Closed pytorch-bot[bot] closed 1 month ago
Another case of trunk flakiness has been found here. The list of platforms [linux] appears to contain all the recently affected platforms [linux]. Either the change didn't propogate fast enough or disable bot might be broken.
Another case of trunk flakiness has been found here. The list of platforms [linux] appears to contain all the recently affected platforms [linux]. Either the change didn't propogate fast enough or disable bot might be broken.
Resolving the issue because the test is not flaky anymore after 1005 reruns without any failures and the issue hasn't been updated in 14 days. Please reopen the issue to re-disable the test if you think this is a false positive
Platforms: linux
This test was disabled because it is failing in CI. See recent examples and the most recent trunk workflow logs.
Over the past 3 hours, it has been determined flaky in 4 workflow(s) with 4 failures and 4 successes.
Debugging instructions (after clicking on the recent samples link): DO NOT ASSUME THINGS ARE OKAY IF THE CI IS GREEN. We now shield flaky tests from developers so CI will thus be green but it will be harder to parse the logs. To find relevant log snippets:
test_retrace_export_while_loop_simple_cpu_float32
Sample error message
Truncated for length ``` gs = (actual_args, args[1]) else: tokens = args[: len(meta.tokens)] args = args[len(meta.tokens) :] assert all(token.numel() == 0 for token in tokens) with disable_above: # Wrap inputs into functional wrappers f_args = pytree.tree_map(to_fun, args) f_tokens = pytree.tree_map(to_fun, tokens) # Populate the current FunctionalTensorMode with the tokens per # operator. See Note [FunctionalTensorMode is Stateful] functional_tensor_mode = torch.utils._python_dispatch._detect_infra_mode( torch._C._TorchDispatchModeKey.FUNCTIONAL ) assert functional_tensor_mode is not None for i, k in enumerate(meta.tokens.keys()): functional_tensor_mode._tokens[k] = f_tokens[i] # Run the joint > f_outs = fn(*f_args) aot_config = AOTConfig(fw_compiler=None, bw_compiler=None, partition_fn=None, decompositions={}, num_params_buffers=0, aot_id=10, keep_inference_input_mutations=False, is_export=True, no_tangents=True, dynamic_shapes=True, aot_autograd_arg_pos_to_source=None, static_input_indices=None, inference_compiler=None, enable_log=True, pre_dispatch=True, cache_key=None) args = (FakeTensor(..., size=(), dtype=torch.int64), FakeTensor(..., size=(2, 3, 4))) disable_above =Test file path:
export/test_hop.py
cc @clee2000 @avikchaudhuri @gmagogsfm @zhxchen17 @tugsbayasgalan @angelayi @suo @ydwu4 @penguinwu