search

pytorch / pytorch

Tensors and Dynamic neural networks in Python with strong GPU acceleration
https://pytorch.org
Other
83.56k stars 22.54k forks source link

torch.sspaddmm: self.is_sparse() INTERNAL ASSERT FAILED on torch 2.5.0.dev20240708+cu121 #132028

Open wangzhen0518 opened 3 months ago

wangzhen0518 commented 3 months ago

🐛 Describe the bug

I met an error RuntimeError self.is_sparse() INTERNAL ASSERT FAILED at "../aten/src/ATen/native/SparseTensorUtils.h":28, please report a bug to PyTorch. _internal_get_SparseTensorImpl: not a sparse tensor when running the following code.

import torch
a = torch.randn(3, 3)
b = torch.randn(3, 3)
c = torch.randn(3, 3)
# Convert tensors to sparse tensors
a_sparse = a.to_sparse()
b_sparse = b.to_sparse()
# Perform sparse matrix multiplication
torch.sspaddmm(c, a_sparse, b_sparse, beta=2.5, alpha=0.1)
torch.sspaddmm(c, a_sparse, b_sparse, beta=1.5, alpha=0.3)
torch.sspaddmm(c, a_sparse, b_sparse, beta=0.75, alpha=0.6)
In [4]: import torch
   ...: a = torch.randn(3, 3)
   ...: b = torch.randn(3, 3)
   ...: c = torch.randn(3, 3)
   ...: # Convert tensors to sparse tensors
   ...: a_sparse = a.to_sparse()
   ...: b_sparse = b.to_sparse()
   ...: # Perform sparse matrix multiplication
   ...: torch.sspaddmm(c, a_sparse, b_sparse, beta=2.5, alpha=0.1)
   ...: torch.sspaddmm(c, a_sparse, b_sparse, beta=1.5, alpha=0.3)
   ...: torch.sspaddmm(c, a_sparse, b_sparse, beta=0.75, alpha=0.6)
---------------------------------------------------------------------------
RuntimeError                              Traceback (most recent call last)
Cell In[4], line 9
      7 b_sparse = b.to_sparse()
      8 # Perform sparse matrix multiplication
----> 9 torch.sspaddmm(c, a_sparse, b_sparse, beta=2.5, alpha=0.1)
     10 torch.sspaddmm(c, a_sparse, b_sparse, beta=1.5, alpha=0.3)
     11 torch.sspaddmm(c, a_sparse, b_sparse, beta=0.75, alpha=0.6)

RuntimeError: self.is_sparse() INTERNAL ASSERT FAILED at "../aten/src/ATen/native/SparseTensorUtils.h":28, please report a bug to PyTorch. _internal_get_SparseTensorImpl: not a sparse tensor

Versions

Collecting environment information...
PyTorch version: 2.5.0.dev20240708+cu121
Is debug build: False
CUDA used to build PyTorch: 12.1
ROCM used to build PyTorch: N/A

OS: Ubuntu 22.04.4 LTS (x86_64)
GCC version: (Ubuntu 11.4.0-1ubuntu1~22.04) 11.4.0
Clang version: Could not collect
CMake version: version 3.22.1
Libc version: glibc-2.35

