feat: add 4d blocked kernel

---
type: pre_commit_static_analysis_report
description: Results of running static analysis checks when committing changes.
report:
  - task: lint_filenames
    status: passed
  - task: lint_editorconfig
    status: passed
  - task: lint_markdown
    status: na
  - task: lint_package_json
    status: na
  - task: lint_repl_help
    status: na
  - task: lint_javascript_src
    status: passed
  - task: lint_javascript_cli
    status: na
  - task: lint_javascript_examples
    status: na
  - task: lint_javascript_tests
    status: na
  - task: lint_javascript_benchmarks
    status: na
  - task: lint_python
    status: na
  - task: lint_r
    status: na
  - task: lint_c_src
    status: na
  - task: lint_c_examples
    status: na
  - task: lint_c_benchmarks
    status: na
  - task: lint_c_tests_fixtures
    status: na
  - task: lint_shell
    status: na
  - task: lint_typescript_declarations
    status: na
  - task: lint_typescript_tests
    status: na
  - task: lint_license_headers
    status: passed
---

Author: kgryte

lib/node_modules/@stdlib/ndarray/base/binary/lib/4d_blocked.js

@@ -0,0 +1,281 @@
+/**
+* @license Apache-2.0
+*
+* Copyright (c) 2025 The Stdlib Authors.
+*
+* Licensed under the Apache License, Version 2.0 (the "License");
+* you may not use this file except in compliance with the License.
+* You may obtain a copy of the License at
+*
+*    http://www.apache.org/licenses/LICENSE-2.0
+*
+* Unless required by applicable law or agreed to in writing, software
+* distributed under the License is distributed on an "AS IS" BASIS,
+* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+* See the License for the specific language governing permissions and
+* limitations under the License.
+*/
+
+/* eslint-disable max-depth, max-statements */
+
+'use strict';
+
+// MODULES //
+
+var loopOrder = require( '@stdlib/ndarray/base/binary-loop-interchange-order' );
+var blockSize = require( '@stdlib/ndarray/base/binary-tiling-block-size' );
+
+
+// MAIN //
+
+/**
+* Applies a binary callback to elements in four-dimensional input ndarrays and assigns results to elements in an equivalently shaped output ndarray via loop blocking.
+*
+* @private
+* @param {Object} x - object containing input ndarray meta data
+* @param {string} x.dtype - data type
+* @param {Collection} x.data - data buffer
+* @param {NonNegativeIntegerArray} x.shape - dimensions
+* @param {IntegerArray} x.strides - stride lengths
+* @param {NonNegativeInteger} x.offset - index offset
+* @param {string} x.order - specifies whether `x` is row-major (C-style) or column-major (Fortran-style)
+* @param {Object} y - object containing input ndarray meta data
+* @param {string} y.dtype - data type
+* @param {Collection} y.data - data buffer
+* @param {NonNegativeIntegerArray} y.shape - dimensions
+* @param {IntegerArray} y.strides - stride lengths
+* @param {NonNegativeInteger} y.offset - index offset
+* @param {string} y.order - specifies whether `y` is row-major (C-style) or column-major (Fortran-style)
+* @param {Object} z - object containing output ndarray meta data
+* @param {string} z.dtype - data type
+* @param {Collection} z.data - data buffer
+* @param {NonNegativeIntegerArray} z.shape - dimensions
+* @param {IntegerArray} z.strides - stride lengths
+* @param {NonNegativeInteger} z.offset - index offset
+* @param {string} z.order - specifies whether `z` is row-major (C-style) or column-major (Fortran-style)
+* @param {Callback} fcn - binary callback
+*
+* @example
+* var Float64Array = require( '@stdlib/array/float64' );
+*
+* function fcn( x, y ) {
+*     return x + y;
+* }
+*
+* // Create data buffers:
+* var xbuf = new Float64Array( [ 1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0, 9.0, 10.0, 11.0, 12.0 ] );
+* var ybuf = new Float64Array( [ 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0 ] );
+* var zbuf = new Float64Array( 12 );
+*
+* // Define the shape of the input and output arrays:
+* var shape = [ 1, 3, 2, 2 ];
+*
+* // Define the array strides:
+* var sx = [ 12, 4, 2, 1 ];
+* var sy = [ 12, 4, 2, 1 ];
+* var sz = [ 12, 4, 2, 1 ];
+*
