Add sdl3_gpu.odin

4 files changed, 923 insertions, 38 deletions

diff --git a/vendor/sdl3/sdl3_audio.odin b/vendor/sdl3/sdl3_audio.odin
index 698193b32..43a8a34d2 100644
--- a/vendor/sdl3/sdl3_audio.odin
+++ b/vendor/sdl3/sdl3_audio.odin
@@ -14,28 +14,28 @@ DEFINE_AUDIO_FORMAT :: #force_inline proc "c" (signed, bigendian, flt, size: Uin
 
 
 AudioFormat :: enum c.int {
-    UNKNOWN   = 0x0000,  /**< Unspecified audio format */
-    U8        = 0x0008,  /**< Unsigned 8-bit samples */
-        /* DEFINE_AUDIO_FORMAT(0, 0, 0, 8), */
-    S8        = 0x8008,  /**< Signed 8-bit samples */
-        /* DEFINE_AUDIO_FORMAT(1, 0, 0, 8), */
-    S16LE     = 0x8010,  /**< Signed 16-bit samples */
-        /* DEFINE_AUDIO_FORMAT(1, 0, 0, 16), */
-    S16BE     = 0x9010,  /**< As above, but big-endian byte order */
-        /* DEFINE_AUDIO_FORMAT(1, 1, 0, 16), */
-    S32LE     = 0x8020,  /**< 32-bit integer samples */
-        /* DEFINE_AUDIO_FORMAT(1, 0, 0, 32), */
-    S32BE     = 0x9020,  /**< As above, but big-endian byte order */
-        /* DEFINE_AUDIO_FORMAT(1, 1, 0, 32), */
-    F32LE     = 0x8120,  /**< 32-bit floating point samples */
-        /* DEFINE_AUDIO_FORMAT(1, 0, 1, 32), */
-    F32BE     = 0x9120,  /**< As above, but big-endian byte order */
-        /* DEFINE_AUDIO_FORMAT(1, 1, 1, 32), */
-
-    /* These represent the current system's byteorder. */
-    S16 = S16LE when BYTEORDER == LIL_ENDIAN else S16BE,
-    S32 = S32LE when BYTEORDER == LIL_ENDIAN else S32BE,
-    F32 = F32LE when BYTEORDER == LIL_ENDIAN else F32BE,
+	UNKNOWN   = 0x0000,  /**< Unspecified audio format */
+	U8        = 0x0008,  /**< Unsigned 8-bit samples */
+	                     /* DEFINE_AUDIO_FORMAT(0, 0, 0, 8), */
+	S8        = 0x8008,  /**< Signed 8-bit samples */
+	                     /* DEFINE_AUDIO_FORMAT(1, 0, 0, 8), */
+	S16LE     = 0x8010,  /**< Signed 16-bit samples */
+	                     /* DEFINE_AUDIO_FORMAT(1, 0, 0, 16), */
+	S16BE     = 0x9010,  /**< As above, but big-endian byte order */
+	                     /* DEFINE_AUDIO_FORMAT(1, 1, 0, 16), */
+	S32LE     = 0x8020,  /**< 32-bit integer samples */
+	                     /* DEFINE_AUDIO_FORMAT(1, 0, 0, 32), */
+	S32BE     = 0x9020,  /**< As above, but big-endian byte order */
+	                     /* DEFINE_AUDIO_FORMAT(1, 1, 0, 32), */
+	F32LE     = 0x8120,  /**< 32-bit floating point samples */
+	                     /* DEFINE_AUDIO_FORMAT(1, 0, 1, 32), */
+	F32BE     = 0x9120,  /**< As above, but big-endian byte order */
+	                     /* DEFINE_AUDIO_FORMAT(1, 1, 1, 32), */
+
+	/* These represent the current system's byteorder. */
+	S16 = S16LE when BYTEORDER == LIL_ENDIAN else S16BE,
+	S32 = S32LE when BYTEORDER == LIL_ENDIAN else S32BE,
+	F32 = F32LE when BYTEORDER == LIL_ENDIAN else F32BE,
 }
 
 @(require_results) AUDIO_BITSIZE        :: proc "c" (x: AudioFormat) -> Uint16 { return (Uint16(x) & AUDIO_MASK_BITSIZE)       }
diff --git a/vendor/sdl3/sdl3_gamepad.odin b/vendor/sdl3/sdl3_gamepad.odin
index d5778b446..225accaee 100644
--- a/vendor/sdl3/sdl3_gamepad.odin
+++ b/vendor/sdl3/sdl3_gamepad.odin
@@ -81,10 +81,10 @@ GamepadBindingType :: enum c.int {
 GamepadBinding :: struct {
 	input_type: GamepadBindingType,
 	input: struct #raw_union {
-        	button: c.int,
+		button: c.int,
 
-        	axis: struct {
-        		axis: c.int,
+		axis: struct {
+			axis: c.int,
 			axis_min: c.int,
 			axis_max: c.int,
 		},
diff --git a/vendor/sdl3/sdl3_gpu.odin b/vendor/sdl3/sdl3_gpu.odin
new file mode 100644
index 000000000..1e2a1fd19
--- /dev/null
+++ b/vendor/sdl3/sdl3_gpu.odin
@@ -0,0 +1,885 @@
+package sdl3
+
+import "core:c"
+
+GPUDevice           :: struct {}
+GPUBuffer           :: struct {}
+GPUTransferBuffer   :: struct {}
+GPUTexture          :: struct {}
+GPUSampler          :: struct {}
+GPUShader           :: struct {}
+GPUComputePipeline  :: struct {}
+GPUGraphicsPipeline :: struct {}
+GPUCommandBuffer    :: struct {}
+GPURenderPass       :: struct {}
+GPUComputePass      :: struct {}
+GPUCopyPass         :: struct {}
+GPUFence            :: struct {}
+
+
+GPUPrimitiveType :: enum c.int {
+	TRIANGLELIST,  /**< A series of separate triangles. */
+	TRIANGLESTRIP, /**< A series of connected triangles. */
+	LINELIST,      /**< A series of separate lines. */
+	LINESTRIP,     /**< A series of connected lines. */
+	POINTLIST,     /**< A series of separate points. */
+}
+
+GPULoadOp :: enum c.int {
+	LOAD,      /**< The previous contents of the texture will be preserved. */
+	CLEAR,     /**< The contents of the texture will be cleared to a color. */
+	DONT_CARE, /**< The previous contents of the texture need not be preserved. The contents will be undefined. */
+}
+
+GPUStoreOp :: enum c.int {
+	STORE,             /**< The contents generated during the render pass will be written to memory. */
+	DONT_CARE,         /**< The contents generated during the render pass are not needed and may be discarded. The contents will be undefined. */
+	RESOLVE,           /**< The multisample contents generated during the render pass will be resolved to a non-multisample texture. The contents in the multisample texture may then be discarded and will be undefined. */
+	RESOLVE_AND_STORE, /**< The multisample contents generated during the render pass will be resolved to a non-multisample texture. The contents in the multisample texture will be written to memory. */
+}
+
+GPUIndexElementSize :: enum c.int {
+	_16BIT, /**< The index elements are 16-bit. */
+	_32BIT, /**< The index elements are 32-bit. */
+}
+
+GPUTextureFormat :: enum c.int {
+	INVALID,
+
+    /* Unsigned Normalized Float Color Formats */
+	A8_UNORM,
+	R8_UNORM,
+	R8G8_UNORM,
+	R8G8B8A8_UNORM,
+	R16_UNORM,
+	R16G16_UNORM,
+	R16G16B16A16_UNORM,
+	R10G10B10A2_UNORM,
+	B5G6R5_UNORM,
+	B5G5R5A1_UNORM,
+	B4G4R4A4_UNORM,
+	B8G8R8A8_UNORM,
+    /* Compressed Unsigned Normalized Float Color Formats */
+	BC1_RGBA_UNORM,
+	BC2_RGBA_UNORM,
+	BC3_RGBA_UNORM,
+	BC4_R_UNORM,
+	BC5_RG_UNORM,
+	BC7_RGBA_UNORM,
+    /* Compressed Signed Float Color Formats */
+	BC6H_RGB_FLOAT,
+    /* Compressed Unsigned Float Color Formats */
+	BC6H_RGB_UFLOAT,
+    /* Signed Normalized Float Color Formats  */
+	R8_SNORM,
+	R8G8_SNORM,
+	R8G8B8A8_SNORM,
+	R16_SNORM,
+	R16G16_SNORM,
+	R16G16B16A16_SNORM,
+    /* Signed Float Color Formats */
+	R16_FLOAT,
+	R16G16_FLOAT,
+	R16G16B16A16_FLOAT,
+	R32_FLOAT,
+	R32G32_FLOAT,
+	R32G32B32A32_FLOAT,
+    /* Unsigned Float Color Formats */
+	R11G11B10_UFLOAT,
+    /* Unsigned Integer Color Formats */
+	R8_UINT,
+	R8G8_UINT,
+	R8G8B8A8_UINT,
+	R16_UINT,
+	R16G16_UINT,
+	R16G16B16A16_UINT,
+	R32_UINT,
+	R32G32_UINT,
+	R32G32B32A32_UINT,
+    /* Signed Integer Color Formats */
+	R8_INT,
+	R8G8_INT,
+	R8G8B8A8_INT,
+	R16_INT,
+	R16G16_INT,
+	R16G16B16A16_INT,
+	R32_INT,
+	R32G32_INT,
+	R32G32B32A32_INT,
+    /* SRGB Unsigned Normalized Color Formats */
