This thread has been locked.
If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.
Tool/software:
Could you provide an easy example of CPSW based L2 packet transmission?
A simplified example with driver initialization and TX RX would be perfect.
Thanks for all
M
Hi Marco Messori,
Thanks for your query.
Please refer below example and let me know if you need further help.
AM243x MCU+ SDK: Enet Layer 2 CPSW Example (ti.com)
Regards
Ashwani
Hi Ashwani
I know all the example in the MCU SDK but to start is better spmething easy. For example trasmit only a packet and not use the queue could be better. Also a more clear initialization could be better.
Then all example in different MCU sdk regarding CPSW L2 show this error:UDMA RX Channel open failed: 0xffffffff
How can I solve these issues ?
Thanks for all
M
Then all example in different MCU sdk regarding CPSW L2 show this error:UDMA RX Channel open failed: 0xffffffff
Do you mean, you are getting this error, while running out of box SDK example?
Any specific example you can suggest me to reproduce the issue on my setup?
Please attach a screenshot of you workspace/ logs as well.
Regards
Ashwani
Hi Ashwani
yes I get the error running out of box enet_layer2_cpsw examples from any MCU-SDK (mcu_plus_sdk_am243x_08_02_00_31, ............mcu_plus_sdk_am243x_10_00_00_20).
I have installed CCS v.12.8 and all MCU-SDK since mcu_plus_sdk_am243x_08_02_00_31.
Following the console message:
[MAIN_Cortex_R5_0_0] ==========================
Layer 2 CPSW Test
==========================
Init Enet's OSAL and utils to use defaults
Init memory utils
Create clock and task for periodic tick
Create periodic tick task
Create periodic tick clock
Open Main UDMA driver
Init all peripheral clocks
----------------------------------------------
Enabling clocks!
Init all configs
----------------------------------------------
cpsw-3g: init config
Create RX tasks
----------------------------------------------
cpsw-3g: Create RX task
Open all peripherals
----------------------------------------------
cpsw-3g: Open enet
EnetUdma_openRxCh: [Enet UDMA] UDMA RX Channel open failed: 0xffffffff
EnetHostPortDma_open: Failed to open Enet DMA RX channel: -1
Cpsw_openInternal: CPSW: Failed to open CPSW DMA
Cpsw_closeInternal: Assertion @ Line: 921 in /home/gtbldadm/nightlybuilds/repo_manifests/scripts/jenkins/mcu_plus_sdk_am243x_08_02_00_31/source/networking/enet/core/src/per/V1/cpsw.c: hCpsw->hRxRsvdFlow != NULL
Any ideas for solve the issues?
Are there easy example to use CPSW to tx and Rx single Ethernet Layer2 frame?
Thanks for all
M
Hi I quickly run "/home/sitara_mpu_apps/ti/mcu_plus_sdk_am64x_10_00_00_20/examples/networking/enet_layer2_cpsw_switch" and got it working perfectly fine without any local changes.
Pelase check your setup and settings as per MCU+SDK documentaion
CCS v.12.8
software-dl.ti.com/.../GETTING_STARTED.html
Regards
Ashwani
Hi Ashwani
now the exsmple from mcu_plus_sdk_am243x_10_00_00_20 compile correctly and open DNA channels.
I send packet to AM2434 LP board by means colasoft packet, I see the packet on wire shark but interrup on DMA receptions is not triggered at all.
Also I Try to trasmit, callback trasmission fire correctly but I don't trigger nothing on wire shark.
I connected the AM243X to my PC directly or using a switch but the result not change.
Do you have some ideas?
Can you provide me a easy script to anly trasmit a Layer 2 Ener frame?
Thanks for the support
M
Hi Marco Messori,
I am assuming you are following steps mentioned in section "Run the example" in "AM64x MCU+ SDK: Enet Layer 2 CPSW Example (ti.com)
Please share full UART console logs.
Try with broadcast frames as well.
Regards
Ashwani
Hi Ashwani
yes I'm following the step you mentioned. The behaviour (no receptions , no packet trasmission) remain the same with all MCU-SDK I tested.
I try to send broadcast telegram but the result do not change,
Following UART log;
[MAIN_Cortex_R5_0_0] ==========================
Layer 2 CPSW Test
==========================
Init all peripheral clocks
----------------------------------------------
Enabling clocks!
