Searched refs:multiplication (Results 1 – 25 of 332) sorted by relevance
12345678910>>...14
| /external/llvm/test/CodeGen/SystemZ/ |
| D | vec-mul-01.ll | 1 ; Test vector multiplication.
5 ; Test a v16i8 multiplication.
14 ; Test a v8i16 multiplication.
23 ; Test a v4i32 multiplication.
32 ; Test a v2i64 multiplication. There's no vector equivalent.
41 ; Test a v2f64 multiplication.
51 ; Test an f64 multiplication that uses vector registers.
|
| D | int-mul-06.ll | 1 ; Test 64-bit multiplication in which the second operand is constant.
5 ; Check multiplication by 2, which should use shifts.
14 ; Check multiplication by 3.
77 ; Check multiplication by -1, which is a negation.
86 ; Check multiplication by -2, which should use shifts.
96 ; Check multiplication by -3.
152 ; Check the next value down again, which cannot use constant multiplication
|
| D | int-mul-05.ll | 1 ; Test 32-bit multiplication in which the second operand is constant.
5 ; Check multiplication by 2, which should use shifts.
14 ; Check multiplication by 3.
77 ; Check multiplication by -1, which is a negation.
86 ; Check multiplication by -2, which should use shifts.
96 ; Check multiplication by -3.
|
| D | int-mul-07.ll | 8 ; Check zero-extended multiplication in which only the high part is used.
21 ; Check sign-extended multiplication in which only the high part is used.
34 ; Check zero-extended multiplication in which the result is split into
50 ; Check sign-extended multiplication in which the result is split into
|
| D | fp-mul-05.ll | 1 ; Test multiplication of two f128s.
5 ; There is no memory form of 128-bit multiplication.
|
| /external/swiftshader/third_party/llvm-7.0/llvm/test/CodeGen/SystemZ/ |
| D | vec-mul-01.ll | 1 ; Test vector multiplication.
5 ; Test a v16i8 multiplication.
14 ; Test a v8i16 multiplication.
23 ; Test a v4i32 multiplication.
32 ; Test a v2i64 multiplication. There's no vector equivalent.
41 ; Test a v2f64 multiplication.
51 ; Test an f64 multiplication that uses vector registers.
|
| D | int-mul-06.ll | 1 ; Test 64-bit multiplication in which the second operand is constant.
5 ; Check multiplication by 2, which should use shifts.
14 ; Check multiplication by 3.
77 ; Check multiplication by -1, which is a negation.
86 ; Check multiplication by -2, which should use shifts.
96 ; Check multiplication by -3.
152 ; Check the next value down again, which cannot use constant multiplication
|
| D | vec-mul-03.ll | 1 ; Test vector multiplication on z14.
5 ; Test a v4f32 multiplication.
15 ; Test an f32 multiplication that uses vector registers.
|
| D | int-mul-05.ll | 1 ; Test 32-bit multiplication in which the second operand is constant.
5 ; Check multiplication by 2, which should use shifts.
14 ; Check multiplication by 3.
77 ; Check multiplication by -1, which is a negation.
86 ; Check multiplication by -2, which should use shifts.
96 ; Check multiplication by -3.
|
| D | int-mul-07.ll | 8 ; Check zero-extended multiplication in which only the high part is used.
21 ; Check sign-extended multiplication in which only the high part is used.
34 ; Check zero-extended multiplication in which the result is split into
50 ; Check sign-extended multiplication in which the result is split into
|
| D | fp-mul-05.ll | 1 ; Test multiplication of two f128s.
5 ; There is no memory form of 128-bit multiplication.
|
| /external/tensorflow/tensorflow/core/api_def/base_api/ |
| D | api_def_TridiagonalMatMul.pbtxt | 8 tri-diagonal matrices to the left of multiplication. Last element is ingored.
15 matrices to the left of multiplication.
22 matrices to the left of multiplication. First element is ingored.
29 multiplication.
|
| D | api_def_SparseMatrixSparseMatMul.pbtxt | 34 Performs a matrix multiplication of a sparse matrix `a` with a sparse matrix
83 # Compute the CSR SparseMatrix matrix multiplication
106 transpose_a: If True, `a` transposed before multiplication.
107 transpose_b: If True, `b` transposed before multiplication.
108 adjoint_a: If True, `a` adjointed before multiplication.
109 adjoint_b: If True, `b` adjointed before multiplication.
|
| D | api_def_QuantizedMatMul.pbtxt | 54 If true, `a` is transposed before multiplication.
60 If true, `b` is transposed before multiplication.
70 summary: "Perform a quantized matrix multiplication of `a` by the matrix `b`."
|
| D | api_def_MatMul.pbtxt | 6 If true, "a" is transposed before multiplication.
12 If true, "b" is transposed before multiplication.
|
| D | api_def_QuantizedMatMulWithBiasAndRelu.pbtxt | 61 description: "If true, `a` is transposed before multiplication."
65 description: "If true, `b` is transposed before multiplication."
74 Perform a quantized matrix multiplication of `a` by the matrix `b` with bias
|
| /external/llvm/test/CodeGen/X86/ |
| D | fdiv.ll | 5 ; Exact division by a constant converted to multiplication.
15 ; Inexact division by a constant converted to multiplication.
25 ; No conversion to multiplication if too funky.
36 ; Don't generate multiplication by a denormal.
46 ; Don't generate multiplication by a denormal.
|
| /external/swiftshader/third_party/llvm-7.0/llvm/test/CodeGen/X86/ |
| D | fdiv.ll | 5 ; Exact division by a constant converted to multiplication.
15 ; Inexact division by a constant converted to multiplication.
25 ; No conversion to multiplication if too funky.
36 ; Don't generate multiplication by a denormal.
46 ; Don't generate multiplication by a denormal.
|
| /external/FXdiv/ |
| D | METADATA | 3 "Header-only library for division via fixed-point multiplication by inverse "
6 "multiplication. FXdiv implements an algorithm to replace an integer "
7 "division with a multiplication and two shifts. This algorithm improves "
|
| D | README.md | 7 Header-only library for division via fixed-point multiplication by inverse
9 … several times slower than multiplication. FXdiv implements an algorithm to replace an integer div…
60 - Granlund, Torbjörn, and Peter L. Montgomery. "Division by invariant integers using multiplication…
|
| /external/gemmlowp/meta/ |
| D | README | 5 - providing a new matrix multiplication kernel.
14 The multiplication kernel - the most inner loop of the matrix multiplication,
18 specialized 8bit multiplication, aggregating to vector aggregators and then
47 some python scripts. 9 Versions of the multiplication kernel were prepared.
52 Using those multiplication and packing/unpacking primitives 144 gemm function
|
| /external/tensorflow/tensorflow/lite/experimental/ruy/ |
| D | README.md | 1 # The ruy matrix multiplication library
5 ruy is a matrix multiplication library. Its focus is to cover the matrix
6 multiplication needs of neural network inference engines. Its initial user has
|
| /external/boringssl/src/third_party/wycheproof_testvectors/ |
| D | x25519_test.txt | 37 # If a point multiplication is performed then it is important that the result is
49 # If a point multiplication is performed then it is important that the result is
61 # If a point multiplication is performed then it is important that the result is
73 # If a point multiplication is performed then it is important that the result is
85 # If a point multiplication is performed then it is important that the result is
117 # If a point multiplication is performed then it is important that the result is
129 # If a point multiplication is performed then it is important that the result is
141 # If a point multiplication is performed then it is important that the result is
153 # If a point multiplication is performed then it is important that the result is
165 # If a point multiplication is performed then it is important that the result is
[all …]
|
| /external/gemmlowp/doc/ |
| D | quantization.md | 136 ## Quantizing a matrix multiplication
140 rewriting a multiplication of matrices of real numbers, by the equivalent
141 multiplication of matrices of quantized values.
182 matrix multiplication should do, i.e. how to compute `result_quantized_value`.
203 "quantized matrix multiplication" should actually compute, in order to be able
206 ## Implementation of quantized matrix multiplication
208 Having obtained the mathematical form (5) of quantized matrix multiplication, we
263 So how do we implement the multiplication of a int32 value by a positive real
325 multiplication with a int32 multiplier, which may overflow. By contrast, in
327 integer multiplication becomes a fixed-point multiplication and cannot
[all …]
|
| D | output.md | 15 multiplication feeds into other operations applied to its result, without
18 activation. One then wants to feed the result of the matrix multiplication
51 quantized matrix multiplication with a sounds quantization paradigm, is here:
|
12345678910>>...14