Remove redundant ISR test.

[mprse]

Description

As a result of discussion with @bulislaw we agreed that ISR test is redundant and can be removed since ticker test provides scenario (#5233) which proves that interrupts are correctly handled:

• ticker ISR is executed in the expected point of time and it is executed only once.

Status

READY

Migrations

NO

Related PRs

ARMmbed:feature-hal-spec-ticker | #5233

[0xc0170]

I restarted jenkins CI (not relevant failure there).

@mprse Once approved this will be tested. Is the destination branch up to date (expected to have CI green!) ?

[mprse]

I restarted jenkins CI (not relevant failure there).

@mprse Once approved this will be tested. Is the destination branch up to date (expected to have CI green!) ?

Honestly I don't know how to check if this branch (ARMmbed:feature-hal-spec-ticker) is up to date and what does it mean. I have created a few PRs to this branch which are related to new ticker implementation and testing. Could you please provide more information what you expect me to do?

[tommikas]

@0xc0170

Is the destination branch up to date (expected to have CI green!) ?

I'm pretty sure it's out of date. jenkins/pr-head is failing on a build that requires mbed OS 5.6.1 or newer.

Specifically it's failing with IAR on this line:
extern SDBlockDevice sd(MBED_CONF_SD_SPI_MOSI, MBED_CONF_SD_SPI_MISO, MBED_CONF_SD_SPI_CLK, MBED_CONF_SD_SPI_CS);

with error message:
Error[Pe322]: object of abstract class type "SDBlockDevice" is not allowed:
pure virtual function "BlockDevice::erase" has no overrider
pure virtual function "BlockDevice::get_erase_size" has no overrider

[0xc0170]

@tommikas The destination branch was fast forwarded yesterday. It should be up to date and green. I restarted the job. Does jenkins CI merge with master (I cant locate the command in the logs).

[0xc0170]

/morph build

[tommikas]

Does jenkins CI merge with master (I cant locate the command in the logs).

It does. The pr head is passed as a pull/<change ID>/head reference which is then used to do the merge like this:

git fetch pull/5326/head
git checkout origin/feature-hal-spec-ticker
git merge FETCH_HEAD

It's done the same way both in the main os-cliapp build as well as the failing sub-build. Just verified that too. I also noticed that it's actually not just IAR that is failing. gcc and armcc failed too.

I'm seeing this warning in the logs as well: https://github.com/ARMmbed/sd-driver/blob/master/SDBlockDevice.cpp#L156

It's possible that Jenkins missed the notification from the latest push. Lets see what the current build says.

[mbed-ci]

Build : SUCCESS

Build number : 242
Build artifacts/logs : http://mbed-os.s3-website-eu-west-1.amazonaws.com/?prefix=builds/5326/

Skipping test trigger, missing label 'NEED CI'

[studavekar]

/morph build

[mbed-ci]

Build : SUCCESS

Build number : 281
Build artifacts/logs : http://mbed-os.s3-website-eu-west-1.amazonaws.com/?prefix=builds/5326/

Triggering tests

/morph test
/morph uvisor-test

[mbed-ci]

Test : SUCCESS

Build number : 127
Test logs :http://mbed-os-logs.s3-website-us-west-1.amazonaws.com/?prefix=logs/5326/127

[mbed-ci]

Build : FAILURE

Build number : 297
Build artifacts/logs : http://mbed-os.s3-website-eu-west-1.amazonaws.com/?prefix=builds/5326/

[c1728p9]

Shouldn't this wait until #5233 is merged?

[bulislaw]

Ideally, but it doesn't really matter that much. Let's not revert it.

[c1728p9]

This was pulled into the release-candidate branch since it has the tag release-version: 5.6.3, but it isn't on master. This should either be removed from the release-candidate branch or moved into master.

[c1728p9]

CC @0xc0170 @adbridge

[0xc0170]

I'll create a new patch now

[adbridge]

Arggh completely missed the fact this went to a feature branch! It could have just been removed from the release-candidate though....