[etherlab-users] ecrt function

[Chun-Yi Huang]

Hi:
  Recently we use ethercat master on a rtai os to control a CNC
machine. There are 4 (devices) motors in it. We face a problem that
the execution time of some ecrt functions are too long to complete the
cycle.
  The max time recorded of those functions,
ecrt_domain_process
ecrt_domain_receive
ecrt_domain_queue
ecrt_domain_send,
is around 0.4 ms individually.
  So, what is the usual time of doing those functions? We are concerned
that something is not normal.
  We check the /var/log/messages in which the log is this,
Dec  7 20:57:11 wcut kernel: [    8.592727] EtherCAT: Master driver 1.5.2
2eff7c993a63
Dec  7 20:57:11 wcut kernel: [    8.592924] EtherCAT: 1 master waiting for
devices.
Dec  7 20:57:11 wcut kernel: [    8.641659] e1000e 0000:03:00.0: PCI INT A
disabled
Dec  7 20:57:11 wcut kernel: [    8.689607] e1000e 0000:00:19.0: PCI INT A
disabled
Dec  7 20:57:11 wcut kernel: [    8.700943] ec_e1000e: Ethercat-capable
Intel(R) PRO/1000 Network Driver - 1.0.2-k2(ethercat)
Dec  7 20:57:11 wcut kernel: [    8.700953] ec_e1000e: Copyright (c)
1999-2008 Intel Corporation.
Dec  7 20:57:11 wcut kernel: [    8.701033] ec_e1000e 0000:00:19.0: PCI INT
A -> GSI 23 (level, low) -> IRQ 23
Dec  7 20:57:12 wcut kernel: [    8.809526] ec_0000:00:19.0: eth0: (PCI
Express:2.5GB/s:Width x1) 00:0b:ab:80:a0:d1
Dec  7 20:57:12 wcut kernel: [    8.809536] ec_0000:00:19.0: eth0: Intel(R)
PRO/1000 Network Connection
Dec  7 20:57:12 wcut kernel: [    8.809579] ec_0000:00:19.0: eth0: MAC: 6,
PHY: 8, PBA No: ffffff-0ff
Dec  7 20:57:12 wcut kernel: [    8.809673] ec_e1000e 0000:03:00.0: PCI INT
A -> GSI 16 (level, low) -> IRQ 16
Dec  7 20:57:12 wcut kernel: [    8.840766] udev[387]: renamed network
interface eth0 to eth2
Dec  7 20:57:12 wcut kernel: [    8.928468] EtherCAT: Accepting
00:0B:AB:80:A0:D2 as main device for master 0.
Dec  7 20:57:12 wcut kernel: [    8.949472] ADDRCONF(NETDEV_UP): eth2: link
is not ready
Dec  7 20:57:12 wcut kernel: [    9.001048] ec_0000:03:00.0: ecm0: MSI
interrupt test failed!
Dec  7 20:57:12 wcut kernel: [    9.001091] ec_0000:03:00.0: ecm0: MSI
interrupt test failed, using legacy interrupt.
Dec  7 20:57:12 wcut kernel: [    9.001103] EtherCAT 0: Starting
EtherCAT-IDLE thread.
Dec  7 20:57:12 wcut kernel: [    9.001191] ec_0000:03:00.0: ecm0: (PCI
Express:2.5GB/s:Width x1) 00:0b:ab:80:a0:d2
Dec  7 20:57:12 wcut kernel: [    9.001199] ec_0000:03:00.0: ecm0: Intel(R)
PRO/1000 Network Connection
Dec  7 20:57:12 wcut kernel: [    9.001294] ec_0000:03:00.0: ecm0: MAC: 4,
PHY: 8, PBA No: ffffff-0ff
Dec  7 20:57:14 wcut kernel: [   11.394277] e1000e: eth2 NIC Link is Up
1000 Mbps Full Duplex, Flow Control: RX/TX
Dec  7 20:57:14 wcut kernel: [   11.394595] ADDRCONF(NETDEV_CHANGE): eth2:
link becomes ready
Dec  7 20:57:16 wcut kernel: [   13.001037] e1000e: ecm0 NIC Link is Up
1000 Mbps Full Duplex, Flow Control: RX/TX
Dec  7 20:57:16 wcut kernel: [   13.001049] EtherCAT 0: Link state of ecm0
changed to UP.
Dec  7 20:57:16 wcut kernel: [   13.009075] EtherCAT 0: 4 slave(s)
responding on main device.
Dec  7 20:57:16 wcut kernel: [   13.009086] EtherCAT 0: Slave states on
main device: INIT.
Dec  7 20:57:16 wcut kernel: [   13.009164] EtherCAT 0: Scanning bus.
Dec  7 20:57:16 wcut kernel: [   13.180771] EtherCAT 0: Bus scanning
completed in 172 ms.
Dec  7 20:57:16 wcut kernel: [   13.180780] EtherCAT 0: Using slave 0 as DC
reference clock.
Dec  7 20:57:16 wcut kernel: [   13.185049] EtherCAT 0: Slave states on
main device: PREOP.

best Regards
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.etherlab.org/pipermail/etherlab-users/attachments/20141207/1db2fcc5/attachment-0002.htm>