Module dsl 

DSL marker functions for WGSL code generation.

These functions are compile-time markers that the transpiler recognizes and converts to WGSL intrinsics. They do nothing at Rust compile time.

Common functions (thread indices, basic math, synchronization) are shared with the CUDA backend via ringkernel_codegen::dsl_common.

Thread/Workgroup Indices

Function	WGSL Equivalent
thread_idx_x()	local_invocation_id.x
thread_idx_y()	local_invocation_id.y
thread_idx_z()	local_invocation_id.z
block_idx_x()	workgroup_id.x
block_idx_y()	workgroup_id.y
block_idx_z()	workgroup_id.z
block_dim_x()	WORKGROUP_SIZE_X (constant)
global_thread_id()	global_invocation_id.x

Functions

abs

Absolute value. Maps to abs(x) in WGSL.

atomic_add

Atomic add. Maps to atomicAdd(&var, value) in WGSL.

atomic_cas

Atomic compare-and-swap. Maps to atomicCompareExchangeWeak(&var, compare, value) in WGSL. Returns the old value.

atomic_exchange

Atomic exchange. Maps to atomicExchange(&var, value) in WGSL.

atomic_load

Atomic load. Maps to atomicLoad(&var) in WGSL.

atomic_max

Atomic maximum. Maps to atomicMax(&var, value) in WGSL.

atomic_min

Atomic minimum. Maps to atomicMin(&var, value) in WGSL.

atomic_store

Atomic store. Maps to atomicStore(&var, value) in WGSL.

atomic_sub

Atomic subtract. Maps to atomicSub(&var, value) in WGSL.

block_dim_x

Get the block/workgroup dimension (x dimension).

block_dim_y

Get the block/workgroup dimension (y dimension).

block_dim_z

Get the block/workgroup dimension (z dimension).

block_idx_x

Get the block/workgroup index (x dimension).

block_idx_y

Get the block/workgroup index (y dimension).

block_idx_z

Get the block/workgroup index (z dimension).

ceil

Ceiling.

clamp

Clamp. Maps to clamp(x, lo, hi) in WGSL.

cos

Cosine.

exp

Exponential (base e).

floor

Floor.

fma

Fused multiply-add: a * b + c.

global_thread_id

Get the global thread ID (x-component). Maps to global_invocation_id.x in WGSL.

global_thread_id_y

Get the global thread ID in the y-dimension. Maps to global_invocation_id.y in WGSL.

global_thread_id_z

Get the global thread ID in the z-dimension. Maps to global_invocation_id.z in WGSL.

grid_dim_x

Get the grid/dispatch dimension (x dimension).

grid_dim_y

Get the grid/dispatch dimension (y dimension).

grid_dim_z

Get the grid/dispatch dimension (z dimension).

lane_id

Get the lane ID within the subgroup. Maps to subgroup_invocation_id builtin in WGSL.

log

Natural logarithm (base e).

max

Maximum. Maps to max(a, b) in WGSL.

min

Minimum. Maps to min(a, b) in WGSL.

mix

Linear interpolation. Maps to mix(a, b, t) in WGSL.

powf

Power: x^y.

round

Round to nearest.

rsqrt

Reciprocal square root.

sin

Sine.

sqrt

Square root.

sync_threads

Synchronize all threads in a block/workgroup.

tan

Tangent.

thread_fence

Thread/storage memory fence.

thread_fence_block

Block-level/workgroup memory fence.

thread_idx_x

Get the thread index within a block/workgroup (x dimension).

thread_idx_y

Get the thread index within a block/workgroup (y dimension).

thread_idx_z

Get the thread index within a block/workgroup (z dimension).

warp_all

Check if all lanes in subgroup satisfy predicate. Maps to subgroupAll(predicate) in WGSL.

warp_any

Check if any lane in subgroup satisfies predicate. Maps to subgroupAny(predicate) in WGSL.

warp_ballot

Ballot within subgroup. Maps to subgroupBallot(predicate) in WGSL. Returns a vec4 where each bit represents a lane’s predicate.

warp_shuffle

Shuffle within subgroup. Maps to subgroupShuffle(value, lane) in WGSL. Requires: enable chromium_experimental_subgroups;

warp_shuffle_down

Shuffle down within subgroup. Maps to subgroupShuffleDown(value, delta) in WGSL.

warp_shuffle_up

Shuffle up within subgroup. Maps to subgroupShuffleUp(value, delta) in WGSL.

warp_shuffle_xor

Shuffle XOR within subgroup. Maps to subgroupShuffleXor(value, mask) in WGSL.

warp_size

Get the subgroup size. Maps to subgroup_size builtin in WGSL.