Debugger erroneously enters `BigFloat` constructor

[krsteffen]

Minimum working example (MWE)

function placeholder_function()
    sin(1.0)
end

function function_we_want_to_inspect()
    return ([1.0 -1.0; 1.0 1.0]/sqrt(2.0))*[1.0, 0.0]
end

function f()
    x = placeholder_function()
    y = BigFloat(1)
    function_we_want_to_inspect()
end

@enter f()

Instructions to replicate

• Save MWE to a file, say, mwe.jl
• Open the Julia REPL
• Invoke include("mwe.jl") in the REPL
• In the Debugger prompt, invoke bp add function_we_want_to_inspect
• Next (in Debugger prompt), invoke c

Expected result

In function_we_want_to_inspect() at /home/steffen/mwe.jl:7
 7  function function_we_want_to_inspect()
>8      return ([1.0 -1.0; 1.0 1.0]/sqrt(2.0))*[1.0, 0.0]
 9  end

About to run: (tuple)(2, 2)

Actual result

In BigFloat(, ) at mpfr.jl:122
>122  function BigFloat(; precision::Integer=DEFAULT_PRECISION[])
 123      precision < 1 && throw(DomainError(precision, "`precision` cannot be less than 1."))
 124      nb = ccall((:mpfr_custom_get_size,libmpfr), Csize_t, (Clong,), precision)
 125      nb = (nb + Core.sizeof(Limb) - 1) ÷ Core.sizeof(Limb) # align to number of Limb allocations required for this
 126      #d = Vector{Limb}(undef, nb)

About to run: Core.NewvarNode(:(_4))

Version info

• Julia 1.10.2
• [31a5f54b] Debugger v0.7.8
• I verified the (erroneous) result in a completely new Julia environment with no packages installed except ]add Debugger
• I also downgraded to [31a5f54b] Debugger v0.7.0, restarted the Julia REPL, and reproduced the same erroneous result. Downgrading further was not possible due to unsatisfiable requirements detected