Conversation
|
Diff looks good to me, can merge once the tests are fixed. I've bumped the opendbc ref too. |
aubsw
left a comment
aubsw
left a comment
There was a problem hiding this comment.
@robbederks not sure why, but these changes have exposed some new cppcheck failures.
Mostly just conditionally unused functions (i.e. some functions only get used when ENABLE_SPI is true). Pls advise (see comments).
tests/misra/test_misra.sh
Outdated
| PANDA_SOURCES="$PANDA_DIR/board/ --enable=all --file-filter=*.c \ | ||
| -i $PANDA_DIR/board/safety.c \ | ||
| -i $PANDA_DIR/board/jungle/main.c \ | ||
| -i $PANDA_DIR/board/bootstub.c" |
There was a problem hiding this comment.
@robbederks should we ignore all non-main.c files?
If we don't ignore, we'll have a lot more cppchecks failures.
The old behavior was to only check board/main.c.
There was a problem hiding this comment.
Ideally we check everything I think, but we should especially check safety.c and main.c if that covers all of the panda app code
There was a problem hiding this comment.
Kk. Looking into it. It may require some more cahnges to opendbc.
There was a problem hiding this comment.
I cleared up board/main.c and safety.c.
Currently both bootstub.c and jungle/main.c are in a MISRA-incompliant state. I think we should address those later for the sake of progress.
aubsw
changed the title
aubsw
force-pushed
the
delete-dup-panda-code-1
branch
from
b1d9ec6 to
6610a3d
Compare
aubsw
force-pushed
the
delete-dup-panda-code-1
branch
from
b6b1b8d to
718f5db
Compare
aubsw
force-pushed
the
delete-dup-panda-code-1
branch
from
8892d9e to
658e5c6
Compare
|
Rebased, cleaned up a few odds and ends. |
|
Anything for me to do regarding this failed hitl run? I don't think I have access to https://jenkins.comma.life/job/panda/job/tmp-jenkins-2186/2/display |
|
Jungle v2 doesn't seem to boot with this PR (gets stuck resetting after flashing). Do you have any hardware to test this out on? |
|
oh no! 🫠 I only have comma 3x. I would need https://comma.ai/shop/panda-jungle to test out the jungle boot, yeah? Does the non-jungle panda boot without issues? |
|
I found a lead for what the issue might be... Not sure if I can trigger jenkins tho |
|
Sorry for the n00b question, but do you think I could debug this with a blue pill stm board? |
|
Sorry, I'm OOO this week so reply time might be a bit slower |
|
trigger-jenkins |
|
@robbederks is the HITL test still failing? |
|
Adeeb was unhappy with this PR in the dev call today, which is understandable as it's not the ideal solution. I think that the ideal solution would probably be to make opendbc use source/header style so that we can easily link the safety code into panda. |
aubsw
force-pushed
the
delete-dup-panda-code-1
branch
from
e1552af to
4b121f5
Compare
aubsw
force-pushed
the
delete-dup-panda-code-1
branch
from
4b121f5 to
fa019c5
Compare
|
OK I took another hard look. Found a waaay simpler way to do this step. We may still need to make another PR to deal with linking safety code properly for the final PR, but if we can land this one first, that PR will be way less risky. |
aubsw
force-pushed
the
delete-dup-panda-code-1
branch
from
95ce85a to
6e937bb
Compare
| # TODO: this should be a "pip install" or not even in this repo at all | ||
| RUN git config --global --add safe.directory $PYTHONPATH/panda | ||
| ENV OPENDBC_REF="da0a5e3d2b3984b56ebf5e25d9769f5c77807e4d" | ||
| ENV OPENDBC_REF="d631299b8d49253fdeb5b709684af031ab1d18be" |
There was a problem hiding this comment.
Needs to be updated after commaai/opendbc#2248 lands
aubsw
changed the title
2 participants
details
utils.handcan.hare approximately duplicated from opendbc, which is making [$500 Bounty] Refactor panda to use source + header file structure #2171 hard to land because it results in duplicated symbols