Conversation
|
@Igorshp cool! |
dc9027a to
0674e83
Compare
|
Pretty much all compilation errors resolved. having trouble with Some tests are also failing, expected as the code has diverged since. |
|
@Igorshp the file is at |
hmm, intresting point. Thanks. I'll check it can be toggled with flags |
* buildtest branches too * minor artifact name fix * workflow bash([[ == ]]), not posix([ = ]), nor broken mix.
```report_cell_voltage``` is not used currently for crossfire. This request fixes it.
* i don't think its broken, but lets just say we fixed it. * new horizon mode stuff :) * update defaults -- i think this is the defaults that PJC wants * NFE mode as a switch
…t#417) * CLI: allow vtx power to be changed by a switch. backported from Betaflight: more info on usage: https://github.com/betaflight/betaflight/wiki/VTX-CLI-Settings * Update cli.c * add unit tests for 'vtx' cli command * unittest: test that activation conditions do take effect Co-authored-by: Igor Shpakov <igor@sliide.com>
Status updateAll but one tests compile (sensor_gyro_unittest requires arm_math.h library and I haven't yet figured out how to provide it) 3 tests compile but are failing since code base have moved on. The github action step fails to run the tests. Looks like environment setup issue. |
|
@Igorshp thanks!
it's a compiler issue - the action will be fine. the current state of the affair is like this:
we should just wait a couple of days for the upstreams to settle. 🍵 |
|
ok got the test to run to completion. credits to @mikeller 😗 @Igorshp now, i can either PR this up as new or you rebase my branch on your open PR.. i'm fine both ways. @Quick-Flash the tests show some interesting results including the |
|
@gretel interesting, thanks for the insight! I've had a fair share of issues with the compiler myself and considering building a docker container to run the tests in. It'll help normalize the testing environment for everyone and from what I understand github actions can use containers. Then there's matter of broken tests. crsf ones dont look to bad. pid controller stuff though is a bit out of my league at the moment. |
no more docker please 🙈 i think it's working fine now.
there is also the matter of broken implemenations 👯 |
…light#10265; backport of betaflight/betaflight#10288; correction of issues showing up using the betaflights improvements; fix unusued variable in 'crsf.c' to get the test to run to completion
|
Merged in your changes, let's wait for the build EDIT: hmm, couple of other commits got pulled in. I'll rebase on master when we're done |
|
i think the |
|
raw logs: |
|
Tried getting this to work with clang-11 and/or clang-12 for modern OS. if this can be fixed, then master merged in, it will need an --- a/src/main/sensors/gyro.h
+++ b/src/main/sensors/gyro.h
@@ -195,7 +195,9 @@ bool gyroOverflowDetected(void);
bool gyroYawSpinDetected(void);
uint16_t gyroAbsRateDps(int axis);
uint8_t gyroReadRegister(uint8_t whichSensor, uint8_t reg);
+#ifdef USE_SMITH_PREDICTOR
float applySmithPredictor(smithPredictor_t *smithPredictor, float gyroFiltered);
+#endif |
Trying to get the testsuite to run.
Will be updating this PR over time.
Help is much appreciated.
To run the tests:
you may need to have
clang,llvmandlibblocksruntime-devinstalledStatus
arm_math.hlibrary)❌ Tests running in github action step (currently failing to compile)