Revert "Enable stats reporting with a flag in targets.json"

[VeijoPesonen]

This reverts commit 9dbdbe8 - PR #9072

The reason for reverting the change is that MBED_ROM_START and MBED_ROM_SIZE are needed for configuring bootloader for NRF52840_DK. The reverted change is a workaround for an issue with online compiler when using MCU_NRF52832 and MCU_NRF52840 based boards. The workaround prevents generation of those two defines.

The workaround should be removed by this commit but the original issue with the online tool should be fixed first, if it's still valid. Please see the original workaround commit for further details.

High possibility that this breaks online compiler when the forementioned boards are used.

Description

Pull request type

[ ] Fix
[ ] Refactor
[ ] Target update
[ ] Functionality change
[ ] Docs update
[ ] Test update
[X] Breaking change

Reviewers

@kjbracey-arm
@theotherjimmy
@SeppoTakalo

Release Notes

[ciarmcom]

@VeijoPesonen, thank you for your changes.
@theotherjimmy @SeppoTakalo @kjbracey-arm @ARMmbed/mbed-os-maintainers @ARMmbed/mbed-os-tools please review.

[mark-edgeworth]

FYI @theotherjimmy is no longer working on this stuff so should not be expected to review here.

[0xc0170]

@VeijoPesonen have you tested this patch with the use case it was described in the old issue - softdevice related ?

[VeijoPesonen]

@VeijoPesonen have you tested this patch with the use case it was described in the old issue - softdevice related ?

No, I haven't. I'm able to work around the issue PR #9072 is causing for me so I'm not eager to continue investigation any further.

[0xc0170]

@ARMmbed/mbed-os-tools Please review. I am not able to review this one as do not understand the implications in the online tools. However if this was a fix back in a day, revert needs to be tested if it works or this workaround is still needed.

[adbridge]

@ARMmbed/mbed-os-tools bump (@jamesbeyond fyi)

[mark-edgeworth]

I'm not really convinced by the arguments to include this patch, especially if it breaks the online compiler.

[mark-edgeworth]

The code looks ok, but further testing work is required to ensure that the online compiler still works afterwards. The original description here could imply that these boards will fail with this change in place.

[VeijoPesonen]

I assume the original issue with Online Compiler occurs when the Nordic Semiconductor's Softdevice firmware is in use. But the current configuration of Mbed OS disables SoftDevice - sets it as SOFTDEVICE_NONE. Earlier it was enabled

[adbridge]

@VeijoPesonen what is the state of this PR now? Is it still required ?

[VeijoPesonen]

I can live without this as I'm able to restore the functionality back to original by overwriting the settings modified by the workaround PR #9072. In general I would like to see this revert to be merged as it would remove one of those piling up workarounds which the build system gathers over time. Especially when the workaround is done only for MCU_NRF52832 and MCU_NRF52840 based targets and when SoftDevice is in use. @madchutney for your information.

[adbridge]

@VeijoPesonen Unfortunately I am not willing to accept this without knowing in advance if it breaks the online compiler. We cannot take a PR in knowing it is going to break something....

[0xc0170]

Setting as needs: work. Only @ARMmbed/mbed-os-tools can you help here ?

[0xc0170]

@VeijoPesonen Can you please create an issue for this rather ? Describing the bug as you did above. Once an issue is opened and references this revert, we can close this as is it not going anywhere at the moment.

[VeijoPesonen]

Created issue #11737