Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

【major-refactoring】Need help about lanch run #66

Closed
wangguan1995 opened this issue Sep 24, 2024 · 2 comments
Closed

【major-refactoring】Need help about lanch run #66

wangguan1995 opened this issue Sep 24, 2024 · 2 comments

Comments

@wangguan1995
Copy link

wangguan1995 commented Sep 24, 2024

python          3.9
warp-lang           1.0.2
jax                 0.4.20
jaxlib              0.4.20
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2023 NVIDIA Corporation
Built on Fri_Nov__3_17:16:49_PDT_2023
Cuda compilation tools, release 12.3, V12.3.103
Build cuda_12.3.r12.3/compiler.33492891_0
python examples/cfd/windtunnel_3d.py 

==================================================

Simulation Configuration:
Grid size: 512 x 128 x 128
Backend: ComputeBackend.WARP
Velocity set: D3Q27
Precision policy: PrecisionPolicy.FP32FP32
Prescribed velocity: 0.02
Reynolds number: 50000.0
Max iterations: 100000

==================================================

Warp 1.0.2 initialized:
   CUDA Toolkit 11.5, Driver 11.4
   Devices:
     "cpu"      : "x86_64"
     "cuda:0"   : "Tesla V100-SXM2-32GB" (32 GiB, sm_70, mempool enabled)
     "cuda:1"   : "Tesla V100-SXM2-32GB" (32 GiB, sm_70, mempool enabled)
     "cuda:2"   : "Tesla V100-SXM2-32GB" (32 GiB, sm_70, mempool enabled)
     "cuda:3"   : "Tesla V100-SXM2-32GB" (32 GiB, sm_70, mempool enabled)
     "cuda:4"   : "Tesla V100-SXM2-32GB" (32 GiB, sm_70, mempool enabled)
     "cuda:5"   : "Tesla V100-SXM2-32GB" (32 GiB, sm_70, mempool enabled)
     "cuda:6"   : "Tesla V100-SXM2-32GB" (32 GiB, sm_70, mempool enabled)
     "cuda:7"   : "Tesla V100-SXM2-32GB" (32 GiB, sm_70, mempool enabled)
   CUDA peer access:
     Supported fully (all-directional)
   Kernel cache:
     /root/.cache/warp/1.0.2
Module xlb.operator.boundary_masker.indices_boundary_masker load on device 'cuda:0' took 26.24 ms
Module xlb.operator.boundary_masker.mesh_boundary_masker load on device 'cuda:0' took 7.66 ms
Module xlb.operator.equilibrium.quadratic_equilibrium load on device 'cuda:0' took 245.91 ms
Warp NVRTC compilation error 6: NVRTC_ERROR_COMPILATION (/buildAgent/work/a9ae500d09a78409/warp/native/warp.cu:2514)
default_program(145): warning #550-D: variable "var_21" was set but never used

default_program(149): warning #550-D: variable "var_25" was set but never used

default_program(701): warning #550-D: variable "adj_14" was set but never used

default_program(714): warning #550-D: variable "adj_27" was set but never used

default_program(723): warning #550-D: variable "adj_36" was set but never used

default_program(735): warning #550-D: variable "adj_48" was set but never used

default_program(744): warning #550-D: variable "adj_57" was set but never used

default_program(756): warning #550-D: variable "adj_69" was set but never used

default_program(1195): warning #550-D: variable "adj_6" was set but never used

default_program(1207): warning #550-D: variable "adj_18" was set but never used

default_program(1208): warning #550-D: variable "adj_19" was set but never used

default_program(1489): warning #177-D: variable "adj_2" was declared but never referenced

default_program(1498): warning #550-D: variable "adj_11" was set but never used

default_program(2464): warning #550-D: variable "adj_7" was set but never used

default_program(2611): warning #550-D: variable "adj_7" was set but never used

default_program(3467): warning #550-D: variable "adj_9" was set but never used

default_program(3474): warning #550-D: variable "adj_16" was set but never used

default_program(3482): warning #550-D: variable "adj_24" was set but never used

default_program(3489): warning #550-D: variable "adj_31" was set but never used

default_program(3497): warning #550-D: variable "adj_39" was set but never used

default_program(3504): warning #550-D: variable "adj_46" was set but never used

default_program(3809): error: function "QuadraticEquilibrium___construct_warp__locals__functional" has already been defined

default_program(3998): error: function "adj_QuadraticEquilibrium___construct_warp__locals__functional" has already been defined

default_program(6322): warning #550-D: variable "adj_7" was set but never used

default_program(68): warning #177-D: function "adj_IncompressibleNavierStokesStepper___construct_warp__locals__BoundaryConditionIDStruct" was declared but never referenced

2 errors detected in the compilation of "default_program".
Module xlb.operator.stepper.nse_stepper load on device 'cuda:0' took 480.77 ms
Traceback (most recent call last):
  File "/workspace/XLB/examples/cfd/windtunnel_3d.py", line 248, in <module>
    simulation.run(num_steps, print_interval, post_process_interval=1000)
  File "/workspace/XLB/examples/cfd/windtunnel_3d.py", line 138, in run
    self.f_1 = self.stepper(self.f_0, self.f_1, self.bc_mask, self.missing_mask, i)
  File "/workspace/XLB/xlb/operator/operator.py", line 74, in __call__
    raise Exception(f"Error captured for backend with key {key} for operator {self.__class__.__name__}: {error}\n {traceback_str}")
Exception: Error captured for backend with key ('IncompressibleNavierStokesStepper', <ComputeBackend.WARP: 2>, '(self, f_0, f_1, bc_mask, missing_mask, timestep)') for operator IncompressibleNavierStokesStepper: CUDA kernel build failed with error code 6
 Traceback (most recent call last):
  File "/workspace/XLB/xlb/operator/operator.py", line 64, in __call__
    result = backend_method(self, *args, **kwargs)
  File "/workspace/XLB/xlb/operator/stepper/nse_stepper.py", line 404, in warp_implementation
    wp.launch(
  File "/root/miniconda3/envs/py3.9/lib/python3.9/site-packages/warp/context.py", line 4234, in launch
    if not module.load(device):
  File "/root/miniconda3/envs/py3.9/lib/python3.9/site-packages/warp/context.py", line 1691, in load
    raise (e)
  File "/root/miniconda3/envs/py3.9/lib/python3.9/site-packages/warp/context.py", line 1669, in load
    warp.build.build_cuda(
  File "/root/miniconda3/envs/py3.9/lib/python3.9/site-packages/warp/build.py", line 30, in build_cuda
    raise Exception(f"CUDA kernel build failed with error code {err}")
Exception: CUDA kernel build failed with error code 6
@wangguan1995
Copy link
Author

I've seen this one:
#64

@mehdiataei
Copy link
Contributor

Please install Warp-lang from source to mitigate this issue. The next release of Warp will have this problem fixed.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants