Hi, Sergey,

thanks for the patch! See my comments.

Sergey

On 10/24/25 13:50, Sergey Kaplun wrote:
> This patch adjusts the aforementioned test to use the benchmark
> framework introduced before. The default arguments are adjusted
> according to the <PARAM_x86.txt> file. The arguments to the script still
> can be provided in the command line run.
> ---
>   perf/LuaJIT-benches/recursive-fib.lua | 28 +++++++++++++++++++++++++--
>   1 file changed, 26 insertions(+), 2 deletions(-)
>
> diff --git a/perf/LuaJIT-benches/recursive-fib.lua b/perf/LuaJIT-benches/recursive-fib.lua
> index ef9950de..99af3f9e 100644
> --- a/perf/LuaJIT-benches/recursive-fib.lua
> +++ b/perf/LuaJIT-benches/recursive-fib.lua
> @@ -1,7 +1,31 @@
> +local bench = require("bench").new(arg)
> +
>   local function fib(n)
>     if n < 2 then return 1 end
>     return fib(n-2) + fib(n-1)
>   end
>   
> -local n = tonumber(arg[1]) or 10
> -io.write(string.format("Fib(%d): %d\n", n, fib(n)))
debug print was lost, is it intentional?
> +local n = tonumber(arg[1]) or 40
Why 40?
> +
> +local benchmark
> +benchmark = {
> +  name = "recursive_fib",
> +  checker = function(res)
> +    local km1, k = 1, 1
> +    for i = 2, n do
> +      local tmp = k + km1
> +      km1 = k
> +      k = tmp
> +    end
> +    return k == res
> +  end,
> +  payload = function()
> +    local res = fib(n)
> +    -- Number of calls.
> +    benchmark.items = res * 2 - 1
> +    return res
> +  end,
> +}
> +
> +bench:add(benchmark)
> +bench:run_and_report()