feat: add C ndarray API and refactor blas/ext/base/scusumkbn

PR-URL: #4782
Reviewed-by: Athan Reines <[email protected]>

Author: headlessNode

lib/node_modules/@stdlib/blas/ext/base/scusumkbn/README.md

@@ -61,11 +61,11 @@ The function has the following parameters:
 -   **N**: number of indexed elements.
 -   **sum**: initial sum.
 -   **x**: input [`Float32Array`][@stdlib/array/float32].
--   **strideX**: index increment for `x`.
+-   **strideX**: stride length for `x`.
 -   **y**: output [`Float32Array`][@stdlib/array/float32].
--   **strideY**: index increment for `y`.
+-   **strideY**: stride length for `y`.
 
-The `N` and stride parameters determine which elements in the strided arrays are accessed at runtime. For example, to compute the cumulative sum of every other element in `x`,
+The `N` and stride parameters determine which elements in the strided arrays are accessed at runtime. For example, to compute the cumulative sum of every other element:
 
 ```javascript
 var Float32Array = require( '@stdlib/array/float32' );
@@ -114,7 +114,7 @@ The function has the following additional parameters:
 -   **offsetX**: starting index for `x`.
 -   **offsetY**: starting index for `y`.
 
-While [`typed array`][mdn-typed-array] views mandate a view offset based on the underlying `buffer`, offset parameters support indexing semantics based on a starting indices. For example, to calculate the cumulative sum of every other value in `x` starting from the second value and to store in the last `N` elements of `y` starting from the last element
+While [`typed array`][mdn-typed-array] views mandate a view offset based on the underlying buffer, the offset parameters support indexing semantics based on starting indices. For example, to calculate the cumulative sum of every other element starting from the second element and to store in the last `N` elements of `y` starting from the last element:
 
 ```javascript
 var Float32Array = require( '@stdlib/array/float32' );
@@ -147,14 +147,17 @@ scusumkbn.ndarray( 4, 0.0, x, 2, 1, y, -1, y.length-1 );
 <!-- eslint no-undef: "error" -->
 
 ```javascript
-var discreteUniform = require( '@stdlib/random/base/discrete-uniform' ).factory;
-var filledarrayBy = require( '@stdlib/array/filled-by' );
+var discreteUniform = require( '@stdlib/random/array/discrete-uniform' );
 var scusumkbn = require( '@stdlib/blas/ext/base/scusumkbn' );
 
-var x = filledarrayBy( 10, 'float32', discreteUniform( 0, 100 ) );
+var x = discreteUniform( 10, -100, 100, {
+    'dtype': 'float32'
+});
 console.log( x );
 
-var y = filledarrayBy( x.length, 'float32', discreteUniform( 0, 10 ) );
+var y = discreteUniform( 10, -100, 100, {
+    'dtype': 'float32'
+});
 console.log( y );
 
 scusumkbn( x.length, 0.0, x, 1, y, -1 );
@@ -165,8 +168,138 @@ console.log( y );
 
 <!-- /.examples -->
 
+<!-- C interface documentation. -->
+
 * * *
 
+<section class="c">
+
+## C APIs
+
+<!-- Section to include introductory text. Make sure to keep an empty line after the intro `section` element and another before the `/section` close. -->
+
+<section class="intro">
+
+</section>
+
+<!-- /.intro -->
+
+<!-- C usage documentation. -->
+
+<section class="usage">
+
+### Usage
+
+```c
+#include "stdlib/blas/ext/base/scusumkbn.h"
+```
+
+#### stdlib_strided_scusumkbn( N, sum, \*X, strideX, \*Y, strideY )
+
+Computes the cumulative sum of single-precision floating-point strided array elements using an improved Kahan–Babuška algorithm.
+
+```c
+const float x[] = { 1.0f, 2.0f, 3.0f, 4.0f };
+float y[] = { 0.0f, 0.0f, 0.0f, 0.0f };
+
+stdlib_strided_scusumkbn( 4, 0.0f, x, 1, y, 1 );
+```
+
+The function accepts the following arguments:
+
+-   **N**: `[in] CBLAS_INT` number of indexed elements.
+-   **sum**: `[in] float` initial sum.
+-   **X**: `[in] float*` input array.
+-   **strideX**: `[in] CBLAS_INT` stride length for `X`.
+-   **Y**: `[out] float*` output array.
+-   **strideY**: `[in] CBLAS_INT` stride length for `Y`.
+
+```c
+void stdlib_strided_scusumkbn( const CBLAS_INT N, const float sum, const float *X, const CBLAS_INT strideX, float *Y, const CBLAS_INT strideY );
+```
+
+<!-- lint disable maximum-heading-length -->
+
+#### stdlib_strided_scusumkbn_ndarray( N, sum, \*X, strideX, offsetX, \*Y, strideY, offsetY )
+
+<!-- lint enable maximum-heading-length -->
+
+Computes the cumulative sum of single-precision floating-point strided array elements using an improved Kahan–Babuška algorithm and alternative indexing semantics.
+
+```c
+const float x[] = { 1.0f, 2.0f, 3.0f, 4.0f };
+float y[] = { 0.0f, 0.0f, 0.0f, 0.0f };
+
+stdlib_strided_scusumkbn_ndarray( 4, 0.0f, x, 1, 0, y, 1, 0 );
+```
+
+The function accepts the following arguments:
+
+-   **N**: `[in] CBLAS_INT` number of indexed elements.
+-   **sum**: `[in] float` initial sum.
+-   **X**: `[in] float*` input array.
+-   **strideX**: `[in] CBLAS_INT` stride length for `X`.
+-   **offsetX**: `[in] CBLAS_INT` starting index for `X`.
+-   **Y**: `[out] float*` output array.
+-   **strideY**: `[in] CBLAS_INT` stride length for `Y`.
+-   **offsetY**: `[in] CBLAS_INT` starting index for `Y`.
+
+```c
+void stdlib_strided_scusumkbn_ndarray( const CBLAS_INT N, const float sum, const float *X, const CBLAS_INT strideX, const CBLAS_INT offsetX, float *Y, const CBLAS_INT strideY, const CBLAS_INT offsetY );
+```
+
+</section>
+
+<!-- /.usage -->
+
+<!-- C API usage notes. Make sure to keep an empty line after the `section` element and another before the `/section` close. -->
+
+<section class="notes">
+
+</section>
+
+<!-- /.notes -->
+
+<!-- C API usage examples. -->
+
+<section class="examples">
+
+### Examples
+
+```c
+#include "stdlib/blas/ext/base/scusumkbn.h"
+#include <stdio.h>
+
+int main( void ) {
+    // Create strided arrays:
+    const float x[] = { 1.0f, 2.0f, 3.0f, 4.0f, 5.0f, 6.0f, 7.0f, 8.0f };
+    float y[] = { 0.0f, 0.0f, 0.0f, 0.0f, 0.0f, 0.0f, 0.0f, 0.0f };
+
+    // Specify the number of elements:
+    const int N = 4;
+
+    // Specify stride lengths:
+    const int strideX = 2;
+    const int strideY = -2;
+
+    // Compute the cumulative sum:
+    stdlib_strided_scusumkbn( N, 0.0f, x, strideX, y, strideY );
+
+    // Print the result:
+    for ( int i = 0; i < 8; i++ ) {
+        printf( "y[ %d ] = %f\n", i, y[ i ] );
+    }
+}
+```
+
+</section>
+
+<!-- /.examples -->
+
+</section>
+
+<!-- /.c -->
+
 <section class="references">
 
 ## References

lib/node_modules/@stdlib/blas/ext/base/scusumkbn/benchmark/benchmark.js

@@ -21,8 +21,7 @@
 // MODULES //
 
 var bench = require( '@stdlib/bench' );
-var uniform = require( '@stdlib/random/base/uniform' ).factory;
-var filledarrayBy = require( '@stdlib/array/filled-by' );
+var uniform = require( '@stdlib/random/array/uniform' );
 var isnanf = require( '@stdlib/math/base/assert/is-nanf' );
 var pow = require( '@stdlib/math/base/special/pow' );
 var pkg = require( './../package.json' ).name;
@@ -31,7 +30,9 @@ var scusumkbn = require( './../lib/scusumkbn.js' );
 
 // VARIABLES //
 
-var rand = uniform( -100.0, 100.0 );
+var options = {
+	'dtype': 'float32'
+};
 
 
 // FUNCTIONS //
@@ -44,8 +45,8 @@ var rand = uniform( -100.0, 100.0 );
 * @returns {Function} benchmark function
 */
 function createBenchmark( len ) {
-	var x = filledarrayBy( len, 'float32', rand );
-	var y = filledarrayBy( len, 'float32', rand );
+	var x = uniform( len, -100, 100, options );
+	var y = uniform( len, -100, 100, options );
 	return benchmark;
 
 	function benchmark( b ) {

lib/node_modules/@stdlib/blas/ext/base/scusumkbn/benchmark/benchmark.native.js

@@ -22,8 +22,7 @@
 
 var resolve = require( 'path' ).resolve;
 var bench = require( '@stdlib/bench' );
-var uniform = require( '@stdlib/random/base/uniform' ).factory;
-var filledarrayBy = require( '@stdlib/array/filled-by' );
+var uniform = require( '@stdlib/random/array/uniform' );
 var isnanf = require( '@stdlib/math/base/assert/is-nanf' );
 var pow = require( '@stdlib/math/base/special/pow' );
 var tryRequire = require( '@stdlib/utils/try-require' );
@@ -36,7 +35,9 @@ var scusumkbn = tryRequire( resolve( __dirname, './../lib/scusumkbn.native.js' )
 var opts = {
 	'skip': ( scusumkbn instanceof Error )
 };
-var rand = uniform( -100.0, 100.0 );
+var options = {
+	'dtype': 'float32'
+};
 
 
 // FUNCTIONS //
@@ -49,8 +50,8 @@ var rand = uniform( -100.0, 100.0 );
 * @returns {Function} benchmark function
 */
 function createBenchmark( len ) {
-	var x = filledarrayBy( len, 'float32', rand );
-	var y = filledarrayBy( len, 'float32', rand );
+	var x = uniform( len, -100, 100, options );
+	var y = uniform( len, -100, 100, options );
 	return benchmark;
 
 	function benchmark( b ) {

lib/node_modules/@stdlib/blas/ext/base/scusumkbn/benchmark/benchmark.ndarray.js

@@ -21,8 +21,7 @@
 // MODULES //
 
 var bench = require( '@stdlib/bench' );
-var uniform = require( '@stdlib/random/base/uniform' ).factory;
-var filledarrayBy = require( '@stdlib/array/filled-by' );
+var uniform = require( '@stdlib/random/array/uniform' );
 var isnanf = require( '@stdlib/math/base/assert/is-nanf' );
 var pow = require( '@stdlib/math/base/special/pow' );
 var pkg = require( './../package.json' ).name;
@@ -31,7 +30,9 @@ var scusumkbn = require( './../lib/ndarray.js' );
 
 // VARIABLES //
 
-var rand = uniform( -100.0, 100.0 );
+var options = {
+	'dtype': 'float32'
+};
 
 
 // FUNCTIONS //
@@ -44,8 +45,8 @@ var rand = uniform( -100.0, 100.0 );
 * @returns {Function} benchmark function
 */
 function createBenchmark( len ) {
-	var x = filledarrayBy( len, 'float32', rand );
-	var y = filledarrayBy( len, 'float32', rand );
+	var x = uniform( len, -100, 100, options );
+	var y = uniform( len, -100, 100, options );
 	return benchmark;
 
 	function benchmark( b ) {

lib/node_modules/@stdlib/blas/ext/base/scusumkbn/benchmark/benchmark.ndarray.native.js

@@ -22,8 +22,7 @@
 
 var resolve = require( 'path' ).resolve;
 var bench = require( '@stdlib/bench' );
-var uniform = require( '@stdlib/random/base/uniform' ).factory;
-var filledarrayBy = require( '@stdlib/array/filled-by' );
+var uniform = require( '@stdlib/random/array/uniform' );
 var isnanf = require( '@stdlib/math/base/assert/is-nanf' );
 var pow = require( '@stdlib/math/base/special/pow' );
 var tryRequire = require( '@stdlib/utils/try-require' );
@@ -36,7 +35,9 @@ var scusumkbn = tryRequire( resolve( __dirname, './../lib/ndarray.native.js' ) )
 var opts = {
 	'skip': ( scusumkbn instanceof Error )
 };
-var rand = uniform( -100.0, 100.0 );
+var options = {
+	'dtype': 'float32'
+};
 
 
 // FUNCTIONS //
@@ -49,8 +50,8 @@ var rand = uniform( -100.0, 100.0 );
 * @returns {Function} benchmark function
 */
 function createBenchmark( len ) {
-	var x = filledarrayBy( len, 'float32', rand );
-	var y = filledarrayBy( len, 'float32', rand );
+	var x = uniform( len, -100, 100, options );
+	var y = uniform( len, -100, 100, options );
 	return benchmark;
 
 	function benchmark( b ) {

lib/node_modules/@stdlib/blas/ext/base/scusumkbn/benchmark/c/benchmark.length.c

@@ -94,7 +94,7 @@ static float rand_float( void ) {
 * @param len          array length
 * @return             elapsed time in seconds
 */
-static double benchmark( int iterations, int len ) {
+static double benchmark1( int iterations, int len ) {
 	double elapsed;
 	float x[ len ];
 	float y[ len ];
@@ -121,6 +121,40 @@ static double benchmark( int iterations, int len ) {
 	return elapsed;
 }
 
+/**
+* Runs a benchmark.
+*
+* @param iterations   number of iterations
+* @param len          array length
+* @return             elapsed time in seconds
+*/
+static double benchmark2( int iterations, int len ) {
+	double elapsed;
+	float x[ len ];
+	float y[ len ];
+	double t;
+	int i;
+
+	for ( i = 0; i < len; i++ ) {
+		x[ i ] = ( rand_float() * 20000.0f ) - 10000.0f;
+		y[ i ] = 0.0f;
+	}
+	t = tic();
+	for ( i = 0; i < iterations; i++ ) {
+		x[ 0 ] += 1.0f;
+		stdlib_strided_scusumkbn_ndarray( len, 0.0f, x, 1, 0, y, 1, 0 );
+		if ( y[ 0 ] != y[ 0 ] ) {
+			printf( "should not return NaN\n" );
+			break;
+		}
+	}
+	elapsed = tic() - t;
+	if ( y[ len-1 ] != y[ len-1 ] ) {
+		printf( "should not return NaN\n" );
+	}
+	return elapsed;
+}
+
 /**
 * Main execution sequence.
 */
@@ -143,7 +177,18 @@ int main( void ) {
 		for ( j = 0; j < REPEATS; j++ ) {
 			count += 1;
 			printf( "# c::%s:len=%d\n", NAME, len );
-			elapsed = benchmark( iter, len );
+			elapsed = benchmark1( iter, len );
+			print_results( iter, elapsed );
+			printf( "ok %d benchmark finished\n", count );
+		}
+	}
+	for ( i = MIN; i <= MAX; i++ ) {
+		len = pow( 10, i );
+		iter = ITERATIONS / pow( 10, i-1 );
+		for ( j = 0; j < REPEATS; j++ ) {
+			count += 1;
+			printf( "# c::%s:ndarray:len=%d\n", NAME, len );
+			elapsed = benchmark2( iter, len );
 			print_results( iter, elapsed );
 			printf( "ok %d benchmark finished\n", count );
 		}