Skip to content
This repository has been archived by the owner on Mar 28, 2023. It is now read-only.

XFAIL: cuda tracker #249

Closed
steffenlarsen opened this issue Apr 22, 2021 · 6 comments
Closed

XFAIL: cuda tracker #249

steffenlarsen opened this issue Apr 22, 2021 · 6 comments
Labels
compiler Compiler related issue cuda Issues related to execution on CUDA backend. runtime Runtime library related issue

Comments

@steffenlarsen
Copy link

steffenlarsen commented Apr 22, 2021 •
edited
Loading

This issue is for tracking XFAIL:cuda tests. This is a continuation of intel/llvm#1542.

The following test currently have XFAIL:cuda in them:
SYCL/Basic/reqd_work_group_size.cpp (Enabled in #278)
SYCL/Config/select_device.cpp
SYCL/DeviceLib/built-ins/nan.cpp (Disabled #336)
SYCL/HierPar/hier_par_wgscope_O0.cpp (Disabled in #290)
SYCL/KernelAndProgram/basic-program.cpp
SYCL/KernelAndProgram/build-log.cpp
SYCL/KernelAndProgram/cache-build-result.cpp
SYCL/KernelAndProgram/get-options.cpp
SYCL/KernelAndProgram/kernel-and-program.cpp
SYCL/Regression/atomic_load.cpp (Enabled in #339)
SYCL/Regression/implicit_offset_debug_info.cpp (Enabled with #264)
SYCL/Sampler/normalized-clamp-linear-float.cpp
SYCL/Sampler/normalized-clampedge-linear-float.cpp
SYCL/Sampler/normalized-mirror-linear-float.cpp
SYCL/Sampler/normalized-mirror-nearest.cpp
SYCL/Sampler/normalized-none-linear-float.cpp
SYCL/Sampler/normalized-repeat-linear-float.cpp
SYCL/Sampler/normalized-repeat-nearest.cpp
SYCL/Sampler/unnormalized-clamp-linear-float.cpp
SYCL/Sampler/unnormalized-clampedge-linear-float.cpp
SYCL/Sampler/unnormalized-none-linear-float.cpp
SYCL/Scheduler/CommandCleanupThreadSafety.cpp
SYCL/Scheduler/MemObjRemapping.cpp (Enabled in #339)
SYCL/USM/prefetch.cpp
@steffenlarsen steffenlarsen mentioned this issue Apr 22, 2021
Closed
@vmaksimo
Copy link

SYCL/HierPar/hier_par_wgscope_O0.cpp
(disabled in #290)

@vmaksimo vmaksimo mentioned this issue May 25, 2021
Merged
@bader bader added the cuda Issues related to execution on CUDA backend. label Aug 3, 2021
@cperkinsintel cperkinsintel mentioned this issue Mar 15, 2022
Open
@AerialMantis AerialMantis added compiler Compiler related issue runtime Runtime library related issue labels May 8, 2022
@t4c1
Copy link

t4c1 commented Jun 20, 2022 •
edited
Loading

Here is an updated list of tests with XFAIL: cuda:

AtomicRef/*_generic*  (PR opened: intel/llvm#5849, but blocked by the remangler)
Config/select_device.cpp
Regression/kernel_bundle_ignore_sycl_external.cpp
Sampler/normalized-mirror-linear-float.cpp
Sampler/unnormalized-clamp-linear-float.cpp
Sampler/unnormalized-clampedge-linear-float.cpp
Sampler/unnormalized-none-linear-float.cpp
Sampler/normalized-clamp-linear-float.cpp
Sampler/normalized-repeat-linear-float.cpp
Sampler/normalized-none-linear-float.cpp
Sampler/normalized-clampedge-linear-float.cpp
Sampler/normalized-repeat-nearest.cpp
Sampler/normalized-mirror-nearest.cpp
KernelAndProgram/undefined-symbol.cpp
KernelAndProgram/cache-build-result.cpp
KernelAndProgram/build-log.cpp
Printf/int.cpp
Scheduler/CommandCleanupThreadSafety.cpp

Here is also a list of tests with UNSUPPORTED: cuda (ignoring ESIMD and DeprecatedFeatures folders) that in fact pass:

SpecConstants/2020/host_apis.cpp
SpecConstants/2020/handler-api.cpp
SpecConstants/2020/vector-convolution-demo.cpp
SpecConstants/2020/non_native/cuda.cpp
SpecConstants/2020/kernel-bundle-api.cpp
GroupAlgorithm/reduce_sycl2020.cpp
GroupAlgorithm/inclusive_scan_sycl2020.cpp
GroupAlgorithm/exclusive_scan_sycl2020.cpp
GroupAlgorithm/exclusive_scan_over_group.cpp
Assert/assert_in_kernels_ndebug.cpp
DeviceLib/ITTAnnotations/atomic.cpp
DeviceLib/ITTAnnotations/barrier.cpp
Regression/image_access.cpp
Basic/image/image_write.cpp
Basic/image/image_sample.cpp
Basic/image/image_array.cpp

Here is also a list of tests with UNSUPPORTED: cuda (ignoring ESIMD and DeprecatedFeatures folders) that seem to need some work:

Assert/assert_in_kernels.cpp
Assert/assert_in_multiple_tus.cpp
Assert/assert_in_multiple_tus_one_ndebug.cpp
Assert/assert_in_one_kernel.cpp
Assert/assert_in_simultaneous_kernels.cpp
Assert/assert_in_simultaneously_multiple_tus.cpp
Assert/assert_in_simultaneously_multiple_tus_one_ndebug.cpp
BFloat16/bfloat16_type.cpp
Basic/image/image_accessor_readsampler.cpp
Basic/image/image_accessor_readwrite.cpp
Basic/image/image_accessor_readwrite_half.cpp
Basic/image/image_max_size.cpp
Basic/image/srgba-read.cpp
DeviceLib/built-ins/printf.cpp
DiscardEvents/discard_events_using_assert.cpp
KernelAndProgram/kernel-bundle-merge-options-env.cpp
KernelAndProgram/kernel-bundle-merge-options.cpp
KernelAndProgram/multiple-kernel-linking.cpp
Plugin/sycl-ls-gpu-default-any.cpp
Regression/commandlist/gpu.cpp
Regression/device_num.cpp
Sampler/unnormalized-clamp-nearest.cpp
Sampler/unnormalized-clampedge-nearest.cpp
Sampler/unnormalized-none-nearest.cpp
Scheduler/HostAccDestruction.cpp
SeparateCompile/test.cpp
SubGroup/reduce_spirv13.cpp
SubGroup/reduce_spirv13_fp16.cpp
SubGroup/scan_spirv13.cpp
SubGroup/scan_spirv13_fp16.cpp

