Ethernet Products
Determine ramifications of Intel® Ethernet products and technologies
4931 Discussions

PTP not working for 10GBASE-KR interface with E823-L Ethernet Controller and ice 1.9.11 driver

kelly_8613
Beginner
865 Views

I am trying to run PTP tests with HW timestamping but receiving the errors below.  The test is running on the 10GBASE-KR control plane interface from the embedded NAC Quad 1 on the Xeon D 1700 Intel Ice Lake SoC processor.

 

Information on the adapter and driver:

# lspci -v | grep "Ethernet" -A 1
02:00.0 Ethernet controller: Intel Corporation I210 Gigabit Network Connection (rev 03)
DeviceName: Onboard LAN
--
f4:00.0 Ethernet controller: Intel Corporation Ethernet Connection E823-L for backplane
DeviceName: Onboard LAN
--
f4:00.1 Ethernet controller: Intel Corporation Ethernet Connection E823-L for backplane
DeviceName: Onboard LAN

 

# ethtool -i eno3
driver: cw_ice
version: 1.9.11
firmware-version: 2.28 0x8000fd37 1.3200.0
expansion-rom-version:
bus-info: 0000:f4:00.1
supports-statistics: yes
supports-test: yes
supports-eeprom-access: yes
supports-register-dump: yes
supports-priv-flags: yes

# ethtool -i eno3
driver: ice
version: 4.18.0-372.9.1.el8.x86_64
firmware-version: 2.28 0x8000fd37 1.3200.0
expansion-rom-version:
bus-info: 0000:f4:00.1
supports-statistics: yes
supports-test: yes
supports-eeprom-access: yes
supports-register-dump: yes
supports-priv-flags: yes

 

* We have modified the ice 1.9.11 driver and renamed it to cw_ice.  I also tried testing with the native ice driver from Almalinux 8.6 and saw the same results.

 

Failing PTP log:

# ptp4l -f /etc/ptp4l.conf -i eno3 -m
ptp4l[220.937]: selected /dev/ptp0 as PTP clock
ptp4l[220.942]: port 1: INITIALIZING to LISTENING on INIT_COMPLETE
ptp4l[220.942]: port 0: INITIALIZING to LISTENING on INIT_COMPLETE
ptp4l[220.942]: port 0: INITIALIZING to LISTENING on INIT_COMPLETE
ptp4l[227.492]: port 1: LISTENING to MASTER on ANNOUNCE_RECEIPT_TIMEOUT_EXPIRES
ptp4l[227.492]: selected local clock 000000.fffe.000101 as best master
ptp4l[227.492]: port 1: assuming the grand master role
ptp4l[238.502]: timed out while polling for tx timestamp
ptp4l[238.502]: increasing tx_timestamp_timeout may correct this issue, but it is likely caused by a driver bug
ptp4l[238.502]: port 1: send sync failed
ptp4l[238.502]: port 1: MASTER to FAULTY on FAULT_DETECTED (FT_UNSPECIFIED)
ptp4l[254.502]: port 1: FAULTY to LISTENING on INIT_COMPLETE
ptp4l[261.299]: port 1: LISTENING to MASTER on ANNOUNCE_RECEIPT_TIMEOUT_EXPIRES
ptp4l[261.299]: port 1: assuming the grand master role
ptp4l[272.309]: timed out while polling for tx timestamp
ptp4l[272.309]: increasing tx_timestamp_timeout may correct this issue, but it is likely caused by a driver bug
ptp4l[272.309]: port 1: send sync failed
ptp4l[272.309]: port 1: MASTER to FAULTY on FAULT_DETECTED (FT_UNSPECIFIED)
ptp4l[288.309]: port 1: FAULTY to LISTENING on INIT_COMPLETE
ptp4l[295.591]: port 1: LISTENING to MASTER on ANNOUNCE_RECEIPT_TIMEOUT_EXPIRES
ptp4l[295.591]: port 1: assuming the grand master role
ptp4l[306.601]: timed out while polling for tx timestamp
ptp4l[306.601]: increasing tx_timestamp_timeout may correct this issue, but it is likely caused by a driver bug
ptp4l[306.601]: port 1: send sync failed
ptp4l[306.601]: port 1: MASTER to FAULTY on FAULT_DETECTED (FT_UNSPECIFIED)
ptp4l[322.602]: port 1: FAULTY to LISTENING on INIT_COMPLETE
ptp4l[330.247]: port 1: LISTENING to MASTER on ANNOUNCE_RECEIPT_TIMEOUT_EXPIRES
ptp4l[330.247]: port 1: assuming the grand master role

 

 

I have additional test logs and tshark logs if needed. Thanks.

