Sylvain Rochet
ee752ab1ce
PPP, PPPoS, renamed PPP_INPROC_MULTITHREADED to PPP_INPROC_IRQ_SAFE
...
Follow-up of the #44565 bug fix, renamed the misnamed
PPP_INPROC_MULTITHREADED to PPP_INPROC_IRQ_SAFE because it is
IRQ safe but not thread safe.
Updated PPP documentation which now clearly state when and how
this feature can be used.
2015-03-19 21:43:55 +01:00
sg
9eb900c448
fixed race conditions in assigning netconn->last_err (fixed bugs #38121 and #37676 )
2015-03-19 21:20:29 +01:00
Sylvain Rochet
0e919d25e9
PPP, PPPoS, improved tcpip input path, fixed bug #44565
...
New input type TCPIP_MSG_INPKT_PPPOS. Removed the netif input pointer usage
which was actually broken by design.
Fixed bug #44565 .
2015-03-19 21:04:21 +01:00
Ivan Delamer
e0fe83740a
fixed bug #44565 : PPPOS support in tcpip thread breaks SLIP
2015-03-18 10:16:14 -06:00
Sylvain Rochet
9778b1411c
PPP, PPPoS, TCPIP: add packet input path for point to point interfaces (only PPPoS for now) through the TCPIP API
...
!NO_SYS users may now use as well the TCPIP API for PPPoS input data,
this way they can disable PPP_INPROC_MULTITHREADED and run pppos_input()
inside the lwIP thread, which fixes, at least for them, all the
threading issues related to PPP_INPROC_MULTITHREADED.
2015-03-11 00:58:09 +01:00
Sylvain Rochet
969306f0f1
API: netdb: replaced deprecated ip_ntoa() to ipaddr_ntoa()
2015-03-08 13:57:13 +01:00
Sylvain Rochet
3ce6dd166c
PPP, L2TP, added link-level IPv6 support
2015-03-01 22:04:24 +01:00
Sylvain Rochet
2731976a95
PPP, renamed ppp_open to ppp_connect
...
Makes it clear we are initiating the PPP session with ppp_connect
(i.e. acting as a PPP client) so there is no confusion possible
between ppp_connect and ppp_listen.
2015-03-01 10:53:08 +01:00
Sylvain Rochet
371bc91d73
PPP, SERVER: added PPPoS server support
...
New function: ppp_listen(), listen for an incoming PPP connection.
2015-02-28 22:41:18 +01:00
Joel Cunningham
9004554da3
fixed bug #43028 (IP_MULTICAST_TTL affects unicast datagrams)
2015-02-25 21:59:21 +01:00
sg
83740d81f2
fixed comment
2015-02-25 21:32:33 +01:00
sg
c8581e4cd9
fixed bug #38165 (socket with mulicast): ensure igmp membership are dropped when socket (not netconn!) is closed.
2015-02-25 21:30:22 +01:00
sg
3e8ac30940
Fixed bug #44297 (CORE_LOCKING was broken some days ago); fixed that netconn_connect still used message passing for LWIP_TCPIP_CORE_LOCKING==1
2015-02-25 20:34:18 +01:00
Sylvain Rochet
00bb70a62d
PPP, CORE, functions ppp_set_netif_statuscallback() and ppp_set_netif_linkcallback() replaced with defines
...
PPP is now pointerful for a while, we don't need anymore accessor functions
for the unique PPP local and static control block. Replaced
ppp_set_netif_statuscallback() and ppp_set_netif_linkcallback() functions to
defines.
Removed pppapi_do_ppp_set_netif_statuscallback() and
pppapi_do_ppp_set_netif_linkcallback(), they were useless because
netif_set_status_callback() and netif_set_link_callback() can be
safely called while PPP status is in dead (= non open) state
and even before the PPP session is actually created at all.
