---
include: hipblaslt_common.yaml
include: known_bugs.yaml
include: matmul_common.yaml

Definitions:
  - &alpha_beta_range
    - { alpha:  5, beta:  0 }
    - { alpha:  0, beta:  3 }
    - { alpha:  1, beta:  3 }
    - { alpha:  1, beta:  1 }

  - &alpha_beta_range_small
    - { alpha: 2, alphai: 2, beta: -1.0, betai: 2.0 }

  - &transA_transB_range
    - { transA: N, transB: N }
    - { transA: N, transB: T }
    - { transA: T, transB: N }
    - { transA: T, transB: T }

Tests:
- name: matmul_smoke
  category: smoke
  function:
    matmul: *real_precisions
  matrix_size: *smoke_matrix_size_range
  transA_transB: *transA_transB_range
  alpha_beta: *alpha_beta_range

- name: matmul_bias_relu_SAV_smoke
  category: smoke
  function:
    matmul: *real_precisions
  matrix_size: *smoke_matrix_size_range
  transA_transB: *transA_transB_range
  alpha: 1
  beta: [ 0.0, 2.0 ]
  scaleAlpha_vector: [0, 1]
  activation_type: relu
  bias_vector: [0, 1]
  unit_check: 1

- name: matmul_bias_sigmoid_smoke
  category: smoke
  function:
    matmul: *real_precisions
  matrix_size: *smoke_matrix_size_range
  transA_transB: *transA_transB_range
  alpha: 1
  beta: [ 0.0, 2.0 ]
  activation_type: sigmoid
  bias_vector: [0, 1]
  unit_check: 1

- name: matmul_bias_gelu_smoke
  category: smoke
  function:
    matmul: *real_precisions
  matrix_size: *smoke_matrix_size_range
  transA_transB: *transA_transB_range
  alpha: 1
  beta: [ 0.0, 2.0 ]
  activation_type: gelu
  bias_vector: [0, 1]
  unit_check: 0
  norm_check: 1

- name: matmul_bias_only_smoke
  category: smoke
  function:
    matmul: *real_precisions
  matrix_size: *smoke_matrix_size_range
  transA_transB: *transA_transB_range
  alpha: 1
  beta: [ 0.0, 2.0 ]
  bias_vector: 1
  unit_check: 1

- name: matmul_bias_type_smoke
  category: smoke
  function:
    matmul: *real_precisions_2b
  matrix_size: *smoke_matrix_size_range
  transA_transB: *transA_transB_range
  alpha: 1
  beta: [ 0.0, 2.0 ]
  bias_vector: 1
  bias_type: [default, f32_r]
  unit_check: 1

- name: matmul_f8_bf8_fnuz_dst_fp16_smoke
  category: smoke
  function:
    matmul: *real_precisions_1b_fnuz_dst_f16
  matrix_size:
    - { M:  128,  N:  128,  K:  128  }
  transA_transB: *transA_transB_range
  alpha: 1
  beta: [ 0.0, 2.0 ]
  scaleA: [ 0, 1]
  scaleB: [ 0, 1]
  bias_vector: [0, 1]
  bias_type: f16_r
  unit_check: 1
  gpu_arch: '942'

- name: matmul_f8_bf8_fnuz_dst_bf16_smoke
  category: smoke
  function:
    matmul: *real_precisions_1b_fnuz_dst_bf16
  matrix_size:
    - { M:  128,  N:  128,  K:  128  }
  transA_transB: *transA_transB_range
  alpha: 1
  beta: [ 0.0, 2.0 ]
  scaleA: [ 0, 1]
  scaleB: [ 0, 1]
  bias_vector: [0, 1]
  bias_type: bf16_r
  unit_check: 1
  gpu_arch: '942'

### GFX12 FP8
- name: matmul_f8_bf8_dst_fp32_gfx12_smoke
  category: smoke
  function:
    matmul: *real_precisions_1b_dst_f32
  matrix_size:
    - { M:  128,  N:  128,  K:  128  }
  transA_transB: *transA_transB_range
  alpha: 1
  beta: [ 0.0, 2.0 ]
  scaleA: [ 0, 1]
  scaleB: [ 0, 1]
  bias_vector: [0, 1]
  bias_type: bf16_r
  unit_check: 0
  norm_check: 1
  gpu_arch: '12(0[0-1]|50)'

