Open drago87 opened 8 months ago
probebly a copy of issu-1707644920
You have AMD RX 5700 XT but your guest Virtualbox system has VMSVGA or VboxVGA graphics card. Maybe that is the case.
in the VM under Display i have acces to VBoxVGA, VMSVGA, VBoxSVGA and None VBoxVGA and VBoxSVGA i get a picture but cant interact with enything VMSVGA is what i have been using and "works" None gets a black screen
I have caved and installed Ubuntu but now i get this error
Traceback (most recent call last):
File "/home/drago87/AI/ComfyUI/main.py", line 72, in <module>
import execution
File "/home/drago87/AI/ComfyUI/execution.py", line 12, in <module>
import nodes
File "/home/drago87/AI/ComfyUI/nodes.py", line 20, in <module>
import comfy.diffusers_load
File "/home/drago87/AI/ComfyUI/comfy/diffusers_load.py", line 4, in <module>
import comfy.sd
File "/home/drago87/AI/ComfyUI/comfy/sd.py", line 5, in <module>
from comfy import model_management
File "/home/drago87/AI/ComfyUI/comfy/model_management.py", line 114, in <module>
total_vram = get_total_memory(get_torch_device()) / (1024 * 1024)
File "/home/drago87/AI/ComfyUI/comfy/model_management.py", line 83, in get_torch_device
return torch.device(torch.cuda.current_device())
File "/home/drago87/.local/lib/python3.10/site-packages/torch/cuda/__init__.py", line 785, in current_device
_lazy_init()
File "/home/drago87/.local/lib/python3.10/site-packages/torch/cuda/__init__.py", line 298, in _lazy_init
torch._C._cuda_init()
RuntimeError: No HIP GPUs are available
after i get this fixed i only need to make a file on my Desktop to run it.
Can someone point me in the right direction
ROCm only works on native linux, it won't work on virtualbox or any other VM unless you do PCI passthrough.
that was what i was afraid of. now i have installed it on a proper ubuntu install and installed everything but get the error
Traceback (most recent call last):
File "/home/drago87/AI/ComfyUI/main.py", line 72, in <module>
import execution
File "/home/drago87/AI/ComfyUI/execution.py", line 12, in <module>
import nodes
File "/home/drago87/AI/ComfyUI/nodes.py", line 20, in <module>
import comfy.diffusers_load
File "/home/drago87/AI/ComfyUI/comfy/diffusers_load.py", line 4, in <module>
import comfy.sd
File "/home/drago87/AI/ComfyUI/comfy/sd.py", line 5, in <module>
from comfy import model_management
File "/home/drago87/AI/ComfyUI/comfy/model_management.py", line 33, in <module>
import torch_directml
File "/home/drago87/.local/lib/python3.10/site-packages/torch_directml/__init__.py", line 13, in <module>
torch.ops.load_library(directml_dll)
File "/home/drago87/.local/lib/python3.10/site-packages/torch/_ops.py", line 852, in load_library
ctypes.CDLL(path)
File "/usr/lib/python3.10/ctypes/__init__.py", line 374, in __init__
self._handle = _dlopen(self._name, mode)
OSError: libd3d12.so: cannot open shared object file: No such file or directory
Did you follow the instructions to set up ROCm? https://rocm.docs.amd.com/en/latest/deploy/linux/install_overview.html
As a first step you could try and see what the rocminfo
command reports back in order to check if ROCm has been set up correctly.
You can then also try to add the user executing the python script to the video and render usergroups:
sudo usermod -a -G video $LOGNAME
sudo usermod -a -G render $LOGNAME
This is what seems to have worked for me in the end.
i have follow that now and no difference. and a search on the computer gives nothing on libd3d12 a search on d3d12 gives
d3d12_dri.so
d3d12_drv_video.so
libvdpau_d3d12.so
libvdpau_d3d12.so.1
libvdpau_d3d12.so.1.0
libvdpau_d3d12.so.1.0.0
in folders usr/lib/x86_64-linux-gnu
and subfolders
and 3 d3d12_dri.so
in some snap/gnome*
folders
with 1 d3d12_drv_video.so
in snap/gnome-42-2204/141/usr/lib/x86_64-linux-gnu/dri
Just to double check, you are not accidentally trying to run the directml version? 5700XT should run via ROCm under native linux. It will probably need the override, though:
HSA_OVERRIDE_GFX_VERSION=10.3.0 python main.py
i was running the --directml
HSA_OVERRIDE_GFX_VERSION=10.3.0 python3 main.py
gives
Traceback (most recent call last):
File "/home/drago87/AI/ComfyUI/main.py", line 72, in <module>
import execution
File "/home/drago87/AI/ComfyUI/execution.py", line 12, in <module>
import nodes
File "/home/drago87/AI/ComfyUI/nodes.py", line 20, in <module>
import comfy.diffusers_load
File "/home/drago87/AI/ComfyUI/comfy/diffusers_load.py", line 4, in <module>
import comfy.sd
File "/home/drago87/AI/ComfyUI/comfy/sd.py", line 5, in <module>
from comfy import model_management
File "/home/drago87/AI/ComfyUI/comfy/model_management.py", line 114, in <module>
total_vram = get_total_memory(get_torch_device()) / (1024 * 1024)
File "/home/drago87/AI/ComfyUI/comfy/model_management.py", line 83, in get_torch_device
return torch.device(torch.cuda.current_device())
File "/home/drago87/.local/lib/python3.10/site-packages/torch/cuda/__init__.py", line 769, in current_device
_lazy_init()
File "/home/drago87/.local/lib/python3.10/site-packages/torch/cuda/__init__.py", line 298, in _lazy_init
torch._C._cuda_init()
RuntimeError: Found no NVIDIA driver on your system. Please check that you have an NVIDIA GPU and installed a driver from http://www.nvidia.com/Download/index.aspx
Running lspci | grep VGA
gives VGA compatible controller: Advanced Micro Devices, Inc. [AMD/ATI] Navi 10 [Radeon RX 5600 OEM/5600 XT / 5700/5700 XT] (rev c1)
This might be the source of the error. Try installing the ROCm PyTorch (below is the rocm 5.7 nightly from the ComfyUI installation guide)
pip install --pre torch torchvision torchaudio --index-url https://download.pytorch.org/whl/nightly/rocm5.7
And run ComfyUI with the override
HSA_OVERRIDE_GFX_VERSION=10.3.0 python main.py
hmm i have already installed that one but if i try doing it agen i get this
Defaulting to user installation because normal site-packages is not writeable
Looking in indexes: https://download.pytorch.org/whl/nightly/rocm5.7
Requirement already satisfied: torch in /home/drago87/.local/lib/python3.10/site-packages (2.1.0)
Requirement already satisfied: torchvision in /home/drago87/.local/lib/python3.10/site-packages (0.15.1)
Requirement already satisfied: torchaudio in /home/drago87/.local/lib/python3.10/site-packages (2.2.0.dev20231101+rocm5.7)
Requirement already satisfied: sympy in /home/drago87/.local/lib/python3.10/site-packages (from torch) (1.11.1)
Requirement already satisfied: nvidia-cusolver-cu12==11.4.5.107 in /home/drago87/.local/lib/python3.10/site-packages (from torch) (11.4.5.107)
Requirement already satisfied: nvidia-cublas-cu12==12.1.3.1 in /home/drago87/.local/lib/python3.10/site-packages (from torch) (12.1.3.1)
Requirement already satisfied: nvidia-cufft-cu12==11.0.2.54 in /home/drago87/.local/lib/python3.10/site-packages (from torch) (11.0.2.54)
Requirement already satisfied: nvidia-cuda-nvrtc-cu12==12.1.105 in /home/drago87/.local/lib/python3.10/site-packages (from torch) (12.1.105)
Requirement already satisfied: fsspec in /home/drago87/.local/lib/python3.10/site-packages (from torch) (2023.10.0)
Requirement already satisfied: nvidia-cuda-runtime-cu12==12.1.105 in /home/drago87/.local/lib/python3.10/site-packages (from torch) (12.1.105)