@t4c1 t4c1 mentioned this issue Jun 20, 2022
Merged
pvchupin pushed a commit that referenced this issue Jun 21, 2022
@t4c1
enable working tests (#1059)
6b9e418 
Enables tests that were marked with UNSUPPORTED: cuda, but are in fact passing.

Result of work on #249
@AerialMantis
Copy link

AerialMantis commented Aug 5, 2022 •
edited
Loading

Here is a list of the pull requests for addressing issues in the UNSUPPORTED: cuda tests which can be fixed:

The Printf/int.cpp was not fixed as it is failing due to an Nvidia bug, which has been reported here - https://docs.nvidia.com/cuda/cuda-c-programming-guide/index.html#format-specifiers

After some investigation we also found that the following tests are not expected to pass with the CUDA backend:

  • BFloat16/bfloat16_type.cpp
  • SeparateCompile/test.cpp
  • KernelAndProgram/cache-build-result.cpp
  • KernelAndProgram/build-log.cpp
  • KernelAndProgram/undefined-symbol.cpp

@pgorlani pgorlani mentioned this issue Aug 5, 2022
Merged
@pvchupin
Copy link

pvchupin commented Aug 5, 2022

Here is a list of the pull requests ...

@AerialMantis, feel free to push individual PRs. It seems some a stuck for no apparent reason...

@jle-quel
Copy link

jle-quel commented Oct 5, 2022

Here is an updated list of the tests marked as XFAIL and UNSUPPORTED for CUDA:

[+] XFAIL
SYCL/AtomicRef/load_generic_local.cpp
SYCL/AtomicRef/store_generic_local.cpp
SYCL/AtomicRef/add_generic_native_fp.cpp
SYCL/AtomicRef/or_generic_local.cpp
SYCL/AtomicRef/add_generic.cpp
SYCL/AtomicRef/add_generic_local.cpp
SYCL/AtomicRef/add_generic_local_native_fp.cpp
SYCL/AtomicRef/max_generic.cpp
SYCL/AtomicRef/max_generic_local.cpp
SYCL/AtomicRef/max_generic_local_native_fp.cpp
SYCL/AtomicRef/max_generic_native_fp.cpp
SYCL/AtomicRef/exchange_generic_local.cpp
SYCL/AtomicRef/sub_generic_native_fp.cpp
SYCL/AtomicRef/sub_generic_local_native_fp.cpp
SYCL/AtomicRef/and_generic_local.cpp
SYCL/AtomicRef/and_generic.cpp
SYCL/AtomicRef/sub_generic_local.cpp
SYCL/AtomicRef/assignment_atomic64_generic.cpp
SYCL/AtomicRef/or_generic.cpp
SYCL/AtomicRef/compare_exchange_generic_local.cpp
SYCL/AtomicRef/store_generic.cpp
SYCL/AtomicRef/compare_exchange_generic.cpp
SYCL/AtomicRef/load_generic.cpp
SYCL/AtomicRef/min_generic.cpp
SYCL/AtomicRef/min_generic_local.cpp
SYCL/AtomicRef/min_generic_native_fp.cpp
SYCL/AtomicRef/min_generic_local_native_fp.cpp
SYCL/AtomicRef/xor_generic.cpp
SYCL/AtomicRef/sub_generic.cpp
SYCL/AtomicRef/exchange_generic.cpp
SYCL/AtomicRef/assignment_generic.cpp
SYCL/AtomicRef/xor_generic_local.cpp
SYCL/Regression/kernel_bundle_ignore_sycl_external.cpp
SYCL/Printf/int.cpp
SYCL/Sampler/normalized-clampedge-linear-float.cpp
SYCL/Sampler/normalized-mirror-linear-float.cpp
SYCL/Sampler/unnormalized-clampedge-linear-float.cpp
SYCL/Sampler/normalized-repeat-linear-float.cpp
SYCL/Sampler/unnormalized-none-linear-float.cpp
SYCL/Sampler/normalized-mirror-nearest.cpp
SYCL/Sampler/normalized-clamp-linear-float.cpp
SYCL/Sampler/normalized-none-linear-float.cpp
SYCL/Sampler/unnormalized-clamp-linear-float.cpp
SYCL/Sampler/normalized-repeat-nearest.cpp

