Menu
▾
▴
linuxptp-users
|
From: David G. <dav...@po...> - 2013-08-11 22:21:57
Attachments:
signature.asc
|
Hi, I'm having trouble getting Linuxptp-1.3 to sync with my switch. switch: Netgear GS716Tv2 FW=5.4.1.13 with AVB (GS716TAV-10000S) license comp: Linux Mint 15 with Intel I210 NIC $ lspci |grep I210 02:00.0 Ethernet controller: Intel Corporation I210 Gigabit Network Connection (rev 03) $ ethtool -T eth2 Time stamping parameters for eth2: Capabilities: hardware-transmit (SOF_TIMESTAMPING_TX_HARDWARE) hardware-receive (SOF_TIMESTAMPING_RX_HARDWARE) hardware-raw-clock (SOF_TIMESTAMPING_RAW_HARDWARE) PTP Hardware Clock: 0 Hardware Transmit Timestamp Modes: off (HWTSTAMP_TX_OFF) on (HWTSTAMP_TX_ON) Hardware Receive Filter Modes: none (HWTSTAMP_FILTER_NONE) all (HWTSTAMP_FILTER_ALL) $ lsmod |grep ptp ptp 18621 1 igb pps_core 14080 1 ptp =============== $ sudo ptp4l -mP2sl 7 -i eth2 ptp4l[162693.120]: selected /dev/ptp0 as PTP clock ptp4l[162693.121]: PI servo: sync interval 1.000 kp 0.700 ki 0.300000 ptp4l[162693.139]: driver changed our HWTSTAMP options ptp4l[162693.139]: tx_type 1 not 1 ptp4l[162693.139]: rx_filter 1 not 12 ptp4l[162693.139]: port 1: INITIALIZING to LISTENING on INITIALIZE ptp4l[162693.139]: port 0: INITIALIZING to LISTENING on INITIALIZE ptp4l[162693.266]: port 1: setting asCapable ptp4l[162694.139]: port 1: delay timeout ptp4l[162695.139]: port 1: delay timeout ptp4l[162696.139]: port 1: delay timeout ptp4l[162697.139]: port 1: delay timeout ptp4l[162698.140]: port 1: delay timeout ptp4l[162699.139]: port 1: announce timeout ptp4l[162699.153]: driver changed our HWTSTAMP options ptp4l[162699.153]: tx_type 1 not 1 ptp4l[162699.153]: rx_filter 1 not 12 ptp4l[162699.153]: selected best master clock a0369f.fffe.1cdd3b ptp4l[162699.153]: port 1: delay timeout ptp4l[162700.153]: port 1: delay timeout ptp4l[162701.153]: port 1: delay timeout ptp4l[162702.153]: port 1: delay timeout ptp4l[162703.153]: port 1: delay timeout ptp4l[162704.153]: port 1: delay timeout ptp4l[162705.139]: port 1: announce timeout ptp4l[162705.156]: driver changed our HWTSTAMP options ptp4l[162705.156]: tx_type 1 not 1 ptp4l[162705.156]: rx_filter 1 not 12 ptp4l[162705.156]: selected best master clock a0369f.fffe.1cdd3b ptp4l[162705.156]: port 1: delay timeout ^Cptp4l[162705.651]: caught signal 2 The line 'ptp4l[162693.139]: rx_filter 1 not 12' I think means HWTSTAMP_FILTER_PTP_V2_L2_EVENT is not a capability. Wireshark shows the switch responding, but apparently ptp4l isn't hearing it. Help walk me through debugging this. Tnks. -- David Gravereaux <dav...@po...> |
|
From: David G. <dav...@po...> - 2013-08-11 23:26:32
Attachments:
signature.asc
|
On 08/11/2013 03:22 PM, David Gravereaux wrote: > $ lsmod |grep ptp > ptp 18621 1 igb > pps_core 14080 1 ptp $ modinfo igb |grep -e ^version: version: 4.1.2-k Looks like the latest is 4.3 from https://downloadcenter.intel.com/Detail_Desc.aspx?DwnldID=13663 Is there a recommended PPA I should follow? -- David Gravereaux <dav...@po...> |
|
From: David G. <dav...@po...> - 2013-08-12 00:14:58
Attachments:
signature.asc
wireshark_packets.gzip
|
I built and installed 4.3 of igb, but no joy. Here's a few packets, it helps. The switch responds, but it appears that ptp4l doesn't hear it. -- David Gravereaux <dav...@po...> |
|
From: David G. <dav...@po...> - 2013-08-12 04:33:38
Attachments:
signature.asc
wireshark_packets.gzip
|