Requirement already satisfied: nvidia-curand-cu12==10.3.2.106 in /home/drago87/.local/lib/python3.10/site-packages (from torch) (10.3.2.106)
Requirement already satisfied: nvidia-cusparse-cu12==12.1.0.106 in /home/drago87/.local/lib/python3.10/site-packages (from torch) (12.1.0.106)
Requirement already satisfied: filelock in /home/drago87/.local/lib/python3.10/site-packages (from torch) (3.9.0)
Requirement already satisfied: triton==2.1.0 in /home/drago87/.local/lib/python3.10/site-packages (from torch) (2.1.0)
Requirement already satisfied: networkx in /home/drago87/.local/lib/python3.10/site-packages (from torch) (3.0rc1)
Requirement already satisfied: nvidia-nccl-cu12==2.18.1 in /home/drago87/.local/lib/python3.10/site-packages (from torch) (2.18.1)
Requirement already satisfied: nvidia-cudnn-cu12==8.9.2.26 in /home/drago87/.local/lib/python3.10/site-packages (from torch) (8.9.2.26)
Requirement already satisfied: typing-extensions in /home/drago87/.local/lib/python3.10/site-packages (from torch) (4.4.0)
Requirement already satisfied: nvidia-nvtx-cu12==12.1.105 in /home/drago87/.local/lib/python3.10/site-packages (from torch) (12.1.105)
Requirement already satisfied: nvidia-cuda-cupti-cu12==12.1.105 in /home/drago87/.local/lib/python3.10/site-packages (from torch) (12.1.105)
Requirement already satisfied: jinja2 in /home/drago87/.local/lib/python3.10/site-packages (from torch) (3.1.2)
Requirement already satisfied: nvidia-nvjitlink-cu12 in /home/drago87/.local/lib/python3.10/site-packages (from nvidia-cusolver-cu12==11.4.5.107->torch) (12.3.52)
Requirement already satisfied: numpy in /home/drago87/.local/lib/python3.10/site-packages (from torchvision) (1.24.1)
Requirement already satisfied: requests in /usr/lib/python3/dist-packages (from torchvision) (2.25.1)
Collecting torchvision
Using cached https://download.pytorch.org/whl/nightly/rocm5.7/torchvision-0.17.0.dev20231102%2Brocm5.7-cp310-cp310-linux_x86_64.whl (65.6 MB)
Collecting torch
Using cached https://download.pytorch.org/whl/nightly/rocm5.7/torch-2.2.0.dev20231102%2Brocm5.7-cp310-cp310-linux_x86_64.whl (1701.2 MB)
Requirement already satisfied: pillow!=8.3.*,>=5.3.0 in /usr/lib/python3/dist-packages (from torchvision) (9.0.1)
Requirement already satisfied: pytorch-triton-rocm==2.1.0+34f8189eae in /home/drago87/.local/lib/python3.10/site-packages (from torch) (2.1.0+34f8189eae)
Collecting torchaudio
Using cached https://download.pytorch.org/whl/nightly/rocm5.7/torchaudio-2.2.0.dev20231102%2Brocm5.7-cp310-cp310-linux_x86_64.whl (1.7 MB)
Requirement already satisfied: MarkupSafe>=2.0 in /home/drago87/.local/lib/python3.10/site-packages (from jinja2->torch) (2.1.2)
Requirement already satisfied: mpmath>=0.19 in /home/drago87/.local/lib/python3.10/site-packages (from sympy->torch) (1.2.1)
ERROR: Exception:
Traceback (most recent call last):
File "/usr/lib/python3/dist-packages/pip/_internal/cli/base_command.py", line 165, in exc_logging_wrapper
status = run_func(*args)
File "/usr/lib/python3/dist-packages/pip/_internal/cli/req_command.py", line 205, in wrapper
return func(self, options, args)
File "/usr/lib/python3/dist-packages/pip/_internal/commands/install.py", line 389, in run
to_install = resolver.get_installation_order(requirement_set)
File "/usr/lib/python3/dist-packages/pip/_internal/resolution/resolvelib/resolver.py", line 188, in get_installation_order
weights = get_topological_weights(
File "/usr/lib/python3/dist-packages/pip/_internal/resolution/resolvelib/resolver.py", line 276, in get_topological_weights
assert len(weights) == expected_node_count
AssertionError
Sadly I am no expert in the Linux environment. Looks like sth might be wrong with the local Python installation. Someone more adept might be able to help. Maybe a force-reinstall could help.
