Commit Graph

1681 Commits

Author SHA1 Message Date
Gregory Nutt
c4dfb76b0d Fix some warnings found in build testing. 2018-12-30 16:05:28 -06:00
Anthony Merlino
5e1055dbfa Merged in antmerlino/nuttx/sixlowpan-d_buf-fix (pull request #795)
net/sixlowpan: Record and restore d_buf on entry/exit of sixlowpan_input. This avoids an issue where d_buf gets set to NULL by sixlowpan, and then is used by forwarding logic from another network interface before getting set back to the drivers internal b

Approved-by: GregoryN <gnutt@nuttx.org>
2018-12-26 16:33:44 +00:00
Anthony Merlino
33a4ec969a Merged in antmerlino/nuttx/avoid-double-context-lookup (pull request #788)
sixlowpan: Address comment regarding avoiding double lookup of address context.

Approved-by: GregoryN <gnutt@nuttx.org>
2018-12-19 18:41:27 +00:00
Anthony Merlino
c897790e55 Merged in antmerlino/nuttx/sixlowpan-ieee802154-address-endianness (pull request #790)
sixlowpan/ieee802154: IEEE802.15.4 stores and uses addresses in little endian order.

# Conflicts:
#	drivers/wireless/ieee802154/xbee/xbee_netdev.c

Approved-by: GregoryN <gnutt@nuttx.org>
2018-12-18 15:06:41 +00:00
Anthony Merlino
b48a62a7d4 Merged in antmerlino/nuttx/sixlowpan-ul-bit (pull request #787)
sixlowpan: Fixes logic surrounding the Universal/Local bit. This bit represents whether the IID is locally/globally administered. The U/L bit is bit 1 of the MSB of the EUI-64. It should only be inverted in cases where there is a full EUI-64. In cases whe

* sixlowpan: Fixes logic surrounding the Universal/Local bit. This bit represents whether the IID is locally/globally administered. The U/L bit is bit 1 of the MSB of the EUI-64. It should only be inverted in cases where there is a full EUI-64. In cases where the IID is derived from say, a short address, this bit should be forced to 0, indicating that it is locally administered.

    See:

    https://tools.ietf.org/html/rfc4291#section-2.5.1
    https://tools.ietf.org/html/rfc4944#section-6
    https://tools.ietf.org/html/rfc2464#section-4

* sixlowpan: Account for endianness with U/L bit.

Approved-by: GregoryN <gnutt@nuttx.org>
2018-12-18 13:34:41 +00:00
Anthony Merlino
35232d4997 Merged in antmerlino/nuttx/uncompress-addr-fix (pull request #786)
net/sixlowpan: Fixes decompression of ipaddr from MAC address. The logic used to populate the IP from the radio address should match sixlowpan_ipfromsaddr/sixlowpan_ipfromeaddr

Approved-by: GregoryN <gnutt@nuttx.org>
2018-12-16 21:39:40 +00:00
Anthony Merlino
67c609e343 Merged in antmerlino/nuttx/big-endian (pull request #784)
Fixes usage of CONFIG_BIG_ENDIAN; should be CONFIG_ENDIAN_BIG

