feat: add C implementation for blas/base/dsyr2

PR-URL: #6572
Ref: #2039
Co-authored-by: Athan Reines <[email protected]>
Reviewed-by: Athan Reines <[email protected]> 
Co-authored-by: stdlib-bot <[email protected]>
Co-authored-by: Aman Bhansali <[email protected]>
Reviewed-by: Aman Bhansali <[email protected]>

Author: 4 people

lib/node_modules/@stdlib/blas/base/dsyr2/README.md

@@ -37,12 +37,12 @@ Performs the symmetric rank 2 operation `A = α*x*y^T + α*y*x^T + A`, where `α
 ```javascript
 var Float64Array = require( '@stdlib/array/float64' );
 
-var A = new Float64Array( [ 1.0, 2.0, 3.0, 0.0, 1.0, 2.0, 0.0, 0.0, 1.0 ] );
+var A = new Float64Array( [ 1.0, 2.0, 3.0, 2.0, 1.0, 2.0, 3.0, 2.0, 1.0 ] );
 var x = new Float64Array( [ 1.0, 2.0, 3.0 ] );
 var y = new Float64Array( [ 1.0, 2.0, 3.0 ] );
 
 dsyr2( 'row-major', 'upper', 3, 1.0, x, 1, y, 1, A, 3 );
-// A => <Float64Array>[ 3.0, 6.0, 9.0, 0.0, 9.0, 14.0, 0.0, 0.0, 19.0 ]
+// A => <Float64Array>[ 3.0, 6.0, 9.0, 2.0, 9.0, 14.0, 3.0, 2.0, 19.0 ]
 ```
 
 The function has the following parameters:
@@ -52,9 +52,9 @@ The function has the following parameters:
 -   **N**: number of elements along each dimension of `A`.
 -   **α**: scalar constant.
 -   **x**: first input [`Float64Array`][mdn-float64array].
--   **sx**: index increment for `x`.
+-   **sx**: stride length for `x`.
 -   **y**: second input [`Float64Array`][mdn-float64array].
--   **sy**: index increment for `y`.
+-   **sy**: stride length for `y`.
 -   **A**: input matrix stored in linear memory as a [`Float64Array`][mdn-float64array].
 -   **lda**: stride of the first dimension of `A` (a.k.a., leading dimension of the matrix `A`).
 
@@ -63,12 +63,12 @@ The stride parameters determine how elements in the input arrays are accessed at
 ```javascript
 var Float64Array = require( '@stdlib/array/float64' );
 
-var A = new Float64Array( [ 1.0, 2.0, 3.0, 0.0, 1.0, 2.0, 0.0, 0.0, 1.0 ] );
+var A = new Float64Array( [ 1.0, 2.0, 3.0, 2.0, 1.0, 2.0, 3.0, 2.0, 1.0 ] );
 var x = new Float64Array( [ 1.0, 2.0, 3.0, 4.0, 5.0 ] );
 var y = new Float64Array( [ 1.0, 2.0, 3.0 ] );
 
 dsyr2( 'row-major', 'upper', 3, 1.0, x, 2, y, 1, A, 3 );
-// A => <Float64Array>[ 3.0, 7.0, 11.0, 0.0, 13.0, 21.0, 0.0, 0.0, 31.0 ]
+// A => <Float64Array>[ 3.0, 7.0, 11.0, 2.0, 13.0, 21.0, 3.0, 2.0, 31.0 ]
 ```
 
 Note that indexing is relative to the first index. To introduce an offset, use [`typed array`][mdn-typed-array] views.
@@ -81,14 +81,14 @@ var Float64Array = require( '@stdlib/array/float64' );
 // Initial arrays...
 var x0 = new Float64Array( [ 0.0, 1.0, 1.0, 1.0 ] );
 var y0 = new Float64Array( [ 0.0, 1.0, 2.0, 3.0 ] );
-var A = new Float64Array( [ 1.0, 2.0, 3.0, 0.0, 1.0, 2.0, 0.0, 0.0, 1.0 ] );
+var A = new Float64Array( [ 1.0, 2.0, 3.0, 2.0, 1.0, 2.0, 3.0, 2.0, 1.0 ] );
 
