4566 Commits (62d7192a5059626396d6f0aa6b65380fde79db69)

Author	SHA1	Message	Date
Jeff Becker	7c6bb9a24c	simplify loop and correct logic, we want to bail out of the loop on first match.	3 years ago
Jeff Becker	ae97b87111	exits don't rewrite ip addresses, fix this and add note	3 years ago
Jeff Becker	f1efcc29ea	mark inbound traffic as convotag active	3 years ago
Jeff	bd9db35975	Merge pull request #1643 from majestrate/win32-rpc-fixups-2021-05-13 ... win32 fixups	3 years ago
Jeff Becker	499bb38e6f	fix route poking via rpc: ... * immediately poke routes when we are told to use an exit so that packets get pushed which makes an exit path happen * fix up cmake oddity in nsis section	3 years ago
Jeff Becker	209bcc39dd	make liblokinet target work on nsis cpack	3 years ago
Jeff Becker	8c0f448e12	add liblokinet option to nsis cpack installer	3 years ago
Jeff Becker	2458b5fd71	rpc server fixups for win32: ... * RoutePoker::Enable calls RoutePoker::Up so remove additional call to RoutePoker::Up * allow specifying null exit via rpc	3 years ago
Jeff Becker	3c2334112c	when we stop a path builder we want to expire all of their paths so they go away	3 years ago
Jeff Becker	51b7566a46	if we look deregistered we will now: ... * not gossip our rc * not explore the network to prevent outbound session attempts * not establish sessions to other service nodes * close all open sessions we have to tell clients we don't want them * catch exceptions flushing peerdb in disk thread * don't connect out to non allowed routers * simplify logic in RCLookupHandler::RemoteIsAllowed() * add HaveReceivedWhitelist to I_RCLookupHandler base type * add LooksDeregistered to Router type that tells us if we think we are deregistered * don't allow building paths over us if we are deregistered	3 years ago
Jeff Becker	c5350ab2c4	DefaultPathAlignmentTimeout -> PathAlignmentTimeout	3 years ago
Jeff Becker	c834414b47	when we have to reestablish an outbound session use a much much higher timeout for restablishment ... so it doesn't time out and get into a state that's totally screwed. add virtual function service::Endpont::DefaultPathAlignmentTimeout() to get the timeout for path alignment and use it for resetablishing outbound sessions	3 years ago
Jeff Becker	38cc130dc3	shift off bad intros when we get a discard message, this happens if the ... other side's intro expires or the pivot restarts for whatever reason.	3 years ago
Jeff Becker	3c22e01d7c	in the event that a session is removed and then gets more traffic we would re-add the session with everything blank if there was more traffic in the same tick. ... this remedies this behavior by only increment usage timestamps on sessions if they exist.	3 years ago
Jeff Becker	6bb31468d7	don't send nx if we have an address already mapped	3 years ago
Jeff Becker	d3d929efa7	fixups in service endpoint ... * increase publish introset timeout so that it does not time out on the network * remove pedantic log warn * make sure the path we are using for replying on inbound sessions is alive * include convotag in log message so we know wtf is going on * appease tom's autism, improve log message text	3 years ago
Jeff	d53945b011	Merge pull request #1621 from majestrate/path-ptr-leak-2021-05-02 ... try fixing std::shared_ptr leak with paths	3 years ago
Jeff	34eb254959	Merge pull request #1626 from majestrate/edge-limiter-2021-05-05 ... limit path builds across all builders	3 years ago
Jeff	ea4529af92	Merge pull request #1628 from majestrate/unfug-network-path-builds-2021-05-05 ... always do path tests	3 years ago
Jeff	2bb5a7ccea	Merge pull request #1622 from majestrate/stricter-profiling-2021-05-03 ... stricter profiling	3 years ago
Jeff Becker	b1afe0f596	always do path tests ... this reverts some stupid bullshit that broke 0.9.0	3 years ago
Jeff Becker	ec62228149	limit path builds across all builders	3 years ago
Jeff Becker	955071ba5c	make profiling far stricter for path builds	3 years ago
Jeff Becker	e916c9610c	format	3 years ago
Jeff Becker	554a44c8bf	report block height reported by oxend in systemd status	3 years ago
Jeff Becker	f955bec5da	re-add parans	3 years ago
Jeff Becker	0826a557d6	use correct intro	3 years ago
Jeff Becker	f108af3d8b	only use alive paths for lookups ... only track usage from sessions and don't care about intro timeouts	3 years ago
Jeff Becker	0005bee196	allow more than one outbound context per endpoint for fallback in case one outbound context takes a shit	3 years ago
Jeff Becker	83e70d95e0	dont invalidate cache on lns timeout, only on explicit negative	3 years ago
Jeff Becker	d563e3b340	if a path's latency is zero dont use it because it's not actually a zero latency path it's probably about to be failed or timed out ... increase default path alignment timeout	3 years ago