[+] UNSUPPORTED
SYCL/DeviceLib/imf_simd_emulate_test.cpp
SYCL/DeviceLib/imf_fp64_test.cpp
SYCL/DeviceLib/ITTAnnotations/atomic.cpp
SYCL/DeviceLib/ITTAnnotations/barrier.cpp
SYCL/DeviceLib/imf_fp32_test.cpp
SYCL/DeprecatedFeatures/deprecated_intel_ext_device.cpp
SYCL/Plugin/sycl-ls-gpu-default-any.cpp
SYCL/Plugin/sycl-ls.cpp
SYCL/DiscardEvents/discard_events_using_assert.cpp
SYCL/InlineAsm/asm_float_neg.cpp
SYCL/InlineAsm/asm_mul.cpp
SYCL/InlineAsm/asm_no_operands.cpp
SYCL/InlineAsm/Negative/asm_wrong_declare.cpp
SYCL/InlineAsm/Negative/asm_illegal_exec_size.cpp
SYCL/InlineAsm/Negative/asm_bad_opcode.cpp
SYCL/InlineAsm/Negative/asm_bad_operand_syntax.cpp
SYCL/InlineAsm/Negative/asm_duplicate_label.cpp
SYCL/InlineAsm/Negative/asm_undefined_decl.cpp
SYCL/InlineAsm/Negative/asm_undefined_pred.cpp
SYCL/InlineAsm/Negative/asm_missing_label.cpp
SYCL/InlineAsm/Negative/asm_missing_region.cpp
SYCL/InlineAsm/Negative/asm_simple.cpp
SYCL/InlineAsm/malloc_shared_in_out_dif.cpp
SYCL/InlineAsm/asm_no_output.cpp
SYCL/InlineAsm/asm_16_no_opts.cpp
SYCL/InlineAsm/asm_16_matrix_mult.cpp
SYCL/InlineAsm/asm_16_empty.cpp
SYCL/InlineAsm/asm_if.cpp
SYCL/InlineAsm/malloc_shared_no_input.cpp
SYCL/InlineAsm/asm_multiple_instructions.cpp
SYCL/InlineAsm/asm_decl_in_scope.cpp
SYCL/InlineAsm/asm_16_no_input_int.cpp
SYCL/InlineAsm/asm_plus_mod.cpp
SYCL/InlineAsm/asm_arbitrary_ops_order.cpp
SYCL/InlineAsm/malloc_shared_32.cpp
SYCL/InlineAsm/asm_loop.cpp
SYCL/InlineAsm/asm_8_empty.cpp
SYCL/InlineAsm/asm_switch.cpp
SYCL/InlineAsm/letter_example.cpp
SYCL/InlineAsm/asm_float_add.cpp
SYCL/InlineAsm/asm_8_no_input_int.cpp
SYCL/InlineAsm/asm_float_imm_arg.cpp
SYCL/InlineAsm/asm_imm_arg.cpp
SYCL/Assert/assert_in_simultaneously_multiple_tus_one_ndebug.cpp
SYCL/Assert/assert_in_simultaneous_kernels.cpp
SYCL/Assert/assert_in_simultaneously_multiple_tus.cpp
SYCL/SpecConstants/2020/non_native/gpu.cpp
SYCL/SpecConstants/2020/non_native/multiple-targets.cpp
SYCL/USM/dep_events.cpp
SYCL/USM/prefetch.cpp
SYCL/SeparateCompile/test.cpp
SYCL/BFloat16/bfloat16_type.cpp
SYCL/AOT/gpu.cpp
SYCL/AOT/multiple-devices.cpp
SYCL/ESIMD/acc_gather_scatter_rgba.cpp
SYCL/ESIMD/regression/sycl_esimd_mixed_unnamed.cpp
SYCL/ESIMD/regression/usm_gather_scatter_32.cpp
SYCL/ESIMD/regression/big_const_initializer.cpp
SYCL/ESIMD/regression/unused_load.cpp
SYCL/ESIMD/regression/operator_decrement.cpp
SYCL/ESIMD/regression/abs_fix_test.cpp
SYCL/ESIMD/regression/dgetrf_ref.cpp
SYCL/ESIMD/regression/copyto_char_test.cpp
SYCL/ESIMD/regression/replicate_bug.cpp
SYCL/ESIMD/regression/variable_gather_mask.cpp
SYCL/ESIMD/regression/rol_ror_compilation_test.cpp
SYCL/ESIMD/regression/globals.cpp
SYCL/ESIMD/regression/fmod_compatibility_test.cpp
SYCL/ESIMD/regression/complex-lib-lin.cpp
SYCL/ESIMD/regression/double_conversion.cpp
SYCL/ESIMD/regression/half_conversion_test.cpp
SYCL/ESIMD/regression/dgetrf.cpp
SYCL/ESIMD/regression/store_zero_const.cpp
SYCL/ESIMD/regression/noinline_bypointers_vadd.cpp
SYCL/ESIMD/regression/dgetrf_8x8.cpp
SYCL/ESIMD/regression/bit_shift_compilation_test.cpp
SYCL/ESIMD/stencil2.cpp
SYCL/ESIMD/noinline_call_from_func.cpp
SYCL/ESIMD/vadd_usm.cpp
SYCL/ESIMD/preemption.cpp
SYCL/ESIMD/fp_call_recursive.cpp
SYCL/ESIMD/api/svm_gather_scatter.cpp
SYCL/ESIMD/api/esimd_merge.cpp
SYCL/ESIMD/api/simd_binop_integer_promotion.cpp
SYCL/ESIMD/api/simd_any_all.cpp