Once I set the 'transportSpecific' option to 1 that matched the packets being sent to me from the switch ptp4l tried attempting a sync, but now ptp4l spits these malformed packets according to wireshark. And I'm back to the switch complaining about 'bad headers' again. -- David Gravereaux <dav...@po...> |
|
From: Richard C. <ric...@gm...> - 2013-08-12 07:55:46
|
On Sun, Aug 11, 2013 at 09:34:07PM -0700, David Gravereaux wrote: > Once I set the 'transportSpecific' option to 1 that matched the packets > being sent to me from the switch ptp4l tried attempting a sync, Can you please try it using the AV configuration file? ptp4l -f gPTP.cfg > but now > ptp4l spits these malformed packets according to wireshark. And I'm > back to the switch complaining about 'bad headers' again. I guess wireshark is noticing the missing follow up information TLV. Using the configuration file (or setting option follow_up_info) should cure this. Thanks, Richard |
|
From: David G. <dav...@po...> - 2013-08-12 08:40:19
Attachments:
signature.asc
|
On 08/12/2013 12:55 AM, Richard Cochran wrote: > On Sun, Aug 11, 2013 at 09:34:07PM -0700, David Gravereaux wrote: >> Once I set the 'transportSpecific' option to 1 that matched the packets >> being sent to me from the switch ptp4l tried attempting a sync, > > Can you please try it using the AV configuration file? > > ptp4l -f gPTP.cfg No luck with that. >> but now >> ptp4l spits these malformed packets according to wireshark. And I'm >> back to the switch complaining about 'bad headers' again. > > I guess wireshark is noticing the missing follow up information TLV. > Using the configuration file (or setting option follow_up_info) should > cure this. Yes and no. With follow_up_info=0, it attempts to sync. No attempt to sync without it. Guess I'll be doing more phone calls to Netgear tomorrow -- David Gravereaux <dav...@po...> |
|
From: Richard C. <ric...@gm...> - 2013-08-12 09:23:18
|
On Mon, Aug 12, 2013 at 01:40:51AM -0700, David Gravereaux wrote: > On 08/12/2013 12:55 AM, Richard Cochran wrote: > > > > I guess wireshark is noticing the missing follow up information TLV. > > Using the configuration file (or setting option follow_up_info) should > > cure this. > > > Yes and no. With follow_up_info=0, it attempts to sync. No attempt to > sync without it. Guess I'll be doing more phone calls to Netgear tomorrow But is wireshark happy now? Can you post the output from ptp4l? BTW, - If ptp4l is the master, it sends follow ups (plus info). - If ptp4l is slave, it consumes follow ups (plus info). Thanks, Richard |
|
From: David G. <dav...@po...> - 2013-08-12 09:36:20
Attachments:
signature.asc
|
On 08/12/2013 02:22 AM, Richard Cochran wrote: > On Mon, Aug 12, 2013 at 01:40:51AM -0700, David Gravereaux wrote: >> On 08/12/2013 12:55 AM, Richard Cochran wrote: >>> >>> I guess wireshark is noticing the missing follow up information TLV. >>> Using the configuration file (or setting option follow_up_info) should >>> cure this. >> >> >> Yes and no. With follow_up_info=0, it attempts to sync. No attempt to >> sync without it. Guess I'll be doing more phone calls to Netgear tomorrow > > But is wireshark happy now? Never gets to a Follow_Up_Message (0x08) so I couldn't say. $ sudo ptp4l -i eth2 -f gPTP.cfg [sudo] password for davygrvy: ptp4l[202911.008]: PI servo: sync interval 1.000 kp 0.700 ki 0.300000 ptp4l[202911.016]: driver changed our HWTSTAMP options ptp4l[202911.016]: tx_type 1 not 1 ptp4l[202911.016]: rx_filter 