NFC_EEPROM sample write does not take effect on NUCLEO-L433RC-P + X-NUCLEO-NFC04A1

[Arachnid]

Description of defect

After building the NFC_EEPROM sample for the Nucleo-L433RC-P dev board with the Nucleo-NFC04A1 expansion, the sample compiles and runs, but while it claims to successfully write a message, the message in the EEPROM is not modified.

The following was added to mbed_app.json in target_overrides:

        "NUCLEO_L433RC_P": {
            "target.extra_labels_add": ["ST25DV"],

After compiling and flashing, the following output is observed:

Building an ndef message
message written successfully
Received an ndef message of size 20
message read successfully

However, the message when read by a smartphone remains unchanged - and no matter the URL set in the source, the length read back is always 20.

Target(s) affected by this defect ?

NUCLEO-L433RC-P + X-NUCLEO-NFC04A1

Toolchain(s) (name and version) displaying this defect ?

Arm Compiler 6

What version of Mbed-os are you using (tag or sha) ?

6.15.1

What version(s) of tools are you using. List all that apply (E.g. mbed-cli)

None

How is this defect reproduced ?

1. Download this repository and open it as a workspace in Mbed Studio
2. Select the NFC_EEPROM project as active
3. Make the changes described above to mbed_app.json
4. Compile and run the app.

[Arachnid]

I've captured a trace of I2C bus activity with my 'scope, starting at the point where build_ndef_message returns:

"Tektronix MDO3034, version v1.14, serial number C016322"
"Bus Definition: I2C"
 Time, Repeat Start, Direction, Address, Data, Missing Ack
3.008000e-05,  , Write, 53, 20 03,  
1.221880e-04,  , Read, 53, 00,  
1.961280e-04,  , Write, 53, 20 03,  
3.646720e-04,  , Write, 57, 00 14,  
4.570400e-04,  , Read, 57, 7F,  
5.293840e-04,  , Write, 57, 00 15,  
6.543360e-04,  , Read, 57, 00,  
7.270080e-04,  , Write, 57, 00 16,  
8.193040e-04,  , Read, 57, 03,  
8.938560e-04,  , Write, 53, 00 00,  
9.862320e-04,  , Read, 53, E1 40 40 00,  
1.121664e-03,  , Write, 53, 00 04,  
1.214084e-03,  , Read, 53, 03 14 D1 01,  
1.350696e-03,  , Write, 53, 00 04,  
1.498520e-03,  , Write, 53,,  
1.661860e-03,  , Write, 53, 00 06,  
1.794940e-03,  , Write, 53,,  
1.869776e-03,  , Write, 53, 00 0F,  
2.002596e-03,  , Write, 53,,  
2.175912e-03,  , Write, 53, 20 03,  
2.268072e-03,  , Read, 53, 00,  
2.341896e-03,  , Write, 53, 20 03,  

3.183078e-02,  , Write, 53, 20 03,  
3.192303e-02,  , Read, 53, 00,  
3.199685e-02,  , Write, 53, 20 03,  
3.216452e-02,  , Write, 53, 00 00,  
3.225686e-02,  , Read, 53, E1 40 40 00,  
3.239296e-02,  , Write, 53, 00 04,  
3.248575e-02,  , Read, 53, 03 14 D1 01,  
3.262183e-02,  , Write, 53, 00 04,  
3.271393e-02,  , Read, 53, 03 14 D1 01,  
3.294784e-02,  , Write, 53, 00 00,  
3.303974e-02,  , Read, 53, E1 40 40 00,  
3.317590e-02,  , Write, 53, 00 04,  
3.326878e-02,  , Read, 53, 03 14 D1 01,  
3.340414e-02,  , Write, 53, 00 04,  
3.349654e-02,  , Read, 53, 03 14 D1 01,  
3.363666e-02,  , Write, 53, 00 06,  
3.372966e-02,  , Read, 53, D1 01 10 55 04 73 68 69 62 62 6F 6C 65 74 68 2E 6C 69 6E 6B,  
3.428787e-02,  , Write, 53, 20 03,  
3.438080e-02,  , Read, 53, 00,  
3.445467e-02,  , Write, 53, 20 03,  

I've inserted a linebreak between the set of operations triggered by write_ndef_message and those triggered by read_ndef_message. As you can see, it's not actually writing the NDEF data in the first section; in fact at 1.350696e-03 you can see that it writes the address where the NDEF data should go to the EEPROM, then writes 0 bytes of data.

I've not yet been able to track down where in the code this is happening, but this shows it's definitely an issue with the code.

[Arachnid]

Using the debugger, I'm able to get as far as the HAL interface with the correct data present for a write - so it seems like this may be some kind of HAL/driver issue? I'm not sure where to go from here, although at this point it's probably not related to this particular example project.

[Arachnid]

I am able to write to the EEPROM using the lower-level I2C library, however - so not a hardware or lower level HAL issue.

[saffetblt]

Hello, I can write NDEF messages to the M24SR integration using this project. The device I am developing needs to run at very low power consumption. The NFC integrated itself consumes power at a very micro-ampere level. But when I activate the library, the power consumption rises to around 3.16mA. There seems to be an issue with the I2C line shutting down. Have you had any experience with this subject?