SYCL/ESIMD/api/bin_and_cmp_ops_heavy.cpp
SYCL/ESIMD/api/replicate_smoke.cpp
SYCL/ESIMD/api/slm_gather_scatter_heavy.cpp
SYCL/ESIMD/api/functional/operators/operator_decrement_and_increment_accuracy_fp_extra.cpp
SYCL/ESIMD/api/functional/operators/operator_bitwise_not_uint.cpp
SYCL/ESIMD/api/functional/operators/operator_decrement_and_increment_core.cpp
SYCL/ESIMD/api/functional/operators/operator_decrement_and_increment_fp_extra.cpp
SYCL/ESIMD/api/functional/operators/operator_bitwise_not_sint.cpp
SYCL/ESIMD/api/functional/operators/operator_decrement_and_increment_accuracy_core.cpp
SYCL/ESIMD/api/functional/operators/operator_logical_not.cpp
SYCL/ESIMD/api/functional/ctors/ctor_converting_core.cpp
SYCL/ESIMD/api/functional/ctors/ctor_broadcast_fp_extra.cpp
SYCL/ESIMD/api/functional/ctors/ctor_vector_core.cpp
SYCL/ESIMD/api/functional/ctors/ctor_fill_fp_extra.cpp
SYCL/ESIMD/api/functional/ctors/ctor_load_usm_fp_extra.cpp
SYCL/ESIMD/api/functional/ctors/ctor_default_core.cpp
SYCL/ESIMD/api/functional/ctors/ctor_broadcast_core.cpp
SYCL/ESIMD/api/functional/ctors/ctor_vector_fp_extra.cpp
SYCL/ESIMD/api/functional/ctors/ctor_array_core.cpp
SYCL/ESIMD/api/functional/ctors/ctor_fill_core.cpp
SYCL/ESIMD/api/functional/ctors/ctor_load_acc_core.cpp
SYCL/ESIMD/api/functional/ctors/ctor_copy_fp_extra.cpp
SYCL/ESIMD/api/functional/ctors/ctor_copy_core.cpp
SYCL/ESIMD/api/functional/ctors/ctor_fill_accuracy_fp.cpp
SYCL/ESIMD/api/functional/ctors/ctor_move_core.cpp
SYCL/ESIMD/api/functional/ctors/ctor_array_fp_extra.cpp
SYCL/ESIMD/api/functional/ctors/ctor_move_fp_extra.cpp
SYCL/ESIMD/api/functional/ctors/ctor_fill_accuracy_fp_extra.cpp
SYCL/ESIMD/api/functional/ctors/ctor_load_acc_fp_extra.cpp
SYCL/ESIMD/api/functional/ctors/ctor_load_usm_core.cpp
SYCL/ESIMD/api/functional/ctors/ctor_converting_fp_extra.cpp
SYCL/ESIMD/api/functional/ctors/ctor_default_fp_extra.cpp
SYCL/ESIMD/api/functional/functions/functions_select_simd_view_rval_fp_extra.cpp
SYCL/ESIMD/api/functional/functions/functions_select_2d_core.cpp
SYCL/ESIMD/api/functional/functions/functions_select_rvalue_fp_extra.cpp
SYCL/ESIMD/api/functional/functions/functions_select_lvalue_fp_extra.cpp
SYCL/ESIMD/api/functional/functions/functions_select_simd_view_rval_core.cpp
SYCL/ESIMD/api/functional/functions/functions_select_rvalue_core.cpp
SYCL/ESIMD/api/functional/functions/functions_select_2d_fp_extra.cpp
SYCL/ESIMD/api/functional/functions/functions_select_lvalue_core.cpp
SYCL/ESIMD/api/simd_memory_access.cpp
SYCL/ESIMD/api/slm_gather_scatter_rgba.cpp
SYCL/ESIMD/api/simd_view_select_2d_int.cpp
SYCL/ESIMD/api/simd_view_subscript_operator.cpp
SYCL/ESIMD/api/simd_subscript_operator.cpp
SYCL/ESIMD/api/esimd_bit_ops.cpp
SYCL/ESIMD/api/esimd_pack_unpack_mask.cpp
SYCL/ESIMD/api/slm_gather_scatter.cpp
SYCL/ESIMD/api/simd_view_copy_move_assign.cpp
SYCL/ESIMD/api/saturation_smoke.cpp
SYCL/ESIMD/api/simd_view_negation_operator.cpp
SYCL/ESIMD/api/simd_view_select_2d_fp.cpp
SYCL/ESIMD/api/simd_mask.cpp
SYCL/ESIMD/api/esimd_rgba_smoke.cpp
SYCL/ESIMD/api/ballot.cpp
SYCL/ESIMD/api/simd_copy_to_from.cpp
SYCL/ESIMD/api/unary_ops_heavy.cpp
SYCL/ESIMD/api/simd_negation_operator.cpp
SYCL/ESIMD/aot_mixed.cpp
SYCL/ESIMD/radix_sort.cpp
SYCL/ESIMD/dpas/dpas_tf32.cpp
SYCL/ESIMD/dpas/dpas_test3.cpp
SYCL/ESIMD/dpas/dpasw_test.cpp
SYCL/ESIMD/dpas/dpas_test2.cpp
