Summary
The stable_fast library fails to run the ControlNet module in a Stable Diffusion Image-to-Image pipeline. The error occurs specifically during CUDNN convolution operations.
Steps to Reproduce
- Load a Stable Diffusion Image-to-Image pipeline with ControlNet
- Configure compilation settings with
do_compile_controlnet=True
- Attempt to compile the pipeline
- Observe CUDNN error during ControlNet during execution with some input sizes
Error
The following operation failed in the TorchScript interpreter.
Traceback of TorchScript (most recent call last):
graph(%1, %2, %3, %4, %5, %6, %7, %8, %9, %10, %11, %12, %13):
%x = sfast::cudnn_convolution_bias(%1, %2, %3, %4, %5, %6, %7, %8, %9)
~~~~~ <--- HERE
return (%x)
RuntimeError: cuDNN error: CUDNN_STATUS_NOT_SUPPORTED. This error may appear if you passed in a non-contiguous input.
Investigation Findings
- Error only occurs when compiling the ControlNet module
- Other components (VAE, UNet) compile successfully
- Adding
.contiguous() to ControlNet inputs did not resolve the issue
- The error appears to be related to tensor memory layout during CUDNN operations
Here is the code I used to selectively compile different parts of the pipeline:
def compile(m, config, do_compile_vae=True, do_compile_unet=True, do_compile_controlnet=True):
# attribute `device` is not generally available
device = m.device if hasattr(m, 'device') else torch.device(
'cuda' if torch.cuda.is_available() else 'cpu')
enable_cuda_graph = config.enable_cuda_graph and device.type == 'cuda'
if do_compile_unet:
m.unet = compile_unet(m.unet, config)
if do_compile_controlnet and hasattr(m, 'controlnet'):
m.controlnet = compile_unet(m.controlnet, config)
if do_compile_vae:
m.vae = compile_vae(m.vae, config)
if config.enable_jit:
lazy_trace_ = _build_lazy_trace(config)
if getattr(m, 'text_encoder', None) is not None:
m.text_encoder.forward = lazy_trace_(m.text_encoder.forward)
# for SDXL
if getattr(m, 'text_encoder_2', None) is not None:
m.text_encoder_2.forward = lazy_trace_(m.text_encoder_2.forward)
# for SVD
if getattr(m, 'image_encoder', None) is not None:
m.image_encoder.forward = lazy_trace_(m.image_encoder.forward)
if config.trace_scheduler:
m.scheduler.scale_model_input = lazy_trace_(
m.scheduler.scale_model_input)
m.scheduler.step = lazy_trace_(m.scheduler.step)
if enable_cuda_graph:
if getattr(m, 'text_encoder', None) is not None:
m.text_encoder.forward = make_dynamic_graphed_callable(
m.text_encoder.forward)
if getattr(m, 'text_encoder_2', None) is not None:
m.text_encoder_2.forward = make_dynamic_graphed_callable(
m.text_encoder_2.forward)
if getattr(m, 'image_encoder', None) is not None:
m.image_encoder.forward = make_dynamic_graphed_callable(
m.image_encoder.forward)
if hasattr(m, 'image_processor'):
from sfast.libs.diffusers.image_processor import patch_image_prcessor
patch_image_prcessor(m.image_processor)
return mQuestions
- Are there known issues with CUDNN operations and specific tensor shapes?
- Are there any recommended workarounds?
Environment Details
Collecting environment information...