 // Create offset views...
 var x1 = new Float64Array( x0.buffer, x0.BYTES_PER_ELEMENT*1 ); // start at 2nd element
 var y1 = new Float64Array( y0.buffer, y0.BYTES_PER_ELEMENT*1 ); // start at 2nd element
 
 dsyr2( 'row-major', 'upper', 3, 1.0, x1, 1, y1, 1, A, 3 );
-// A => <Float64Array>[ 3.0, 5.0, 7.0, 0.0, 5.0, 7.0, 0.0, 0.0, 7.0 ]
+// A => <Float64Array>[ 3.0, 5.0, 7.0, 2.0, 5.0, 7.0, 3.0, 2.0, 7.0 ]
 ```
 
 #### dsyr2.ndarray( uplo, N, α, x, sx, ox, y, sy, oy, A, sa1, sa2, oa )
@@ -98,12 +98,12 @@ Performs the symmetric rank 2 operation `A = α*x*y^T + α*y*x^T + A`, using alt
 ```javascript
 var Float64Array = require( '@stdlib/array/float64' );
 
-var A = new Float64Array( [ 1.0, 2.0, 3.0, 0.0, 1.0, 2.0, 0.0, 0.0, 1.0 ] );
+var A = new Float64Array( [ 1.0, 2.0, 3.0, 2.0, 1.0, 2.0, 3.0, 2.0, 1.0 ] );
 var x = new Float64Array( [ 1.0, 2.0, 3.0 ] );
 var y = new Float64Array( [ 1.0, 2.0, 3.0 ] );
 
 dsyr2.ndarray( 'upper', 3, 1.0, x, 1, 0, y, 1, 0, A, 3, 1, 0 );
-// A => <Float64Array>[ 3.0, 6.0, 9.0, 0.0, 9.0, 14.0, 0.0, 0.0, 19.0 ]
+// A => <Float64Array>[ 3.0, 6.0, 9.0, 2.0, 9.0, 14.0, 3.0, 2.0, 19.0 ]
 ```
 
 The function has the following additional parameters:
@@ -119,12 +119,12 @@ While [`typed array`][mdn-typed-array] views mandate a view offset based on the
 ```javascript
 var Float64Array = require( '@stdlib/array/float64' );
 
-var A = new Float64Array( [ 1.0, 2.0, 3.0, 0.0, 1.0, 2.0, 0.0, 0.0, 1.0 ] );
+var A = new Float64Array( [ 1.0, 2.0, 3.0, 2.0, 1.0, 2.0, 3.0, 2.0, 1.0 ] );
 var x = new Float64Array( [ 1.0, 2.0, 3.0, 4.0, 5.0 ] );
 var y = new Float64Array( [ 1.0, 2.0, 3.0 ] );
 
 dsyr2.ndarray( 'upper', 3, 1.0, x, -2, 4, y, 1, 0, A, 3, 1, 0 );
-// A => <Float64Array>[ 11.0, 15.0, 19.0, 0.0, 13.0, 13.0, 0.0, 0.0, 7.0 ]
+// A => <Float64Array>[ 11.0, 15.0, 19.0, 2.0, 13.0, 13.0, 3.0, 2.0, 7.0 ]
 ```
 
 </section>
@@ -158,12 +158,19 @@ var opts = {
 
 var N = 3;
 
-var A = ones( N*N, opts.dtype );
+// Create N-by-N symmetric matrices:
+var A1 = ones( N*N, opts.dtype );
+var A2 = ones( N*N, opts.dtype );
+
+// Create random vectors:
 var x = discreteUniform( N, -10.0, 10.0, opts );
 var y = discreteUniform( N, -10.0, 10.0, opts );
 
-dsyr2( 'row-major', 'upper', 3, 1.0, x, 1, y, 1, A, 3 );
-console.log( A );
+dsyr2( 'row-major', 'upper', 3, 1.0, x, 1, y, 1, A1, 3 );
+console.log( A1 );
+
+dsyr2.ndarray( 'upper', 3, 1.0, x, 1, 0, y, 1, 0, A2, 3, 1, 0 );
+console.log( A2 );
 ```
 
