NRF52 SDK15 update #9019

TacoGrandeTX · 2018-12-07T20:53:56Z

Description

Add support for Nordic nRF5x SDK15.0

Pull request type

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

pan- · 2018-12-09T12:37:38Z

@cmonr We can look at BLE changes included in this PR but I think you should require a review from the Hal team as most changes are in the HAL.

cmonr · 2018-12-10T20:57:18Z

@JanneKiiskila Would you and your team like to review this as well?
This isn't coming into master yet, but figured I'vd give y'all the change to start looking at what's here now.

targets/TARGET_NORDIC/TARGET_NRF5x/TARGET_SDK_15_0/TARGET_SOFTDEVICE_S112/headers/ble_types.h

0xc0170 · 2018-12-14T14:48:18Z

targets/TARGET_NORDIC/TARGET_NRF5x/TARGET_NRF52/common_rtc.h

@@ -0,0 +1,53 @@
+/* mbed Microcontroller Library
+ * Copyright (c) 2015 ARM Limited


add SPDX identifier and fix the year (new files 2018), also the company name

* Copyright (c) 2018, Arm Limited and affiliates. * SPDX-License-Identifier: Apache-2.0

@0xc0170 Is the following commit appropriate: bc4aed0 ? I've excluded any old files we didn't touch such as itm_api.c. Would you want the company name and SPDX added for those files (and not touch the date)? I'll then follow this model for any other Arm source files located under the TARGET_NORDIC tree.

Please follow for any, would be good 👍

Thanks

bulislaw

What testing did you perform? As a minimum I would imagine all the tests on all base nrf boards + BLE testing should be performed locally.

targets/targets.json

bulislaw · 2018-12-18T11:38:01Z

...T_NORDIC/TARGET_NRF5x/TARGET_NRF52/TARGET_MCU_NRF52832/device/TOOLCHAIN_ARM_STD/nRF52832.sct

@@ -9,17 +9,6 @@
  #define MBED_APP_SIZE 0x80000
 #endif

-/* If softdevice is present, set aside space for it */


What has changed that we don't need to block space for softdevice anymore?

The new memory stats collection (specifically #8607) broke Nordic SD builds as it defines MBED_RAM_START so the scatterfile needed to change. We now define MBED_RAM_START and MBED_RAM_SIZE in the library json file when building with the SD:

https://github.com/TacoGrandeTX/mbed-os/blob/chile/targets/TARGET_NORDIC/TARGET_NRF5x/TARGET_SDK_15_0/TARGET_SOFTDEVICE_S132_FULL/mbed_lib.json#L37-L39

This works out good because if one changes the BLE config the RAM requirements might also change and then they only need to touch one file.

We have only done on NRF52_DK and NRF52840_DK for now, I think that means we should add NRF51_DK to the verification, thank you for comments on this.

@desmond-blue SDK15 doesn't include support for the NRF51 so there is no need to verify on the NRF51_DK.

0Grit · 2018-12-18T14:56:41Z

Any chance of TWIM and SPIM support?

pan-

So, I wanted to try and test ble but I faced several issues with that PR:

Serial reception doesn't work with this PR in our test application; looks like the RX interrupt is not triggered.
The following files should enable the BLE feature (reference here):

targets/TARGET_NORDIC/TARGET_NRF5x/TARGET_SDK_15_0/TARGET_SOFTDEVICE_S132_FULL/mbed_lib.json
targets/TARGET_NORDIC/TARGET_NRF5x/TARGET_SDK_15_0/TARGET_SOFTDEVICE_S132_OTA/mbed_lib.json
targets/TARGET_NORDIC/TARGET_NRF5x/TARGET_SDK_15_0/TARGET_SOFTDEVICE_S140_FULL/mbed_lib.json   targets/TARGET_NORDIC/TARGET_NRF5x/TARGET_SDK_15_0/TARGET_SOFTDEVICE_S140_OTA/mbed_lib.json

elm3 · 2019-01-09T17:28:49Z

Hello Vincent, Could you add the a link or reference to the test application where you see the issue? We would really like to get this closed. Thanks, Earl

pan- · 2019-01-11T18:01:55Z

Here is a small snippet that reproduces the issue:

#include <stdio.h>

int main(void)
{
    while (true) {
        int c = getc(stdin);
        putc(c, stdout);
    }
    return 0;
}

@donatieng Maybe it would be useful to have a test as simple as this one to check that UART works.

My own test case was a bit mot complex and involved IRQ and custom RawSerial; you can access it here: https://gist.github.com/pan-/9c836b51116aa8e9c3c262dca60c17b5 .

Will need rereview

naveenkaje · 2019-01-14T21:18:20Z

Addressed the serial issue with https://github.com/TacoGrandeTX/mbed-os/pull/6

TacoGrandeTX · 2019-01-23T00:41:04Z

Performed required rebase after test infrastructure broke resulting in a test failure for commit a770d03.

