Closed leung1010 closed 1 month ago
I got the same problem. I am not sure if that's the warp context problem. It cannot recognize the cuda path.
same problem
When I upgraded the warp version to 0.15.1, this issue was resolved. At the same time, replace the tape. py file with the one mentioned by the author in another issue.
Sorry for the late reply. I tried but I cannot reproduce the error on my end though. Could you attach more details about your workstation? Does cuda:0
works for your torch.device
or it's only broken in warp?
Thanks for your reply. system: Ubuntu 20.04.4 LTS GPU: NVIDIA-SMI 470.57.02 , 0 Quadro RTX 6000 In the "conda list": nvidia-cublas-cu11 11.10.3.66 pypi_0 pypi nvidia-cuda-cupti-cu11 11.7.101 pypi_0 pypi nvidia-cuda-nvrtc-cu11 11.7.99 pypi_0 pypi nvidia-cuda-runtime-cu11 11.7.99 pypi_0 pypi nvidia-cudnn-cu11 8.5.0.96 pypi_0 pypi nvidia-cufft-cu11 10.9.0.58 pypi_0 pypi nvidia-curand-cu11 10.2.10.91 pypi_0 pypi nvidia-cusolver-cu11 11.4.0.1 pypi_0 pypi nvidia-cusparse-cu11 11.7.4.91 pypi_0 pypi nvidia-nccl-cu11 2.14.3 pypi_0 pypi nvidia-nvtx-cu11 11.7.91 pypi_0 pypi python 3.10.14 torch 2.0.1 pypi_0 pypi torchaudio 2.0.2 pypi_0 pypi torchvision 0.15.2 pypi_0 pypi
Cuda:0 works well for torch Python 3.10.14 | packaged by conda-forge | (main, Mar 20 2024, 12:45:18) [GCC 12.3.0] on linux Type "help", "copyright", "credits" or "license" for more information.
import torch if torch.cuda.is_available(): ... print("CUDA is available") ... else: ... print("CUDA is not available") ... CUDA is available device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu") print("Using device:", device) Using device: cuda:0
Actually, I also tried another new DELL workstation(nvidia driver:535, A6000), it reported the same error info.
From: Pingchuan Ma @.> Sent: Wednesday, August 28, 2024 01:55 To: PingchuanMa/NCLaw @.> Cc: @. @.>; Author @.***> Subject: Re: [PingchuanMa/NCLaw] Report a bug: warp?(codes can be run well in my workstation) (Issue #2)
This is an external email.
Sorry for the late reply. I tried but I cannot reproduce the error on my end though. Could you attach more details about your workstation? Does cuda:0 works for your torch.device or it's only broken in warp?
— Reply to this email directly, view it on GitHubhttps://github.com/PingchuanMa/NCLaw/issues/2#issuecomment-2313185786, or unsubscribehttps://github.com/notifications/unsubscribe-auth/ALYH3ZBMCO36AWKYZIYMNYDZTS4R5AVCNFSM6AAAAABJGNCWICVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDGMJTGE4DKNZYGY. You are receiving this because you authored the thread.Message ID: @.***>
did you install warp from scratch or from pypi? could you also attach the output of bash command which nvcc
please?
Yes. Nvcc version is 11.3. But another workstation is 11.8, both report same error.
On 30 Aug 2024, at 00:02, Pingchuan Ma @.***> wrote:
This is an external email.
did you install warp from scratch or from pypi? could you also attach the output of bash command which nvcc please?
— Reply to this email directly, view it on GitHubhttps://github.com/PingchuanMa/NCLaw/issues/2#issuecomment-2318235083, or unsubscribehttps://github.com/notifications/unsubscribe-auth/ALYH3ZFDCS42I2Q67I56DQTZT5AYHAVCNFSM6AAAAABJGNCWICVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDGMJYGIZTKMBYGM. You are receiving this because you authored the thread.Message ID: @.***>
I do presume that this is a warp installation problem. Could you try to compile warp from scratch? The warp version used in this repo is attached in third_party/warp
. I will also try to reproduce this on a computer with cuda 11.x to see if it's a cuda problem. Will let you know.
OK, thanks. l will let you know after once try.
Sent from my iPhone
On 31 Aug 2024, at 07:26, Pingchuan Ma @.***> wrote:
This is an external email.
