Yes, for hardware related problems a hardware verification seems
mandatory. Also we should define a minimal required test sequence
(i.e. run nsh, uname -a, ostest). Maybe this can be done with some
dedicated "selftest" configuration added to all boards?
This PR broke something:
https://github.com
I like this plan of requiring hardware testing. Also, even if the PR may
contain enough information to be justified, it becomes really difficult for
anyone debugging to track down changes if the PR body doesn't have a
description of any testing or the change impact. If all the PRs follow the
enforc
Maybe we shouldn't merge PRs until at least one reviewer tests on real
hardware. It doesn't have to be the same board, but at least the same arch,
if it touches arch code. If it's in sched or something shared on all
architectures then it's okay to test on any arch. We would need to decide
how to ha
On 2025-02-03 22:45:29, Xiang Xiao wrote:
> The ai bot is a soft suggestion, not a rule from Lup introduces it
> initially.
> Many PR already contain the useful information, like these:
> https://github.com/apache/nuttx/pull/15749
> https://github.com/apache/nuttx/pull/15728
> But ai bot still comp
The ai bot is a soft suggestion, not a rule from Lup introduces it
initially.
Many PR already contain the useful information, like these:
https://github.com/apache/nuttx/pull/15749
https://github.com/apache/nuttx/pull/15728
But ai bot still complain the information isn't enough.
Even https://github
Hello
The other thread made the remark that this pull request was merged
https://github.com/apache/nuttx/pull/15437
However it appears that it was undocumented.
this problem was noticed by the ai bot
shortly after this it was approved by two people and merged
this is a breach of your own rul
