Has ctfe-stress-5-check been gradually regressing since 2022-04-09?

[pnkfelix]

Spawned off of #96015 (comment)

Look at this graph:

is the trend there something we should be worried about? (Or is this just a microbenchmark that we should not put too much stock in relative to everything else that is going on in the compiler...)

[RalfJung]

Cc @rust-lang/wg-const-eval

[oli-obk]

cc @lqd who is looking into some improvements here.

[Enselic]

Or is this just a microbenchmark that we should not put too much stock in relative to everything else that is going on in the compiler

Triage: From the age of this issue I think we can conclude that this is not something to put too much effort into relative to other things.

Closing since I think this will remain open forever otherwise, but feel free to reopen if you disagree.