I do presume that this is a warp installation problem. Could you try to compile warp from scratch? The warp version used in this repo is attached in third_party/warp. I will also try to reproduce this on a computer with cuda 11.x to see if it's a cuda problem. Will let you know.
— Reply to this email directly, view it on GitHubhttps://github.com/PingchuanMa/NCLaw/issues/2#issuecomment-2322595692, or unsubscribehttps://github.com/notifications/unsubscribe-auth/ALYH3ZEVEF3M3U5YSYSJO4TZUD5RPAVCNFSM6AAAAABJGNCWICVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDGMRSGU4TKNRZGI. You are receiving this because you authored the thread.Message ID: @.***>
I do presume that this is a warp installation problem. Could you try to compile warp from scratch? The warp version used in this repo is attached in
third_party/warp
. I will also try to reproduce this on a computer with cuda 11.x to see if it's a cuda problem. Will let you know.
Hi pingchuan,
Sry for late reply. I didn't work these days.
Right now, I ran
pip uninstall -y warp warp-lang pip install warp-lang=0.15.1
It works well now.
Many thanks!
when I excute the command:
xxx$python experiments/scripts/dataset/main.py
I got this, as follow: `env: blob: bsdf_pcd: type: diffuse reflectance: type: rgb value:Warp 0.6.1 initialized: CUDA not available Devices: "cpu" | x86_64 Kernel cache: /home/xxxx/.cache/warp/0.6.1 Error executing job with overrides: ['overwrite=False', 'resume=False', 'gpu=0', 'cpu=0', 'env=jelly', 'render=debug', 'sim=low', 'name=jelly/dataset', 'dataset=True'] Traceback (most recent call last): File "/home/xxxx/xxxx_research/NCLaw/experiments/eval.py", line 137, in
main()
File "/home/xxxx/miniconda3/envs/nclaw/lib/python3.10/site-packages/torch/utils/_contextlib.py", line 115, in decorate_context
return func(*args, **kwargs)
File "/home/xxxx/miniconda3/envs/nclaw/lib/python3.10/site-packages/hydra/main.py", line 94, in decorated_main
_run_hydra(
File "/home/xxxx/miniconda3/envs/nclaw/lib/python3.10/site-packages/hydra/_internal/utils.py", line 394, in _run_hydra
_run_app(
File "/home/xxxx/miniconda3/envs/nclaw/lib/python3.10/site-packages/hydra/_internal/utils.py", line 457, in _run_app
run_and_report(
File "/home/xxxx/miniconda3/envs/nclaw/lib/python3.10/site-packages/hydra/_internal/utils.py", line 223, in run_and_report
raise ex
File "/home/xxxx/miniconda3/envs/nclaw/lib/python3.10/site-packages/hydra/_internal/utils.py", line 220, in run_and_report
return func()
File "/home/xxxx/miniconda3/envs/nclaw/lib/python3.10/site-packages/hydra/_internal/utils.py", line 458, in
lambda: hydra.run(
File "/home/xxxx/miniconda3/envs/nclaw/lib/python3.10/site-packages/hydra/internal/hydra.py", line 132, in run
= ret.return_value
File "/home/xxxx/miniconda3/envs/nclaw/lib/python3.10/site-packages/hydra/core/utils.py", line 260, in return_value
raise self._return_value
File "/home/xxxx/miniconda3/envs/nclaw/lib/python3.10/site-packages/hydra/core/utils.py", line 186, in run_job
ret.return_value = task_function(task_cfg)
File "/home/xxxx/xxxx_research/NCLaw/experiments/eval.py", line 35, in main
wp_device = wp.get_device(f'cuda:{cfg.gpu}')
File "/home/xxxx/miniconda3/envs/nclaw/lib/python3.10/site-packages/warp/context.py", line 1676, in get_device
return runtime.get_device(ident)
File "/home/xxxx/miniconda3/envs/nclaw/lib/python3.10/site-packages/warp/context.py", line 1484, in get_device
return self.device_map[ident]
KeyError: 'cuda:0'
env:
blob:
bsdf_pcd:
type: diffuse
reflectance:
type: rgb
value:
Warp 0.6.1 initialized: CUDA not available Devices: "cpu" | x86_64 Kernel cache: /home/xxxx/.cache/warp/0.6.1 Error executing job with overrides: ['overwrite=False', 'resume=False', 'gpu=0', 'cpu=0', 'env=sand', 'render=debug', 'sim=low', 'name=sand/dataset', 'dataset=True'] Traceback (most recent call last): File "/home/xxxx/xxxx_research/NCLaw/experiments/eval.py", line 137, in
main()
File "/home/xxxx/miniconda3/envs/nclaw/lib/python3.10/site-packages/torch/utils/_contextlib.py", line 115, in decorate_context
return func(*args, **kwargs)
File "/home/xxxx/miniconda3/envs/nclaw/lib/python3.10/site-packages/hydra/main.py", line 94, in decorated_main
_run_hydra(
File "/home/xxxx/miniconda3/envs/nclaw/lib/python3.10/site-packages/hydra/_internal/utils.py", line 394, in _run_hydra
_run_app(
File "/home/xxxx/miniconda3/envs/nclaw/lib/python3.10/site-packages/hydra/_internal/utils.py", line 457, in _run_app
run_and_report(
File "/home/xxxx/miniconda3/envs/nclaw/lib/python3.10/site-packages/hydra/_internal/utils.py", line 223, in run_and_report
raise ex
File "/home/xxxx/miniconda3/envs/nclaw/lib/python3.10/site-packages/hydra/_internal/utils.py", line 220, in run_and_report
return func()
File "/home/xxxx/miniconda3/envs/nclaw/lib/python3.10/site-packages/hydra/_internal/utils.py", line 458, in
lambda: hydra.run(
File "/home/xxxx/miniconda3/envs/nclaw/lib/python3.10/site-packages/hydra/internal/hydra.py", line 132, in run
= ret.return_value
File "/home/xxxx/miniconda3/envs/nclaw/lib/python3.10/site-packages/hydra/core/utils.py", line 260, in return_value
raise self._return_value
File "/home/xxxx/miniconda3/envs/nclaw/lib/python3.10/site-packages/hydra/core/utils.py", line 186, in run_job
ret.return_value = task_function(task_cfg)
File "/home/xxxx/xxxx_research/NCLaw/experiments/eval.py", line 35, in main
wp_device = wp.get_device(f'cuda:{cfg.gpu}')
File "/home/xxxx/miniconda3/envs/nclaw/lib/python3.10/site-packages/warp/context.py", line 1676, in get_device
return runtime.get_device(ident)
File "/home/xxxx/miniconda3/envs/nclaw/lib/python3.10/site-packages/warp/context.py", line 1484, in get_device
return self.device_map[ident]
KeyError: 'cuda:0'
env:
blob:
bsdf_pcd:
type: diffuse
reflectance:
type: rgb
value:
Warp 0.6.1 initialized: CUDA not available Devices: "cpu" | x86_64 Kernel cache: /home/xxxx/.cache/warp/0.6.1 Error executing job with overrides: ['overwrite=False', 'resume=False', 'gpu=0', 'cpu=0', 'env=plasticine', 'render=debug', 'sim=low', 'name=plasticine/dataset', 'dataset=True'] Traceback (most recent call last): File "/home/xxxx/xxxx_research/NCLaw/experiments/eval.py", line 137, in
main()
File "/home/xxxx/miniconda3/envs/nclaw/lib/python3.10/site-packages/torch/utils/_contextlib.py", line 115, in decorate_context
return func(*args, **kwargs)
File "/home/xxxx/miniconda3/envs/nclaw/lib/python3.10/site-packages/hydra/main.py", line 94, in decorated_main
_run_hydra(
File "/home/xxxx/miniconda3/envs/nclaw/lib/python3.10/site-packages/hydra/_internal/utils.py", line 394, in _run_hydra
_run_app(
File "/home/xxxx/miniconda3/envs/nclaw/lib/python3.10/site-packages/hydra/_internal/utils.py", line 457, in _run_app
run_and_report(
File "/home/xxxx/miniconda3/envs/nclaw/lib/python3.10/site-packages/hydra/_internal/utils.py", line 223, in run_and_report
raise ex
File "/home/xxxx/miniconda3/envs/nclaw/lib/python3.10/site-packages/hydra/_internal/utils.py", line 220, in run_and_report
