oweals/gnunet.git
7 years agofirst, very rough conversion of SET service to new cadet client API (fails tests...
Christian Grothoff [Thu, 16 Feb 2017 22:24:48 +0000 (23:24 +0100)]
first, very rough conversion of SET service to new cadet client API (fails tests, as before)

7 years agomigrate SET service to new SERVICE API
Christian Grothoff [Thu, 16 Feb 2017 21:59:53 +0000 (22:59 +0100)]
migrate SET service to new SERVICE API

7 years agothis actually happens and is fine
Christian Grothoff [Thu, 16 Feb 2017 21:14:43 +0000 (22:14 +0100)]
this actually happens and is fine

7 years agointroduce loose state, fix channel state machine
Christian Grothoff [Thu, 16 Feb 2017 21:08:55 +0000 (22:08 +0100)]
introduce loose state, fix channel state machine

7 years agoproperly clean up unbound channels
Christian Grothoff [Thu, 16 Feb 2017 20:55:13 +0000 (21:55 +0100)]
properly clean up unbound channels

7 years agofix log
Christian Grothoff [Thu, 16 Feb 2017 20:26:40 +0000 (21:26 +0100)]
fix log

7 years agodo not respond to BROKEN messages with BROKEN messages
Christian Grothoff [Thu, 16 Feb 2017 20:00:07 +0000 (21:00 +0100)]
do not respond to BROKEN messages with BROKEN messages

7 years agofix log message format str
Christian Grothoff [Thu, 16 Feb 2017 14:19:13 +0000 (15:19 +0100)]
fix log message format str

7 years agoset type correctly in nat auto test
Christian Grothoff [Thu, 16 Feb 2017 14:08:42 +0000 (15:08 +0100)]
set type correctly in nat auto test

7 years agocancel existing last_control_qe handle before scheduling a new one, always
Christian Grothoff [Thu, 16 Feb 2017 13:54:14 +0000 (14:54 +0100)]
cancel existing last_control_qe handle before scheduling a new one, always

7 years agoextra asserts
Christian Grothoff [Thu, 16 Feb 2017 13:42:35 +0000 (14:42 +0100)]
extra asserts

7 years agouse force when required
Christian Grothoff [Thu, 16 Feb 2017 13:05:55 +0000 (14:05 +0100)]
use force when required

7 years agofix #4895: set highbit always for the returned cn, not just on the second call
Christian Grothoff [Wed, 15 Feb 2017 16:49:59 +0000 (17:49 +0100)]
fix #4895: set highbit always for the returned cn, not just on the second call

7 years agospell out message types more, use correct conversion direction
Christian Grothoff [Wed, 15 Feb 2017 16:37:17 +0000 (17:37 +0100)]
spell out message types more, use correct conversion direction

7 years agofix typos in log msgs
Christian Grothoff [Wed, 15 Feb 2017 16:24:43 +0000 (17:24 +0100)]
fix typos in log msgs

7 years agofixing #4878 -- by avoiding allocation in the first place
Christian Grothoff [Wed, 15 Feb 2017 15:56:14 +0000 (16:56 +0100)]
fixing #4878 -- by avoiding allocation in the first place

7 years agoupdate ignore file
Christian Grothoff [Wed, 15 Feb 2017 15:55:47 +0000 (16:55 +0100)]
update ignore file

7 years agointroducing GNUNET_CRYPTO_ecdhe_create2() to avoid malloc nonsense
Christian Grothoff [Wed, 15 Feb 2017 15:55:25 +0000 (16:55 +0100)]
introducing GNUNET_CRYPTO_ecdhe_create2() to avoid malloc nonsense

7 years agofix format string bug
Christian Grothoff [Wed, 15 Feb 2017 15:54:04 +0000 (16:54 +0100)]
fix format string bug

7 years agofix #4890
Christian Grothoff [Wed, 15 Feb 2017 14:37:07 +0000 (15:37 +0100)]
fix #4890

7 years agopass correct pointer to GNUNET_NAT_register
Christian Grothoff [Mon, 13 Feb 2017 09:29:11 +0000 (10:29 +0100)]
pass correct pointer to GNUNET_NAT_register

