On 24.07.19 10:45, danwe wrote: > RTnet starts at each boot with script "S51rtnet". I can also stop and start > rtnet by myself. See the output: I have to start rtnet with the script > S51rtnet. > > # /usr/sbin/rtnet -cf /etc/rtnet.conf stop > [ 328.192316] RTmac/TDMA: unloaded > [ 328.200527] RTmac: unloaded > [ 328.205608] RTcfg: unloaded > # /usr/sbin/rtnet -cf /etc/rtnet.conf -v start > Turning on verbose mode > insmod: can't insert '/lib/modules/3.8.13-xeno/kernel/rtnet.ko': File exists
Ah, you are on legacy, out-of-tree RTnet! I forgot that RTnet only moved into Xenomai for version 3.0. For 2.x, you are with the old repo that wasn't touched since 2013. In this particular case, the script fails early because of "insmod <existing-module>", rather than using "modprobe" like in Xenomai upstream. You may run in many more such issues that have been fixed since then, and you will be on your own there now. You really need to update to Xenomai 3. Jan > # /etc/init.d/S51rtnet start> Starting RTnet: rtnet rt_smsc rt_davinci_mdio > rt_ticpsw rtpacket rtipv4 rtudp > rtmac [ > > 349.250660] RTmac: init realtime media access > control > rtcfg [ 350.275735] RTcfg: init real-time configuration distribution protocol > tdma [ 351.301744] RTmac/TDMA: init time division multiple access control > mechanism > OK > [ 352.322560] cpsw 4a100000.ethernet: initializing cpsw version 1.12 (0) > [ 352.332087] cpsw 4a100000.ethernet: phy found : id is : 0x7c0f1 > [ 352.338510] libphy: PHY 4a101000.mdio:01 not found > [ 352.343569] cpsw 4a100000.ethernet: phy 4a101000.mdio:01 not found on > slave 1 > # rtifconfig[ 354.338178] libphy: 4a101000.mdio:00 - Link is Up - 100/Full > > Am Mi., 24. Juli 2019 um 10:30 Uhr schrieb Jan Kiszka <[email protected] > <mailto:[email protected]>>: > > On 24.07.19 10:15, danwe wrote: > > Today I have tried TDMA on rtnet with all the necessary modules/drivers. > > Unfortunately I do not see any different output on rtping. It is the > same > as before. > > This is the output: > > > > # rtping 192.168.127.11 > > Real-time PING 192.168.127.11 56(84) bytes of data. > > 64 bytes from 192.168.127.11 <http://192.168.127.11>: icmp_seq=1 > time=226.7 us > > 64 bytes from 192.168.127.11 <http://192.168.127.11>: icmp_seq=2 > time=142.2 us > > 64 bytes from 192.168.127.11 <http://192.168.127.11>: icmp_seq=3 > time=151.1 us > > 64 bytes from 192.168.127.11 <http://192.168.127.11>: icmp_seq=4 > time=128.8 us > > 64 bytes from 192.168.127.11 <http://192.168.127.11>: icmp_seq=5 > time=135.1 us > > 64 bytes from 192.168.127.11 <http://192.168.127.11>: icmp_seq=6 > time=122.9 us > > > > --- 192.168.127.11 rtping statistics --- > > 6 packets transmitted, 6 received, 0% packet loss > > worst case rtt = 226.7 us > > > > And this are the modules which are loaded: > > > > # lsmod > > Module Size Used by Tainted: G > > tdma 18805 0 > > rtcfg 45592 0 > > rtmac 7219 1 tdma > > rtudp 7640 0 > > rtipv4 18699 2 rtcfg,rtudp > > rtpacket 4179 0 > > rt_ticpsw 26115 0 > > rt_davinci_mdio 4570 0 > > rt_smsc 2165 0 > > rtnet 28479 7 > tdma,rtcfg,rtmac,rtudp,rtipv4,rtpacket,rt_ticpsw > > omap_rng 4174 0 > > > > I would like to ask you if you could take a look at my settings for the > master > > in rtnet.conf and tdma.conf if everything is correct? For slave I have > just > > change "master" to "slave". Should I start realtime loopback device and > > capturing interface? I do not have started the driver for realtime > loopback > > device. Is that a problem? > > I don't recall all details anymore, but I do not spot any obvious mistake. > > What did you call to start rtnet on both sides? What was the console > output? Did > you try adding -v to the start script to see more? > > Jan > > -- > Siemens AG, Corporate Technology, CT RDA IOT SES-DE > Corporate Competence Center Embedded Linux > -- Siemens AG, Corporate Technology, CT RDA IOT SES-DE Corporate Competence Center Embedded Linux