Create RX tasks
----------------------------------------------
cpsw-3g: Create RX task
Open all peripherals
----------------------------------------------
cpsw-3g: Open enet
Init all configs
----------------------------------------------
cpsw-3g: init config
Mdio_open: MDIO Manual_Mode enabled
cpsw-3g: Open port 1
EnetPhy_bindDriver: PHY 3: OUI:080028 Model:0f Ver:01 <-> 'dp83869' : OK
PHY 3 is alive
PHY 15 is alive
Attach core id 1 on all peripherals
----------------------------------------------
cpsw-3g: Attach core
cpsw-3g: Open DMA
initQs() txFreePktInfoQ initialized with 16 pkts
cpsw-3g: Waiting for link up...
Cpsw_handleLinkUp: Port 1: Link up: 1-Gbps Full-Duplex
MAC Port 1: link up
cpsw-3g: MAC port addr: f4:84:4c:f9:98:2c
Can you send me a script with easy packet trasmission to start analyse what is the issue?
Thanks for all
M
Hi Marco Messori,
Can you send me a script with easy packet trasmission to start analyse what is the issue?
You can refer code base in "l2_cpsw_dataflow.c" file for reference and make chnages for local transmission of your custom frame instead of received frames.
====================================================
Refer void EnetApp_rxTask(void *args);
you can create your own packet instead of
/* Wait for packet reception */
/* Get the packets received so far */
/* Consume the received packets and send them back */
====================================================
/* Enqueue the packet for later transmission */
EnetQueue_enq(&txSubmitQ, &txPktInfo->node);
/* Transmit all enqueued packets */
status = EnetDma_submitTxPktQ(perCtxt->hTxCh, &txSubmitQ);
if (status != ENET_SOK)
{
EnetAppUtils_print("%s: Failed to submit TX pkt queue: %d\r\n", perCtxt->name, status);
}
====================================================
Regards
Ashwani
Hi Ashwani
have you ideas regarding the issues in trasmitting and receiving Enet packet?
Any suggestion is appreciate by us
Thanks for all
M
Hi Marco Messori,
I am assuming, default SDK example is working fine for you. correct?
C:\ti\mcu_plus_sdk_am243x_10_00_00_20\examples\networking\enet_layer2_cpsw_switch\am243x-evm
C:\ti\mcu_plus_sdk_am243x_10_00_00_20\examples\networking\enet_layer2_cpsw\am243x-evm
have you ideas regarding the issues in trasmitting and receiving Enet packet?
Can you provide more info about the issue ?
What you change locally and what is the observation?
Some screenshots of the issues will be helpful to reproduce the issue on my setup.
"l2_cpsw_dataflow.c"
Regards
Ashwani
Hi Hi Ashwani
the default SDK example show the same problems.
The Enet is open and the negotiation oprate correctly. The MAC adress il print correctly on the UART and I can read the statistics. But il I send a frame by means colasoft to the AM2434LP the callback DMA rx function never fire.
Statistics using UART commad not change during the operation time. It seem the PHY do not process the packet.
Thanck
Best regards
M
. But il I send a frame by means colasoft
There may be some problem on windows PC side. These tools won’t work under VPN and might need admin privileges.
Can you try with Linux PC and packEth tool?
It seem the PHY do not process the packet.
PHY 3 is alive
PHY 15 is alive
Attach core id 1 on all peripherals
----------------------------------------------
cpsw-3g: Attach core
cpsw-3g: Open DMA
initQs() txFreePktInfoQ initialized with 16 pkts
cpsw-3g: Waiting for link up...
Cpsw_handleLinkUp: Port 1: Link up: 1-Gbps Full-Duplex
MAC Port 1: link up
cpsw-3g: MAC port addr: f4:84:4c:f9:98:2c
But from your above logs, seems that PHY is up and working. Llink negotiation happedned successfully for 1G.
Regards
Ashwani
Hi Ashwani
really I'm able to capture the frames send by windows PC using Wireshark, so I think Colasoft sends correctly the frame.
At the moment I can't use a Linux client all pC are win based client.
I can send you Wureshark log to show the trasmitted frame by PC.
Thanks
M
I can send you Wureshark log to show the trasmitted frame by PC.
Let's see if it give me some pointers to help you Marco Messori
Regards
Ashwanic
ok, next step in my test will be capture WireShark log and send it to you.
Best regards
M
Hi Marco Messori ,
You can also refer below functions in AM64x MCU+ SDK: Enet CPSW EST Example
void EnetApp_txTest(void);
static void EnetApp_rxTask(void *args);
Regards
Ashwani
Hi Ashwani
in the image the packet send to texas LPAM2434
You can see packet send using Colasoft or using Python Scapy.
