• State
  • Locked Locked
  • Replies 6 replies
  • Subscribers 32 subscribers
  • Views 271 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

AM263P4: SDK 11.0.0.19 Indroduced new error messages in Networking Driver. "Failed to set dscp Priority map for Port 1 - -1"

Huey Duong
Intellectual 400 points

Part Number: AM263P4
Other Parts Discussed in Thread: TMDSCNCD263

Tool/software:

With the new SDK 11.0.0.19 there are new errors being generated during initialization.  But the driver seems to continue to run and bring up the link.

Can you help give more insight regarding this error and if we should be concerned?

You can see the issue with enet_lwip_cpsw_am263px-cc_r5fss0-0_freertos example.   Running this on TMDSCNCD263 Rev A board with CCS: 20.3.0.14__1.9.0 and manually installed SDK 11.0.0.19 and SYSCONFIG1.25.0.

==========================
      ENET LWIP App
==========================
EnetAppUtils_reduceCoreMacAllocation: Reduced Mac Address Allocation for CoreId:0 From 4 To 2
Link Status Changed. PHY: 0x0, state: up
Open MAC port 2
EnetPhy_bindDriver: PHY 0: OUI:080028 Model:0f Ver:03 <-> 'DP83869' : OK
PHY 0 is alive
CpswMacPort_ioctl: cpsw3G.macport1: Module is not open
Cpsw_registerIoctlHandler: Failed to register IOCTL handler: -1, 1000502, 70139C91
EnetPer_ioctl: cpsw3g: Failed to do IOCTL cmd 0x01000110: -1
Enet_ioctl: cpsw3g: IOCTL 0x01000110 failed: -1
Failed to set dscp Priority map for Port 1 - -1
Starting lwIP, local interface IP is dhcp-enabled
[LWIPIF_LWIP] NETIF INIT SUCCESS
Host MAC address-0 : 70:**:**:**
[LWIPIF_LWIP] Enet has been started successfully
[0]status_callback==UP, local interface IP is 0.0.0.0
UDP server listening on port 5001
Cpsw_handleLinkUp: Port 2: Link up: 1-Gbps Full-Duplex
MAC Port 2: link up

The culprit is in Cpsw_registerIoctlHandler(), looks like there has been code change here that generates this new error.

This is SDK10:

This is SDK 11:

SDK 11's call to CpswMacPort_ioctl returns failure whereas SDK10's call to EnetMod_registerMacportIoctlHandler() returns success.

We are currently trying to migrate our code to SDK 11 and want to know how to manage this error.  Thanks.

PS.  There are some compile errors in the code that should be fixed.  These are seen when ENETTRACE is set to VERBOSE level:

  • 0
    TI__Mastermind 21332 points

    Hi Huey,

    Let me take a look at this and discuss with the networking team. I'll share an update soon.

    Regards,
    Shaunak

  • 0
    TI__Mastermind 21332 points

    Hi Huey,

    The error we see on the AM263Px-Control Card for Port-1 is due to the ControlCard having only one ethernet port brought out (port-2). The error seen from MAC Port-1 is due to no PHY Add-on-board being present (thus no MAC Port-1) being available. The DSCP priority mapping is attempted from Port-1 and due to no port being available the error is seen. This should not be breaking any functionalities, i believe in the latest release, the default approach is try to set DSCP priority mapping for all ports. I'm discussing this implementation further with the team, meanwhile, this should not cause issues in the example since the MAC Port-2 is opened and linked up correctly and no errors seen for that.

    Regards,
    Shaunak

  • 0
    TI__Mastermind 21332 points
    Huey Duong said:

    We are currently trying to migrate our code to SDK 11 and want to know how to manage this error.  Thanks.

    PS.  There are some compile errors in the code that should be fixed.  These are seen when ENETTRACE is set to VERBOSE level:

    Regarding this,

    I tried building the enet-cpsw lib in the 11.00.00.19 SDK with -DENET_CFG_TRACE_LEVEL=5 (5 is VERBOSE level), i did not encounter any compilation errors weirdly. Can you share the exact change made for enabling ENET_TRACE_VERBOSE, i'll try to replicate based on that. But looking at the code, it does seem to have extra semicolons and potential errors.

    Regards,
    Shaunak

  • 0 in reply to Shaunak Deshpande
    Intellectual 400 points

    thank you for the response.  this makes sense, and I will inform the team to expect a new error message that can be ignored.

  • 0 in reply to Shaunak Deshpande
    Intellectual 400 points

    Only thing i had to do was set ENET_CFG_TRACE_LEVEL=5.  This enabled ENETTRACE_trace(...)

    Which generates this first error.   hMod is not a defined variable.  So this had to be fixed to hStats->name.

    Fixing that issue, the next one is found:

  • 0 in reply to Huey Duong
    TI__Mastermind 21332 points

    Hi Huey,

    Regarding these errors, I have filed a bug for the networking team to fix this. Apologies you run into these build errors which is definitely not expected. I reproduced the issue locally and saw that there are two errors, which are already mentioned (and fixed) by you (one using hMod instead of hStats and another error having an extra semi-colon). 

    The JIRA for internal tracking is: https://jira.itg.ti.com/browse/MCUSDK-15019

    Regards,
    Shaunak