7 years agoadd assertion that we do not release path that is still used by a connection
Christian Grothoff [Mon, 13 Feb 2017 09:24:57 +0000 (10:24 +0100)]
add assertion that we do not release path that is still used by a connection

7 years agoset up multicast group with origin and member
xrs [Sun, 12 Feb 2017 00:44:00 +0000 (01:44 +0100)]
set up multicast group with origin and member

7 years agoNo need to mention HOLE_EXTERNAL twice
David Barksdale [Sat, 11 Feb 2017 16:58:12 +0000 (10:58 -0600)]
No need to mention HOLE_EXTERNAL twice

7 years agofix use-after-free in postgres error message
Florian Dold [Sat, 11 Feb 2017 15:41:58 +0000 (16:41 +0100)]
fix use-after-free in postgres error message

7 years agofixed crash on test_multicast_multipeer
xrs [Thu, 9 Feb 2017 21:28:29 +0000 (22:28 +0100)]
fixed crash on test_multicast_multipeer

7 years agoAdd mq test
Bart Polot [Wed, 8 Feb 2017 18:43:33 +0000 (19:43 +0100)]
Add mq test

7 years agoFix open port, copy handlers for use on incoming channels
Bart Polot [Wed, 8 Feb 2017 18:41:41 +0000 (19:41 +0100)]
Fix open port, copy handlers for use on incoming channels

7 years ago- adjust FWD/BCK to new channel numbering
Bart Polot [Wed, 8 Feb 2017 18:41:06 +0000 (19:41 +0100)]
- adjust FWD/BCK to new channel numbering

7 years ago- code/debug format
Bart Polot [Wed, 8 Feb 2017 18:40:36 +0000 (19:40 +0100)]
- code/debug format

7 years ago- debug, indentation, legibility
Bart Polot [Wed, 8 Feb 2017 16:28:19 +0000 (17:28 +0100)]
- debug, indentation, legibility

7 years agofix cadet mq api, use proper function arguments
Bart Polot [Wed, 8 Feb 2017 16:27:23 +0000 (17:27 +0100)]
fix cadet mq api, use proper function arguments

7 years ago- fix assertion condition
Bart Polot [Wed, 8 Feb 2017 16:23:09 +0000 (17:23 +0100)]
- fix assertion condition

7 years ago- doxygen, debug text
Bart Polot [Wed, 8 Feb 2017 16:22:36 +0000 (17:22 +0100)]
- doxygen, debug text

7 years agoMerge branch 'master' of git+ssh://gnunet.org/gnunet
Bart Polot [Wed, 8 Feb 2017 01:28:25 +0000 (02:28 +0100)]
Merge branch 'master' of git+ssh://gnunet.org/gnunet

7 years agofix psyc service being unwilling to shutdown
lurchi [Tue, 7 Feb 2017 22:44:20 +0000 (23:44 +0100)]
fix psyc service being unwilling to shutdown

7 years agoOnly call window_size on incrementing size, allow ACK+1 messages to be taken
Bart Polot [Thu, 2 Feb 2017 17:13:00 +0000 (18:13 +0100)]
Only call window_size on incrementing size, allow ACK+1 messages to be taken

7 years agoImplement data ack in CADET MQ API
Bart Polot [Thu, 2 Feb 2017 16:16:20 +0000 (17:16 +0100)]
Implement data ack in CADET MQ API

7 years agognunet-social CLI: porcelain status and psyc messages
psyc://loupsycedyglgamf.onion/~lynX [Thu, 2 Feb 2017 09:21:58 +0000 (09:21 +0000)]
gnunet-social CLI: porcelain status and psyc messages

7 years agodon't let further circuits disrupt 'gnunet-cadet -o'
psyc://loupsycedyglgamf.onion/~lynX [Thu, 2 Feb 2017 09:21:56 +0000 (09:21 +0000)]
don't let further circuits disrupt 'gnunet-cadet -o'