1 not 12 ptp4l[202911.016]: port 1: INITIALIZING to LISTENING on INITIALIZE ptp4l[202911.016]: port 0: INITIALIZING to LISTENING on INITIALIZE ptp4l[202911.221]: port 1: setting asCapable ptp4l[202911.221]: port 1: peer port id set to 0026f2.fffe.f25aa0-1 <- Netgear 802.1AS clock. ptp4l[202912.016]: port 1: delay timeout ptp4l[202913.016]: port 1: delay timeout ptp4l[202914.016]: port 1: delay timeout ptp4l[202915.016]: port 1: delay timeout ptp4l[202916.016]: port 1: delay timeout ptp4l[202917.016]: port 1: announce timeout ptp4l[202917.016]: port 1: LISTENING to MASTER on ANNOUNCE_RECEIPT_TIMEOUT_EXPIRES ptp4l[202917.016]: selected best master clock a0369f.fffe.1cdd3b ptp4l[202917.016]: assuming the grand master role <- Netgear ignores this election ptp4l[202917.016]: port 1: delay timeout ptp4l[202917.141]: port 1: master sync timeout ptp4l[202917.266]: port 1: master sync timeout ptp4l[202917.391]: port 1: master sync timeout ptp4l[202917.516]: port 1: master sync timeout ptp4l[202917.641]: port 1: master sync timeout ptp4l[202917.766]: port 1: master sync timeout ptp4l[202917.891]: port 1: master sync timeout ptp4l[202918.016]: port 1: delay timeout ptp4l[202918.018]: port 1: master sync timeout ptp4l[202918.143]: port 1: master sync timeout ptp4l[202918.268]: port 1: master sync timeout ptp4l[202918.393]: port 1: master sync timeout ptp4l[202918.518]: port 1: master sync timeout ptp4l[202918.643]: port 1: master sync timeout ptp4l[202918.768]: port 1: master sync timeout ptp4l[202918.893]: port 1: master sync timeout ptp4l[202919.016]: port 1: master tx announce timeout ptp4l[202919.016]: port 1: delay timeout ptp4l[202919.018]: port 1: master sync timeout ptp4l[202919.143]: port 1: master sync timeout ptp4l[202919.268]: port 1: master sync timeout ptp4l[202919.393]: port 1: master sync timeout ptp4l[202919.518]: port 1: master sync timeout ptp4l[202919.643]: port 1: master sync timeout ptp4l[202919.768]: port 1: master sync timeout ptp4l[202919.893]: port 1: master sync timeout ptp4l[202920.017]: port 1: delay timeout ptp4l[202920.019]: port 1: master sync timeout ptp4l[202920.144]: port 1: master sync timeout ptp4l[202920.269]: port 1: master sync timeout ptp4l[202920.394]: port 1: master sync timeout ptp4l[202920.519]: port 1: master sync timeout ptp4l[202920.644]: port 1: master sync timeout ptp4l[202920.769]: port 1: master sync timeout ptp4l[202920.894]: port 1: master sync timeout ptp4l[202921.016]: port 1: master tx announce timeout ptp4l[202921.017]: port 1: delay timeout ptp4l[202921.019]: port 1: master sync timeout ptp4l[202921.144]: port 1: master sync timeout ptp4l[202921.269]: port 1: master sync timeout ptp4l[202921.394]: port 1: master sync timeout ptp4l[202921.519]: port 1: master sync timeout ptp4l[202921.644]: port 1: master sync timeout ptp4l[202921.769]: port 1: master sync timeout ptp4l[202921.894]: port 1: master sync timeout ptp4l[202922.017]: port 1: delay timeout ptp4l[202922.019]: port 1: master sync timeout ptp4l[202922.145]: port 1: master sync timeout ptp4l[202922.270]: port 1: master sync timeout ptp4l[202922.395]: port 1: master sync timeout ptp4l[202922.520]: port 1: master sync timeout ptp4l[202922.645]: port 1: master sync timeout ptp4l[202922.770]: port 1: master sync timeout ptp4l[202922.895]: port 1: master sync timeout ptp4l[202923.016]: port 1: master tx announce timeout ptp4l[202923.017]: port 1: delay timeout ptp4l[202923.020]: port 1: master sync timeout ptp4l[202923.145]: port 1: master sync timeout ptp4l[202923.270]: port 1: master sync timeout ptp4l[202923.395]: port 1: master sync timeout ptp4l[202923.520]: port 1: master sync timeout ptp4l[202923.645]: port 1: master sync timeout ptp4l[202923.770]: port 1: master sync timeout ptp4l[202923.895]: port 1: master sync timeout ptp4l[202924.017]: port 1: delay timeout ptp4l[202924.020]: port 1: master sync timeout ptp4l[202924.146]: port 1: master sync timeout ptp4l[202924.271]: port 1: master sync timeout ptp4l[202924.396]: port 1: master sync timeout ptp4l[202924.521]: port 1: master sync timeout ptp4l[202924.646]: port 1: master sync timeout ptp4l[202924.771]: port 1: master sync timeout ptp4l[202924.896]: port 1: master sync timeout ptp4l[202925.016]: port 1: master tx announce timeout ptp4l[202925.017]: port 1: delay timeout ptp4l[202925.021]: port 1: master sync timeout ptp4l[202925.146]: port 1: master sync timeout ptp4l[202925.271]: port 1: master sync timeout ptp4l[202925.396]: port 1: master sync timeout ptp4l[202925.521]: port 1: master sync timeout ptp4l[202925.646]: port 1: master sync timeout ptp4l[202925.771]: port 1: master sync timeout ptp4l[202925.896]: port 1: master sync timeout ptp4l[202926.017]: port 1: delay timeout ptp4l[202926.021]: port 1: master sync timeout ptp4l[202926.146]: port 1: master sync timeout ptp4l[202926.272]: port 1: master sync timeout ptp4l[202926.397]: port 1: master sync timeout ptp4l[202926.522]: port 1: master sync timeout ptp4l[202926.647]: port 1: master sync timeout ptp4l[202926.772]: port 1: master sync timeout ptp4l[202926.897]: port 1: master sync timeout ptp4l[202927.016]: port 1: master tx announce timeout ptp4l[202927.017]: port 1: delay timeout ptp4l[202927.022]: port 1: master sync timeout ptp4l[202927.147]: port 1: master sync timeout ptp4l[202927.272]: port 1: master sync timeout ptp4l[202927.397]: port 1: master sync timeout ptp4l[202927.522]: port 1: master sync timeout ptp4l[202927.647]: port 1: master sync timeout ptp4l[202927.772]: port 1: master sync timeout ptp4l[202927.897]: port 1: master sync timeout ptp4l[202928.017]: port 1: delay timeout ptp4l[202928.022]: port 1: master sync timeout ptp4l[202928.147]: port 1: master sync timeout ptp4l[202928.273]: port 1: master sync timeout ptp4l[202928.398]: port 1: master sync timeout ptp4l[202928.523]: port 1: master sync timeout ^Cptp4l[202928.647]: caught signal 2 > Can you post the output from ptp4l? > > BTW, > > - If ptp4l is the master, it sends follow ups (plus info). > - If ptp4l is slave, it consumes follow ups (plus info). I want ptp4l to become master, which is why I have the priority set lower than 248 and slaveOnly=0. Now if the Netgear can just talk to it right... more phone calls to them in the AM. Thanks for the assist. -- David Gravereaux <dav...@po...> |
|
From: Richard C. <ric...@gm...> - 2013-08-12 09:51:41
|
On Mon, Aug 12, 2013 at 02:36:57AM -0700, David Gravereaux wrote: > > I want ptp4l to become master, which is why I have the priority set > lower than 248 and slaveOnly=0. Now if the Netgear can just talk to it > right... more phone calls to them in the AM. It looks like ptp4l thinks the port is asCapable. That means that the P2P mechanism to the Netgear is working (from ptp4l's perspective). Does the Netgear also show that the port is asCapable, like in a web interface? Does ptp4l with slaveOnly=1 synchronize to the Netgear? Thanks, Richard |
