From 1453a50c947e0855a6b0eb4dc2997f03938e9e25 Mon Sep 17 00:00:00 2001
From: Valentin Churavy <v.churavy@gmail.com>
Date: Tue, 25 Feb 2025 13:34:16 +0100
Subject: [PATCH] handle inbounds and insert `julia.simdloop` to every loop

---
 src/Trixi.jl                | 14 +-------------
 src/auxiliary/mock_turbo.jl | 28 ++++++++++++++++++++++++++++
 2 files changed, 29 insertions(+), 13 deletions(-)
 create mode 100644 src/auxiliary/mock_turbo.jl

diff --git a/src/Trixi.jl b/src/Trixi.jl
index 725ede2eb6..7178243c1a 100644
--- a/src/Trixi.jl
+++ b/src/Trixi.jl
@@ -63,19 +63,7 @@ if _PREFERENCE_LOOPVECTORIZATION
     using LoopVectorization: LoopVectorization, @turbo, indices
 else
     using LoopVectorization: LoopVectorization, indices
-    macro turbo(exprs...)
-        body = nothing
-        for expr in exprs
-            if expr.head == :for
-                body = expr
-            end
-        end
-        @assert body !== nothing
-        # TODO: We should insert !loopinfo !julia.ivdep !julia.simd
-        #       but SimdLoop.compile doesn't deal with nested for loops.
-        # esc(Base.SimdLoop.compile(body, Symbol("julia.ivdep")))
-        return esc(body)
-    end
+    include("auxiliary/mock_turbo.jl")
 end
 
 using StaticArrayInterface: static_length # used by LoopVectorization
diff --git a/src/auxiliary/mock_turbo.jl b/src/auxiliary/mock_turbo.jl
new file mode 100644
index 0000000000..dcf1a892ed
--- /dev/null
+++ b/src/auxiliary/mock_turbo.jl
@@ -0,0 +1,28 @@
+macro turbo(exprs...)
+    body = nothing
+    for expr in exprs
+        if Meta.isexpr(expr, :for)
+            body = expr
+        end
+    end
+    @assert body !== nothing
+
+    function insert_loopinfo!(expr)
+        recurse = Meta.isexpr(expr, :for) || Meta.isexpr(expr, :block) ||
+                  Meta.isexpr(expr, :let)
+        if recurse
+            foreach(insert_loopinfo!, expr.args)
+        end
+        if Meta.isexpr(expr, :for)
+            # TODO: Should we insert LLVM loopinfo or `julia.ivdep`?
+            push!(expr.args, Expr(:loopinfo, Symbol("julia.simdloop")))
+        end
+    end
+    insert_loopinfo!(body)
+
+    body = Expr(:block,
+                Expr(:inbounds, true),
+                body,
+                Expr(:inbounds, :pop))
+    return esc(body)
+end