changelog

freescale k40/k60 freertos-lwip例程

    TCP connection caused a crash. Note that using (lwip_)recvfrom
    like this is a bit slow and that using (lwip)getpeername is the
    good lwip way to do it (so, using recv is faster on tcp sockets).

  2008-03-12 Fr閐閞ic Bernon, Jonathan Larmour
  * api_msg.c, contrib/apps/ping.c: Fix bug #22530 "api_msg.c's
    recv_raw() does not consume data", and the ping sample (with
    LWIP_SOCKET=1, the code did the wrong supposition that lwip_recvfrom
    returned the IP payload, without the IP header).

  2008-03-04 Jonathan Larmour
  * mem.c, stats.c, mem.h: apply patch #6414 to avoid compiler errors
  and/or warnings on some systems where mem_size_t and size_t differ.
  * pbuf.c, ppp.c: Fix warnings on some systems with mem_malloc.

  2008-03-04 Kieran Mansley (contributions by others) 
  * Numerous small compiler error/warning fixes from contributions to
    mailing list after 1.3.0 release candidate made.

  2008-01-25 Cui hengbin (integrated by Fr閐閞ic Bernon)
  * dns.c: Fix bug #22108 "DNS problem" caused by unaligned structures.

  2008-01-15 Kieran Mansley
  * tcp_out.c: BUG20511.  Modify persist timer to start when we are
    prevented from sending by a small send window, not just a zero
    send window.

  2008-01-09 Jonathan Larmour
  * opt.h, ip.c: Rename IP_OPTIONS define to IP_OPTIONS_ALLOWED to avoid
    conflict with Linux system headers.

  2008-01-06 Jonathan Larmour
  * dhcp.c: fix bug #19927: "DHCP NACK problem" by clearing any existing set IP
    address entirely on receiving a DHCPNAK, and restarting discovery.

  2007-12-21 Simon Goldschmidt
  * sys.h, api_lib.c, api_msg.c, sockets.c: fix bug #21698: "netconn->recv_avail
    is not protected" by using new macros for interlocked access to modify/test
    netconn->recv_avail.

  2007-12-20 Kieran Mansley (based on patch from Oleg Tyshev)
  * tcp_in.c: fix bug# 21535 (nrtx not reset correctly in SYN_SENT state)

  2007-12-20 Kieran Mansley (based on patch from Per-Henrik Lundbolm)
  * tcp.c, tcp_in.c, tcp_out.c, tcp.h: fix bug #20199 (better handling
    of silly window avoidance and prevent lwIP from shrinking the window)

  2007-12-04 Simon Goldschmidt
  * tcp.c, tcp_in.c: fix bug #21699 (segment leak in ooseq processing when last
    data packet was lost): add assert that all segment lists are empty in
    tcp_pcb_remove before setting pcb to CLOSED state; don't directly set CLOSED
    state from LAST_ACK in tcp_process

  2007-12-02 Simon Goldschmidt
  * sockets.h: fix bug #21654: exclude definition of struct timeval from #ifndef FD_SET
    If including <sys/time.h> for system-struct timeval, LWIP_TIMEVAL_PRIVATE now
    has to be set to 0 in lwipopts.h

  2007-12-02 Simon Goldschmidt
  * api_msg.c, api_lib.c: fix bug #21656 (recvmbox problem in netconn API): always
    allocate a recvmbox in netconn_new_with_proto_and_callback. For a tcp-listen
    netconn, this recvmbox is later freed and a new mbox is allocated for acceptmbox.
    This is a fix for thread-safety and allocates all items needed for a netconn
    when the netconn is created.

  2007-11-30 Simon Goldschmidt
  * udp.c: first attempt to fix bug #21655 (DHCP doesn't work reliably with multiple
    netifs): if LWIP_DHCP is enabled, UDP packets to DHCP_CLIENT_PORT are passed
    to netif->dhcp->pcb only (if that exists) and not to any other pcb for the same
    port (only solution to let UDP pcbs 'bind' to a netif instead of an IP address)

  2007-11-27 Simon Goldschmidt
  * ip.c: fixed bug #21643 (udp_send/raw_send don't fail if netif is down) by
    letting ip_route only use netifs that are up.

  2007-11-27 Simon Goldschmidt
  * err.h, api_lib.c, api_msg.c, sockets.c: Changed error handling: ERR_MEM, ERR_BUF
    and ERR_RTE are seen as non-fatal, all other errors are fatal. netconns and
    sockets block most operations once they have seen a fatal error.

  2007-11-27 Simon Goldschmidt
  * udp.h, udp.c, dhcp.c: Implemented new function udp_sendto_if which takes the
    netif to send as an argument (to be able to send on netifs that are down).