2015-02-24 20:41:47 +01:00
sg
03159254ce
added proper accessor functions for pcb->multicast_ip (previously used by get/setsockopt only)
2015-02-22 21:49:46 +01:00
sg
ec5cf8593e
Continued chrysn's work: changed nearly all functions taking 'ip(X)_addr_t' pointer to take const pointers (changed user callbacks: raw_recv_fn, udp_recv_fn; changed port callbacks: netif_output_fn, netif_igmp_mac_filter_fn)
2015-02-22 21:46:35 +01:00
Sylvain Rochet
4be7fccad3
PPP, CORE, ppp_close() and ppp_sighup() ended up sharing almost everything, merged
...
Merged ppp_sighup() to ppp_close() using an optional argument "nocarrier"
on ppp_close().
2015-02-22 14:25:36 +01:00
Sylvain Rochet
dc2e700057
PPP, CORE, using u8_t on ioctl command instead of int
...
We don't need an int here, all commands are between 0 and 255.
2015-02-21 21:37:08 +01:00
Sylvain Rochet
6b4db944dd
PPP, using err_t return code instead of PPPERR_
...
Standardised PPP API to follow lwIP already used return codes.
PPPERR_ are now used only on link status callback.
2015-02-20 21:24:58 +01:00
Sylvain Rochet
0a761d238a
PPP, using err_t return type on ppp_ioctl()
2015-02-20 21:01:24 +01:00
sg
730529353d
fixed bug #37958 "netconn API doesn't handle correctly connections half-closed by peer"
2015-02-18 22:21:26 +01:00
sg
2cfc9e286e
"Not connected" shouldn't be fatal (as opposed to "closed")
2015-02-18 20:39:20 +01:00
sg
998ed99288
Fixed select not reporting received FIN as 'readable' in certain rare cases (bug #43779 : select(), close(), and TCP retransmission error)
2015-02-18 20:38:42 +01:00
Sylvain Rochet
ee2936ffbf
PPP, all protocols, uniformised naming
...
Uniformised fonction naming between and inside PPPoS, PPPoE, PPPoL2TP.
2015-02-17 23:22:14 +01:00
Sylvain Rochet
9c15ffbb74
PPP: Moved PPPoS from ppp.c to pppos.c
...
Started removing from PPP core most of low level protocol PPPoS
handling to separate files, using the new low level PPP callbacks.
2015-02-17 23:22:13 +01:00
Sylvain Rochet
ee85aaccd2
PPP, removed low level create functions from PPP core
...
ppp_over_ethernet_create() moved from ppp.c to pppoe.c
ppp_over_l2tp_create() moved from ppp.c to pppol2tp.c
2015-02-17 23:22:13 +01:00
Sylvain Rochet
17c6be6a9b
PPP: removed ppp_delete(), merged with ppp_free()
...
The only benefit of ppp_delete() call was about having a persistent
netif interface. netif was moved out of PPP pcb so we don't need
ppp_delete() anymore, second step in simplifying the weird
new/open/free/delete PPP API.
2015-02-17 23:22:12 +01:00
Sylvain Rochet
318e752fd6
PPP: removed ppp_new(), merged with ppp_over_X_create()
...
The only benefit of ppp_new() call was about having a persistent netif
interface. netif was moved out of PPP pcb so we don't need ppp_new()
anymore, first step in simplifying the weird new/open/free/delete PPP
API.
2015-02-17 23:22:12 +01:00
Sylvain Rochet
3dcfd7ba41
PPP: moved netif out of PPP control block
...
Users might want to share a netif control block between an Ethernet
interface and a PPPoS interface (I want actually) in case PPP is just
used as redundancy if Ethernet is down (eg. PPPoS GPRS fail over).
Moved netif out of PPP control block in a similar way it is currently
done for Ethernet interfaces. Furthermore, this is a first step on
removing the "new/create/free/delete" API which is awful but currently
necessary to handle fail over from PPPoX to another PPPoX (eg. from PPoE
on xDSL to PPPoS on GPRS fail over) without free()ing the netif which
might be used on udp_sendto() or L2TP VPN links.
2015-02-17 23:22:12 +01:00
sg
604a92dc3d
fixed bug #38853 "connect() use a wrong errno": return ERR_ALREADY/EALRADY during connect, ERR_ISCONN/EISCONN when already connected
2015-02-17 22:10:23 +01:00
sg
b146ae96a7
sockets.c: removed LWIP_TCPIP_CORE_LOCKING hack in lwip_sendto(): since all functions are static when LWIP_NETCONN==0, the speedup gained by this hack should now be minimal and not worth the duplicate code.