Greentea results are all pass for nRF52_DK and nRF52840_DK with the the exception of mbed_hal-trng. This test also fails for my host system on master... the test performs a physical reset that doesn't play well with my laptop (results in a timeout). Test results are attached for builds with the Arm compiler.

greentea_log-arm_compiler.zip

cmonr · 2019-01-23T04:06:23Z

@TacoGrandeTX Looking at the feature branch itself, since if this PR doesn't modify/update .travis.yml, it and any PRs opened against the feature branch will be hit with the same issues that Mbed OS faced last week.

Fortunately since the feature-branch doesn't have any unique commits, rebasing it was not difficult.

This will need another rebase (sorry), but after that, it should be back to normal, testing-wise.

…feature defines

- top level files ported from TARGET_NORDIC/TARGET_NRF5x/ Also addressed: - fixed linking issue for gcc - added support for nRF52-DK builds, but reverted to using nRF52840 sdk_config.h (must be updated) - introduced "RTC" to targets.json (might need to be removed eventually)

Also addressed: - removed dependency on legacy config (excluded apply_old_config.h) - removed legacy pwm and saadc headers - Arm Compiler 5 linking issue (a band-aid for now... needs to be properly addressed for peripheral sharing) - added missing header in SoftDevice file

- Missed some NRFX defines that needed to changed - Set PWM base clock to 125kHz (needs to be reverted back to 1 MHz) - Updated sdk_config.h for nRF52_DK builds - Brought in updates from PR7779 (fix for nRF52 PWM issues)

- Add MBR, NONE and OTA SoftDevice build options for S132 and S140 - Add S112 SoftDevice (single build option) - Some folder restructuring in TARGET_SOFTDEVICE_COMMON was required

This reverts commit 3d2fa53. This was a breaking change for the "MBR" and "NONE" builds. After testing it was also determined that copying the vector table a second time wasn't required for the "FULL" build.

* Update TARGET_NRF5x/README.md to improve "Changing SoftDevice" section and added section on NRF52840 CryptoCell310 Support * Update the file list in TARGET_SDK_15_0/TARGET_SOFTDEVICE_COMMON/README.md * Add missing CR-LF to Nordic-provided SDK file * Rename a header file in the TARGET_SOFTDEVICE_S112 tree

- Brought in new nrfx APIs - Brought in PPI additions - Removed dead code for RTC

- Adjust memory for SoftDevice - Enable PRIO=5 for interrupt priority check - Change NRF_SD_BLE_API_VERSION to 6 - Add handle and buffer for advertising and scanning - Remove guard for phy update - Change scatter files and mbed_lib.json for PR #8607

- Update (c) year and company name - Add SPDX-License-Identifier - Impacted files: - Files touched for SDK15 port - Files added in 2018

Move the check from preprocessor to the if condition

TacoGrandeTX · 2019-01-23T23:19:09Z

@cmonr OK - thanks for that. New rebase has been performed and testing looks good as before (attaching greentea reports). The serial issue has been resolved so now @pan- can continue his review.

gt_log-arm_compiler.zip.zip

Fix nrf section iter ARMCC compiler macro check

Please re-review. Initial comments addressed.

cmonr · 2019-01-30T22:14:30Z

@TacoGrandeTX Fyi: #9358

0xc0170 · 2019-02-02T08:08:54Z

@elm3 Please if PR needs merging talk to @ARMmbed/mbed-os-maintainers , this should not have been merged.

TacoGrandeTX · 2019-02-06T20:09:33Z

@ARMmbed/mbed-os-maintainers Yes we are waiting for this to be merged. What is the gating item?

TacoGrandeTX · 2019-02-06T22:48:59Z

@cmonr I'm not expecting any impact to the SDK update from #9358. On the driver side the SDK update is largely a change in driver structure (not much has changed in functionality or implementation). Likewise the new SoftDevice won't be impacted either.

TacoGrandeTX requested a review from marcuschangarm December 7, 2018 21:13

cmonr changed the title ~~feature-nrf52-sdk15~~ NRF52 SDK15 update Dec 8, 2018

cmonr added the needs: review label Dec 8, 2018

cmonr requested a review from a team December 9, 2018 01:09

0xc0170 requested a review from a team December 10, 2018 12:57

cmonr added the release-version: feature branch label Dec 10, 2018

0xc0170 previously requested changes Dec 14, 2018

View reviewed changes

0xc0170 added needs: work and removed needs: review labels Dec 14, 2018

bulislaw reviewed Dec 18, 2018

View reviewed changes

pan- previously requested changes Dec 19, 2018

View reviewed changes

cmonr mentioned this pull request Dec 27, 2018

nrf52: Periodic wakeups every 3ms #9093

Closed

3 tasks

0xc0170 approved these changes Jan 17, 2019

View reviewed changes

TacoGrandeTX and others added 8 commits January 23, 2019 10:08

Baseline legacy build using 14.2 drivers.

