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

Merging batch-dpcpp into batch-develop #1298

Closed
wants to merge 140 commits into from
Closed

Merging batch-dpcpp into batch-develop #1298

wants to merge 140 commits into from

Commits on Oct 15, 2023

  1. Log dpcpp version info

    @pratikvn
    pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    653a167 View commit details
    Browse the repository at this point in the history

  2. Add a batch_struct for dpcpp

    @pratikvn
    pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    9e78c4a View commit details
    Browse the repository at this point in the history

  3. update headers in dpcpp kernels

    @pratikvn
    pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    68e512d View commit details
    Browse the repository at this point in the history

  4. implemented spmv and advanced_spmv

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    a1ce184 View commit details
    Browse the repository at this point in the history

  5. spmv compiled, but not batch_scale

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    41d649c View commit details
    Browse the repository at this point in the history

  6. Instantiate for int32 only

    @pratikvn
    pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    e096ed0 View commit details
    Browse the repository at this point in the history

  7. bath_csr compiled but did not pass tests

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    916545b View commit details
    Browse the repository at this point in the history

  8. DiagScale still didn't pass the tests

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    2221d99 View commit details
    Browse the repository at this point in the history

  9. add batch_csr kernels passed the test

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    3d791e4 View commit details
    Browse the repository at this point in the history

  10. saved current state

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    23e1a37 View commit details
    Browse the repository at this point in the history

  11. implemented batch_dense_kernels, all tests passed

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    bb09fff View commit details
    Browse the repository at this point in the history

  12. implemented batch diagonal kernels

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    b750f33 View commit details
    Browse the repository at this point in the history

  13. implemented batch ell kernels

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    cec05a8 View commit details
    Browse the repository at this point in the history

  14. Move batch matrix kernels to header files

    @pratikvn
    pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    d14dc0a View commit details
    Browse the repository at this point in the history

  15. Add batch preconditioner headers

    @pratikvn
    pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    58f39e0 View commit details
    Browse the repository at this point in the history

  16. Enable batch dpcpp solver dispatch

    @pratikvn
    pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    5a81d3f View commit details
    Browse the repository at this point in the history

  17. Add dcppp batch stop criteria

    @pratikvn
    pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    bda56f6 View commit details
    Browse the repository at this point in the history

  18. Add dpcpp batch loggers

    @pratikvn
    pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    63158c0 View commit details
    Browse the repository at this point in the history

  19. Add template batch_cg kernel header

    @pratikvn
    pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    53541c8 View commit details
    Browse the repository at this point in the history

  20. moved item_ct1 as the last argument of device kernels, rm unused head… 

    …er includes
    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    a6e1673 View commit details
    Browse the repository at this point in the history

  21. renamed some device functions

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    47f1c89 View commit details
    Browse the repository at this point in the history

  22. implemented basic preconditioners for batch, did not test them yet

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    83e9682 View commit details
    Browse the repository at this point in the history

  23. implemented batch-cg, did not test them yet

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    b6f82a1 View commit details
    Browse the repository at this point in the history

  24. tmp save

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    6c52440 View commit details
    Browse the repository at this point in the history

  25. pushed batch_vector_kernels.hpp

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    69c9c0e View commit details
    Browse the repository at this point in the history

  26. Move .hpp files to .hpp.inc and remove namespaces

    @pratikvn
    pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    5b4022b View commit details
    Browse the repository at this point in the history

  27. Usae .hpp.inc in correct namespaces

    @pratikvn
    pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    e06776a View commit details
    Browse the repository at this point in the history

  28. Fix function signatures and remove exceptions

    @pratikvn
    pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    ca7858d View commit details
    Browse the repository at this point in the history

  29. Enable batch_cg dpcpp tests

    @pratikvn
    pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    2102c94 View commit details
    Browse the repository at this point in the history

  30. updated header files

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    5c321cb View commit details
    Browse the repository at this point in the history

  31. added a new line after licence text

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    93135e5 View commit details
    Browse the repository at this point in the history

  32. cg passed 4/7 tests

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    534b3e4 View commit details
    Browse the repository at this point in the history

  33. - batch_cg passed 4/7 tests. 