Jeff Becker	397d8b01fc	try fixing std::shared_ptr leak with paths	3 years ago
Jeff	d514f30eaa	Merge pull request #1620 from majestrate/default-log-level-warn-2021-05-01 ... make default log level warn	3 years ago
Jeff Becker	ed707eecf9	change granularity to seconds	3 years ago
Jeff Becker	2968caf7af	make default log level warn fixes #1593	3 years ago
Jeff Becker	54f9e1b44e	make path alignment timeout configuable ... adds [network] section parameter called path-alignment-timeout that allows configring the timeout for optional name lookup + introset lookup + aligned path build, used by tun endpoint dns, provided as milliseconds.	3 years ago
Jeff Becker	f9fe6f7e3b	consmetic fixes ... * print names in lookup jobs instead of yyyyyy * update syntax in construction of objects	3 years ago
Jeff Becker	f2e8b5547d	compat for lokinet 0.8.x ... * don't send messages back that aren't expected	3 years ago
Jeff Becker	e0185bab09	don't rehash decaying hashsets that is no bueno and probably leaks like the pipes in tom's apartment	3 years ago
Jeff Becker	c6320724d3	parameterize traffic alignement timeout in dns lookups	3 years ago
Jeff Becker	f69ccb73a8	limit path reanimation ... * wait for a limited time for dead paths to reanimate and then remove them after that forever	3 years ago
Jeff Becker	b31a484bc9	fix up outbound session ... * prevent introset lookup spam * change return values to void becuase bool doesn't mean fucking shit at all	3 years ago
Jeff Becker	6b115913bc	lokinetmon updates ... * add country flags to lokinetmon * expose hop ip addresses via rpc introspection for geoip in lokinetmon	3 years ago
Thomas Winget	cecbddc912	Fixes subtle memory leak, adds comments ... Fixes a subtle memory leak that was a result of outbound messages which were in the shared queue (not yet sorted into a per-path queue) when a path was removed, resulting in a ghost path queue (and thus round-robin order entry as well). Adds much needed documentation to the outbound message handler class.	3 years ago
Jeff	ef28de8c9f	Merge pull request #1610 from majestrate/android-fixes-2021-04-26 ... Android fixes	3 years ago
Jason Rhinelander	7963cd0181	Fix headers & grammar for non-systemd	3 years ago
Jason Rhinelander	1d48cd6d35	format	3 years ago
Jason Rhinelander	b9363c8d21	Put systemd ifdef around the headers, too	3 years ago
Jason Rhinelander	4ef25ef679	Add systemd-resolved dynamic DNS updating ... Wires up systemd support to configure DNS on startup and when enabling/disabling exit mode. On startup (and when turning off an exit) we tell systemd-resolved to direct .loki and .snode lookups to lokinet (leaving other DNS traffic alone). On exit enabling, we reconfigure it to resolve "." (i.e. the root DNS domain) so that all lookups come into it.	3 years ago
Jason Rhinelander	35e4e8817b	Add missing getIPv6 implementation	3 years ago
Jeff Becker	8a74b55af3	limit calls to service node list updates for when we are synching the chain we dont spam with list updates	3 years ago
Jeff Becker	d750f68328	prepare for ipv6 on android ... * remove 21/8 from ipv4 bogon ranges as it is being sold by DoD * start adding ipv6 bogon ranges	3 years ago
Jeff Becker	3e7137ad96	fix regression: llarp::Context::CallSafe was not deferring call on startup	3 years ago
Jeff Becker	6d12a7a712	fixes for android jni	3 years ago
Jason Rhinelander	86dcdde8d3	Add missing <optional> header ... Reported by TechnicalTumbleweed as needed to fix a build.	3 years ago
Jason Rhinelander	b2b0e904d7	😡	3 years ago
Jason Rhinelander	41e5cf09ed	ngtcp2 compat update	3 years ago
Jeff	4273fb0208	Merge pull request #1602 from majestrate/rpc-block-notify-fix-2021-04-20 ... add_command instead of add_request_command	3 years ago
Jeff Becker	7f93b95080	add_command instead of add_request_command	3 years ago
Jeff Becker	a3d7590fc8	correct header	3 years ago
Jeff Becker	714f6b0183	macos variant fixes	3 years ago
Jason Rhinelander	869b362572	Attempt fix to work around gcc-8 warning	3 years ago
Jeff Becker	76eb2277f8	deconstexpr ip range bogon check	3 years ago
Jeff Becker	71391957c5	set up accept any plainquic handlers	3 years ago
Jeff Becker	83b0f25075	cleanup after rebase	3 years ago
Jeff Becker	f63122272d	fix some compiler errors	3 years ago
Jeff Becker	b721b8d65b	finish wiring up config for #1595	3 years ago
Jeff Becker	0be62f0ad4	win32 fix ... * windows does not have uint	3 years ago
Jeff Becker	fedc56e3f1	initial commit for #1595	3 years ago
Jeff Becker	545021aa3d	temp commit	3 years ago
Jeff Becker	bb7b46c43d	fixes for testnet	3 years ago
Jeff Becker	118bc9c2fb	formatting	3 years ago
Jeff Becker	230fb15b24	ensure dns srv data is valid after decoding	3 years ago
Jeff Becker	a7a886e670	fix infinite iteration	3 years ago
Jeff Becker	07eaeb681a	try fixing deadlock	3 years ago
Jeff Becker	4889b8cddf	fix up client to snode codepath	3 years ago
Jeff Becker	8b2ede5fc5	use rxid	3 years ago
Jeff Becker	1ea77e6a68	more client to snode quic fixes	3 years ago
Jeff Becker	0ee40c1720	use rxid	3 years ago
Jeff Becker	e1f90e5e5a	use rx id on snode side of paths ... set protocol on reply routing messages	3 years ago
Jeff Becker	e058b873b6	Revert "tweak timeout parameters for link layer." ... This reverts commit eef192b37aec070d624e84cedfc94f522fd0a346.	3 years ago
Jeff Becker	1677ed40f6	dont persist sessions if we failed to build	3 years ago
Jeff Becker	657c39a91c	tweak timeout parameters for link layer. ... have different timeouts for service node vs client	3 years ago
Jeff Becker	6b2c311800	cover special case for 1 hop path builders	3 years ago
Jeff Becker	2e8f47a7fa	various cleanups	3 years ago
Jeff Becker	b36f87a4d2	stop crashing when we send dns reply for snode on service nodes	3 years ago
Jeff Becker	b01e20b4cb	add forgotten file again	3 years ago
Jeff Becker	2fa24b5eae	add untracked file	3 years ago
Jeff Becker	5eda4addc2	correctly discard invalid data on transfer traffic message	3 years ago
Jeff Becker	25e338d621	srv records ... * add srv records in RCs if we have any * add mechanism to add SRV records for plainquic exposed ports * resign and republish rc or introset on srv record changes	3 years ago
Jeff Becker	a2285730b8	reduce log level	3 years ago
Jeff Becker	de1b0c9d14	add forgotten header	3 years ago
Jeff Becker	94ecf02a62	add option to run lokinet traffic only with no exit and drop all non lokinet traffic ... triggered by: exit-node=null	3 years ago
Jeff Becker	2ed0ab1ca1	liblokinet and lnproxy updates ... * add lokinet_status function to get the current status of the context, aka are we ready to send or not. * catch some exceptions in lnproxy	3 years ago
Jeff Becker	e005b34169	fix up some retarded bullshit ... * llarp::service::NameIsValid was not checking that the tld was .loki, add this check. * make link layer initial connection timeout 5s not the session activity timeout which happens to be 60 god damn seconds.	3 years ago
Jeff Becker	0046de3e7a	service node logic change: simplify transit hop behavior to send protocol frames directly from a path transfer message	3 years ago
Jeff Becker	3f9e51e391	more liblokinet api functions ... * get/set netid * set log level	3 years ago
Jeff Becker	cf7603f20e	temp commit, closeReset -> close	3 years ago
Jeff Becker	48ce77ed1f	XXX: single threaded mode	3 years ago
Jeff Becker	bac9341ae0	fix: if we have not used a session yet we only check the intro for expiration	3 years ago
Jeff Becker	2188b443e9	more liblokinet api stuff ... * add lokinet_wait_for_ready to wait until we have published introsets and such * optionally expose ports in lnproxy * dont save profiles by default	3 years ago
Jeff Becker	95cd275cdd	liblokinet additions: ... * add lokinet_add_bootstrap_rc function for adding an rc from memory * prevent stack overflow on error closing connection in quic * add in memory nodedb * refactor how convotags are set as active * add initial stubs for endpoint statistics * refactor time stuff to be a bit cleaner * update lnproxy script with more arguments	3 years ago
Jason Rhinelander	354f0f63e5	Reduce max number of per-connection active streams to 32 ... This is the max number of simultaneous open connections to the same port on the remote. 100 was fairly arbitrary and seems a bit high.	3 years ago
Jason Rhinelander	5e912600f8	Fix connection close handling ... Replace stream_reset (which typically isn't called) with a stream_close handler (which is already called whether or not it was a reset). Most importantly, the server side needs to extend the max bidi streams counter during stream_close (otherwise we run out when we hit the limit and new connections just stall).	3 years ago
Jeff Becker	8bc60a59ac	fix up liblokinet api to be more friendly to ffi ... add lnproxy exmaple of use of liblokinet	3 years ago
Jeff Becker	853cc52efb	wire up liblokinet	3 years ago
Jason Rhinelander	2a809c7a30	Add trace debugging of inbound packet type	3 years ago
Jason Rhinelander	35ead9dc6b	Remove superfluous block (FIXME: squash this with earlier "FIXME")	3 years ago
Jason Rhinelander	a2045cd966	Fix debug message	3 years ago
Jason Rhinelander	80d3738531	Increase null_iv to 8 bytes to avoid ngtcp2 stack corruption	3 years ago
Jeff Becker	309a899c9f	mark loopback convotags as forever ... make null handler mark discard data as activity for convotag	3 years ago
Jeff Becker	f310778f81	dont regenerate convotag on loopback when we have one already	3 years ago
Jeff Becker	abf8d2fb17	fix loopback convotag	3 years ago
Jeff Becker	5b05d22bad	refactors ... * add path sequence numbers on routing messages * reduce log level in debug mode * wire up loopback style sending to ourself	3 years ago
Jeff Becker	f86a2daf83	fixes ... * Add service::Endpoint::HasOutboundConvo * dont mark outbound convos as inbound * order quic packets	3 years ago
Jeff Becker	59c9e997f2	build paths faster and limit path builds at edge router instead of via a time counter for all edges	3 years ago
Jason Rhinelander	ff2e79ce38	Remove superfluous log statement ... The send_packet() we call here already prints the buffer.	3 years ago
Jason Rhinelander	312222e881	Properly handle ngtcp2 returning UINT64_MAX	3 years ago
Jason Rhinelander	2ea84a61f9	Replace timer->again() with stop/start() ... repeat()/again() sets a repeating timer, which we don't want; stop and start it instead so that it fires just once.	3 years ago
Jason Rhinelander	68e96a4e5d	Remove obsolete fixme ... Remote addr and ecn are handled.	3 years ago
Jason Rhinelander	7982581cfd	Revisit/reduce quic logging ... Demote many things to Trace.	3 years ago
Jason Rhinelander	3c630b260a	Don't install stream forwarding here ... Stream forward on the client-side TCP connection gets set up within in initial_client_data_handler (which also handles reading the initial stream version byte).	3 years ago
Jason Rhinelander	f534103971	Simplify/fix expiry timer logic ... Also fix a potential transmission delay because `again()` wasn't being called when the expiry is already passed (i.e. meaning we should run immediately).	3 years ago
Jason Rhinelander	ac34835c12	Fix/refactor stream closing ... Make stream closing with expiring connections work better. Fixes an issue where the stream's uv_async could outlive the stream and/or connection and segfault.	3 years ago
Jason Rhinelander	60c813d306	Schedule retransmit after blocked send ... In the standalone plainquic code we triggered a retransmit when the socket became writeable again, but that doesn't work here, so just schedule it right away to let ngtcp2 worry about retrying.	3 years ago
Jason Rhinelander	71beae5923	Stop retransmit timer when destroying a connection ... Otherwise we segfault because it outlives the Connection.	3 years ago
Jason Rhinelander	027243cbcc	Remove obsolete fixme	3 years ago
Jason Rhinelander	233cb86191	Sever tunnel links earlier ... When we get an error on the tcp connection immediately sever the link to the quic tunnel so that it doesn't keep trying to forward data to it.	3 years ago
Jeff Becker	73572b317b	stuff	3 years ago
Jason Rhinelander	4cea33a139	Fix Server's missing opt-out of ring buffer ... For now we still steal buffers from uvw. In the future I'd like to change that, but it's still uvw work-in-progress to support custom data allocators, and so for now we still steal data buffers from uvw.	3 years ago
Jason Rhinelander	99954f7501	Add some more useful debug logging	3 years ago
Jason Rhinelander	b8be889291	Add missing client TCP accept/forwarding handlers ... Somehow the TCP client connection accept and stream forwarding got dropped in the quic refactor.	3 years ago
Jason Rhinelander	738f16366b	Try outbound convos if no inbound convos work (FIXME) ... FIXME - eliminate this block of code that used to be an `else` before PR'ing.	3 years ago
Jason Rhinelander	183abd58aa	Add more checks and logging ... Most of the logging here is Trace level so needs a Debug build to not get compiled away.	3 years ago
Jason Rhinelander	44fc941c32	Set port properly in server-to-client reply packets ... ngtcp2 was rejecting them because we have the port when constructing, but then it was 0 on the return packet (which ngtcp2 drops because it's coming from an unknown/invalid path).	3 years ago
Jeff Becker	6bb23aa956	patch	3 years ago
Jeff Becker	3977c12da6	dont try to build untracked file	3 years ago
Jeff Becker	708e408c30	various fixups and cleanups ... * wire up last of the quic stuff * clean up udp packet generation code * pass EndpointBase not quic tunnel for quic stuff * add {n,h}uint16_t::FromString * add nuint_t::FromString * make AlignedBuffer::IsZero non constant time call for speed	3 years ago
Jeff Becker	100a953a23	make the lokinet endpoint code work ... still needs the quic sides to work	3 years ago
Jeff Becker	7098f46d60	rpc endpoint for triggering quic tunnel	3 years ago
Jeff Becker	838b968d8f	re add static build for all internal lokinet libs	3 years ago
Jeff Becker	c88406fb27	move lokinet-quic into the big giant amalgum of liblokinet	3 years ago
Jeff Becker	4dafe973c2	comment out member	3 years ago
Jeff Becker	883005d7db	variable is unused for now but i dont want to remove it yet	3 years ago
Jeff Becker	2360700911	add virtual default destructor to appease clang	3 years ago
Jeff Becker	8b95f9945a	fix compiler errors	3 years ago
Jeff Becker	f5700c560e	try inbound paths first	3 years ago
Jeff Becker	59891d5d5f	wire up snode traffic to quic	3 years ago
Jeff Becker	58bec09232	fix compile error	3 years ago
Jason Rhinelander	752879d712	QUIC lokinet integration refactor ... Refactors how quic packets get handled: the actual tunnels now live in tunnel.hpp's TunnelManager which holds and manages all the quic<->tcp tunnelling. service::Endpoint now holds a TunnelManager rather than a quic::Server. We only need one quic server, but we need a separate quic client instance per outgoing quic tunnel, and TunnelManager handles all that glue now. Adds QUIC packet handling to get to the right tunnel code. This required multiplexing incoming quic packets, as follows: Adds a very small quic tunnel packet header of 4 bytes: [1, SPORT, ECN] for client->server packets, where SPORT is our source "port" (really: just a uint16_t unique quic instance identifier) or [2, DPORT, ECN] for server->client packets where the DPORT is the SPORT from above. (This also reworks ECN bits to get properly carried over lokinet.) We don't need a destination/source port for the server-side because there is only ever one quic server (and we know we're going to it when the first byte of the header is 1). Removes the config option for quic exposing ports; a full lokinet will simply accept anything incoming on quic and tunnel it to the requested port on the the local endpoint IP (this handler will come in a following commit). Replace ConvoTags with full addresses: we need to carry the port, as well, which the ConvoTag can't give us, so change those to more general SockAddrs from which we can extract both the ConvoTag *and* the port. Add a pending connection queue along with new quic-side handlers to call when a stream becomes available (TunnelManager uses this to wire up pending incoming conns with quic streams as streams open up). Completely get rid of tunnel_server/tunnel_client.cpp code; it is now moved to tunnel.hpp. Add listen()/forget() methods in TunnelManager for setting up quic listening sockets (for liblokinet usage). Add open()/close() methods in TunnelManager for spinning up new quic clients for outgoing quic connections.	3 years ago
Jason Rhinelander	158ea4a951	Add isIPv4/isIPv6 to SockAddr	3 years ago
Jason Rhinelander	2ca7ef7f5f	Rename isRouter -> isSNode ... The isRouter wording was quite confusing, especially in a call such as: router->Configure(config, opts.isRouter, nodedb)	3 years ago
Jason Rhinelander	d588f777c7	Fix logging of std::byte values ... They were failing to compile if output because implicit integer promotion doesn't work for std::byte's (but rather needs an explicit std::to_integer call).	3 years ago
Jason Rhinelander	e27fdf4c9a	Simplify/unify SendTo* methods ... Overload them to the same name to make it easier to call and/or visit them with either a RouterID, Address, Variant, or ConvoTag.	3 years ago
Jason Rhinelander	104f63543f	Change MaybeGetLoop to a virtual member func ... Minor simplification.	3 years ago
Jeff Becker	a1a1b2e2bc	remove superflous include	3 years ago
Jeff Becker	a61f846d33	* add convotags to ensure path to snode ... * add service::Endpoint::EnsurePathTo which gives you a std::optional<Convotag>	3 years ago
Jeff Becker	fe32475cad	use more variants	3 years ago
Jeff Becker	9d483a12db	* add protocol type to snode traffic ... * make path::PathSet::SendPacketToRemote know about protocol type	3 years ago
Jeff Becker	e4ca7ce4de	get rid of uneeded type	3 years ago
Jeff Becker	618881e65c	update ngtcp2 address/path structs	3 years ago
Jeff Becker	fc9b09bdbc	clean up address/ip management code to use std::variant and std::optional	3 years ago
Jeff Becker	1885b1cae9	more	3 years ago
Jeff Becker	4446f2fc16	fix and format. ... * start moving quic code to use lokinet internals	3 years ago
Jason Rhinelander	39d31df059	no-op LogTrace(); print chars as ints ... - LogTrace() (and LogTraceTag, etc.) are now no-ops for release builds. (hoping there are no side effects in trace logging!) - renamed llarp::_Log to llarp::_log because _Log is a reserved keyword - change logging code to implicitly convert 1-byte types (char, unsigned char, uint8_t) to ints so that we print them as numeric values rather than raw chars because, more often than not, printing a single char is trying to log an 8-bit value.	3 years ago
Jason Rhinelander	b61bd82b4b	Add llarp::buffer_printer for pretty buffer logs	3 years ago
Jason Rhinelander	818b4042e9	Migrate plainquic logging to llarp logging	3 years ago