+	R8G8B8A8_UNORM_SRGB,
+	B8G8R8A8_UNORM_SRGB,
+    /* Compressed SRGB Unsigned Normalized Color Formats */
+	BC1_RGBA_UNORM_SRGB,
+	BC2_RGBA_UNORM_SRGB,
+	BC3_RGBA_UNORM_SRGB,
+	BC7_RGBA_UNORM_SRGB,
+    /* Depth Formats */
+	D16_UNORM,
+	D24_UNORM,
+	D32_FLOAT,
+	D24_UNORM_S8_UINT,
+	D32_FLOAT_S8_UINT,
+    /* Compressed ASTC Normalized Float Color Formats*/
+	ASTC_4x4_UNORM,
+	ASTC_5x4_UNORM,
+	ASTC_5x5_UNORM,
+	ASTC_6x5_UNORM,
+	ASTC_6x6_UNORM,
+	ASTC_8x5_UNORM,
+	ASTC_8x6_UNORM,
+	ASTC_8x8_UNORM,
+	ASTC_10x5_UNORM,
+	ASTC_10x6_UNORM,
+	ASTC_10x8_UNORM,
+	ASTC_10x10_UNORM,
+	ASTC_12x10_UNORM,
+	ASTC_12x12_UNORM,
+    /* Compressed SRGB ASTC Normalized Float Color Formats*/
+	ASTC_4x4_UNORM_SRGB,
+	ASTC_5x4_UNORM_SRGB,
+	ASTC_5x5_UNORM_SRGB,
+	ASTC_6x5_UNORM_SRGB,
+	ASTC_6x6_UNORM_SRGB,
+	ASTC_8x5_UNORM_SRGB,
+	ASTC_8x6_UNORM_SRGB,
+	ASTC_8x8_UNORM_SRGB,
+	ASTC_10x5_UNORM_SRGB,
+	ASTC_10x6_UNORM_SRGB,
+	ASTC_10x8_UNORM_SRGB,
+	ASTC_10x10_UNORM_SRGB,
+	ASTC_12x10_UNORM_SRGB,
+	ASTC_12x12_UNORM_SRGB,
+    /* Compressed ASTC Signed Float Color Formats*/
+	ASTC_4x4_FLOAT,
+	ASTC_5x4_FLOAT,
+	ASTC_5x5_FLOAT,
+	ASTC_6x5_FLOAT,
+	ASTC_6x6_FLOAT,
+	ASTC_8x5_FLOAT,
+	ASTC_8x6_FLOAT,
+	ASTC_8x8_FLOAT,
+	ASTC_10x5_FLOAT,
+	ASTC_10x6_FLOAT,
+	ASTC_10x8_FLOAT,
+	ASTC_10x10_FLOAT,
+	ASTC_12x10_FLOAT,
+	ASTC_12x12_FLOAT,
+}
+
+GPUTextureUsageFlags :: distinct bit_set[GPUTextureUsageFlag; Uint32]
+GPUTextureUsageFlag :: enum Uint32 {
+	SAMPLER                                 = 0, /**< Texture supports sampling. */
+	COLOR_TARGET                            = 1, /**< Texture is a color render target. */
+	DEPTH_STENCIL_TARGET                    = 2, /**< Texture is a depth stencil target. */
+	GRAPHICS_STORAGE_READ                   = 3, /**< Texture supports storage reads in graphics stages. */
+	COMPUTE_STORAGE_READ                    = 4, /**< Texture supports storage reads in the compute stage. */
+	COMPUTE_STORAGE_WRITE                   = 5, /**< Texture supports storage writes in the compute stage. */
+	COMPUTE_STORAGE_SIMULTANEOUS_READ_WRITE = 6, /**< Texture supports reads and writes in the same compute shader. This is NOT equivalent to READ | WRITE. */
+}
+
+GPUTextureType :: enum c.int {
+	D2,         /**< The texture is a 2-dimensional image. */
+	D2_ARRAY,   /**< The texture is a 2-dimensional array image. */
+	D3,         /**< The texture is a 3-dimensional image. */
+	CUBE,       /**< The texture is a cube image. */
+	CUBE_ARRAY, /**< The texture is a cube array image. */
+}
+
+GPUSampleCount :: enum c.int {
+	_1,  /**< No multisampling. */
+	_2,  /**< MSAA 2x */
+	_4,  /**< MSAA 4x */
+	_8,  /**< MSAA 8x */
+}
+
+GPUCubeMapFace :: enum c.int {
+	POSITIVEX,
+	NEGATIVEX,
+	POSITIVEY,
+	NEGATIVEY,
+	POSITIVEZ,
+	NEGATIVEZ,
+}
+
+
+GPUBufferUsageFlags :: distinct bit_set[GPUBufferUsageFlag; Uint32]
+GPUBufferUsageFlag :: enum Uint32 {
+	VERTEX                = 0, /**< Buffer is a vertex buffer. */
+	INDEX                 = 1, /**< Buffer is an index buffer. */
+	INDIRECT              = 2, /**< Buffer is an indirect buffer. */
+	GRAPHICS_STORAGE_READ = 3, /**< Buffer supports storage reads in graphics stages. */
+	COMPUTE_STORAGE_READ  = 4, /**< Buffer supports storage reads in the compute stage. */
+	COMPUTE_STORAGE_WRITE = 5, /**< Buffer supports storage writes in the compute stage. */
+}
+
+GPUTransferBufferUsage :: enum c.int {
+	UPLOAD,
+	DOWNLOAD,
+}
+
+GPUShaderStage :: enum c.int {
+	VERTEX,
+	FRAGMENT,
+}
+
+GPUShaderFormat :: distinct bit_set[GPUShaderFormatFlag; Uint32]
+GPUShaderFormatFlag :: enum Uint32 {
+	PRIVATE  = 0, /**< Shaders for NDA'd platforms. */
+	SPIRV    = 1, /**< SPIR-V shaders for Vulkan. */
+	DXBC     = 2, /**< DXBC SM5_1 shaders for D3D12. */
+	DXIL     = 3, /**< DXIL SM6_0 shaders for D3D12. */
+	MSL      = 4, /**< MSL shaders for Metal. */
+	METALLIB = 5, /**< Precompiled metallib shaders for Metal. */
+}
+GPU_SHADERFORMAT_INVALID :: GPUShaderFormat{}
+
+GPUVertexElementFormat :: enum c.int {
+	INVALID,
+
+	/* 32-bit Signed Integers */
+	INT,
+	INT2,
+	INT3,
+	INT4,
+
+	/* 32-bit Unsigned Integers */
+	UINT,
+	UINT2,
+	UINT3,
+	UINT4,
+
+	/* 32-bit Floats */
+	FLOAT,
+	FLOAT2,
+	FLOAT3,
+	FLOAT4,
+
+	/* 8-bit Signed Integers */
+	BYTE2,
+	BYTE4,
+
+	/* 8-bit Unsigned Integers */
+	UBYTE2,
+	UBYTE4,
+
+	/* 8-bit Signed Normalized */
+	BYTE2_NORM,
+	BYTE4_NORM,
+
+	/* 8-bit Unsigned Normalized */
+	UBYTE2_NORM,
+	UBYTE4_NORM,
+
+	/* 16-bit Signed Integers */
+	SHORT2,
+	SHORT4,
+
+	/* 16-bit Unsigned Integers */
+	USHORT2,
+	USHORT4,
+
+	/* 16-bit Signed Normalized */
+	SHORT2_NORM,
+	SHORT4_NORM,
+
+	/* 16-bit Unsigned Normalized */
+	USHORT2_NORM,
+	USHORT4_NORM,
+
+	/* 16-bit Floats */
+	HALF2,
+	HALF4,
+}
+
+GPUVertexInputRate :: enum c.int {
+	VERTEX,   /**< Attribute addressing is a function of the vertex index. */
+	INSTANCE, /**< Attribute addressing is a function of the instance index. */
+}
+
+
+GPUFillMode :: enum c.int {
+	FILL,  /**< Polygons will be rendered via rasterization. */
+	LINE,  /**< Polygon edges will be drawn as line segments. */
+}
+
+GPUCullMode :: enum c.int {
+	NONE,   /**< No triangles are culled. */
+	FRONT,  /**< Front-facing triangles are culled. */
+	BACK,   /**< Back-facing triangles are culled. */
+}
+
+GPUFrontFace :: enum c.int {
+	COUNTER_CLOCKWISE,  /**< A triangle with counter-clockwise vertex winding will be considered front-facing. */
+	CLOCKWISE,          /**< A triangle with clockwise vertex winding will be considered front-facing. */
+}
+
+GPUCompareOp :: enum c.int {
+	INVALID,
+	NEVER,             /**< The comparison always evaluates false. */
+	LESS,              /**< The comparison evaluates reference < test. */
+	EQUAL,             /**< The comparison evaluates reference == test. */
+	LESS_OR_EQUAL,     /**< The comparison evaluates reference <= test. */
+	GREATER,           /**< The comparison evaluates reference > test. */
+	NOT_EQUAL,         /**< The comparison evaluates reference != test. */
+	GREATER_OR_EQUAL,  /**< The comparison evalutes reference >= test. */
+	ALWAYS,            /**< The comparison always evaluates true. */
+}
+
+GPUStencilOp :: enum c.int {
+	INVALID,
+	KEEP,                 /**< Keeps the current value. */
+	ZERO,                 /**< Sets the value to 0. */
+	REPLACE,              /**< Sets the value to reference. */
+	INCREMENT_AND_CLAMP,  /**< Increments the current value and clamps to the maximum value. */
+	DECREMENT_AND_CLAMP,  /**< Decrements the current value and clamps to 0. */
+	INVERT,               /**< Bitwise-inverts the current value. */
+	INCREMENT_AND_WRAP,   /**< Increments the current value and wraps back to 0. */
+	DECREMENT_AND_WRAP,   /**< Decrements the current value and wraps to the maximum value. */
+}
+
+GPUBlendOp :: enum c.int {
+	INVALID,
+	ADD,               /**< (source * source_factor) + (destination * destination_factor) */
