Christian Grothoff [Wed, 25 Jan 2017 19:37:30 +0000 (20:37 +0100)]
build new test
Christian Grothoff [Wed, 25 Jan 2017 19:06:39 +0000 (20:06 +0100)]
cleanups
Christian Grothoff [Wed, 25 Jan 2017 19:04:32 +0000 (20:04 +0100)]
handle ACKs past client down
Christian Grothoff [Wed, 25 Jan 2017 18:59:29 +0000 (19:59 +0100)]
fix insert_sorted macro
Christian Grothoff [Wed, 25 Jan 2017 18:42:06 +0000 (19:42 +0100)]
typo in comment
Christian Grothoff [Wed, 25 Jan 2017 18:36:35 +0000 (19:36 +0100)]
fix confusing indentation
Christian Grothoff [Wed, 25 Jan 2017 18:29:54 +0000 (19:29 +0100)]
towards proper DATA_ACK handling
Christian Grothoff [Wed, 25 Jan 2017 18:29:45 +0000 (19:29 +0100)]
towards proper DATA_ACK handling
Christian Grothoff [Wed, 25 Jan 2017 17:26:27 +0000 (18:26 +0100)]
handle ancient/future duplicate payload properly
Christian Grothoff [Wed, 25 Jan 2017 16:39:28 +0000 (17:39 +0100)]
fix comment, this was implemented
Christian Grothoff [Wed, 25 Jan 2017 16:38:22 +0000 (17:38 +0100)]
fix double-transmission scheduling, implement client-client loopback flow control
Christian Grothoff [Wed, 25 Jan 2017 15:11:27 +0000 (16:11 +0100)]
fix leak
Christian Grothoff [Wed, 25 Jan 2017 14:57:52 +0000 (15:57 +0100)]
remove legacy commented out code
Christian Grothoff [Wed, 25 Jan 2017 14:53:43 +0000 (15:53 +0100)]
fix leak
Christian Grothoff [Wed, 25 Jan 2017 14:50:27 +0000 (15:50 +0100)]
fix leak
Christian Grothoff [Wed, 25 Jan 2017 13:51:40 +0000 (14:51 +0100)]
handle duplicate DATA packets
Christian Grothoff [Wed, 25 Jan 2017 13:41:39 +0000 (14:41 +0100)]
use new generic insertion sort logic
Christian Grothoff [Wed, 25 Jan 2017 13:41:05 +0000 (14:41 +0100)]
add generic insertion sort logic
Christian Grothoff [Wed, 25 Jan 2017 13:41:00 +0000 (14:41 +0100)]
add generic insertion sort logic
Bart Polot [Wed, 25 Jan 2017 12:17:22 +0000 (13:17 +0100)]
New CADET MQ API draft
Christian Grothoff [Wed, 25 Jan 2017 09:16:35 +0000 (10:16 +0100)]
nicer logging
Christian Grothoff [Wed, 25 Jan 2017 08:51:54 +0000 (09:51 +0100)]
fix crash if this end closed connection and other still sends data
Christian Grothoff [Tue, 24 Jan 2017 21:06:34 +0000 (22:06 +0100)]
init mid_recv properly
Christian Grothoff [Tue, 24 Jan 2017 20:57:58 +0000 (21:57 +0100)]
use proper message type constants
Christian Grothoff [Tue, 24 Jan 2017 20:57:40 +0000 (21:57 +0100)]
fix bug from compiler inserting padding if we try to allocate data at the end of the struct instead of in a new area
Christian Grothoff [Tue, 24 Jan 2017 20:57:00 +0000 (21:57 +0100)]
nicer function names
Christian Grothoff [Tue, 24 Jan 2017 20:56:42 +0000 (21:56 +0100)]
bitch and continue if we do not have a handler for a message
Christian Grothoff [Mon, 23 Jan 2017 22:54:09 +0000 (23:54 +0100)]
simplify logic
Christian Grothoff [Tue, 24 Jan 2017 20:00:23 +0000 (21:00 +0100)]
fix client-client loopback flow control
Christian Grothoff [Mon, 23 Jan 2017 22:50:55 +0000 (23:50 +0100)]
better logging, prep for having LOCAL_ACKs
Christian Grothoff [Mon, 23 Jan 2017 22:35:40 +0000 (23:35 +0100)]
fix LOCAL_ACK transmission, do not duplicate it
Christian Grothoff [Mon, 23 Jan 2017 22:17:38 +0000 (23:17 +0100)]
avoid cmp pointers, cmp ccn instead, as client may literally be the same
Christian Grothoff [Mon, 23 Jan 2017 20:15:10 +0000 (21:15 +0100)]
fix log invocation
Christian Grothoff [Mon, 23 Jan 2017 20:03:51 +0000 (21:03 +0100)]
fix log invocation
Christian Grothoff [Mon, 23 Jan 2017 19:13:35 +0000 (20:13 +0100)]
fix testcase handling service disconnect
