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.

Sign up for GitHub

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

Jump to bottom

gpu stanmathcl [WIP] #655

Closed
wants to merge 112 commits into from
Closed

gpu stanmathcl [WIP] #655

wants to merge 112 commits into from

Commits on Dec 5, 2017

  1. initial commit

    @SteveBronder
    SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    ba4dd04 View commit details
    Browse the repository at this point in the history

  2. random fixings

    @SteveBronder
    SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    2eef1f2 View commit details
    Browse the repository at this point in the history

  3. lintr things

    @SteveBronder
    SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    0c6efac View commit details
    Browse the repository at this point in the history

  4. Started adding templating for the gpu functions

    @SteveBronder
    SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    b0f6dc4 View commit details
    Browse the repository at this point in the history

  5. Built tests and rev cholesk with new code. Seg Fault on test

    @SteveBronder
    SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    ac0a834 View commit details
    Browse the repository at this point in the history

  6. moved basic_matrix_gpu funcs to funcs

    @SteveBronder
    SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    8db822c View commit details
    Browse the repository at this point in the history

  7. simple test for chol segfault

    @SteveBronder
    SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    f3a300c View commit details
    Browse the repository at this point in the history

  8. added GPU tests, failing

    @SteveBronder
    SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    f96cb58 View commit details
    Browse the repository at this point in the history

  9. Trying to run tests on tests

    @SteveBronder
    SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    264d0bf View commit details
    Browse the repository at this point in the history

  10. added some things for checking test vals

    @SteveBronder
    SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    598fdb0 View commit details
    Browse the repository at this point in the history

  11. Update code to only return lower left triangular for cholesky

    @SteveBronder
    SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    96a89ad View commit details
    Browse the repository at this point in the history

  12. gpu_chol now gives back the original matrix when the size if zero. Te… 

    …sts still failing
    @SteveBronder
    SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    83916c7 View commit details
    Browse the repository at this point in the history

  13. tests pass if the second condition of check_pos_definite() is comment… 

    …ed out
    @SteveBronder
    SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    3cf7021 View commit details
    Browse the repository at this point in the history

  14. tests pass if the second condition of check_pos_definite() is comment… 

    …ed out
    @SteveBronder
    SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    af87eb3 View commit details
    Browse the repository at this point in the history

  15. Inital commit with the following changes: 

    - replaced rows()*cols() with size() from Eigen, added size() to matrix_gpu
- pulled the enums in the stan::math namespace
- removed stan::math qualifiers when inside the namespace
- suffixed the matrix_gpu class variables (rows_,cols_), added rows() and cols() getter function
- suffixed the ocl class variables
- const added to catch (cl::Error&)
- changed the conditions for matrix dimension checks
- added spaces after for and if, starting { to the condition line
- added initializers for the constructor
- returning context, queue & buffer by reference
- change dynamicly allocated arrays to std::vector
    @rok-cesnovar @SteveBronder
    rok-cesnovar authored and SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    f02225f View commit details
    Browse the repository at this point in the history

  16. - reading kernels from files

    @rok-cesnovar @SteveBronder
    rok-cesnovar authored and SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    e91a870 View commit details
    Browse the repository at this point in the history

  17. all tests pass on smaller block sizes for inverse

    @rok-cesnovar @SteveBronder
    rok-cesnovar authored and SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    618f145 View commit details
    Browse the repository at this point in the history

  18. - fixed the inverse semifrequent nan error, all test pass

    @rok-cesnovar @SteveBronder
    rok-cesnovar authored and SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    e851141 View commit details
    Browse the repository at this point in the history

  19. - missing kernel changes from the previous commit

    @rok-cesnovar @SteveBronder
    rok-cesnovar authored and SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    f4cb8eb View commit details
    Browse the repository at this point in the history

  20. - added nan value checks for the GPU 

    - added zeros on diagonal checks for the GPU
- copy of lower to upper triangular is made on the GPU, not sure if still needed
    @rok-cesnovar @SteveBronder
    rok-cesnovar authored and SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    cfa79f0 View commit details
    Browse the repository at this point in the history

  21. - changed the kernel source reading

    @rok-cesnovar @SteveBronder
    rok-cesnovar authored and SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    3385baf View commit details
    Browse the repository at this point in the history

  22. - use of a more safe transpose, minor cleanup

    @rok-cesnovar @SteveBronder
    rok-cesnovar authored and SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    66da004 View commit details
    Browse the repository at this point in the history

  23. code passes most of cpplint

    @SteveBronder
    SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    103d92b View commit details
    Browse the repository at this point in the history

