remove :boundscheck argument from Core._svec_ref (#50561)

`Core._svec_ref` has accepted `boundscheck`-value as the first argument
since it was added in #45062. Nonetheless, `Core._svec_ref` simply calls
`jl_svec_ref` in either the interpreter or the codegen, and thus the
`boundscheck` value isn't utilized in any optimizations. Rather, even
worse, this `boundscheck`-argument negatively influences the effect
analysis (xref #50167 for details) and has caused type inference
regressions as reported in #50544.

For these reasons, this commit simply eliminates the `boundscheck`
argument from `Core._svec_ref`. Consequently,
`getindex(::SimpleVector, ::Int)` is now being concrete-eval eligible.

closes #50544

base/compiler/ssair/inlining.jl

@@ -1808,7 +1808,7 @@ end
 
 function insert_spval!(insert_node!::Inserter, spvals_ssa::SSAValue, spidx::Int, do_isdefined::Bool)
     ret = insert_node!(
-        effect_free_and_nothrow(NewInstruction(Expr(:call, Core._svec_ref, false, spvals_ssa, spidx), Any)))
+        effect_free_and_nothrow(NewInstruction(Expr(:call, Core._svec_ref, spvals_ssa, spidx), Any)))
     tcheck_not = nothing
     if do_isdefined
         tcheck = insert_node!(

base/compiler/ssair/passes.jl

@@ -811,10 +811,10 @@ function perform_lifting!(compact::IncrementalCompact,
 end
 
 function lift_svec_ref!(compact::IncrementalCompact, idx::Int, stmt::Expr)
-    length(stmt.args) != 4 && return
+    length(stmt.args) != 3 && return
 
-    vec = stmt.args[3]
-    val = stmt.args[4]
+    vec = stmt.args[2]
+    val = stmt.args[3]
     valT = argextype(val, compact)
     (isa(valT, Const) && isa(valT.val, Int)) || return
     valI = valT.val::Int

base/essentials.jl

@@ -761,7 +761,7 @@ end
 
 # SimpleVector
 
-@eval getindex(v::SimpleVector, i::Int) = (@_foldable_meta; Core._svec_ref($(Expr(:boundscheck)), v, i))
+getindex(v::SimpleVector, i::Int) = (@_foldable_meta; Core._svec_ref(v, i))
 function length(v::SimpleVector)
     @_total_meta
     t = @_gc_preserve_begin v

src/builtins.c

@@ -1655,11 +1655,9 @@ JL_CALLABLE(jl_f__compute_sparams)
 
 JL_CALLABLE(jl_f__svec_ref)
 {
-    JL_NARGS(_svec_ref, 3, 3);
-    jl_value_t *b = args[0];
-    jl_svec_t *s = (jl_svec_t*)args[1];
-    jl_value_t *i = (jl_value_t*)args[2];
-    JL_TYPECHK(_svec_ref, bool, b);
+    JL_NARGS(_svec_ref, 2, 2);
+    jl_svec_t *s = (jl_svec_t*)args[0];
+    jl_value_t *i = (jl_value_t*)args[1];
     JL_TYPECHK(_svec_ref, simplevector, (jl_value_t*)s);
     JL_TYPECHK(_svec_ref, long, i);
     size_t len = jl_svec_len(s);

test/compiler/inference.jl

@@ -5014,3 +5014,19 @@ let 𝕃 = Core.Compiler.SimpleInferenceLattice.instance
         map(T -> Issue49785{S,T}, (a = S,))
     end isa Vector
 end
+
+# `getindex(::SimpleVector, ::Int)` shuold be concrete-evaluated
+@eval Base.return_types() do
+    $(Core.svec(1,Int,nothing))[2]
+end |> only == Type{Int}
+# https://github.com/JuliaLang/julia/issues/50544
+struct Issue50544{T<:Tuple}
+    t::T
+end
+Base.@propagate_inbounds f_issue50544(x, i, ii...) = f_issue50544(f_issue50544(x, i), ii...)
+Base.@propagate_inbounds f_issue50544(::Type{Issue50544{T}}, i) where T = T.parameters[i]
+g_issue50544(T...) = Issue50544{Tuple{T...}}
+h_issue50544(x::T) where T = g_issue50544(f_issue50544(T, 1), f_issue50544(T, 2, 1))
+let x = Issue50544((1, Issue50544((2.0, 'x'))))
+    @test only(Base.return_types(h_issue50544, (typeof(x),))) == Type{Issue50544{Tuple{Int,Float64}}}
+end

test/core.jl

@@ -8040,3 +8040,7 @@ bar50293(@nospecialize(u)) = (Base.issingletontype(u.a), baz50293(u.a))
 let u = Union{Type{Union{}}, Type{Any}}, ab = bar50293(u)
     @test ab[1] == ab[2] == false
 end
+
+# `SimpleVector`-operations should be concrete-eval eligible
+@test Core.Compiler.is_foldable(Base.infer_effects(length, (Core.SimpleVector,)))
+@test Core.Compiler.is_foldable(Base.infer_effects(getindex, (Core.SimpleVector,Int)))