fixing example performance issues

.gitignore

@@ -2,3 +2,5 @@
 zig-out/
 .gdb_history
 *.o
+vgcore*
+callgrind*

build.zig

@@ -23,6 +23,11 @@ pub fn build(b: *std.Build) void {
 
     // Zig example setup
 
+    const sdl3 = b.lazyDependency("sdl3", .{
+        .target = target,
+        .optimize = optimize,
+    }) orelse return;
+
     const example_exe = b.addExecutable(.{
         .name = "spirv_interpreter_example",
         .root_module = b.createModule(.{
@@ -31,17 +36,12 @@ pub fn build(b: *std.Build) void {
             .optimize = optimize,
             .imports = &.{
                 .{ .name = "spv", .module = mod },
-                .{ .name = "pretty", .module = pretty.module("pretty") },
+                .{ .name = "sdl3", .module = sdl3.module("sdl3") },
+                //.{ .name = "pretty", .module = pretty.module("pretty") },
             },
         }),
     });
 
-    const sdl3 = b.lazyDependency("sdl3", .{
-        .target = target,
-        .optimize = optimize,
-    }) orelse return;
-    example_exe.root_module.addImport("sdl3", sdl3.module("sdl3"));
-
     const example_install = b.addInstallArtifact(example_exe, .{});
     example_install.step.dependOn(&lib_install.step);
 

example/main.zig

@@ -9,8 +9,11 @@ const screen_height = 480;
 
 pub fn main() !void {
     {
-        var gpa: std.heap.DebugAllocator(.{}) = .init;
+        //var gpa: std.heap.DebugAllocator(.{}) = .init;
-        defer _ = gpa.deinit();
+        //defer _ = gpa.deinit();
+
+        var gpa = std.heap.ArenaAllocator.init(std.heap.page_allocator);
+        defer gpa.deinit();
 
         defer sdl3.shutdown();
         const init_flags = sdl3.InitFlags{ .video = true };
@@ -22,7 +25,7 @@ pub fn main() !void {
 
         const allocator = gpa.allocator();
 
-        var module = try spv.Module.init(allocator, @ptrCast(@alignCast(shader_source)));
+        var module = try spv.Module.init(allocator, @ptrCast(@alignCast(shader_source)), .{});
         defer module.deinit(allocator);
 
         const surface = try window.getSurface();
@@ -72,10 +75,10 @@ pub fn main() !void {
                     try rt.readOutput(f32, output[0..], color);
 
                     const rgba = surface.mapRgba(
-                        @intFromFloat(output[0] * 255.0),
+                        @truncate(@as(u32, @intFromFloat(output[0] * 255.0))),
-                        @intFromFloat(output[1] * 255.0),
+                        @truncate(@as(u32, @intFromFloat(output[1] * 255.0))),
-                        @intFromFloat(output[2] * 255.0),
+                        @truncate(@as(u32, @intFromFloat(output[2] * 255.0))),
-                        @intFromFloat(output[3] * 255.0),
+                        @truncate(@as(u32, @intFromFloat(output[3] * 255.0))),
                     );
 
                     pixel_map[(y * surface.getWidth()) + x] = rgba.value;

src/Runtime.zig

@@ -270,39 +270,39 @@ fn writeValue(self: *const Self, comptime T: type, input: []const T, value: *Res
                 inline else => return RuntimeError.InvalidValueType,
             }
         },
-        .Vector4f32 => |vec| inline for (0..4) |i| switch (T) {
+        .Vector4f32 => |*vec| inline for (0..4) |i| switch (T) {
             f32 => vec[i] = input[i],
             inline else => return RuntimeError.InvalidValueType,
         },
-        .Vector3f32 => |vec| inline for (0..3) |i| switch (T) {
+        .Vector3f32 => |*vec| inline for (0..3) |i| switch (T) {
             f32 => vec[i] = input[i],
             inline else => return RuntimeError.InvalidValueType,
         },
-        .Vector2f32 => |vec| inline for (0..2) |i| switch (T) {
+        .Vector2f32 => |*vec| inline for (0..2) |i| switch (T) {
             f32 => vec[i] = input[i],
             inline else => return RuntimeError.InvalidValueType,
         },
-        .Vector4i32 => |vec| inline for (0..4) |i| switch (T) {
+        .Vector4i32 => |*vec| inline for (0..4) |i| switch (T) {
             i32 => vec[i] = input[i],
             inline else => return RuntimeError.InvalidValueType,
         },
-        .Vector3i32 => |vec| inline for (0..3) |i| switch (T) {
+        .Vector3i32 => |*vec| inline for (0..3) |i| switch (T) {
             i32 => vec[i] = input[i],
             inline else => return RuntimeError.InvalidValueType,
         },
-        .Vector2i32 => |vec| inline for (0..2) |i| switch (T) {
+        .Vector2i32 => |*vec| inline for (0..2) |i| switch (T) {
             i32 => vec[i] = input[i],
             inline else => return RuntimeError.InvalidValueType,
         },
-        .Vector4u32 => |vec| inline for (0..4) |i| switch (T) {
+        .Vector4u32 => |*vec| inline for (0..4) |i| switch (T) {
             u32 => vec[i] = input[i],
             inline else => return RuntimeError.InvalidValueType,
         },
-        .Vector3u32 => |vec| inline for (0..3) |i| switch (T) {
+        .Vector3u32 => |*vec| inline for (0..3) |i| switch (T) {
             u32 => vec[i] = input[i],
             inline else => return RuntimeError.InvalidValueType,
         },
-        .Vector2u32 => |vec| inline for (0..2) |i| switch (T) {
+        .Vector2u32 => |*vec| inline for (0..2) |i| switch (T) {
             u32 => vec[i] = input[i],
             inline else => return RuntimeError.InvalidValueType,
         },