10000 Segmentation Fault in torch.lu_unpack() with bfloat16 Tensor · Issue #153232 · pytorch/pytorch · GitHub
[go: up one dir, main page]
Skip to content

Segmentation Fault in torch.lu_unpack() with bfloat16 Tensor #153232

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom
Closed
SilentTester73 opened this issue May 9, 2025 · 3 comments
Closed

Segmentation Fault in torch.lu_unpack() with bfloat16 Tensor #153232

SilentTester73 opened this issue May 9, 2025 · 3 comments
Labels
module: edge cases Adversarial inputs unlikely to occur in practice module: linear algebra Issues related to specialized linear algebra operations in PyTorch; includes matrix multiply matmul triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module

Comments

@SilentTester73
Copy link
SilentTester73 commented May 9, 2025 •
edited by pytorch-bot bot
Loading

🐛 Describe the bug

When calling torch.lu_unpack() with a bfloat16 tensor containing extreme values and an empty pivots tensor, the program crashes with a segmentation fault.

Steps to Reproduce

import torch

# Create a bfloat16 tensor with extreme values
lu_data = torch.tensor([[[-8.2838e+07,  5.8933e-32, -3.4757e+25,  1.1114e+21,  1.6466e+06,
          -9.5569e+29,  1.1391e+21],
         [ 3.3389e+21,  1.5088e-21, -1.3733e-10, -4.9472e-06,  3.8554e+21,
           1.0156e+00, -6.2604e-26],
         [-2.6258e-19,  9.3174e-20,  1.4809e+25, -8.0220e+15, -1.1482e-11,
           3.0264e+19,  3.3499e+22],
         [-1.0695e+09, -1.9173e+31, -1.6516e-09, -1.8507e-34,  5.5320e+27,
          -5.8620e-13,  1.1511e+22]],
], dtype=torch.bfloat16)

# Empty pivots tensor
lu_pivots = torch.tensor([], dtype=torch.int32)

# Call lu_unpack which triggers the segmentation fault
torch.lu_unpack(lu_data, lu_pivots, unpack_data=True, unpack_pivots=True)

Output

2.7.0+cu126
Segmentation fault (core dumped)

Colab link

https://colab.research.google.com/drive/1LbOcnXC878lkB28gnpHKOfY-tk9mEwkj?usp=sharing

Versions

Collecting environment information...
PyTorch version: 2.7.0+cu126
Is debug build: False
CUDA used to build PyTorch: 12.6
ROCM used to build PyTorch: N/A

OS: Ubuntu 24.04.1 LTS (x86_64)
GCC version: (Ubuntu 13.3.0-6ubuntu2~24.04) 13.3.0
Clang version: 18.1.8 (++20240731025043+3b5b5c1ec4a3-1~exp1~20240731145144.92)
CMake version: version 4.0.0
Libc version: glibc-2.39

Python version: 3.12.3 (main, Feb  4 2025, 14:48:35) [GCC 13.3.0] (64-bit runtime)
Python platform: Linux-6.8.0-58-generic-x86_64-with-glibc2.39
Is CUDA available: False
CUDA runtime version: 12.8.93
CUDA_MODULE_LOADING set to: N/A
GPU models and configuration: Could not collect
Nvidia driver version: Could not collect
cuDNN version: Could not collect
HIP runtime version: N/A
MIOpen runtime version: N/A
Is XNNPACK available: True

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 9684X 96-Core Processor
CPU family:                           25
Model:                                17
Thread(s) per core:                   2
Core(s) per socket:                   96
Socket(s):                            2
Stepping:                             2
BogoMIPS:                             5099.98
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 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 user_shstk avx512_bf16 clzero irperf xsaveerptr rdpru wbnoinvd amd_ppin cppc amd_ibpb_ret 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 la57 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:                             2.3 GiB (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:   Mitigation; Safe RET
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; STIBP always-on; RSB filling; PBRSB-eIBRS Not affected; BHI Not affected
Vulnerability Srbds:                  Not affected
Vulnerability Tsx async abort:        Not affected

Versions of relevant libraries:
[pip3] numpy==2.2.5
[pip3] nvidia-cublas-cu12==12.6.4.1
[pip3] nvidia-cuda-cupti-cu12==12.6.80
[pip3] nvidia-cuda-nvrtc-cu12==12.6.77
[pip3] nvidia-cuda-runtime-cu12==12.6.77
[pip3] nvidia-cudnn-cu12==9.5.1.17
[pip3] nvidia-cufft-cu12==11.3.0.4
[pip3] nvidia-curand-cu12==10.3.7.77
[pip3] nvidia-cusolver-cu12==11.7.1.2
[pip3] nvidia-cusparse-cu12==12.5.4.2
[pip3] nvidia-cusparselt-cu12==0.6.3
[pip3] nvidia-nccl-cu12==2.26.2
[pip3] nvidia-nvjitlink-cu12==12.6.85
[pip3] nvidia-nvtx-cu12==12.6.77
[pip3] optree==0.15.0
[pip3] torch==2.7.0
[pip3] triton==3.3.0
[conda] Could not collect

cc @jianyuh @nikitaved @mruberry @walterddr @xwang233 @lezcano
@ngimel ngimel added the module: edge cases Adversarial inputs unlikely to occur in practice label May 9, 2025
@colesbury colesbury added triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module module: linear algebra Issues related to specialized linear algebra operations in PyTorch; includes matrix multiply matmul labels May 13, 2025
@lezcano
Copy link
Collaborator
lezcano commented May 13, 2025

This data does not come from lu_factor, and as per the docs, it's UB, so the crash is expected.
If this data did come from an lu_factor call, please feel free to reopen and add an end-to-end repro using lu_factor.

@lezcano lezcano closed this as completed May 13, 2025
@SilentTester73
Copy link
Author

Hi @lezcano Thanks for the clarification on the expected behavior.

Unpacks the LU decomposition returned by lu_factor() into the P, L, U matrices.

While it states that it "Unpacks the LU decomposition returned by lu_factor()", it doesn't explicitly warn that providing inputs from other sources, can lead to UB.

To prevent future confusion, would it be possible to update the documentation for torch.lu_unpack?

Perhaps a note to clarify that providing arbitrarily constructed tensors, or tensors that do not satisfy the implicit structural requirements from lu_factor, will result in undefined behavior.

This would make the API contract more explicit. Thanks for considering this suggestion!

@lezcano
Copy link
Collaborator
lezcano commented May 13, 2025

This is documented here: https://docs.pytorch.org/docs/stable/notes/numerical_accuracy.html#non-finite-values

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
module: edge cases Adversarial inputs unlikely to occur in practice module: linear algebra Issues related to specialized linear algebra operations in PyTorch; includes matrix multiply matmul triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
4 participants
@colesbury @lezcano @ngimel @SilentTester73
0