Thingy53 - peripheral_lbs sample fails to build for target thingy53/cpuapp/ns

[zpm1066]

Hello,

Attempts to build the "peripheral_lbs" sample for target thingy53/cpuapp/ns in the latest nRF Connect SDK v3.2.1 fails.

The build for target thingy53/cpuapp works fine.

The build failed with error  "‘FLASH overflowed when building TFM image".

/opt/nordic/ncs/toolchains/322ac893fe/opt/zephyr-sdk/arm-zephyr-eabi/bin/../lib/gcc/arm-zephyr-eabi/12.2.0/../../../../arm-zephyr-eabi/bin/ld: address 0x241e0 of bin/tfm_s.axf section `.TFM_UNPRIV_CODE' is not within region `FLASH'
/opt/nordic/ncs/toolchains/322ac893fe/opt/zephyr-sdk/arm-zephyr-eabi/bin/../lib/gcc/arm-zephyr-eabi/12.2.0/../../../../arm-zephyr-eabi/bin/ld: address 0x27c40 of bin/tfm_s.axf section `.gnu.sgstubs' is not within region `FLASH'
/opt/nordic/ncs/toolchains/322ac893fe/opt/zephyr-sdk/arm-zephyr-eabi/bin/../lib/gcc/arm-zephyr-eabi/12.2.0/../../../../arm-zephyr-eabi/bin/ld: bin/tfm_s.axf section `.TFM_PSA_ROT_LINKER_DATA' will not fit in region `FLASH'
/opt/nordic/ncs/toolchains/322ac893fe/opt/zephyr-sdk/arm-zephyr-eabi/bin/../lib/gcc/arm-zephyr-eabi/12.2.0/../../../../arm-zephyr-eabi/bin/ld: address 0x241e0 of bin/tfm_s.axf section `.TFM_UNPRIV_CODE' is not within region `FLASH'
/opt/nordic/ncs/toolchains/322ac893fe/opt/zephyr-sdk/arm-zephyr-eabi/bin/../lib/gcc/arm-zephyr-eabi/12.2.0/../../../../arm-zephyr-eabi/bin/ld: address 0x27c40 of bin/tfm_s.axf section `.gnu.sgstubs' is not within region `FLASH'
/opt/nordic/ncs/toolchains/322ac893fe/opt/zephyr-sdk/arm-zephyr-eabi/bin/../lib/gcc/arm-zephyr-eabi/12.2.0/../../../../arm-zephyr-eabi/bin/ld: region `FLASH' overflowed by 48668 bytes
Memory region         Used Size  Region Size  %age Used
           FLASH:       97308 B      48640 B    200.06%
             RAM:       48404 B       192 KB     24.62%
collect2: error: ld returned 1 exit status
ninja: build stopped: subcommand failed.

I've used the following temporary workaround:

• Disabled “CONFIG_BT_LBS_SECURITY_ENABLED=n" in prj.conf
• Used SB_CONFIG_NETCORE_HCI_IPC=y in sysbuild.conf 
• Removed Kconfig.sysbuild
• Built for thingy53/cpuapp/ns
• Application built and ran fine

However, I'd like to have the BT_LBS_SECURITY_ENABLED set.

What is the fix for thingy53/cpuapp/ns build with the default Kconfig “BT_LBS_SECURITY_ENABLED=y"?


Thank you.

[zpm1066]

Update:

I've re-built the peripheral_lbs for target thingy53/cpuapp/ns for both NETCORE_IPC_RADIO and NETWORE_HCI_IPC, with “CONFIG_BT_LBS_SECURITY_ENABLED=n" in prj.conf.

Configurations and results as follow.

NETCORE_IPC_RADIO

• Disabled “CONFIG_BT_LBS_SECURITY_ENABLED=n" in prj.conf
• Sample Kconfig.sysbuild used (NETCORE_IPC_RADIO_BT_HCI_IPC)
• Built with thingy53/cpuapp/ns
• Application built and ran fine
• In Kconfig.sysbuild, “NRF_DEFAULT_IPC_RADIO” is defined. 
• Where is "NRF_DEFAULT_IPC_RADIO" used? “Kconfig search” in docs.nordicsemi.com has no entry for it.

Console output:

[Disconnected]
[Connected]
[00:00:00.000,793] <inf> udc_nrf: Preinit
*** Booting My Application v3.2.1 - unknown commit ***
*** Using nRF Connect SDK v3.2.1-d8887f6f32df ***
*** Using Zephyr OS v4.2.99-ec78104f1569 ***
[00:00:00.001,342] <inf> udc_nrf: Initialized
Starting Bluetooth Peripheral LBS sample
[00:00:00.005,584] <inf> udc_nrf: SUSPEND state detected
[00:00:00.015,502] <inf> bt_hci_core: HW Platform: Nordic Semiconductor (0x0002)
[00:00:00.015,533] <inf> bt_hci_core: HW Variant: nRF53x (0x0003)
[00:00:00.015,563] <inf> bt_hci_core: Firmware: Standard Bluetooth controller (0x00) Version 155.44175 Build 2923568579
[00:00:00.018,432] <inf> bt_hci_core: HCI transport: IPC
[00:00:00.018,524] <inf> bt_hci_core: Identity: C3:A6:C8:C4:25:42 (random)
[00:00:00.018,554] <inf> bt_hci_core: HCI: version 6.2 (0x10) revision 0x2020, manufacturer 0x0059
[00:00:00.018,585] <inf> bt_hci_core: LMP: version 6.2 (0x10) subver 0x2020
Bluetooth initialized
Advertising successfully started
[00:00:00.103,851] <inf> udc_nrf: Reset
[00:00:00.103,912] <inf> udc: ep 0x80 is not halted|disabled
[00:00:00.103,942] <inf> udc_nrf: RESUMING from suspend
Connected
[00:01:55.334,228] <wrn> bt_l2cap: Ignoring data for unknown channel ID 0x003a
Disconnected, reason 0x13 
Connection object available from previous conn. Disconnect is complete!
Advertising successfully started

NETWORE_HCI_IPC

• Disabled “CONFIG_BT_LBS_SECURITY_ENABLED=n" in prj.conf
• Used SB_CONFIG_NETCORE_HCI_IPC=y in sysbuild.conf 
• Removed Kconfig.sysbuild
• Built with thingy53/cpuapp/ns
• Application built and ran fine

Console Output

Note that this configuration threw error:

[00:00:00.028,961] <err> bt_gatt: Unable to import the key for AES CMAC -134

[Disconnected]
[Connected]
[00:00:00.000,762] <inf> udc_nrf: Preinit
*** Booting My Application v3.2.1 - unknown commit ***
*** Using nRF Connect SDK v3.2.1-d8887f6f32df ***
*** Using Zephyr OS v4.2.99-ec78104f1569 ***
[00:00:00.001,525] <inf> udc_nrf: Initialized
Starting Bluetooth Peripheral LBS sample
[00:00:00.005,767] <inf> udc_nrf: SUSPEND state detected
[00:00:00.015,838] <inf> bt_hci_core: HW Platform: Nordic Semiconductor (0x0002)
[00:00:00.015,869] <inf> bt_hci_core: HW Variant: nRF53x (0x0003)
[00:00:00.015,869] <inf> bt_hci_core: Firmware: Standard Bluetooth controller (0x00) Version 155.44175 Build 2923568579
[00:00:00.018,829] <inf> bt_hci_core: HCI transport: IPC
[00:00:00.018,920] <inf> bt_hci_core: Identity: C3:A6:C8:C4:25:42 (random)
[00:00:00.018,951] <inf> bt_hci_core: HCI: version 6.2 (0x10) revision 0x2020, manufacturer 0x0059
[00:00:00.018,981] <inf> bt_hci_core: LMP: version 6.2 (0x10) subver 0x2020
Bluetooth initialized
Advertising successfully started
[00:00:00.028,961] <err> bt_gatt: Unable to import the key for AES CMAC -134
[00:00:00.102,935] <inf> udc_nrf: Reset
[00:00:00.102,996] <inf> udc: ep 0x80 is not halted|disabled
[00:00:00.103,027] <inf> udc_nrf: RESUMING from suspend
Connected
[00:00:22.549,224] <wrn> bt_l2cap: Ignoring data for unknown channel ID 0x003a
Disconnected, reason 0x13 
Connection object available from previous conn. Disconnect is complete!
Advertising successfully started
Connected
[00:01:00.380,340] <wrn> bt_l2cap: Ignoring data for unknown channel ID 0x003a
Disconnected, reason 0x13 
Connection object available from previous conn. Disconnect is complete!
Advertising successfully started
[Disconnected]

What is the fix for thingy53/cpuapp/ns build with the default Kconfig “BT_LBS_SECURITY_ENABLED=y"?

Is it a case that thingy53/cpuapp/ns build doesn't support BT_LBS_SECURITY_ENABLED?

Thank you.

[Priyanka]

Hi,

I am able to reproduce this issue. I am working on this and will get back to you with a response.

-Priyanka

[zpm1066]

Hi Priyanka,

Thank you. Do you have any updates?

Regards,
Ravi

[Priyanka]

Hi Ravi,

Thingy53 sample with ns support has been removed due to flash overflows for a lot of samples.

And that's the case for this sample as well.

-Priyanka

[zpm1066]

HI Priyanka,

I can perhaps understand this for the samples. However, the Thingy53 has 8MB of external flash, so "thingy53/cpuapp/ns" targets can be run.

The Nordic docs state that thingy53/cpuapp/ns targets are supported.

https://docs.nordicsemi.com/bundle/ncs-latest/page/nrf/app_dev/device_guides/thingy53/building_thingy53.html#board-targets

Can you or one of your colleagues advise on the recommended method to use the Thingy53 external flash to avoid flash overflows, when using nRF Connect SDK 3.2.1 for thingy53/cpuapp/ns? Thanks.

Regards,
Ravi

[Priyanka]

Hi Ravi,

Since the size taken by TF-M has increased a lot, support for certain samples, (especially for the Thingy:53 ns ) has been removed. You can see it in this PR: https://github.com/nrfconnect/sdk-nrf/pull/24059/commits/bd3dc1f94c0d48b2aa0dead466b16ac8a729dc21 

Maybe extending the size of TF-M might help. but this is not something that we have tested, or is in the plans for the time being.

-Priyanka

[zpm1066]

HI Priyanka,

Yes, extending the size of TF-M may help. I'll dig further. Thanks.

Regards,
Ravi

[zpm1066]

Hi Priyanka,

Please keep this ticket open until it has been verified that TF-M size can be extended by using the Thing53 external FLASH. 

I'll give the above a go this week. Thank you.


Regards,
Ravi