[RFC] add olsr2 routing algorithm by benpicco · Pull Request #765 · RIOT-OS/RIOT

benpicco · 2014-02-22T22:21:54Z

This adds an implementation of the OLSRv2 routing protocol (without metrics), based on oonf_api.
The interface offered to applications currently consists of only a olsr_init function which will set up and start a sender and receiver thread for RFC5444 based (UDP on port 269) HELLO and Topology Control messages.
That means that every HELLO_REFRESH_INTERVAL (2s + random jitter) a non-forwarded HELLO message is send for neighborhood discovery. Every TC_REFRESH_INTERVAL (5s + random jitter) a topology control message is broadcast (only by selected flooding MPRs) through the network.

This happens transparent to the application, IP packets are now routed to known hosts within the MANET once a path has been established.

print_topology_set will output the forwarding table with all known mesh nodes.

Does work well on native with #680, but still crashes on the testbed after some time.
Each known MANET router requires 72 byte of memory + 12 bytes for each alternative route to the node.

Depends on #630

benpicco · 2014-05-12T11:29:46Z

What is necessary to make the example application work with buildtest?

LudwigKnuepfer · 2014-05-12T13:07:52Z

I thought you would just need to fill out the whitelist/blacklist .. you can take a look at the other examples/buildtest output for reference.

But looking at the output, it appears that the application does not build at all.. Therefore I'd say you need to fix the bugs ;-)

    Building application: olsr2

failed: avsextrem, chronos, mbed_lpc1768, msb-430, msb-430h, msba2, native, pttu, redbee-econotag, telosb, wsn430-v1_3b, wsn430-v1_4

LudwigKnuepfer · 2014-05-12T14:55:10Z

~~you also need to model the oonf_api dependency~~

benpicco · 2014-05-12T14:57:15Z

#1142 should fix buildtest, it triggered a problem with the oonf_api Makefile

LudwigKnuepfer · 2014-05-12T15:10:57Z

you also need to model the oonf_api dependency

I thought it failed the first time I built it, now this does not happen anymore even if I delete everything ... not sure what I did there..

LudwigKnuepfer · 2014-05-12T15:39:29Z

I get errors when building for msba2:

"make" -C /home/ludwig/RIOT/pkg/oonf_api
make[1]: Entering directory '/home/ludwig/RIOT/pkg/oonf_api'
make -C /home/ludwig/RIOT/pkg/oonf_api/oonf_api
make[2]: Entering directory '/home/ludwig/RIOT/pkg/oonf_api/oonf_api'
mkdir -p /home/ludwig/RIOT/examples/olsr2/bin/msba2/
make[3]: Entering directory '/home/ludwig/RIOT/pkg/oonf_api/oonf_api/src-api/common'
make[3]: '/home/ludwig/RIOT/examples/olsr2/bin/msba2/oonf_common.a' is up to date.
make[3]: Leaving directory '/home/ludwig/RIOT/pkg/oonf_api/oonf_api/src-api/common'
make[3]: Entering directory '/home/ludwig/RIOT/pkg/oonf_api/oonf_api/src-api/rfc5444'
arm-none-eabi-gcc -DRIOT -DENABLE_NAME -DBOARD_MSBA2 -DINIT_ON_START -DENABLE_LEDS -DBOARD_MSBA2 -DCPU_LPC2387 -DMODULE_ARM_COMMON -DMODULE_AUTO_INIT -DMODULE_CC110X_NG -DMODULE_CC110X_SPI -DMODULE_CONFIG -DMODULE_CORE -DMODULE_CPU -DMODULE_DESTINY -DMODULE_GPIOINT -DMODULE_IEEE802154 -DMODULE_LIB -DMODULE_LPC_COMMON -DMODULE_NET_HELP -DMODULE_NET_IF -DMODULE_OLSR2 -DMODULE_OONF_API -DMODULE_OONF_COMMON -DMODULE_OONF_RFC5444 -DMODULE_POSIX -DMODULE_PROTOCOL_MULTIPLEX -DMODULE_PS -DMODULE_RANDOM -DMODULE_RTC -DMODULE_SHELL -DMODULE_SHELL_COMMANDS -DMODULE_SIXLOWPAN -DMODULE_SYS -DMODULE_TIMEX -DMODULE_TRANSCEIVER -DMODULE_UART0 -DMODULE_VTIMER -O2 -Wall -Wstrict-prototypes -mcpu=arm7tdmi-s -gdwarf-2 -std=gnu99 -fno-delete-null-pointer-checks -O2 -Wall -Wstrict-prototypes -mcpu=arm7tdmi-s -gdwarf-2 -DVERSION=\"v0.3.0-3-gf2be\" -I/home/ludwig/RIOT/core/include -I/home/ludwig/RIOT/drivers/include -I/home/ludwig/RIOT/sys/include -I/home/ludwig/RIOT/boards/msba2/include -I/home/ludwig/RIOT/boards/msba2-common/include -I/home/ludwig/RIOT/boards/msba2-common/drivers/include -I/home/ludwig/RIOT/cpu/lpc2387/include -I/home/ludwig/RIOT/cpu/arm_common/include/ -I/home/ludwig/RIOT/cpu/lpc_common/include -I/home/ludwig/RIOT/sys/net/include -I/home/ludwig/RIOT/drivers/cc110x -I/home/ludwig/RIOT/drivers/cc110x_ng/include -I/home/ludwig/RIOT/sys/posix/include -I /home/ludwig/RIOT/pkg/oonf_api/oonf_api/src-api -I/home/ludwig/RIOT/boards/msba2/include -I/home/ludwig/RIOT/boards/msba2-common/include -I/home/ludwig/RIOT/boards/msba2-common/drivers/include -I/home/ludwig/RIOT/sys/include -I/home/ludwig/RIOT/sys/net/include -I/home/ludwig/RIOT/sys/posix/include -I/home/ludwig/RIOT/sys/posix/pnet/include -c rfc5444_reader.c -o /home/ludwig/RIOT/examples/olsr2/bin/msba2/oonf_rfc5444/rfc5444_reader.o
rfc5444_reader.c:129:1: error: conflicting types for 'rfc5444_reader_handle_packet'
 rfc5444_reader_handle_packet(struct rfc5444_reader *parser, uint8_t *buffer, size_t length) {
 ^
In file included from rfc5444_reader.c:48:0:
/home/ludwig/RIOT/pkg/oonf_api/oonf_api/src-api/rfc5444/rfc5444_reader.h:312:12: note: previous declaration of 'rfc5444_reader_handle_packet' was here
 EXPORT int rfc5444_reader_handle_packet(
            ^
rfc5444_reader.c:439:1: error: conflicting types for '_parse_tlv'
 _parse_tlv(struct rfc5444_reader_tlvblock_entry *entry, uint8_t **ptr, uint8_t *eob, uint8_t addr_count) {
 ^
rfc5444_reader.c:65:12: note: previous declaration of '_parse_tlv' was here
 static int _parse_tlv(struct rfc5444_reader_tlvblock_entry *entry, uint8_t **ptr,
            ^
rfc5444_reader.c:542:1: error: conflicting types for '_parse_tlvblock'
 _parse_tlvblock(struct rfc5444_reader *parser,
 ^
rfc5444_reader.c:67:12: note: previous declaration of '_parse_tlvblock' was here
 static int _parse_tlvblock(struct rfc5444_reader *parser,
            ^
rfc5444_reader.c:605:1: error: conflicting types for '_schedule_tlvblock'
 _schedule_tlvblock(struct rfc5444_reader_tlvblock_consumer *consumer, struct rfc5444_reader_tlvblock_context *context,
 ^
rfc5444_reader.c:69:12: note: previous declaration of '_schedule_tlvblock' was here
 static int _schedule_tlvblock(struct rfc5444_reader_tlvblock_consumer *consumer,
            ^
rfc5444_reader.c:791:1: error: conflicting types for '_parse_addrblock'
 _parse_addrblock(struct rfc5444_reader_addrblock_entry *addr_entry,
 ^
rfc5444_reader.c:71:12: note: previous declaration of '_parse_addrblock' was here
 static int _parse_addrblock(struct rfc5444_reader_addrblock_entry *addr_entry,
            ^
rfc5444_reader.c:1034:1: error: conflicting types for '_handle_message'
 _handle_message(struct rfc5444_reader *parser,
 ^
rfc5444_reader.c:73:12: note: previous declaration of '_handle_message' was here
 static int _handle_message(struct rfc5444_reader *parser,
            ^
rfc5444_reader.c:67:12: warning: '_parse_tlvblock' used but never defined
 static int _parse_tlvblock(struct rfc5444_reader *parser,
            ^
rfc5444_reader.c:69:12: warning: '_schedule_tlvblock' used but never defined
 static int _schedule_tlvblock(struct rfc5444_reader_tlvblock_consumer *consumer,
            ^
rfc5444_reader.c:73:12: warning: '_handle_message' used but never defined
 static int _handle_message(struct rfc5444_reader *parser,
            ^
rfc5444_reader.c:1034:1: warning: '_handle_message' defined but not used [-Wunused-function]
 _handle_message(struct rfc5444_reader *parser,
 ^
/home/ludwig/RIOT/Makefile.base:33: recipe for target '/home/ludwig/RIOT/examples/olsr2/bin/msba2/oonf_rfc5444/rfc5444_reader.o' failed
make[3]: *** [/home/ludwig/RIOT/examples/olsr2/bin/msba2/oonf_rfc5444/rfc5444_reader.o] Error 1
make[3]: Leaving directory '/home/ludwig/RIOT/pkg/oonf_api/oonf_api/src-api/rfc5444'
Makefile:26: recipe for target 'all' failed
make[2]: *** [all] Error 1
make[2]: Leaving directory '/home/ludwig/RIOT/pkg/oonf_api/oonf_api'
Makefile:19: recipe for target 'all' failed
make[1]: *** [all] Error 2
make[1]: Leaving directory '/home/ludwig/RIOT/pkg/oonf_api'
/home/ludwig/RIOT/examples/olsr2/../../Makefile.include:143: recipe for target '/home/ludwig/RIOT/examples/olsr2/bin/msba2/oonf_api.a' failed
make: *** [/home/ludwig/RIOT/examples/olsr2/bin/msba2/oonf_api.a] Error 2