- name: matmul_f8_bf8_dst_fp16_gfx12_smoke
  category: smoke
  function:
    matmul: *real_precisions_1b_dst_f16
  matrix_size:
    - { M:  128,  N:  128,  K:  128  }
  transA_transB: *transA_transB_range
  alpha: 1
  beta: [ 0.0, 2.0 ]
  scaleA: [ 0, 1]
  scaleB: [ 0, 1]
  bias_vector: [0, 1]
  bias_type: f16_r
  unit_check: 1
  gpu_arch: '(120[0-1]|1250)'

- name: matmul_f8_bf8_dst_bf16_gfx12_smoke
  category: smoke
  function:
    matmul: *real_precisions_1b_dst_bf16
  matrix_size:
    - { M:  128,  N:  128,  K:  128  }
  transA_transB: *transA_transB_range
  alpha: 1
  beta: [ 0.0, 2.0 ]
  scaleA: [ 0, 1]
  scaleB: [ 0, 1]
  bias_vector: [0, 1]
  bias_type: bf16_r
  unit_check: 1
  gpu_arch: '(120[0-1]|1250)'

- name: matmul_real_1b_dst_f8_SCDInt1_gfx12_smoke
  category: smoke
  function:
    matmul: *real_precisions_1b_dst_f8
  matrix_size:
    - { M:  128,  N:  128,  K:  128  }
  transA_transB: *transA_transB_range
  alpha: 1
  beta: [ 0.0, 2.0 ]
  scaleA: [0, 1]
  scaleB: [0, 1]
  scaleC: [0] # will use default value '1'
  scaleD: [0] # will use default value '1'
  bias_vector: [0, 1]
  bias_type: f16_r
  unit_check: 1
  gpu_arch: '(120[0-1]|1250)'

- name: matmul_real_1b_dst_f8_SCDNotInt_gfx12_smoke
  category: smoke
  function:
    matmul: *real_precisions_1b_dst_f8
  matrix_size:
    - { M:  128,  N:  128,  K:  128  }
  transA_transB: *transA_transB_range
  alpha: 1
  beta: [ 0.0, 2.0 ]
  scaleA: [0, 1]
  scaleB: [0, 1]
  scaleC: [1] # will use random value from 0.0 to 1.0
  scaleD: [1] # will use random value from 0.0 to 1.0
  bias_vector: [0, 1]
  bias_type: f16_r
  unit_check: 1
  gpu_arch: '(120[0-1]|1250)'

- name: matmul_one_integer_precisions_i8_gfx12_smoke
  category: smoke
  function:
    matmul: *integer_precisions_i8
  matrix_size: *one_matrix_size_range
  transA_transB: *transA_transB_range
  alpha: [ 0.0, 1.0 ]
  beta: [ 0.0, 2.0 ]
  gpu_arch: '12(0[0-1]|50)'

- name: matmul_real_1b_fnuz_dst_f8_fnuz_SCDInt1_smoke
  category: smoke
  function:
    matmul: *real_precisions_1b_fnuz_dst_f8_fnuz
  matrix_size:
    - { M:  128,  N:  128,  K:  128  }
  transA_transB: *transA_transB_range
  alpha: 1
  beta: [ 0.0, 2.0 ]
  scaleA: [0, 1]
  scaleB: [0, 1]
  scaleC: [0] # will use default value '1'
  scaleD: [0] # will use default value '1'
  bias_vector: [0, 1]
  bias_type: f16_r
  unit_check: 1
  gpu_arch: '942'

- name: matmul_real_1b_fnuz_dst_1b_fnuz_smallsize_smoke
  category: smoke
  function:
    matmul: *real_precisions_1b_fnuz_dst_1b_fnuz
  matrix_size:
    - { M:  128,  N:  128,  K:  128  }
  transA_transB: *transA_transB_range
  alpha: 1
  beta: [ 0.0, 2.0 ]
  scaleA: [0, 1]
  scaleB: [0, 1]
  scaleC: [0]
  scaleD: [0]
  bias_vector: [0, 1]
  bias_type: f32_r
  unit_check: 1
  gpu_arch: '942'

#TODO: extend to all f8 transpose and datatype if necessary
- name: matmul_real_f8_fnuz_dst_fp32_smoke
  category: smoke
  function:
    matmul: *f8_fnuz_precision_dst_fp32
  matrix_size:
    - { M:  128,  N:  128,  K:  128  }
  transA_transB: *transA_transB_range
  alpha: 1
  beta: [ 0.0, 2.0 ]
  scaleA: [0, 1]
  scaleB: [0, 1]
  scaleC: [0]
  scaleD: [0]
  bias_vector: [0, 1]
  bias_type: f32_r
  unit_check: 1
  gpu_arch: '942'