Python version: 3.11.0 | packaged by conda-forge | (main, Jan 14 2023, 12:27:40) [GCC 11.3.0] (64-bit runtime)
Python platform: Linux-5.15.0-116-generic-x86_64-with-glibc2.35
Is CUDA available: True
CUDA runtime version: 12.4.131
CUDA_MODULE_LOADING set to: LAZY
GPU models and configuration: GPU 0: NVIDIA GeForce RTX 3060 Ti
Nvidia driver version: 550.76
cuDNN version: Probably one of the following:
/usr/lib/x86_64-linux-gnu/libcudnn.so.8.9.7
/usr/lib/x86_64-linux-gnu/libcudnn.so.9.2.1
/usr/lib/x86_64-linux-gnu/libcudnn_adv.so.9.2.1
/usr/lib/x86_64-linux-gnu/libcudnn_adv_infer.so.8.9.7
/usr/lib/x86_64-linux-gnu/libcudnn_adv_train.so.8.9.7
/usr/lib/x86_64-linux-gnu/libcudnn_cnn.so.9.2.1
/usr/lib/x86_64-linux-gnu/libcudnn_cnn_infer.so.8.9.7
/usr/lib/x86_64-linux-gnu/libcudnn_cnn_train.so.8.9.7
/usr/lib/x86_64-linux-gnu/libcudnn_engines_precompiled.so.9.2.1
/usr/lib/x86_64-linux-gnu/libcudnn_engines_runtime_compiled.so.9.2.1
/usr/lib/x86_64-linux-gnu/libcudnn_graph.so.9.2.1
/usr/lib/x86_64-linux-gnu/libcudnn_heuristic.so.9.2.1
/usr/lib/x86_64-linux-gnu/libcudnn_ops.so.9.2.1
/usr/lib/x86_64-linux-gnu/libcudnn_ops_infer.so.8.9.7
/usr/lib/x86_64-linux-gnu/libcudnn_ops_train.so.8.9.7
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:                        39 bits physical, 48 bits virtual
Byte Order:                           Little Endian
CPU(s):                               24
On-line CPU(s) list:                  0-23
Vendor ID:                            GenuineIntel
Model name:                           13th Gen Intel(R) Core(TM) i7-13700F
CPU family:                           6
Model:                                183
Thread(s) per core:                   2
Core(s) per socket:                   16
Socket(s):                            1
Stepping:                             1
CPU max MHz:                          5200.0000
CPU min MHz:                          800.0000
BogoMIPS:                             4224.00
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 tsc_known_freq pni pclmulqdq dtes64 monitor ds_cpl vmx est tm2 ssse3 sdbg fma cx16 xtpr pdcm sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand lahf_lm abm 3dnowprefetch cpuid_fault epb ssbd ibrs ibpb stibp ibrs_enhanced tpr_shadow vnmi flexpriority ept vpid ept_ad fsgsbase tsc_adjust bmi1 avx2 smep bmi2 erms invpcid rdseed adx smap clflushopt clwb intel_pt sha_ni xsaveopt xsavec xgetbv1 xsaves split_lock_detect avx_vnni dtherm ida arat pln pts hwp hwp_notify hwp_act_window hwp_epp hwp_pkg_req umip pku ospke waitpkg gfni vaes vpclmulqdq rdpid movdiri movdir64b fsrm md_clear serialize arch_lbr flush_l1d arch_capabilities
Virtualization:                       VT-x
L1d cache:                            640 KiB (16 instances)
L1i cache:                            768 KiB (16 instances)
L2 cache:                             24 MiB (10 instances)
L3 cache:                             30 MiB (1 instance)
NUMA node(s):                         1
NUMA node0 CPU(s):                    0-23
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: Mitigation; Clear Register File
Vulnerability Retbleed:               Not affected
Vulnerability Spec rstack overflow:   Not affected
Vulnerability Spec store bypass:      Mitigation; Speculative Store Bypass disabled via prctl and seccomp
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; BHI BHI_DIS_S
Vulnerability Srbds:                  Not affected
Vulnerability Tsx async abort:        Not affected

Versions of relevant libraries:
[pip3] numpy==1.26.4
[pip3] pytorch-triton==3.0.0+dedb7bdf33
[pip3] torch==2.5.0.dev20240708+cu121
[pip3] torchaudio==2.4.0.dev20240708+cu121
[pip3] torchvision==0.20.0.dev20240708+cu121
[conda] numpy                     1.26.4                   pypi_0    pypi
[conda] pytorch-triton            3.0.0+dedb7bdf33          pypi_0    pypi
[conda] torch                     2.5.0.dev20240708+cu121          pypi_0    pypi
[conda] torchaudio                2.4.0.dev20240708+cu121          pypi_0    pypi
[conda] torchvision               0.20.0.dev20240708+cu121          pypi_0    pypi

cc @alexsamardzic @nikitaved @pearu @cpuhrsch @amjames @bhosmer @jcaip

amjames commented 2 months ago

sspaddmm performs an addmm with the LHS of the matrix multiply (a in your example) and the addition factor (c in your example) expected sparse. The RHS of the multiply (b in your example) should be dense. (see the docs here )

I don't know off the top of my head if we have a specific API for dense += sparse @ sparse which you are trying to do here. We might have support for some combination of compressed layouts for the MM and a dense result.

In general though dispatch though addmm will find the appropriate implementation to call for the combination of layouts for a, b, c provided if it exists so I would suggest using that API and trying coo, csr and csc layouts for a/b.