Just keep in mind that the torch packages for CUDA, ROCm, and DirectML are not the same within the same python environment and they will replace some components with one another.
BTW, checking rocminfo
might still be worthwhile to check if the GPU side of things is working correctly.
ROCk module is loaded
=====================
HSA System Attributes
=====================
Runtime Version: 1.1
System Timestamp Freq.: 1000.000000MHz
Sig. Max Wait Duration: 18446744073709551615 (0xFFFFFFFFFFFFFFFF) (timestamp count)
Machine Model: LARGE
System Endianness: LITTLE
Mwaitx: DISABLED
DMAbuf Support: YES
==========
HSA Agents
==========
*******
Agent 1
*******
Name: AMD Ryzen 5 3600X 6-Core Processor
Uuid: CPU-XX
Marketing Name: AMD Ryzen 5 3600X 6-Core Processor
Vendor Name: CPU
Feature: None specified
Profile: FULL_PROFILE
Float Round Mode: NEAR
Max Queue Number: 0(0x0)
Queue Min Size: 0(0x0)
Queue Max Size: 0(0x0)
Queue Type: MULTI
Node: 0
Device Type: CPU
Cache Info:
L1: 32768(0x8000) KB
Chip ID: 0(0x0)
ASIC Revision: 0(0x0)
Cacheline Size: 64(0x40)
Max Clock Freq. (MHz): 3800
BDFID: 0
Internal Node ID: 0
Compute Unit: 12
SIMDs per CU: 0
Shader Engines: 0
Shader Arrs. per Eng.: 0
WatchPts on Addr. Ranges:1
Features: None
Pool Info:
Pool 1
Segment: GLOBAL; FLAGS: FINE GRAINED
Size: 32766012(0x1f3f83c) KB
Allocatable: TRUE
Alloc Granule: 4KB
Alloc Alignment: 4KB
Accessible by all: TRUE
Pool 2
Segment: GLOBAL; FLAGS: KERNARG, FINE GRAINED
Size: 32766012(0x1f3f83c) KB
Allocatable: TRUE
Alloc Granule: 4KB
Alloc Alignment: 4KB
Accessible by all: TRUE
Pool 3
Segment: GLOBAL; FLAGS: COARSE GRAINED
Size: 32766012(0x1f3f83c) KB
Allocatable: TRUE
Alloc Granule: 4KB
Alloc Alignment: 4KB
Accessible by all: TRUE
ISA Info:
*******
Agent 2
*******
Name: gfx1010
Uuid: GPU-XX
Marketing Name: AMD Radeon RX 5700 XT
Vendor Name: AMD
Feature: KERNEL_DISPATCH
Profile: BASE_PROFILE
Float Round Mode: NEAR
Max Queue Number: 128(0x80)
Queue Min Size: 64(0x40)
Queue Max Size: 131072(0x20000)
Queue Type: MULTI
Node: 1
Device Type: GPU
Cache Info:
L1: 16(0x10) KB
L2: 4096(0x1000) KB
Chip ID: 29471(0x731f)
ASIC Revision: 2(0x2)
Cacheline Size: 64(0x40)
Max Clock Freq. (MHz): 2100
BDFID: 3328
Internal Node ID: 1
Compute Unit: 40
SIMDs per CU: 2
Shader Engines: 2
Shader Arrs. per Eng.: 2
WatchPts on Addr. Ranges:4
Features: KERNEL_DISPATCH
Fast F16 Operation: TRUE
Wavefront Size: 32(0x20)
Workgroup Max Size: 1024(0x400)
Workgroup Max Size per Dimension:
x 1024(0x400)
y 1024(0x400)
z 1024(0x400)
Max Waves Per CU: 40(0x28)
Max Work-item Per CU: 1280(0x500)
Grid Max Size: 4294967295(0xffffffff)
Grid Max Size per Dimension:
x 4294967295(0xffffffff)
y 4294967295(0xffffffff)
z 4294967295(0xffffffff)
Max fbarriers/Workgrp: 32
Packet Processor uCode:: 146
SDMA engine uCode:: 35
IOMMU Support:: None
Pool Info:
Pool 1
Segment: GLOBAL; FLAGS: COARSE GRAINED
Size: 8372224(0x7fc000) KB
Allocatable: TRUE
Alloc Granule: 4KB
Alloc Alignment: 4KB
Accessible by all: FALSE
Pool 2
Segment: GLOBAL; FLAGS:
Size: 8372224(0x7fc000) KB
Allocatable: TRUE