  24. passing cpplint

    @SteveBronder
    SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    5c6ea69 View commit details
    Browse the repository at this point in the history

  25. - removed the loops for transposing the matrices before copy to GPU

    @rok-cesnovar @SteveBronder
    rok-cesnovar authored and SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    4590b03 View commit details
    Browse the repository at this point in the history

  26. - changed all dimension checks to exsitent /prim/mat/err checks 

    - added overloaded functions for check_square and check_matching_dims
    @rok-cesnovar @SteveBronder
    rok-cesnovar authored and SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    8bd0b8e View commit details
    Browse the repository at this point in the history

  27. - added doxygen comments; only prim and rev cholesky_decompose_gpu.hp… 

    …p are now missing comments
    @rok-cesnovar @SteveBronder
    rok-cesnovar authored and SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    b3386c5 View commit details
    Browse the repository at this point in the history

  28. - added the rest of doxygen comments

    @rok-cesnovar @SteveBronder
    rok-cesnovar authored and SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    38b65c3 View commit details
    Browse the repository at this point in the history

  29. - added the copy_submatrix function

    @rok-cesnovar @SteveBronder
    rok-cesnovar authored and SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    64e1ff4 View commit details
    Browse the repository at this point in the history

  30. - passes cpplint

    @rok-cesnovar @SteveBronder
    rok-cesnovar authored and SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    fa56656 View commit details
    Browse the repository at this point in the history

  31. - implemented the blocked version of the rev/cholesky_decompose on th… 

    …e GPU

- added copy_submatrix
    @rok-cesnovar @SteveBronder
    rok-cesnovar authored and SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    17434a9 View commit details
    Browse the repository at this point in the history

  32. shortened multiply_with_scalar to multiply. Started builing tests

    @SteveBronder
    SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    c1a8e5e View commit details
    Browse the repository at this point in the history

  33. GPU matrix tests all passing

    @SteveBronder
    SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    a67548a View commit details
    Browse the repository at this point in the history

  34. Added OpenCL headers

    @SteveBronder
    SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    2578605 View commit details
    Browse the repository at this point in the history

  35. fix link to OpenCL from hpp to h

    @SteveBronder
    SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    80c41cf View commit details
    Browse the repository at this point in the history

  36. fix link to OpenCL from hpp to h

    @SteveBronder
    SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    a7fe86a View commit details
    Browse the repository at this point in the history

  37. Removed old OpenCL, using C++ opencl header file

    @SteveBronder
    SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    e7e8b3e View commit details
    Browse the repository at this point in the history

  38. Change OpenCL header

    @SteveBronder
    SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    6686a4f View commit details
    Browse the repository at this point in the history

  39. opencl headers

    @SteveBronder
    SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    760daaf View commit details
    Browse the repository at this point in the history

  40. placed tranpose test in wrong place

    @SteveBronder
    SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    6047f22 View commit details
    Browse the repository at this point in the history

  41. update check_diagonal_zero test

    @SteveBronder
    SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    9684668 View commit details
    Browse the repository at this point in the history

  42. fix check_gpu_tests

    @SteveBronder
    SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    5d7cfd6 View commit details
    Browse the repository at this point in the history

  43. Forgot : and ; in check_gpu_test

    @SteveBronder
    SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    22f3ded View commit details
    Browse the repository at this point in the history

  44. add test for zero matrix, identity matrix, and subtract

    @SteveBronder
    SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    31d110a View commit details
    Browse the repository at this point in the history

  45. - fixed seg fault when creating matrices with size 0

    @rok-cesnovar @SteveBronder
    rok-cesnovar authored and SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    e0fafde View commit details
    Browse the repository at this point in the history

  46. - order of parameters in subtract now matches the add/multiply order … 

    …of parameters
    @rok-cesnovar @SteveBronder
    rok-cesnovar authored and SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    3052592 View commit details
    Browse the repository at this point in the history

  47. - cpplint errors cleanup

    @rok-cesnovar @SteveBronder
    rok-cesnovar authored and SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    af7fa93 View commit details
    Browse the repository at this point in the history

  48. - changed add to copy on output, tests changed

    @rok-cesnovar @SteveBronder
    rok-cesnovar authored and SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    0802f69 View commit details
    Browse the repository at this point in the history

  49. - changed subtract to copy on output, test fixed

    @rok-cesnovar @SteveBronder
    rok-cesnovar authored and SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    7ae8263 View commit details
    Browse the repository at this point in the history