 </section>
@@ -193,21 +200,74 @@ console.log( A );
 ### Usage
 
 ```c
-TODO
+#include "stdlib/blas/base/dsyr2.h"
+```
+
+#### c_dsyr2( order, uplo, N, alpha, \*X, sx, \*Y, sy, \*A, LDA )
+
+Performs the symmetric rank 2 operation `A = α*x*y^T + α*y*x^T + A` where `α` is a scalar, `x` and `y` are `N` element vectors, and `A` is an `N` by `N` symmetric matrix.
+
+```c
+#include "stdlib/blas/base/shared.h"
+
+double A[] = { 1.0, 2.0, 3.0, 2.0, 1.0, 2.0, 3.0, 2.0, 1.0 };
+const double x[] = { 1.0, 2.0, 3.0 };
+const double y[] = { 1.0, 2.0, 3.0 };
+
+c_dsyr2( CblasColMajor, CblasUpper, 3, 1.0, x, 1, y, 1, A, 3 );
 ```
 
-#### TODO
+The function accepts the following arguments:
 
-TODO.
+-   **order**: `[in] CBLAS_LAYOUT` storage layout.
+-   **uplo**: `[in] CBLAS_UPLO` specifies whether the upper or lower triangular part of the symmetric matrix `A` should be referenced.
+-   **N**: `[in] CBLAS_INT` number of elements along each dimension of `A`.
+-   **alpha**: `[in] double` scalar constant.
+-   **X**: `[in] double*` first input array.
+-   **strideX**: `[in] CBLAS_INT` stride length for `X`.
+-   **Y**: `[in] double*` second input array.
+-   **strideY**: `[in] CBLAS_INT` stride length for `Y`.
+-   **A**: `[inout] double*` input matrix.
+-   **LDA**: `[in] CBLAS_INT` stride of the first dimension of `A` (a.k.a., leading dimension of the matrix `A`).
 
 ```c
-TODO
+void c_dsyr2( const CBLAS_LAYOUT order, const CBLAS_UPLO uplo, const CBLAS_INT N, const double alpha, const double *X, const CBLAS_INT strideX, const double *Y, const CBLAS_INT strideY, double *A, const CBLAS_INT LDA )
+```
+
+<!-- lint disable maximum-heading-length -->
+
+#### c_dsyr2_ndarray( uplo, N, alpha, \*X, sx, ox, \*Y, sy, oy, \*A, sa1, sa2, oa )
+
+Performs the symmetric rank 2 operation `A = α*x*y^T + α*y*x^T + A`, using alternative indexing semantics and where `α` is a scalar, `x` and `y` are `N` element vectors, and `A` is an `N` by `N` symmetric matrix.
+
+```c
+#include "stdlib/blas/base/shared.h"
+
+double A[] = { 1.0, 2.0, 3.0, 2.0, 1.0, 2.0, 3.0, 2.0, 1.0 };
+const double x[] = { 1.0, 2.0, 3.0 };
+const double y[] = { 1.0, 2.0, 3.0 };
+
+c_dsyr2_ndarray( CblasUpper, 3, 1.0, x, 1, 0, y, 1, 0, A, 3, 1, 0 );
 ```
 
-TODO
+The function accepts the following arguments:
+
+-   **uplo**: `[in] CBLAS_UPLO` specifies whether the upper or lower triangular part of the symmetric matrix `A` should be referenced.
+-   **N**: `[in] CBLAS_INT` number of elements along each dimension of `A`.
+-   **alpha**: `[in] double` scalar constant.
+-   **X**: `[in] double*` first input array.
+-   **sx**: `[in] CBLAS_INT` stride length for `X`.
+-   **ox**: `[in] CBLAS_INT` starting index for `X`.
+-   **Y**: `[in] double` second input array.
+-   **sy**: `[in] CBLAS_INT` stride length for `Y`.
+-   **oy**: `[in] CBLAS_INT` starting index for `Y`.
+-   **A**: `[inout] double*` input matrix.
+-   **sa1**: `[in] CBLAS_INT` stride of the first dimension of `A`.
+-   **sa2**: `[in] CBLAS_INT` stride of the second dimension of `A`.
+-   **oa**: `[in] CBLAS_INT` starting index for `A`.
 
 ```c
-TODO
+void c_dsyr2_ndarray( const CBLAS_UPLO uplo, const CBLAS_INT N, const double alpha, const double *X, const CBLAS_INT strideX, const CBLAS_INT offsetX, const double *Y, CBLAS_INT strideY, const CBLAS_INT offsetY, double *A, const CBLAS_INT strideA1, const CBLAS_INT strideA2, const CBLAS_INT offsetA )
 ```
 
