adjustments based on superluminal

Still not sure how hybrid I'm going to go (away from his more purist stance on typed info embedding).

If I were to codegen this and wanted typeless vs typed for debug it would be easier to malleablly switch between...
If I decide for this sample to go full asm (no C features almost we can still keep the type info for debug...

.gitignore

@@ -5,3 +5,4 @@ C/watl.v0.msvc.c.listing.txt
 watl_exercise.user
 watl_exercise.proj
 Odin/watl.v0.win32.odin.listing.txt
+C/watl.v0.lottes.c.listing.txt

C/watl.v0.msvc.c

@@ -1367,19 +1367,19 @@ void arena_allocator_proc(AllocatorProc_In in, AllocatorProc_Out* out)
 #pragma endregion Arena
 
 #pragma region Key Table 1-Layer Linear (KT1L)
-void kt1l__populate_slice_a2(KT1L_Byte* kt, AllocatorInfo backing, KT1L_Meta m, Slice_Byte values, SSIZE num_values ) {
+void kt1l__populate_slice_a2(KT1L_Byte*restrict kt, AllocatorInfo backing, KT1L_Meta m, Slice_Byte values, SSIZE num_values ) {
 	assert(kt != nullptr);
 	if (num_values == 0) { return; }
 	* kt = alloc_slice(backing, Byte, m.slot_size * num_values );
 	slice_assert(* kt);
 	for (span_iter(SSIZE, iter, 0, <, num_values)) {
 		SSIZE         slot_offset = iter.cursor * m.slot_size;                         // slot id
-		Byte*      slot_cursor = & kt->ptr[slot_offset];                            // slots[id]            type: KT1L_<Type>
-		U64*       slot_key    = (U64*)slot_cursor;                                 // slots[id].key        type: U64
+		Byte*restrict slot_cursor = & kt->ptr[slot_offset];                            // slots[id]            type: KT1L_<Type>
+		U64*restrict  slot_key    = (U64*restrict)slot_cursor;                         // slots[id].key        type: U64
 		Slice_Byte    slot_value  = { slot_cursor + m.kt_value_offset, m.type_width }; // slots[id].value      type: <Type>
 		SSIZE         a2_offset   = iter.cursor * m.type_width * 2;                    // a2 entry id
-		Byte*      a2_cursor   = & values.ptr[a2_offset];                           // a2_entries[id]       type: A2_<Type>
-		Slice_Byte a2_key      = * cast(Slice_Byte*, a2_cursor);                    // a2_entries[id].key   type: <Type>
+		Byte*restrict a2_cursor   = & values.ptr[a2_offset];                           // a2_entries[id]       type: A2_<Type>
+		Slice_Byte    a2_key      = * cast(Slice_Byte*restrict, a2_cursor);            // a2_entries[id].key   type: <Type>
 		Slice_Byte    a2_value    = { a2_cursor + m.type_width, m.type_width };        // a2_entries[id].value type: <Type>
 		slice_copy(slot_value, a2_value);                                              // slots[id].value = a2_entries[id].value
 		* slot_key = 0; hash64_djb8(slot_key,  a2_key);                                // slots[id].key   = hash64_djb8(a2_entries[id].key)

scripts/build.c.ps1

@@ -115,10 +115,10 @@ $compiler_args += $flag_full_src_path
 # $compiler_args += $flag_asm_listing_file
 
 # $compiler_args += $flag_optimize_speed_max
-$compiler_args += $flag_optimize_fast
+# $compiler_args += $flag_optimize_fast
 # $compiler_args += $flag_optimize_size
 # $compiler_args += $flag_optimize_intrinsics
-# $compiler_args += $flag_no_optimization
+$compiler_args += $flag_no_optimization
 
 # Debug setup
 $compiler_args += ($flag_define + 'BUILD_DEBUG')

scripts/build.c_lottes.ps1

@@ -93,7 +93,7 @@ $compiler_args += $flag_c11
 # Constraints on C program code-gen 
 $compiler_args += $flag_exceptions_disabled
 $compiler_args += $flag_RTTI_disabled
-$compiler_args += $flag_preprocess_conform
+# $compiler_args += $flag_preprocess_conform
 # $compiler_args += $flag_sanitize_address
 
 $compiler_args += $flag_wall
@@ -103,7 +103,7 @@ $compiler_args += $flag_charset_utf8
 
 # Specifing output pathing
 $compiler_args += ( $flag_path_interm + $path_build + '\' )
-$compiler_args += ( $flag_path_output + $path_build + '\' )
+# $compiler_args += ( $flag_path_output + $path_build + '\' )
 
 # Dump preprocess file
 if ($false) {
@@ -141,7 +141,10 @@ $compiler_args += $flag_compile, $unit
 $compiler_args | ForEach-Object { Write-Host $_ }
 
 # Compile the unit
+$compilation_time = Measure-Command { 
 	& $compiler $compiler_args 
+}
+write-host "Compilation took $($compilation_time.TotalMilliseconds)ms"
 write-host
 
 $binary = join-path $path_build "$unit_name.exe"
@@ -174,8 +177,9 @@ if ($true) {
 	# Diagnoistc print for the args
 	$linker_args | ForEach-Object { Write-Host $_ }
 
-	& $linker $linker_args
+	$linking_time = Measure-Command { & $linker $linker_args }
 	# & $radlink $linker_args
+	write-host "Linking took $($linking_time.TotalMilliseconds)ms"
 	write-host
 }