Closed emersonknapp closed 3 years ago
Merging #290 (34c03b1) into master (b929773) will not change coverage. The diff coverage is
100.00%
.
@@ Coverage Diff @@
## master #290 +/- ##
=======================================
Coverage 93.06% 93.06%
=======================================
Files 10 10
Lines 404 404
=======================================
Hits 376 376
Misses 28 28
Flag | Coverage Δ | |
---|---|---|
unittests | 93.06% <100.00%> (ø) |
Flags with carried forward coverage won't be shown. Click here to find out more.
Impacted Files | Coverage Δ | |
---|---|---|
ros_cross_compile/platform.py | 100.00% <100.00%> (ø) |
Continue to review full report at Codecov.
Legend - Click here to learn more
Δ = absolute <relative> (impact)
,ø = not affected
,? = missing data
Powered by Codecov. Last update b929773...34c03b1. Read the comment docs.
run_e2e_test.sh for Foxy doesn't do anything on ARMHF. Should this be duplicated for Noetic?
Noetic does support armhf https://ros.org/reps/rep-0003.html#noetic-ninjemys-may-2020-may-2025 and release binaries for that platform Foxy does not https://ros.org/reps/rep-2000.html#foxy-fitzroy-may-2020-may-2023 (tier 3 basically means no support)
The build problem here seems to be a bug in CMake < 3.17, when used in a 32 bit QEMU environment https://gitlab.kitware.com/cmake/cmake/-/issues/20568 https://gitlab.kitware.com/cmake/cmake/-/merge_requests/5258 fixes it, but we don't get that newer release in Focal
build.ros.org uses https://hub.docker.com/r/osrf/ubuntu_armhf as a base image, which seems to not have this problem because they've baked in some custom QEMU version, so I am investigating using those base images to not reinvent the wheel.
Note https://build.ros.org/view/Nbin_ufhf_uFhf/job/Nbin_ufhf_uFhf__actionlib__ubuntu_focal_armhf__binary/18/consoleFull#console-section-3 for an example build
This was implemented by #319
Closes #100