SPIF: fixed _address_size to always initialize as 3_BYTES in ::init()

[ndrs-pst]

Summary of changes

Fixed _address_size attribute in blockdevice::COMPONENT_SPIF when using with device that has size bigger than 16 MiB. (e.g. 32 MiB)

In this scenario, during SPIFBlockDevice::init() _address_size will be set to SPIF_ADDR_SIZE_4_BYTES after enumerated by SFDP command. Then calling to deinit() will not restore _address_size back to initial value SPIF_ADDR_SIZE_3_BYTES again.

Result that the following calling sequence will be failed on the 2nd init()
static SPIFBlockDevice spif_bd;
rc = spif_bd.init(); // rc : SPIF_BD_ERROR_OK
rc = spif_bd.deinit(); // rc : SPIF_BD_ERROR_OK
rc = spif_bd.init(); // rc : SPIF_BD_ERROR_PARSING_FAILED

Impact of changes

Migration actions required

Documentation

None

---

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

---

Reviewers

---

[ciarmcom]

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

[ciarmcom]

This pull request has automatically been marked as stale because it has had no recent activity. @ARMmbed/mbed-os-maintainers, @ARMmbed/mbed-os-core, please complete review of the changes to move the PR forward. Thank you for your contributions.

[LDong-Arm]

@ndrs-pst Thanks for the fix. It looks like we always need to start with 3-byte until 4-byte addressing is enabled in _sfdp_detect_and_enable_4byte_addressing(). This change looks good to me.

Given that 3-byte addressing is the "initial state", what about moving _address_size = QSPI_CFG_ADDR_SIZE_24 from constructors into ::init() to start this way every time, rather than using deinit() to change it back?

[ndrs-pst]

@LDong-Arm That's also OK too.
By the way I just realized that the implementation on detect and enable 4 bytes between SPIF vs (QSPIF/OSPIF) are different as follows
SPIF : relying on _address_size value so affect by this fixed.
QSPIF/OSPIF : hard code to use QSPI_CFG_ADDR_SIZE_24 during read SFDP so not affect by this fixed.

[LDong-Arm]

@LDong-Arm That's also OK too.
By the way I just realized that the implementation on detect and enable 4 bytes between SPIF vs (QSPIF/OSPIF) are different as follows
SPIF : relying on _address_size value so affect by this fixed.
QSPIF/OSPIF : hard code to use QSPI_CFG_ADDR_SIZE_24 during read SFDP so not affect by this fixed.

If we can use the constant QSPI_CFG_ADDR_SIZE_24 for operations that only support 3-byte and _address_size elsewhere, that would be better. But this requires some testing. Always starting with 3-byte is a relatively safe approach.

What target and what type of storage are you using?

[ndrs-pst]

@LDong-Arm Your approach would be safer then.
About your question :
We are using W25R256JWEIQ (1.8V), W25R256JVEIQ (3.3V) 256 Mb NOR Flash from Winbond via SPIF in our custom target (STM32F411CE).
So it would be limit to test QSPIF (4-bytes addressing) for now.

[LDong-Arm]

@LDong-Arm Your approach would be safer then.
About your question :
We are using W25R256JWEIQ (1.8V), W25R256JVEIQ (3.3V) 256 Mb NOR Flash from Winbond via SPIF in our custom target (STM32F411CE).
So it would be limit to test QSPIF (4-bytes addressing) for now.

Thanks for the info. Could you please either make the suggested change (set _address_size in init()), or use the constant QSPI_CFG_ADDR_SIZE_24 in SPIF (which you can test) in a similar way to QSPIF/OSPIF if it works and update the PR? Thanks in advance.

[LDong-Arm]

(Requesting change)

Pull request has been modified.

[ndrs-pst]

@LDong-Arm Just push more commit, please review. 😃
🚀 🗼

[LDong-Arm]

@LDong-Arm Just push more commit, please review. 😃
🚀 🗼

Thanks @ndrs-pst, it looks good now. Could you squash the two commits, and make sure QSPIF/OSPIF are not touched (currently just an empty line is removed from each) so only SPIF is changed? Also the PR description can be updated to include SPIF only.

[ndrs-pst]

@LDong-Arm Just squash the commit and change PR comment

[LDong-Arm]

LGTM

[0xc0170]

CI started

[mbed-ci]

Jenkins CI Test : ✔️ SUCCESS

Build Number: 1 | 🔒 Jenkins CI Job | 🌐 Logs & Artifacts

CLICK for Detailed Summary

jobs	Status
jenkins-ci/mbed-os-ci_unittests	✔️
jenkins-ci/mbed-os-ci_cmake-cloud-example-ARM	✔️
jenkins-ci/mbed-os-ci_build-greentea-ARM	✔️
jenkins-ci/mbed-os-ci_build-cloud-example-ARM	✔️
jenkins-ci/mbed-os-ci_cmake-cloud-example-GCC_ARM	✔️
jenkins-ci/mbed-os-ci_build-cloud-example-GCC_ARM	✔️
jenkins-ci/mbed-os-ci_build-greentea-GCC_ARM	✔️
jenkins-ci/mbed-os-ci_cmake-example-ARM	✔️
jenkins-ci/mbed-os-ci_build-example-ARM	✔️
jenkins-ci/mbed-os-ci_cmake-example-GCC_ARM	✔️
jenkins-ci/mbed-os-ci_build-example-GCC_ARM	✔️
jenkins-ci/mbed-os-ci_greentea-test	✔️