KVStore: Remove incorrect check with non-default configuration of FLASHIAP

[ccli8]

Summary of changes

This PR tries to fix invalid check in KVstore/FLASHIAP configuration. With non-default configuration of FLASHIAP, the defined FLASHIAP region can be in front of application, not default post-application. The post-application (FLASHIAP_APP_ROM_END_ADDR) check must be skipped.

---

Pull request type

[x] Patch update (Bug fix / Target update / Docs update / Test update / Refactor)
[] Feature update (New feature / Functionality change / New API)
[] Major update (Breaking change E.g. Return code change / API behaviour change)

---

Test results

[] No Tests required for this change (E.g docs only update)
[x] Covered by existing mbed-os tests (Greentea or Unittest)
[] Tests / results supplied as part of this PR

[ciarmcom]

@ccli8, thank you for your changes.
@ARMmbed/mbed-os-storage @ARMmbed/mbed-os-maintainers please review.

[SeppoTakalo]

Why the checks are not fixed, instead of just skipping?

[mbed-ci]

Test run: SUCCESS

Summary: 6 of 6 test jobs passed
Build number : 1
Build artifacts

[ccli8]

Take flash layout of Pelion client application for examples. The FLASHIAP_APP_ROM_END_ADDR check is valid for default flash layout, where KVSTORE is located in back of FLASHIAP_APP_ROM_END_ADDR.

Default flash layout

    +--------------------------+
    |                          |
    |         KVSTORE          |
    |                          |
    +--------------------------+ <-+ storage_tdb_internal.internal_base_address
    |                          |
    |        Free space        |
    |                          |
    +--------------------------+
    |                          |
    |                          |
    |                          |
    |        Active App        |
    |                          |
    |                          |
    |                          |
    +--------------------------+ <-+ mbed-bootloader.application-start-address
    |Active App Metadata Header|
    +--------------------------+ <-+ update-client.application-details
    |                          |
    |        Bootloader        |
    |                          |
    +--------------------------+ <-+ 0

But invalid for non-default flash layout, where KVSTORE is located in front of FLASHIAP_APP_ROM_END_ADDR:

Non-default flash layout

    +---------------------------+ <-+ MBED_ROM_START + MBED_ROM_SIZE
    |        FREE SPACE         |
    +---------------------------+ <-+ FLASHIAP_APP_ROM_END_ADDR
    |                           |
    |                           |
    |        ACTIVE APP         |
    |                           |
    |                           |
    +---------------------------+ <-+ MBED_APP_START =
    +                           |     MBED_ROM_START + target.app_offset
    |ACTIVE APP METADATA HEADER |
    |                           |
    +---------------------------+ <-+ MBED_ROM_START + target.header_offse
    |                           |
    |    KVSTORE (internal)     |
    |                           |
    +---------------------------+ <-+ MBED_ROM_START + MBED_BOOTLOADER_SIZE
    |                           |
    |        BOOTLOADER         |
    |                           |
    +---------------------------+ <-+ MBED_ROM_START

Compared to mbed-os-5.15.1, its _get_blockdevice_FLASHIAP(...) also skips the FLASHIAP_APP_ROM_END_ADDR check for non-default configuration, so it is OK.

[ccli8]

Update?

[0xc0170]

@ARMmbed/mbed-os-storage Please review

[0xc0170]

CI restarted

[mbed-ci]

Test run: SUCCESS

Summary: 6 of 6 test jobs passed
Build number : 2
Build artifacts

[ccli8]

The bug this PR tries to fix can cause mbed-os-example-pelion to fail on all Nuvoton targets. Hopefully, this PR can also merge into mbed-os 5.15 release, in addition to 6.0 release. Per check, the bug has introduced into 5.15.2 and 5.15.3.

[0xc0170]

@ccli8 Thanks for the heads up. Can you backport this to 5.15 branch as well?

[ccli8]

Thanks for the heads up. Can you backport this to 5.15 branch as well?

After further look, this related bug doesn't get into 5.15. Backport is unnecessary.