+	SUBTRACT,          /**< (source * source_factor) - (destination * destination_factor) */
+	REVERSE_SUBTRACT,  /**< (destination * destination_factor) - (source * source_factor) */
+	MIN,               /**< min(source, destination) */
+	MAX,               /**< max(source, destination) */
+}
+
+GPUBlendFactor :: enum c.int {
+	INVALID,
+	ZERO,                      /**< 0 */
+	ONE,                       /**< 1 */
+	SRC_COLOR,                 /**< source color */
+	ONE_MINUS_SRC_COLOR,       /**< 1 - source color */
+	DST_COLOR,                 /**< destination color */
+	ONE_MINUS_DST_COLOR,       /**< 1 - destination color */
+	SRC_ALPHA,                 /**< source alpha */
+	ONE_MINUS_SRC_ALPHA,       /**< 1 - source alpha */
+	DST_ALPHA,                 /**< destination alpha */
+	ONE_MINUS_DST_ALPHA,       /**< 1 - destination alpha */
+	CONSTANT_COLOR,            /**< blend constant */
+	ONE_MINUS_CONSTANT_COLOR,  /**< 1 - blend constant */
+	SRC_ALPHA_SATURATE,        /**< min(source alpha, 1 - destination alpha) */
+}
+
+
+GPUColorComponentFlags :: distinct bit_set[GPUColorComponentFlag; Uint8]
+GPUColorComponentFlag :: enum Uint8 {
+	R = 0, /**< the red component */
+	G = 1, /**< the green component */
+	B = 2, /**< the blue component */
+	A = 3, /**< the alpha component */
+}
+
+GPUFilter :: enum c.int {
+	NEAREST,  /**< Point filtering. */
+	LINEAR,   /**< Linear filtering. */
+}
+
+GPUSamplerMipmapMode :: enum c.int {
+	NEAREST,  /**< Point filtering. */
+	LINEAR,   /**< Linear filtering. */
+}
+
+GPUSamplerAddressMode :: enum c.int {
+	REPEAT,           /**< Specifies that the coordinates will wrap around. */
+	MIRRORED_REPEAT,  /**< Specifies that the coordinates will wrap around mirrored. */
+	CLAMP_TO_EDGE,    /**< Specifies that the coordinates will clamp to the 0-1 range. */
+}
+
+GPUPresentMode :: enum c.int {
+	VSYNC,
+	IMMEDIATE,
+	MAILBOX,
+}
+
+GPUSwapchainComposition :: enum c.int {
+	SDR,
+	SDR_LINEAR,
+	HDR_EXTENDED_LINEAR,
+	HDR10_ST2084,
+}
+
+
+GPUViewport :: struct {
+	x: f32,          /**< The left offset of the viewport. */
+	y: f32,          /**< The top offset of the viewport. */
+	w: f32,          /**< The width of the viewport. */
+	h: f32,          /**< The height of the viewport. */
+	min_depth: f32,  /**< The minimum depth of the viewport. */
+	max_depth: f32,  /**< The maximum depth of the viewport. */
+}
+
+
+GPUTextureTransferInfo :: struct {
+	transfer_buffer: ^GPUTransferBuffer,  /**< The transfer buffer used in the transfer operation. */
+	offset:          Uint32,              /**< The starting byte of the image data in the transfer buffer. */
+	pixels_per_row:  Uint32,              /**< The number of pixels from one row to the next. */
+	rows_per_layer:  Uint32,              /**< The number of rows from one layer/depth-slice to the next. */
+}
+
+
+GPUTransferBufferLocation :: struct {
+	transfer_buffer: ^GPUTransferBuffer,  /**< The transfer buffer used in the transfer operation. */
+	offset:          Uint32,              /**< The starting byte of the buffer data in the transfer buffer. */
+}
+
+GPUTextureLocation :: struct {
+	texture:   ^GPUTexture,  /**< The texture used in the copy operation. */
+	mip_level: Uint32,       /**< The mip level index of the location. */
+	layer:     Uint32,       /**< The layer index of the location. */
+	x:         Uint32,       /**< The left offset of the location. */
+	y:         Uint32,       /**< The top offset of the location. */
+	z:         Uint32,       /**< The front offset of the location. */
+}
+
+GPUTextureRegion :: struct {
+	texture:   ^GPUTexture,  /**< The texture used in the copy operation. */
+	mip_level: Uint32,       /**< The mip level index to transfer. */
+	layer:     Uint32,       /**< The layer index to transfer. */
+	x:         Uint32,       /**< The left offset of the region. */
+	y:         Uint32,       /**< The top offset of the region. */
+	z:         Uint32,       /**< The front offset of the region. */
+	w:         Uint32,       /**< The width of the region. */
+	h:         Uint32,       /**< The height of the region. */
+	d:         Uint32,       /**< The depth of the region. */
+}
+
+GPUBlitRegion :: struct {
+	texture:              ^GPUTexture,  /**< The texture. */
+	mip_level:            Uint32,       /**< The mip level index of the region. */
+	layer_or_depth_plane: Uint32,       /**< The layer index or depth plane of the region. This value is treated as a layer index on 2D array and cube textures, and as a depth plane on 3D textures. */
+	x:                    Uint32,       /**< The left offset of the region. */
+	y:                    Uint32,       /**< The top offset of the region.  */
+	w:                    Uint32,       /**< The width of the region. */
+	h:                    Uint32,       /**< The height of the region. */
+}
+
+GPUBufferLocation :: struct {
+	buffer: ^GPUBuffer,  /**< The buffer. */
+	offset: Uint32,      /**< The starting byte within the buffer. */
+}
+
+
+GPUBufferRegion :: struct {
+	buffer: ^GPUBuffer,  /**< The buffer. */
+	offset: Uint32,      /**< The starting byte within the buffer. */
+	size:   Uint32,      /**< The size in bytes of the region. */
+}
+
+
+GPUIndirectDrawCommand :: struct {
+	num_vertices:   Uint32,  /**< The number of vertices to draw. */
+	num_instances:  Uint32,  /**< The number of instances to draw. */
+	first_vertex:   Uint32,  /**< The index of the first vertex to draw. */
+	first_instance: Uint32,  /**< The ID of the first instance to draw. */
+}
+
+GPUIndexedIndirectDrawCommand :: struct {
+	num_indices:    Uint32,  /**< The number of indices to draw per instance. */
+	num_instances:  Uint32,  /**< The number of instances to draw. */
+	first_index:    Uint32,  /**< The base index within the index buffer. */
+	vertex_offset:  Sint32,  /**< The value added to the vertex index before indexing into the vertex buffer. */
+	first_instance: Uint32,  /**< The ID of the first instance to draw. */
+}
+
+GPUIndirectDispatchCommand :: struct {
+	groupcount_x: Uint32,  /**< The number of local workgroups to dispatch in the X dimension. */
+	groupcount_y: Uint32,  /**< The number of local workgroups to dispatch in the Y dimension. */
+	groupcount_z: Uint32,  /**< The number of local workgroups to dispatch in the Z dimension. */
+}
+
+
+GPUSamplerCreateInfo :: struct {
+	min_filter:        GPUFilter,              /**< The minification filter to apply to lookups. */
+	mag_filter:        GPUFilter,              /**< The magnification filter to apply to lookups. */
+	mipmap_mode:       GPUSamplerMipmapMode,   /**< The mipmap filter to apply to lookups. */
+	address_mode_u:    GPUSamplerAddressMode,  /**< The addressing mode for U coordinates outside [0, 1). */
+	address_mode_v:    GPUSamplerAddressMode,  /**< The addressing mode for V coordinates outside [0, 1). */
+	address_mode_w:    GPUSamplerAddressMode,  /**< The addressing mode for W coordinates outside [0, 1). */