Wireshark show the packet trasmitted by packet are not received from LPAM243X and not trasmitted.
Thanks Marco Messori for wireshark capture.
Looks like AM243x is not getting IP address from DHCP. correct ?
Can you insert a DHCP enabled switch in between PC and AM243x ?
Regards
Ashwani
Hi Ashwani
correct we don't want a TCP-IP layer but only a L2 trasmission so raw frame trasmitted in deasy chain between some (up to 10) client based on Am243x.
I can try enable DHCP but this is out of the design scope.
Tanks for all
Best regards
M
Hi Ashwani
to use TCP/IO layer I have to introduce LwIP so before change design specifications I try to use an old MCU-SDK. I performed same test with mcu_plus_sdk_am243x_08_02_00_31 out of box and I find communication between PC and LPAM2434 board.
Wireshark log show the datagram trasmitted by LPAM2434 board evry time it receives a packet from PC.
Have you an idea regarding this behaviour? It's quite strange I'am able to communicate with an old Enet driver and I 'm not able to mede the same with the last driver released by Ti.
Thanks for all
Best regards
M
I performed same test with mcu_plus_sdk_am243x_08_02_00_31 out of box and I find communication between PC and LPAM2434 board.
Thanks for pointing it...I forwarded this to concern team. Tracking as a bug with SITSW-5861.
Regards
Ashwani
Hi Ashwani
I checked some other projects related to network l2_cpsw and I verified
the correct receive and transmit packets only for the example related to
MCU_SDK_08_02_00_31. The example in this MCU_SDK perform correctly.
All design starting from MCU_SDK_08_03_00_18 to MCU_SDK_10_00_00_20 show the
same behaviour.
The driver is opened and configured, the DMA RX and TX channels are configured,
the reception task is started, the link is UP but there is no transmission from
the LP_AM2434 to the PC nor reception from the PC to the board.
I use Whireshark to check the communications.
I verified the difference in the configuration functions in file l2_cpsw_cfg
coused by the EnetApp_driverOpen() API that in these MCU_SDK takes care of the Ethernet
configuration operations.
Since the LP_AM2434 seems to be correctly initialized, do you think the communication
problems may be due to the UDMA driver?
Have you ever experienced similar behavior? I see the some result using two different
client windows PC with Win7 and Win10 OS installed.
Packet Builder is used to send ethernet packets and Wireshark to supervise the LAN.
Thanks foe tge support
Best regards
M
Hi Ashwani
I made some network test using ICSS-EMAC.
I performed the test with the example enet_layer2_icssg on MCU-SDK_08_02_00_31 and MCU_SDK_10_00_00_20 with the same result.
In both cases program block during the DMA opening phase of second istance.
You can see the log in the following:
[MAIN_Cortex_R5_0_0] ==========================
MULTIPORT TEST
==========================
Init all peripheral clocks
----------------------------------------------
Open all peripherals
----------------------------------------------
icssg1-p1: Register async IOCTL callback
icssg1-p1: Register TX timestamp callback
icssg1-p2: Register async IOCTL callback
icssg1-p2: Register TX timestamp callback
Attach core id 1 on all peripherals
----------------------------------------------
icssg1-p1: Attach core
icssg1-p2: Attach core
Create RX tasks
----------------------------------------------
icssg1-p1: Create RX task
icssg1-p1: Waiting for link up...
icssg1-p2: Create RX task
icssg1-p2: Waiting for link up...
icssg1-p2: Port 2 link is up
icssg1-p2: Set port state to 'Forward'
icssg1-p2: Open DMA
icssg1-p2: Set MAC addr: icssg1-p1: Port 1 link is up
icssg1-p1: Set port state to 'Forward'
icssg1-p1: Open DMA
I have the same behaviour for both MCU.
ICSS-EMAC is configured as DUAL_MAC. Can I leave only one istance ti avoid this error?
Have you some ideas to solve the issue?