PyTorch version: 2.2.2+cu118
Is debug build: False
CUDA used to build PyTorch: 11.8
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: Could not collect
Libc version: glibc-2.35
Python version: 3.10.16 | packaged by conda-forge | (main, Dec 5 2024, 14:16:10) [GCC 13.3.0] (64-bit runtime)
Python platform: Linux-5.15.0-86-generic-x86_64-with-glibc2.35
Is CUDA available: True
CUDA runtime version: Could not collect
CUDA_MODULE_LOADING set to: LAZY
GPU models and configuration: GPU 0: NVIDIA A100 80GB PCIe
Nvidia driver version: 535.104.12
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: 46 bits physical, 57 bits virtual
Byte Order: Little Endian
CPU(s): 12
On-line CPU(s) list: 0-11
Vendor ID: GenuineIntel
Model name: Intel(R) Xeon(R) Gold 6338 CPU @ 2.00GHz
CPU family: 6
Model: 106
Thread(s) per core: 2
Core(s) per socket: 6
Socket(s): 1
Stepping: 6
BogoMIPS: 3999.99
Flags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ss ht syscall nx pdpe1gb rdtscp lm constant_tsc arch_perfmon rep_good nopl xtopology nonstop_tsc cpuid tsc_known_freq pni pclmulqdq vmx ssse3 fma cx16 pdcm pcid sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand hypervisor lahf_lm abm 3dnowprefetch cpuid_fault invpcid_single ssbd ibrs ibpb stibp ibrs_enhanced tpr_shadow vnmi flexpriority ept vpid ept_ad fsgsbase tsc_adjust bmi1 avx2 smep bmi2 erms invpcid avx512f avx512dq rdseed adx smap avx512ifma clflushopt clwb avx512cd sha_ni avx512bw avx512vl xsaveopt xsavec xgetbv1 xsaves wbnoinvd arat avx512vbmi umip pku ospke avx512_vbmi2 gfni vaes vpclmulqdq avx512_vnni avx512_bitalg avx512_vpopcntdq la57 rdpid fsrm md_clear arch_capabilities
Virtualization: VT-x
Hypervisor vendor: KVM
Virtualization type: full
L1d cache: 288 KiB (6 instances)
L1i cache: 192 KiB (6 instances)
L2 cache: 7.5 MiB (6 instances)
L3 cache: 48 MiB (1 instance)
NUMA node(s): 1
NUMA node0 CPU(s): 0-11
Vulnerability Gather data sampling: Unknown: Dependent on hypervisor status
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 Host state unknown
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 IBRS, IBPB conditional, RSB filling, PBRSB-eIBRS SW sequence
Vulnerability Srbds: Not affected
Vulnerability Tsx async abort: Mitigation; TSX disabled
Versions of relevant libraries:
[pip3] numpy==1.26.3
[pip3] nvidia-cublas-cu11==11.11.3.6
[pip3] nvidia-cuda-cupti-cu11==11.8.87
[pip3] nvidia-cuda-nvrtc-cu11==11.8.89
[pip3] nvidia-cuda-runtime-cu11==11.8.89
[pip3] nvidia-cudnn-cu11==8.7.0.84
[pip3] nvidia-cufft-cu11==10.9.0.58
[pip3] nvidia-curand-cu11==10.3.0.86
[pip3] nvidia-cusolver-cu11==11.4.1.48
[pip3] nvidia-cusparse-cu11==11.7.5.86
[pip3] nvidia-nccl-cu11==2.19.3
[pip3] nvidia-nvtx-cu11==11.8.86
[pip3] stable-fast==1.0.5+torch222cu118
[pip3] torch==2.2.2+cu118
[pip3] torchaudio==2.2.2+cu118
[pip3] torchvision==0.17.2+cu118
[pip3] triton==2.2.0
[conda] numpy 1.26.3 pypi_0 pypi
[conda] nvidia-cublas-cu11 11.11.3.6 pypi_0 pypi
[conda] nvidia-cuda-cupti-cu11 11.8.87 pypi_0 pypi
[conda] nvidia-cuda-nvrtc-cu11 11.8.89 pypi_0 pypi
[conda] nvidia-cuda-runtime-cu11 11.8.89 pypi_0 pypi
[conda] nvidia-cudnn-cu11 8.7.0.84 pypi_0 pypi
[conda] nvidia-cufft-cu11 10.9.0.58 pypi_0 pypi
[conda] nvidia-curand-cu11 10.3.0.86 pypi_0 pypi
[conda] nvidia-cusolver-cu11 11.4.1.48 pypi_0 pypi
[conda] nvidia-cusparse-cu11 11.7.5.86 pypi_0 pypi
[conda] nvidia-nccl-cu11 2.19.3 pypi_0 pypi
[conda] nvidia-nvtx-cu11 11.8.86 pypi_0 pypi
[conda] stable-fast 1.0.5+torch222cu118 pypi_0 pypi
[conda] torch 2.2.2+cu118 pypi_0 pypi
[conda] torchaudio 2.2.2+cu118 pypi_0 pypi
[conda] torchvision 0.17.2+cu118 pypi_0 pypi
[conda] triton 2.2.0 pypi_0 pypi
[conda] diffusers 0.32.2 pypi_0 pypi
[conda] transformers 4.48.1 pypi_0 pypi
Summary
The
stable_fastlibrary fails to run the ControlNet module in a Stable Diffusion Image-to-Image pipeline. The error occurs specifically during CUDNN convolution operations.Steps to Reproduce
do_compile_controlnet=TrueError
Investigation Findings
.contiguous()to ControlNet inputs did not resolve the issueHere is the code I used to selectively compile different parts of the pipeline:
Questions
Environment Details