SYCL/ESIMD/dpas/dpas_test1.cpp
SYCL/ESIMD/dpas/dpas_bf16.cpp
SYCL/ESIMD/dpas/dpas_fp16.cpp
SYCL/ESIMD/dpas/dpas_int.cpp
SYCL/ESIMD/accessor_stateless.cpp
SYCL/ESIMD/matrix_transpose.cpp
SYCL/ESIMD/esimd_check_vc_codegen.cpp
SYCL/ESIMD/BitonicSortK.cpp
SYCL/ESIMD/accessor.cpp
SYCL/ESIMD/vec_arg_call_conv_smoke.cpp
SYCL/ESIMD/vadd_2d.cpp
SYCL/ESIMD/accessor_gather_scatter_stateless.cpp
SYCL/ESIMD/vadd_1d.cpp
SYCL/ESIMD/fp_call_from_func.cpp
SYCL/ESIMD/ext_math.cpp
SYCL/ESIMD/noinline_args_size/noinline_args_char_int_size_96.cpp
SYCL/ESIMD/noinline_args_size/noinline_args_char_int_size_64.cpp
SYCL/ESIMD/noinline_args_size/noinline_args_int_size_512.cpp
SYCL/ESIMD/noinline_args_size/noinline_args_char_int_size_192.cpp
SYCL/ESIMD/noinline_args_size/noinline_args_int_size_96.cpp
SYCL/ESIMD/noinline_args_size/noinline_args_int_size_192.cpp
SYCL/ESIMD/noinline_args_size/noinline_args_char_int_size_256.cpp
SYCL/ESIMD/noinline_args_size/noinline_args_int_size_256.cpp
SYCL/ESIMD/noinline_args_size/noinline_args_char_int_size_512.cpp
SYCL/ESIMD/noinline_args_size/noinline_args_int_size_64.cpp
SYCL/ESIMD/vadd_raw_send.cpp
SYCL/ESIMD/mandelbrot/mandelbrot_spec.cpp
SYCL/ESIMD/mandelbrot/mandelbrot.cpp
SYCL/ESIMD/fp_in_phi.cpp
SYCL/ESIMD/Prefix_Local_sum2.cpp
SYCL/ESIMD/matrix_transpose2.cpp
SYCL/ESIMD/Prefix_Local_sum1.cpp
SYCL/ESIMD/spec_const/spec_const_uint.cpp
SYCL/ESIMD/spec_const/spec_const_char.cpp
SYCL/ESIMD/spec_const/spec_const_int64.cpp
SYCL/ESIMD/spec_const/spec_const_float.cpp
SYCL/ESIMD/spec_const/spec_const_bool.cpp
SYCL/ESIMD/spec_const/spec_const_ushort.cpp
SYCL/ESIMD/spec_const/spec_const_short.cpp
SYCL/ESIMD/spec_const/spec_const_double.cpp
SYCL/ESIMD/spec_const/unused_spec_const.cpp
SYCL/ESIMD/spec_const/spec_const_int.cpp
SYCL/ESIMD/spec_const/spec_const_uint64.cpp
SYCL/ESIMD/spec_const/spec_const_uchar.cpp
SYCL/ESIMD/fp_args_size/fp_args_char_int_size_96.cpp
SYCL/ESIMD/fp_args_size/fp_args_char_int_size_256.cpp
SYCL/ESIMD/fp_args_size/fp_args_int_size_256.cpp
SYCL/ESIMD/fp_args_size/fp_args_int_size_512.cpp
SYCL/ESIMD/fp_args_size/fp_args_char_int_size_192.cpp
SYCL/ESIMD/fp_args_size/fp_args_int_size_64.cpp
SYCL/ESIMD/fp_args_size/fp_args_int_size_96.cpp
SYCL/ESIMD/fp_args_size/fp_args_char_int_size_64.cpp
SYCL/ESIMD/fp_args_size/fp_args_char_int_size_512.cpp
SYCL/ESIMD/fp_args_size/fp_args_int_size_192.cpp
SYCL/ESIMD/printf.cpp
SYCL/ESIMD/matrix_transpose_glb.cpp
SYCL/ESIMD/usm_gather_scatter_rgba.cpp
SYCL/ESIMD/fp_in_select.cpp
SYCL/ESIMD/accessor_load_store_stateless.cpp
SYCL/ESIMD/dword_atomic_cmpxchg.cpp
SYCL/ESIMD/histogram_raw_send.cpp
SYCL/ESIMD/histogram_256_slm_spec_2020.cpp
SYCL/ESIMD/accessor_gather_scatter.cpp
SYCL/ESIMD/Prefix_Local_sum3.cpp
SYCL/ESIMD/slm_barrier.cpp
SYCL/ESIMD/vec_arg_call_conv_ext.cpp
SYCL/ESIMD/reduction.cpp
SYCL/ESIMD/lsc/lsc_slm_scatter.cpp
SYCL/ESIMD/lsc/lsc_surf_load_u16u32.cpp
SYCL/ESIMD/lsc/lsc_block_store_u32.cpp
SYCL/ESIMD/lsc/lsc_surf_store_u8_u16.cpp
SYCL/ESIMD/lsc/lsc_block_store_u64.cpp
SYCL/ESIMD/lsc/lsc_surf_prefetch_u64.cpp
SYCL/ESIMD/lsc/lsc_block_load_u64.cpp
SYCL/ESIMD/lsc/lsc_block_prefetch_u64.cpp
SYCL/ESIMD/lsc/lsc_usm_store_u8_u16.cpp
SYCL/ESIMD/lsc/lsc_usm_store_u8u32.cpp
SYCL/ESIMD/lsc/lsc_surf.cpp
SYCL/ESIMD/lsc/lsc_surf_load_u64.cpp
SYCL/ESIMD/lsc/lsc_surf_stateless.cpp
SYCL/ESIMD/lsc/lsc_block_load_u8.cpp
