summaryrefslogtreecommitdiffstats
path: root/gatchat
AgeCommit message (Collapse)AuthorFilesLines
2011-05-24gatppp: Rename guard_timeout_src to _sourceDenis Kenzior1-5/+4
2011-05-24gatppp: Do not call disconnect func if suspendedGuillaume Zajac1-2/+3
2011-05-24gatppp: Add GAtPPP suspend APIGuillaume Zajac2-0/+41
2011-05-24gatppp: Add GAtPPP resume functionalityGuillaume Zajac2-0/+20
2011-05-24ppp_net: add ppp_net_resume_interface() APIGuillaume Zajac2-0/+11
2011-05-24gathdlc: Squash unneeded functionDenis Kenzior1-6/+1
2011-05-24gathdlc: add g_at_hdlc_resume() APIGuillaume Zajac2-0/+17
2011-05-24gatppp: Make ppp->fd handling a bit easier to grokDenis Kenzior1-1/+6
2011-05-24ppp_net: streamline ppp_net_new logicDenis Kenzior1-16/+9
2011-05-24gatppp: Ensure created GAtPPP object is not nullDenis Kenzior1-1/+2
2011-05-24gatppp: Add new contructor to use external fdGuillaume Zajac4-17/+66
2011-05-24gatppp: Don't go over 80 characters / lineDenis Kenzior1-1/+1
2011-05-24test-server: Remove legacy codeDenis Kenzior1-12/+2
The previous thinking was to use the PPP stack to send Conf-Naks with some delay to client requests until the IP configuration was setup correctly. However, we now hide this by using the delay during the processing of the ATD. This code is no longer relevant.
2011-05-11gatppp: add name field when using chap authCaiwen Zhang1-0/+9
Some systems require name field, otherwise chap authentication will fail.
2011-05-11gatppp: fix ppp protocol-reject constructing errorCaiwen Zhang1-3/+3
In PPP Protocol-Reject package the 'rejected data field' should be a copy of the rejected package, the copy starts from the data field. besides, Protocol-Reject package include a 'rejected protocol field', it is the same as the protocol field of the rejected package. Protocol-Reject package structure is: | PPP_header | rejected protocol | rejected data The rejected package structure is: | Addr | Control | protocol | data So the Protocol-Reject package data field is copied from the 3rd byte of the rejected package.
2011-05-08gatppp: Add API for setting suspend callbackGuillaume Zajac2-0/+26
In situations where the PPP stream can be suspended by the peer, e.g. in the server role using '+++'
2011-05-08ppp_net: Reset the watch to zero when removingDenis Kenzior1-3/+10
2011-05-08ppp_net: add ppp_net_suspend_interface() functionGuillaume Zajac2-0/+10
2011-05-08gathdlc: delete read/write handler in hdlc_suspendGuillaume Zajac1-0/+3
2011-05-08gathdlc: Add public suspend functionGuillaume Zajac2-0/+11
2011-05-06gatchat: Fix memory leak with notify nodes listMarcel Holtmann1-0/+1
The list itself needs to be freed as well and not only its elements.
2011-05-02gathdlc: Make sure to set in_read_handler properlyDenis Kenzior1-4/+4
2011-05-02gathdlc: Add support for detecting +++ escapesDenis Kenzior2-0/+110
2011-05-02gatio: add ability to drain GAtIO read bufferGuillaume Zajac2-0/+7
2011-04-29gat: add GAtSuspendFunc CB typedefGuillaume Zajac1-0/+1
2011-04-26gatppp: Fix crash related to not stopping timersMartin Xu1-0/+2
In case of offline modem when GPRS data connection is connected, if gprs atom is removed before PPP termination process is complete, the terminate_timer will not be stop. It will cause ofonod crash when the timer times out.
2011-03-29gatserver: add tone and pulse dialing AT commandsOlivier Guiter1-0/+38
2011-03-28gatchat: Add support for short prompt send variantDenis Kenzior1-13/+40
2011-03-28gatsyntax: support short prompt in v1 parserDenis Kenzior1-0/+24
2011-03-28gatsyntax: support short prompt for permissiveDenis Kenzior1-0/+25
2011-03-28gatchat: new hint to handle +CPOSJarko Poutiainen1-1/+2
2011-03-28gatchat: introduce send for +CPOSJarko Poutiainen1-0/+8
2011-03-28gatsyntax: Support echo of empty linesDenis Kenzior1-1/+4
2011-03-28gatsyntax: Re-name enum a proper wayDenis Kenzior1-3/+3
2011-03-25gatserver: add v250 ATL and ATM commandsOlivier Guiter1-0/+18
2011-03-24gatserver: add AT&F commandOlivier Guiter1-2/+27
2011-03-23gatserver: add ATZ commandOlivier Guiter1-18/+35
2011-03-16ringbuffer: Switch ringbuffer to g_slice allocatorPatrick Porlan1-4/+4
This should allow for more efficient handling of equally sized buffers, in terms of alignment and recycling.
2011-03-15PPP: Optimize write buffer managementPatrick Porlan1-22/+71
Extend the write buffer handling in gathdlc.c to minimize stalling and process switching during large PPP transfers. The single write buffer is replaced by a queue of buffers, allowing for much larger emission windows without hugely impacting memory consumption. This reduces the time required to send 50 MB between a couple of local PPP interfaces on my laptop from ~53s to ~3s.
2011-03-02gatchat: Optimize ringbuffer modulo operationsPatrick Porlan1-6/+8
Replace modulo operations in ringbuffer.c by masking operations. This is possible because the size of the ring buffers is always a power of two, and yields a small performance improvement. The improvement should be mostly visible on processors that implement division in microcode (Atom) or lack a division instruction (ARM).
2011-02-28gsmdial: Introduce data_modeDenis Kenzior1-0/+7
If we're in PPP mode, then don't send unsolicited notifications. Only a NO CARRIER is really allowed.
2011-02-28gatserver: Tweak use of g_at_server_suspend/resumeDenis Kenzior1-11/+18
2011-02-28test-server: Make sure to send CONNECTDenis Kenzior1-6/+9
Before starting the ppp session we should make sure that the CONNECT response has been sent on the wire.
2011-02-28gatserver: Exit garbage state on HDLC frame endDenis Kenzior1-1/+2
2011-02-28gatppp: Try to detect no carrier conditionsDenis Kenzior1-0/+1
2011-02-28gatppp: Delay signaling ppp_downDenis Kenzior1-2/+14
Delay signaling of ppp_down until we exit the GAtHDLC read handler. Otherwise exchanging the read handler on GAtIO might lead to funny double consumption problems.
2011-02-28gathdlc: Try to detect no carrier conditionsDenis Kenzior2-0/+21
Sometimes we receive the no carrier embedded in a stream following the PPP packets. This might be due to write scheduling on the remote side or read scheduling locally. Try not to consume the no carrier condition and assume the previous hdlc frames will result in closing of the ppp stack.
2011-02-28gatio: Remove stray empty lineDenis Kenzior1-1/+0
2011-02-28gatio: update commentDenis Kenzior1-2/+2
2011-02-28gathdlc: Unregister read / write handlersDenis Kenzior1-0/+3
When GAtHDLC is being unregistered make sure to unregister read / write handlers so as not to crash.