AM6442: PRG1 (ICSSG) eth is not stable

jacky lin

Part Number: AM6442

Tool/software:

Hello engineer!

We follow the PROC101 to design ower board. But we fount the eth from PRG1 have some problem. As described in the attached log.

Fullscreen eth error.txt Download

root@AM64x-Tronlong:~# ifconfig
docker0: flags=4099<UP,BROADCAST,MULTICAST>  mtu 1500  metric 1
        inet 172.17.0.1  netmask 255.255.0.0  broadcast 172.17.255.255
        ether 02:42:70:76:7e:19  txqueuelen 0  (Ethernet)
        RX packets 0  bytes 0 (0.0 B)
        RX errors 0  dropped 0  overruns 0  frame 0
        TX packets 0  bytes 0 (0.0 B)
        TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0

eth0: flags=4099<UP,BROADCAST,MULTICAST>  mtu 1500  metric 1
        inet6 fe80::2ab5:e8ff:fee3:80f4  prefixlen 64  scopeid 0x20<link>
        ether 28:b5:e8:e3:80:f4  txqueuelen 1000  (Ethernet)
        RX packets 0  bytes 0 (0.0 B)
        RX errors 0  dropped 0  overruns 0  frame 0
        TX packets 41  bytes 7410 (7.2 KiB)
        TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0

eth1: flags=4163<UP,BROADCAST,RUNNING,MULTICAST>  mtu 1500  metric 1
        inet 192.168.1.123  netmask 255.255.255.0  broadcast 192.168.1.255
        inet6 fe80::b04a:22ff:fede:d0eb  prefixlen 64  scopeid 0x20<link>
        ether b2:4a:22:de:d0:eb  txqueuelen 1000  (Ethernet)
        RX packets 26  bytes 1560 (1.5 KiB)
        RX errors 0  dropped 25  overruns 0  frame 0
        TX packets 24  bytes 4436 (4.3 KiB)
        TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0

eth2: flags=4099<UP,BROADCAST,MULTICAST>  mtu 1500  metric 1
        inet6 fe80::8864:f2ff:fed6:c173  prefixlen 64  scopeid 0x20<link>
        ether 8a:64:f2:d6:c1:73  txqueuelen 1000  (Ethernet)
        RX packets 0  bytes 0 (0.0 B)
        RX errors 0  dropped 0  overruns 0  frame 0
        TX packets 42  bytes 6318 (6.1 KiB)
        TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0

lo: flags=73<UP,LOOPBACK,RUNNING>  mtu 65536  metric 1
        inet 127.0.0.1  netmask 255.0.0.0
        inet6 ::1  prefixlen 128  scopeid 0x10<host>
        loop  txqueuelen 1000  (Local Loopback)
        RX packets 82  bytes 6220 (6.0 KiB)
        RX errors 0  dropped 0  overruns 0  frame 0
        TX packets 82  bytes 6220 (6.0 KiB)
        TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0

root@AM64x-Tronlong:~# 
root@AM64x-Tronlong:~# 
root@AM64x-Tronlong:~# 
root@AM64x-Tronlong:~# [  307.270120] icssg-prueth icssg1-eth eth1: Link is Down
[  310.214461] icssg-prueth icssg1-eth eth2: Link is Up - 1Gbps/Full - flow control off

