Hi Andrej,
On 3/26/20 10:54 AM, Andrej Butok via TF-A wrote:
But I am worried that a self-review is rarely as good as a peer review
On practice, unfortunately, some TF-M tasks are waiting weeks and even months for review and following approvals. If I were a maintainer & owner of my own TFM area, I do not want to wait & push & remind somebody else. Better to have a post-merge review for these cases, which does not limit and slow down the development.
Thanks for the feedback. That's not good, patches can't realistically stay in review for weeks and even months, that's just not workable. Worse, it might discourage developers to contribute to the project.
I can see that cumulating maintainer & owner roles would solve the problem here but perhaps enlarging the pool of maintainers would as well? Presumably, the situation is like that today because the current maintainers of the project are overloaded and cannot get all reviews done in a timely manner?
I am skeptical about a post-merge review process... Once a patch is merged there is less urge and motivation (if any) for people to take a look at it. I am worried that patches might never get reviewed that way.
Regards, Sandrine