return func()
File "/home/xxxx/miniconda3/envs/nclaw/lib/python3.10/site-packages/hydra/_internal/utils.py", line 458, in
lambda: hydra.run(
File "/home/xxxx/miniconda3/envs/nclaw/lib/python3.10/site-packages/hydra/internal/hydra.py", line 132, in run
= ret.return_value
File "/home/xxxx/miniconda3/envs/nclaw/lib/python3.10/site-packages/hydra/core/utils.py", line 260, in return_value
raise self._return_value
File "/home/xxxx/miniconda3/envs/nclaw/lib/python3.10/site-packages/hydra/core/utils.py", line 186, in run_job
ret.return_value = task_function(task_cfg)
File "/home/xxxx/xxxx_research/NCLaw/experiments/eval.py", line 35, in main
wp_device = wp.get_device(f'cuda:{cfg.gpu}')
File "/home/xxxx/miniconda3/envs/nclaw/lib/python3.10/site-packages/warp/context.py", line 1676, in get_device
return runtime.get_device(ident)
File "/home/xxxx/miniconda3/envs/nclaw/lib/python3.10/site-packages/warp/context.py", line 1484, in get_device
return self.device_map[ident]
KeyError: 'cuda:0'
env:
blob:
bsdf_pcd:
type: diffuse
reflectance:
type: rgb
value:
Warp 0.6.1 initialized: CUDA not available Devices: "cpu" | x86_64 Kernel cache: /home/xxxx/.cache/warp/0.6.1 Error executing job with overrides: ['overwrite=False', 'resume=False', 'gpu=0', 'cpu=0', 'env=water', 'render=debug', 'sim=low', 'name=water/dataset', 'dataset=True'] Traceback (most recent call last): File "/home/xxxx/xxxx_research/NCLaw/experiments/eval.py", line 137, in
main()
File "/home/xxxx/miniconda3/envs/nclaw/lib/python3.10/site-packages/torch/utils/_contextlib.py", line 115, in decorate_context
return func(*args, **kwargs)
File "/home/xxxx/miniconda3/envs/nclaw/lib/python3.10/site-packages/hydra/main.py", line 94, in decorated_main
_run_hydra(
File "/home/xxxx/miniconda3/envs/nclaw/lib/python3.10/site-packages/hydra/_internal/utils.py", line 394, in _run_hydra
_run_app(
File "/home/xxxx/miniconda3/envs/nclaw/lib/python3.10/site-packages/hydra/_internal/utils.py", line 457, in _run_app
run_and_report(
File "/home/xxxx/miniconda3/envs/nclaw/lib/python3.10/site-packages/hydra/_internal/utils.py", line 223, in run_and_report
raise ex
File "/home/xxxx/miniconda3/envs/nclaw/lib/python3.10/site-packages/hydra/_internal/utils.py", line 220, in run_and_report
return func()
File "/home/xxxx/miniconda3/envs/nclaw/lib/python3.10/site-packages/hydra/_internal/utils.py", line 458, in
lambda: hydra.run(
File "/home/xxxx/miniconda3/envs/nclaw/lib/python3.10/site-packages/hydra/internal/hydra.py", line 132, in run
= ret.return_value
File "/home/xxxx/miniconda3/envs/nclaw/lib/python3.10/site-packages/hydra/core/utils.py", line 260, in return_value
raise self._return_value
File "/home/xxxx/miniconda3/envs/nclaw/lib/python3.10/site-packages/hydra/core/utils.py", line 186, in run_job
ret.return_value = task_function(task_cfg)
File "/home/xxxx/xxxx_research/NCLaw/experiments/eval.py", line 35, in main
wp_device = wp.get_device(f'cuda:{cfg.gpu}')
File "/home/xxxx/miniconda3/envs/nclaw/lib/python3.10/site-packages/warp/context.py", line 1676, in get_device
return runtime.get_device(ident)
File "/home/xxxx/miniconda3/envs/nclaw/lib/python3.10/site-packages/warp/context.py", line 1484, in get_device
return self.device_map[ident]
KeyError: 'cuda:0'`
Btw, I'm sure the CUDA has been installed correctly follow your provided instruction, so could your give me some advices to tackle this issue? Maybe the warp is the trouble maker?
Looking forward to your reply!!!