jamesjwu

loading model: 0it [00:00, ?it/s]
loading model: 0it [00:03, ?it/s]
ERROR:common:Backend dynamo failed in warmup()
Traceback (most recent call last):
  File "/data/users/jjwu/a/pytorch/benchmarks/dynamo/common.py", line 2561, in warmup
    fn(model, example_inputs)
  File "/data/users/jjwu/a/pytorch/torch/_dynamo/eval_frame.py", line 804, in compile_wrapper
    return fn(*args, **kwargs)
  File "/data/users/jjwu/a/pytorch/benchmarks/dynamo/torchbench.py", line 469, in forward_and_backward_pass

jamesjwu

import time

import torch
import triton
from torch._dynamo.device_interface import get_interface_for_device
from torch._inductor.runtime.static_cuda_launcher import StaticallyLaunchedCudaKernel

import sys
import os

jamesjwu

import time

import torch
import torch._inductor.config as config
from torch import Tensor
from torch._dynamo.device_interface import get_interface_for_device
from torch._inductor.runtime.static_cuda_launcher import StaticallyLaunchedCudaKernel
from torch._inductor.runtime.triton_compat import tl, triton

# Constants

jamesjwu

{
  "cache_state": "miss",
  "cache_status_detailed": "guard_miss",
  "cache_status_guard_expr": "L['t1']*L['t0'] < 2147483648 and L['t2']*L['t3'] < 2147483648 and 2 <= L['t0'] and 2 <= L['t1'] and 2 <= L['t2'] and 2 <= L['t3']",
  "components": [
    "[f5mdxkfzriesfvzu264hheibpxmhcat7d3an75ymc4edjb5bagj] aot_config: (0, True, False, False, False, [TensorPropertySource(base=LocalSource(local_name='x', is_input=True, dynamism=None, is_derefed_cell_contents=False), prop=<TensorProperty.SIZE: 0>, idx=0), TensorPropertySource(base=LocalSource(local_name='x', is_input=True, dynamism=None, is_derefed_cell_contents=False), prop=<TensorProperty.SIZE: 0>, idx=1), LocalSource(local_name='x', is_input=True, dynamism=None, is_derefed_cell_contents=False), TensorPropertySource(base=LocalSource(local_name='y', is_input=True, dynamism=None, is_derefed_cell_contents=False), prop=<TensorProperty.SIZE: 0>, idx=0), TensorPropertySource(base=LocalSource(local_name='y', is_input=True, dynamism=None, is_derefed_cell_contents=False

jamesjwu

{
  "compilation_metrics_runtime": {
    "compile_id": "0/0",
    "frame_key": null,
    "co_name": null,
    "co_filename": null,
    "co_firstlineno": null,
    "cache_size": null,
    "accumulated_cache_size": null,
    "guard_count": null,

jamesjwu

import torch
import time
import torch._inductor.config as config
from torch import Tensor
from torch._inductor.runtime.triton_compat import tl, triton

sp = time.time()
N = 100
template = """
def nop_kernel_{i}(arg0, arg1):

jamesjwu

Triton compilation failed: triton_poi_fused_0
defb844cc147', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': False, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
    min_elem_per_thread=0
)
@triton.jit
def triton_(in_ptr0, out_ptr0, ynumel, xnumel, YBLOCK : tl.constexpr, XBLOCK : tl.constexpr):
    ynumel = 192
    xnumel = 49
    yoffset = tl.program_id(1) * (tl.program_id(2) + 1) * YBLOCK
    yindex = yoffset + tl.arange(0, YBLOCK)[None, :]

jamesjwu

# If not running interactively, don't do anything
case $- in
    *i*) ;;
      *) return;;
esac

# don't put duplicate lines or lines starting with space in the history.
# See bash(1) for more options
HISTCONTROL=ignoreboth

jamesjwu

# If not running interactively, don't do anything
case $- in
    *i*) ;;
      *) return;;
esac

# don't put duplicate lines or lines starting with space in the history.
# See bash(1) for more options
HISTCONTROL=ignoreboth