Commits on Jul 9, 2021

1. add simple kernel support

   

   upsj committed Jul 9, 2021
   Copy the full SHA

   2d7f0ea View commit details

   Browse the repository at this point in the history

2. Revert "add simple kernel support" …

   This reverts commit da6ed9d.

   

   upsj committed Jul 9, 2021
   Copy the full SHA

   7392892 View commit details

   Browse the repository at this point in the history

3. add direct kernel launch wrappers

   

   upsj committed Jul 9, 2021
   Copy the full SHA

   e65792d View commit details

   Browse the repository at this point in the history

4. set correct nvcc flags

   

   upsj committed Jul 9, 2021
   Copy the full SHA

   3e08abf View commit details

   Browse the repository at this point in the history

5. work around icpc internal compiler error

   

   upsj committed Jul 9, 2021
   Copy the full SHA

   1e8fae1 View commit details

   Browse the repository at this point in the history

6. workaround for MSVC extended-lambda issue

   

   upsj committed Jul 9, 2021
   Copy the full SHA

   4ea1bf0 View commit details

   Browse the repository at this point in the history

7. unify cg and precision_conversion kernel files

   

   upsj committed Jul 9, 2021
   Copy the full SHA

   9785d51 View commit details

   Browse the repository at this point in the history

8. review updates and formatting …

   Co-authored-by: Yuhsiang M. Tsai <yhmtsai@gmail.com>

   

   upsj and yhmtsai committed Jul 9, 2021
   Copy the full SHA

   3b878b1 View commit details

   Browse the repository at this point in the history

9. unify kernel files

   

   upsj committed Jul 9, 2021
   Copy the full SHA

   de7a8f4 View commit details

   Browse the repository at this point in the history

10. make simple kernel launchers free functions

    

    upsj committed Jul 9, 2021
    Copy the full SHA

    225a78d View commit details

    Browse the repository at this point in the history

11. make device mapping more flexible

    

    upsj committed Jul 9, 2021
    Copy the full SHA

    dfe928e View commit details

    Browse the repository at this point in the history

12. move dense operations to simple interface

    

    upsj committed Jul 9, 2021
    Copy the full SHA

    c46e94d View commit details

    Browse the repository at this point in the history

13. add simple Krylov kernels

    

    upsj committed Jul 9, 2021
    Copy the full SHA

    3359b75 View commit details

    Browse the repository at this point in the history

14. eliminate unnecessary stride parameters

    

    upsj committed Jul 9, 2021
    Copy the full SHA

    fff356c View commit details

    Browse the repository at this point in the history

15. replace BLAS axpy and scal by simple kernels

    

    upsj committed Jul 9, 2021
    Copy the full SHA

    4863109 View commit details

    Browse the repository at this point in the history

16. don't preserve stride on create_with_same_config

    

    upsj committed Jul 9, 2021
    Copy the full SHA

    c84c707 View commit details

    Browse the repository at this point in the history

17. review updates …

    * make behavior equivalent to reference
    * rename device_map_impl to to_device_type_impl
    * add comments
    
    Co-authored-by: Yuhsiang Tsai <yhmtsai@gmail.com>
    Co-authored-by: Terry Cojean <terry.cojean@kit.edu>

    

    3 people committed Jul 9, 2021
    Copy the full SHA

    1bb3100 View commit details

    Browse the repository at this point in the history

18. unify simple solver kernel tests

    

    upsj committed Jul 9, 2021
    Copy the full SHA

    147745f View commit details

    Browse the repository at this point in the history

19. add tests for stride handling and zero division

    

    upsj committed Jul 9, 2021
    Copy the full SHA

    d8724ed View commit details

    Browse the repository at this point in the history

20. review updates …

    * remove 1d device_unpack
    * use 2D indices in OpenMP and SYCL
    
    Co-authored-by: Aditya Kashi <aditya.kashi@kit.edu>

    

    upsj and Slaedr committed Jul 9, 2021
    Copy the full SHA

    cd8143b View commit details

    Browse the repository at this point in the history

21. use consistent zero assignment in common kernels

    

    upsj committed Jul 9, 2021
    Copy the full SHA

    b4225bf View commit details

    Browse the repository at this point in the history

22. remove unstable bicg test

    

    upsj committed Jul 9, 2021
    Copy the full SHA

    1ac5aa8 View commit details

    Browse the repository at this point in the history

23. add kernel launch tests

    

    upsj committed Jul 9, 2021
    Copy the full SHA

    33254fb View commit details

    Browse the repository at this point in the history

24. simplify device_unpack_2d_impl parameters

    

    upsj committed Jul 9, 2021
    Copy the full SHA

    415c41b View commit details

    Browse the repository at this point in the history

25. adapt test_install paths

    

    upsj committed Jul 9, 2021
    Copy the full SHA

    6e8ec19 View commit details

    Browse the repository at this point in the history

26. fix kernel launch tests

    

    upsj committed Jul 9, 2021
    Copy the full SHA

    cf11409 View commit details

    Browse the repository at this point in the history

27. add support for DPCPP_SINGLE_MODE to tests

    

    upsj committed Jul 9, 2021
    Copy the full SHA

    4863501 View commit details

    Browse the repository at this point in the history

28. fix clang-cuda tests

    

    upsj committed Jul 9, 2021
    Copy the full SHA

    e2dd2aa View commit details

    Browse the repository at this point in the history

29. fix OpenMP collapse compilation issue

    

    upsj committed Jul 9, 2021
    Copy the full SHA

    a1e93c6 View commit details

    Browse the repository at this point in the history

30. relax solver test precision

    

    upsj committed Jul 9, 2021
    Copy the full SHA

    6de0b69 View commit details

    Browse the repository at this point in the history

31. use SYCL_DEVICE_TYPE/FILTER to make sure using gpu/cpu …

    data cannot be passed between different backends on discrete gpu.

