TMS570LC4357: TI FEE driver reading block after formatting or invalidating gets into strange mode

Usha Jogappanaik

Part Number: TMS570LC4357
Other Parts Discussed in Thread: HALCOGEN

I am testing FEE read and write example code from halcogen with little modification. On first run I am formatting bank7 , reading data from from block 1 throws me an error of INVALID_BLOCK and values are 0's . which is correct since no data is written yet. Then I am writing data asynchronously of length 32 into block 1 , then reading data which looks good. then in next power cycle or soft reset when I just try to read the data at block 1 throws me an error of INVALID_BLOCK which is strange. I see same behavior when data is written, invalidated and read in first run . then just read in second cycle leads to INVALID_BLOCK error. Looking for some support here.

over 2 years ago

0 jagadish gundavarapu over 2 years ago

TI__Guru 69891 points

Hi Usha,

We saw similar issue in the past, please refer below thread

(+) TMS570LS1227: FEE synchronous mode doubt - Arm-based microcontrollers forum - Arm-based microcontrollers - TI E2E support forums

Thanks & regards,
Jagadish.

0 Usha Jogappanaik over 2 years ago in reply to jagadish gundavarapu

Prodigy 110 points

eeprom_bank_rw_example.zip

yes sure. I am using Hercules TMS570LS12x LaunchPad for testing

+1 jagadish gundavarapu over 2 years ago in reply to Usha Jogappanaik

TI__Guru 69891 points

Hi Usha,

Can you share your code for easy debugging at our end?

0 Usha Jogappanaik over 2 years ago in reply to Usha Jogappanaik

Prodigy 110 points

sorry its Hercules TMS570LC43x LaunchPad

0 jagadish gundavarapu over 2 years ago in reply to Usha Jogappanaik

TI__Guru 69891 points

Hi Usha,

Your issue not rectified right? I guess you clicked on Resoled button.

0 Usha Jogappanaik over 2 years ago in reply to jagadish gundavarapu

Prodigy 110 points

yes not yet rectified ! it was by mistake

0 jagadish gundavarapu over 2 years ago in reply to Usha Jogappanaik

TI__Guru 69891 points

okay i will check the issue.

0 Usha Jogappanaik over 2 years ago in reply to jagadish gundavarapu

Prodigy 110 points

also one more update , I see this strange mode only when format is done and read. reading after Invalidate and erase block seems fine even after reset

0 jagadish gundavarapu over 2 years ago in reply to Usha Jogappanaik

TI__Guru 69891 points

Hi Usha,

Understood the root cause.

Actually, whenever we do formatting, we should not call write function directly after formatting done. Because after formatting our all virtual sectors in FEE will be in Invalid state(0xFFFF)

And there is no active virtual sector selected for the write or read operations in this case.

In this situation we should call the initialization function before calling the write or read functions, because initialization function will select a active sector again.

In your code i added one switch-case with initialization function and i called this after formatting and before calling write function

And i confirmed working without any issues. Please test it at your end and confirm me.

Thanks & regards,
Jagadish.

0 Usha Jogappanaik over 2 years ago in reply to jagadish gundavarapu

Prodigy 110 points

Thanks Jagadish for the support. calling TI_Fee_Init() after formatting fixed the issue.