pytorch - ✅(Solved) Fix torch.compile(dynamic=True) ignores incompatible `out` tensor for `torch.empty` and returns wrong shape [1 pull requests, 8 comments, 3 participants]

pytorch2026-03-26 06:22:07

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#178482•Fetched 2026-04-08 01:30:17

View on GitHub

Comments

Participants

Timeline

Reactions

Author

Participants

Assignees

Timeline (top)

mentioned ×30subscribed ×30labeled ×9commented ×8

torch.empty(..., out=...) behaves inconsistently under torch.compile(dynamic=True): an incompatible out tensor that should raise an error in eager mode is accepted and returned unchanged.

Error Message

import torch

def f(size, out): return torch.empty(size, out=out, dtype=torch.float32)

size = [2, 3] out = torch.empty([1])

Eager: expected to fail

try: print(f(size, out)) except Exception as e: print('eager error:', e)

Compiled: unexpectedly succeeds

cf = torch.compile(f, dynamic=True) print('compiled result shape:', cf(size, out).shape)

Root Cause

torch.empty(..., out=...) behaves inconsistently under torch.compile(dynamic=True): an incompatible out tensor that should raise an error in eager mode is accepted and returned unchanged.

Fix Action

Fix / Workaround

CPU: Architecture: x86_64 CPU op-mode(s): 32-bit, 64-bit Address sizes: 46 bits physical, 57 bits virtual Byte Order: Little Endian CPU(s): 112 On-line CPU(s) list: 0-111 Vendor ID: GenuineIntel Model name: Intel(R) Xeon(R) Gold 6330 CPU @ 2.00GHz CPU family: 6 Model: 106 Thread(s) per core: 2 Core(s) per socket: 28 Socket(s): 2 Stepping: 6 CPU(s) scaling MHz: 28% CPU max MHz: 3100.0000 CPU min MHz: 800.0000 BogoMIPS: 4000.00 Flags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx pdpe1gb rdtscp lm constant_tsc art arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc cpuid aperfmperf pni pclmulqdq dtes64 monitor ds_cpl vmx smx est tm2 ssse3 sdbg fma cx16 xtpr pdcm pcid dca sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand lahf_lm abm 3dnowprefetch cpuid_fault epb cat_l3 invpcid_single intel_ppin ssbd mba ibrs ibpb stibp ibrs_enhanced tpr_shadow flexpriority ept vpid ept_ad fsgsbase tsc_adjust bmi1 avx2 smep bmi2 erms invpcid cqm rdt_a avx512f avx512dq rdseed adx smap avx512ifma clflushopt clwb intel_pt avx512cd sha_ni avx512bw avx512vl xsaveopt xsavec xgetbv1 xsaves cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local split_lock_detect wbnoinvd dtherm ida arat pln pts hwp hwp_act_window hwp_epp hwp_pkg_req vnmi avx512vbmi umip pku ospke avx512_vbmi2 gfni vaes vpclmulqdq avx512_vnni avx512_bitalg tme avx512_vpopcntdq la57 rdpid fsrm md_clear pconfig flush_l1d arch_capabilities Virtualization: VT-x L1d cache: 2.6 MiB (56 instances) L1i cache: 1.8 MiB (56 instances) L2 cache: 70 MiB (56 instances) L3 cache: 84 MiB (2 instances) NUMA node(s): 2 NUMA node0 CPU(s): 0,2,4,6,8,10,12,14,16,18,20,22,24,26,28,30,32,34,36,38,40,42,44,46,48,50,52,54,56,58,60,62,64,66,68,70,72,74,76,78,80,82,84,86,88,90,92,94,96,98,100,102,104,106,108,110 NUMA node1 CPU(s): 1,3,5,7,9,11,13,15,17,19,21,23,25,27,29,31,33,35,37,39,41,43,45,47,49,51,53,55,57,59,61,63,65,67,69,71,73,75,77,79,81,83,85,87,89,91,93,95,97,99,101,103,105,107,109,111 Vulnerability Gather data sampling: Mitigation; Microcode Vulnerability Itlb multihit: Not affected Vulnerability L1tf: Not affected Vulnerability Mds: Not affected Vulnerability Meltdown: Not affected Vulnerability Mmio stale data: Mitigation; Clear CPU buffers; SMT vulnerable Vulnerability Retbleed: Not affected Vulnerability Spec rstack overflow: Not affected Vulnerability Spec store bypass: Mitigation; Speculative Store Bypass disabled via prctl Vulnerability Spectre v1: Mitigation; usercopy/swapgs barriers and __user pointer sanitization Vulnerability Spectre v2: Mitigation; Enhanced / Automatic IBRS, IBPB conditional, RSB filling, PBRSB-eIBRS SW sequence Vulnerability Srbds: Not affected Vulnerability Tsx async abort: Not affected