Christian Grothoff [Mon, 23 Jan 2017 19:09:37 +0000 (20:09 +0100)]
handle loopback in GCCH_2s
Christian Grothoff [Mon, 23 Jan 2017 14:24:20 +0000 (15:24 +0100)]
give more complete message types in log output
Christian Grothoff [Mon, 23 Jan 2017 14:04:23 +0000 (15:04 +0100)]
nicer log messages
Christian Grothoff [Mon, 23 Jan 2017 14:03:29 +0000 (15:03 +0100)]
do not send duplicate LOCAL_CHANNEL_CREATE message
Christian Grothoff [Mon, 23 Jan 2017 13:57:36 +0000 (14:57 +0100)]
handle case of MQM being already NULL
Christian Grothoff [Mon, 23 Jan 2017 13:42:23 +0000 (14:42 +0100)]
more work on loopback, this time payload transmission (for now without flow control)
Christian Grothoff [Mon, 23 Jan 2017 13:41:52 +0000 (14:41 +0100)]
avoid asserting in API on timeout destruction
Christian Grothoff [Mon, 23 Jan 2017 13:31:40 +0000 (14:31 +0100)]
add sender argument to GCCH_handle_local_data so we can implement loopback
Christian Grothoff [Mon, 23 Jan 2017 13:28:30 +0000 (14:28 +0100)]
use correct header size calculation in test
Christian Grothoff [Mon, 23 Jan 2017 13:19:24 +0000 (14:19 +0100)]
towards path desirability calculations
Christian Grothoff [Mon, 23 Jan 2017 13:19:06 +0000 (14:19 +0100)]
another special case for loopback
Christian Grothoff [Mon, 23 Jan 2017 13:18:48 +0000 (14:18 +0100)]
beautify disconnect logic a bit
Christian Grothoff [Mon, 23 Jan 2017 13:18:35 +0000 (14:18 +0100)]
handle disconnect properly in test
Christian Grothoff [Mon, 23 Jan 2017 13:06:52 +0000 (14:06 +0100)]
add generated file to ignore list
Christian Grothoff [Mon, 23 Jan 2017 13:05:16 +0000 (14:05 +0100)]
get rid of redundant, confusing state
Christian Grothoff [Mon, 23 Jan 2017 13:02:18 +0000 (14:02 +0100)]
more implementation of loopback handling
Christian Grothoff [Mon, 23 Jan 2017 13:01:34 +0000 (14:01 +0100)]
fix unaligned access
Christian Grothoff [Mon, 23 Jan 2017 10:14:37 +0000 (11:14 +0100)]
fix shutdown sequence, start to handle loopback
Christian Grothoff [Mon, 23 Jan 2017 10:13:55 +0000 (11:13 +0100)]
allow more than one ACK being given to us at a time
Christian Grothoff [Sun, 22 Jan 2017 22:55:00 +0000 (23:55 +0100)]
remove actually problematic client_allowed check
Christian Grothoff [Sun, 22 Jan 2017 22:54:35 +0000 (23:54 +0100)]
log precise error if we receive malformed data
Christian Grothoff [Sun, 22 Jan 2017 22:53:27 +0000 (23:53 +0100)]
do not send malformed payload in test
Christian Grothoff [Sun, 22 Jan 2017 22:25:25 +0000 (23:25 +0100)]
simplify check_data logic
Christian Grothoff [Sun, 22 Jan 2017 22:09:50 +0000 (23:09 +0100)]
guard against 0
Christian Grothoff [Sun, 22 Jan 2017 22:07:10 +0000 (23:07 +0100)]
allow client's payload to include more than one message
Christian Grothoff [Sun, 22 Jan 2017 21:50:38 +0000 (22:50 +0100)]
API documentation, logging
Christian Grothoff [Sun, 22 Jan 2017 21:50:14 +0000 (22:50 +0100)]
API code cleanup
Christian Grothoff [Sun, 22 Jan 2017 21:49:30 +0000 (22:49 +0100)]
fix stray log calls that lagged the core-api prefix
Christian Grothoff [Sun, 22 Jan 2017 21:30:58 +0000 (22:30 +0100)]
correctly handle assignment of cid during channel open, send channel open ack, and detect duplicate open
Christian Grothoff [Sun, 22 Jan 2017 21:29:57 +0000 (22:29 +0100)]
spell out correct constants
Christian Grothoff [Sun, 22 Jan 2017 21:28:56 +0000 (22:28 +0100)]
cosmetics for test
Christian Grothoff [Sun, 22 Jan 2017 19:24:41 +0000 (20:24 +0100)]
fix off-by-one
Christian Grothoff [Sun, 22 Jan 2017 19:21:35 +0000 (20:21 +0100)]