Approved-by: GregoryN <gnutt@nuttx.org>
2018-12-16 20:33:54 +00:00
Gregory Nutt
cacbe062ad Fix a new warning found in build tesing. 2018-12-01 17:27:32 -06:00
Gregory Nutt
6941f916ce net/socket: For consistency in naming, change the name of sock_release() to psock_release(). 2018-11-27 07:50:09 -06:00
Gregory Nutt
b8cc5ef757 net/socket/net_clone.c: On a failure to start the TCP monitor, the clone socket is closed but is not properly freed, leaving it attached to a TCP connection structure with the wrong reference count. 2018-11-27 07:40:55 -06:00
Gregory Nutt
41d5e31390 configs/stm32f429i-disco/highpri/defconfig and net/procfs/netdev_statistics.c: Eliminate some warnings found in build testing. 2018-11-25 18:42:39 -06:00
Gregory Nutt
22de66d553 arch/arm/src/max326xx/max32660/max32660_wdt.c: Fix alarm delay calculation. 2018-11-21 17:54:36 -06:00
Gregory Nutt
09f4dee6bc All network drivers! Change pre-processor logic that selects the high priority work queue or gives preferential treatment to the high priority work. All network logic must run on the low priority work queue! Or suffer the consequences. 2018-11-21 07:57:26 -06:00
Gregory Nutt
a588d64b45 net/procfs/netdev_statistics.c: Fix a error found in build testing. 2018-11-20 08:09:03 -06:00
Gregory Nutt
1fd9eb6069 net/pkt: pkt_input() should not report an error using the nerr() macro when the PKT tap does not need the packet. That is not an error. Use ninfo() instead. 2018-11-16 13:48:30 -06:00
Xiang Xiao
77098f8736 Move NETDEV_LATEINIT from drivers/net/Kconfig to net/Kconfig so that we can select NETDEV_LATEINIT without enabling NETDEVICES since the net driver in arch folder may need to initialize later too. 2018-11-12 06:36:26 -06:00
Gregory Nutt
d94bd49b78 net/mld: Fix a typo in conditional compilation. Clean up return values from mld_ngroups(). 2018-11-11 16:23:18 -06:00
Gregory Nutt
9b1025ca1b net/mld/mld_group.c: Fixed failure to stop queries. This turned out to be an uninitialized variable. I wonder why there was no warning from GCC? 2018-11-11 15:25:16 -06:00
Gregory Nutt
610c187e96 net/procfs/netdev_statistics.c: Eliminate warning: array subscript is above array bounds. 2018-11-11 14:28:15 -06:00
Gregory Nutt
002f09f2da net/mld: Fix logic error when testing for the case where all members have left. Still does not work. The end result is that the query timer no longer stops. Not when another another querier with a lower IP is present or when all of the members have left. Basically just can't stop querying under any condition. 2018-11-11 12:43:12 -06:00
Gregory Nutt
bfa8c362c6 net/devif/devif_loopback.c: Fix warning: implicit declaration of memcmp. 2018-11-11 11:56:37 -06:00
Gregory Nutt
8a3fc26b74 Squashed commit of the following:
The MLD implementation did not follow the RFC correctly when it is the Querier.  The Querier should use a general query and get query messages from all members of all groups.  This would be driven by a single timer per sub-nset since all groups are queried at once. Instead, the design used a Multicast Address Specific Query with one timer per group and ignores groups that we are not members of.

    Similary, the MLDv1 compatibility timer should be a single, separate timer, not a per-group timer.

    net/mld: Group may be NULL when sending a general query
2018-11-11 11:38:29 -06:00
Gregory Nutt
591b03aa7d net/mld/mld_query.c: Eliminate a warning: implicit definitino of memcmp. 2018-11-11 07:36:52 -06:00
Gregory Nutt
1a56229386 net/mld: Resolve and issue with sending reports from multiple groups in the same polling cycle. 2018-11-11 07:31:59 -06:00
Gregory Nutt
8cfb376f81 net/mld: There is an issue with the general query timer design. There should be a single instance, not a per-group instance of the timer (ditto for the v1 compatibility timer). Fall back for now and use the per-group MAX query. 2018-11-10 12:13:54 -06:00
Gregory Nutt
f8d39e2c84 net/MLD: The MAS query should be sent on the group multicast address. 2018-11-10 11:41:50 -06:00
Gregory Nutt
854046a931 /net/devif/ipv6_input.c: Correct handling of IPv6 extension headers. The main confusion was that the payload length in the IPv6 header does not include its extension headers.
net/icmpv6/icmpv6_input.c:  Correct caculation of the ICMPv6 header address
    net/mld/mld_query.c:  Correct back test for group found.
    net/mld/mld_report.c: Fix host vs. network order problem.
    net/mld/mld_send.c: Correct the address used in sending the General Query.  It should be the unspecified address in that case.
    net/mld:  Querying workaround.  The MLD implementation does not follow the RFC correct when it is the Querier.  The Querier should use a general query and get query messages from all members of all groups.  This would be driven by a single timer since all groups are queried at once.  Instead, the design currently uses a Multicast Address Specific Query with one timer per group and ignores groups that we are not members of.
2018-11-10 11:29:02 -06:00
Gregory Nutt
b0ba5b69c4 net/udp/Kconfig: Auto-select CONFIG_WQUEUE_NOTIFIER when CONFIG_UDP_READAHEAD_NOTIFIER is selected. 2018-11-10 09:43:44 -06:00
Gregory Nutt
68a115aed9 net/icmpv6/icmpv6_autoconfig.c and icmpv6_rnotify.c: Don't take the network device down when reconfiguring only the IP address from within ICMPv6 logic. Recommended by Xiang Xiao in order to avoid the long delays of bringing some networks back up.
Normally it is required that the network be in the "down" state when re-configuring the network interface.  This is thought not to be a necessary here because.

