pytorch - ✅(Solved) Fix [Dynamo][Aot Compile] Builtin causes failure in AOT Compile [1 pull requests, 1 comments, 1 participants]

pytorch2026-03-16 18:39:50

ON THIS PAGE

Recommended Tools

×6

Utilities matched from this issue’s tags and category — try them while you read without losing context.

GitHub issue graph ai analysis

Paste a GitHub issue URL. We fetch that issue, discover linked issues from bodies/comments/timeline, collect linked pull requests, and produce a structured English report.

The report is written in English Markdown for sharing and archival.

GitHub issue URL

Helpful · Quick feedback

GitHub stats

pytorch/pytorch#177556•Fetched 2026-04-08 00:47:30

View on GitHub

Comments

Participants

Timeline

Reactions

Author

Lucaskabela

Participants

Lucaskabela

Assignees

Lucaskabela

Timeline (top)

mentioned ×32subscribed ×32labeled ×5referenced ×3

Error Message

File "/home/lucaskabela/pytorch/repo.py", line 15, in <module> loaded = AOTCompiledFunction.deserialize(data.serialized_data) # RuntimeError! ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/home/lucaskabela/pytorch/torch/_dynamo/aot_compile.py", line 304, in deserialize return cls(artifacts, _extra_globals=f_globals) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "<string>", line 6, in init File "/home/lucaskabela/pytorch/torch/_dynamo/aot_compile.py", line 220, in post_init self.fn = self._artifacts.runtime_env.forward_callable( ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/home/lucaskabela/pytorch/torch/_dynamo/convert_frame.py", line 981, in forward_callable self._check_external_refs(f_globals) File "/home/lucaskabela/pytorch/torch/_dynamo/convert_frame.py", line 1001, in _check_external_refs raise RuntimeError( RuntimeError: Missing required external references: ['__builtins_dict___0']. Please load AOT compiled function with f_globals=<enclosing global scope>

Root Cause

This is because while __builtins_dict__ is installed in f_globals while tracing, it is not preserved in serialization, so using a builtin that is not folded will result in error when loading from serialization with an error like:

  File "/home/lucaskabela/pytorch/repo.py", line 15, in <module>
    loaded = AOTCompiledFunction.deserialize(data.serialized_data)  # RuntimeError!
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/lucaskabela/pytorch/torch/_dynamo/aot_compile.py", line 304, in deserialize
    return cls(artifacts, _extra_globals=f_globals)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "<string>", line 6, in __init__
  File "/home/lucaskabela/pytorch/torch/_dynamo/aot_compile.py", line 220, in __post_init__
    self.fn = self._artifacts.runtime_env.forward_callable(
              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/lucaskabela/pytorch/torch/_dynamo/convert_frame.py", line 981, in forward_callable
    self._check_external_refs(f_globals)
  File "/home/lucaskabela/pytorch/torch/_dynamo/convert_frame.py", line 1001, in _check_external_refs
    raise RuntimeError(
RuntimeError: Missing required external references: ['__builtins_dict___0']. Please load AOT compiled function with `f_globals=<enclosing global scope>`

Fix Action

Fixed

Fixed by PR: [Bugfix] Keep builtin for serialization when used in dynamo (https://github.com/pytorch/pytorch/pull/177558)

PR fix notes

PR #177558: [Bugfix] Keep builtin for serialization when used in dynamo

Repository: pytorch/pytorch
Author: Lucaskabela
State: closed | merged: False
Link: https://github.com/pytorch/pytorch/pull/177558

Description (problem / solution / changelog)

Fixes https://github.com/pytorch/pytorch/issues/177556

Summary

When we have a builtin used in code dynamo is tracing that is not folded, we recieve an error when loading from the serialized function.