SYCL/ESIMD/lsc/lsc_surf_prefetch_u16u32.cpp
SYCL/ESIMD/lsc/lsc_usm_prefetch_u64.cpp
SYCL/ESIMD/lsc/lsc_usm_atomic_cachehint.cpp
SYCL/ESIMD/lsc/lsc_usm_2d.cpp
SYCL/ESIMD/lsc/lsc_slm_block_store.cpp
SYCL/ESIMD/lsc/lsc_usm_prefetch_u8u32.cpp
SYCL/ESIMD/lsc/lsc_surf_load_u8u32.cpp
SYCL/ESIMD/lsc/lsc_surf_prefetch_u32.cpp
SYCL/ESIMD/lsc/lsc_usm_prefetch_u32.cpp
SYCL/ESIMD/lsc/lsc_surf_store_u16u32.cpp
SYCL/ESIMD/lsc/lsc_surf_prefetch_u8u32.cpp
SYCL/ESIMD/lsc/lsc_usm_store_u16u32.cpp
SYCL/ESIMD/lsc/lsc_usm_load_u32.cpp
SYCL/ESIMD/lsc/lsc_slm_block_load.cpp
SYCL/ESIMD/lsc/lsc_usm_load_u16u32.cpp
SYCL/ESIMD/lsc/lsc_surf_load_stateless.cpp
SYCL/ESIMD/lsc/lsc_usm_load_u8u32.cpp
SYCL/ESIMD/lsc/lsc_block_load_u32.cpp
SYCL/ESIMD/lsc/lsc_usm.cpp
SYCL/ESIMD/lsc/lsc_surf_load_u8_u16.cpp
SYCL/ESIMD/lsc/lsc_slm.cpp
SYCL/ESIMD/lsc/lsc_block_load_u16.cpp
SYCL/ESIMD/lsc/lsc_block_prefetch_u16.cpp
SYCL/ESIMD/lsc/lsc_slm_gather.cpp
SYCL/ESIMD/lsc/lsc_usm_load_u64.cpp
SYCL/ESIMD/lsc/lsc_usm_store_u64.cpp
SYCL/ESIMD/lsc/lsc_block_store_u8.cpp
SYCL/ESIMD/lsc/lsc_surf_store_u64.cpp
SYCL/ESIMD/lsc/lsc_block_store_u16.cpp
SYCL/ESIMD/lsc/lsc_usm_prefetch_u16u32.cpp
SYCL/ESIMD/lsc/lsc_block_prefetch_u32.cpp
SYCL/ESIMD/lsc/lsc_surf_load_u32.cpp
SYCL/ESIMD/lsc/lsc_surf_store_u32.cpp
SYCL/ESIMD/lsc/lsc_block_prefetch_u8.cpp
SYCL/ESIMD/lsc/lsc_fence.cpp
SYCL/ESIMD/lsc/lsc_usm_load_u8_u16.cpp
SYCL/ESIMD/lsc/lsc_surf_store_u8u32.cpp
SYCL/ESIMD/lsc/lsc_surf_store_stateless.cpp
SYCL/ESIMD/lsc/lsc_usm_store_u32.cpp
SYCL/ESIMD/histogram.cpp
SYCL/ESIMD/private_memory/pm_access_3.cpp
SYCL/ESIMD/private_memory/pm_access_2.cpp
SYCL/ESIMD/private_memory/pm_access_1.cpp
SYCL/ESIMD/vadd_2d_acc.cpp
SYCL/ESIMD/slm_split_barrier.cpp
SYCL/ESIMD/noinline_call_recursive.cpp
SYCL/ESIMD/acc_gather_scatter_rgba_stateless.cpp
SYCL/ESIMD/histogram_2d.cpp
SYCL/ESIMD/PrefixSum.cpp
SYCL/ESIMD/kmeans/kmeans.cpp
SYCL/ESIMD/dword_atomic_smoke.cpp
SYCL/ESIMD/accessor_load_store.cpp
SYCL/ESIMD/vadd_half.cpp
SYCL/ESIMD/linear/linear.cpp
SYCL/ESIMD/matrix_transpose_usm.cpp
SYCL/ESIMD/test_id_3d.cpp
SYCL/ESIMD/BitonicSortKv2.cpp
SYCL/ESIMD/vadd_usm_opqptr.cpp
SYCL/ESIMD/Stencil.cpp
SYCL/ESIMD/sycl_esimd_mix.cpp
SYCL/ESIMD/histogram_256_slm.cpp
SYCL/HostInteropTask/interop-task.cpp
SYCL/Basic/code_location_e2e.cpp
SYCL/Basic/image/image_accessor_readwrite_half.cpp
SYCL/Basic/image/image_max_size.cpp
SYCL/Basic/image/image_write_fp16.cpp
SYCL/Basic/image/image_accessor_range.cpp
SYCL/Basic/image/image_write.cpp
SYCL/Basic/image/srgba-read.cpp
SYCL/Basic/image/image_accessor_readwrite.cpp
SYCL/Basic/image/image_accessor_readsampler.cpp
SYCL/Basic/query_emulate_subdevice.cpp
SYCL/Basic/event_profiling_info.cpp
SYCL/Basic/kernel_bundle/kernel_bundle_api.cpp
SYCL/Basic/intel-ext-device.cpp
SYCL/Basic/handler/handler_mem_op.cpp
SYCL/Basic/free_function_queries/free_function_queries_sub_group.cpp
SYCL/Basic/built-ins.cpp
SYCL/SubGroup/scan_spirv13_fp16.cpp
SYCL/SubGroup/reduce_spirv13_fp16.cpp
SYCL/SubGroup/scan_spirv13_fp64.cpp
SYCL/SubGroup/reduce_spirv13_fp64.cpp
SYCL/Config/kernel_from_file.cpp
SYCL/DeviceCodeSplit/aot-gpu.cpp
SYCL/Regression/implicit_kernel_bundle_image_filtering.cpp
SYCL/KernelAndProgram/cache-build-result.cpp
SYCL/KernelAndProgram/undefined-symbol.cpp
SYCL/KernelAndProgram/build-log.cpp
SYCL/InvokeSimd/invoke_simd_conv.cpp
SYCL/InvokeSimd/invoke_simd_smoke.cpp