+	mip_lod_bias:      f32,                    /**< The bias to be added to mipmap LOD calculation. */
+	max_anisotropy:    f32,                    /**< The anisotropy value clamp used by the sampler. If enable_anisotropy is false, this is ignored. */
+	compare_op:        GPUCompareOp,           /**< The comparison operator to apply to fetched data before filtering. */
+	min_lod:           f32,                    /**< Clamps the minimum of the computed LOD value. */
+	max_lod:           f32,                    /**< Clamps the maximum of the computed LOD value. */
+	enable_anisotropy: bool,                   /**< true to enable anisotropic filtering. */
+	enable_compare:    bool,                   /**< true to enable comparison against a reference value during lookups. */
+	_:                 Uint8,
+	_:                 Uint8,
+
+	props:             PropertiesID,           /**< A properties ID for extensions. Should be 0 if no extensions are needed. */
+}
+
+GPUVertexBufferDescription :: struct {
+	slot:               Uint32,              /**< The binding slot of the vertex buffer. */
+	pitch:              Uint32,              /**< The byte pitch between consecutive elements of the vertex buffer. */
+	input_rate:         GPUVertexInputRate,  /**< Whether attribute addressing is a function of the vertex index or instance index. */
+	instance_step_rate: Uint32,              /**< The number of instances to draw using the same per-instance data before advancing in the instance buffer by one element. Ignored unless input_rate is GPU_VERTEXINPUTRATE_INSTANCE */
+}
+
+GPUVertexAttribute :: struct {
+	location:    Uint32,                  /**< The shader input location index. */
+	buffer_slot: Uint32,                  /**< The binding slot of the associated vertex buffer. */
+	format:      GPUVertexElementFormat,  /**< The size and type of the attribute data. */
+	offset:      Uint32,                  /**< The byte offset of this attribute relative to the start of the vertex element. */
+}
+
+GPUVertexInputState :: struct {
+	vertex_buffer_descriptions: [^]GPUVertexBufferDescription `fmt:"v,num_vertex_buffers"`,    /**< A pointer to an array of vertex buffer descriptions. */
+	num_vertex_buffers:         Uint32,                                                        /**< The number of vertex buffer descriptions in the above array. */
+	vertex_attributes:          [^]GPUVertexAttribute         `fmt:"v,num_vertex_attributes"`, /**< A pointer to an array of vertex attribute descriptions. */
+	num_vertex_attributes:      Uint32,                                                        /**< The number of vertex attribute descriptions in the above array. */
+}
+
+GPUStencilOpState :: struct {
+	fail_op:       GPUStencilOp, /**< The action performed on samples that fail the stencil test. */
+	pass_op:       GPUStencilOp, /**< The action performed on samples that pass the depth and stencil tests. */
+	depth_fail_op: GPUStencilOp, /**< The action performed on samples that pass the stencil test and fail the depth test. */
+	compare_op:    GPUCompareOp, /**< The comparison operator used in the stencil test. */
+}
+
+GPUColorTargetBlendState :: struct {
+	src_color_blendfactor:   GPUBlendFactor,          /**< The value to be multiplied by the source RGB value. */
+	dst_color_blendfactor:   GPUBlendFactor,          /**< The value to be multiplied by the destination RGB value. */
+	color_blend_op:          GPUBlendOp,              /**< The blend operation for the RGB components. */
+	src_alpha_blendfactor:   GPUBlendFactor,          /**< The value to be multiplied by the source alpha. */
+	dst_alpha_blendfactor:   GPUBlendFactor,          /**< The value to be multiplied by the destination alpha. */
+	alpha_blend_op:          GPUBlendOp,              /**< The blend operation for the alpha component. */
+	color_write_mask:        GPUColorComponentFlags,  /**< A bitmask specifying which of the RGBA components are enabled for writing. Writes to all channels if enable_color_write_mask is false. */
+	enable_blend:            bool,                    /**< Whether blending is enabled for the color target. */
+	enable_color_write_mask: bool,                    /**< Whether the color write mask is enabled. */
+	_: Uint8,
+	_: Uint8,
+}
+
+GPUShaderCreateInfo :: struct {
+	code_size:            uint,             /**< The size in bytes of the code pointed to. */
+	code:                 [^]Uint8,         /**< A pointer to shader code. */
+	entrypoint:           cstring,          /**< A pointer to a null-terminated UTF-8 string specifying the entry point function name for the shader. */
+	format:               GPUShaderFormat,  /**< The format of the shader code. */
+	stage:                GPUShaderStage,   /**< The stage the shader program corresponds to. */
+	num_samplers:         Uint32,           /**< The number of samplers defined in the shader. */
+	num_storage_textures: Uint32,           /**< The number of storage textures defined in the shader. */
+	num_storage_buffers:  Uint32,           /**< The number of storage buffers defined in the shader. */
+	num_uniform_buffers:  Uint32,           /**< The number of uniform buffers defined in the shader. */
+
+	props:                PropertiesID,     /**< A properties ID for extensions. Should be 0 if no extensions are needed. */
+}
+
+GPUTextureCreateInfo :: struct {
+	type:                 GPUTextureType,        /**< The base dimensionality of the texture. */
+	format:               GPUTextureFormat,      /**< The pixel format of the texture. */
+	usage:                GPUTextureUsageFlags,  /**< How the texture is intended to be used by the client. */
+	width:                Uint32,                /**< The width of the texture. */
+	height:               Uint32,                /**< The height of the texture. */
+	layer_count_or_depth: Uint32,                /**< The layer count or depth of the texture. This value is treated as a layer count on 2D array textures, and as a depth value on 3D textures. */
+	num_levels:           Uint32,                /**< The number of mip levels in the texture. */
+	sample_count:         GPUSampleCount,        /**< The number of samples per texel. Only applies if the texture is used as a render target. */
+
+	props:                PropertiesID,          /**< A properties ID for extensions. Should be 0 if no extensions are needed. */
+}
+
+GPUBufferCreateInfo :: struct {
+	usage: GPUBufferUsageFlags,  /**< How the buffer is intended to be used by the client. */
+	size:  Uint32,               /**< The size in bytes of the buffer. */
+
+	props: PropertiesID,         /**< A properties ID for extensions. Should be 0 if no extensions are needed. */
+}
+
+GPUTransferBufferCreateInfo :: struct {
+	usage: GPUTransferBufferUsage,  /**< How the transfer buffer is intended to be used by the client. */