0 Kudos
3 Replies
B_Y
Employee
835 Views

Hi Kelly,

 

Thank you for posting in Intel Ethernet Communities.

You may give a try to install with following drivers:

https://www.intel.com/content/www/us/en/download/19630/intel-network-adapter-driver-for-e810-series-devices-under-linux.html

https://sourceforge.net/projects/e1000/files/ice%20stable/


Kindly share NIC pba number for us to validate the controller:

https://www.intel.com/content/www/us/en/support/articles/000007022/ethernet-products/500-series-network-adapters-up-to-10gbe.html


Best regards,

BY_Intel

Intel Customer Support


0 Kudos
kelly_8613
Beginner
802 Views

Hello,

 

Thanks for the recommendation, I will try the 1.11.14 ice driver and see if that helps. 

 

As for the NIC pba #, we don't have a NIC but a NAC that is in the SoC.  I ran Intel's SSU script and extracted the relevant ethernet information (attached), does this help?  If not, please let me know how I can gather information on the NAC device.

 

Thank you,

Kelly

 

 

-- Update: I tried the PTP tests with the ice 1.11.14 driver and saw similar failures.

 

[root@vpx3-484 ~]# ethtool -i eno3
driver: ice
version: 1.11.14
firmware-version: 2.28 0x8000fd37 1.3200.0
expansion-rom-version:
bus-info: 0000:f4:00.1
supports-statistics: yes
supports-test: yes
supports-eeprom-access: yes
supports-register-dump: yes
supports-priv-flags: yes

 

[root@vpx3-484 ~]# ptp4l -i eno3 -m

ptp4l[2091.650]: selected /dev/ptp0 as PTP clock

ptp4l[2091.656]: port 1: INITIALIZING to LISTENING on INIT_COMPLETE

ptp4l[2091.656]: port 0: INITIALIZING to LISTENING on INIT_COMPLETE

ptp4l[2091.656]: port 0: INITIALIZING to LISTENING on INIT_COMPLETE

ptp4l[2097.852]: port 1: LISTENING to MASTER on ANNOUNCE_RECEIPT_TIMEOUT_EXPIRES

ptp4l[2097.852]: selected local clock 000000.fffe.000101 as best master

ptp4l[2097.852]: port 1: assuming the grand master role

ptp4l[2098.862]: timed out while polling for tx timestamp

ptp4l[2098.862]: increasing tx_timestamp_timeout may correct this issue, but it is likely caused by a driver bug

ptp4l[2098.862]: port 1: send sync failed

ptp4l[2098.862]: port 1: MASTER to FAULTY on FAULT_DETECTED (FT_UNSPECIFIED)

ptp4l[2114.863]: port 1: FAULTY to LISTENING on INIT_COMPLETE

ptp4l[2122.390]: port 1: LISTENING to MASTER on ANNOUNCE_RECEIPT_TIMEOUT_EXPIRES

ptp4l[2122.390]: port 1: assuming the grand master role

ptp4l[2123.400]: timed out while polling for tx timestamp

ptp4l[2123.400]: increasing tx_timestamp_timeout may correct this issue, but it is likely caused by a driver bug

ptp4l[2123.400]: port 1: send sync failed

0 Kudos
Irwan_Intel
Moderator
749 Views

Hi kelly_8613


Sorry for late respond,based on product information, this requires assistance from the embedded communities. You may post this question at: https://community.intel.com/t5/Embedded-Products/ct-p/embedded-products


For any developing question you may goes to the RDC website:

https://www.intel.com/content/www/us/en/resources-documentation/developer.html


This thread will be close and no longer monitored. Thank you.


Regards,


Irwan_Intel

Intel Customer Support


0 Kudos
Reply