GeezerVison is showing github.com — Tensile/pytest.ini at develop_deprecated · ROCm/Tensile

Expand file tree

pytest.ini

Copy path

More file actions

Latest commit

History

130 lines (128 loc) · 3.08 KB

pytest.ini

Copy path

File metadata and controls

130 lines (128 loc) · 3.08 KB

Raw

Copy raw file

Download raw file

Open symbols panel

Edit and raw actions

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

[pytest]

junit_family = xunit2

junit_logging = all

junit_log_passing_tests = False

xfail_strict = True

markers =

extended: Longer tests, formerly nightly.

pre_checkin: All pre-checkin tests.

disabled: Disabled tests.

regression: directory: tests for some bug hotspots

bufferload_offset: directory

bugs: directory

dot: directory

nightly: directory

no_load_loop: Directory, tests for the OptNoLoadLoop option.

special: directory

unit: Unit tests

use_initial_strides: directory

use_initial_strides_cd: directory

vega_20: directory

weekly: directory

yaml_only: directory, all tests which simply run a YAML file.

integration:

denorm: directory, tests for denorm behavior

emulation: Quick set of essential tests to run on emulation.

emulation1: Emulation tests part 1

emulation2: Emulation tests part 2

emulation3: Emulation tests part 3

emulation4: Emulation tests part 4

emulation5: Emulation tests part 5

validate: All tests which validate the results.

validateAll: All tests which validate all data points.

GEMM: All tests which run standard GEMMs

TensorContraction: All tests which do more complex tensor contractions

ConvolutionForward: Forward weights

Assembly: Tests with assembly kernels.

Source: Tests with source kernels.

bfloat16: Data type

bfloat8: Data type

bfloat8Float8: Data type

complexDouble : Data type

complexSingle : Data type

double : Data type

float8 : Data type

float8Bfloat8 : Data type

half : Data type

int8x4 : Data type

int8 : Data type

single: Data type

assertions

big_tensor

classic

classic_source

convolution

convolution-vs-contraction

cov

custom_kernel

direct_to_lds

direct_to_vgpr

dot2

double_complex

fast

flat

float_complex

fractional

global_split_u

global_split_u_src

hpa_source

igemm

local_split_u

mfma

multi_sum

multi_sum_psd

nonbatched

pack_tensor_dims

pbd

source

stagger_u

stream_k

syntax_error

tensor_contraction

vector_width

wip

wmma

zeropad

bufferload_offset

mirror_dims

xfail-gfx000: architecture

xfail-gfx900: architecture

xfail-gfx906: architecture

xfail-gfx908: architecture

xfail-gfx90a: architecture

xfail-gfx940: architecture

xfail-gfx941: architecture

xfail-gfx942: architecture

xfail-gfx950: architecture

xfail-gfx1010: architecture

xfail-gfx1011: architecture

xfail-gfx1012: architecture

xfail-gfx1030: architecture

xfail-gfx1031: architecture

xfail-gfx1032: architecture

xfail-gfx1034: architecture

xfail-gfx1035: architecture

xfail-gfx1100: architecture

xfail-gfx1101: architecture

xfail-gfx1102: architecture

skip-gfx000: architecture

skip-gfx900: architecture

skip-gfx906: architecture

skip-gfx908: architecture

skip-gfx90a: architecture

skip-gfx940: architecture

skip-gfx941: architecture

skip-gfx942: architecture

skip-gfx950: architecture

skip-gfx1010: architecture

skip-gfx1011: architecture

skip-gfx1012: architecture

skip-gfx1030: architecture

skip-gfx1031: architecture

skip-gfx1032: architecture

skip-gfx1034: architecture

skip-gfx1035: architecture

skip-gfx1100: architecture

skip-gfx1101: architecture

skip-gfx1102: architecture

skip-gfx1151: architecture