root@AM64x-Tronlong:~# 
root@AM64x-Tronlong:~# 
root@AM64x-Tronlong:~# 
root@AM64x-Tronlong:~# 
root@AM64x-Tronlong:~# 
root@AM64x-Tronlong:~# [  315.973623] ------------[ cut here ]------------
[  315.973713] NETDEV WATCHDOG: eth2 (icssg-prueth): transmit queue 0 timed out
[  315.973793] WARNING: CPU: 1 PID: 0 at net/sched/sch_generic.c:467 dev_watchdog+0x350/0x358
[  315.973821] Modules linked in: xt_conntrack xt_MASQUERADE xt_addrtype iptable_filter iptable_nat nf_nat nf_conntrack nf_defrag_ipv6 nf_defrag_ipv4 libcrc32c ip_tables x_tables br_netfilter bridge stp llc overlay xfrm_user xfrm_algo xhci_plat_hcd xhci_hcd cdns3 udc_core roles usbcore usb_common rpmsg_char irq_pruss_intc ti_am335x_adc kfifo_buf crct10dif_ce icssg_prueth pru_rproc icss_iep ti_k3_r5_remoteproc pci_endpoint_test cdns3_ti virtio_rpmsg_bus rti_wdt ti_k3_m4_remoteproc sa2ul pruss sha512_generic authenc ti_am335x_tscadc at24 rtc_ds1307 m_can_platform m_can can_dev optee_rng rng_core sch_fq_codel ipv6
[  315.973982] CPU: 1 PID: 0 Comm: swapper/1 Not tainted 5.10.168-rt83-g2cc603147413 #1
[  315.973991] Hardware name: TL64x-EVM (DT)
[  315.973995] pstate: 20000005 (nzCv daif -PAN -UAO -TCO BTYPE=--)
[  315.974004] pc : dev_watchdog+0x350/0x358
[  315.974012] lr : dev_watchdog+0x350/0x358
[  315.974020] sp : ffff800011263db0
[  315.974023] x29: ffff800011263db0 x28: ffff000003c61440 
[  315.974032] x27: 0000000000000004 x26: 0000000000000140 
[  315.974040] x25: 00000000ffffffff x24: 0000000000000001 
[  315.974048] x23: ffff000004eec3dc x22: ffff000004eec000 
[  315.974056] x21: ffff000004eec480 x20: ffff8000110b6000 
[  315.974064] x19: 0000000000000000 x18: 0000000000000000 
[  315.974071] x17: 0000000000000000 x16: 0000000000000000 
[  315.974078] x15: ffff0000000dcde0 x14: ffffffffffffffff 
[  315.974086] x13: ffff8000111e7fbc x12: ffff8000111e7faf 
[  315.974095] x11: 0000000000000040 x10: ffff8000110d9168 
[  315.974102] x9 : 00000000fffffffe x8 : 3020657565757120 
[  315.974110] x7 : 74696d736e617274 x6 : 0000000000000003 
[  315.974117] x5 : 0000000000000000 x4 : 0000000000000000 
[  315.974124] x3 : 0000000000000100 x2 : 0000000000000100 
[  315.974131] x1 : b330a4409a208600 x0 : 0000000000000000 
[  315.974139] Call trace:
[  315.974144]  dev_watchdog+0x350/0x358
[  315.974153]  call_timer_fn.isra.0+0x24/0x80
[  315.974168]  run_timer_softirq+0x400/0x438
[  315.974177]  efi_header_end+0x120/0x284
[  315.974186]  irq_exit+0xd8/0xf8
[  315.974194]  __handle_domain_irq+0x68/0xc0
[  315.974203]  gic_handle_irq+0x58/0x128
[  315.974216]  el1_irq+0xcc/0x180
[  315.974222]  arch_cpu_idle+0x18/0x28
[  315.974231]  default_idle_call+0x20/0x68
[  315.974240]  do_idle+0xc0/0x128
[  315.974248]  cpu_startup_entry+0x24/0x48
[  315.974254]  secondary_start_kernel+0x15c/0x188
[  315.974267] ---[ end trace 4b3cbb528e413431 ]---
[  316.374631] icssg-prueth icssg1-eth eth2: xmit timeout

root@AM64x-Tronlong:~# [  322.117638] icssg-prueth icssg1-eth eth2: xmit timeout
[  326.981618] icssg-prueth icssg1-eth eth2: xmit timeout
[  332.101625] icssg-prueth icssg1-eth eth2: xmit timeout

root@AM64x-Tronlong:~# 
root@AM64x-Tronlong:~# 
root@AM64x-Tronlong:~# [  336.965619] icssg-prueth icssg1-eth eth2: xmit timeout
[  342.085644] icssg-prueth icssg1-eth eth2: xmit timeout

root@AM64x-Tronlong:~# 
root@AM64x-Tronlong:~# 
root@AM64x-Tronlong:~# [  346.949634] icssg-prueth icssg1-eth eth2: xmit timeout

root@AM64x-Tronlong:~# 
root@AM64x-Tronlong:~# [  352.069635] icssg-prueth icssg1-eth eth2: xmit timeout
[  356.933617] icssg-prueth icssg1-eth eth2: xmit timeout
^C
root@AM64x-Tronlong:~# [  362.053618] icssg-prueth icssg1-eth eth2: xmit timeout

And sometimes they keep up and down.

