`Array.reverse_each` is faster than `Array.reverse.each` because:
* reverse.each creates a new array then loops each element
* reverse_each loops in reverse order (no intermediate array created)
This adds the backtrace to a table to show exactly where the Gitaly call
was made to make it easier to understand where the call originated.
This change also collapses the details in the same row to improve the
usability when there is a backtrace.
Previously, we used a personal access token. This had a couple of
problems:
1. If the user didn't have a PAT, we couldn't impersonate them.
2. It depended on reading the raw PAT from the database.
Instead, we can monkey-patch the authentication methods on
ApplicationController (overriding the Devise ones), and remove them once
we're done. This does mean that profiles will not profile auth
correctly, so for that, use a PAT directly.
This method uses Gitlab::Profiler::TotalTimeFlatPrinter internally, which is the
same as RubyProf::FlatPrinter, but the min_percent option (and new max_percent
option) applies to the _total_ time, not the _self_ time. This helps us figure
out if we're calling a library inefficiently, for instance.
Gitlab::Profiler.profile takes a URL to profile, and returns the ruby-prof
profile result.
This is currently only for interactive use (in a console) or from the script
runner. Note that when using this interactively, changes to the application
within that console session will be used, so this can be used to test proposed
performance changes.