+	size:  Uint32,                  /**< The size in bytes of the transfer buffer. */
+
+	props: PropertiesID,            /**< A properties ID for extensions. Should be 0 if no extensions are needed. */
+}
+
+GPURasterizerState :: struct {
+	fill_mode:                  GPUFillMode,   /**< Whether polygons will be filled in or drawn as lines. */
+	cull_mode:                  GPUCullMode,   /**< The facing direction in which triangles will be culled. */
+	front_face:                 GPUFrontFace,  /**< The vertex winding that will cause a triangle to be determined as front-facing. */
+	depth_bias_constant_factor: f32,           /**< A scalar factor controlling the depth value added to each fragment. */
+	depth_bias_clamp:           f32,           /**< The maximum depth bias of a fragment. */
+	depth_bias_slope_factor:    f32,           /**< A scalar factor applied to a fragment's slope in depth calculations. */
+	enable_depth_bias:          bool,          /**< true to bias fragment depth values. */
+	enable_depth_clip:          bool,          /**< true to enable depth clip, false to enable depth clamp. */
+	_: Uint8,
+	_: Uint8,
+}
+
+
+GPUMultisampleState :: struct {
+	sample_count: GPUSampleCount,  /**< The number of samples to be used in rasterization. */
+	sample_mask:  Uint32,          /**< Determines which samples get updated in the render targets. Treated as 0xFFFFFFFF if enable_mask is false. */
+	enable_mask:  bool,            /**< Enables sample masking. */
+	_: Uint8,
+	_: Uint8,
+	_: Uint8,
+}
+
+
+GPUDepthStencilState :: struct {
+	compare_op:          GPUCompareOp,       /**< The comparison operator used for depth testing. */
+	back_stencil_state:  GPUStencilOpState,  /**< The stencil op state for back-facing triangles. */
+	front_stencil_state: GPUStencilOpState,  /**< The stencil op state for front-facing triangles. */
+	compare_mask:        Uint8,              /**< Selects the bits of the stencil values participating in the stencil test. */
+	write_mask:          Uint8,              /**< Selects the bits of the stencil values updated by the stencil test. */
+	enable_depth_test:   bool,               /**< true enables the depth test. */
+	enable_depth_write:  bool,               /**< true enables depth writes. Depth writes are always disabled when enable_depth_test is false. */
+	enable_stencil_test: bool,               /**< true enables the stencil test. */
+	_: Uint8,
+	_: Uint8,
+	_: Uint8,
+}
+
+GPUColorTargetDescription :: struct {
+	format:      GPUTextureFormat,          /**< The pixel format of the texture to be used as a color target. */
+	blend_state: GPUColorTargetBlendState,  /**< The blend state to be used for the color target. */
+}
+
+GPUGraphicsPipelineTargetInfo :: struct {
+	color_target_descriptions: [^]GPUColorTargetDescription `fmt:"v,num_color_targets"`,  /**< A pointer to an array of color target descriptions. */
+	num_color_targets:         Uint32,                                                    /**< The number of color target descriptions in the above array. */
+	depth_stencil_format:      GPUTextureFormat,                                          /**< The pixel format of the depth-stencil target. Ignored if has_depth_stencil_target is false. */
+	has_depth_stencil_target:  bool,                                                      /**< true specifies that the pipeline uses a depth-stencil target. */
+	_: Uint8,
+	_: Uint8,
+	_: Uint8,
+}
+
+GPUGraphicsPipelineCreateInfo :: struct {
+	vertex_shader:       ^GPUShader,                     /**< The vertex shader used by the graphics pipeline. */
+	fragment_shader:     ^GPUShader,                     /**< The fragment shader used by the graphics pipeline. */
+	vertex_input_state:  GPUVertexInputState,            /**< The vertex layout of the graphics pipeline. */
+	primitive_type:      GPUPrimitiveType,               /**< The primitive topology of the graphics pipeline. */
+	rasterizer_state:    GPURasterizerState,             /**< The rasterizer state of the graphics pipeline. */
+	multisample_state:   GPUMultisampleState,            /**< The multisample state of the graphics pipeline. */
+	depth_stencil_state: GPUDepthStencilState,           /**< The depth-stencil state of the graphics pipeline. */
+	target_info:         GPUGraphicsPipelineTargetInfo,  /**< Formats and blend modes for the render targets of the graphics pipeline. */
+
+	props:               PropertiesID,                   /**< A properties ID for extensions. Should be 0 if no extensions are needed. */
+}
+
+GPUComputePipelineCreateInfo :: struct {
+	code_size:                      uint,             /**< The size in bytes of the compute shader code pointed to. */
+	code:                           [^]Uint8,         /**< A pointer to compute shader code. */
+	entrypoint:                     cstring,          /**< A pointer to a null-terminated UTF-8 string specifying the entry point function name for the shader. */
+	format:                         GPUShaderFormat,  /**< The format of the compute shader code. */
+	num_samplers:                   Uint32,           /**< The number of samplers defined in the shader. */
+	num_readonly_storage_textures:  Uint32,           /**< The number of readonly storage textures defined in the shader. */
+	num_readonly_storage_buffers:   Uint32,           /**< The number of readonly storage buffers defined in the shader. */
+	num_readwrite_storage_textures: Uint32,           /**< The number of read-write storage textures defined in the shader. */
+	num_readwrite_storage_buffers:  Uint32,           /**< The number of read-write storage buffers defined in the shader. */
+	num_uniform_buffers:            Uint32,           /**< The number of uniform buffers defined in the shader. */
+	threadcount_x:                  Uint32,           /**< The number of threads in the X dimension. This should match the value in the shader. */
+	threadcount_y:                  Uint32,           /**< The number of threads in the Y dimension. This should match the value in the shader. */
+	threadcount_z:                  Uint32,           /**< The number of threads in the Z dimension. This should match the value in the shader. */
+
+	props:                          PropertiesID,     /**< A properties ID for extensions. Should be 0 if no extensions are needed. */
+}
+
+GPUColorTargetInfo :: struct {
+	texture:               ^GPUTexture,  /**< The texture that will be used as a color target by a render pass. */
+	mip_level:             Uint32,       /**< The mip level to use as a color target. */
+	layer_or_depth_plane:  Uint32,       /**< The layer index or depth plane to use as a color target. This value is treated as a layer index on 2D array and cube textures, and as a depth plane on 3D textures. */
+	clear_color:           FColor,       /**< The color to clear the color target to at the start of the render pass. Ignored if GPU_LOADOP_CLEAR is not used. */
