#[repr(C)]pub struct ImageDataLayout {
pub offset: BufferAddress,
pub bytes_per_row: Option<u32>,
pub rows_per_image: Option<u32>,
}
Expand description
Layout of a texture in a buffer’s memory.
The bytes per row and rows per image can be hard to figure out so here are some examples:
Resolution | Format | Bytes per block | Pixels per block | Bytes per row | Rows per image |
---|---|---|---|---|---|
256x256 | RGBA8 | 4 | 1 * 1 * 1 | 256 * 4 = Some(1024) | None |
32x16x8 | RGBA8 | 4 | 1 * 1 * 1 | 32 * 4 = 128 padded to 256 = Some(256) | None |
256x256 | BC3 | 16 | 4 * 4 * 1 | 16 * (256 / 4) = 1024 = Some(1024) | None |
64x64x8 | BC3 | 16 | 4 * 4 * 1 | 16 * (64 / 4) = 256 = Some(256) | 64 / 4 = 16 = Some(16) |
Corresponds to WebGPU GPUImageDataLayout
.
Fields§
§offset: BufferAddress
Offset into the buffer that is the start of the texture. Must be a multiple of texture block size. For non-compressed textures, this is 1.
bytes_per_row: Option<u32>
Bytes per “row” in an image.
A row is one row of pixels or of compressed blocks in the x direction.
This value is required if there are multiple rows (i.e. height or depth is more than one pixel or pixel block for compressed textures)
Must be a multiple of 256 for CommandEncoder::copy_buffer_to_texture
and CommandEncoder::copy_texture_to_buffer
. You must manually pad the
image such that this is a multiple of 256. It will not affect the image data.
Queue::write_texture
does not have this requirement.
Must be a multiple of the texture block size. For non-compressed textures, this is 1.
rows_per_image: Option<u32>
“Rows” that make up a single “image”.
A row is one row of pixels or of compressed blocks in the x direction.
An image is one layer in the z direction of a 3D image or 2DArray texture.
The amount of rows per image may be larger than the actual amount of rows of data.
Required if there are multiple images (i.e. the depth is more than one).
Trait Implementations§
Source§impl Clone for ImageDataLayout
impl Clone for ImageDataLayout
Source§fn clone(&self) -> ImageDataLayout
fn clone(&self) -> ImageDataLayout
1.0.0 · Source§fn clone_from(&mut self, source: &Self)
fn clone_from(&mut self, source: &Self)
source
. Read moreSource§impl Debug for ImageDataLayout
impl Debug for ImageDataLayout
Source§impl Default for ImageDataLayout
impl Default for ImageDataLayout
Source§fn default() -> ImageDataLayout
fn default() -> ImageDataLayout
Source§impl<'de> Deserialize<'de> for ImageDataLayout
impl<'de> Deserialize<'de> for ImageDataLayout
Source§fn deserialize<__D>(__deserializer: __D) -> Result<Self, __D::Error>where
__D: Deserializer<'de>,
fn deserialize<__D>(__deserializer: __D) -> Result<Self, __D::Error>where
__D: Deserializer<'de>,
Source§impl Serialize for ImageDataLayout
impl Serialize for ImageDataLayout
impl Copy for ImageDataLayout
Auto Trait Implementations§
impl Freeze for ImageDataLayout
impl RefUnwindSafe for ImageDataLayout
impl Send for ImageDataLayout
impl Sync for ImageDataLayout
impl Unpin for ImageDataLayout
impl UnwindSafe for ImageDataLayout
Blanket Implementations§
Source§impl<T> BorrowMut<T> for Twhere
T: ?Sized,
impl<T> BorrowMut<T> for Twhere
T: ?Sized,
Source§fn borrow_mut(&mut self) -> &mut T
fn borrow_mut(&mut self) -> &mut T
Source§impl<T> CloneToUninit for Twhere
T: Clone,
impl<T> CloneToUninit for Twhere
T: Clone,
Source§unsafe fn clone_to_uninit(&self, dst: *mut T)
unsafe fn clone_to_uninit(&self, dst: *mut T)
clone_to_uninit
)