    

    yhmtsai authored and upsj committed Jul 9, 2021
    Copy the full SHA

    1e9f716 View commit details

    Browse the repository at this point in the history

32. remove OpenMP collapse(2)

    

    upsj committed Jul 9, 2021
    Copy the full SHA

    f6c6d98 View commit details

    Browse the repository at this point in the history

33. explicitly unroll OpenMP simple kernels

    

    upsj committed Jul 9, 2021
    Copy the full SHA

    70c16d7 View commit details

    Browse the repository at this point in the history

34. avoid unrolling remainder loop

    

    upsj committed Jul 9, 2021
    Copy the full SHA

    e4879b6 View commit details

    Browse the repository at this point in the history

35. unroll width 4

    

    upsj committed Jul 9, 2021
    Copy the full SHA

    39b7ed3 View commit details

    Browse the repository at this point in the history

36. explicitly unroll remainder with width 4

    

    upsj committed Jul 9, 2021
    Copy the full SHA

    64cccd7 View commit details

    Browse the repository at this point in the history

37. add kernel_launch to format_header.sh config

    

    upsj committed Jul 9, 2021
    Copy the full SHA

    000b973 View commit details

    Browse the repository at this point in the history

38. keep stride in solver create_with_same_config …

    This means we need to provide a default stride
    to the simple solver kernels, which we implement using
    a separate run_kernel_solver function

    

    upsj committed Jul 9, 2021
    Copy the full SHA

    ef37edb View commit details

    Browse the repository at this point in the history

39. update documentation and assertions

    

    upsj committed Jul 9, 2021
    Copy the full SHA

    2aa080f View commit details

    Browse the repository at this point in the history

40. fix remaining create_with_same_config test

    

    upsj committed Jul 9, 2021
    Copy the full SHA

    3630a7a View commit details

    Browse the repository at this point in the history

41. review updates

    

    upsj committed Jul 9, 2021
    Copy the full SHA

    1c87a35 View commit details

    Browse the repository at this point in the history

42. extract common code for simple kernel launch

    

    upsj committed Jul 9, 2021
    Copy the full SHA

    87287a4 View commit details

    Browse the repository at this point in the history

43. work around DPC++ use-after-free

    

    upsj committed Jul 9, 2021
    Copy the full SHA

    a4d6278 View commit details

    Browse the repository at this point in the history

44. improve documentation for kernel_launch

    

    upsj committed Jul 9, 2021
    Copy the full SHA

    1d29893 View commit details

    Browse the repository at this point in the history

45. move simple dense kernel tests to common tests

    

    upsj committed Jul 9, 2021
    Copy the full SHA

    1cb4705 View commit details

    Browse the repository at this point in the history

46. increase BiCGSTAB test tolerances

    

    upsj committed Jul 9, 2021
    Copy the full SHA

    9c5df14 View commit details

    Browse the repository at this point in the history

47. fix formatting

    

    upsj committed Jul 9, 2021
    Copy the full SHA

    5dc8aac View commit details

    Browse the repository at this point in the history

48. move kernel_launch defines into test CMakeLists

    

    upsj committed Jul 9, 2021
    Copy the full SHA

    8e3a2bb View commit details

    Browse the repository at this point in the history

49. update test_install path for formatting job

    

    upsj committed Jul 9, 2021
    Copy the full SHA

    56948b3 View commit details

    Browse the repository at this point in the history

50. rename compile tests inside test/

    

    upsj committed Jul 9, 2021
    Copy the full SHA

    dc6f4d7 View commit details

    Browse the repository at this point in the history

51. remove circular dependency in kernel_launch header

    

    upsj committed Jul 9, 2021
    Copy the full SHA

    382f501 View commit details

    Browse the repository at this point in the history

52. review updates …

    add missing includes for more IWYU-like structure
    
    Co-authored-by: Terry Cojean <terry.cojean@kit.edu>
    Co-authored-by: Thomas Grützmacher <thomas.gruetzmacher@kit.edu>

    

    3 people committed Jul 9, 2021
    Copy the full SHA

    c29627e View commit details

    Browse the repository at this point in the history

53. remove magic numbers for OpenMP unrolling

    

    upsj committed Jul 9, 2021
    Copy the full SHA

    0ef51f1 View commit details

    Browse the repository at this point in the history

54. fix include position for kernel_launch tests

    

    upsj committed Jul 9, 2021
    Copy the full SHA

    65a0a65 View commit details

    Browse the repository at this point in the history

55. fix single-precision dpcpp dense kernel tests

    

    upsj committed Jul 9, 2021
    Copy the full SHA

    8828510 View commit details

    Browse the repository at this point in the history

56. skip DPCPP solver apply tests

    

    upsj committed Jul 9, 2021
    Copy the full SHA

    29af274 View commit details

    Browse the repository at this point in the history