Startup-speed regression

[PallHaraldsson]

$ hyperfine "julia +1.11"
Benchmark 1: julia +1.11
  Time (mean ± σ):     566.4 ms ±  33.2 ms    [User: 1378.7 ms, System: 83.1 ms]
  Range (min … max):   508.1 ms … 612.6 ms    10 runs

It's very strange since:
 
$ hyperfine "julia +1.11.0-rc4 -e ''"
Benchmark 1: julia +1.11.0-rc4 -e ''
  Time (mean ± σ):     175.6 ms ±  16.7 ms    [User: 210.7 ms, System: 66.0 ms]
  Range (min … max):   155.9 ms … 206.9 ms    14 runs

$ hyperfine "julia +1.10 -e ''"
Benchmark 1: julia +1.10 -e ''
  Time (mean ± σ):     214.0 ms ±  14.7 ms    [User: 174.0 ms, System: 121.6 ms]
  Range (min … max):   191.7 ms … 236.7 ms    12 runs

Remembering the speedup in RCs I've already asked for the Debian Benchmark Game to rerun on 1.11.0, but now regret it...

Is 1.11.1 going to be released soon, and might it already have the fix?

This might be a simple thing to debug, or bisect (I'm not good with that), or anyone has a guess why?

[KristofferC]

#56063

[nsajko]

Duplicate of #56063

(trying to get Github to recognize this)