  50. - transpose copies on output

    @rok-cesnovar @SteveBronder
    rok-cesnovar authored and SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    4c7bb3a View commit details
    Browse the repository at this point in the history

  51. - all multiplies except for diagonal copy on output

    @rok-cesnovar @SteveBronder
    rok-cesnovar authored and SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    c43a944 View commit details
    Browse the repository at this point in the history

  52. - changed the use of multiply in rev/cholesky_decompose

    @rok-cesnovar @SteveBronder
    rok-cesnovar authored and SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    8757bb5 View commit details
    Browse the repository at this point in the history

  53. - inverse copy on output

    @rok-cesnovar @SteveBronder
    rok-cesnovar authored and SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    c4501b4 View commit details
    Browse the repository at this point in the history

  54. - fixed /rev/cholesky_decompose_gpu with the copy on output functions

    @rok-cesnovar @SteveBronder
    rok-cesnovar authored and SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    a1ff9ce View commit details
    Browse the repository at this point in the history

  55. added tests for matrix_gpu, the copy functions, and inverse function.

    @SteveBronder
    SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    6109dc8 View commit details
    Browse the repository at this point in the history

  56. Fix up tests for add and subtract

    @SteveBronder
    SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    13b8410 View commit details
    Browse the repository at this point in the history

  57. update check_nan tests. We have nothing to test infinite values which… 

    … we probably should.
    @SteveBronder
    SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    51f688d View commit details
    Browse the repository at this point in the history

  58. add AMD OpenCL install to travis.yml so that CPU version of OpenCL is… 

    … used for testing on travis
    @SteveBronder
    SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    1751f7e View commit details
    Browse the repository at this point in the history

  59. - set device filter to CPU to run tests

    @rok-cesnovar @SteveBronder
    rok-cesnovar authored and SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    db6f798 View commit details
    Browse the repository at this point in the history

  60. - minor test fixes, added the AMD double pragma to the kernels

    @rok-cesnovar @SteveBronder
    rok-cesnovar authored and SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    81f6249 View commit details
    Browse the repository at this point in the history

  61. - cpplint cleanup, kernel init test

    @rok-cesnovar @SteveBronder
    rok-cesnovar authored and SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    df71123 View commit details
    Browse the repository at this point in the history

  62. - checking symmetric in /prim/cholesky_decompose

    @rok-cesnovar @SteveBronder
    rok-cesnovar authored and SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    153fd31 View commit details
    Browse the repository at this point in the history

  63. add POST_LDLIBS when compiling .o file, this avoids warnings in clang… 

    … and allows g++ to compile
    @SteveBronder
    SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    c4e48f2 View commit details
    Browse the repository at this point in the history

  64. Update AMD SDK to 3.0

    @SteveBronder
    SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    c8ca938 View commit details
    Browse the repository at this point in the history

  65. move libopencl.so symbolic link line in .travis.yml

    @SteveBronder
    SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    f0c4a61 View commit details
    Browse the repository at this point in the history

  66. - context passed by reference 

    - changes to the cholesky_decompose 1st step kernel
- CPU passes precision tests
    @rok-cesnovar @SteveBronder
    rok-cesnovar authored and SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    2133c14 View commit details
    Browse the repository at this point in the history

  67. move libopencl.so symbolic link line in .travis.yml (reverted from co… 

    …mmit e4816cc)
    @rok-cesnovar @SteveBronder
    rok-cesnovar authored and SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    f35359b View commit details
    Browse the repository at this point in the history

  68. Update AMD SDK to 3.0 (reverted from commit e1ed076)

    @rok-cesnovar @SteveBronder
    rok-cesnovar authored and SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    0ad3dc7 View commit details
    Browse the repository at this point in the history

  69. - default work-group size is 16x16 to support all CPUs 

    - cpplint cleanup of .cl files
    @rok-cesnovar @SteveBronder
    rok-cesnovar authored and SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    bac8c78 View commit details
    Browse the repository at this point in the history

  70. - block in cholesky step to 32

    @rok-cesnovar @SteveBronder
    rok-cesnovar authored and SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    a93f5bb View commit details
    Browse the repository at this point in the history

  71. - added function name to check_opencl for easier debugging

    @rok-cesnovar @SteveBronder
    rok-cesnovar authored and SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    d1d9d0d View commit details
    Browse the repository at this point in the history

  72. - testing the Travis OpenCL device

    @rok-cesnovar @SteveBronder
    rok-cesnovar authored and SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    80fd359 View commit details
    Browse the repository at this point in the history

