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.cpp

@@ -0,0 +1,2 @@
+#pragma once
+

C/watl.v0.msvc.c

@@ -383,7 +383,7 @@ cast(type*, varena__push(arena, 1, size_of(type), opt_args(Opts_varena, lit(stri
 
 #define varena_push_array(arena, type, amount, ...) \
 (tmpl(Slice,type)){ varena__push(arena, size_of(type), amount, opt_args(Opts_varena, lit(stringify(type)), __VA_ARGS__)).ptr, amount }
-#pragma endregion VArnea
+#pragma endregion VArena
 
 #pragma region Arena (Casey-Ryan Composite Arenas)
 typedef Opts_varena Opts_arena;
@@ -1367,22 +1367,22 @@ 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
-		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>
-		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)
+		SSIZE         slot_offset = iter.cursor * m.slot_size;                         // slot id
+		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*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)
 	}
 	kt->len = num_values;
 }
@@ -1790,14 +1790,14 @@ void str8gen_append_str8(Str8Gen* gen, Str8 str){
 }
 void str8gen__append_fmt(Str8Gen* gen, Str8 fmt_template, Slice_A2_Str8* entries){
 	local_persist Byte tbl_mem[kilo(32)]; FArena tbl_arena = farena_make(slice_fmem(tbl_mem));
-	KT1L_Str8 kt = {0}; kt1l_populate_slice_a2(Str8, & kt, ainfo_farena(tbl_arena), *entries );
+	KT1L_Str8  kt     = {0}; kt1l_populate_slice_a2(Str8, & kt, ainfo_farena(tbl_arena), *entries );
 	Slice_Byte buffer = { gen->ptr + gen->len, gen->cap - gen->len };
 	if (buffer.len < kilo(16)) {
 		Slice_Byte result = mem_grow(gen->backing, str8gen_slice_byte(* gen), kilo(16) + gen->cap );
 		slice_assert(result);
 		gen->ptr  = result.ptr;
 		gen->cap += kilo(16);
-		buffer = (Slice_Byte){ cast(Byte*, gen->ptr + gen->len), gen->cap - gen->len };
+		buffer    = (Slice_Byte){ cast(Byte*, gen->ptr + gen->len), gen->cap - gen->len };
 	}
 	Str8 result = str8__fmt_kt1l(gen->backing, & buffer, kt, fmt_template);
 	gen->len += result.len;

scripts/build.c.ps1

@@ -42,8 +42,8 @@ $flag_link_win_machine_64          = '/MACHINE:X64'
 $flag_link_win_path_output         = '/OUT:'
 $flag_link_win_rt_dll              = '/MD'
 $flag_link_win_rt_dll_debug        = '/MDd'
-$flag_link_win_rt_static           = '/MT'
-$flag_link_win_rt_static_debug     = '/MTd'
+$flag_link_win_rt_static           = '/MT:STATIC'
+$flag_link_win_rt_static_debug     = '/MTd:STATIC'
 $flag_link_win_subsystem_console   = '/SUBSYSTEM:CONSOLE'
 $flag_link_win_subsystem_windows   = '/SUBSYSTEM:WINDOWS'
 $flag_no_optimization              = '/Od'
@@ -115,6 +115,7 @@ $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_size
 # $compiler_args += $flag_optimize_intrinsics
 $compiler_args += $flag_no_optimization

