TRNG enabled. TRNG APIs implemented. REV A/B/C/D flags removed. Warnings removed

[pradeep-gr]

Notes:

• Pull requests will not be accepted until the submitter has agreed to the contributer agreement.
• This is just a template, so feel free to use/remove the unnecessary things

Description

A few sentences describing the overall goals of the pull request's commits.

Status

READY/IN DEVELOPMENT/HOLD

Migrations

If this PR changes any APIs or behaviors, give a short description of what API users should do when this PR is merged.

YES | NO

Related PRs

List related PRs against other branches:

branch	PR
other_pr_production	link
other_pr_master	link

Todos

• Tests
• Documentation

Deploy notes

Notes regarding the deployment of this PR. These should note any
required changes in the build environment, tools, compilers, etc.

Steps to test or reproduce

Outline the steps to test or reproduce the PR here.

…ngs removed.

[sg-]

/morph test-nightly

[mbed-bot]

Result: FAILURE

Your command has finished executing! Here's what you wrote!

/morph test-nightly

Output

mbed Build Number: 0

Build failed!

[sg-]

/morph test-nightly

[mbed-bot]

Result: FAILURE

Your command has finished executing! Here's what you wrote!

/morph test-nightly

Output

mbed Build Number: 0

Test failed!

[bridadan]

The previous test looked good, however a new commit has since been pushed. I'll run another nightly on it

/morph test-nightly

[mbed-bot]

Result: FAILURE

Your command has finished executing! Here's what you wrote!

/morph test-nightly

Output

mbed Build Number: 0

Test failed!

[pradeep-gr]

@mbed-bot Any action needed from my side?

[bridadan]

@mbed-bot: TEST

HOST_OSES=ALL
BUILD_TOOLCHAINS=ALL
TARGETS=NCS36510

[mbed-bot]

[Build 1048]
SUCCESS: Building succeeded and tests were run! Be sure to check the test results

[bridadan]

@pradeep-gr nope, all tests look ok. Just need someone to review now! @0xc0170 Do you have a second to take a look?

[sg-]

@pradeep-gr also need a conflict resolved.

[pradeep-gr]

@sg Conflict resolved

[0xc0170]

Thanks for rebasing.

Note for future: please refrain from mixing non-related changes (removing warnings, adding TRNG, etc). It's more difficult to review/integrate, easier to end up with conflicts.

[0xc0170]

/morph test-nightly

[mbed-bot]

Result: FAILURE

Your command has finished executing! Here's what you wrote!

/morph test-nightly

Output

mbed Build Number: 924

Test failed!

[bridadan]

Tests look OK

[pradeep-gr]

@0xc0170 That was not intentional. Due to difficulty in re-basing.

[pradeep-gr]

Last commit will solve issue #3038

[pradeep-gr]

mbedtls_entropy_self_test is failing on new re-based fork. Find snapshot attached.

[0xc0170]

mbedtls_entropy_self_test is failing on new re-based fork. Find snapshot attached.

Any details for this?

[pradeep-gr]

mbed-os-tests-mbedtls-selftest_result.txt
Attached is test result.

[0xc0170]

@pradeep-gr Have you investigated why it fails?

cc @andresag01 @sbutcher-arm - please look at the test results above (failure)

[pradeep-gr]

@0xc0170 I tried to investigate. But i do not think i have access to some of the files. Eg: This test uses self_test_function() function where i do not have access. Please let me know if there is any HAL dependency. I do not think this test reaching till trng APIs.

[andresag01]

@pradeep-gr: The mbedtls_entropy_self_test() functionality is in https://github.com/ARMmbed/mbed-os/blob/master/features/mbedtls/src/entropy.c#L578. They test the mbedtls_hardware_poll() function which, as far as I understand, is implemented using the TRGN API in mbed OS https://github.com/ARMmbed/mbed-os/blob/master/features/mbedtls/platform/src/mbed_trng.c#L21.

[andresag01]

@pradeep-gr, @0xc0170: Unfortunately, the mbedgt test output contains very little information related to the cause of the failure. The only text I can see related to the failure is:

[1477376794.62][CONN][RXD] :67::FAIL: Expected 1 Was 0
[1477376794.67][CONN][INF] found KV pair in stream: {{__testcase_finish;mbedtls_entropy_self_test;0;
1}}, queued...
[1477376794.76][CONN][RXD] >>> 'mbedtls_entropy_self_test': 0 passed, 1 failed with reason 'Assertio
n Failed'

Could you please provide some more information? I would suggest running a stand-alone application that calls mbedtls_entropy_self_test() and mbedtls_entropy_source_self_test() with argument 1. Then start a debugger or put some print statements inside the self_test() functions to see where it fails.

[pradeep-gr]

@andresag01 @0xc0170 As I mentioned i do not have access to some of the functions used by mbedtls_entropy_self_test() I cannot fetch more debug details. Please try to reproduce at your side. And this test not reaching HAL trng module also.

[andresag01]

@pradeep-gr, @0xc0170:

i do not have access to some of the functions used by mbedtls_entropy_self_test()

I am not sure I understand this, as I posted sometime ago, the mbedtls code is available under mbed-os/features/mbedtls and the functions of interest are https://github.com/ARMmbed/mbed-os/blob/master/features/mbedtls/src/entropy.c#L578.

this test not reaching HAL trng module also

I do not think this is correct. As I explained before, the function mbedtls_hardware_poll() from mbedtls is implemented in mbed OS using the TRNG API. The source code that does this is available at https://github.com/ARMmbed/mbed-os/blob/master/features/mbedtls/platform/src/mbed_trng.c#L21. Furthermore, I manually ran the entropy self tests, and I can confirm that control reaches your TRNG implementation at targets/TARGET_ONSEMI/TARGET_NCS36510/ncs36510_trng_api.c.

The mbedtls entropy self test is divided in two parts. The one that seems to be failing in this PR is mbedtls_entropy_source_self_test(). This part does some simple checks directly in the output of the output of the TRNG API in the case of mbed OS. I printed the entropy gathered during the test and got the following output:

First gather: 00 00 30 a8 00 00 00 80 00 00 00 00 00 00 00 00 
Second gather: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 
Third gather: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 

Clearly, the data gathered does not contain much entropy, and this is the reason why the test is failing.

[pradeep-gr]

@andresag01 @0xc0170 mbed-os-tests-mbedtls-selftest issue fixed. Issue was due to optimization. This test was passing in IAR as IAR is using different optimization level for debug.
Please verify with commit bb824c6

[0xc0170]

@andresag01 Thanks for the help. Can you please verify that now this all passing mbedtls tests?

[bridadan]

/morph test-nightly

[andresag01]

@0xc0170: I have run the mbedtls self test with the latest version of this PR and it seems to pass. However, I would like to clarify that I have NOT reviewed this code and cannot comment on anything regarding its quality and correctness other than the mbedtls_entropy_self_test passes as reported by mbedgt.

[mbed-bot]

Result: SUCCESS

Your command has finished executing! Here's what you wrote!

/morph test-nightly

Output

mbed Build Number: 952

All builds and test passed!

[bridadan]

Yay, 100% passing nightly again! 😄 Can we get this PR merged soon?