    - implemented `batch_jacobi_kernels`. All `scalar_jacobi` tests passed
  but not the `block_jacobi`.
- changed all `.hpp` file into `.hpp.inc` and changes their namespaces
  accordingly.
    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    23002f6 View commit details
    Browse the repository at this point in the history

  34. rm old files

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    ecba3df View commit details
    Browse the repository at this point in the history

  35. batch_cg passed tests

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    7670da1 View commit details
    Browse the repository at this point in the history

  36. minor changes, these changes need to be redo in the future

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    3d2ca3f View commit details
    Browse the repository at this point in the history

  37. saved current state

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    7fe1198 View commit details
    Browse the repository at this point in the history

  38. batch_cg passed all tests when using slm

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    1ec5ba0 View commit details
    Browse the repository at this point in the history

  39. passing the accessors with their ptrs

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    6337228 View commit details
    Browse the repository at this point in the history

  40. slm calculation does not involve a matrix_size substraction

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    056f002 View commit details
    Browse the repository at this point in the history

  41. Revert "passing the accessors with their ptrs" 

    This reverts commit d24a473.
    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    a164182 View commit details
    Browse the repository at this point in the history

  42. Revert "slm calculation does not involve a matrix_size substraction" 

    This reverts commit 8ffdde1.
    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    de45f31 View commit details
    Browse the repository at this point in the history

  43. passing accessors via its pointers

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    4e35f57 View commit details
    Browse the repository at this point in the history

  44. enforcing subgroup_size for all block_jacobi rountines

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    ab0c84c View commit details
    Browse the repository at this point in the history

  45. fixed

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    50498f3 View commit details
    Browse the repository at this point in the history

  46. fixed

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    656ba38 View commit details
    Browse the repository at this point in the history

  47. - Batch Block Jacobi does not work yet. The issue might come from the 

      fact that dpcpp does not allow subgroup size smaller than 16, thus
when block_size is smaller than 16, the kernel gives wrong results.
    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    e8da1d5 View commit details
    Browse the repository at this point in the history

  48. minor changes for csr single matvec

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    1488e5c View commit details
    Browse the repository at this point in the history

  49. marked BatchBlockJacobi as NOT_IMPLEMENTED

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    415cd02 View commit details
    Browse the repository at this point in the history

  50. implemented batch_bicgstab, passed all tests

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    63b9280 View commit details
    Browse the repository at this point in the history

  51. corrected include path for .hpp.in with cmake preprocessor

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    15ea037 View commit details
    Browse the repository at this point in the history

  52. inline all 'device' functions

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    d2888af View commit details
    Browse the repository at this point in the history

  53. Format files 

    Co-authored-by: Pratik Nayak <pratikvn@pm.me>
    @ginkgo-bot @pratikvn
    ginkgo-bot and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    443d268 View commit details
    Browse the repository at this point in the history

  54. implemented batch_richardson, passed all tests

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    d2ac213 View commit details
    Browse the repository at this point in the history

  55. enabled test compilations for dpcpp

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    489c180 View commit details
    Browse the repository at this point in the history

  56. - implemented batch_lower_trs_kernels. For now, the kernel calls the 

      naive_lower_trsv. The warp_synchronous alg needs to be re-write in the
way that it does not invole the  while loop.
    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    4e136aa View commit details
    Browse the repository at this point in the history

  57. implemented batch_upper_trs

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    4b97df1 View commit details
    Browse the repository at this point in the history

  58. enabling batch_idr test

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    c923247 View commit details
    Browse the repository at this point in the history

  59. wrote onemkl bindings for getrf and getrs

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    b78895b View commit details
    Browse the repository at this point in the history

  60. batch_direct is implemented but not compilable yet

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    7f1b5c1 View commit details
    Browse the repository at this point in the history

  61. - wrote getrs_batch and getrf_batch in onemkl_bindings.cpp. 

    - batch_direct compiled, no test yet.
    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    b00cd94 View commit details
    Browse the repository at this point in the history

  62. tmp saved

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    42fc377 View commit details
    Browse the repository at this point in the history

  63. fixed batch_cg and batch_bicgstab, both kernels should work with larg… 

    …e sizes now
    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    0f17ec5 View commit details
    Browse the repository at this point in the history

  64. Add batch_direct tests for dpcpp

    @pratikvn
    pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    310dcdf View commit details
    Browse the repository at this point in the history

  65. Enable dpcpp batch_richardson tests

    @pratikvn
    pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    df67ec5 View commit details
    Browse the repository at this point in the history

  66. batch_block_jacobi passed test for float and double, but not for cmplx

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    da5e544 View commit details
    Browse the repository at this point in the history

  67. added tests for component functions of batch_block_jacobi

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    5e35df2 View commit details
    Browse the repository at this point in the history

  68. Some fixes for nan output with complex on dpcpp

    @pratikvn
    pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    2b83e07 View commit details
    Browse the repository at this point in the history

  69. fixed batch_direct, passed all tests now

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    57756bb View commit details
    Browse the repository at this point in the history

  70. implemented batch_gmres, passed 6/7 tests on iris

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    afa88d5 View commit details
    Browse the repository at this point in the history

  71. fixed slm allocation issue

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    ee1e042 View commit details
    Browse the repository at this point in the history

  72. implemented batch_idr, passed 5/6 tests

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    225061e View commit details
    Browse the repository at this point in the history

  73. fixed slm allocation in batch_block_jacobi

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    4f4e3da View commit details
    Browse the repository at this point in the history

  74. impl batch_ilu, batch_isai, batch_ilu_isai, compilation does not go t… 

    …hrough yet
    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    400560c View commit details
    Browse the repository at this point in the history

