v2 SURVEY: v1 Engine support is only for stream_engine_t at this stage. Signed-off-by: Martin Lucina <martin@lucina.net>
2012-04-06Allow pipe_t::flush() to be called from pipe_t::terminate()Martin Sustrik
Previous patch have broken flushing a delimiter in pipe_t::terminate() function. This patch allows the flush to happen even if pipe state is set to 'terminated'. Signed-off-by: Martin Sustrik <sustrik@250bpm.com>
2012-04-06Fix bogus typo in xs::pipe_t::flush()Andrew W. Nosenko
* src/pipe.cpp (flush): Fix bogus typo: '&&' was used instead of '||'
2012-04-040MQ/2.1 wire format compatibility implementedMartin Sustrik
- XS_PROTOCOL option added - libxs ignores when unused flags are set to 1 (0MQ/2.1 bug) - compatibility tests added Signed-off-by: Martin Sustrik <sustrik@250bpm.com>
2012-02-16Full name of the project used in license headersMartin Sustrik
Signed-off-by: Martin Sustrik <sustrik@250bpm.com>
2012-02-16ZeroMQ renamed to CrossroadsMartin Sustrik
Signed-off-by: Martin Sustrik <sustrik@250bpm.com>
2011-11-04Identities re-introducedMartin Sustrik
However, the "durable socket" behaviour wasn't re-added. Identities are used solely for routing in REQ/REP pattern. Signed-off-by: Martin Sustrik <sustrik@250bpm.com>
2011-11-01Copyright dates adjusted to reflect realityMartin Sustrik
Signed-off-by: Martin Sustrik <sustrik@250bpm.com>
2011-11-01LABELS and COMMANDs removedMartin Sustrik
Signed-off-by: Martin Sustrik <sustrik@250bpm.com>
2011-10-31250bpm copyrights addedMartin Sustrik
Signed-off-by: Martin Sustrik <sustrik@250bpm.com>
2011-06-23Fix MSVC buildMartin Sustrik
Signed-off-by: Martin Sustrik <sustrik@250bpm.com>
2011-06-22O(1) fair-queueing in XREP implementedMartin Sustrik
Up to now the complexity of fair-queueing in XREP was O(n). Signed-off-by: Martin Sustrik <sustrik@250bpm.com>
2011-06-20LABEL flag added to the wire formatMartin Sustrik
So far there was no distinction between message parts used by 0MQ and message parts used by user. Now, the message parts used by 0MQ are marked as 'LABEL'. Signed-off-by: Martin Sustrik <sustrik@250bpm.com>
2011-06-20Bug fixed in pipe terminationMartin Sustrik
Signed-off-by: Martin Sustrik <sustrik@250bpm.com>
2011-06-19Race condition in pipe_t fixed.Martin Sustrik
pipe_t now correctly drops pointer to the underlying pipe when sending pipe_term_ack command. Signed-off-by: Martin Sustrik <sustrik@250bpm.com>
2011-05-31Session termination induced by socket fixedMartin Sustrik
Signed-off-by: Martin Sustrik <sustrik@250bpm.com>
2011-05-30Patches from sub-forward branch incorporatedMartin Sustrik
Signed-off-by: Martin Sustrik <sustrik@250bpm.com>
2011-05-26Pending messages are delivered even if connection doesn't exist yetMartin Sustrik
Bug in previous refactoring fixed. Signed-off-by: Martin Sustrik <sustrik@250bpm.com>
2011-05-25All pipe termination code moved to pipe_tMartin Sustrik
Till now the code was spread over mutliple locations. Additionally, the code was made more formally correct, with explicit pipe state machine etc. Signed-off-by: Martin Sustrik <sustrik@250bpm.com>
2011-05-23Move the pipe termination code to socket_base_tMartin Sustrik
So far, the pipe termination code was spread among socket type classes, fair queuer, load balancer, etc. This patch moves all the associated logic to a single place. Signed-off-by: Martin Sustrik <sustrik@250bpm.com>
2011-05-22Introduces bi-directional pipesMartin Sustrik
So far, there was a pair of unidirectional pipes between a socket and a session (or an inproc peer). This resulted in complex problems with half-closed states and tracking which inpipe corresponds to which outpipe. This patch doesn't add any functionality in itself, but is essential for further work on features like subscription forwarding. Signed-off-by: Martin Sustrik <sustrik@250bpm.com>
2011-04-21Message-related functionality factored out into msg_t class.Martin Sustrik