This is because the __builtin_dict__ which dynamo uses is not captured; so in this PR, we update to selectively capture/preserve these so that loading succeeds

Testing

python -m pytest test/dynamo/test_aot_compile.py::TestAOTCompile::test_builtins_dict_survives_serialization -xvs

Results in:

test/dynamo/test_aot_compile.py::TestAOTCompile::test_builtins_dict_survives_serialization PASSED [13.3021s]

cc @voznesenskym @penguinwu @EikanWang @jgong5 @Guobing-Chen @XiaobingSuper @zhuhaozhe @blzheng @wenzhe-nrv @jiayisunx @kadeng @chauhang @amjames @jataylo

Changed files

test/dynamo/test_aot_compile.py (modified, +21/-0)
torch/_dynamo/convert_frame.py (modified, +25/-0)

Code Example

import torch
from torch._dynamo.aot_compile import AOTCompiledFunction

torch._dynamo.config.enable_aot_compile = True

def fn(x):
    return x + 1, type

x = torch.randn(4)
compiled = torch.compile(fn, fullgraph=True, backend="inductor")
aot_fn = compiled.aot_compile(((x,), {}))

# Serialize and deserialize without passing f_globals
data = AOTCompiledFunction.serialize(aot_fn)
loaded = AOTCompiledFunction.deserialize(data.serialized_data)  # RuntimeError!
loaded_out = loaded(x)
fn_out = fn(x)
assert loaded_out[0].allclose(fn_out[0])
assert loaded_out[1] == fn_out[1]

---

File "/home/lucaskabela/pytorch/repo.py", line 15, in <module>
    loaded = AOTCompiledFunction.deserialize(data.serialized_data)  # RuntimeError!
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/lucaskabela/pytorch/torch/_dynamo/aot_compile.py", line 304, in deserialize
    return cls(artifacts, _extra_globals=f_globals)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "<string>", line 6, in __init__
  File "/home/lucaskabela/pytorch/torch/_dynamo/aot_compile.py", line 220, in __post_init__
    self.fn = self._artifacts.runtime_env.forward_callable(
              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/lucaskabela/pytorch/torch/_dynamo/convert_frame.py", line 981, in forward_callable
    self._check_external_refs(f_globals)
  File "/home/lucaskabela/pytorch/torch/_dynamo/convert_frame.py", line 1001, in _check_external_refs
    raise RuntimeError(
RuntimeError: Missing required external references: ['__builtins_dict___0']. Please load AOT compiled function with `f_globals=<enclosing global scope>`

RAW_BUFFERClick to expand / collapse

🐛 Describe the bug

Given the following code, serialization will fail:

import torch
from torch._dynamo.aot_compile import AOTCompiledFunction

torch._dynamo.config.enable_aot_compile = True

def fn(x):
    return x + 1, type

x = torch.randn(4)
compiled = torch.compile(fn, fullgraph=True, backend="inductor")
aot_fn = compiled.aot_compile(((x,), {}))

# Serialize and deserialize without passing f_globals
data = AOTCompiledFunction.serialize(aot_fn)
loaded = AOTCompiledFunction.deserialize(data.serialized_data)  # RuntimeError!
loaded_out = loaded(x)
fn_out = fn(x)
assert loaded_out[0].allclose(fn_out[0])
assert loaded_out[1] == fn_out[1]

  File "/home/lucaskabela/pytorch/repo.py", line 15, in <module>
    loaded = AOTCompiledFunction.deserialize(data.serialized_data)  # RuntimeError!
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/lucaskabela/pytorch/torch/_dynamo/aot_compile.py", line 304, in deserialize
    return cls(artifacts, _extra_globals=f_globals)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "<string>", line 6, in __init__
  File "/home/lucaskabela/pytorch/torch/_dynamo/aot_compile.py", line 220, in __post_init__
    self.fn = self._artifacts.runtime_env.forward_callable(
              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/lucaskabela/pytorch/torch/_dynamo/convert_frame.py", line 981, in forward_callable
    self._check_external_refs(f_globals)
  File "/home/lucaskabela/pytorch/torch/_dynamo/convert_frame.py", line 1001, in _check_external_refs
    raise RuntimeError(
RuntimeError: Missing required external references: ['__builtins_dict___0']. Please load AOT compiled function with `f_globals=<enclosing global scope>`