7 years agofind /bin/ip
psyc://loupsycedyglgamf.onion/~lynX [Thu, 2 Feb 2017 09:21:55 +0000 (09:21 +0000)]
find /bin/ip

7 years agofix sanity check in dht_api: we are passing 32-byte PIDs, not 64-byte hashes
Christian Grothoff [Thu, 2 Feb 2017 09:21:54 +0000 (10:21 +0100)]
fix sanity check in dht_api: we are passing 32-byte PIDs, not 64-byte hashes

7 years agohave sanity checks for DHT path construction
Christian Grothoff [Thu, 2 Feb 2017 09:17:38 +0000 (10:17 +0100)]
have sanity checks for DHT path construction

7 years agoset PATH during testcase
Christian Grothoff [Thu, 2 Feb 2017 09:16:12 +0000 (10:16 +0100)]
set PATH during testcase

7 years agoforce attachment if connection is still there
Christian Grothoff [Wed, 1 Feb 2017 20:07:52 +0000 (21:07 +0100)]
force attachment if connection is still there

7 years agopreserve invariants while calling GCP_detatch_path while reorging paths
Christian Grothoff [Wed, 1 Feb 2017 19:49:52 +0000 (20:49 +0100)]
preserve invariants while calling GCP_detatch_path while reorging paths

7 years agouse NULL as flag for evaluation of query, ensure we pass non-NULL for reply_block...
Christian Grothoff [Wed, 1 Feb 2017 15:57:36 +0000 (16:57 +0100)]
use NULL as flag for evaluation of query, ensure we pass non-NULL for reply_block to evaluate reply even if reply is zero bytes long

7 years agolog if we get a redundant KX_AUTH, do not assert
Christian Grothoff [Wed, 1 Feb 2017 15:49:06 +0000 (16:49 +0100)]
log if we get a redundant KX_AUTH, do not assert

7 years agopath heap could be null here during shutdown
Christian Grothoff [Wed, 1 Feb 2017 15:44:13 +0000 (16:44 +0100)]
path heap could be null here during shutdown

7 years agoinit worst, to avoid valgrind warning (otherwise harmless)
Christian Grothoff [Wed, 1 Feb 2017 15:34:30 +0000 (16:34 +0100)]
init worst, to avoid valgrind warning (otherwise harmless)

7 years agofix peer destruction triggers
Christian Grothoff [Wed, 1 Feb 2017 12:35:48 +0000 (13:35 +0100)]
fix peer destruction triggers

7 years agoanother channel-add destroy-cancel fix
Christian Grothoff [Wed, 1 Feb 2017 08:36:14 +0000 (09:36 +0100)]
another channel-add destroy-cancel fix

7 years agostop destroy task if connection goes up
Christian Grothoff [Tue, 31 Jan 2017 18:09:57 +0000 (19:09 +0100)]
stop destroy task if connection goes up

7 years agouse observed RTT as starting point for retransmissions
Christian Grothoff [Tue, 31 Jan 2017 16:54:48 +0000 (17:54 +0100)]
use observed RTT as starting point for retransmissions

7 years agoimplement no-buffer option at route level
Christian Grothoff [Tue, 31 Jan 2017 16:45:25 +0000 (17:45 +0100)]
implement no-buffer option at route level

7 years agofix lower_rung() logic
Christian Grothoff [Tue, 31 Jan 2017 16:41:09 +0000 (17:41 +0100)]
fix lower_rung() logic

7 years agofix rung prev injection
Christian Grothoff [Tue, 31 Jan 2017 16:04:56 +0000 (17:04 +0100)]
fix rung prev injection

7 years agofix loop detection logic
Christian Grothoff [Tue, 31 Jan 2017 16:02:24 +0000 (17:02 +0100)]
fix loop detection logic

7 years agoinitialize direction in rung DLL
Christian Grothoff [Tue, 31 Jan 2017 14:54:11 +0000 (15:54 +0100)]
initialize direction in rung DLL

7 years agouse modern time specification for expiration
Christian Grothoff [Tue, 31 Jan 2017 13:46:57 +0000 (14:46 +0100)]
use modern time specification for expiration

7 years agoadd method to record and print DHT routes to DHT command-line tools
Christian Grothoff [Tue, 31 Jan 2017 13:44:41 +0000 (14:44 +0100)]
add method to record and print DHT routes to DHT command-line tools

7 years agoupdating gitignores and POTFILES
Christian Grothoff [Tue, 31 Jan 2017 13:31:25 +0000 (14:31 +0100)]
updating gitignores and POTFILES

7 years agoFix off-by-one
Christian Grothoff [Tue, 31 Jan 2017 13:16:20 +0000 (14:16 +0100)]
Fix off-by-one

7 years agomore statistics
Christian Grothoff [Tue, 31 Jan 2017 12:59:33 +0000 (13:59 +0100)]
more statistics

7 years agoinit rung field
Christian Grothoff [Tue, 31 Jan 2017 12:56:21 +0000 (13:56 +0100)]
init rung field

7 years agodo not recompute off if we can just pass it in
Christian Grothoff [Tue, 31 Jan 2017 10:44:08 +0000 (11:44 +0100)]
do not recompute off if we can just pass it in

7 years agodo not test against unset cpath values that we skipped
Christian Grothoff [Tue, 31 Jan 2017 10:28:33 +0000 (11:28 +0100)]
do not test against unset cpath values that we skipped

7 years agocancel tunnel destruction if we get another channel during the grace period
Christian Grothoff [Tue, 31 Jan 2017 10:24:17 +0000 (11:24 +0100)]
cancel tunnel destruction if we get another channel during the grace period

7 years agoImplement incoming traffic handling on MQ
Bart Polot [Tue, 31 Jan 2017 04:06:36 +0000 (05:06 +0100)]
Implement incoming traffic handling on MQ

7 years agoReconnect to service instead of aborting on a malformed data message
Bart Polot [Tue, 31 Jan 2017 03:58:05 +0000 (04:58 +0100)]
Reconnect to service instead of aborting on a malformed data message

7 years agoRefactor reconnect code
Bart Polot [Tue, 31 Jan 2017 03:57:50 +0000 (04:57 +0100)]
Refactor reconnect code

7 years agoImplement channel cleanup in MQ API, simplify destroy channel
Bart Polot [Tue, 31 Jan 2017 03:33:36 +0000 (04:33 +0100)]
Implement channel cleanup in MQ API, simplify destroy channel

7 years agoFix free of NULL pointer
Bart Polot [Tue, 31 Jan 2017 03:22:07 +0000 (04:22 +0100)]
Fix free of NULL pointer

7 years agoImplementation of port opening and handling in MQ
Bart Polot [Tue, 31 Jan 2017 03:17:58 +0000 (04:17 +0100)]
Implementation of port opening and handling in MQ

7 years agoAdd note for clarification
Bart Polot [Tue, 31 Jan 2017 03:17:27 +0000 (04:17 +0100)]
Add note for clarification

7 years agoImplement the connect and create_channel call for mq api
Bart Polot [Tue, 31 Jan 2017 01:58:54 +0000 (02:58 +0100)]
Implement the connect and create_channel call for mq api

7 years agoCADET MQ API documentation improvements
Bart Polot [Tue, 31 Jan 2017 01:57:46 +0000 (02:57 +0100)]
CADET MQ API documentation improvements

7 years agofair, global message buffer implemented
Christian Grothoff [Mon, 30 Jan 2017 20:14:30 +0000 (21:14 +0100)]
fair, global message buffer implemented

7 years agoputting into place the data structures for a global buffer pool shared across routes
Christian Grothoff [Mon, 30 Jan 2017 16:43:00 +0000 (17:43 +0100)]
putting into place the data structures for a global buffer pool shared across routes

7 years agoallow external code to use DLL pointers of envelopes, under very particular circumstances
Christian Grothoff [Mon, 30 Jan 2017 16:42:00 +0000 (17:42 +0100)]
allow external code to use DLL pointers of envelopes, under very particular circumstances