Thanks for all
Best regards
M
Hi,
I quickly tried example with SDK 10.0 on AM243-evm
example enet_layer2_icssg on MCU-SDK_08_02_00_31 and MCU_SDK_10_00_00_20
Setup:
Windows-PC <=> D-link-switch <=> AM243x-evm-ICSSG1-Port-1
Logs:
========================== MULTIPORT TEST ========================== Init all peripheral clocks ---------------------------------------------- Enabling clocks! Enabling clocks! Open all peripherals ---------------------------------------------- Init configs EnetType:1, InstId :2 ---------------------------------------------- Mdio_open: MDIO Manual_Mode enabled Open MAC port 1 EnetPhy_bindDriver: PHY 15: OUI:080028 Model:0f Ver:01 <-> 'dp83869' : OK PHY 3 is alive PHY 15 is alive Init configs EnetType:1, InstId :3 ---------------------------------------------- Mdio_open: MDIO Manual_Mode enabled Open MAC port 2 EnetPhy_bindDriver: PHY 3: OUI:080028 Model:0f Ver:01 <-> 'dp83869' : OK PHY 3 is alive PHY 15 is alive icssg1-p1: Register async IOCTL callback icssg1-p1: Register TX timestamp callback icssg1-p2: Register async IOCTL callback icssg1-p2: Register TX timestamp callback Attach core id 1 on all peripherals ---------------------------------------------- icssg1-p1: Attach core icssg1-p2: Attach core Create RX tasks ---------------------------------------------- icssg1-p1: Create RX task icssg1-p1: Waiting for link up... icssg1-p2: Create RX task icssg1-p2: Waiting for link up... Enet Multiport Menu: 'T' - Enable timestamp prints 't' - Disable timestamp prints 's' - Print statistics 'r' - Reset statistics 'm' - Show allocated MAC addresses 'd' - Enable dscp based priority mapping 'f' - Read all ICSSG FDB slots 'x' - Stop the test Icssg_handleLinkUp: icssg1-1: Port 1: Link up: 1-Gbps Full-Duplex icssg1-p1: Port 1 link is up icssg1-p1: Set port state to 'Forward' icssg1-p1: Async IOCTL completed icssg1-p1: Open DMA initQs() txFreePktInfoQ initialized with 8 pkts 34:08:e1:80:a9:86 icssg1-p1: Set MAC addr: 34:08:e1:80:a9:86
Regards
Ashwani
Hi Ashwani
I wrote you directly on email because the Tii website is under maintenance-
Now I made some test using PRU-ICSS with LLD and the example on MCU-SDH-10_00_00_20 operate correctly.
I had problems with EnetAppUtils_print() that block the correct execution. Replacing EnetAppUtils_print() with printf() solve issues in enet_layer2_icssg……. example.
Now I would want to test PRU-ICSSG with EMAC driver. But I read the following on the MCU-SDK documentation:
ICSS-EMAC will be supported for use in Profinet Device, EtherNet/IP Adapter and PRP(Parallel Redundancy Protocol) examples only.
Can you provide an example of layer2 communication using PRU-ICSS and EMAC similar to the examples in the MCU-SDK implemented usng PRU-ICSS and LLD driver?
Thanks for the support
M
Hi Ashwani
can I use one of the example to design an Ethernet switch for connecting LPAM2434 board in daisy chain?
I need a communication between 4 board in daisy chain mode without predefined Ethernet Protocol but only trasmit MAC frame using EMAC driver istead of Enet LLD.
Can I found or modify one of the example in sdk?
Thganks for the support
M
Hi,
Currently, we do not support generic Ehernet using icss-emac.
We recommend using enet-lld for this purpose.
I need a communication between 4 board in daisy chain mode without predefined Ethernet Protocol but only trasmit MAC frame using EMAC driver istead of Enet LLD.
Can you help me with the reason (you have specific use case or system limitation) for not using enet-lld ?
Regards
Ashwani
ok Aswani
I understand.
But what do you think could be the best example in industrial ethernet sdk to understand the EMAC driver?
Thanks
M
But what do you think could be the best example in industrial ethernet sdk to understand the EMAC driver?
You can refer EIP or HSR but those are NOT equivalent to standard ethernet.
Can you help me with the reason (you have specific use case or system limitation) for not using enet-lld ?
We recommnd to use enet-lld for standard ethernet use case.
Regards
Ashwani
Hi Ashwani
I made other test using MCU-SDK following the version 08_02_00_31 and I find the same issues in ol MCU-SDK. Both single MAC or switch CPSW example open correctly the port but I do not see any communication.