|
From: David G. <dav...@po...> - 2013-08-12 10:18:59
Attachments:
signature.asc
|
On 08/12/2013 02:51 AM, Richard Cochran wrote: > Does the Netgear also show that the port is asCapable, like in a web > interface? From the web interface it says it isn't capable. The web interface also says my two Xmos AVB dev boards are also not capable. > Does ptp4l with slaveOnly=1 synchronize to the Netgear? Nothing synchronizes with it.. ever. -- David Gravereaux <dav...@po...> |
|
From: Richard C. <ric...@gm...> - 2013-08-12 10:39:41
|
On Mon, Aug 12, 2013 at 03:19:30AM -0700, David Gravereaux wrote: > On 08/12/2013 02:51 AM, Richard Cochran wrote: > > Does the Netgear also show that the port is asCapable, like in a web > > interface? > > From the web interface it says it isn't capable. The web interface also > says my two Xmos AVB dev boards are also not capable. Okay, then it looks like Netgear has an issue. FWIW, I tested ptp4l together with a LabX Titanium 411 bridge and the XMOS AVB Demo Board, and it works fine with those. Thanks, Richard |
|
From: David G. <dav...@po...> - 2013-08-12 10:51:22
Attachments:
signature.asc
|
On 08/12/2013 03:39 AM, Richard Cochran wrote: > FWIW, I tested ptp4l together with a LabX Titanium 411 bridge and the > XMOS AVB Demo Board, and it works fine with those. I believe it! This hobbyist can't drop $1800. I'm thinking a bunch of the 5-port DSP4YOU ones @ $100 each. Or maybe the Extreme 24-port XA440 for $1K if I can swing it -- David Gravereaux <dav...@po...> |
|
From: Keller, J. E <jac...@in...> - 2013-08-12 17:47:04
|
On Mon, 2013-08-12 at 03:51 -0700, David Gravereaux wrote: > On 08/12/2013 03:39 AM, Richard Cochran wrote: > > > FWIW, I tested ptp4l together with a LabX Titanium 411 bridge and the > > XMOS AVB Demo Board, and it works fine with those. > > I believe it! This hobbyist can't drop $1800. I'm thinking a bunch of > the 5-port DSP4YOU ones @ $100 each. Or maybe the Extreme 24-port XA440 > for $1K if I can swing it This may seem dumb, but have you checked to ensure that the firewall was disabled or the correct PTP ports were opened? That is a pretty common issue with symptoms like this. - Jake |
|
From: David G. <dav...@po...> - 2013-08-12 18:40:03
Attachments:
signature.asc
|
On 08/12/2013 10:46 AM, Keller, Jacob E wrote: > This may seem dumb, but have you checked to ensure that the firewall was > disabled or the correct PTP ports were opened? > > That is a pretty common issue with symptoms like this. Thanks Jake. Not a dumb question. It's all off. The 802.1Qav part of the firmware for this switch has it's roots from a project started by Harman/BSS Audio. http://www.bssaudio.com/en-US/products/gs724t Although they say you can't modify a stock switch, that's incorrect, the current GS724Tv3(-300) has the same guts as what Harman did. The Harman is a modified v2 hardware which became the v3. According to the helpful persons I got in touch with, their work doesn't properly support the ratified 802.1Q-2011, but a pre-release. Now that their work is owned by Netgear with regards to the EAV license, they just haven't finished it. The last few days playing around with wireshark has pretty much confirmed it. -- David Gravereaux <dav...@po...> |
×
Want the latest updates on software, tech news, and AI?
Get latest updates about software, tech news, and AI from SourceForge directly in your inbox once a month.
Thanks for helping keep SourceForge clean.
X