  73. - smaller multiply size for CPU OpenCL

    @rok-cesnovar @SteveBronder
    rok-cesnovar authored and SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    53a5ca6 View commit details
    Browse the repository at this point in the history

  74. - missing kernel size

    @rok-cesnovar @SteveBronder
    rok-cesnovar authored and SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    8cb7c76 View commit details
    Browse the repository at this point in the history

  75. - added missing checks in copy_submatrix

    @rok-cesnovar @SteveBronder
    rok-cesnovar authored and SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    f1afe40 View commit details
    Browse the repository at this point in the history

  76. - matrix multiply with matrix of size() 0 returns immediately 

    - cpplint and doxygen cleanup
    @rok-cesnovar @SteveBronder
    rok-cesnovar authored and SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    96ac7df View commit details
    Browse the repository at this point in the history

  77. - added multiply with self transposed 

    - rewrote the cholesky 2nd and 3rd step using existing /prim GPU functions
- speedup of choleskyy
    @rok-cesnovar @SteveBronder
    rok-cesnovar authored and SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    ee2c22c View commit details
    Browse the repository at this point in the history

  78. - removed test for removed kernels 

    - tests for multiply_with_self_transposed
    @rok-cesnovar @SteveBronder
    rok-cesnovar authored and SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    b39b9b1 View commit details
    Browse the repository at this point in the history

  79. - cholesky now reuses the inverse and zeros kernels

    @rok-cesnovar @SteveBronder
    rok-cesnovar authored and SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    c4a5f78 View commit details
    Browse the repository at this point in the history

  80. - two level cholesky decompose

    @rok-cesnovar @SteveBronder
    rok-cesnovar authored and SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    c54ba4a View commit details
    Browse the repository at this point in the history

  81. - added check_symmetric_gpu, tests 

    - set block sizes that give best overall times for different sizes
    @rok-cesnovar @SteveBronder
    rok-cesnovar authored and SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    a7f0436 View commit details
    Browse the repository at this point in the history

  82. removed mac specific compilations for OpenCL

    @SteveBronder
    SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    4a91a88 View commit details
    Browse the repository at this point in the history

  83. fix cpplint errors

    @SteveBronder
    SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    599d11b View commit details
    Browse the repository at this point in the history

  84. - updated the GPU kernels for readability 

    - removed the unused kernels
    @rok-cesnovar @SteveBronder
    rok-cesnovar authored and SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    103a6f7 View commit details
    Browse the repository at this point in the history

  85. - cleanup of tests due to removed kernels

    @rok-cesnovar @SteveBronder
    rok-cesnovar authored and SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    801a065 View commit details
    Browse the repository at this point in the history

  86. - switched all kernels to column-major

    @rok-cesnovar @SteveBronder
    rok-cesnovar authored and SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    51861d4 View commit details
    Browse the repository at this point in the history

  87. - optimizations on the matrix multiplies

    @rok-cesnovar @SteveBronder
    rok-cesnovar authored and SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    1227672 View commit details
    Browse the repository at this point in the history

  88. - cleanup of check_opencl, does not throws on 0 

    - changed matrix_gpu try/catch scopes
- changed size checks to use size()
- rows_, cols_ moved to private, const added
    @rok-cesnovar @SteveBronder
    rok-cesnovar authored and SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    754ed8d View commit details
    Browse the repository at this point in the history

  89. - more changes for matrices with const rows and cols

    @rok-cesnovar @SteveBronder
    rok-cesnovar authored and SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    1f9fbe6 View commit details
    Browse the repository at this point in the history

  90. - appropriate scopes for openCL try/catch for all added functions 

    - removed _gpu form check functions
    @rok-cesnovar @SteveBronder
    rok-cesnovar authored and SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    8aca5b1 View commit details
    Browse the repository at this point in the history

  91. - removed the use of enqueueCopyBuffer

    @rok-cesnovar @SteveBronder
    rok-cesnovar authored and SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    70deee5 View commit details
    Browse the repository at this point in the history

  92. Add docs for pos_def_check in check_diagonal_zeros and check_nan for … 

    …GPU funcs
    @SteveBronder
    SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    ab33aef View commit details
    Browse the repository at this point in the history

  93. Attempt to merge / rebase fork with origin/stanMathcl

    @SteveBronder
    SteveBronder committed Dec 5, 2017
    Configuration menu
    Copy the full SHA
    baf39a1 View commit details
    Browse the repository at this point in the history