  2007-11-26 Simon Goldschmidt
  * tcp_in.c: Fixed bug #21582: pcb->acked accounting can be wrong when ACKs
    arrive out-of-order

  2007-11-21 Simon Goldschmidt
  * tcp.h, tcp_out.c, api_msg.c: Fixed bug #20287: tcp_output_nagle sends too early
    Fixed the nagle algorithm; nagle now also works for all raw API applications
    and has to be explicitly disabled with 'tcp_pcb->flags |= TF_NODELAY'

  2007-11-12 Fr閐閞ic Bernon
  * sockets.c, api.h, api_lib.c, api_msg.h, api_msg.c: Fixed bug #20900. Now, most
    of the netconn_peer and netconn_addr processing is done inside tcpip_thread
    context in do_getaddr.

  2007-11-10 Simon Goldschmidt
  * etharp.c: Fixed bug: assert fired when MEMP_ARP_QUEUE was empty (which can
    happen any time). Now the packet simply isn't enqueued when out of memory.

  2007-11-01 Simon Goldschmidt
  * tcp.c, tcp_in.c: 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 Simon Goldschmidt
  * tcp.h, tcp.c, tcp_in.c: 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 Simon Goldschmidt
  * tcp_in.c: 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.

  2007-10-09 Simon Goldschmidt
  * udp.c, inet.c, inet.h: Fixed UDPLite: send: Checksum was always generated too
    short and also was generated wrong if checksum coverage != tot_len;
    receive: checksum was calculated wrong if checksum coverage != tot_len

  2007-10-08 Simon Goldschmidt
  * mem.c: lfree was not updated in mem_realloc!

  2007-10-07 Fr閐閞ic Bernon
  * 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_addr, 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).

  2007-09-20 Simon Goldschmidt
  * tcp.c: Fixed bug #21080 (tcp_bind without check pcbs in TIME_WAIT state)
    by checking  tcp_tw_pcbs also

  2007-09-19 Simon Goldschmidt
  * icmp.c: Fixed bug #21107 (didn't reset IP TTL in ICMP echo replies)

  2007-09-15 Mike Kleshov
  * mem.c: Fixed bug #21077 (inaccuracy in calculation of lwip_stat.mem.used)

  2007-09-06 Fr閐閞ic Bernon
  * several-files: replace some #include "arch/cc.h" by "lwip/arch.h", or simply remove
    it as long as "lwip/opt.h" is included before (this one include "lwip/debug.h" which
    already include "lwip/arch.h"). Like that, default defines are provided by "lwip/arch.h"
    if they are not defined in cc.h, in the same spirit than "lwip/opt.h" for lwipopts.h.

  2007-08-30 Fr閐閞ic Bernon
  * igmp.h, igmp.c: Some changes to remove some redundant code, add some traces, 
    and fix some coding style.

  2007-08-28 Fr閐閞ic Bernon
  * tcpip.c: Fix TCPIP_MSG_INPKT processing: now, tcpip_input can be used for any
    kind of packets. These packets are considered like Ethernet packets (payload 
    pointing to ethhdr) if the netif got the NETIF_FLAG_ETHARP flag. Else, packets 
    are considered like IP packets (payload pointing to iphdr).

  2007-08-27 Fr閐閞ic Bernon
  * api.h, api_lib.c, api_msg.c: First fix for "bug #20900 : Potential crash error
    problem with netconn_peer & netconn_addr". Introduce NETCONN_LISTEN netconn_state
    and remove obsolete ones (NETCONN_RECV & NETCONN_ACCEPT).

  2007-08-24 Kieran Mansley
  * inet.c Modify (acc >> 16) test to ((acc >> 16) != 0) to help buggy
    compiler (Paradigm C++)

  2007-08-09 Fr閐閞ic Bernon, Bill Florac
  * stats.h, stats.c, igmp.h, igmp.c, opt.h: Fix for bug #20503 : IGMP Improvement.
    Introduce IGMP_STATS to centralize statistics management.

  2007-08-09 Fr閐閞ic Bernon, Bill Florac
  * udp.c: Fix for bug #20503 : IGMP Improvement. Enable to receive a multicast
    packet on a udp pcb binded on an netif's IP address, and not on "any".

  2007-08-09 Fr閐閞ic Bernon, Bill Florac
  * igmp.h, igmp.c, ip.c: Fix minor changes from bug #20503 : IGMP Improvement.
    This is mainly on using lookup/lookfor, and some coding styles...

  2007-07-26 Fr閐閞ic Bernon (and "thedoctor")
  * igmp.c: Fix bug #20595 to accept IGMPv3 "Query" messages.

  2007-07-25 Simon Goldschmidt
  * api_msg.c, tcp.c: Another fix for bug #20021: by not returning an error if
    tcp_output fails in tcp_close, the code in do_close_internal gets simpler
    (tcp_output is called again later from tcp timers).

  2007-07-25 Simon Goldschmidt
  * ip_frag.c: Fixed bug #20429: use the new pbuf_copy_partial instead of the old
    copy_from_pbuf, which illegally modified the given pbuf.

  2007-07-25 Simon Goldschmidt
  * tcp_out.c: tcp_enqueue: pcb->snd_queuelen didn't work for chaine PBUF_RAMs:
    changed snd_queuelen++ to snd_queuelen += pbuf_clen(p).