benpicco · 2014-05-12T16:43:47Z

Now this is strange, rfc5444_reader_handle_packet really is declared as returning int in rfc5444_reader.h but defined as returning enum rfc5444_result in rfc5444_reader.c.

Why did this previously work and why is it still working for native?

LudwigKnuepfer · 2014-05-12T19:59:17Z

Different toolchains..

benpicco · 2014-05-12T22:47:47Z

It's the same version of gcc

arm-none-eabi-gcc (4.8.2-14ubuntu1+6) 4.8.2

vs

gcc (Ubuntu 4.8.2-19ubuntu1) 4.8.2

LudwigKnuepfer · 2014-05-13T08:02:32Z

I meant architectures.. I mean - it's a shot in the dark anyway.. still - gcc has different code for different architectures, right?

LudwigKnuepfer · 2014-05-13T08:50:39Z

examples/olsr2/main.c

Compare #854

I'll change that when #854 is merged

I don't know what this comment is about, but #854 was merged.

Ok, but currently it's only implemented for native, stm32f1 and nrf51822. I'll drop the getpid() bit and replace it with cpuid_get(), but keep the sysconfig.id for the msba2.

benpicco · 2014-05-20T12:22:02Z

After #1210 it should build on the msba2 again

OlegHahm · 2014-06-17T21:45:07Z

A lot of warnings.

OlegHahm · 2014-06-22T17:24:02Z

And fails for MSP based platforms. Please fix or blacklist.

benpicco · 2014-07-07T19:43:14Z

Is there some way to find out why the build fails in travis?

OlegHahm · 2014-07-07T20:35:56Z

@Kijewski, you had a hint for that, but I cannot remember. Can you provide some information and maybe put it also in the wiki?

Kijewski · 2014-07-07T21:23:42Z

Prepend .travis.yml's section script with the test that you actually want to run:

    - make -C examples/olsr2 BOARD=msba2
    - make -C examples/olsr2 BOARD=native

OlegHahm · 2014-11-10T22:29:30Z

@benpicco, Can you fix the cppcheck warnings and add the extern C to the headers?

OlegHahm · 2014-11-18T18:08:26Z

@benpicco, ping.

