Commit Graph

1248 Commits

Author SHA1 Message Date
fbernon
ecce865cfe api.h, api_lib.c, api_msg.c, sockets.c, opt.h: add SO_RCVBUF option for UDP & RAW netconn. You need to set LWIP_SO_RCVBUF=1 in your lwipopts.h (it's disabled by default). Netconn API users can use the netconn_recv_bufsize macro to access it. This is a first release which have to be improve for TCP. Note it used the netconn::recv_avail which need to be more "thread-safe" (note there is already the problem for FIONREAD with lwip_ioctl/ioctlsocket). 2007-11-02 10:37:08 +00:00
fbernon
4a6f60937a Minor change, add comment for "apiflags" parameters. 2007-11-01 19:38:01 +00:00
fbernon
cbe9b050a9 sockets.h, sockets.c, api.h, api_lib.c, api_msg.h, api_msg.c, tcp.h, tcp_out.c: Integrate "patch #6250 : MSG_MORE flag for send". MSG_MORE is used at socket api layer, NETCONN_MORE at netconn api layer, and TCP_WRITE_FLAG_MORE at raw api layer. This option enable to delayed TCP PUSH flag on multiple "write" calls. Note that previous "copy" parameter for "write" APIs is now called "apiflags". 2007-11-01 17:37:50 +00:00
goldsimon
2d5908f4de Fixed bug #21494: The send mss (pcb->mss) is set to 536 (or TCP_MSS if that is smaller) as long as no MSS option is received from the remote host. 2007-11-01 16:53:43 +00:00
goldsimon
298d5cf042 Fixed bug #21491: The MSS option sent (with SYN) is now based on TCP_MSS instead of pcb->mss (on passive open now effectively sending our configured TCP_MSS instead of the one received). 2007-11-01 16:23:32 +00:00
goldsimon
853765954e Fixed bug #21181: On active open, the initial congestion window was calculated based on the configured TCP_MSS, not on the MSS option received with SYN+ACK. 2007-11-01 15:51:43 +00:00
goldsimon
7a99d73092 Added some explaining comments. 2007-10-28 17:43:21 +00:00
fbernon
7077d51f1f api.h, api_lib.c, api_msg.c: Add macro API_EVENT in the same spirit than TCP_EVENT_xxx macros to get a code more readable. It could also help to remove some code (like we have talk in "patch #5919 : Create compile switch to remove select code"), but it could be done later. 2007-10-24 12:09:18 +00:00
goldsimon
270c7c1110 Removed check for conn->err for functions whose counterparts in api_msg.c also do the check (1. remove redundant code, 2. check is better done in tcpip thread context) 2007-10-22 20:01:16 +00:00
goldsimon
2d3a64815f Minor coding style fixes, added comment 2007-10-22 19:59:52 +00:00
fbernon
86dc53b72a Minor change (coding style, #if/#endif to include icmp.h is not necessary) 2007-10-16 07:11:35 +00:00
fbernon
649d43c2c5 Minor change (define DHCP_COARSE_TIMER_MSECS - using DHCP_COARSE_TIMER_SECS - to use milliseconds like all others timers) 2007-10-15 21:31:42 +00:00
goldsimon
6d9b44e243 IP_REASSEMBLY: send ICMP time exceeded when discarding datagrams for which the first fragment was received; try to keep the header of the first (octet 0) fragment; combined code to make it smaller; fixed bug in timer: when freeing packets, counter was not updated correctly 2007-10-15 19:33:16 +00:00
goldsimon
e864ba6c91 IP_REASSEMBLY: Changed 'packet' to 'datagram' to match the names used in the RFCs 2007-10-15 19:28:09 +00:00
goldsimon
0686380302 Removed bug on memory failure introduced 3 months ago in rev 1.34 (input pbuf was freed on puf_alloc failure) 2007-10-15 19:18:02 +00:00
fbernon
282e9ead17 Minor change (sanity check for MEMP_NUM_REASSDATA only done if IP_REASSEMBLY=1) 2007-10-12 06:50:16 +00:00
goldsimon
1f08a5e5f5 Added code to free the oldest packets if the IP_REASS_MAX_PBUFS limit is reached or MEMP_REASSDATA is empty. 2007-10-11 19:49:06 +00:00
goldsimon
c41508cc8d Added check that MEMP_NUM_REASSDATA is bigger than IP_REASS_MAX_PBUFS (because each MEMP_NUM_REASSDATA has one pbuf at least!) 2007-10-11 19:26:04 +00:00
goldsimon
187df4464e Added description for MEMP_NUM_REASSDATA and changed the default value to 5. 2007-10-11 19:24:59 +00:00
goldsimon
c94d640e2a Gave the parameters of pbuf_cat() and pbuf_chain() better names to make the use more clear. 2007-10-11 19:24:23 +00:00
fbernon
cf998b7bf8 Minor fix (remove warning) 2007-10-10 08:13:58 +00:00
fbernon
1a104762ea Update documentation for the SNMP uptime timestamp. 2007-10-10 08:12:19 +00:00
fbernon
2a5209bb5a Update documentation for the SNMP uptime timestamp. 2007-10-10 08:05:34 +00:00