2015-02-17 21:26:40 +01:00
goldsimon
0963e91c34
api_msg_c: fixed compiler warning (added brackets in if statement)
2015-02-17 09:38:20 +01:00
goldsimon
5d13b5a2fb
fixed bug #37614 "Errors from ipX_output are not processed". Now tcp_output(_segment) checks for the return value of ipX_output and does not try to send more on error. A netif driver can call tcp_txnow() (from tcpip_thread!) to try to send again if TX buffers are available again.
2015-02-17 08:02:34 +01:00
sg
8155b8cfb3
patch #7702 "Include ability to increase the socket number with defined offset"
2015-02-11 23:05:52 +01:00
sg
276e35ecfb
Fixed a bug in linger-closing when LWIP_TCPIP_CORE_LOCKING==1
2015-02-11 22:53:19 +01:00
sg
060a1fc727
patch #8481 Fix a couple of c&p mistakes in the error prints
2015-02-11 21:39:45 +01:00
sg
28783abbe2
fixed bug #43094 "The function tcpip_input() forget to handle IPv6"
2015-02-11 21:36:46 +01:00
sg
7ff9825f55
started to implement fullduplex sockets/netconns (note that this is highly unstable yet!)
2015-02-11 21:18:15 +01:00
sg
6c3f6cfd89
netconn/socket api: fixed bug #44225 "closing TCP socket should time out eventually", implemented task #6930 "Implement SO_LINGER": closing TCP sockets times out after 20 seconds or after the configured SND_TIMEOUT or depending on the linger settings; fixed that netconn_close/netconn_delete still used message passing for LWIP_TCPIP_CORE_LOCKING==1
2015-02-10 22:15:54 +01:00
sg
ec68aaf43b
fixed that SHUT_RD followed by SHUT_WR was different to SHUT_RDWR, fixed return value of lwip_netconn_do_close on unconnected netconns
2015-01-27 21:28:39 +01:00
goldsimon
24df78bcbc
fixed that lwip_netconn_do_delconn() did not set msg->err on success (just introduced that bug this weekend...)
2015-01-20 14:01:16 +01:00
sg
05aa0ad114
netconn_delete(): check for errors returned by lwip_netconn_do_delconn() and don't call netconn_free() on error
2015-01-17 21:54:17 +01:00
sg
5ceaed291f
fixed bug #43361 select() crashes with stale FDs
2015-01-17 21:02:58 +01:00
sg
c1804810d8
allow enabling socket API without (public) netconn API - netconn API is still used by sockets, but keeping it private (static) should allow better compiler optimizations
2015-01-17 14:42:50 +01:00
sg
5d2e93e5f0
fixed bug #40788 "lwip_setsockopt_internal() crashes" by rewriting set/getsockopt functions to combine checks with the actual code and add more NULL checks; this also fixes that CORE_LOCKING used message passing for set/getsockopt.
2015-01-17 14:18:27 +01:00
Sylvain Rochet
91356d2d03
PPPAPI, added const modifier on auth strings
2014-12-24 19:05:19 +01:00
goldsimon
a883fe7b8e
sockets.c: fixed assert check after changing type of sock->select_waiting to unsigned
2014-12-17 16:21:44 +01:00
sg
7ca4fd817e
- fixed bug #43797 set/getsockopt: SO_SNDTIMEO/SO_RCVTIMEO take int as option but should take timeval (LWIP_SO_SNDRCVTIMEO_STANDARD==0 can be used to revert to the old 'winsock' style behaviour);
...
- Fixed implementation of SO_ACCEPTCONN to just look at the pcb state;
- cleaned up the SO/SOF defines (only 3 left a ip_pcb level);
2014-12-10 21:46:49 +01:00
sg
8e13bcd43d
fixed indentation
2014-12-10 21:24:02 +01:00
sg
cacdbb5262
added option LWIP_NETCONN_SEM_PER_THREAD to use a semaphore per thread instead of using one per netconn and per select call
2014-12-10 20:45:01 +01:00