Alloc Granule: 4KB
Alloc Alignment: 4KB
Accessible by all: FALSE
Pool 3
Segment: GROUP
Size: 64(0x40) KB
Allocatable: FALSE
Alloc Granule: 0KB
Alloc Alignment: 0KB
Accessible by all: FALSE
ISA Info:
ISA 1
Name: amdgcn-amd-amdhsa--gfx1010:xnack-
Machine Models: HSA_MACHINE_MODEL_LARGE
Profiles: HSA_PROFILE_BASE
Default Rounding Mode: NEAR
Default Rounding Mode: NEAR
Fast f16: TRUE
Workgroup Max Size: 1024(0x400)
Workgroup Max Size per Dimension:
x 1024(0x400)
y 1024(0x400)
z 1024(0x400)
Grid Max Size: 4294967295(0xffffffff)
Grid Max Size per Dimension:
x 4294967295(0xffffffff)
y 4294967295(0xffffffff)
z 4294967295(0xffffffff)
FBarrier Max Size: 32
*** Done ***
This part looks good.
HSA_OVERRIDE_GFX_VERSION=10.3.0 python main.py
should work now after getting the python package errors sorted.
i manage to get it to work i was missing allot of package but i just followed it and installed them one by one.
Do you know somewhere i can follow to make a executable to run it from the desktop?
First thing that comes to my mind is creating a .desktop entry. But there are probably more elegant ways.
But bottom line to summarize the issue: You are now able to generate images using your 5700XT via ROCm 5.7 under native Ubuntu, correct?
well i'm able to start ComfyUI i'm getting this error with the bare minimum nodes
got prompt
ERROR:root:Failed to validate prompt for output 18:
ERROR:root:* KSampler 15:
ERROR:root: - Required input is missing: latent_image
ERROR:root:Output will be ignored
invalid prompt: {'type': 'prompt_outputs_failed_validation', 'message': 'Prompt outputs failed validation', 'details': '', 'extra_info': {}}
got prompt
model_type EPS
adm 2816
Using split attention in VAE
Working with z of shape (1, 4, 32, 32) = 4096 dimensions.
Using split attention in VAE
missing {'cond_stage_model.clip_l.logit_scale', 'cond_stage_model.clip_l.text_projection'}
left over keys: dict_keys(['conditioner.embedders.1.model.transformer.text_model.embeddings.position_ids', 'cond_stage_model.clip_l.transformer.text_model.embeddings.position_ids'])
Requested to load SDXLClipModel
Loading 1 new model
Memory access fault by GPU node-1 (Agent handle: 0x72e3f20) on address 0x7f3cc4ffd000. Reason: Page not present or supervisor privilege.
Aborted (core dumped)
The first error i got i only forgot the empty latent
If it would help i can let you remote control my pc with AnyDesk i'm usually up from 10-11 in the morning to 3 at night cet (gmt+1) i'm not home this Saturday but else i should be home
I am sorry but I won't give any support on that scale.
As for some ad hoc advice after a quick glance over the error msg: Start with a fresh ComfyUI without custom nodes. And I suggest trying a SD 1.5 based model first, before going for SDXL.
Also in this git issue there are some possible mitigation steps based on Automatic1111
Ok will try that
I'm trying to run ComfyUI on Ubunto with Virtual box but when i run the main.py file with
python3 main.py
from within the comfyui folder i geti have a AMD RX 5700 XT and have installed the rocm5.7