Struct GpuContext 

pub struct GpuContext { /* private fields */ }

GPU context for compute operations

This is the main entry point for GPU-accelerated operations. It manages device selection, resource allocation, and command submission.

Implementations

impl GpuContext

pub fn new() -> Result<Self>

Create a new GPU context with automatic device selection

This will select the most suitable GPU device available on the system. If no GPU is available, an error is returned.

Errors

Returns an error if no suitable GPU device is found or if device initialization fails.

pub fn with_device(device_index: usize) -> Result<Self>

Create a new GPU context with a specific device

Arguments

• device_index - Index of the device to use (from list_devices)

Errors

Returns an error if the device index is invalid or if device initialization fails.

pub fn list_devices() -> Result<Vec<GpuDeviceInfo>>

List available GPU devices

Returns information about all GPU devices available on the system.

pub fn device_info(&self) -> &GpuDeviceInfo

Get information about the current device

pub fn rgb_to_yuv(&self, input: &[u8], output: &mut [u8]) -> Result<()>

Convert RGB to YUV (BT.601)

Arguments

• input - Input RGB buffer (packed RGBA format)
• output - Output YUV buffer (packed YUVA format)

Errors

Returns an error if buffer sizes are invalid or if the GPU operation fails.

pub fn yuv_to_rgb(&self, input: &[u8], output: &mut [u8]) -> Result<()>

Convert YUV to RGB (BT.601)

Arguments

• input - Input YUV buffer (packed YUVA format)
• output - Output RGB buffer (packed RGBA format)

Errors

Returns an error if buffer sizes are invalid or if the GPU operation fails.

pub fn scale_bilinear( &self, input: &[u8], src_width: u32, src_height: u32, output: &mut [u8], dst_width: u32, dst_height: u32, ) -> Result<()>

Scale an image using bilinear interpolation

Arguments

• input - Input image buffer (packed RGBA format)
• src_width - Source image width
• src_height - Source image height
• output - Output image buffer (packed RGBA format)
• dst_width - Destination image width
• dst_height - Destination image height

Errors

Returns an error if buffer sizes are invalid or if the GPU operation fails.

pub fn scale_bicubic( &self, input: &[u8], src_width: u32, src_height: u32, output: &mut [u8], dst_width: u32, dst_height: u32, ) -> Result<()>

Scale an image using bicubic interpolation

Arguments

• input - Input image buffer (packed RGBA format)
• src_width - Source image width
• src_height - Source image height
• output - Output image buffer (packed RGBA format)
• dst_width - Destination image width
• dst_height - Destination image height

Errors

Returns an error if buffer sizes are invalid or if the GPU operation fails.

pub fn scale_lanczos( &self, input: &[u8], src_width: u32, src_height: u32, output: &mut [u8], dst_width: u32, dst_height: u32, ) -> Result<()>

Scale an image using Lanczos-3 interpolation (highest quality)

Arguments

• input - Input image buffer (packed RGBA format)
• src_width - Source image width
• src_height - Source image height
• output - Output image buffer (packed RGBA format)
• dst_width - Destination image width
• dst_height - Destination image height

Errors

Returns an error if buffer sizes are invalid or if the GPU operation fails.

pub fn gaussian_blur( &self, input: &[u8], output: &mut [u8], width: u32, height: u32, sigma: f32, ) -> Result<()>

Apply Gaussian blur

Arguments

• input - Input image buffer (packed RGBA format)
• output - Output image buffer (packed RGBA format)
• width - Image width
• height - Image height
• sigma - Blur radius (standard deviation)

Errors

Returns an error if buffer sizes are invalid or if the GPU operation fails.

pub fn sharpen( &self, input: &[u8], output: &mut [u8], width: u32, height: u32, amount: f32, ) -> Result<()>

Apply sharpening filter

Arguments

• input - Input image buffer (packed RGBA format)
• output - Output image buffer (packed RGBA format)
• width - Image width
• height - Image height
• amount - Sharpening strength

Errors

Returns an error if buffer sizes are invalid or if the GPU operation fails.

pub fn edge_detect( &self, input: &[u8], output: &mut [u8], width: u32, height: u32, ) -> Result<()>

Detect edges using Sobel operator

Arguments

• input - Input image buffer (packed RGBA format)
• output - Output image buffer (packed RGBA format)
• width - Image width
• height - Image height

Errors

Returns an error if buffer sizes are invalid or if the GPU operation fails.

pub fn dct_2d( &self, input: &[f32], output: &mut [f32], width: u32, height: u32, ) -> Result<()>

Compute 2D DCT (Discrete Cosine Transform)

Arguments

• input - Input data (f32 values)
• output - Output DCT coefficients
• width - Data width (must be multiple of 8)
• height - Data height (must be multiple of 8)

Errors

Returns an error if dimensions are invalid or if the GPU operation fails.

pub fn idct_2d( &self, input: &[f32], output: &mut [f32], width: u32, height: u32, ) -> Result<()>

Compute 2D IDCT (Inverse Discrete Cosine Transform)

Arguments

• input - Input DCT coefficients
• output - Output reconstructed data
• width - Data width (must be multiple of 8)
• height - Data height (must be multiple of 8)

Errors

Returns an error if dimensions are invalid or if the GPU operation fails.

pub fn wait(&self)

Wait for all GPU operations to complete

This is useful for synchronization and benchmarking.