+	load_op:               GPULoadOp,    /**< What is done with the contents of the color target at the beginning of the render pass. */
+	store_op:              GPUStoreOp,   /**< What is done with the results of the render pass. */
+	resolve_texture:       ^GPUTexture,  /**< The texture that will receive the results of a multisample resolve operation. Ignored if a RESOLVE* store_op is not used. */
+	resolve_mip_level:     Uint32,       /**< The mip level of the resolve texture to use for the resolve operation. Ignored if a RESOLVE* store_op is not used. */
+	resolve_layer:         Uint32,       /**< The layer index of the resolve texture to use for the resolve operation. Ignored if a RESOLVE* store_op is not used. */
+	cycle:                 bool,         /**< true cycles the texture if the texture is bound and load_op is not LOAD */
+	cycle_resolve_texture: bool,         /**< true cycles the resolve texture if the resolve texture is bound. Ignored if a RESOLVE* store_op is not used. */
+	_: Uint8,
+	_: Uint8,
+}
+
+GPUDepthStencilTargetInfo :: struct {
+	texture:          ^GPUTexture,  /**< The texture that will be used as the depth stencil target by the render pass. */
+	clear_depth:      f32,          /**< The value to clear the depth component to at the beginning of the render pass. Ignored if GPU_LOADOP_CLEAR is not used. */
+	load_op:          GPULoadOp,    /**< What is done with the depth contents at the beginning of the render pass. */
+	store_op:         GPUStoreOp,   /**< What is done with the depth results of the render pass. */
+	stencil_load_op:  GPULoadOp,    /**< What is done with the stencil contents at the beginning of the render pass. */
+	stencil_store_op: GPUStoreOp,   /**< What is done with the stencil results of the render pass. */
+	cycle:            bool,         /**< true cycles the texture if the texture is bound and any load ops are not LOAD */
+	clear_stencil:    Uint8,        /**< The value to clear the stencil component to at the beginning of the render pass. Ignored if GPU_LOADOP_CLEAR is not used. */
+	_: Uint8,
+	_: Uint8,
+}
+
+
+GPUBlitInfo :: struct {
+	source:      GPUBlitRegion,  /**< The source region for the blit. */
+	destination: GPUBlitRegion,  /**< The destination region for the blit. */
+	load_op:     GPULoadOp,      /**< What is done with the contents of the destination before the blit. */
+	clear_color: FColor,         /**< The color to clear the destination region to before the blit. Ignored if load_op is not GPU_LOADOP_CLEAR. */
+	flip_mode:   FlipMode,       /**< The flip mode for the source region. */
+	filter:      GPUFilter,      /**< The filter mode used when blitting. */
+	cycle:       bool,           /**< true cycles the destination texture if it is already bound. */
+	_: Uint8,
+	_: Uint8,
+	_: Uint8,
+}
+
+GPUBufferBinding :: struct {
+	buffer: ^GPUBuffer,  /**< The buffer to bind. Must have been created with GPU_BUFFERUSAGE_VERTEX for BindGPUVertexBuffers, or GPU_BUFFERUSAGE_INDEX for BindGPUIndexBuffer. */
+	offset: Uint32,      /**< The starting byte of the data to bind in the buffer. */
+}
+
+GPUTextureSamplerBinding :: struct {
+	texture: ^GPUTexture,  /**< The texture to bind. Must have been created with GPU_TEXTUREUSAGE_SAMPLER. */
+	sampler: ^GPUSampler,  /**< The sampler to bind. */
+}
+
+GPUStorageBufferReadWriteBinding :: struct {
+	buffer: ^GPUBuffer,  /**< The buffer to bind. Must have been created with GPU_BUFFERUSAGE_COMPUTE_STORAGE_WRITE. */
+	cycle:  bool,        /**< true cycles the buffer if it is already bound. */
+	_: Uint8,
+	_: Uint8,
+	_: Uint8,
+}
+
+GPUStorageTextureReadWriteBinding :: struct {
+	texture:   ^GPUTexture,  /**< The texture to bind. Must have been created with GPU_TEXTUREUSAGE_COMPUTE_STORAGE_WRITE or GPU_TEXTUREUSAGE_COMPUTE_STORAGE_SIMULTANEOUS_READ_WRITE. */
+	mip_level: Uint32,       /**< The mip level index to bind. */
+	layer:     Uint32,       /**< The layer index to bind. */
+	cycle:     bool,         /**< true cycles the texture if it is already bound. */
+	_: Uint8,
+	_: Uint8,
+	_: Uint8,
+}
+
+
+PROP_GPU_DEVICE_CREATE_DEBUGMODE_BOOLEAN          :: "SDL.gpu.device.create.debugmode"
+PROP_GPU_DEVICE_CREATE_PREFERLOWPOWER_BOOLEAN     :: "SDL.gpu.device.create.preferlowpower"
+PROP_GPU_DEVICE_CREATE_NAME_STRING                :: "SDL.gpu.device.create.name"
+PROP_GPU_DEVICE_CREATE_SHADERS_PRIVATE_BOOLEAN    :: "SDL.gpu.device.create.shaders.private"
+PROP_GPU_DEVICE_CREATE_SHADERS_SPIRV_BOOLEAN      :: "SDL.gpu.device.create.shaders.spirv"
+PROP_GPU_DEVICE_CREATE_SHADERS_DXBC_BOOLEAN       :: "SDL.gpu.device.create.shaders.dxbc"
+PROP_GPU_DEVICE_CREATE_SHADERS_DXIL_BOOLEAN       :: "SDL.gpu.device.create.shaders.dxil"
+PROP_GPU_DEVICE_CREATE_SHADERS_MSL_BOOLEAN        :: "SDL.gpu.device.create.shaders.msl"
+PROP_GPU_DEVICE_CREATE_SHADERS_METALLIB_BOOLEAN   :: "SDL.gpu.device.create.shaders.metallib"
+PROP_GPU_DEVICE_CREATE_D3D12_SEMANTIC_NAME_STRING :: "SDL.gpu.device.create.d3d12.semantic"
+
+PROP_GPU_COMPUTEPIPELINE_CREATE_NAME_STRING       :: "SDL.gpu.computepipeline.create.name"
+PROP_GPU_GRAPHICSPIPELINE_CREATE_NAME_STRING      :: "SDL.gpu.graphicspipeline.create.name"
+PROP_GPU_SAMPLER_CREATE_NAME_STRING               :: "SDL.gpu.sampler.create.name"
+PROP_GPU_SHADER_CREATE_NAME_STRING                :: "SDL.gpu.shader.create.name"
+PROP_GPU_TEXTURE_CREATE_D3D12_CLEAR_R_FLOAT       :: "SDL.gpu.texture.create.d3d12.clear.r"
+PROP_GPU_TEXTURE_CREATE_D3D12_CLEAR_G_FLOAT       :: "SDL.gpu.texture.create.d3d12.clear.g"
+PROP_GPU_TEXTURE_CREATE_D3D12_CLEAR_B_FLOAT       :: "SDL.gpu.texture.create.d3d12.clear.b"
+PROP_GPU_TEXTURE_CREATE_D3D12_CLEAR_A_FLOAT       :: "SDL.gpu.texture.create.d3d12.clear.a"
+PROP_GPU_TEXTURE_CREATE_D3D12_CLEAR_DEPTH_FLOAT   :: "SDL.gpu.texture.create.d3d12.clear.depth"
+PROP_GPU_TEXTURE_CREATE_D3D12_CLEAR_STENCIL_UINT8 :: "SDL.gpu.texture.create.d3d12.clear.stencil"
+PROP_GPU_TEXTURE_CREATE_NAME_STRING               :: "SDL.gpu.texture.create.name"
+PROP_GPU_BUFFER_CREATE_NAME_STRING                :: "SDL.gpu.buffer.create.name"
+PROP_GPU_TRANSFERBUFFER_CREATE_NAME_STRING        :: "SDL.gpu.transferbuffer.create.name"
+
+@(default_calling_convention="c", link_prefix="", require_results)
+foreign lib {
+	GPUSupportsShaderFormats              :: proc(format_flags: GPUShaderFormat, name: cstring) -> bool ---
+	GPUSupportsProperties                 :: proc(props: PropertiesID) -> bool ---
+	CreateGPUDevice                       :: proc(format_flags: GPUShaderFormat, debug_mode: bool, name: cstring) -> ^GPUDevice ---
+	CreateGPUDeviceWithProperties         :: proc(props: PropertiesID) -> ^GPUDevice ---
+	DestroyGPUDevice                      :: proc(device: ^GPUDevice) ---
+	GetNumGPUDrivers                      :: proc() -> c.int ---
+	GetGPUDriver                          :: proc(index: c.int) -> cstring ---