Commits on Dec 6, 2017

  1. change deevice type to cpu

    @SteveBronder
    SteveBronder committed Dec 6, 2017
    Configuration menu
    Copy the full SHA
    70ce68b View commit details
    Browse the repository at this point in the history

  2. fix lint warnings in test and header files

    @SteveBronder
    SteveBronder committed Dec 6, 2017
    Configuration menu
    Copy the full SHA
    242b823 View commit details
    Browse the repository at this point in the history

  3. add -framework OpenCL and a check for mac in the default_compiler_opt… 

    …ions
    @SteveBronder
    SteveBronder committed Dec 6, 2017
    Configuration menu
    Copy the full SHA
    77d331c View commit details
    Browse the repository at this point in the history

  4. Change device type to CPU

    @SteveBronder
    SteveBronder committed Dec 6, 2017
    Configuration menu
    Copy the full SHA
    9010fb6 View commit details
    Browse the repository at this point in the history

Commits on Dec 7, 2017

  1. added define for OpenCL expections in check_opencl. In a previous com… 

    …mit I was auto changing all spaces to tabs and accidentally did that to boost. This is corrected here
    @SteveBronder
    SteveBronder committed Dec 7, 2017
    Configuration menu
    Copy the full SHA
    0717822 View commit details
    Browse the repository at this point in the history

  2. Add POST_LDLIBS to multiple_translation_units test

    @SteveBronder
    SteveBronder committed Dec 7, 2017
    Configuration menu
    Copy the full SHA
    1cd8ab6 View commit details
    Browse the repository at this point in the history

Commits on Dec 9, 2017

  1. inlined functions to avoid multiple translation unit error

    @SteveBronder
    SteveBronder committed Dec 9, 2017
    Configuration menu
    Copy the full SHA
    cde6ff7 View commit details
    Browse the repository at this point in the history

  2. ...

    @SteveBronder
    SteveBronder committed Dec 9, 2017
    Configuration menu
    Copy the full SHA
    58e380b View commit details
    Browse the repository at this point in the history

  3. add static to global kernel maps and inlined copy function

    @SteveBronder
    SteveBronder committed Dec 9, 2017
    Configuration menu
    Copy the full SHA
    2c43e0e View commit details
    Browse the repository at this point in the history

  4. include <exception> for check_opencl

    @SteveBronder
    SteveBronder committed Dec 9, 2017
    Configuration menu
    Copy the full SHA
    26f79dc View commit details
    Browse the repository at this point in the history

  5. add scale file for domain error in check_opencl

    @SteveBronder
    SteveBronder committed Dec 9, 2017
    Configuration menu
    Copy the full SHA
    2b1a6a3 View commit details
    Browse the repository at this point in the history

  6. add char* to check_ocl_error

    @SteveBronder
    SteveBronder committed Dec 9, 2017
    Configuration menu
    Copy the full SHA
    e686c52 View commit details
    Browse the repository at this point in the history

Commits on Dec 11, 2017

  1. Use a string to throw a domain error for OpenCL failures

    @SteveBronder
    SteveBronder committed Dec 11, 2017
    Configuration menu
    Copy the full SHA
    895ce30 View commit details
    Browse the repository at this point in the history

  2. ...

    @SteveBronder
    SteveBronder committed Dec 11, 2017
    Configuration menu
    Copy the full SHA
    8077b41 View commit details
    Browse the repository at this point in the history

  3. Add constant tolerance header file to check_gpu.hpp

    @SteveBronder
    SteveBronder committed Dec 11, 2017
    Configuration menu
    Copy the full SHA
    54d9c19 View commit details
    Browse the repository at this point in the history

  4. add basic_matrix_gpu to headers for multipy_gpu

    @SteveBronder
    SteveBronder committed Dec 11, 2017
    Configuration menu
    Copy the full SHA
    eb174f2 View commit details
    Browse the repository at this point in the history

Commits on Dec 28, 2017

  1. update AMD SDK download script

    @SteveBronder
    SteveBronder committed Dec 28, 2017
    Configuration menu
    Copy the full SHA
    fa0376b View commit details
    Browse the repository at this point in the history

Commits on Jan 6, 2018

  1. Max workgroup size is considered for determining workgroup sizes

    @rok-cesnovar
    rok-cesnovar committed Jan 6, 2018
    Configuration menu
    Copy the full SHA
    64e0317 View commit details
    Browse the repository at this point in the history

  2. - removed residual debug info

    @rok-cesnovar
    rok-cesnovar committed Jan 6, 2018
    Configuration menu
    Copy the full SHA
    b2be485 View commit details
    Browse the repository at this point in the history