I saw the MCU_SDK_08_02_00_31 is the last one with the CPSW driver configuration made into the user functions while from the following MCU_SDK version (MCU_SDK_08_03_00_18 ...) the API EnetApp_driverOpen(Enet_Type enetType, uint32_t instId) in file ti_enet_open_close.c contains most of the initialization code. I also noticed the function EnetApp_driverOpen() invoke the function EnetApp_getCpswInitCfg() and then invoke EnetApp_updateCpswInitCfg() that update the CPSW parameter into user function. There are other point non so clear in the newer MCU-SDK but what I don't understand is the reason of the lake in the communication (both RX and TX) in all MCU-SDK except MCU-SDK-08_02_00_31.
Hav eyou some idea regarding the issue
Best refgards
M
Hi Ashwani
I test the example enet_layer2_timesync into MCU_SDK_08_02_00_31 and it operate correctly.
What I have to understand is the mechanism to allow the MAC to manage the 2 different task in reception. I saw that changing the Ethernet Type in Colasoft Packet Builder result in the activation of one of the 2 Rx Task.
If I set the Ethernet Type 0x88F7 the EnetApp_rxTaskPTP execute after the packet reception.
If I set a different Ethernet Type (0x806) the EnetApp_rxTask execute after the packet reception.
I see the firmware create both Task (EnetApp_rxTaskPTP and EnetApp_rxTask) but where the dependecy by the ethernet Type is made? When PHY receive the packet where is the point to manage Enet Type and call one of the 2 different callback function (EnetApp_rxIsrFxn and EnetApp_rxIsrFxnPtp) to post the proper semaphore?
Can you halp me to understand the mechanism?
And what are the difference between PTP frame and standard layer 2 frame?
Thanks for the support
Best regards
M
Hi Ashwani
I'm using the example l2_cpsw on MCU_SDK_08_02_00_31 the only I'm able to operate correctly.
In the design a Tak is used to both receive and trasmit back the Ethernet frame.
Is it possible to use the driver without freeRTOS OS in bare metal fashion?
I try several time modify the example but I find error douring the initialization stage?
Can you send me a bare metal example? I can use OS in real application.
Thanks for the support
Best regards
M
If I set the Ethernet Type 0x88F7 the EnetApp_rxTaskPTP execute after the packet reception.
If I set a different Ethernet Type (0x806) the EnetApp_rxTask execute after the packet reception.
Is it possible to use the driver without freeRTOS OS in bare metal fashion?
Currently, we do not support baremetal development for Ethernet
Regards
Ashwani
Hi Ashwani
I summarize the topics not clear from my side.
1) I have tried all the examples in the MCU-SDKs related to CPSW or PRU-ICSS based Ethernet communication. I m able to establish a proper communication (verified with Wire Shark) only using the MCU-SDK_08_02_00_31 (with the driver configured as a switch single MAC). All examples relating to MCU-SDK from the subsequent versions up to the latest available on the site (MCU-SDK-10_00_00_20) show neither transmission nor reception of frames sniffing by Wirw Shark. The driver opens correctly but no frame transmission is detected via Wireshark. Why does only the 2 year old framework work fine and the new ones don't? I tried with PC Win7 and Win10 but the result remain the same. I tried connecting the LP2364 directly to the PC or via a local switch. This poin is now open.
2) Is it possible to have an example without using FreeRTOS. For the application I'm trying to analyze a bare metal would be better. I know you answered to me that there are no examples made without RTOS. I tried to modify the working one supplied with MCU-SDK-08_02_00_31 but I found errors during the initialization phase. Program flow freezes. Can you suggest me a way to design a system without RTOS? I not found any other support on the web.
3) I need a support for a feasybility analysis involved Sitara microcontroller in Daisy Chain connection. I need to share the schematics of the design involved. Wjhat is the better way to follow to have support?
Thanks for all
Best regards
M
2) Is it possible to have an example without using FreeRTOS. For the application I'm trying to analyze a bare metal would be better.
Currently, we do not support baremetal development for Ethernet
3) I need a support for a feasybility analysis involved Sitara microcontroller in Daisy Chain connection. I need to share the schematics of the design involved. Wjhat is the better way to follow to have support?
You can connect with TI-field team (FAE).
You can also raise schematic review with HW expert team.
Regards
Ashwani
Hi Ashwani
I performed new test changing the board. I try the same example using a LP_AM263X board and obviously I used the MCU_SDK_10_00_00_35 last version delivery for AM263x by Texas.
The Example perform correctly, driver open correctly and I see the communication on Wire Shark.
Also I'm able to flash the application Image via QSPI using Ti Uniflash.