- name: matmul_real_1b_fnuz_dst_f8_fnuz_SCDNotInt_smoke
  category: smoke
  function:
    matmul: *real_precisions_1b_fnuz_dst_f8_fnuz
  matrix_size:
    - { M:  128,  N:  128,  K:  128  }
  transA_transB: *transA_transB_range
  alpha: 1
  beta: [ 0.0, 2.0 ]
  scaleA: [0, 1]
  scaleB: [0, 1]
  scaleC: [1] # will use random value from 0.0 to 1.0
  scaleD: [1] # will use random value from 0.0 to 1.0
  bias_vector: [0, 1]
  bias_type: f16_r
  unit_check: 0
  norm_check: 1
  gpu_arch: '942'

- name: matmul_real_1b_fnuz_dst_bf8_fnuz_SCDInt1_smoke
  category: smoke
  function:
    matmul: *real_precisions_1b_fnuz_dst_bf8_fnuz
  matrix_size:
    - { M:  128,  N:  128,  K:  128  }
  transA_transB: *transA_transB_range
  alpha: 1
  beta: [ 0.0, 2.0 ]
  scaleA: [0, 1]
  scaleB: [0, 1]
  scaleC: [0] # will use default value '1'
  scaleD: [0] # will use default value '1'
  bias_vector: [0,1]
  bias_type: f16_r
  unit_check: 1
  gpu_arch: '942'

- name: matmul_real_1b_fnuz_dst_bf8_fnuz_SCDNotInt_smoke
  category: smoke
  function:
    matmul: *real_precisions_1b_fnuz_dst_bf8_fnuz
  matrix_size:
    - { M:  128,  N:  128,  K:  128  }
  transA_transB: *transA_transB_range
  alpha: 1
  beta: [ 0.0, 2.0 ]
  scaleA: [0, 1]
  scaleB: [0, 1]
  scaleC: [1] # will use random value from 0.0 to 1.0
  scaleD: [1] # will use random value from 0.0 to 1.0
  bias_vector: [0, 1]
  bias_type: f16_r
  unit_check: 0
  norm_check: 1
  gpu_arch: '942'

- name: matmul_gemm_xf32_smoke
  category: smoke
  function:
    matmul: *xf32_precision
  matrix_size: *smoke_matrix_size_range
  transA_transB: *transA_transB_range
  alpha: 1
  beta: [ 0.0, 2.0 ]
  unit_check: 1
  gpu_arch: '(942|1250)'

- name: matmul_gemm_double_smoke
  category: smoke
  function:
    matmul: *double_precision
  matrix_size: *smoke_matrix_size_range
  transA_transB: *transA_transB_range
  alpha: 1
  beta: [ 0.0, 2.0 ]
  unit_check: 1
  gpu_arch: '9(0a|4[0-2])'

- name: matmul_gemm_i8_dst_i32_smoke
  category: smoke
  function:
    matmul: *i8_precision_dst_i32
  matrix_size:
    - { M:  128,  N:  128,  K:  128  }
  transA_transB: *transA_transB_range
  alpha: 1
  beta: [ 0, 2 ]
  unit_check: 1
  gpu_arch: '90a'

- name: matmul_gemm_i8_dst_i32_94x_smoke
  category: smoke
  function:
    matmul: *i8_precision_dst_i32
  matrix_size: *smoke_matrix_size_range
  transA_transB: *transA_transB_range
  alpha: 1
  beta: [ 0, 2 ]
  scaleAlpha_vector: [0, 1]
  activation_type: [relu, sigmoid]
  unit_check: 1
  gpu_arch: '942'

- name: matmul_gemm_i8_dst_i8_1xxx_smoke
  category: smoke
  function:
    matmul: *i8_precision_dst_i8
  matrix_size:
    - { M:  128,  N:  128,  K:  128  }
  transA_transB: *transA_transB_range
  alpha: 1
  beta: [ 0, 2 ]
  scaleAlpha_vector: [0, 1]
  activation_type: [relu, sigmoid]
  unit_check: 1
  gpu_arch: '1[1-2]\d{2}'

- name: matmul_gemm_i8_dst_i32_1xxx_smoke
  category: smoke
  function:
    matmul: *i8_precision_dst_i32
  matrix_size:
    - { M:  128,  N:  128,  K:  128  }
  transA_transB: *transA_transB_range
  alpha: 1
  beta: [ 0, 2 ]
  scaleAlpha_vector: [0, 1]
  activation_type: [relu, sigmoid]
  unit_check: 1
  gpu_arch: '1[1-2]\d{2}'

