v.pass()

v.pass() connects a Kernel to its data.

const node = v.pass(kernel, {
    input,
    output,
    multiplier
});

It does not submit GPU work by itself. It prepares work that can be executed later with v.run() or v.wait().

The returned value is a node.

Signature

const node = v.pass(kernel, bindings, options);

Argument	Meaning
kernel	A Kernel instance
bindings	Buffer, RawBuffer, Uniform, or Handle values
options	Optional pass-level overrides

Pass Options

const node = v.pass(kernel, { input, output }, {
    label: 'scale pass',
    threads: input.count
});

threads overrides the kernel’s thread configuration for this pass.

label gives the node a readable name in debug output and browser GPU tools.

debug enables shader debugging for this pass.

const node = v.pass(kernel, { input, output }, {
    debug: true
});

What v.pass() Does

When you call v.pass(), Volten:

• validates the kernel and bindings
• resolves thread bounds and dispatch size
• generates WGSL bindings
• assembles the final shader
• creates or reuses a compute pipeline
• returns a runnable node

Note

v.pass() prepares work. v.run() and v.wait() submit it.