fbernon
ebd167b658 Add sanity checking on TCP_MAXRTX and TCP_SYNMAXRTX 2007-10-09 20:48:28 +00:00
goldsimon
c13ec867c9 Forgot two explicit initializations... 2007-10-09 20:00:55 +00:00
goldsimon
199648ff37 Changed initialization: many init functions are not needed any more since we now rely on the compiler initializing global and static variables to zero! 2007-10-09 19:59:56 +00:00
goldsimon
536f2e42d2 Fixed UDPLite receive: checksum was calculated wrong if checksum coverage != tot_len 2007-10-09 19:43:29 +00:00
goldsimon
0f9799d7e0 Added todo: free the oldest entry when the MEMP_REASSDATA pool is empty or the IP_REASS_MAX_PBUFS limit is reached 2007-10-09 19:31:54 +00:00
goldsimon
b1a4b00091 Preprocessor statements have to start at the beginning of a line for old compilers! 2007-10-09 19:30:17 +00:00
goldsimon
c1c9983cfe Removed <warning: potential uninitialized reference to
"buf" in function "lwip_recvfrom">
2007-10-09 19:29:54 +00:00
goldsimon
8980c3f74e Typo in chain_frag_into_packet_and_validate (but obviously worked anyway...) 2007-10-09 06:13:12 +00:00
goldsimon
418f57dd6a Removed type conversion warning in debug statement. 2007-10-08 19:06:58 +00:00
goldsimon
9ef005454a lfree was not updated in mem_realloc! 2007-10-08 18:39:44 +00:00
fbernon
afad35a8ac Minor change (add a default value to tune for MEMP_NUM_REASSDATA in opt.h) 2007-10-08 07:39:24 +00:00
goldsimon
2ca113a218 Fixed UDPLite SENDING: Checksum was always generated too short and also was generated wrong if checksum coverage != tot_len. 2007-10-07 20:19:23 +00:00
fbernon
911ee4d9f9 Minor change (coding style) 2007-10-07 17:41:21 +00:00
fbernon
9804b25acc Fix CHANGELOG entry 2007-10-07 17:30:50 +00:00
fbernon
87e16a8f47 sockets.c, api.h, api_lib.c: First step to fix "bug #20900 : Potential crash error problem with netconn_peer & netconn_addr". VERY IMPORTANT: this change cause an API breakage for netconn_peer, since a parameter type change. Any compiler should cause an error without any changes in yours netconn_peer calls (so, it can't be a "silent change"). It also reduce a little bit the footprint for socket layer (lwip_getpeername & lwip_getsockname use now a common lwip_getaddrname function since netconn_peer & netconn_addr have the same parameters). 2007-10-07 17:26:54 +00:00
goldsimon
e561c7b49d Added check for overlapping or duplicate fragments. ip_reass_init() is not needed any more (the check is done in chain_frag_into_packet_and_validate()). 2007-10-07 11:52:32 +00:00
goldsimon
8cd65eaf46 Check for TCP_SND_QUEUELEN overflow. 2007-10-07 11:27:45 +00:00
goldsimon
b7ef6077df Changed IP_REASSEMBLY to enqueue the received pbufs so that multiple packets can be reassembled simultaneously and no static reassembly buffer is needed. 2007-10-06 15:24:43 +00:00
goldsimon
cb71d6d393 Moved ethernet_input from tcpip.c to etharp.c so all netifs (or ports) can use it. 2007-10-05 14:00:48 +00:00
fbernon
79c00be516 netifapi.h, netifapi.c: add function netifapi_netif_set_default. Change the common function to reduce a little bit the footprint (for all functions using only the "netif" parameter). 2007-10-05 13:34:48 +00:00
fbernon
949efb414a netifapi.h, netifapi.c: add functions netifapi_netif_set_up, netifapi_netif_set_down, netifapi_autoip_start and netifapi_autoip_stop. Use a common function to reduce a little bit the footprint (for all functions using only the "netif" parameter). 2007-10-03 16:54:27 +00:00
jifl
5dd1256769 Small tweak: clarify use of link speed in comments for NETIF_INIT_SNMP. 2007-10-03 09:10:28 +00:00
fbernon
c52ef1d2b3 Minor changes for debug (indent, and set LWIP_DBG_TYPES_ON to LWIP_DBG_ON to have all traces selected display per default, in the same spirit that LWIP_DBG_MIN_LEVEL is set per default to LWIP_DBG_LEVEL_OFF) 2007-09-22 11:16:07 +00:00
fbernon
c7e91d3302 Minor change, sys_prot_t is undefined is NO_SYS=1 2007-09-21 21:23:39 +00:00
jifl
c1bbcf5ed3 Add further checking that MEMP_NUM_SYS_TIMEOUT is high enough 2007-09-21 14:39:03 +00:00
fbernon
87c5a61d07 Minor changes by Andrew Dennison: add sanity check, improve debug messages for memp, fix some warnings... 2007-09-21 12:36:35 +00:00
goldsimon
0aa3d89134 Fixed bug #21080 (tcp_bind without check pcbs in TIME_WAIT state) by checking tcp_tw_pcbs also 2007-09-20 18:27:50 +00:00