Below is our sch.

SOM_AM6442 phy.pdf

Please help analyze it, thank you!

3 months ago

0 Kallikuppa Sreenivasa 3 months ago

TI__Guru**** 190159 points

Hello Jacky Lin,

Thank you for the query.

Do you have issues with both Ethernets?

Can you describe the issue -

Are the EPHs communicating with the MAC (host)

Are the issues seen always of intermittently.

I cant see the clock source and the supplies related information in the attached schematics.

Regards,

Sreenivasa

0 jacky lin 3 months ago in reply to Kallikuppa Sreenivasa

Prodigy 175 points

SOM_AM6442 phy2.pdf

Hi Sreenivasa,

Thanks reply, I add the sch include more infomation, please check if the required information is included.

Do you have issues with both Ethernets?

Yes, both Eth of PRG1 are not good. Keep up and down, or reported an error like previous post.

Are the EPHs communicating with the MAC (host)

I think they are, because "ifconfig" can see the device, and sometimes they can work(ping function).

Looking forward to replies, thanks!

0 Kallikuppa Sreenivasa 3 months ago in reply to jacky lin

TI__Guru**** 190159 points

Hello Jacky Lin,

Thank you.

Can you confirm the RGMII length matching. The AM64x length matching info is available in the below FAQ.

(+) [FAQ] AM625 / AM623 / AM620-Q1 / AM62Ax / AM62Px / AM62D-Q1 / AM62L / AM64x / AM243x Design Recommendations / Custom board hardware design - Queries related to RGMII interface and RGMII TI EPHY - Processors forum - Processors - TI E2E support forums

based on the clock buffer seems like there is a EPHY interfaced to the CPSW3G.

Can you please confirm if that is the case and if the EPHY communicates as expected.

Do you have any EPHYs on PRG0 and are those EPHYs working?

Regards,

Sreenivasa

0 jacky lin 3 months ago in reply to Kallikuppa Sreenivasa

Prodigy 175 points

Hi Sreenivasa,

We total have 3 Eths, 2 are on this board, another one(CPSW_RGMII1) is on another board, return to this core board through the connector.

CPSW3G_RGII1.pdf

And the CPSW3G eth also have some problem, through the ping function, its packet loss rate is very high(80%).

Do you have any EPHYs on PRG0 and are those EPHYs working?

We don't have the phy to PRG0.

Can you confirm the RGMII length matching.

We have considered issues of equal length and impedance when designing, and we will re-examine the hardware design.

If there are any other doubts, please let me know, thanks!

0 Kallikuppa Sreenivasa 3 months ago in reply to jacky lin

TI__Guru**** 190159 points

Hello Jacky Lin,

Thank you.

Let me review the inputs.

In the Mean time can you start a ticket titled DP83867I loop back test for the EPHY team to support performing a loopback test to kind of ensure the PHY is not a problem,

Can you confirm if you are able to access the EPHYs using MDIO?

Regards,

Sreenivasa

0 Kallikuppa Sreenivasa 3 months ago in reply to Kallikuppa Sreenivasa

TI__Guru**** 190159 points

Hello Jacky Lin,

Missed asking.

Would you be able to measure the RGMII RX clock and/or TX clock and confirm the frequency.

Regards,

Sreenivasa

0 jacky lin 3 months ago in reply to Kallikuppa Sreenivasa

Prodigy 175 points

Hi Sreenivasa,

A loopback test, I think it's a good idea! But I'm not sure how to test it, any relevant testing instructions?

Thanks!

0 Kallikuppa Sreenivasa 3 months ago in reply to jacky lin

TI__Guru**** 190159 points

Hello Jacky Lin,

I would recommend doing the PHY loop back initially.

You will have to add a E2E thread as i have mentioned above for the EPHY team to guide or support.

Regards,

Sreenivasa