OlegHahm · 2014-11-26T13:37:17Z

@benpicco, please rebase.

OlegHahm · 2015-01-13T18:51:49Z

Superseeded by #2294.

mehlis added sixlowpan-ietf and removed sixlowpan-ietf labels Feb 24, 2014

This was referenced Feb 24, 2014

[RFC] add oonf_api to pkg #630

Merged

add olsr2 example application des-testbed/RIOT-projects#98

Open

OlegHahm added the PR-award-nominee label Feb 24, 2014

LudwigKnuepfer reviewed May 13, 2014
View reviewed changes

OlegHahm added this to the Release NEXT MAJOR milestone May 16, 2014

OlegHahm assigned LudwigKnuepfer and emmanuelsearch and unassigned LudwigKnuepfer May 16, 2014

OlegHahm modified the milestones: FIX ME FIRST, Release NEXT MAJOR Jun 3, 2014

LudwigKnuepfer mentioned this pull request Jun 3, 2014

pkg/oonf_api: fix conflicting types on msba2 #1210

Merged

benpicco and others added 21 commits October 30, 2014 22:50

explicit check for NULL

4196a4d

style fixes

b2f6c02

use cpuid_get() when availiable

97eaa4a

slist cleanup part 1

dc53b2f

slist cleanup pt2

d482de8

slist cleanup pt3

9a04ead

adapt to net api change

c9b2f5a

routing.h -> routing_table.h

ad03d12

introduce new Makefile options

ad42160

silence warnings

cc25083

fix merge conflict

ca235e4

fix warnings in oonf_api

605f3b9

fix warning about printf type

68cc358

explain IF_ID

a8e2b50

make nbuf static

1efb0f5

prefix local_name

173bd7d

pkg: copy Kijewski's container_of implementation

82ef757

Add patches to fix the rest of the warnings

98cfc03

use msg_try_send

d6c6214

set address mode

d44adc9

should be +=

15b66ba

benpicco force-pushed the olsr2 branch from ab34277 to 15b66ba Compare October 30, 2014 21:57

Lotterleben mentioned this pull request Dec 17, 2014

ipv6_iface_set_routing_provider only usable if MODULE_RPL is defined #2206

Closed

OlegHahm added the Community: Hack'n'ACK candidate This PR is a candidate for review and discussion during one of RIOT's monthly Hack'n'ACK parties label Jan 6, 2015

OlegHahm mentioned this pull request Jan 13, 2015

routing: add OLSRv2 #2294

Closed

OlegHahm closed this Jan 13, 2015

Conversation

benpicco commented Feb 22, 2014

Uh oh!

benpicco commented May 12, 2014

Uh oh!

LudwigKnuepfer commented May 12, 2014

Uh oh!

LudwigKnuepfer commented May 12, 2014

Uh oh!

benpicco commented May 12, 2014

Uh oh!

LudwigKnuepfer commented May 12, 2014

Uh oh!

LudwigKnuepfer commented May 12, 2014

Uh oh!

benpicco commented May 12, 2014

Uh oh!

LudwigKnuepfer commented May 12, 2014

Uh oh!

benpicco commented May 12, 2014

Uh oh!

LudwigKnuepfer commented May 13, 2014

Uh oh!

LudwigKnuepfer May 13, 2014

Choose a reason for hiding this comment

Uh oh!

benpicco May 18, 2014

Choose a reason for hiding this comment

Uh oh!

Kijewski Sep 2, 2014

Choose a reason for hiding this comment

Uh oh!

benpicco Sep 2, 2014

Choose a reason for hiding this comment

Uh oh!

benpicco commented May 20, 2014

Uh oh!

OlegHahm commented Jun 17, 2014

Uh oh!

OlegHahm commented Jun 22, 2014

Uh oh!

benpicco commented Jul 7, 2014

Uh oh!

OlegHahm commented Jul 7, 2014

Uh oh!

Kijewski commented Jul 7, 2014

Uh oh!

OlegHahm commented Nov 10, 2014

Uh oh!

OlegHahm commented Nov 18, 2014

Uh oh!

OlegHahm commented Nov 26, 2014

Uh oh!

OlegHahm commented Jan 13, 2015

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

9 participants