@hackage / accelerate-cuda

Accelerate backend for NVIDIA GPUs

Version Deprecated package0.17.0.0

About

Metadata

  • Last updated , by TrevorMcDonell
  • License BSD-3-Clause
  • Categories Compilers and Interpreters, Parallelism, Concurrency

  • Maintained by: Trevor L. McDonell <tmcdonell@cse.unsw.edu.au>

  • Lottery factor: 0

Links

Installation

This package has been deprecated in favour of

Tested Compilers

  1. 9.12.2
  2. 9.12.1
  3. 9.10.3
  4. 9.10.2
  5. 9.10.1
  6. 9.8.4
  7. 9.8.3
  8. 9.8.2
  9. 9.8.1
  10. 9.6.7
  11. 9.6.6
  12. 9.6.5
  13. 9.6.4
  14. 9.6.3
  15. 9.6.2
  16. 9.6.1
  17. 9.4.8
  18. 9.4.7
  19. 9.4.6
  20. 9.4.5
  21. 9.4.4
  22. 9.4.3
  23. 9.4.2
  24. 9.4.1
  25. 9.2.8
  26. 9.2.7
  27. 9.2.6
  28. 9.2.5
  29. 9.2.4
  30. 9.2.3
  31. 9.2.2
  32. 9.2.1
  33. 9.0.2
  34. 9.0.1
  35. 8.10.7
  36. 8.10.6
  37. 8.10.5
  38. 8.10.4
  39. 8.10.3
  40. 8.10.2
  41. 8.10.1
  42. 8.8.4
  43. 8.8.3
  44. 8.8.2
  45. 8.8.1
  46. 8.6.5
  47. 8.6.4
  48. 8.6.3
  49. 8.6.2
  50. 8.6.1
  51. 8.4.4
  52. 8.4.3
  53. 8.4.2
  54. 8.4.1
  55. 8.2.2
  56. 8.0.2
  57. 7.10.3

Package Flags

Use the -f option with cabal commands to enable flags

    debug (off by default)

    Enable tracing message flags. These are read from the command-line arguments, which is convenient but may cause problems interacting with the user program, so is not enabled by default. The available options:

    • -ddump-cc: print the generated code, kernel table management information, nvcc compiler warnings, and thread & resource statistics

    • -ddebug-cc: compile code with debugging symbols, suitable for 'cuda-gdb'

    • -ddump-exec: print each kernel name as it is invoked

    • -ddump-gc: print memory management information

    • -dverbose: other, uncategorised messages

    • -fflush-cache: delete the persistent kernel cache

    bounds-checks (on by default)

    Enable bounds checking

    unsafe-checks (off by default)

    Enable bounds checking in unsafe operations

    internal-checks (off by default)

    Enable internal consistency checks

Readme

This backend has been deprecated in favour of accelerate-llvm-ptx.

This library implements a backend for the Accelerate language instrumented for parallel execution on CUDA-capable NVIDIA GPUs. For further information, refer to the main Accelerate package: http://hackage.haskell.org/package/accelerate

To use this backend you will need:

  1. A CUDA-enabled NVIDIA GPU with, for full functionality, compute capability 1.3 or greater. See the table on Wikipedia for supported GPUs: http://en.wikipedia.org/wiki/CUDA#Supported_GPUs

  2. The CUDA SDK, available from the NVIDIA Developer Zone: http://developer.nvidia.com/cuda-downloads

See the Haddock documentation for additional information related to using this backend.

Compile modules that use the CUDA backend with the -threaded flag.