Jason Rhinelander	5d897781ef	Replace custom random with llarp/sodium random	3 years ago
Jason Rhinelander	aa0f54fa07	WIP plainquic tunnels	3 years ago
Jason Rhinelander	46c619c49f	Move external/ cmake bits into external/CMakeLists.txt ... - Move uvw target to external instead of llarp/CMakeLists.txt - Add sqlite_orm headers via interface library rather than shoving it into the global include search path. - Find sqlite3 as part of the sqlite_orm target deps rather than linking it separately. (Also don't link it into lokinet-util, which doesn't appear to actually need it).	3 years ago
Jeff Becker	b45a268dbc	fix windows dll link-ism	3 years ago
Jeff Becker	2fa7a54e86	fix pybind layer compile errors from missing llarp.h header	3 years ago
Jeff Becker	e4841917ba	redo cmake parts ... * add liblokinet.so target * make every library we build except liblokinet.so static * wire up parts of liblokinet	3 years ago
Jeff Becker	a6715a1f75	remove old vpnio bits	3 years ago
Jeff Becker	c92894804c	Merge remote-tracking branch 'origin/stable' into dev	3 years ago
Jeff Becker	a0d3ab125a	throw on failed writing and port fs namespace changes from oxen-core	3 years ago
Jeff Becker	e8178a032e	lokinet-bootstrap native binary	3 years ago
Jeff Becker	0c37cc7f60	lokinet-bootstrap native binary	3 years ago
Jeff Becker	7a1ffc2df4	* lessen replay filter window ... * dont have transit hops use a replay filter * formatting	3 years ago
Jeff Becker	9c742b36eb	call delivery status hook on overload	3 years ago
Jeff	6bd53484da	Merge pull request #1541 from majestrate/lns-consensus-2021-02-19 ... lns lookup consensus	3 years ago
Jeff	cb2254ba46	Merge pull request #1573 from jagerman/more-code-refactors ... More code refactors	3 years ago
Jeff Becker	dfa84081c9	forgot a type	3 years ago
Jeff Becker	202c9348fa	use std::variant	3 years ago
Jeff Becker	b7f72601ce	move stuff to var::variant	3 years ago
Jeff Becker	ba9dca6c5d	use var::variant and var::visit shims because apple is sux	3 years ago
Jeff Becker	f10269b03e	remove std::visit because apple is functionally retarded and does not support it despite claiming to support c++14	3 years ago
Jeff Becker	c07358f084	allow failure of setting ipv6 address when it is not allowed by kernel on linux	3 years ago
Jeff Becker	6c1d18412c	seperate logic into IsDone()	3 years ago
Jeff Becker	88ddcceca4	make lookup name async return optional variant	3 years ago
Jeff Becker	4e14567d67	make Endpoint::LookupNameAsync return void	3 years ago
Jeff Becker	8e57572b37	dont reference iterator in lambda.	3 years ago
Jeff Becker	95eba18e40	cache result on success	3 years ago
Jeff Becker	7a11f3b1e3	redunant lookups for lns. ... * request lns from all endpoints we have on our pathset * make sure all snodes agree on the lns name being looked up	3 years ago
Jeff	ec242447a0	Merge pull request #1581 from majestrate/ipv6-configurable-range-2021-03-18 ... make ipv6 range used by exit traffic configurable.	3 years ago
Jeff Becker	cf85ea6f2b	add warning and note in documenation of config option	3 years ago
Jeff Becker	1acea95ced	fix issue #1562 ... * use csrng for std::shuffle	3 years ago
Jeff Becker	b2687a53ee	dont use .value() because macos is absolutely broken, thanks apple.	3 years ago
Jeff Becker	a83428297e	make ipv6 range used by exit traffic configurable. ... * add ip6-range option to network section to control which range we want to use for ipv6 traffic * make ip6-range able to disable ipv6 exits ( i dont like this but eh )	3 years ago
Jeff Becker	ad6dcd683b	fix segfault	3 years ago
Jeff Becker	80a76c9386	fix segfault	3 years ago
Jason Rhinelander	181953b4a6	Replace ::Hash nested structs with std::hash specializations	3 years ago
Jason Rhinelander	d1dadb530c	SockAddr endian cleanups & add uint128 ctors ... - Make SockAddr endian arguments explicit - Consolidate port-less contructors and port constructors into one with a default port of 0. - Add {h,n}uint128_t ctors for construction from IPv6 addrs	3 years ago
Jason Rhinelander	fe0d099e86	Code cleanups ... - De-lambda a lambda that is only called once, immediately after it is defined. - C++17 modernize key-value for loops - Remove unused (and useless) method	3 years ago
Jason Rhinelander	0339bd941a	Convert protocol type to enum class	3 years ago
Jeff	9bf002cc8d	Merge pull request #1574 from tewinget/header_fixup ... standardize include format and pragma once	3 years ago
Jason Rhinelander	f47e0cf7a2	Fix UB -- use after move	3 years ago
Thomas Winget	7caa87862e	standardize include format and pragma once ... All #ifndef guards on headers have been removed, I think, in favor of #pragma once Headers are now included as `#include "filename"` if the included file resides in the same directory as the file including it, or any subdirectory therein. Otherwise they are included as `#include <project/top/dir/relative/path/filename>` The above does not include system/os headers.	3 years ago
Thomas Winget	35fb54d131	add project source dir to util include directories	3 years ago
Jeff Becker	324498fe9a	flip order of parameters so it works	3 years ago
Jeff Becker	3d09b94590	make android work again ... ip packet on read error was returning an uninitialized ip packet with non zero size so it was thinking it was a valid packet. value initialize it to set the packet size to zero by default. this fixed it in an emulator at least...	3 years ago
Jeff Becker	6550aa844c	try to unbreak more	3 years ago
Jeff Becker	7af4807302	correctly set size of OwnedBuffer when using OwnedBuffer::copy_used	3 years ago
Jason Rhinelander	91d6698a9d	Reformat with clang-format-11	3 years ago
Jason Rhinelander	c00de1e4ec	Add wakeup call into transit_hop UpstreamHandler	3 years ago
Jason Rhinelander	89b582bf60	Modernize [[maybe_unused]] attribute	3 years ago
Jason Rhinelander	4b5387a078	Remove unused ShouldFlushNow method	3 years ago
Jason Rhinelander	f9dc308f75	Tick event loop on data queuing	3 years ago
Jason Rhinelander	895acc45ff	EventLoop: add public wakeup() method, and call it from call() ... call(), when invoked from the logic thread, wasn't triggering a wakeup which stalled some traffic (such as client-to-snode packets). Fix it by triggering a wakeup on `call()`, and expose it because this is a useful thing to be able to do.	3 years ago
Jason Rhinelander	f8d5e106a0	Uncurse SockAddr's sockaddr* operator ... The `const sockaddr*` conversion was returning a pointer to the sockaddr_in6, but with its family set to AF_INET, so the receiving code would have treated interpreting the result as a sockaddr_in pointer -- but it isn't, and trying to get the IPv4 from the mis-reinterpreted sockaddr_in6 fails; this then prevented uvw UDP binding (which uses the sockaddr* operator) from working. This changes it to always leave the family value at the proper value and using the ipv6_is_mapped_ipv4 check to determine when we have (and should return) a pointer to the sockaddr_in rather than the sockaddr_in6. Also fixes toString() to compare the full prefix instead of just bytes 10 and 11.	3 years ago
Jason Rhinelander	62fc1bbc48	Make llarp_buffer_t constructible from an rvalue reference	3 years ago
Jason Rhinelander	f8486604c4	Export UDP file descriptor (for android)	3 years ago
Jason Rhinelander	d995766436	Android & hive fixes	3 years ago
Jason Rhinelander	673d065ed8	Remove HasNextPacket() (and fix win32) ... We don't use it anymore (instead we just ReadNextPacket until it returns an empty result).	3 years ago
Jason Rhinelander	6e05dc693f	Remove unused arguments (and ants)	3 years ago
Jason Rhinelander	2c827870c9	Simplify code of many f's ... We don't need to go through the contention killer here, that was mainly for (long ago) debugging reasons.	3 years ago
Jason Rhinelander	36edabd9b0	Rename EventLoop::udp to EventLoop::make_udp ... Makes it consistent with the make_waker, etc. Also removes some debugging.	3 years ago
Jason Rhinelander	afd856bca0	Remove unused junk (also fix win32 compilation fail)	3 years ago
Jason Rhinelander	096f37fe4e	Fix linux compilation & narrowing conversion warning	3 years ago
Jason Rhinelander	e198bfd3e1	Remove unused variable	3 years ago
Jason Rhinelander	c4559d158e	Make format	3 years ago
Jason Rhinelander	ccc7b5c9e9	Merge Logic functions into EventLoop ... loop->call(...) is similar to the old logic->Call(...), but is smart about the current thread: if called from within the event loop it simply runs the argument directly, otherwise it queues it. Similarly most of the other event loop calls are also now thread-aware: for example, `call_later(...)` can queue the job directly when called if in the event loop rather than having to double-queue through the even loop (once to call, then inside the call to initiate the time).	3 years ago
Jason Rhinelander	5b555ee5aa	Replace libuv with uvw & related refactoring ... - removes all the llarp_ev_* functions, replacing with methods/classes/functions in the llarp namespace. - banish ev/ev.h to the void - Passes various things by const lvalue ref, especially shared_ptr's that don't need to be copied (to avoid an atomic refcount increment/decrement). - Add a llarp::UDPHandle abstract class for UDP handling - Removes the UDP tick handler; code that needs tick can just do a separate handler on the event loop outside the UDP socket. - Adds an "OwnedBuffer" which owns its own memory but is implicitly convertible to a llarp_buffer_t. This is mostly needed to take over ownership of buffers from uvw without copying them