7 years agoimprove connection evaluation, include more metrics
Christian Grothoff [Mon, 30 Jan 2017 16:19:22 +0000 (17:19 +0100)]
improve connection evaluation, include more metrics

7 years agomark as done
Christian Grothoff [Mon, 30 Jan 2017 15:03:09 +0000 (16:03 +0100)]
mark as done

7 years agotrack and return performance metrics per cadet connection
Christian Grothoff [Mon, 30 Jan 2017 15:01:30 +0000 (16:01 +0100)]
track and return performance metrics per cadet connection

7 years agohandle KX AUTH failure by retransmitting KX
Christian Grothoff [Mon, 30 Jan 2017 14:19:21 +0000 (15:19 +0100)]
handle KX AUTH failure by retransmitting KX

7 years agopass connection tunnel identifier to channel when receiving messages, so that perform...
Christian Grothoff [Mon, 30 Jan 2017 13:03:46 +0000 (14:03 +0100)]
pass connection tunnel identifier to channel when receiving messages, so that performance can be matched

7 years agohave tunnel tell channel which connection it used for transmission, so we can track...
Christian Grothoff [Mon, 30 Jan 2017 11:27:11 +0000 (12:27 +0100)]
have tunnel tell channel which connection it used for transmission, so we can track connection performance by observing ACKs on the channel-level

7 years agoSave next pointer in case mqm gets freed
David Barksdale [Mon, 30 Jan 2017 02:07:02 +0000 (20:07 -0600)]
Save next pointer in case mqm gets freed

7 years agoDon't pass NULL to destroy_route
David Barksdale [Mon, 30 Jan 2017 01:10:43 +0000 (19:10 -0600)]
Don't pass NULL to destroy_route

7 years agoadded path desirability calculations
Christian Grothoff [Sun, 29 Jan 2017 22:41:49 +0000 (23:41 +0100)]
added path desirability calculations

7 years agoadd path desirability calculations
Christian Grothoff [Sun, 29 Jan 2017 22:41:23 +0000 (23:41 +0100)]
add path desirability calculations

7 years agoremove route also from routes
Christian Grothoff [Sun, 29 Jan 2017 19:33:43 +0000 (20:33 +0100)]
remove route also from routes

7 years agofix shutdown sequence
Christian Grothoff [Sun, 29 Jan 2017 19:20:48 +0000 (20:20 +0100)]
fix shutdown sequence

7 years agofix task cleanup on CTRL-D
Christian Grothoff [Sun, 29 Jan 2017 18:44:30 +0000 (19:44 +0100)]
fix task cleanup on CTRL-D

7 years agoof course we need 'i' here, not 'off'
Christian Grothoff [Sun, 29 Jan 2017 18:39:23 +0000 (19:39 +0100)]
of course we need 'i' here, not 'off'

7 years agosanitize paths to eliminate loops before using them; fix off-by-one causing a peer...
Christian Grothoff [Sun, 29 Jan 2017 18:31:12 +0000 (19:31 +0100)]
sanitize paths to eliminate loops before using them; fix off-by-one causing a peer to be listed twice

7 years agonote on des
Christian Grothoff [Sun, 29 Jan 2017 17:59:57 +0000 (18:59 +0100)]
note on des

7 years agocollect misc. TODO items into one big list
Christian Grothoff [Sun, 29 Jan 2017 17:52:36 +0000 (18:52 +0100)]
collect misc. TODO items into one big list

7 years agointroduce buffering options on the route level
Christian Grothoff [Sun, 29 Jan 2017 17:43:35 +0000 (18:43 +0100)]
introduce buffering options on the route level

7 years agooptimize mqm_head scans by avoiding constantly scanning over definitively non-ready...
Christian Grothoff [Sun, 29 Jan 2017 16:57:58 +0000 (17:57 +0100)]
optimize mqm_head scans by avoiding constantly scanning over definitively non-ready entries

7 years agofix ready_cb notifications to ensure they exactly happen only when needed
Christian Grothoff [Sun, 29 Jan 2017 15:58:21 +0000 (16:58 +0100)]
fix ready_cb notifications to ensure they exactly happen only when needed