HigherOrderCO / Bend

A massively parallel, high-level programming language
https://higherorderco.com
Apache License 2.0
17.48k stars 428 forks source link

Failed to launch kernels. #715

Closed noskill closed 2 months ago

noskill commented 2 months ago

Reproducing the behavior

running example code with cuda backend fails:

sequential_sum.bend

# Defines the function Sum with two parameters: start and target
def Sum(start, target):
  if start == target:
    # If the value of start is the same as target, returns start.
    return start
  else:
    # If start is not equal to target, recursively call Sum with
    # start incremented by 1, and add the result to start.
    return start + Sum(start + 1, target)  

def main():
  # This translates to (1 + (2 + (3 + (...... + (999999 + 1000000)))))
  # Note that this will overflow the maximum value of a number in Bend
  return Sum(1, 1_000_000)

bend run-cu sequential_sum.bend

Failed to launch kernels. Error code: invalid argument. Errors: Failed to parse result from HVM.

System Settings

$ bend --version bend-lang 0.2.36

$ hvm --version hvm 2.0.22

$ nvcc --version nvcc: NVIDIA (R) Cuda compiler driver Copyright (c) 2005-2024 NVIDIA Corporation Built on Wed_Aug_14_10:10:22_PDT_2024 Cuda compilation tools, release 12.6, V12.6.68 Build cuda_12.6.r12.6/compiler.34714021_0

developedby commented 2 months ago

Hi, the 1080 Ti is not currently supported by Bend. You can read a bit more here https://github.com/HigherOrderCO/Bend/blob/main/FAQ.md