- name: matmul_gemm_tn_mxf4_dst_bf16_1250_smoke
  category: smoke
  function:
    matmul: 
      - { a_type: f4_r, b_type: f4_r, c_type: bf16_r, d_type: bf16_r, compute_type: c_f32_r, scaleA: 3, scaleB: 3, scale_type: f32_r}
      - { a_type: f4_r, b_type: f4_r, c_type: bf16_r, d_type: bf16_r, compute_type: c_f32_r, scaleA: 6, scaleB: 6, scale_type: f32_r}
  matrix_size:
    - { M:  128,  N:  128,  K:  512  }
    - { M:  128,  N:  128,  K:  544  }
    - { M:  130,  N:  128,  K:  512  }
    - { M:  128,  N:  130,  K:  512  }
  transA: T
  transB: N
  alpha: 1
  beta: [ 0.0, 2.0 ]
  initialization: uniform_low_precision
  unit_check: 0
  norm_check: 1
  gpu_arch: '1250'

- name: matmul_gemm_tn_mxf4_dst_s_1250_smoke
  category: smoke
  function:
    matmul:
      - { a_type: f4_r, b_type: f4_r, c_type: f32_r, d_type: f32_r, compute_type: c_f32_r, scaleA: 3, scaleB: 3, scale_type: f32_r}
      - { a_type: f4_r, b_type: f4_r, c_type: f32_r, d_type: f32_r, compute_type: c_f32_r, scaleA: 6, scaleB: 6, scale_type: f32_r}
  matrix_size:
    - { M:  128,  N:  128,  K:  512  }
    - { M:  128,  N:  128,  K:  544  }
    - { M:  130,  N:  128,  K:  512  }
    - { M:  128,  N:  130,  K:  512  }
  transA: T
  transB: N
  alpha: 1
  beta: [ 0.0, 2.0 ]
  initialization: uniform_low_precision
  unit_check: 0
  norm_check: 1
  gpu_arch: '1250'

- name: matmul_gemm_tn_mxf8f4_dst_s_1250_bias_smoke
  category: smoke
  function:
    matmul:
      - { a_type: f8_r, b_type: f4_r, c_type: f32_r, d_type: f32_r, compute_type: c_f32_r, compute_input_typeA: f8_r, compute_input_typeB: f4_r, scaleA: 3, scaleB: 3, scale_type: f32_r }
  matrix_size:
    - { M:  128,  N:  128,  K:  512 }
    - { M:  128,  N:  128,  K:  544 }
    - { M:  130,  N:  128,  K:  512 }
    - { M:  128,  N:  130,  K:  512 }
  transA: T
  transB: N
  alpha: 1
  beta: [ 0.0, 2.0 ]
  bias_vector: [ 0, 1 ]
  bias_type: f32_r
  unit_check: 1
  gpu_arch: '1250'

- name: matmul_gemm_tn_mxf8_dst_bf16_1250_smoke
  category: smoke
  function:
    matmul: 
      - { a_type: f8_r, b_type: f8_r, c_type: bf16_r, d_type: bf16_r, compute_type: c_f32_r, scaleA: 3, scaleB: 3, scale_type: f32_r}
  matrix_size:
    - { M:  128,  N:  128,  K:  512  }
    - { M:  128,  N:  128,  K:  544  }
    - { M:  129,  N:  128,  K:  512  }
    - { M:  128,  N:  129,  K:  512  }

  transA: T
  transB: N
  alpha: 1
  beta: [ 0.0, 2.0 ]
  batch_count: [1, 16]
  unit_check: 0
  norm_check: 1
  gpu_arch: '1250'

- name: matmul_gemm_tn_mxf8_dst_s_1250_smoke
  category: smoke
  function:
    matmul: 
      - { a_type: f8_r, b_type: f8_r, c_type: f32_r, d_type: f32_r, compute_type: c_f32_r, scaleA: 3, scaleB: 3, scale_type: f32_r}
  matrix_size:
    - { M:  128,  N:  128,  K:  512  }
    - { M:  128,  N:  128,  K:  544  }
    - { M:  129,  N:  128,  K:  512  }
    - { M:  128,  N:  129,  K:  512  }
  transA: T
  transB: N
  alpha: 1
  beta: [ 0.0, 2.0 ]
  unit_check: 1
  gpu_arch: '1250'
...