+* // Define the index offsets:
+* var ox = 0;
+* var oy = 0;
+* var oz = 0;
+*
+* // Create the input and output ndarray-like objects:
+* var x = {
+*     'dtype': 'float64',
+*     'data': xbuf,
+*     'shape': shape,
+*     'strides': sx,
+*     'offset': ox,
+*     'order': 'row-major'
+* };
+* var y = {
+*     'dtype': 'float64',
+*     'data': ybuf,
+*     'shape': shape,
+*     'strides': sy,
+*     'offset': oy,
+*     'order': 'row-major'
+* };
+* var z = {
+*     'dtype': 'float64',
+*     'data': zbuf,
+*     'shape': shape,
+*     'strides': sz,
+*     'offset': oz,
+*     'order': 'row-major'
+* };
+*
+* // Apply the binary function:
+* blockedbinary4d( x, y, z, fcn );
+*
+* console.log( z.data );
+* // => <Float64Array>[ 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0, 9.0, 10.0, 11.0, 12.0, 13.0 ]
+*/
+function blockedbinary4d( x, y, z, fcn ) {
+	var bsize;
+	var xbuf;
+	var ybuf;
+	var zbuf;
+	var dx0;
+	var dx1;
+	var dx2;
+	var dx3;
+	var dy0;
+	var dy1;
+	var dy2;
+	var dy3;
+	var dz0;
+	var dz1;
+	var dz2;
+	var dz3;
+	var ox1;
+	var ox2;
+	var ox3;
+	var oy1;
+	var oy2;
+	var oy3;
+	var oz1;
+	var oz2;
+	var oz3;
+	var sh;
+	var s0;
+	var s1;
+	var s2;
+	var s3;
+	var sx;
+	var sy;
+	var sz;
+	var ox;
+	var oy;
+	var oz;
+	var ix;
+	var iy;
+	var iz;
+	var i0;
+	var i1;
+	var i2;
+	var i3;
+	var j0;
+	var j1;
+	var j2;
+	var j3;
+	var o;
+
+	// Note on variable naming convention: s#, dx#, dy#, dz#, i#, j# where # corresponds to the loop number, with `0` being the innermost loop...
+
+	// Resolve the loop interchange order:
+	o = loopOrder( x.shape, x.strides, y.strides, z.strides );
+	sh = o.sh;
+	sx = o.sx;
+	sy = o.sy;
+	sz = o.sz;
+
+	// Determine the block size:
+	bsize = blockSize( x.dtype, y.dtype, z.dtype );
+
+	// Cache the indices of the first indexed elements in the respective ndarrays...
+	ox = x.offset;
+	oy = y.offset;
+	oz = z.offset;
+
+	// Cache references to the input and output ndarray buffers...
+	xbuf = x.data;
+	ybuf = y.data;
+	zbuf = z.data;
+
+	// Cache offset increments for the innermost loop...
+	dx0 = sx[0];
+	dy0 = sy[0];
+	dz0 = sz[0];
+
+	// Iterate over blocks...
+	for ( j3 = sh[3]; j3 > 0; ) {
+		if ( j3 < bsize ) {
+			s3 = j3;
+			j3 = 0;
+		} else {
+			s3 = bsize;
+			j3 -= bsize;
+		}
+		ox3 = ox + ( j3*sx[3] );
+		oy3 = oy + ( j3*sy[3] );
+		oz3 = oz + ( j3*sz[3] );
+		for ( j2 = sh[2]; j2 > 0; ) {
+			if ( j2 < bsize ) {
+				s2 = j2;
+				j2 = 0;
+			} else {
+				s2 = bsize;
+				j2 -= bsize;
+			}
+			dx3 = sx[3] - ( s2*sx[2] );
+			dy3 = sy[3] - ( s2*sy[2] );
+			dz3 = sz[3] - ( s2*sz[2] );
+			ox2 = ox3 + ( j2*sx[2] );
+			oy2 = oy3 + ( j2*sy[2] );
+			oz2 = oz3 + ( j2*sz[2] );
+			for ( j1 = sh[1]; j1 > 0; ) {
+				if ( j1 < bsize ) {
+					s1 = j1;
+					j1 = 0;
+				} else {
+					s1 = bsize;
+					j1 -= bsize;
+				}
+				dx2 = sx[2] - ( s1*sx[1] );
+				dy2 = sy[2] - ( s1*sy[1] );
+				dz2 = sz[2] - ( s1*sz[1] );
+				ox1 = ox2 + ( j1*sx[1] );
+				oy1 = oy2 + ( j1*sy[1] );
+				oz1 = oz2 + ( j1*sz[1] );
+				for ( j0 = sh[0]; j0 > 0; ) {
+					if ( j0 < bsize ) {
+						s0 = j0;
+						j0 = 0;
+					} else {
+						s0 = bsize;
+						j0 -= bsize;
+					}
+					// Compute index offsets for the first input and output ndarray elements in the current block...
+					ix = ox1 + ( j0*sx[0] );
+					iy = oy1 + ( j0*sy[0] );
+					iz = oz1 + ( j0*sz[0] );
+
+					// Compute loop offset increments...
+					dx1 = sx[1] - ( s0*sx[0] );
+					dy1 = sy[1] - ( s0*sy[0] );
+					dz1 = sz[1] - ( s0*sz[0] );
+
+					// Iterate over the ndarray dimensions...
+					for ( i3 = 0; i3 < s3; i3++ ) {
+						for ( i2 = 0; i2 < s2; i2++ ) {
+							for ( i1 = 0; i1 < s1; i1++ ) {
+								for ( i0 = 0; i0 < s0; i0++ ) {
+									zbuf[ iz ] = fcn( xbuf[ ix ], ybuf[ iy ] );
+									ix += dx0;
+									iy += dy0;
+									iz += dz0;
+								}
+								ix += dx1;
+								iy += dy1;
+								iz += dz1;
+							}
+							ix += dx2;
+							iy += dy2;
+							iz += dz2;
+						}
+						ix += dx3;
+						iy += dy3;
+						iz += dz3;
+					}
+				}
+			}
+		}
+	}
+}
+
+
+// EXPORTS //
+
+module.exports = blockedbinary4d;