Release candidate for mbed-os-5.13.0-rc3 #10818

adbridge · 2019-06-12T15:22:49Z

No description provided.

The BLE stack from SoftDevice is not actively maintained and has issues when used with Nordic SDK v15.

APP interrupts are masked on SoftDevice BLE stack only, but we have switched to Cordio stack on NRF52 targets.

The config "cryptocell310-acceleration" is set by MCU_NRF52840 but individual targets may have crytocell310 feature disabled.

Minor tweaks to fix ARM C 5 compatibility. Pushing "ns_list.h" include to first makes sure "ns_types.h" is included first, meaning it gets to define `__STDC_LIMIT_MACROS` before the first include of <stdint.h>, which ensures that UINT8_MAX etc are defined.

Prevent compilation issues when someone has included <stdint.h> before a header file that needs to include <ns_list.h>. Some toolchains like ARM C 5 will not provide UINT_FAST8_MAX in C++ unless __STDC_LIMIT_MACROS is defined, and if this was not defined the first time <stdint.h> was included, it's too late. We can get the maximum value for our unsigned list offset by casting -1 to it, thanks to modulo arithmetic.

When using Mbed Crypto's PSA Entropy Injection feature on Mbed OS, it is not required to opt out of having entropy sources added to your entropy contexts by default (via MBEDTLS_NO_DEFAULT_ENTROPY_SOURCES). As integrated in Mbed OS, MBEDTLS_PSA_INJECT_ENTROPY is compatible with actual entropy sources. PSA entropy injection is implemented using the standard Mbed TLS NV Seed feature, and is as compatible with other entropy sources as the standard Mbed TLS NV Seed feature which does support entropy mixing.

Mbed TLS now enables PSA APIs by default on all targets. It's not necessary to explicitly enable MBEDTLS_PSA_CRYPTO_C, as that can be gotten from the Mbed TLS config.h. However, many PSA targets depend on `-DMBEDTLS_PSA_CRYPTO_C` being defined by the Mbed OS json configuration system and are not yet properly including the Mbed TLS configuration; for these PSA targets, warnings may remain until this issue is fixed. Avoiding re-definition will eliminate warnings like the following, when building mbed-os-example-blinky: Compile [ 14.5%]: pal_client_api_empty_intf.c [Warning] pal_client_api_intf.h@35,0: "PSA_SUCCESS" redefined Compile [ 14.6%]: pal_client_api_intf.c Compile [ 14.7%]: DeviceKey.cpp Compile [ 14.9%]: pal_internal_trusted_storage_intf.c [Warning] pal_internal_trusted_storage_intf.c@45,9: 'psa_its_set' is deprecated: PS specific types should not be used [-Wdeprecated-declarations] Compile [ 15.3%]: val_attestation.c [Warning] client.h@40,0: "PSA_VERSION_NONE" redefined <..> Compile [ 33.3%]: asn1parse.c [Warning] config.h@2838,0: "MBEDTLS_PSA_CRYPTO_C" redefined Compile [ 33.5%]: aes.c [Warning] config.h@2838,0: "MBEDTLS_PSA_CRYPTO_C" redefined Compile [ 33.6%]: asn1write.c [Warning] config.h@2838,0: "MBEDTLS_PSA_CRYPTO_C" redefined Compile [ 33.7%]: psa_crypto.c [Warning] config.h@2838,0: "MBEDTLS_PSA_CRYPTO_C" redefined Compile [ 33.8%]: blowfish.c [Warning] config.h@2838,0: "MBEDTLS_PSA_CRYPTO_C" redefined Compile [ 33.9%]: camellia.c [Warning] config.h@2838,0: "MBEDTLS_PSA_CRYPTO_C" redefined Compile [ 34.0%]: base64.c [Warning] config.h@2838,0: "MBEDTLS_PSA_CRYPTO_C" redefined Compile [ 34.2%]: ccm.c [Warning] config.h@2838,0: "MBEDTLS_PSA_CRYPTO_C" redefined Compile [ 34.3%]: chacha20.c [Warning] config.h@2838,0: "MBEDTLS_PSA_CRYPTO_C" redefined Compile [ 34.4%]: chachapoly.c [Warning] config.h@2838,0: "MBEDTLS_PSA_CRYPTO_C" redefined Compile [ 34.5%]: cipher_wrap.c [Warning] config.h@2838,0: "MBEDTLS_PSA_CRYPTO_C" redefined Compile [ 34.6%]: cmac.c [Warning] config.h@2838,0: "MBEDTLS_PSA_CRYPTO_C" redefined Compile [ 34.7%]: cipher.c [Warning] config.h@2838,0: "MBEDTLS_PSA_CRYPTO_C" redefined Compile [ 34.9%]: bignum.c [Warning] config.h@2838,0: "MBEDTLS_PSA_CRYPTO_C" redefined Compile [ 35.0%]: des.c [Warning] config.h@2838,0: "MBEDTLS_PSA_CRYPTO_C" redefined Compile [ 35.1%]: dhm.c [Warning] config.h@2838,0: "MBEDTLS_PSA_CRYPTO_C" redefined Compile [ 35.2%]: ctr_drbg.c <..> Compile [ 70.9%]: EthernetInterface.cpp [Warning] config.h@2838,0: "MBEDTLS_PSA_CRYPTO_C" redefined Compile [ 71.0%]: InternetSocket.cpp Compile [ 71.1%]: L3IPInterface.cpp [Warning] config.h@2838,0: "MBEDTLS_PSA_CRYPTO_C" redefined Compile [ 71.2%]: NetworkInterface.cpp