really fully create path objects before calling even GCP_path_entry_add()
Christian Grothoff [Sun, 22 Jan 2017 19:13:39 +0000 (20:13 +0100)]
fixing misc cleanup issues
Christian Grothoff [Sun, 22 Jan 2017 17:12:57 +0000 (18:12 +0100)]
fix copy & paste
Christian Grothoff [Sun, 22 Jan 2017 17:02:13 +0000 (18:02 +0100)]
misc bugfixes
Christian Grothoff [Sun, 22 Jan 2017 16:28:45 +0000 (17:28 +0100)]
initialize new path fully before trying to attach it
Christian Grothoff [Sun, 22 Jan 2017 15:51:37 +0000 (16:51 +0100)]
pass correct closure
Christian Grothoff [Sun, 22 Jan 2017 15:49:32 +0000 (16:49 +0100)]
logging fixes, notify MQM clients properly
Christian Grothoff [Sun, 22 Jan 2017 15:37:48 +0000 (16:37 +0100)]
add sanity check for GNUNET_MQ_notify_sent
Christian Grothoff [Sun, 22 Jan 2017 15:33:01 +0000 (16:33 +0100)]
fix shutdown logic
Christian Grothoff [Sun, 22 Jan 2017 14:50:34 +0000 (15:50 +0100)]
nicer logging
Christian Grothoff [Sun, 22 Jan 2017 14:49:46 +0000 (15:49 +0100)]
fix comparisson
Christian Grothoff [Sun, 22 Jan 2017 14:23:38 +0000 (15:23 +0100)]
must not move location of -struct CadetPeerPathEntry- in memory when growing path length as those structs are also referenced from a DLL
Christian Grothoff [Sun, 22 Jan 2017 14:22:45 +0000 (15:22 +0100)]
nicer logging
Christian Grothoff [Sun, 22 Jan 2017 14:22:22 +0000 (15:22 +0100)]
must delay iteration over paths until later, as we may be right now creating a connection
Christian Grothoff [Sun, 22 Jan 2017 14:21:42 +0000 (15:21 +0100)]
nicer order of LOG statements
Christian Grothoff [Sun, 22 Jan 2017 14:21:18 +0000 (15:21 +0100)]
minor cleanup
Christian Grothoff [Sun, 22 Jan 2017 13:20:16 +0000 (14:20 +0100)]
fix indentation
Christian Grothoff [Sun, 22 Jan 2017 13:20:01 +0000 (14:20 +0100)]
properly launch DHT GETs based on need for paths
Christian Grothoff [Sun, 22 Jan 2017 13:19:12 +0000 (14:19 +0100)]
fix a few off-by-ones
Christian Grothoff [Sun, 22 Jan 2017 13:13:42 +0000 (14:13 +0100)]
fix buffer size
Christian Grothoff [Sun, 22 Jan 2017 13:06:35 +0000 (14:06 +0100)]
make GCPP_2s also return static string
Christian Grothoff [Sun, 22 Jan 2017 11:35:32 +0000 (12:35 +0100)]
more logging for cadet-core subsystem
Christian Grothoff [Sun, 22 Jan 2017 11:23:08 +0000 (12:23 +0100)]
allow NULL argument for GNUNET_i2s()
Christian Grothoff [Sun, 22 Jan 2017 11:21:24 +0000 (12:21 +0100)]
introcude GNUNET_i2s2()
Christian Grothoff [Sun, 22 Jan 2017 11:12:31 +0000 (12:12 +0100)]
improving logging
Christian Grothoff [Sun, 22 Jan 2017 10:44:22 +0000 (11:44 +0100)]
only send CHANNEL_DESTROY message if we at least send the CHANNEL_CREATE message, and not if the tunnel never went up in the first place
Christian Grothoff [Sun, 22 Jan 2017 10:35:46 +0000 (11:35 +0100)]
fix startup interaction between channel and tunnel to not send CHANNEL_OPEN until tunnel is in KEY_OK state
Christian Grothoff [Sun, 22 Jan 2017 10:34:56 +0000 (11:34 +0100)]
improve logging, log before doing, not after
Christian Grothoff [Sun, 22 Jan 2017 10:34:24 +0000 (11:34 +0100)]
do not run upnpc during tests, produces unnecessary errors if not available
Christian Grothoff [Sun, 22 Jan 2017 10:34:02 +0000 (11:34 +0100)]
optimize startup sequence for when we execute first DHT PUTs
Christian Grothoff [Sat, 21 Jan 2017 23:52:04 +0000 (00:52 +0100)]
still fixing the stupid connect_task cleanup issue
Christian Grothoff [Sat, 21 Jan 2017 23:49:29 +0000 (00:49 +0100)]
adding missing ignores
Christian Grothoff [Sat, 21 Jan 2017 23:46:03 +0000 (00:46 +0100)]
use new SCHEDULER_add_at API where applicable