+	GetGPUDeviceDriver                    :: proc(device: ^GPUDevice) -> cstring ---
+	GetGPUShaderFormats                   :: proc(device: ^GPUDevice) -> GPUShaderFormat ---
+	CreateGPUComputePipeline              :: proc(device: ^GPUDevice, #by_ptr createinfo: GPUComputePipelineCreateInfo) -> ^GPUComputePipeline ---
+	CreateGPUGraphicsPipeline             :: proc(device: ^GPUDevice, #by_ptr createinfo: GPUGraphicsPipelineCreateInfo) -> ^GPUGraphicsPipeline ---
+	CreateGPUSampler                      :: proc(device: ^GPUDevice, #by_ptr createinfo: GPUSamplerCreateInfo) -> ^GPUSampler ---
+	CreateGPUShader                       :: proc(device: ^GPUDevice, #by_ptr createinfo: GPUShaderCreateInfo) -> ^GPUShader ---
+	CreateGPUTexture                      :: proc(device: ^GPUDevice, #by_ptr createinfo: GPUTextureCreateInfo) -> ^GPUTexture ---
+	CreateGPUBuffer                       :: proc(device: ^GPUDevice, #by_ptr createinfo: GPUBufferCreateInfo) -> ^GPUBuffer ---
+	CreateGPUTransferBuffer               :: proc(device: ^GPUDevice, #by_ptr createinfo: GPUTransferBufferCreateInfo) -> ^GPUTransferBuffer ---
+	SetGPUBufferName                      :: proc(device: ^GPUDevice, buffer: ^GPUBuffer, text: cstring) ---
+	SetGPUTextureName                     :: proc(device: ^GPUDevice, texture: ^GPUTexture, text: cstring) ---
+	InsertGPUDebugLabel                   :: proc(command_buffer: ^GPUCommandBuffer, text: cstring) ---
+	PushGPUDebugGroup                     :: proc(command_buffer: ^GPUCommandBuffer, name: cstring) ---
+	PopGPUDebugGroup                      :: proc(command_buffer: ^GPUCommandBuffer) ---
+	ReleaseGPUTexture                     :: proc(device: ^GPUDevice, texture: ^GPUTexture) ---
+	ReleaseGPUSampler                     :: proc(device: ^GPUDevice, sampler: ^GPUSampler) ---
+	ReleaseGPUBuffer                      :: proc(device: ^GPUDevice, buffer: ^GPUBuffer) ---
+	ReleaseGPUTransferBuffer              :: proc(device: ^GPUDevice, transfer_buffer: ^GPUTransferBuffer) ---
+	ReleaseGPUComputePipeline             :: proc(device: ^GPUDevice, compute_pipeline: ^GPUComputePipeline) ---
+	ReleaseGPUShader                      :: proc(device: ^GPUDevice, shader: ^GPUShader) ---
+	ReleaseGPUGraphicsPipeline            :: proc(device: ^GPUDevice, graphics_pipeline: ^GPUGraphicsPipeline) ---
+	AcquireGPUCommandBuffer               :: proc(device: ^GPUDevice) -> ^GPUCommandBuffer ---
+	PushGPUVertexUniformData              :: proc(command_buffer: ^GPUCommandBuffer, slot_index: Uint32, data: rawptr, length: Uint32) ---
+	PushGPUFragmentUniformData            :: proc(command_buffer: ^GPUCommandBuffer, slot_index: Uint32, data: rawptr, length: Uint32) ---
+	PushGPUComputeUniformData             :: proc(command_buffer: ^GPUCommandBuffer, slot_index: Uint32, data: rawptr, length: Uint32) ---
+	BeginGPURenderPass                    :: proc(command_buffer: ^GPUCommandBuffer, color_target_infos: [^]GPUColorTargetInfo, num_color_targets: Uint32, #by_ptr depth_stencil_target_info: GPUDepthStencilTargetInfo) -> ^GPURenderPass ---
+	BindGPUGraphicsPipeline               :: proc(render_pass: ^GPURenderPass, graphics_pipeline: ^GPUGraphicsPipeline) ---
+	SetGPUViewport                        :: proc(render_pass: ^GPURenderPass, #by_ptr viewport: GPUViewport) ---
+	SetGPUScissor                         :: proc(render_pass: ^GPURenderPass, #by_ptr scissor: Rect) ---
+	SetGPUBlendConstants                  :: proc(render_pass: ^GPURenderPass, blend_constants: FColor) ---
+	SetGPUStencilReference                :: proc(render_pass: ^GPURenderPass, reference: Uint8) ---
+	BindGPUVertexBuffers                  :: proc(render_pass: ^GPURenderPass, first_slot: Uint32, bindings: [^]GPUBufferBinding, num_bindings: Uint32) ---
+	BindGPUIndexBuffer                    :: proc(render_pass: ^GPURenderPass, #by_ptr binding: GPUBufferBinding, index_element_size: GPUIndexElementSize) ---
+	BindGPUVertexSamplers                 :: proc(render_pass: ^GPURenderPass, first_slot: Uint32, texture_sampler_bindings: [^]GPUTextureSamplerBinding, num_bindings: Uint32) ---
+	BindGPUVertexStorageTextures          :: proc(render_pass: ^GPURenderPass, first_slot: Uint32, storage_textures: [^]^GPUTexture, num_bindings: Uint32) ---
+	BindGPUVertexStorageBuffers           :: proc(render_pass: ^GPURenderPass, first_slot: Uint32, storage_buffers: [^]^GPUBuffer, num_bindings: Uint32) ---
+	BindGPUFragmentSamplers               :: proc(render_pass: ^GPURenderPass, first_slot: Uint32, texture_sampler_bindings: [^]GPUTextureSamplerBinding, num_bindings: Uint32) ---
+	BindGPUFragmentStorageTextures        :: proc(render_pass: ^GPURenderPass, first_slot: Uint32, storage_textures: [^]^GPUTexture, num_bindings: Uint32) ---
+	BindGPUFragmentStorageBuffers         :: proc(render_pass: ^GPURenderPass, first_slot: Uint32, storage_buffers: [^]^GPUBuffer, num_bindings: Uint32) ---
+	DrawGPUIndexedPrimitives              :: proc(render_pass: ^GPURenderPass, num_indices: Uint32,  num_instances: Uint32, first_index:  Uint32, vertex_offset: Sint32, first_instance: Uint32) ---
+	DrawGPUPrimitives                     :: proc(render_pass: ^GPURenderPass, num_vertices: Uint32, num_instances: Uint32, first_vertex: Uint32, first_instance: Uint32) ---
+	DrawGPUPrimitivesIndirect             :: proc(render_pass: ^GPURenderPass, buffer: ^GPUBuffer, offset: Uint32, draw_count: Uint32) ---
+	DrawGPUIndexedPrimitivesIndirect      :: proc(render_pass: ^GPURenderPass, buffer: ^GPUBuffer, offset: Uint32, draw_count: Uint32) ---
+	EndGPURenderPass                      :: proc(render_pass: ^GPURenderPass) ---
+	BeginGPUComputePass                   :: proc(command_buffer: ^GPUCommandBuffer, storage_texture_bindings: [^]GPUStorageTextureReadWriteBinding, num_storage_texture_bindings: Uint32, storage_buffer_bindings: [^]GPUStorageBufferReadWriteBinding, num_storage_buffer_bindings: Uint32) -> ^GPUComputePass ---
+	BindGPUComputePipeline                :: proc(compute_pass: ^GPUComputePass, compute_pipeline: ^GPUComputePipeline) ---
+	BindGPUComputeSamplers                :: proc(compute_pass: ^GPUComputePass, first_slot: Uint32, texture_sampler_bindings: [^]GPUTextureSamplerBinding, num_bindings: Uint32) ---
+	BindGPUComputeStorageTextures         :: proc(compute_pass: ^GPUComputePass, first_slot: Uint32, storage_textures: [^]^GPUTexture, num_bindings: Uint32) ---
+	BindGPUComputeStorageBuffers          :: proc(compute_pass: ^GPUComputePass, first_slot: Uint32, storage_buffers: [^]^GPUBuffer, num_bindings: Uint32) ---
+	DispatchGPUCompute                    :: proc(compute_pass: ^GPUComputePass, groupcount_x, groupcount_y, groupcount_z: Uint32) ---
+	DispatchGPUComputeIndirect            :: proc(compute_pass: ^GPUComputePass, buffer: ^GPUBuffer, offset: Uint32) ---
+	EndGPUComputePass                     :: proc(compute_pass: ^GPUComputePass) ---
