MPC5745B FCCU questions(about STCU)

[zszcyx]

Dear NXP Engineers： 

I have some questions about the 2 NCFs below  , sincerely ask for your help。

Backgrounds : 

CPU : MPC 5745B

Specification ：MPC5746C_RM_Rev4.pdf

NCF[5] ： Configuration error during reset while loading STCU configuration

 I assume it happens in the phase3  during reset，In my opinion，there may be 2 reasons that can cause this NCF。

• DCF Recores error ：If it happens，Software can do nothing to recover it ， the device is unuseable anymore.
• DCF Records are correct， but some problems happen during flash read。
  • If it happens，it is possiable to be recovered by a long reset。
  • since FCCU configurations are saved in the NVM， if I set the reaction of NCF[5] as long reset，It is able to long reset at once 。

Please kindly correct me if I’m wrong。 

I’m also wondering ，If I do nothing when the error occurs ，what will happen？

       -Will the CPU continue to excute BIST ，and set a BIST error？

       -Or Jump over the BIST ，run the application directly？

       -Or hold in the Phase 3？

NCF[8] ： Critical fault indication from STCU in case selftest control signals go to wrong condition during user  application

What’s the meaning of this error ?Does it mean that the BIST run in the application mode unexpectedly?

Since the online BIST is started by the user application , I’m not sure  how the cpu differs the application BIST  from shutdown BIST.

What will happen if this error occurs , how shall I resolve it?

Thans a lot for your support.

Best Regrads

Vera 

[zszcyx]

Dear Peter

Thanks a lot for your answers , I had read AN5259 before, It helps me a lot. 

But I still have some questions about the 2 NCFs.

NCF[5] ：

Question1 :

   When STCU configuration loading fail , what will happen？

       -Will the CPU continue to excute BIST ，and set a BIST error？

       -Or Jump over the BIST ，run the application directly？

       -Or hold in the Phase 3？

Question2 :

In case the safety DCF client is wrong you will get CERS bit set and NCF[4].

MPC5745B doesn't support any DTC client safety features , isn't it ? 

NCF[8] ： 

LBIST/MBIST control signals are getting into an unexpected state during  application runtime , potentially putting a LBIST partition or a RAM array into the BIST mode during application execution. 

Question 1:

Which is the correct sequence of the error ?

• application runtime -> user start Online BIST -> Enter online BIST mode ->BIST control signals state error ->NCF[8] happen
• application runtime -> Enter online BIST mode  unexpectly (not user's intention) ->NCF[8] happen

Question 2:

   What should I do to recover this error ? 

Sorry , I've got too many questions ,again ,thank you very much.

[petervlna]

Hello,

MPC5745B doesn't support any DTC client safety features , isn't it ? 

No, it does not apply the safety records. Threfore there is no CERS bit SSCM to report safety client error.

Question 1:

Which is the correct sequence of the error ?

• application runtime -> user start Online BIST -> Enter online BIST mode ->BIST control signals state error ->NCF[8] happen
• application runtime -> Enter online BIST mode  unexpectly (not user's intention) ->NCF[8] happen

Th second scenario is for NCF[8].

The probability of this fault is very low, so you will see it only when there is a real issue on micro. I do not expect any fake event.

Question 2:

   What should I do to recover this error ? 

It really depend on the safety level we are talking about. Safe reaction would be reset and than conditional program execution. But this is defined by the standard according to you are developing your application.

My advice would be to put micro into the safe state a sent the car to garage for check.

regards,

Peter

[zszcyx]

Dear Peter

Thanks a lot ，Peter， you really help a lot , I'm really apperciate.

Here is my last question :

When STCU configuration loading fail , what will happen？

       -Will the CPU continue to excute BIST ，and set a BIST error？

       -Or Jump over the BIST ，run the application directly？

       -Or hold in the Phase 3？

Thank you ~

Vera

[petervlna]

MIcro will run into application and will report NCF and also that BIST was bypassed/ wasn't successful (not sure here).

Even if the BIST fail you will still get into application.

regards,
Perter

[zszcyx]

Thanks a lot for your professional answers, Peter 

All my quesions are answered, Thank you !

[petervlna]

Hi,

lets explain this. I have already explained majority of the faults in application note for MPC5744P. FCCU is the same only some faults signals are different. But in general is applicable for whole MPC57xx family.

https://www.nxp.com/docs/en/application-note/AN5259.pdf 

NCF[5] -

Yes, STCU records are loaded in reset phase 3.

The default chip configuration is stored in the DCF records. While the STCU configuration is being
loaded from the DCF record flash memory, a fault can occur. To signal that the STCU loading is not
done correctly, the SSCM module sends out an error signal to the FCCU. 

n my opinion，there may be 2 reasons that can cause this NCF。

• DCF Recores error ：If it happens，Software can do nothing to recover it ， the device is unuseable anymore.
• DCF Records are correct， but some problems happen during flash read。

Well if DCF record is not correct you will also see NCF[5] (wrong data or address, etc...) you will get set SSCM bit CES.

In case the safety DCF client is wrong you will get CERS bit set and NCF[4].

NCF[8] -

5.7.  STCU2 fault condition (run in application mode)
The STCU module is the source of this signal. The STCU2 fault condition (run in the application mode)  is triggered when the LBIST/MBIST control signals are getting into an unexpected state during  application runtime, potentially putting a LBIST partition or a RAM array into the BIST mode during application execution. 

best regards,
Peter