  1. The ICMPv6 logic here runs with the network locked so there can be no outgoing packets with bad source IP addresses from any asynchronous network activity using the device being reconfigured.
  2. Incoming packets depend only upon the MAC filtering.  Network drivers do not use the IP address; they filter incoming packets using only the MAC address which is not being changed here.
2018-11-10 07:13:54 -06:00
Gregory Nutt
ae29df7445 net/devif/ipv6_input.c: Remove an old, outdated comment. 2018-11-09 17:53:31 -06:00
Gregory Nutt
a583488829 net/devif/ipv6_input.c: Fix a logic error when the received IPv6 srcaddr is a multicast address. Exiting logic only supported UDP multicast. But MLD and certain other ICMPv6 packets also require acceptance of multicast packets. 2018-11-09 16:26:49 -06:00
Gregory Nutt
8527fcf4c5 net/icmpv6/icmpv6_input.c: Fix some ICMPv6 introduced with recent changes. 2018-11-09 16:10:19 -06:00
Xiang Xiao
71e1056e6d net/icmpv6, neighbor, and netdev: Rename netdev_dev_lladdrsize() to netdev_lladdrsize() and move the prototype to include/nuttx/nex/netdev.h, giving is global scope within the OS. Reviewer: This degrades the integrity of the modular architecture by introducing a new coupling via C function call. Not a good thing at all. 2018-11-09 16:05:11 -06:00
Xiang Xiao
a139b7f9b1 net/arp and icmpv6: Double the delay time on each iteration for ARP/ICMP to fight work jitter better. 2018-11-09 15:55:38 -06:00
Xiang Xiao
4f25b83f8f et/icmpv6: Remove the coupling between ICMPv6 and Ethernet by moving all Ethernet stuff to neighbor_ethernet_out.c and make the mac address as first field. 2018-11-09 15:41:23 -06:00
Xiang Xiao
a3cda3867b net/icmpv6/icmpv6_linkipaddr.c: Encapsulate the link scope address generation into icmpv6_linkipaddr() 2018-11-09 15:28:58 -06:00
Xiang Xiao
52021a5300 net/icmpv6/icmpv6_conn.c: icmpv6_free() remove reference count operation since it's already done in icmpv6_close(). 2018-11-09 15:22:25 -06:00
Xiang Xiao
688ccb156a net/icmpv6/icmpv6_neighbor.c: icmpv6_neighbor shouldn't use the out of scope variable(dripaddr) 2018-11-09 15:18:17 -06:00
Gregory Nutt
b93c3c13c4 net/: Fix some compile errors introduce with recent networking patches. Found in build testing. 2018-11-09 14:25:44 -06:00
Xiang Xiao
e294162697 net/neighbor: Support neighbor_out() for multiple link layer at the same time 2018-11-09 14:21:49 -06:00
Xiang Xiao
0a6e234962 net/neighbor: neighbor_lookup() checks if the target IP belongs to one of the local network devices. 2018-11-09 14:14:15 -06:00
Xiang Xiao
0a673d78dc net/neighbor/neighbor_dumpentry.c: Simplify neighbor_dumpentry() implementation 2018-11-09 14:10:18 -06:00
Xiang Xiao
7f8496c643 net/neighbor: Simplify the neighbor table aging process 2018-11-09 14:08:18 -06:00
Xiang Xiao
eb31dc69ac net/: Remove the unnecessary arp.h inclusion 2018-11-09 14:01:12 -06:00
Xiang Xiao
bdf93a8abf net/arp/arp_send.c: Check ifname match before sending packet 2018-11-09 13:57:16 -06:00
Xiang Xiao
43706cd797 net/arp: Redesign ARP table aging to simplify the net initialization 2018-11-09 13:54:55 -06:00
Gregory Nutt
6ca2bcb577 net/icmpv6/icmpv6_input.c: Fix yet another build issue in this file found by build testing. 2018-11-09 11:44:01 -06:00
Xiang Xiao
b8b90b5c4f net/usrsock/usrsock_accept.c: Mark newconn ready before issue accept request to avoid the event get discarded due to the socket in the invalid state 2018-11-09 11:28:19 -06:00
Xiang Xiao
44bcd2c44d net/: add NET_ICMP[v6]_NO_STACK for usrsock case 2018-11-09 11:25:57 -06:00
Xiang Xiao
26ae4176f0 net/arp and net/icmpv6: don't call d_txavail directly since it may point to NULL 2018-11-09 11:22:58 -06:00