Now my question related to the LP_AM2434 board are the folowung:
1) Why I not able to program the application Image in any way using this demo board? I verify the same error using Uniflash or uart_uniflash_gui.py. I can only program the RAM by CCS studio but I'm not able to flash appimage in FLASH memory so I have to download the image any time I reset the application.
2) Why the newer driver foe AM263X (MCU_SDK_10_00_00_35) that sem very similar to the MCU_SDK_10_00_00_20 tested for AM2434 operate correctly? It seem a problem coused by LP_AM2434 because a very similar example operate correctly on AM263X. Are all my issues strictly connected with AM2434 board?. Can I solve this issue in any way?
Pleae note that I have only a LP_AM2434 board and a LP_AM263X board and I neet test the possible operating way in a 4 NIC cluster so I have to connect the 2 board with 2 PC to simulate the complete cluster. So I need to use also AM2434 board.
Some suggestion for me?
Thanks for all
best regards
M
The Example perform correctly, driver open correctly and I see the communication on Wire Shark.
Thanks for confimration.
Why I not able to program the application Image in any way using this demo board
Can you please raise separate query for flashing issue to connect you domain expert?
2) Why the newer driver foe AM263X (MCU_SDK_10_00_00_35) that sem very similar to the MCU_SDK_10_00_00_20 tested for AM2434 operate correctly? It seem a problem coused by LP_AM2434 because a very similar example operate correctly on AM263X. Are all my issues strictly connected with AM2434 board?. Can I solve this issue in any way?
Can you raise separate query on this ?
Regards
Ashwani
Hi Ashwani
I try to separate the queries related the way to flashing application image on Sitara LP_AM263 and Sitara LP_AM2434 demo boards.
As in my previous message I performed some tests using a new LP_AM263 board instead the old LP_AM2434 board.
I tested the same example (mcu_plus_sdk_am263x_10_00_00_35\examples\networking\enet_layer2_cpsw) previously tested with LP_AM2434 board (mcu_plus_sdk_am243x_10_00_00_20\examples\networking\enet_layer2_cpsw)
The Example for LP_AM263 perform correctly, driver open correctly and I see the communication on Wire Shark.
Also I'm able to flash the application Image via QSPI using Ti Uniflash.
My question related the way to flash image application on the LP_AM2434 board are the following:
1) Why I not able to program the application Image in any way using the LP_AM2434 demo board? I verify the same error using Ti Uniflash GUI or uart_uniflash_gui.py. I can only program the RAM by CCS studio but I'm not able to flash appimage in FLASH memory so I have to download the image any time I reset the application.
2) Why the mcu_plus_sdk for AM263 (MCU_SDK_10_00_00_35) that seem very similar to the MCU_SDK_10_00_00_20 tested for AM2434 operate correctly? It seem the flashing issues coused by LP_AM2434 because a very similar example operate correctly on AM263. Are all my issues strictly connected with AM2434 board?. Can I solve this issue in any way?
Summerizing I note a different behaviour using LP_AM263 and LP_AM2434 flashing the application image on FLASH.
I follow the line guide in the link file:///H:/Texas_Sitara_AM243X/mcu_sdk_am243x/mcu_plus_sdk_am243x_10_00_00_20/docs/api_guide_am243x/GETTING_STARTED_FLASH.html
I hope to be clear, in case you need more information please ask them to me and I try to made a quick response
Thanks for allM
flash image application on the LP_AM2434 board
Hi Marco,
Can you please open a separate e2e-thread to discuss flashing issue on Am243x-lp ?
I will send that to flashing expert.
Regards
Ashwani
Hi Hi Ashwani
just another question.
What are the API to read the speed of the link and the duplexity?
I configured as the example Link Speed Capability as ENET_SPEED_AUTO and Link Duplexity Capability as ENET_DUPLEX_AUTO.
But what are the IOCTL command to read these parameter from the PHY?
Thanks
M
Hi Ashwani
Iìm able to operate correctly using LP_AM363x and MCU_plus_SDK_10_00_0_35. I sniff the net using Wire Shark and I see the frame trasmittend and received from the NICs.
I have a question for tou.
What are the API to read the speed of the link and the duplexity from the driver?
I configured as in the following:
- Link Speed Capability as ENET_SPEED_AUTO
- Link Duplexity Capability as ENET_DUPLEX_AUTO.
But what are the IOCTL command to read these parameter from the PHY?
Thanks for all
Best regards
M