+	MapGPUTransferBuffer                  :: proc(device: ^GPUDevice, transfer_buffer: ^GPUTransferBuffer, cycle: bool) -> rawptr ---
+	UnmapGPUTransferBuffer                :: proc(device: ^GPUDevice, transfer_buffer: ^GPUTransferBuffer) ---
+	BeginGPUCopyPass                      :: proc(command_buffer: ^GPUCommandBuffer) -> ^GPUCopyPass ---
+	UploadToGPUTexture                    :: proc(copy_pass: ^GPUCopyPass, #by_ptr source: GPUTextureTransferInfo,    #by_ptr destination: GPUTextureRegion, cycle: bool) ---
+	UploadToGPUBuffer                     :: proc(copy_pass: ^GPUCopyPass, #by_ptr source: GPUTransferBufferLocation, #by_ptr destination: GPUBufferRegion, cycle: bool) ---
+	CopyGPUTextureToTexture               :: proc(copy_pass: ^GPUCopyPass, #by_ptr source: GPUTextureLocation,        #by_ptr destination: GPUTextureLocation, w, h, d: Uint32, cycle: bool) ---
+	CopyGPUBufferToBuffer                 :: proc(copy_pass: ^GPUCopyPass, #by_ptr source: GPUBufferLocation,         #by_ptr destination: GPUBufferLocation, size: Uint32, cycle: bool) ---
+	DownloadFromGPUTexture                :: proc(copy_pass: ^GPUCopyPass, #by_ptr source: GPUTextureRegion,          #by_ptr destination: GPUTextureTransferInfo) ---
+	DownloadFromGPUBuffer                 :: proc(copy_pass: ^GPUCopyPass, #by_ptr source: GPUBufferRegion,           #by_ptr destination: GPUTransferBufferLocation) ---
+	EndGPUCopyPass                        :: proc(copy_pass: ^GPUCopyPass) ---
+	GenerateMipmapsForGPUTexture          :: proc(command_buffer: ^GPUCommandBuffer, texture: ^GPUTexture) ---
+	BlitGPUTexture                        :: proc(command_buffer: ^GPUCommandBuffer, #by_ptr info: GPUBlitInfo) ---
+	WindowSupportsGPUSwapchainComposition :: proc(device: ^GPUDevice, window: ^Window, swapchain_composition: GPUSwapchainComposition) -> bool ---
+	WindowSupportsGPUPresentMode          :: proc(device: ^GPUDevice, window: ^Window, present_mode: GPUPresentMode) -> bool ---
+	ClaimWindowForGPUDevice               :: proc(device: ^GPUDevice, window: ^Window) -> bool ---
+	ReleaseWindowFromGPUDevice            :: proc(device: ^GPUDevice, window: ^Window) ---
+	SetGPUSwapchainParameters             :: proc(device: ^GPUDevice, window: ^Window, swapchain_composition: GPUSwapchainComposition, present_mode: GPUPresentMode) -> bool ---
+	SetGPUAllowedFramesInFlight           :: proc(device: ^GPUDevice, allowed_frames_in_flight: Uint32) -> bool ---
+	GetGPUSwapchainTextureFormat          :: proc(device: ^GPUDevice, window: ^Window) -> GPUTextureFormat ---
+	AcquireGPUSwapchainTexture            :: proc(command_buffer: ^GPUCommandBuffer, window: ^Window, swapchain_texture: ^^GPUTexture, swapchain_texture_width, swapchain_texture_height: ^Uint32) -> bool ---
+	WaitForGPUSwapchain                   :: proc(device: ^GPUDevice, window: ^Window) -> bool ---
+	WaitAndAcquireGPUSwapchainTexture     :: proc(command_buffer: ^GPUCommandBuffer, window: ^Window, swapchain_texture: ^^GPUTexture, swapchain_texture_width, swapchain_texture_height: ^Uint32) -> bool ---
+	SubmitGPUCommandBuffer                :: proc(command_buffer: ^GPUCommandBuffer) -> bool ---
+	SubmitGPUCommandBufferAndAcquireFence :: proc(command_buffer: ^GPUCommandBuffer) -> ^GPUFence ---
+	CancelGPUCommandBuffer                :: proc(command_buffer: ^GPUCommandBuffer) -> bool ---
+	WaitForGPUIdle                        :: proc(device: ^GPUDevice) -> bool ---
+	WaitForGPUFences                      :: proc(device: ^GPUDevice, wait_all: bool, fences: [^]^GPUFence, num_fences: Uint32) -> bool ---
+	QueryGPUFence                         :: proc(device: ^GPUDevice, fence: ^GPUFence) -> bool ---
+	ReleaseGPUFence                       :: proc(device: ^GPUDevice, fence: ^GPUFence) ---
+	GPUTextureFormatTexelBlockSize        :: proc(format: GPUTextureFormat) -> Uint32 ---
+	GPUTextureSupportsFormat              :: proc(device: ^GPUDevice, format: GPUTextureFormat, type: GPUTextureType, usage: GPUTextureUsageFlags) -> bool ---
+	GPUTextureSupportsSampleCount         :: proc(device: ^GPUDevice, format: GPUTextureFormat, sample_count: GPUSampleCount) -> bool ---
+	CalculateGPUTextureFormatSize         :: proc(format: GPUTextureFormat, width, height: Uint32, depth_or_layer_count: Uint32) -> Uint32 ---
+}
+
+
+
+
+// GDK
+@(default_calling_convention="c", link_prefix="")
+foreign lib {
+	GDKSuspendGPU :: proc(device: ^GPUDevice) ---
+	GDKResumeGPU  :: proc(device: ^GPUDevice) ---
+}
+
+
diff --git a/vendor/sdl3/sdl3_video.odin b/vendor/sdl3/sdl3_video.odin
index 06ea38286..b7cfab726 100644
--- a/vendor/sdl3/sdl3_video.odin
+++ b/vendor/sdl3/sdl3_video.odin
@@ -128,9 +128,9 @@ WINDOWPOS_ISCENTERED :: proc "c" (X: c.int) -> bool {
 
 
 FlashOperation :: enum c.int {
-    	CANCEL,                   /**< Cancel any window flash state */
-    	BRIEFLY,                  /**< Flash the window briefly to get attention */
-    	UNTIL_FOCUSED,            /**< Flash the window until it gets focus */
+	CANCEL,                   /**< Cancel any window flash state */
+	BRIEFLY,                  /**< Flash the window briefly to get attention */
+	UNTIL_FOCUSED,            /**< Flash the window until it gets focus */
 }
 
 GLContextState :: struct {}
@@ -408,16 +408,16 @@ foreign lib {
 }
 
 HitTestResult :: enum c.int {
-    	NORMAL,             /**< Region is normal. No special properties. */
-    	DRAGGABLE,          /**< Region can drag entire window. */
-    	RESIZE_TOPLEFT,     /**< Region is the resizable top-left corner border. */
-    	RESIZE_TOP,         /**< Region is the resizable top border. */
-    	RESIZE_TOPRIGHT,    /**< Region is the resizable top-right corner border. */
-    	RESIZE_RIGHT,       /**< Region is the resizable right border. */
-    	RESIZE_BOTTOMRIGHT, /**< Region is the resizable bottom-right corner border. */
-    	RESIZE_BOTTOM,      /**< Region is the resizable bottom border. */
-    	RESIZE_BOTTOMLEFT,  /**< Region is the resizable bottom-left corner border. */
-    	RESIZE_LEFT,        /**< Region is the resizable left border. */
+	NORMAL,             /**< Region is normal. No special properties. */
+	DRAGGABLE,          /**< Region can drag entire window. */
+	RESIZE_TOPLEFT,     /**< Region is the resizable top-left corner border. */
+	RESIZE_TOP,         /**< Region is the resizable top border. */
+	RESIZE_TOPRIGHT,    /**< Region is the resizable top-right corner border. */
+	RESIZE_RIGHT,       /**< Region is the resizable right border. */
+	RESIZE_BOTTOMRIGHT, /**< Region is the resizable bottom-right corner border. */
+	RESIZE_BOTTOM,      /**< Region is the resizable bottom border. */
+	RESIZE_BOTTOMLEFT,  /**< Region is the resizable bottom-left corner border. */
+	RESIZE_LEFT,        /**< Region is the resizable left border. */
 }
 
 HitTest :: #type proc "c" (win: ^Window, area: ^Point, data: rawptr) -> HitTestResult