This patch addresses serveral issues: 1. It gathers message related functionality scattered over whole codebase into a single class. 2. It makes zmq_msg_t an opaque datatype. Internals of the class don't pollute zmq.h header file. 3. zmq_msg_t size decreases from 48 to 32 bytes. That saves ~33% of memory in scenarios with large amount of small messages. Signed-off-by: Martin Sustrik <sustrik@250bpm.com>
2011-03-24ZMQ_HWM type changed to intMartin Sustrik
Signed-off-by: Martin Sustrik <sustrik@250bpm.com>
2011-03-24SWAP functionality removedMartin Sustrik
On-disk storage should be implemented in devices rather than in 0MQ core. 0MQ is a networking library and there's no point in storing network buffers on disk. Signed-off-by: Martin Sustrik <sustrik@250bpm.com>
2011-03-02The copyrights in file headers updated.Martin Sustrik
Signed-off-by: Martin Sustrik <sustrik@250bpm.com>
2011-02-22Added explicit error message in case of memory exhaustionMartin Sustrik
Signed-off-by: Martin Sustrik <sustrik@250bpm.com>
2011-01-07Fix pipe writer terminationDhammika Pathirana
Signed-off-by: Dhammika Pathirana <dhammika@gmail.com>
2010-12-15Fixes assertion on pipe.cpp:237 when swap fills up.Mikko Koppanen
Fixes swap::full () functionality Signed-off-by: Mikko Koppanen <mkoppanen@php.net>
2010-10-30Name of "GNU Lesser Public License" corrected.Martin Sustrik
Signed-off-by: Martin Sustrik <sustrik@250bpm.com>
2010-10-08Clean-up of session termination processMartin Sustrik
Specifically, shutdown of child objects is initiated *before* termination handshake with socket object. Signed-off-by: Martin Sustrik <sustrik@250bpm.com>
2010-09-13zmq::writer_t: Add missing test for swapMartin Lucina
2010-08-28msg_store_t renamed to swap_tMartin Sustrik
2010-08-28refactoring of pipe/swap interactionMartin Sustrik
2010-08-28terminology unified: revive->activateMartin Sustrik
2010-08-28bug in pipe deactivation fixedMartin Sustrik
2010-08-25REQ socket implementation is layered on top of XREQMartin Sustrik
2010-08-25Destruction of session is delayed till both in & out pipes are closedMartin Sustrik
2010-08-25WIP: Socket migration between threads, new zmq_close() semanticsMartin Sustrik
Sockets may now be migrated between OS threads; sockets may not be used by more than one thread at any time. To migrate a socket to another thread the caller must ensure that a full memory barrier is called before using the socket from the target thread. The new zmq_close() semantics implement the behaviour discussed at: http://lists.zeromq.org/pipermail/zeromq-dev/2010-July/004244.html Specifically, zmq_close() is now deterministic and while it still returns immediately, it does not discard any data that may still be queued for sending. Further, zmq_term() will now block until all outstanding data has been sent. TODO: Many bugs have been introduced, needs testing. Further, SO_LINGER or an equivalent mechanism (possibly a configurable timeout to zmq_term()) needs to be implemented.
2010-07-24pipe: check_read() should check for message delimiterMartin Hurton
2010-07-24Make sure lwm > 0 when hwm > 0Martin Hurton
2010-06-21Add SWAP supportMartin Hurton
2010-06-19Revive reader on pipe terminationMartin Hurton
2010-06-17pipe: fix bug in rollback() methodMartin Hurton
The msgs_written variable keeps track how many complete messages have been written so far. The rollback operation drops all fragments of the last incomplete message so it shouldn't change this variable at all.
2010-05-25LWM is computed rather than explicitly specified by userMartin Sustrik
2010-05-19incomplete messages can be stored in ypipeMartin Sustrik
2010-03-27ZMQ_TBC renamed to ZMQ_MOREMartin Sustrik
2010-03-27multi-part messages work with PUB/SUBMartin Sustrik
2010-03-13pipe_t::rollback removes only unfinished message from the pipe rather than ↵Martin Sustrik
all unflushed messages