Versions

Collecting environment information... PyTorch version: 2.12.0a0+gita01976a Is debug build: False CUDA used to build PyTorch: 12.9 ROCM used to build PyTorch: N/A

OS: CentOS Stream 9 (x86_64) GCC version: (GCC) 11.5.0 20240719 (Red Hat 11.5.0-14) Clang version: 21.1.7 (CentOS 21.1.7-1.el9) CMake version: version 4.2.1 Libc version: glibc-2.34

Python version: 3.12.12 | packaged by Anaconda, Inc. | (main, Oct 21 2025, 20:16:04) [GCC 11.2.0] (64-bit runtime) Python platform: Linux-6.13.2-0_fbk8_0_g8695f611147d-x86_64-with-glibc2.34 Is CUDA available: True CUDA runtime version: 12.9.86 CUDA_MODULE_LOADING set to: GPU models and configuration: GPU 0: NVIDIA H100 GPU 1: NVIDIA H100 GPU 2: NVIDIA H100 GPU 3: NVIDIA H100 GPU 4: NVIDIA H100 GPU 5: NVIDIA H100 GPU 6: NVIDIA H100 GPU 7: NVIDIA H100

Nvidia driver version: 580.82.07 cuDNN version: Could not collect Is XPU available: False HIP runtime version: N/A MIOpen runtime version: N/A Is XNNPACK available: True Caching allocator config: N/A

CPU: Architecture: x86_64 CPU op-mode(s): 32-bit, 64-bit Address sizes: 52 bits physical, 57 bits virtual Byte Order: Little Endian CPU(s): 384 On-line CPU(s) list: 0-383 Vendor ID: AuthenticAMD Model name: AMD EPYC 9654 96-Core Processor CPU family: 25 Model: 17 Thread(s) per core: 2 Core(s) per socket: 96 Socket(s): 2 Stepping: 1 Frequency boost: enabled CPU(s) scaling MHz: 100% CPU max MHz: 2400.0000 CPU min MHz: 1500.0000 BogoMIPS: 4792.60 Flags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc rep_good amd_lbr_v2 nopl xtopology nonstop_tsc cpuid extd_apicid aperfmperf rapl pni pclmulqdq monitor ssse3 fma cx16 pcid sse4_1 sse4_2 x2apic movbe popcnt aes xsave avx f16c rdrand lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt tce topoext perfctr_core perfctr_nb bpext perfctr_llc mwaitx cpb cat_l3 cdp_l3 hw_pstate ssbd mba perfmon_v2 ibrs ibpb stibp ibrs_enhanced vmmcall fsgsbase bmi1 avx2 smep bmi2 erms invpcid cqm rdt_a avx512f avx512dq rdseed adx smap avx512ifma clflushopt clwb avx512cd sha_ni avx512bw avx512vl xsaveopt xsavec xgetbv1 xsaves cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local avx512_bf16 clzero irperf xsaveerptr rdpru wbnoinvd amd_ppin cppc arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold avic v_vmsave_vmload vgif x2avic v_spec_ctrl vnmi avx512vbmi umip pku ospke avx512_vbmi2 gfni vaes vpclmulqdq avx512_vnni avx512_bitalg avx512_vpopcntdq rdpid overflow_recov succor smca fsrm flush_l1d debug_swap Virtualization: AMD-V L1d cache: 6 MiB (192 instances) L1i cache: 6 MiB (192 instances) L2 cache: 192 MiB (192 instances) L3 cache: 768 MiB (24 instances) NUMA node(s): 2 NUMA node0 CPU(s): 0-95,192-287 NUMA node1 CPU(s): 96-191,288-383 Vulnerability Gather data sampling: Not affected Vulnerability Itlb multihit: Not affected Vulnerability L1tf: Not affected Vulnerability Mds: Not affected Vulnerability Meltdown: Not affected Vulnerability Mmio stale data: Not affected Vulnerability Reg file data sampling: Not affected Vulnerability Retbleed: Not affected Vulnerability Spec rstack overflow: Vulnerable Vulnerability Spec store bypass: Vulnerable Vulnerability Spectre v1: Vulnerable: __user pointer sanitization and usercopy barriers only; no swapgs barriers Vulnerability Spectre v2: Vulnerable; IBPB: disabled; STIBP: disabled; PBRSB-eIBRS: Not affected; BHI: Not affected Vulnerability Srbds: Not affected Vulnerability Tsx async abort: Not affected

Versions of relevant libraries: [pip3] flake8==7.3.0 [pip3] mypy==1.19.1 [pip3] mypy_extensions==1.1.0 [pip3] numpy==2.2.6 [pip3] nvidia-cublas-cu12==12.8.4.1 [pip3] nvidia-cuda-cupti-cu12==12.8.90 [pip3] nvidia-cuda-nvrtc-cu12==12.8.93 [pip3] nvidia-cuda-runtime-cu12==12.8.90 [pip3] nvidia-cudnn-cu12==9.10.2.21 [pip3] nvidia-cudnn-frontend==1.17.0 [pip3] nvidia-cufft-cu12==11.3.3.83 [pip3] nvidia-curand-cu12==10.3.9.90 [pip3] nvidia-cusolver-cu12==11.7.3.90 [pip3] nvidia-cusparse-cu12==12.5.8.93 [pip3] nvidia-cusparselt-cu12==0.7.1 [pip3] nvidia-nccl-cu12==2.27.5 [pip3] nvidia-nvjitlink-cu12==12.8.93 [pip3] nvidia-nvtx-cu12==12.8.90 [pip3] optree==0.18.0 [pip3] torch==2.12.0a0+gita01976a [pip3] torch_c_dlpack_ext==0.1.5 [pip3] torchdata==0.11.0 [pip3] torchmonarch==0.3.0 [pip3] torchtitan==0.2.2 [pip3] torchvision==0.26.0a0+4729419 [pip3] torchx-nightly==2026.1.26 [pip3] triton==3.5.1 [conda] numpy 2.2.6 pypi_0 pypi [conda] nvidia-cublas-cu12 12.8.4.1 pypi_0 pypi [conda] nvidia-cuda-cupti-cu12 12.8.90 pypi_0 pypi [conda] nvidia-cuda-nvrtc-cu12 12.8.93 pypi_0 pypi [conda] nvidia-cuda-runtime-cu12 12.8.90 pypi_0 pypi [conda] nvidia-cudnn-cu12 9.10.2.21 pypi_0 pypi [conda] nvidia-cudnn-frontend 1.17.0 pypi_0 pypi [conda] nvidia-cufft-cu12 11.3.3.83 pypi_0 pypi [conda] nvidia-curand-cu12 10.3.9.90 pypi_0 pypi [conda] nvidia-cusolver-cu12 11.7.3.90 pypi_0 pypi [conda] nvidia-cusparse-cu12 12.5.8.93 pypi_0 pypi [conda] nvidia-cusparselt-cu12 0.7.1 pypi_0 pypi [conda] nvidia-nccl-cu12 2.27.5 pypi_0 pypi [conda] nvidia-nvjitlink-cu12 12.8.93 pypi_0 pypi [conda] nvidia-nvtx-cu12 12.8.90 pypi_0 pypi [conda] optree 0.18.0 pypi_0 pypi [conda] torch 2.12.0a0+gita01976a pypi_0 pypi [conda] torch-c-dlpack-ext 0.1.5 pypi_0 pypi [conda] torchdata 0.11.0 pypi_0 pypi [conda] torchmonarch 0.3.0 pypi_0 pypi [conda] torchtitan 0.2.2 pypi_0 pypi [conda] torchvision 0.26.0a0+4729419 pypi_0 pypi [conda] torchx-nightly 2026.1.26 pypi_0 pypi [conda] triton 3.5.1 pypi_0 pypi