Here is an updated list of the tests marked as XFAIL and UNSUPPORTED for CUDA,
which have been enabled, ran, and resulted in failures:

[+] XFAIL

N/A (All tests marked as XFAIL, which have been enabled, ran, and failed)

[+] UNSUPPORTED
AOT/gpu.cpp
BFloat16/bfloat16_type.cpp
Basic/code_location_e2e.cpp
Basic/image/image_accessor_readsampler.cpp
Basic/image/image_accessor_readwrite.cpp
Basic/image/image_accessor_readwrite_half.cpp
Basic/image/image_max_size.cpp
Basic/image/srgba-read.cpp
Config/kernel_from_file.cpp
DeviceCodeSplit/aot-gpu.cpp
DeviceLib/imf_fp32_test.cpp
DeviceLib/imf_fp64_test.cpp
DeviceLib/imf_simd_emulate_test.cpp
DiscardEvents/discard_events_using_assert.cpp
ESIMD/BitonicSortK.cpp
ESIMD/BitonicSortKv2.cpp
ESIMD/Prefix_Local_sum1.cpp
ESIMD/Prefix_Local_sum2.cpp
ESIMD/Prefix_Local_sum3.cpp
ESIMD/Stencil.cpp
ESIMD/acc_gather_scatter_rgba.cpp
ESIMD/acc_gather_scatter_rgba_stateless.cpp
ESIMD/accessor.cpp
ESIMD/accessor_gather_scatter.cpp
ESIMD/accessor_gather_scatter_stateless.cpp
ESIMD/accessor_load_store.cpp
ESIMD/accessor_load_store_stateless.cpp
ESIMD/accessor_stateless.cpp
ESIMD/aot_mixed.cpp
ESIMD/api/ballot.cpp
ESIMD/api/bin_and_cmp_ops_heavy.cpp
ESIMD/api/esimd_bit_ops.cpp
ESIMD/api/esimd_merge.cpp
ESIMD/api/esimd_pack_unpack_mask.cpp
ESIMD/api/replicate_smoke.cpp
ESIMD/api/saturation_smoke.cpp
ESIMD/api/simd_any_all.cpp
ESIMD/api/simd_binop_integer_promotion.cpp
ESIMD/api/simd_copy_to_from.cpp
ESIMD/api/simd_mask.cpp
ESIMD/api/simd_memory_access.cpp
ESIMD/api/simd_negation_operator.cpp
ESIMD/api/simd_view_copy_move_assign.cpp
ESIMD/api/simd_view_negation_operator.cpp
ESIMD/api/simd_view_select_2d_fp.cpp
ESIMD/api/simd_view_select_2d_int.cpp
ESIMD/api/slm_gather_scatter_heavy.cpp
ESIMD/api/svm_gather_scatter.cpp
ESIMD/api/unary_ops_heavy.cpp
ESIMD/dword_atomic_cmpxchg.cpp
ESIMD/dword_atomic_smoke.cpp
ESIMD/esimd_check_vc_codegen.cpp
ESIMD/ext_math.cpp
ESIMD/fp_args_size/fp_args_char_int_size_192.cpp
ESIMD/fp_args_size/fp_args_char_int_size_256.cpp
ESIMD/fp_args_size/fp_args_char_int_size_512.cpp
ESIMD/fp_args_size/fp_args_char_int_size_64.cpp
ESIMD/fp_args_size/fp_args_char_int_size_96.cpp
ESIMD/fp_args_size/fp_args_int_size_192.cpp
ESIMD/fp_args_size/fp_args_int_size_256.cpp
ESIMD/fp_args_size/fp_args_int_size_512.cpp
ESIMD/fp_args_size/fp_args_int_size_64.cpp
ESIMD/fp_args_size/fp_args_int_size_96.cpp
ESIMD/fp_in_phi.cpp
ESIMD/histogram.cpp
ESIMD/histogram_256_slm.cpp
ESIMD/histogram_256_slm_spec_2020.cpp
ESIMD/histogram_2d.cpp
ESIMD/kmeans/kmeans.cpp
ESIMD/linear/linear.cpp
ESIMD/mandelbrot/mandelbrot.cpp
ESIMD/mandelbrot/mandelbrot_spec.cpp
ESIMD/matrix_transpose_glb.cpp
ESIMD/matrix_transpose_usm.cpp
ESIMD/noinline_args_size/noinline_args_char_int_size_192.cpp
ESIMD/noinline_args_size/noinline_args_char_int_size_256.cpp
ESIMD/noinline_args_size/noinline_args_char_int_size_512.cpp
ESIMD/noinline_args_size/noinline_args_char_int_size_64.cpp
ESIMD/noinline_args_size/noinline_args_char_int_size_96.cpp
ESIMD/noinline_args_size/noinline_args_int_size_192.cpp
ESIMD/noinline_args_size/noinline_args_int_size_256.cpp
ESIMD/noinline_args_size/noinline_args_int_size_512.cpp
ESIMD/noinline_args_size/noinline_args_int_size_64.cpp
ESIMD/noinline_args_size/noinline_args_int_size_96.cpp
ESIMD/preemption.cpp
ESIMD/printf.cpp
ESIMD/radix_sort.cpp
ESIMD/reduction.cpp
ESIMD/regression/abs_fix_test.cpp
ESIMD/regression/big_const_initializer.cpp
ESIMD/regression/bit_shift_compilation_test.cpp
ESIMD/regression/complex-lib-lin.cpp
ESIMD/regression/copyto_char_test.cpp
ESIMD/regression/dgetrf.cpp
ESIMD/regression/dgetrf_8x8.cpp
ESIMD/regression/dgetrf_ref.cpp
ESIMD/regression/fmod_compatibility_test.cpp
ESIMD/regression/globals.cpp
ESIMD/regression/half_conversion_test.cpp
ESIMD/regression/operator_decrement.cpp
ESIMD/regression/replicate_bug.cpp
ESIMD/regression/store_zero_const.cpp
ESIMD/regression/sycl_esimd_mixed_unnamed.cpp
ESIMD/regression/unused_load.cpp
ESIMD/regression/usm_gather_scatter_32.cpp
ESIMD/regression/variable_gather_mask.cpp
ESIMD/slm_barrier.cpp
ESIMD/slm_split_barrier.cpp
ESIMD/spec_const/spec_const_bool.cpp
ESIMD/spec_const/spec_const_char.cpp
ESIMD/spec_const/spec_const_double.cpp
ESIMD/spec_const/spec_const_float.cpp
ESIMD/spec_const/spec_const_int.cpp
ESIMD/spec_const/spec_const_int64.cpp
ESIMD/spec_const/spec_const_short.cpp
ESIMD/spec_const/spec_const_uchar.cpp
ESIMD/spec_const/spec_const_uint.cpp
ESIMD/spec_const/spec_const_uint64.cpp
ESIMD/spec_const/spec_const_ushort.cpp
ESIMD/spec_const/unused_spec_const.cpp
ESIMD/stencil2.cpp
ESIMD/sycl_esimd_mix.cpp
ESIMD/test_id_3d.cpp
ESIMD/usm_gather_scatter_rgba.cpp
ESIMD/vadd_1d.cpp
ESIMD/vadd_2d.cpp
ESIMD/vadd_2d_acc.cpp
ESIMD/vadd_half.cpp
ESIMD/vadd_usm.cpp
ESIMD/vadd_usm_opqptr.cpp
InvokeSimd/invoke_simd_conv.cpp
InvokeSimd/invoke_simd_smoke.cpp
KernelAndProgram/build-log.cpp
KernelAndProgram/cache-build-result.cpp
KernelAndProgram/undefined-symbol.cpp
Matrix/element_wise_all_ops_cuda.cpp
Matrix/joint_matrix_tensorcore.cpp
Plugin/sycl-ls-gpu-default-any.cpp
Regression/fsycl-host-compiler.cpp
SeparateCompile/test.cpp
SubGroup/reduce_spirv13_fp16.cpp
SubGroup/scan_spirv13_fp16.cpp

Finally, the following list is the tests marked as UNSUPPORTED, which have been
enabled, ran, and resulted in success:

SYCL/Assert/assert_in_simultaneous_kernels.cpp
SYCL/Assert/assert_in_simultaneously_multiple_tus.cpp
SYCL/Assert/assert_in_simultaneously_multiple_tus_one_ndebug.cpp
SYCL/Basic/built-ins.cpp
SYCL/Basic/event_profiling_info.cpp
SYCL/Basic/handler/handler_mem_op.cpp
SYCL/Basic/image/image_write.cpp
SYCL/Basic/image/image_write_fp16.cpp
SYCL/Basic/intel-ext-device.cpp
SYCL/Basic/kernel_bundle/kernel_bundle_api.cpp
SYCL/Basic/query_emulate_subdevice.cpp
SYCL/DeprecatedFeatures/deprecated_intel_ext_device.cpp
SYCL/DeviceLib/ITTAnnotations/atomic.cpp
SYCL/DeviceLib/ITTAnnotations/barrier.cpp
SYCL/ESIMD/api/esimd_rgba_smoke.cpp
SYCL/ESIMD/api/simd_subscript_operator.cpp
SYCL/ESIMD/api/simd_view_subscript_operator.cpp
SYCL/ESIMD/api/slm_gather_scatter.cpp
SYCL/ESIMD/api/slm_gather_scatter_rgba.cpp
SYCL/ESIMD/PrefixSum.cpp
SYCL/ESIMD/vec_arg_call_conv_ext.cpp
SYCL/ESIMD/vec_arg_call_conv_smoke.cpp
SYCL/InlineAsm/asm_16_empty.cpp
SYCL/InlineAsm/asm_16_matrix_mult.cpp
SYCL/InlineAsm/asm_16_no_input_int.cpp
SYCL/InlineAsm/asm_16_no_opts.cpp
SYCL/InlineAsm/asm_8_empty.cpp
SYCL/InlineAsm/asm_8_no_input_int.cpp
SYCL/InlineAsm/asm_arbitrary_ops_order.cpp
SYCL/InlineAsm/asm_decl_in_scope.cpp
SYCL/InlineAsm/asm_float_add.cpp
SYCL/InlineAsm/asm_float_imm_arg.cpp
SYCL/InlineAsm/asm_float_neg.cpp
SYCL/InlineAsm/asm_if.cpp
SYCL/InlineAsm/asm_imm_arg.cpp
SYCL/InlineAsm/asm_loop.cpp
SYCL/InlineAsm/asm_mul.cpp
SYCL/InlineAsm/asm_multiple_instructions.cpp (SHOULD HAVE FAILED, REQUIRES GPU; XFAIL GPU)
SYCL/InlineAsm/asm_no_operands.cpp
SYCL/InlineAsm/asm_no_output.cpp
SYCL/InlineAsm/asm_plus_mod.cpp
SYCL/InlineAsm/asm_switch.cpp
SYCL/InlineAsm/letter_example.cpp
SYCL/InlineAsm/malloc_shared_32.cpp
SYCL/InlineAsm/malloc_shared_in_out_dif.cpp
SYCL/InlineAsm/malloc_shared_no_input.cpp
SYCL/InlineAsm/Negative/asm_bad_opcode.cpp
SYCL/InlineAsm/Negative/asm_bad_operand_syntax.cpp
SYCL/InlineAsm/Negative/asm_duplicate_label.cpp
SYCL/InlineAsm/Negative/asm_illegal_exec_size.cpp
SYCL/InlineAsm/Negative/asm_missing_label.cpp
SYCL/InlineAsm/Negative/asm_missing_region.cpp
SYCL/InlineAsm/Negative/asm_simple.cpp
SYCL/InlineAsm/Negative/asm_undefined_decl.cpp
SYCL/InlineAsm/Negative/asm_undefined_pred.cpp
SYCL/InlineAsm/Negative/asm_wrong_declare.cpp
SYCL/Plugin/sycl-ls.cpp
SYCL/Regression/implicit_kernel_bundle_image_filtering.cpp
SYCL/SubGroup/reduce_spirv13_fp64.cpp
SYCL/SubGroup/scan_spirv13_fp64.cpp
SYCL/USM/dep_events.cpp
SYCL/USM/prefetch.cpp

@jle-quel jle-quel mentioned this issue Oct 11, 2022
Merged
@AerialMantis
Copy link

We've re-evaluated all the current XFAIL and UNSUPPORTED tests for the CUDA backend, tests which were marked as UNSUPPORTED but were actually passing have been re-enabled in #1317 and some tests which were marked as XFAIL which should have been UNSUPPORTED have been updated in #1340.

We are investigate all remaining XFAIL and UNSUPPORTED, so we will close this ticket now.

steffenlarsen pushed a commit that referenced this issue Feb 27, 2023
@jle-quel
[SYCL][CUDA] Re-enable tests that were marked as UNSUPPORTED for the …
281559a 
…CUDA backend (#1317)

This PR re-enables the test marked as UNSUPPORTED for the CUDA backend.

As a result of these tests ran to update this issue: #249; Specifically, the part which mentions the tests that passed but are marked as unsupported;

I'm proposing this PR to re-enable them.
myler pushed a commit to myler/llvm-test-suite that referenced this issue Mar 22, 2023
@t4c1 @bb-sycl
enable working tests (intel#1059)
aa64cd1 
Enables tests that were marked with UNSUPPORTED: cuda, but are in fact passing.

Result of work on intel#249
aelovikov-intel pushed a commit to aelovikov-intel/llvm that referenced this issue Mar 27, 2023
@jle-quel
[SYCL][CUDA] Re-enable tests that were marked as UNSUPPORTED for the …
7037605 
…CUDA backend (intel/llvm-test-suite#1317)

This PR re-enables the test marked as UNSUPPORTED for the CUDA backend.

As a result of these tests ran to update this issue: intel/llvm-test-suite#249; Specifically, the part which mentions the tests that passed but are marked as unsupported;

I'm proposing this PR to re-enable them.
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Assignees
No one assigned
Labels
compiler Compiler related issue cuda Issues related to execution on CUDA backend. runtime Runtime library related issue
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
7 participants
@bader @t4c1 @AerialMantis @jle-quel @vmaksimo @pvchupin @steffenlarsen