Struct Conv2dBackwardDw 

pub struct Conv2dBackwardDw<D: Num> {

    pub name: &'static str,
    pub id: String,
    pub kh: i32,
    pub kw: i32,
    pub stride_h: i32,
    pub stride_w: i32,
    pub pad_h: i32,
    pub pad_w: i32,
    pub g: i32,
    pub block_ow: i32,
    pub kernel_source: String,
    pub entry_point_source: String,
    pub source: String,
    /* private fields */
}

2-D convolution backward pass — gradient with respect to weights (dw).

Uses the same grid as the forward pass. Each CTA at (b, c_out, oh, ow_tile) accumulates partial sums into dw via atomic_add (one per (c_in_local, kh, kw) combination within the group).

G — number of groups. Weight layout: [C_OUT, C_IN/G, KH, KW]. Grid: pid = ((b * C_OUT + c_out) * OH + oh) * num_ow_tiles + ow_tile

dw must be zero-initialised before launch.

Fields

name: &'static str

The kernel function’s name (e.g. "flash_attention2_forward").

id: String

Unique kernel identifier: fn_name + dtype(s) + const values joined by “__”.

kh: i32

Compile-time kernel constant, from the annotated fn’s const generics.

kw: i32

Compile-time kernel constant, from the annotated fn’s const generics.

stride_h: i32

Compile-time kernel constant, from the annotated fn’s const generics.

stride_w: i32

Compile-time kernel constant, from the annotated fn’s const generics.

pad_h: i32

Compile-time kernel constant, from the annotated fn’s const generics.

pad_w: i32

Compile-time kernel constant, from the annotated fn’s const generics.

g: i32

Compile-time kernel constant, from the annotated fn’s const generics.

block_ow: i32

Compile-time kernel constant, from the annotated fn’s const generics.

kernel_source: String

The original kernel function source.

entry_point_source: String

The Rust source of the generated C-ABI entry-point wrapper function.

source: String

Combined source (kernel_source + "\n\n" + entry_point_source); used by the Kernel trait.

Implementations

impl<D: Num> Conv2dBackwardDw<D>

pub fn new( kh: i32, kw: i32, stride_h: i32, stride_w: i32, pad_h: i32, pad_w: i32, g: i32, block_ow: i32, ) -> Self

Constructs a new kernel instance for these compile-time parameters.

Trait Implementations

impl<D: Num> Kernel for Conv2dBackwardDw<D>

type Args<'__a> = (*mut D, *mut D, *mut D, i32, i32, i32, i32, i32, i32, i32)

This kernel’s launch-argument tuple type.

fn id(&self) -> String

A content hash of this kernel’s source, used as a cache key.

fn name(&self) -> &str

This kernel’s name.

fn source(&self) -> &str

This kernel’s full source (DSL + kernel body + entry point wrapper).

fn kernel_source(&self) -> &str

This kernel’s body source, without the entry-point wrapper.

fn entry_point_source(&self) -> &str

Returns the Rust source of the generated C-ABI entry-point wrapper function.

fn entry_point_name(&self) -> String

Returns the PTX symbol name for this kernel: "{name}_entry_point".