PR fix notes

PR #178825: fix for uncompatible tensor shape being ignored in torch.compile

Repository: pytorch/pytorch
Author: khushali9
State: open | merged: False
Link: https://github.com/pytorch/pytorch/pull/178825

Description (problem / solution / changelog)

Fixes 178482

Interesting bug, here empty tensor without value works normally, without any error raised.

The py_impl stub for aten.empty.out in init.py was returning out unchanged instead of resizing it, causing torch.compile(dynamic=True) to silently ignore incompatible out= tensor shapes. Fixed by implementing the stub to call out.resize_(size) (matching the C++ kernel) basically.

Here is my output matching current Eager , where without mamory_format it reshapes, and with memory_format it returns error. I do agree in both case we need to raise error sounds much better solution.

Eager mode: result shape: torch.Size([2, 3])

torch.compile(dynamic=True): result shape: torch.Size([2, 3])

Eager mode with memory_format: error: 'memory_format' argument is incompatible with 'out' tensor argument

torch.compile(dynamic=True) with memory_format: error: 'memory_format' argument is incompatible with 'out' tensor argument

cc @voznesenskym @penguinwu @EikanWang @jgong5 @Guobing-Chen @XiaobingSuper @zhuhaozhe @blzheng @wenzhe-nrv @jiayisunx @kadeng @chauhang @amjames @Lucaskabela @jataylo @azahed98

Changed files

test/dynamo/test_repros.py (modified, +43/-0)
torch/_refs/__init__.py (modified, +4/-0)

Code Example

import torch

def f(size, out):
    return torch.empty(size, out=out, dtype=torch.float32)

size = [2, 3]
out = torch.empty([1])

# Eager: expected to fail
try:
    print(f(size, out))
except Exception as e:
    print('eager error:', e)

# Compiled: unexpectedly succeeds
cf = torch.compile(f, dynamic=True)
print('compiled result shape:', cf(size, out).shape)

RAW_BUFFERClick to expand / collapse

🐛 Describe the bug

Summary

torch.empty(..., out=...) behaves inconsistently under torch.compile(dynamic=True): an incompatible out tensor that should raise an error in eager mode is accepted and returned unchanged.

Repro

import torch

def f(size, out):
    return torch.empty(size, out=out, dtype=torch.float32)

size = [2, 3]
out = torch.empty([1])

# Eager: expected to fail
try:
    print(f(size, out))
except Exception as e:
    print('eager error:', e)

# Compiled: unexpectedly succeeds
cf = torch.compile(f, dynamic=True)
print('compiled result shape:', cf(size, out).shape)

Expected

torch.empty should reject an out tensor whose shape/layout is incompatible with the requested size, both in eager and compiled modes.

Actual

Eager mode raises an error.
torch.compile(dynamic=True) returns the provided out tensor with shape [1] instead of raising.