as, currently, uvw does its own allocation (pending some open upstream issues/PRs). - Logic: - add `make_caller`/`call_forever`/`call_every` utility functions to abstract Call wrapping and dependent timed tasks. - Add inLogicThread() so that code can tell its inside the logic thread (typically for debugging assertions). - get rid of janky integer returns and dealing with cancellations on call_later: the other methods added here and the event loop code remove the need for them. - Event loop: - redo everything with uvw instead of libuv - rename EventLoopWakeup::Wakeup to EventLoopWakeup::Trigger to better reflect what it does. - add EventLoopRepeater for repeated events, and replace the code that reschedules itself every time it is called with a repeater. - Split up `EventLoop::run()` into a non-virtual base method and abstract `run_loop()` methods; the base method does a couple extra setup/teardown things that don't need to be in the derived class. - udp_listen is replaced with ev->udp(...) which returns a new UDPHandle object rather that needing gross C-style-but-not-actually-C-compatible structs. - Remove unused register_poll_fd_(un)readable - Use shared_ptr for EventLoopWakeup rather than returning a raw pointer; uvw lets us not have to worry about having the event loop class maintain ownership of it. - Add factory EventLoop::create() function to create a default (uvw-based) event loop (previously this was one of the llarp_ev_blahblah unnamespaced functions). - ev_libuv: this is mostly rewritten; all of the glue code/structs, in particular, are gone as they are no longer needed with uvw. - DNS: - Rename DnsHandler to DnsInterceptor to better describe what it does (this is the code that intercepts all DNS to the tun IP range for Android). - endpoint: - remove unused "isolated network" code - remove distinct (but actually always the same) variables for router/endpoint logic objects - llarp_buffer_t - make constructors type-safe against being called with points to non-size-1 values - tun packet reading: - read all available packets off the device/file descriptor; previously we were reading one packet at a time then returning to the event loop to poll again. - ReadNextPacket() now returns a 0-size packet if the read would block (so that we can implement the previous point). - ReadNextPacket() now throws on I/O error - Miscellaneous code cleanups/simplifications	3 years ago
Jason Rhinelander	c71d3527bd	C++17 simplification	3 years ago
Jason Rhinelander	8b00eacabf	Fix signed/unsigned char comparison bug ... These functions would not working properly if given a high bit char value.	3 years ago
Jeff	abf057d6e3	Merge pull request #1536 from majestrate/memleak-fix-2021-02-15 ... plug super slow memory leak	3 years ago
Jeff Becker	ec0f45eeb7	plug memory leak in outbound message queue. ... sometimes we use rxid in our pathid so we need to clear those queues too. if we don't it'll leak from never clearing those queues.	3 years ago
Jeff Becker	7dbd25f271	don't leave old multi ack codepath in	3 years ago
Jeff Becker	c9ff917e0d	revert priority queue ordering change	3 years ago
Jeff Becker	ad3c23ba2b	simplify call	3 years ago
Jeff Becker	5cffc3b0f8	consolidate rx message handling in iwp ... * add toggle flag for using multi ack * check replay filter before processing message	3 years ago
Jeff Becker	e6ac7e721d	use a std::map instead of a std::unordered_map for holding message buffers by their sequentially increasing message id. ... when retransmissions happen, they will be executed in a deterministic order by their delivery id. this reduces the jitter from delayed message delivery on link layer as with an unordered map the order of iteration is "random", so when we iterate all messages for retransmission we do it lowest id first (the messages queued first).	3 years ago
Jeff Becker	74f707ee01	only do single ack packets	3 years ago
Jeff	4992629f20	Merge pull request #1563 from majestrate/dont-multiplex-from-outbound-context-2021-03-03 ... use lowest latency path for send context	3 years ago
Jeff Becker	7060a60c84	enable route poker poking before going "up"	3 years ago
Jeff Becker	f77cbfb48e	fixes for ipv4 traffic: dont use new range because that breaks backwards comapt	3 years ago
Jeff Becker	f67668f0a6	use lowest latency path for send context	3 years ago
Jeff	f2b234d6c3	Merge pull request #1539 from majestrate/path-algorithm-flavors-2021-02-18 ... add option to enforce unique netblocks per path.	3 years ago
Jeff	b4b8b84e57	Merge pull request #1548 from majestrate/lns-mx-records-2021-02-25 ... add mx records to lns names	3 years ago
Jeff Becker	d6567614e3	use mutable lambda instead of shared_ptr for handling mx dns reply for lns	3 years ago