name | about | labels |
---|---|---|
Bug Report | Use this template for reporting a bug | kind/bug |
fftwithsize mode = "irfft"时signal_sizes空tuple时ascend报错RuntimeError: Launch kernel failed, name:Default/FFTWithSize-op2
Ascend
/GPU
/CPU
) / 硬件环境:Please delete the backend not involved / 请删除不涉及的后端:
/device ascend/
Software Environment / 软件环境 (Mandatory / 必填):
-- MindSpore version (e.g., 1.7.0.Bxxx) :
-- Python version (e.g., Python 3.7.5) :
-- OS platform and distribution (e.g., Linux Ubuntu 16.04):
-- GCC/Compiler version (if compiled from source):
Excute Mode / 执行模式 (Mandatory / 必填)(PyNative
/Graph
):
Please delete the mode not involved / 请删除不涉及的模式:
/mode pynative
/mode graph
test_p_fftwithsize_irfft
用例执行通过
[ERROR] KERNEL(3565036,fffecb7fe070,python):2024-05-12-13:59:49.913.395 [mindspore/ccsrc/plugin/device/ascend/kernel/acl/acl_kernel_mod.cc:261] Launch] Kernel launch failed, msg: Acl compile and execute failed, op_type_:CustFFTWithSize
----------------------------------------------------
- Ascend Error Message:
----------------------------------------------------
E19014: 2024-05-12-13:59:49.909.944 Value [input __input0 shape] for Op [online_Node_Output] is invalid. Reason: contains negative or zero dimension.
TraceBack (most recent call last):
online_Node_Output(NetOutput) Verify failed.[FUNC:Verify][FILE:node_utils_ex.cc][LINE:127]
Verifying online_Node_Output failed.[FUNC:InferShapeAndType][FILE:infershape_pass.cc][LINE:132]
Call InferShapeAndType for node:online_Node_Output(NetOutput) failed[FUNC:Infer][FILE:infershape_pass.cc][LINE:120]
process pass InferShapePass on node:online_Node_Output failed, ret:4294967295[FUNC:RunPassesOnNode][FILE:base_pass.cc][LINE:570]
build graph failed, graph id:2, ret:1343242270[FUNC:BuildModelWithGraphId][FILE:ge_generator.cc][LINE:1615]
[Build][SingleOpModel]call ge interface generator.BuildSingleOpModel failed. ge result = 1343242270[FUNC:ReportCallError][FILE:log_inner.cpp][LINE:161]
[Build][Op]Fail to build op model[FUNC:ReportInnerError][FILE:log_inner.cpp][LINE:145]
build op model failed, result = 500002[FUNC:ReportInnerError][FILE:log_inner.cpp][LINE:145]
(Please search "CANN Common Error Analysis" at https://www.mindspore.cn for error code description)
----------------------------------------------------
- C++ Call Stack: (For framework developers)
----------------------------------------------------
mindspore/ccsrc/transform/acl_ir/acl_utils.cc:379 Run
[ERROR] DEVICE(3565036,fffecb7fe070,python):2024-05-12-13:59:49.913.441 [mindspore/ccsrc/plugin/device/ascend/hal/hardware/ge_kernel_executor.cc:951] LaunchKernel] Launch kernel failed, kernel full name: Default/FFTWithSize-op2
FAILED
=================================== FAILURES ===================================
___________________________ test_p_fftwithsize_irfft ___________________________
@Level2
def test_p_fftwithsize_irfft():
mode = "irfft"
onesided_list = random.sample([True, True, False, False], 3)
for i, norm_mode in enumerate(['backward', 'forward', 'ortho']):
# choice 1/2/3 oneside True/False
onesided = onesided_list[i]
signal_ndim = signal_list[i]
