subsys: usb: shell: Added pytest#6
subsys: usb: shell: Added pytest#6roma-jam wants to merge 6 commits intopr-add-dwc2-uhc-channels-initfrom
Conversation
Added: - Driver API signatures without implementation - DT miltiinstance definition of uhc dwc2 - Stubbed vendor quirks calls for esp32 - Added the esp32s3 overlay for usb shell example for testing purpose Signed-off-by: Roman Leonov <jam_roma@yahoo.com>
- Added esp32 specific vendor PHY configuration - Added specific esp32 IRQ configuration Signed-off-by: Roman Leonov <jam_roma@yahoo.com>
- Added port and channel events handler - Added logic for port conn/dconn debouncing - Added stubs for the future higher layer notification Signed-off-by: Roman Leonov <jam_roma@yahoo.com>
Added: - thread channel events CPLT, STALL, DO_RELEASE, ERROR - channel structure, configuration: claim, submit_ctrl, release - FIFO pre-calculaction on init and configuration on port reset Signed-off-by: Roman Leonov <jam_roma@yahoo.com>
roma-jam
force-pushed
the
pr-add-subsys-usb-shell-test
branch
from
d33aa8c to
f92073d
Compare
roma-jam
changed the title
roma-jam
changed the title
roma-jam
force-pushed
the
pr-add-subsys-usb-shell-test
branch
4 times, most recently
from
801c8b0 to
0a028a5
Compare
roma-jam
force-pushed
the
pr-add-subsys-usb-shell-test
branch
2 times, most recently
from
9933511 to
13ee027
Compare
Added: - host_prj.conf for the applicaiton (to enable usbh driver on esp32) - test_usbh_shell.py with basic logical commands Changed: - Added the sample.usbh.shell.hil entry in the usbh subsys sample manifest Signed-off-by: Roman Leonov <jam_roma@yahoo.com>
roma-jam
force-pushed
the
pr-add-subsys-usb-shell-test
branch
2 times, most recently
from
7271103 to
40afb3a
Compare
Added: - Workflow with zephyr container for one subsys usbh sample Signed-off-by: Roman Leonov <jam_roma@yahoo.com>
roma-jam
force-pushed
the
pr-add-subsys-usb-shell-test
branch
from
40afb3a to
7f16b15
Compare
roma-jam
changed the title
roma-jam
changed the title
roma-jam
force-pushed
the
pr-add-dwc2-uhc-channels-init
branch
from
aae36b7 to
0f267aa
Compare
There was a problem hiding this comment.
FYI, for convenience, this file got submitted here:
There was a problem hiding this comment.
Cool, thanks!
I also want to check the ztest to know what is it better.
Seems it might be worth switch to them to test uhc on hardware instead of running the shell example.
There was a problem hiding this comment.
Twister works the same way for the shell sample and USB test, so it might not need a lot of effort or learning.
I am testing the pytest and for now it did not get the prompt, but very nice setup nonetheless, first time I use pytest with twister.
I am now trying a HIL test that calls the usb_init/enable() and will report here.
Hopefully this can be proposed upstream then, so that we can have it on any PR.
Thanks for the infrastructure effort!
There was a problem hiding this comment.
However interactively it all works without glitch so far.
There was a problem hiding this comment.
| shell: bash | ||
| run: | | ||
| cd "$GITHUB_WORKSPACE" | ||
| tar -czf twister-out.tgz twister-out |
There was a problem hiding this comment.
maybe you'd like to try twisters option to prepare packed artifacts? It should store only what is needed for execution --package-artifacts https://github.com/zephyrproject-rtos/zephyr/blob/main/scripts/pylib/twister/twisterlib/environment.py#L377C10-L377C29. If you need some extra files you can add --keep-artifacts https://github.com/zephyrproject-rtos/zephyr/blob/main/scripts/pylib/twister/twisterlib/environment.py#L654C10-L654C26
There was a problem hiding this comment.
Cool, thanks a lot for a note!
Will definitely try it.
| -p esp32s3_devkitc/esp32s3/procpu \ | ||
| -s sample.usbh.shell.hil \ | ||
| --build-only \ | ||
| -j 1 -v \ |
There was a problem hiding this comment.
Interesting question actually....I thought that if I build only one app and I don't need to make it faster, I decided to do it via single worker only.
But there was no error or anything, so it might be removed.
roma-jam
force-pushed
the
pr-add-dwc2-uhc-channels-init
branch
2 times, most recently
from
b21a310 to
6c8ebae
Compare
PerMac
left a comment
PerMac
left a comment
There was a problem hiding this comment.
I just realized I got this comment pending for 2 weeks
| . "$GITHUB_WORKSPACE/ws/.venv/bin/activate" | ||
| cd "$GITHUB_WORKSPACE/ws/zephyr" | ||
|
|
||
| west twister \ |
There was a problem hiding this comment.
I believe you not using the artifacts, but rebuilding the image. To skip building you'd need also --test-only https://github.com/zephyrproject-rtos/zephyr/blob/main/scripts/pylib/twister/twisterlib/environment.py#L382 I suggest you add --log-level debug which will print all the cmake command etc to verify if it was rebuild or not
3 participants
Description
Automation of the verification process during
uhc_dwc2driver development.samples/subsys/usb/shellprovides the possibility to send driver commands, such as:init,enable,device listetc.To verify the current working state of the driver on a hardware usually the test requires manual launch and typing the commands.
To automate the process, the approach of building the twister artifacts on the ubuntu-latest and running them on the self-hosted runner is used.
Changes
This PR adds: