Press n or j to go to the next uncovered block, b, p or k for the previous block.
| 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 | 2x 2x 2x 2x 2x 2x 2x 2x 2x 2x 2x 2x 2x 2x 2x 2x 2x 2x 2x 2x 2x 2x 2x 2x 2x 2x 2x 2x 2x 2x 2x 2x 2x 2x 2x 2x 2x 2x 2x 2x 2x 2x 2x 2x 2x 2x 2x 2x 2x 2x 2x 2x 2x 2x 2x 2x 2x 2x 2x 2x 2x 2x 2x 2x 2x 2x 2x 2x 2x 2x 2x 2x 2x 2x 76x 76x 76x 76x 76x 76x 76x 76x 76x 76x 76x 76x 76x 76x 76x 76x 76x 4x 4x 72x 76x 4x 4x 68x 76x 4x 4x 64x 76x 4x 4x 76x 3x 3x 76x 3x 3x 76x 3x 3x 51x 51x 51x 76x 28x 76x 35x 76x 16x 16x 51x 76x 28x 76x 35x 76x 16x 16x 76x 5x 5x 76x 6x 6x 76x 24x 76x 16x 16x 76x 6x 6x 76x 16x 16x 16x 16x 16x 16x 76x 18x 18x 18x 18x 18x 18x 18x 34x 76x 2x 2x 2x 2x 2x | /**
* @license Apache-2.0
*
* Copyright (c) 2026 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 max = require( '@stdlib/math/base/special/fast/max' );
var isLayout = require( '@stdlib/blas/base/assert/is-layout' );
var resolveTrans = require( '@stdlib/blas/base/transpose-operation-resolve-str' );
var resolveUplo = require( '@stdlib/blas/base/matrix-triangle-resolve-str' );
var isRowMajor = require( '@stdlib/ndarray/base/assert/is-row-major-string' );
var isColumnMajor = require( '@stdlib/ndarray/base/assert/is-column-major-string' );
var format = require( '@stdlib/string/format' );
var base = require( './base.js' );
// MAIN //
/**
* Performs the matrix-matrix operation `uplo(C) = uplo(α*op(A)*op(B) + β*C)` where `op(X)` is either `op(X) = X` or `op(X) = X^T`, `α` and `β` are scalars, `A`, `B`, and `C` are matrices, with `op(A)` an `M` by `K` matrix, `op(B)` a `K` by `N` matrix, and `C` an `M` by `N` matrix, and only the elements in the upper or lower triangle of `C` are updated.
*
* @param {string} order - storage layout
* @param {(integer|string)} uplo - specifies whether to update the upper or lower triangular part of `C`
* @param {(integer|string)} transA - specifies whether `A` should be transposed, conjugate-transposed, or not transposed
* @param {(integer|string)} transB - specifies whether `B` should be transposed, conjugate-transposed, or not transposed
* @param {NonNegativeInteger} M - number of rows in the matrix `op(A)` and in the matrix `C`
* @param {NonNegativeInteger} N - number of columns in the matrix `op(B)` and in the matrix `C`
* @param {NonNegativeInteger} K - number of columns in the matrix `op(A)` and number of rows in the matrix `op(B)`
* @param {number} alpha - scalar constant
* @param {Float32Array} A - first matrix
* @param {PositiveInteger} LDA - stride of the first dimension of `A` (a.k.a., leading dimension of the matrix `A`)
* @param {Float32Array} B - second matrix
* @param {PositiveInteger} LDB - stride of the first dimension of `B` (a.k.a., leading dimension of the matrix `B`)
* @param {number} beta - scalar constant
* @param {Float32Array} C - third matrix
* @param {PositiveInteger} LDC - stride of the first dimension of `C` (a.k.a., leading dimension of the matrix `C`)
* @throws {TypeError} first argument must be a valid order
* @throws {TypeError} second argument must specify whether to reference the lower or upper triangular matrix
* @throws {TypeError} third argument must be a valid transpose operation
* @throws {TypeError} fourth argument must be a valid transpose operation
* @throws {RangeError} fifth argument must be a nonnegative integer
* @throws {RangeError} sixth argument must be a nonnegative integer
* @throws {RangeError} seventh argument must be a nonnegative integer
* @throws {RangeError} tenth argument must be greater than or equal to max(1,M) when `A` is not transposed and max(1,K) otherwise
* @throws {RangeError} twelfth argument must be greater than or equal to max(1,K) when `B` is not transposed and max(1,N) otherwise
* @throws {RangeError} fifteenth argument must be greater than or equal to max(1,M)
* @returns {Float32Array} `C`
*
* @example
* var Float32Array = require( '@stdlib/array/float32' );
*
* var A = new Float32Array( [ 1.0, 2.0, 3.0, 4.0 ] );
* var B = new Float32Array( [ 1.0, 1.0, 0.0, 1.0 ] );
* var C = new Float32Array( [ 1.0, 2.0, 3.0, 4.0 ] );
*
* sgemmtr( 'row-major', 'upper', 'no-transpose', 'no-transpose', 2, 2, 2, 1.0, A, 2, B, 2, 1.0, C, 2 );
* // C => <Float32Array>[ 2.0, 5.0, 3.0, 11.0 ]
*/
function sgemmtr( order, uplo, transA, transB, M, N, K, alpha, A, LDA, B, LDB, beta, C, LDC ) { // eslint-disable-line max-params, max-len
var nrowsa;
var nrowsb;
var valc;
var isrm;
var iscm;
var sa1;
var sa2;
var sb1;
var sb2;
var sc1;
var sc2;
var ta;
var tb;
var u;
if ( !isLayout( order ) ) {
throw new TypeError( format( 'invalid argument. First argument must be a valid order. Value: `%s`.', order ) );
}
u = resolveUplo( uplo );
if ( u === null ) {
throw new TypeError( format( 'invalid argument. Second argument must specify whether to reference the lower or upper triangular matrix. Value: `%s`.', uplo ) );
}
ta = resolveTrans( transA );
if ( ta === null ) {
throw new TypeError( format( 'invalid argument. Third argument must be a valid transpose operation. Value: `%s`.', transA ) );
}
tb = resolveTrans( transB );
if ( tb === null ) {
throw new TypeError( format( 'invalid argument. Fourth argument must be a valid transpose operation. Value: `%s`.', transB ) );
}
if ( M < 0 ) {
throw new RangeError( format( 'invalid argument. Fifth argument must be a nonnegative integer. Value: `%d`.', M ) );
}
if ( N < 0 ) {
throw new RangeError( format( 'invalid argument. Sixth argument must be a nonnegative integer. Value: `%d`.', N ) );
}
if ( K < 0 ) {
throw new RangeError( format( 'invalid argument. Seventh argument must be a nonnegative integer. Value: `%d`.', K ) );
}
isrm = isRowMajor( order );
iscm = isColumnMajor( order );
if (
( isrm && transA === 'no-transpose' ) ||
( iscm && transA === 'transpose' )
) {
nrowsa = K;
} else {
nrowsa = M;
}
if (
( isrm && transB === 'no-transpose' ) ||
( iscm && transB === 'transpose' )
) {
nrowsb = N;
} else {
nrowsb = K;
}
if ( LDA < max( 1, nrowsa ) ) {
throw new RangeError( format( 'invalid argument. Tenth argument must be greater than or equal to max(1,%d). Value: `%d`.', nrowsa, LDA ) );
}
if ( LDB < max( 1, nrowsb ) ) {
throw new RangeError( format( 'invalid argument. Twelfth argument must be greater than or equal to max(1,%d). Value: `%d`.', nrowsb, LDB ) );
}
if ( isrm ) {
valc = N;
} else {
valc = M;
}
if ( LDC < max( 1, valc ) ) {
throw new RangeError( format( 'invalid argument. Fifteenth argument must be greater than or equal to max(1,%d). Value: `%d`.', valc, LDC ) );
}
if ( iscm ) {
sa1 = 1;
sa2 = LDA;
sb1 = 1;
sb2 = LDB;
sc1 = 1;
sc2 = LDC;
} else { // order === 'row-major'
sa1 = LDA;
sa2 = 1;
sb1 = LDB;
sb2 = 1;
sc1 = LDC;
sc2 = 1;
}
return base( u, ta, tb, M, N, K, alpha, A, sa1, sa2, 0, B, sb1, sb2, 0, beta, C, sc1, sc2, 0 ); // eslint-disable-line max-len
}
// EXPORTS //
module.exports = sgemmtr;
|