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

Algorithm of Unique doesn't work with Array over 1 million on Cuda #1300

Open
geobimmodeling opened this issue Nov 11, 2024 · 1 comment
Open
Labels

Comments

@geobimmodeling
Copy link

Describe the bug

It throws following message: (test project is attached)
AlgorithmsUnique.zip

ILGPU.Runtime.Cuda.CudaException
HResult=0x80131500
Message=a PTX JIT compilation failed
Source=ILGPU
StackTrace:
at ILGPU.Runtime.Cuda.CudaException.ThrowIfFailed(CudaError cudaStatus) in C:\Users\17058\Documents\ILGPU\Src\ILGPU\Runtime\Cuda\CudaException.cs:line 130
at ILGPU.Runtime.Cuda.CudaKernel..ctor(CudaAccelerator accelerator, PTXCompiledKernel kernel, MethodInfo launcher) in C:\Users\17058\Documents\ILGPU\Src\ILGPU\Runtime\Cuda\CudaKernel.cs:line 63
at ILGPU.Runtime.Cuda.CudaAccelerator.CreateKernel(PTXCompiledKernel compiledKernel, MethodInfo launcher) in C:\Users\17058\Documents\ILGPU\Src\ILGPU\Runtime\Cuda\CudaAccelerator.cs:line 388
at ILGPU.Runtime.KernelAccelerator2.LoadKernelInternal(CompiledKernel kernel) in C:\Users\17058\Documents\ILGPU\Src\ILGPU\Runtime\KernelAccelerator.cs:line 52 at ILGPU.Runtime.Accelerator.LoadKernel(CompiledKernel kernel) in C:\Users\17058\Documents\ILGPU\Src\ILGPU\Runtime\KernelLoading.cs:line 78 at ILGPU.Runtime.Accelerator.DefaultKernelLoader.LoadKernel(Accelerator accelerator, CompiledKernel compiledKernel, KernelInfo& kernelInfo) in C:\Users\17058\Documents\ILGPU\Src\ILGPU\Runtime\KernelLoading.cs:line 234 at ILGPU.Runtime.SpecializationCache3.SpecializeKernel(TArgs& args) in C:\Users\17058\Documents\ILGPU\Src\ILGPU\Runtime\SpecializationCache.cs:line 196
at ILGPU.Runtime.SpecializationCache3.GetOrCreateKernel(TArgs args) in C:\Users\17058\Documents\ILGPU\Src\ILGPU\Runtime\SpecializationCache.cs:line 216 at ILGPU.Algorithms.UniqueExtensions.<>c__DisplayClass2_02.b__0(AcceleratorStream stream, ArrayView1 input, ArrayView1 output, ArrayView`1 temp) in C:\Users\17058\Documents\ILGPU\Src\ILGPU.Algorithms\UniqueExtensions.cs:line 186
at AlgorithmsUnique.Program.UniqueKernel(Accelerator accelerator, Int32[] values) in C:\Users\17058\Documents\ILGPU\Samples\AlgorithmsUnique\Program.cs:line 90
at AlgorithmsUnique.Program.Main() in C:\Users\17058\Documents\ILGPU\Samples\AlgorithmsUnique\Program.cs:line 170

Environment

  • ILGPU version: [e.g., 1.5.1]
  • .NET version: [e.g., .NET 7]
  • Operating system: [e.g., Windows 10]
  • Hardware (if GPU-related): [e.g., NVIDIA A5000 mobile]

Steps to reproduce

AlgorithmsUnique.zip

Expected behavior

Can you please fix up this bug?

Additional context

No response

@MoFtZ
Copy link
Collaborator

MoFtZ commented Nov 12, 2024

ptxas error : Entry function 'Kernel_UniqueKernel' uses too much shared data (0x42000 bytes, 0xc000 max)

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

No branches or pull requests

2 participants