cc @mruberry @mikaylagawarecki @voznesenskym @penguinwu @EikanWang @jgong5 @Guobing-Chen @XiaobingSuper @zhuhaozhe @blzheng @wenzhe-nrv @jiayisunx @kadeng @chauhang @amjames @jataylo

extent analysis

Fix Plan

To fix the serialization issue, you need to pass the f_globals when deserializing the AOTCompiledFunction.

Here are the steps:

When serializing the AOTCompiledFunction, also store the f_globals used during compilation.
When deserializing, pass the stored f_globals to the deserialize method.

Example code:

import torch
from torch._dynamo.aot_compile import AOTCompiledFunction

torch._dynamo.config.enable_aot_compile = True

def fn(x):
    return x + 1, type

x = torch.randn(4)
compiled = torch.compile(fn, fullgraph=True, backend="inductor")
aot_fn = compiled.aot_compile(((x,), {}))

# Serialize and deserialize with f_globals
data = AOTCompiledFunction.serialize(aot_fn)
f_globals = globals()  # Store the f_globals
loaded = AOTCompiledFunction.deserialize(data.serialized_data, f_globals=f_globals)

loaded_out = loaded(x)
fn_out = fn(x)
assert loaded_out[0].allclose(fn_out[0])
assert loaded_out[1] == fn_out[1]

Verification

After applying the fix, verify that the deserialization works correctly and the output of the original function and the deserialized function are the same.

Extra Tips

Make sure to store the correct f_globals during serialization.
When deserializing, use the same f_globals that were used during serialization.
This fix assumes that the f_globals are available during deserialization. If the f_globals are not available, you may need to modify the serialization process to store the required information.

Vote matrix · Quick signals

Works

Did the solution work? Tap to confirm.

Easy Fix

Was it a quick fix?

Time Saver

Did it save you time?

Blocking

Was it severely blocking?

Common Issue

Are others likely hitting this too?

Flaky / Intermittent

Is it intermittent?

Verified / Reproducible

Can you reproduce it reliably?

#api #ssr #installation #tensor shape #autograd error #serialization error #model compatibility #GPU setup #container setup

Still need to ship something?

×6

Another batch ranked right after the header list — different links, same matching logic.

Data

Security

Network

Code

UI/UX

Text

System

Multimedia

Protocol

API

Engineering

pytorch - ✅(Solved) Fix [Dynamo][Aot Compile] Builtin causes failure in AOT Compile [1 pull requests, 1 comments, 1 participants]

Recommended Tools

GitHub issue graph ai analysis

Error Message

Root Cause

Fix Action

Fixed

PR fix notes

PR #177558: [Bugfix] Keep builtin for serialization when used in dynamo

Description (problem / solution / changelog)

Summary

Testing

Changed files

Code Example

🐛 Describe the bug

Versions

extent analysis

Fix Plan

Verification

Extra Tips

Still need to ship something?

TRENDING

pytorch - ✅(Solved) Fix [Dynamo][Aot Compile] Builtin causes failure in AOT Compile [1 pull requests, 1 comments, 1 participants]

Recommended Tools

GitHub issue graph ai analysis

Error Message

Root Cause

Fix Action

Fixed

PR fix notes

PR #177558: [Bugfix] Keep builtin for serialization when used in dynamo

Description (problem / solution / changelog)

Summary

Testing

Changed files

Code Example

🐛 Describe the bug

Versions

extent analysis

Fix Plan

Verification

Extra Tips

Still need to ship something?

RELATED_DISCOVERY

TRENDING