Versions

PyTorch version: 2.10.0+cu128 Is debug build: False CUDA used to build PyTorch: 12.8 ROCM used to build PyTorch: N/A

OS: Ubuntu 24.04.4 LTS (x86_64) GCC version: (Ubuntu 13.3.0-6ubuntu2~24.04.1) 13.3.0 Clang version: Could not collect CMake version: Could not collect Libc version: glibc-2.39

Python version: 3.12.0 | packaged by Anaconda, Inc. | (main, Oct 2 2023, 17:29:18) [GCC 11.2.0] (64-bit runtime) Python platform: Linux-6.5.11-060511-generic-x86_64-with-glibc2.39 Is CUDA available: True CUDA runtime version: Could not collect CUDA_MODULE_LOADING set to: GPU models and configuration: GPU 0: NVIDIA RTX A6000 GPU 1: NVIDIA RTX A6000

Nvidia driver version: 580.65.06 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

Versions of relevant libraries: [pip3] numpy==2.4.1 [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-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==2.10.0 [pip3] triton==3.6.0 [conda] numpy 2.4.1 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-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 2.10.0 pypi_0 pypi [conda] triton 3.6.0 pypi_0 pypi

cc @ezyang @gchanan @kadeng @msaroufim @mruberry @chauhang @penguinwu @bobrenjc93 @aditvenk @laithsakka @voznesenskym @EikanWang @jgong5 @Guobing-Chen @XiaobingSuper @zhuhaozhe @blzheng @wenzhe-nrv @jiayisunx @amjames @Lucaskabela @jataylo

extent analysis

Fix Plan

To address the inconsistent behavior of torch.empty under torch.compile(dynamic=True), we need to ensure that the out tensor is compatible with the requested size.

Here are the steps to fix the issue:

Check the shape and layout of the out tensor before passing it to torch.empty.
Verify that the shape of the out tensor matches the requested size.
If the shapes do not match, create a new tensor with the correct shape instead of using the incompatible out tensor.

Example code:

import torch

def f(size, out):
    # Check if out tensor has compatible shape
    if out.shape != tuple(size):
        # Create a new tensor with the correct shape
        out = torch.empty(size, dtype=torch.float32)
    return torch.empty(size, out=out, dtype=torch.float32)

size = [2, 3]
out = torch.empty([1])

# Eager: expected to fail
try:
    print(f(size, out))
except Exception as e:
    print('eager error:', e)

# Compiled: should now raise an error
cf = torch.compile(f, dynamic=True)
try:
    print('compiled result shape:', cf(size, out).shape)
except Exception as e:
    print('compiled error:', e)

Verification

To verify that the fix worked, run the example code and check that it raises an error in both eager and compiled modes when the out tensor has an incompatible shape.

Extra Tips

Always verify the shape and layout of tensors before performing operations on them.
Use torch.empty with caution and ensure that the out tensor is compatible with the requested size.
Consider adding error checking code to handle incompatible tensors and provide informative error messages.

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 #GPU compatibility #latency issue #model loading #dependency error

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 torch.compile(dynamic=True) ignores incompatible `out` tensor for `torch.empty` and returns wrong shape [1 pull requests, 8 comments, 3 participants]

Recommended Tools

GitHub issue graph ai analysis

Error Message

Eager: expected to fail

Compiled: unexpectedly succeeds

Root Cause

Fix Action

Fix / Workaround

PR fix notes

PR #178825: fix for uncompatible tensor shape being ignored in torch.compile

Description (problem / solution / changelog)

Changed files

Code Example

🐛 Describe the bug

Summary

Repro

Expected

Actual

Versions

extent analysis

Fix Plan

Verification

Extra Tips

Still need to ship something?

TRENDING

pytorch - ✅(Solved) Fix torch.compile(dynamic=True) ignores incompatible `out` tensor for `torch.empty` and returns wrong shape [1 pull requests, 8 comments, 3 participants]

Recommended Tools

GitHub issue graph ai analysis

Error Message

Eager: expected to fail

Compiled: unexpectedly succeeds

Root Cause

Fix Action

Fix / Workaround

PR fix notes

PR #178825: fix for uncompatible tensor shape being ignored in torch.compile

Description (problem / solution / changelog)

Changed files

Code Example

🐛 Describe the bug

Summary

Repro

Expected

Actual

Versions

extent analysis

Fix Plan

Verification

Extra Tips

Still need to ship something?

RELATED_DISCOVERY

TRENDING