5a921a4

Mods for baseline 15.0 support

0ec3da7

Bringing in Softdevice but excluding BLE feature and improvements in …

94b6791

…feature defines

Updates to PWM for IOTDEV-1573

cb9a842

- Missed some NRFX defines that needed to changed - Set PWM base clock to 125kHz (needs to be reverted back to 1 MHz) - Updated sdk_config.h for nRF52_DK builds - Brought in updates from PR7779 (fix for nRF52 PWM issues)

serial_api: Fixes to compile with SDK15.0

6220290

serial_api: add changes to port from Softdevice 14.2 to 15.0

cd0867a

Ralph F and others added 16 commits January 23, 2019 10:12

Add missing SoftDevice build support (MBR, NONE, OTA)

fa09de9

- Add MBR, NONE and OTA SoftDevice build options for S132 and S140 - Add S112 SoftDevice (single build option) - Some folder restructuring in TARGET_SOFTDEVICE_COMMON was required

Revert "NRF52: Fix vector table"

a3ab4eb

This reverts commit 3d2fa53. This was a breaking change for the "MBR" and "NONE" builds. After testing it was also determined that copying the vector table a second time wasn't required for the "FULL" build.

Enabling NRFX_PPI_ENABLED for serial_api updates

b2ce612

Changes required for rebase:

074d5f0

- Brought in new nrfx APIs - Brought in PPI additions - Removed dead code for RTC

Set the PWM clock prescaler back to default (1MHz clock)

8d97ab6

NRF52: serial: fix build warnings

56134b4

NRF52: Cleanup unused code

f484ebb

Fix timer1 irq configuration

5e9cb64

Fix SoftDevice configuration on NRF52_DK

fa4c1b8

Remove duplicate file

0fe61d3

Remove RTC from nRF52840_DK and DELTA_DFCM_NNN50 (was testing)

9098477

Update file header at TARGET_NRF52 level

73e490e

- Update (c) year and company name - Add SPDX-License-Identifier - Impacted files: - Files touched for SDK15 port - Files added in 2018

Add BLE feature to mbed_lib.json of SoftDevice

a177fe7

NRF52: reloc_vector_table: check the RTS pin settings in code

ef632c8

Move the check from preprocessor to the if condition

desmond-blue and others added 2 commits January 24, 2019 16:53

Fix nrf section iter macro check

a883bf7

Merge pull request #7 from desmond-blue/fix-nrf-section-iter

360b099

Fix nrf section iter ARMCC compiler macro check

kjbracey mentioned this pull request Jan 28, 2019

NRF5x: rationalise sleep+critical #9358

Closed

cmonr added needs: review and removed needs: work labels Jan 30, 2019

elm3 merged commit d179c9d into ARMmbed:feature-nrf52-sdk15 Feb 1, 2019

studavekar removed the needs: review label Feb 1, 2019

bearsh mentioned this pull request Feb 11, 2019

nrf52: after disconnect, device doesn't go into sleep again #9276

Closed

3 tasks

		@@ -0,0 +1,53 @@
		/* mbed Microcontroller Library
		* Copyright (c) 2015 ARM Limited

NRF52 SDK15 update #9019

NRF52 SDK15 update #9019

Uh oh!

Conversation

TacoGrandeTX commented Dec 7, 2018

Description

Pull request type

Uh oh!

pan- commented Dec 9, 2018

Uh oh!

cmonr commented Dec 10, 2018

Uh oh!

Uh oh!

0xc0170 Dec 14, 2018

Choose a reason for hiding this comment

Uh oh!

TacoGrandeTX Dec 19, 2018 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

0xc0170 Dec 20, 2018

Choose a reason for hiding this comment

Uh oh!

bulislaw left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

bulislaw Dec 18, 2018

Choose a reason for hiding this comment

Uh oh!

TacoGrandeTX Dec 18, 2018

Choose a reason for hiding this comment

Uh oh!

desmond-blue Dec 18, 2018

Choose a reason for hiding this comment

Uh oh!

TacoGrandeTX Dec 19, 2018

Choose a reason for hiding this comment

Uh oh!

0Grit commented Dec 18, 2018

Uh oh!

pan- left a comment

Choose a reason for hiding this comment

Uh oh!

elm3 commented Jan 9, 2019 via email • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

pan- commented Jan 11, 2019

Uh oh!

naveenkaje commented Jan 14, 2019

Uh oh!

TacoGrandeTX commented Jan 23, 2019

Uh oh!

cmonr commented Jan 23, 2019

Uh oh!

TacoGrandeTX commented Jan 23, 2019

Uh oh!

cmonr commented Jan 30, 2019

Uh oh!

0xc0170 commented Feb 2, 2019

Uh oh!

TacoGrandeTX commented Feb 6, 2019

Uh oh!

TacoGrandeTX commented Feb 6, 2019

Uh oh!

Uh oh!

TacoGrandeTX Dec 19, 2018 •

edited

Loading

elm3 commented Jan 9, 2019 via email •

edited

Loading