[Bug] version libcudnn_cnn_infer.so.8 not defined in file libcudnn_cnn_infer.so.8 with link time reference

[leya516]

Your current environment information

PyTorch version: 2.3.1+cu118
Is debug build: False
CUDA used to build PyTorch: 11.8
ROCM used to build PyTorch: N/A

OneFlow version: none
Nexfort version: none
OneDiff version: 1.2.1.dev28+g424c81a.d20241125
OneDiffX version: none

Vulnerability Itlb multihit: Not affected
Vulnerability L1tf: Not affected
Vulnerability Mds: Not affected
Vulnerability Meltdown: Not affected
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
Vulnerability Tsx async abort: Not affected
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 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 cat_l2 cdp_l3 invpcid_single cdp_l2 ssbd mba ibrs ibpb stibp ibrs_enhanced tpr_shadow vnmi flexpriority ept vpid fsgsbase tsc_adjust bmi1 hle avx2 smep bmi2 erms invpcid rtm 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 wbnoinvd dtherm ida arat pln pts hwp hwp_act_window hwp_epp hwp_pkg_req avx512vbmi umip pku ospke avx512_vbmi2 gfni vaes vpclmulqdq avx512_vnni avx512_bitalg tme avx512_vpopcntdq la57 rdpid cldemote movdiri movdir64b md_clear pconfig flush_l1d arch_capabilities

Versions of relevant libraries:
[pip3] diffusers==0.31.0
[pip3] numpy==2.1.3
[pip3] torch==2.3.1+cu118
[pip3] torchvision==0.18.1+cu118
[pip3] transformers==4.46.3
[pip3] triton==2.3.1
[conda] numpy 2.1.3 pypi_0 pypi
[conda] torch 2.3.1+cu118 pypi_0 pypi
[conda] torchvision 0.18.1+cu118 pypi_0 pypi
[conda] triton 2.3.1 pypi_0 pypi

🐛 Describe the bug

File "/home/hadoop-platcv/workspace/codebase/onediff/src/onediff/infer_compiler/backends/oneflow/init.py", line 3, in
from .deployable_module import OneflowDeployableModule
File "/home/hadoop-platcv/workspace/codebase/onediff/src/onediff/infer_compiler/backends/oneflow/deployable_module.py", line 6, in
import oneflow as flow # usort: skip
File "/home/hadoop-platcv/anaconda3/envs/one/lib/python3.10/site-packages/oneflow/init.py", line 26, in
import oneflow._oneflow_internal
ImportError: /home/hadoop-platcv/anaconda3/envs/one/lib/python3.10/site-packages/oneflow/../oneflow.libs/libcudnn_cnn_train-374a60c5.so.8.9.6: symbol _ZN5cudnn6fusion35decompress_byte_arr_n_append_to_strEPKhmmRSs, version libcudnn_cnn_infer.so.8 not defined in file libcudnn_cnn_infer.so.8 with link time reference

[leya516]

from onediff.infer_compiler import oneflow_compile
pipe.unet = oneflow_compile(pipe.unet)
can cause the problem mentioned above

[lilyzlt]

same, my env is 2.3.1+cu12.1

[1586358665]

same, my env is 2.4.1+cu12.1，import ondflow always cause some cudnn problems，

[owu-1]

If using the prebuilt oneflow package from this repo, try the older versions. Using latest for me gives

ImportError: /opt/conda/lib/python3.11/site-packages/oneflow/../oneflow.libs/libcudnn_cnn_train-fb9074d0.so.8.9.0: undefined symbol: _ZN5cudnn3cnn34layerNormFwd_execute_internal_implERKNS_7backend11VariantPackEP11CUstream_stRNS0_18LayerNormFwdParamsERNS1_12OperationSetERP12cudnnContextmb, version libcudnn_cnn_infer.so.8

and 0.9.1.dev20241114+cu121 works