  2007-07-24 Simon Goldschmidt
  * api_msg.c, tcp.c: Fix bug #20480: Check the pcb passed to tcp_listen() for the
    correct state (must be CLOSED).

  2007-07-13 Thomas Taranowski (commited by Jared Grubb)
  * memp.c: Fix bug #20478: memp_malloc returned NULL+MEMP_SIZE on failed
    allocation. It now returns NULL.

  2007-07-13 Fr閐閞ic Bernon
  * api_msg.c: Fix bug #20318: api_msg "recv" callbacks don't call pbuf_free in
    all error cases.

  2007-07-13 Fr閐閞ic Bernon
  * api_msg.c: Fix bug #20315: possible memory leak problem if tcp_listen failed,
    because current code doesn't follow rawapi.txt documentation.

  2007-07-13 Kieran Mansley
  * src/core/tcp_in.c Apply patch#5741 from Oleg Tyshev to fix bug in
    out of sequence processing of received packets

  2007-07-03 Simon Goldschmidt
  * nearly-all-files: Added assertions where PBUF_RAM pbufs are used and an
    assumption is made that this pbuf is in one piece (i.e. not chained). These
    assumptions clash with the possibility of converting to fully pool-based
    pbuf implementations, where PBUF_RAM pbufs might be chained.

  2007-07-03 Simon Goldschmidt
  * api.h, api_lib.c, api_msg.c: Final fix for bug #20021 and some other problems
    when closing tcp netconns: removed conn->sem, less context switches when
    closing, both netconn_close and netconn_delete should safely close tcp
    connections.

  2007-07-02 Simon Goldschmidt
  * ipv4/ip.h, ipv6/ip.h, opt.h, netif.h, etharp.h, ipv4/ip.c, netif.c, raw.c,
    tcp_out.c, udp.c, etharp.c: Added option LWIP_NETIF_HWADDRHINT (default=off)
    to cache ARP table indices with each pcb instead of single-entry cache for
    the complete stack.

  2007-07-02 Simon Goldschmidt
  * tcp.h, tcp.c, tcp_in.c, tcp_out.c: Added some ASSERTS and casts to prevent
    warnings when assigning to smaller types.

  2007-06-28 Simon Goldschmidt
  * tcp_out.c: Added check to prevent tcp_pcb->snd_queuelen from overflowing.

  2007-06-28 Simon Goldschmidt
  * tcp.h: Fixed bug #20287: Fixed nagle algorithm (sending was done too early if
    a segment contained chained pbufs)

  2007-06-28 Fr閐閞ic Bernon
  * autoip.c: replace most of rand() calls by a macro LWIP_AUTOIP_RAND which compute
    a "pseudo-random" value based on netif's MAC and some autoip fields. It's always
    possible to define this macro in your own lwipopts.h to always use C library's
    rand(). Note that autoip_create_rand_addr doesn't use this macro.

  2007-06-28 Fr閐閞ic Bernon
  * netifapi.h, netifapi.c, tcpip.h, tcpip.c: Update code to handle the option
    LWIP_TCPIP_CORE_LOCKING, and do some changes to be coherent with last modifications
    in api_lib/api_msg (use pointers and not type with table, etc...) 

  2007-06-26 Simon Goldschmidt
  * udp.h: Fixed bug #20259: struct udp_hdr was lacking the packin defines.

  2007-06-25 Simon Goldschmidt
  * udp.c: Fixed bug #20253: icmp_dest_unreach was called with a wrong p->payload
    for udp packets with no matching pcb.

  2007-06-25 Simon Goldschmidt
  * udp.c: Fixed bug #20220: UDP PCB search in udp_input(): a non-local match
    could get udp input packets if the remote side matched.

  2007-06-13 Simon Goldschmidt
  * netif.c: Fixed bug #20180 (TCP pcbs listening on IP_ADDR_ANY could get
    changed in netif_set_ipaddr if previous netif->ip_addr.addr was 0.

  2007-06-13 Simon Goldschmidt
  * api_msg.c: pcb_new sets conn->err if protocol is not implemented
    -> netconn_new_..() does not allocate a new connection for unsupported
    protocols.

  2007-06-13 Fr閐閞ic Bernon, Simon Goldschmidt
  * api_lib.c: change return expression in netconn_addr and netconn_peer, because
    conn->err was reset to ERR_OK without any reasons (and error was lost)...

  2007-06-13 Fr閐閞ic Bernon, Matthias Weisser
  * opt.h, mem.h, mem.c, memp.c, pbuf.c, ip_frag.c, vj.c: Fix bug #20162. Rename
    MEM_ALIGN in LWIP_MEM_ALIGN and MEM_ALIGN_SIZE in LWIP_MEM_ALIGN_SIZE to avoid
    some macro names collision with some OS macros.

  2007-06-11 Simon Goldschmidt
  * udp.c: UDP Lite: corrected the u