On 08/12/2025 23:16, Tomas Volf wrote:
Out of curiosity, how does git-bisect work with merges? Is it still usable to pin-point the offending commits? It is long time since I tried to use it on non-linear history, so I am unsure how well it works these days.
The bisect algorithm works fine with merges. It even has some advantages; for instance you can restrict the merge history to the first parent only, to quickly identify which topic branch caused the problem or skip intermediate unbuildable commits.
There is some suggestion that bisect did not handle merges correctly on the early versions of git.
Topic branch merges allow a high level overview of the history of the project, something which is difficult to determine from a linear history.
Chris