scripts/build.c_lottes.ps1

@@ -0,0 +1,200 @@
+$misc = Join-Path $PSScriptRoot 'helpers/misc.psm1'
+import-module $misc
+
+# This script now uses the LLVM toolchain (clang-cl, lld-link).
+# Ensure these tools are available in your PATH.
+# The original call to the MSVC devshell has been removed.
+# & (join-path $PSScriptRoot 'helpers/devshell.ps1') -arch amd64
+
+$path_root = Get-ScriptRepoRoot
+
+$path_root      = split-path -Path $PSScriptRoot -Parent
+$path_toolchain = join-path $path_root      'toolchain'
+$path_rad       = join-path $path_toolchain 'rad'
+
+# https://learn.microsoft.com/en-us/cpp/build/reference/compiler-options-listed-by-category?view=msvc-170
+# Most cl.exe flags are compatible with clang-cl.exe
+$flag_all_c                        = '/TC'
+$flag_c11                          = '/std:c11'
+$flag_c23                          = '/std:c23'
+$flag_all_cpp                      = '/TP'
+$flag_compile                      = '/c'
+$flag_charset_utf8                 = '/utf-8'
+$flag_debug                        = '/Zi'
+$flag_define                       = '/D'
+$flag_exceptions_disabled          = '/EHsc-'
+$flag_RTTI_disabled                = '/GR-'
+$flag_include                      = '/I'
+$flag_full_src_path                = '/FC'
+$flag_asm_listing_file             = '/FAs'
+$flag_nologo                       = '/nologo'
+$flag_dll                          = '/LD'
+$flag_dll_debug                    = '/LDd'
+$flag_linker                       = '/link'
+# $flag_link_lib                     = '/lib'
+$flag_link_dll                     = '/DLL'
+$flag_link_no_incremental          = '/INCREMENTAL:NO'
+$flag_link_mapfile                 = '/MAP:'
+$flag_link_optimize_references     = '/OPT:REF'
+$flag_link_win_debug               = '/DEBUG'
+$flag_link_win_pdb                 = '/PDB:'
+$flag_link_win_machine_32          = '/MACHINE:X86'
+$flag_link_win_machine_64          = '/MACHINE:X64'
+$flag_link_win_path_output         = '/OUT:'
+$flag_link_win_rt_dll              = '/MD'
+$flag_link_win_rt_dll_debug        = '/MDd'
+$flag_link_win_rt_static           = '/MT'
+$flag_link_win_rt_static_debug     = '/MTd'
+$flag_link_win_subsystem_console   = '/SUBSYSTEM:CONSOLE'
+$flag_link_win_subsystem_windows   = '/SUBSYSTEM:WINDOWS'
+$flag_no_optimization              = '/Od'
+$flag_optimize_speed_max           = '/Ox'
+$flag_optimize_fast                = '/O2'
+$flag_optimize_size                = '/O1'
+$flag_optimize_intrinsics          = '/Oi'
+$flag_optimized_debug_forceinline  = '/d2Obforceinline'
+$flag_optimized_debug              = '/Zo'
+$flag_preprocess_to_file           = '/P'
+$flag_preprocess_preserve_comments = '/C'
+# $flag_out_name                     = '/OUT:'
+$flag_path_interm                  = '/Fo'
+$flag_path_debug                   = '/Fd'
+$flag_path_output                  = '/Fe'
+$flag_preprocess_conform           = '/Zc:preprocessor'
+$flag_sanitize_address             = '/fsanitize=address'
+$flag_updated_cpp_macro            = "/Zc:__cplusplus"
+$flag_set_stack_size               = '/F'
+$flag_syntax_only                  = '/Zs'
+$flag_wall                         = '/Wall'
+$flag_warnings_as_errors           = '/WX'
+$flag_lib_list                     = '/LIST'
+ 
+$archiver = 'llvm-lib'
+$compiler = 'clang-cl'
+$linker   = 'lld-link'
+$radbin   = join-path $path_rad 'radbin.exe'
+$radlink  = join-path $path_rad 'radlink.exe'
+
+$path_build = join-path $path_root 'build'
+if ( -not(test-path -Path $path_build) ) {
+	new-item -ItemType Directory -Path $path_build
+}
+
+push-location $path_build
+
+write-host "Compiling with clang-cl"
+
+$compiler_args = @()
+$compiler_args += $flag_nologo
+
+# Constraints on interpeting all files as C code
+$compiler_args += $flag_all_c
+$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 += $flag_sanitize_address
+
+$compiler_args += $flag_wall
+
+# Set charset encoding for both execution and source to UTF-8
+$compiler_args += $flag_charset_utf8
+
+# Specifing output pathing
+$compiler_args += ( $flag_path_interm + $path_build + '\' )
+# $compiler_args += ( $flag_path_output + $path_build + '\' )
+
+# Dump preprocess file
+if ($false) {
+	$compiler_args += $flag_preprocess_to_file
+	$compiler_args += $flag_preprocess_preserve_comments
+}
+
+# Diagnostic logging
+$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_size
+# $compiler_args += $flag_optimize_intrinsics
+$compiler_args += $flag_no_optimization
+
+# Debug setup
+$compiler_args += ($flag_define + 'BUILD_DEBUG')
+$compiler_args += $flag_debug
+$compiler_args += ( $flag_path_debug + $path_build + '\' )
+# Use the static, multithreaded, debug runtime library
+$compiler_args += $flag_link_win_rt_static_debug
+
+# Include setup
+$compiler_args += ($flag_include + $path_root)
+
+$unit_name   = "watl.v0.lottes"
+
+# Specify unit to compile
+$unit           = join-path $path_root "C\$unit_name.c"
+$compiler_args += $flag_compile, $unit
+
+# Diagnoistc print for the args
+$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"
+$object = join-path $path_build "$unit_name.obj"
+
+$pdb         = join-path $path_build "$unit_name.pdb"
+$map         = join-path $path_build "$unit_name.map"
+$rdi         = join-path $path_build "$unit_name.rdi"
+$rdi_listing = join-path $path_build "$unit_name.rdi.list"
+
+if ($true) {
+	write-host "Linking with lld-link"
+
+	$linker_args = @()
+	$linker_args += $flag_nologo
+	$linker_args += $flag_link_win_machine_64
+	$linker_args += $flag_link_no_incremental
+	$linker_args += ($flag_link_win_path_output + $binary)
+
+	$linker_args += "$flag_link_win_debug"
+	$linker_args += $flag_link_win_pdb + $pdb
+	$linker_args += $flag_link_mapfile + $map
+    $linker_args += $flag_link_win_subsystem_console
+
+	$linker_args += $object
+    
+    # Add necessary libraries for a basic Windows application
+    $linker_args += "kernel32.lib", "user32.lib", "gdi32.lib"
+
+	# Diagnoistc print for the args
+	$linker_args | ForEach-Object { Write-Host $_ }
+
+	$linking_time = Measure-Command { & $linker $linker_args }
+	# & $radlink $linker_args
+	write-host "Linking took $($linking_time.TotalMilliseconds)ms"
+	write-host
+}
+
+if ($false) {
+	write-host "Dumping Debug Info"
+
+	$rbin_out  = '--out:'
+	$rbin_dump = '--dump'
+
+	$nargs = @($pdb, ($rbin_out + $rdi))
+	& $radbin $nargs
+
+	$nargs = @($rbin_dump, $rdi)
+	$dump = & $radbin $nargs
+	$dump > $rdi_listing
+}
+
+Pop-Location

scripts/build.odin.odin

@@ -106,7 +106,8 @@ main :: proc() {
 		file_source,
 		flag_file,
 		join_str(flag_output_path, file_exe),
-		flag_optimize_none,
+		// flag_optimize_none,
+		falg_optimize_aggressive,
 		flag_default_allocator_nil,
 		flag_debug,
 		flag_microarch_zen5,