transpiler: generates valid PTX and evaluates expressions correctly

package/test/PerformanceTests.jl

@@ -1,6 +1,8 @@
 using .Transpiler
 using .Interpreter
 
+# University setup at 10.20.1.7 if needed
+
 @testset "CPU performance" begin
 	function test_cpu_interpreter(nrows; parallel = false)
 		exprs = [
@@ -39,9 +41,23 @@ LinearAlgebra.BLAS.set_num_threads(1) # only use a single thread for peakflops
 end
 
 @testset "Interpreter Performance" begin
-	
+	# Put data in shared memory: 
+	# https://cuda.juliagpu.org/v2.6/api/kernel/#Shared-memory
+
+	# Make array const:
+	# https://cuda.juliagpu.org/v2.6/api/kernel/#Device-arrays
+
+	# Memory management like in C++ might help with performance improvements
+	# https://cuda.juliagpu.org/v2.6/lib/driver/#Memory-Management
 end
 
 @testset "Transpiler Performance" begin
-	
+	# Put data in shared memory: 
+	# https://cuda.juliagpu.org/v2.6/api/kernel/#Shared-memory
+
+	# Make array const:
+	# https://cuda.juliagpu.org/v2.6/api/kernel/#Device-arrays
+
+	# Memory management like in C++ might help with performance improvements
+	# https://cuda.juliagpu.org/v2.6/lib/driver/#Memory-Management
 end