 </section>
@@ -229,7 +289,47 @@ TODO
 ### Examples
 
 ```c
-TODO
+#include "stdlib/blas/base/dsyr2.h"
+#include "stdlib/blas/base/shared.h"
+#include <stdio.h>
+
+int main( void ) {
+    // Define 3x3 symmetric matrices stored in row-major layout:
+    double A1[ 3*3 ] = {
+        1.0, 2.0, 3.0,
+        2.0, 1.0, 2.0,
+        3.0, 2.0, 1.0
+    };
+
+    double A2[ 3*3 ] = {
+        1.0, 2.0, 3.0,
+        2.0, 1.0, 2.0,
+        3.0, 2.0, 1.0
+    };
+
+    // Define `x` and `y` vectors:
+    const double x[ 3 ] = { 1.0, 2.0, 3.0 };
+    const double y[ 3 ] = { 1.0, 2.0, 3.0 };
+
+    // Specify the number of elements along each dimension of `A1` and `A2`:
+    const int N = 3;
+
+    // Perform the symmetric rank 2 operation `A = α*x*y^T + α*y*x^T + A`:
+    c_dsyr2( CblasColMajor, CblasUpper, N, 1.0, x, 1, y, 1, A1, N );
+
+    // Print the result:
+    for ( int i = 0; i < N*N; i++ ) {
+        printf( "A1[ %i ] = %lf\n", i, A1[ i ] );
+    }
+
+    // Perform the symmetric rank 2 operation `A = α*x*y^T + α*y*x^T + A` using alternative indexing semantics:
+    c_dsyr2_ndarray( CblasUpper, N, 1.0, x, 1, 0, y, 1, 0, A2, N, 1, 0 );
+
+    // Print the result:
+    for ( int i = 0; i < N*N; i++ ) {
+        printf( "A2[ %i ] = %lf\n", i, A2[ i ] );
+    }
+}
 ```
 
 </section>

lib/node_modules/@stdlib/blas/base/dsyr2/benchmark/benchmark.native.js

@@ -0,0 +1,110 @@
+/**
+* @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.
+*/
+
+'use strict';
+
+// MODULES //
+
+var resolve = require( 'path' ).resolve;
+var bench = require( '@stdlib/bench' );
+var isnan = require( '@stdlib/math/base/assert/is-nan' );
+var ones = require( '@stdlib/array/ones' );
+var pow = require( '@stdlib/math/base/special/pow' );
+var floor = require( '@stdlib/math/base/special/floor' );
+var tryRequire = require( '@stdlib/utils/try-require' );
+var pkg = require( './../package.json' ).name;
+
+
+// VARIABLES //
+
+var dsyr2 = tryRequire( resolve( __dirname, './../lib/dsyr2.native.js' ) );
+var opts = {
+	'skip': ( dsyr2 instanceof Error )
+};
+var options = {
+	'dtype': 'float64'
+};
+
+
+// FUNCTIONS //
+
+/**
+* Creates a benchmark function.
+*
+* @private
+* @param {PositiveInteger} N - number of elements along each dimension
+* @returns {Function} benchmark function
+*/
+function createBenchmark( N ) {
+	var x = ones( N, options.dtype );
+	var y = ones( N, options.dtype );
+	var A = ones( N*N, options.dtype );
+	return benchmark;
+
+	/**
+	* Benchmark function.
+	*
+	* @private
+	* @param {Benchmark} b - benchmark instance
+	*/
+	function benchmark( b ) {
+		var z;
+		var i;
+
+		b.tic();
+		for ( i = 0; i < b.iterations; i++ ) {
+			z = dsyr2( 'row-major', 'upper', N, 1.0, x, 1, y, 1, A, N );
+			if ( isnan( z[ i%z.length ] ) ) {
+				b.fail( 'should not return NaN' );
+			}
+		}
+		b.toc();
+		if ( isnan( z[ i%z.length ] ) ) {
+			b.fail( 'should not return NaN' );
+		}
+		b.pass( 'benchmark finished' );
+		b.end();
+	}
+}
+
+
+// MAIN //
+
+/**
+* Main execution sequence.
+*
+* @private
+*/
+function main() {
+	var min;
+	var max;
+	var N;
+	var f;
+	var i;
+
+	min = 1; // 10^min
+	max = 6; // 10^max
+
+	for ( i = min; i <= max; i++ ) {
+		N = floor( pow( pow( 10, i ), 1.0/2.0 ) );
+		f = createBenchmark( N );
+		bench( pkg+'::native:size='+(N*N), opts, f );
+	}
+}
+
+main();