adbridge · 2019-06-12T15:29:38Z

CI started

mbed-ci · 2019-06-12T18:22:02Z

Test run: FAILED

Summary: 1 of 11 test jobs failed
Build number : 1
Build artifacts

Failed test jobs:

jenkins-ci/mbed-os-ci_greentea-test

0xc0170 · 2019-06-13T10:09:02Z

restarted test job

softagram-bot · 2019-06-13T12:28:41Z

This impact report was requested by @TommiTallgren - get yours at https://softagram.com/pull-request-bot

Softagram Impact Report for pull/10818 (head commit: `bf676c1`)

Target: K64F

⭐ Change Overview

(Open in Softagram Desktop for full details)

⭐ Details of Dependency Changes

(Open in Softagram Desktop for full details)

📄 Full report

Permalink: Full report for pull/10818

Give feedback on this report to [email protected]

0xc0170 · 2019-06-13T13:13:03Z

@LDong-Arm the only nrf52 change here is uart enablement, could this affect lpticker? Can you retest quickly the lpticker test (can also fetch the binaries from this PR). It's only ARMCC6

The retest failed as well the same test case :/

LDong-Arm · 2019-06-13T15:11:02Z

@LDong-Arm the only nrf52 change here is uart enablement, could this affect lpticker? Can you retest quickly the lpticker test (can also fetch the binaries from this PR). It's only ARMCC6
The retest failed as well the same test case :/

Yes the failure can be reproduced locally on NRF52_DK and NRF52840_DK with ARMC6

LDong-Arm · 2019-06-13T15:33:19Z

@0xc0170 Yes, the issue is gone when the UART change (#10753) is reverted. I'm trying to figure out the reason.

0xc0170 · 2019-06-14T06:36:33Z

@desmond-blue Any updates?

desmond-blue · 2019-06-14T07:18:10Z

I am able to reproduce it, but have no idea so far...still investigating.

adbridge · 2019-06-14T10:43:42Z

@desmond-blue any progress?

kimlep01 · 2019-06-14T14:14:12Z

I tested locally with NRF52840_DK and ARMC6:
with latest master test passes
with this release-candidate branch test fails

desmond-blue · 2019-06-14T15:31:13Z

Tried adding timer.reset() before start the timer and the test passes on my local test. It looks like not related to UART configuration, still digging.

desmond-blue · 2019-06-17T06:51:14Z

I changed the alignment bytes of delay_loop_code refer here, and the test passes locally.
Since the comments say "Timing seems to depend on alignment", would anyone know if this change is acceptable and how this affects the timing, thanks.

adbridge · 2019-06-18T10:17:16Z

Restarted ci

OPpuolitaival · 2019-06-18T12:35:58Z

I needed to restart the greentea test

mbed-ci · 2019-06-18T12:39:22Z

Test run: FAILED

Summary: 2 of 11 test jobs failed
Build number : 2
Build artifacts

Failed test jobs:

jenkins-ci/mbed-os-ci_exporter
jenkins-ci/mbed-os-ci_greentea-test

adbridge · 2019-06-18T13:54:53Z

Restarted exporters

adbridge · 2019-06-19T08:51:01Z

Note: Test results for this PR contain one issue #10849 .
This will be taken as a known issue for the release. CI results have been updated accordingly to allow this PR to be merged.