# choice complex64/128
x_dtype = dtype_list[i]
shape_rank = np.random.randint(signal_ndim, 8)
x_shape = np.random.randint(2, 11, shape_rank)
x, input_loss, output_loss = get_x(mode, norm_mode, x_dtype, x_shape)
norm = get_norm_(mode, signal_ndim, norm_mode, x_shape, onesided)
cmp_obj = get_cmp_(mode, signal_ndim, onesided)
fact = FFTWithSizeMock(attributes={
"signal_ndim": signal_ndim,
"inverse": True,
"real": True,
"norm": norm_mode,
"onesided": onesided,
'signal_sizes': (),
},
inputs=[x])
info_str = "signal_ndim={},norm_mode={},x_shape={},x_dtype={},cmp_obj={},loss={}".format(
signal_ndim, norm_mode, x_shape, x_dtype, cmp_obj, [input_loss, output_loss])
logger.info(info_str)
# tf not support IRFFT 3d gradien
if signal_ndim == 3:
fact.loss = 1e-3
> fact.forward_cmp(cmp_obj, norm)
../test_fftwithsize.py:343:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
../../share/ops/primitive/fftwithsize_ops.py:173: in forward_cmp
out_mindspore = self.forward_mindspore_impl()
../../share/ops/primitive/fftwithsize_ops.py:168: in forward_mindspore_impl
out = net(self.input_x)
../../share/utils.py:253: in __call__
out = super().__call__(*args, **kwargs)
/home/ci/miniconda3/envs/ci3.9/lib/python3.9/site-packages/mindspore/nn/cell.py:713: in __call__
raise err
/home/ci/miniconda3/envs/ci3.9/lib/python3.9/site-packages/mindspore/nn/cell.py:710: in __call__
_pynative_executor.end_graph(self, output, *args, **kwargs)
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
self = <mindspore.common.api._PyNativeExecutor object at 0xfffe32436b80>
obj = WrapOp<>
output = Tensor(shape=[8, 9, 6, 10, 8, 3, 12], dtype=Float32, value=
[[[[[[[ 0.00000000e+00, 0.00000000e+00, 0.00000000e+00 ....
[ 0.00000000e+00, 0.00000000e+00, 0.00000000e+00 ... 0.00000000e+00, 0.00000000e+00, 0.00000000e+00]]]]]]])
args = (Tensor(shape=[8, 9, 6, 10, 8, 3, 7], dtype=Complex64, value=
[[[[[[[0.848174+1.85267j, -0.303879-0.211966j, 0.290996+...42+0.577635j, 1.68219-1.23387j, -0.836732+0.131673j ... 0.795906-0.647839j, 0.705827+1.48806j, 1.6735+1.1798j]]]]]]]),)
kwargs = {}
def end_graph(self, obj, output, *args, **kwargs):
"""
Clean resources after building forward and backward graph.
Args:
obj (Function/Cell): The function or cell instance.
output (Tensor/tuple/list): Function or cell output object.
args (tuple): Function or cell input arguments.
kwargs (dict): keyword arguments.
Return:
None.
"""
> self._executor.end_graph(obj, output, *args, *(kwargs.values()))
E RuntimeError: Launch kernel failed, name:Default/FFTWithSize-op2
E
E ----------------------------------------------------
E - C++ Call Stack: (For framework developers)
E ----------------------------------------------------
E mindspore/ccsrc/runtime/pynative/op_runner.cc:632 LaunchKernels
/home/ci/miniconda3/envs/ci3.9/lib/python3.9/site-packages/mindspore/common/api.py:1303: RuntimeError
Please assign maintainer to check this issue.
请为此issue分配处理人。
@tanxinglian
此处可能存在不合适展示的内容,页面不予展示。您可通过相关编辑功能自查并修改。
如您确认内容无涉及 不当用语 / 纯广告导流 / 暴力 / 低俗色情 / 侵权 / 盗版 / 虚假 / 无价值内容或违法国家有关法律法规的内容,可点击提交进行申诉,我们将尽快为您处理。
感谢您的提问,您可以评论//mindspore-assistant更快获取帮助:
登录 后才可以发表评论