Re: [PATCH 0/6] Speed up mirror-fetches with many refs

On 8/20/2021 6:08 AM, Patrick Steinhardt wrote:
...

As it turns out, many of the issues are again caused by loading and
dereferencing refs. This patch series thus mostly focusses on optimizing
the patterns there, where the biggest win is to opportunistically load
refs via commit-graphs.

You caught my attention at "commit-graph" and I found your use of them
to be interesting. You strike a balance in checking the commit-graph
when it is likely to be helpful, and skip the commit-graph when it is
not. (For example, PATCH 2 is unlikely to benefit from checking the
commit-graph at that point, because we are looking for objects that
were just downloaded.)

I read all the patches and checked the full context of the functions
to see if there were any issues, but found none. My only comments are
about the case of many annotated tags (do we slow down?) and some
nitpicks.

Thanks,
-Stolee