Patater and others added 13 commits June 12, 2019 16:15

mbed-crypto: Update to Mbed Crypto 1.1.0d1

465944b

Re-enable complilation for wifi and nanostack examples

14ed96e

Default to Cordio BLE stack for NRF52* targets

2944654

The BLE stack from SoftDevice is not actively maintained and has issues when used with Nordic SDK v15.

critical_section tests: do not check APP interrupts on NRF52

ad91527

APP interrupts are masked on SoftDevice BLE stack only, but we have switched to Cordio stack on NRF52 targets.

NORDIC_CORDIC pal_crypto: check if cryptocell310 is enabled

0607340

The config "cryptocell310-acceleration" is set by MCU_NRF52840 but individual targets may have crytocell310 feature disabled.

exculde test

3b66065

Fix ARMC5 compilation

3a2bd24

Minor tweaks to fix ARM C 5 compatibility. Pushing "ns_list.h" include to first makes sure "ns_types.h" is included first, meaning it gets to define `__STDC_LIMIT_MACROS` before the first include of <stdint.h>, which ensures that UINT8_MAX etc are defined.

Fix nrf52 enabled uart count and enable uart0/1

0ebc053

mbedtls: Update to Mbed Crypto 1.1.0d2

0689c58

mbedtls: Update to Mbed TLS 2.18.0-rc3

c0b4343

adbridge requested a review from 0xc0170 June 12, 2019 15:28

adbridge added the needs: CI label Jun 12, 2019

0xc0170 approved these changes Jun 13, 2019

View reviewed changes

pan- mentioned this pull request Jun 17, 2019

TEST: Mbed_platform_wait_ns fail on NRF52* when compiled with ARMC6 #10849

Closed

ARMmbed deleted a comment from OPpuolitaival Jun 19, 2019

adbridge merged commit 92a58df into mbed-os-5.13 Jun 19, 2019

0xc0170 removed the needs: CI label Jul 3, 2019

Release candidate for mbed-os-5.13.0-rc3 #10818

Release candidate for mbed-os-5.13.0-rc3 #10818

Uh oh!

Conversation

adbridge commented Jun 12, 2019

Uh oh!

adbridge commented Jun 12, 2019

Uh oh!

mbed-ci commented Jun 12, 2019

Test run: FAILED

Uh oh!

0xc0170 commented Jun 13, 2019

Uh oh!

softagram-bot commented Jun 13, 2019

Softagram Impact Report for pull/10818 (head commit: bf676c1)

Target: K64F

⭐ Change Overview

⭐ Details of Dependency Changes

📄 Full report

Uh oh!

0xc0170 commented Jun 13, 2019 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

LDong-Arm commented Jun 13, 2019

Uh oh!

LDong-Arm commented Jun 13, 2019

Uh oh!

0xc0170 commented Jun 14, 2019

Uh oh!

desmond-blue commented Jun 14, 2019

Uh oh!

adbridge commented Jun 14, 2019

Uh oh!

kimlep01 commented Jun 14, 2019 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

desmond-blue commented Jun 14, 2019

Uh oh!

desmond-blue commented Jun 17, 2019

Uh oh!

adbridge commented Jun 18, 2019

Uh oh!

OPpuolitaival commented Jun 18, 2019

Uh oh!

mbed-ci commented Jun 18, 2019

Test run: FAILED

Uh oh!

adbridge commented Jun 18, 2019

Uh oh!

adbridge commented Jun 19, 2019

Uh oh!

Uh oh!

Softagram Impact Report for pull/10818 (head commit: `bf676c1`)

0xc0170 commented Jun 13, 2019 •

edited

Loading

kimlep01 commented Jun 14, 2019 •

edited

Loading