  75. added .inc files

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    902a1fd View commit details
    Browse the repository at this point in the history

  76. batch_ilu passed all tests. batch_isai and batch_ilu_isai only passed… 

    … tests with small matrices size
    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    7072b8a View commit details
    Browse the repository at this point in the history

  77. batch_isai and batch_ilu_isai passed all tests

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    ede82da View commit details
    Browse the repository at this point in the history

  78. Add a run script

    @pratikvn
    pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    1be2895 View commit details
    Browse the repository at this point in the history

  79. improved performance of batch_cg by ~3x for matsize bigger than 1024

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    392db00 View commit details
    Browse the repository at this point in the history

  80. Remove branching to improve perf for small mats

    @pratikvn
    pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    2584e2e View commit details
    Browse the repository at this point in the history

  81. changes in example: nreps + time in msecs

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    f694ec4 View commit details
    Browse the repository at this point in the history

  82. implemented dynamic group_size + clean up

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    2625440 View commit details
    Browse the repository at this point in the history

  83. implemented branching for small and big matsize with different group … 

    …and subgroup size
    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    875d45d View commit details
    Browse the repository at this point in the history

  84. fixed sync

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    1cec8c9 View commit details
    Browse the repository at this point in the history

  85. implemented kernel for each mat_size range

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    a913280 View commit details
    Browse the repository at this point in the history

  86. minor changes in order of calculations

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    fc65a53 View commit details
    Browse the repository at this point in the history

  87. optimization for batchcg medium kernel

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    5be03f0 View commit details
    Browse the repository at this point in the history

  88. resolved conflicts

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    3d79d9e View commit details
    Browse the repository at this point in the history

  89. allocated alpha once outside of iteration steps

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    06e5fa4 View commit details
    Browse the repository at this point in the history

  90. templated vecs_shared_all and sg_kernel_only, using lambda to contain… 

    … apply_kernel launching which requires c++20 but sycl bit_mask breaks
    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    3c509b4 View commit details
    Browse the repository at this point in the history

  91. clean batch_vector_kernels

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    e2c36a9 View commit details
    Browse the repository at this point in the history

  92. templated batch_cg for all optimization parameters

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    bfdcf93 View commit details
    Browse the repository at this point in the history

  93. added icpx for sycl compilation into CMakefiles.txt

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    31145ea View commit details
    Browse the repository at this point in the history

  94. renamed template parameters, add group_size cals

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    fd32fd4 View commit details
    Browse the repository at this point in the history

  95. impled optimizations for batch_gmres and batch_cg, passed all tests

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    d1cb464 View commit details
    Browse the repository at this point in the history

  96. cleanup

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    51eec85 View commit details
    Browse the repository at this point in the history

  97. rm a few barriers

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    2f047f3 View commit details
    Browse the repository at this point in the history

  98. fixed batch_gmres for checking rot and hess storages

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    3d590b0 View commit details
    Browse the repository at this point in the history

  99. resolved conflicts

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    b00d7c9 View commit details
    Browse the repository at this point in the history

  100. - Splitting rot storage into individual vectors. Now n_global is 9 (5 

      main vecs and 4 rotation vecs).
- Changing the order of those vecs and H and V allocations.
- Template all n_shared and prec_storage parameters.
    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    e0d9281 View commit details
    Browse the repository at this point in the history

  101. adjusted init workgroup size + size range for selecting the kernels

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    f5dd9ca View commit details
    Browse the repository at this point in the history

  102. fixed slm overuse issue

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    3d0c0c9 View commit details
    Browse the repository at this point in the history

  103. - Templated batch_bicgstab and batch_cg. 

    - Changed the shared_vecs allocation order.
    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    9012f49 View commit details
    Browse the repository at this point in the history

  104. added batch_ell to example

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    ee94478 View commit details
    Browse the repository at this point in the history

  105. templated batch_gmres cuda

    @pratikvn
    Phuong Nguyen authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    1ed08df View commit details
    Browse the repository at this point in the history

  106. added c++17 std

    @pratikvn
    Phuong Nguyen authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    70559bb View commit details
    Browse the repository at this point in the history

  107. added cuda templating

    @pratikvn
    Phuong Nguyen authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    778a1e4 View commit details
    Browse the repository at this point in the history

  108. rm timer in CMakeLists

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    0d86327 View commit details
    Browse the repository at this point in the history

  109. fixed CMakeLists so that it includes batched kernels

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    92b0866 View commit details
    Browse the repository at this point in the history

  110. undo CMakeLists

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    f56bcac View commit details
    Browse the repository at this point in the history

  111. fixed rot vectors' and hess vector' sizes

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    c4b027b View commit details
    Browse the repository at this point in the history

  112. make sure num_rot_vecs_shared does not have negative value

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    bb67f02 View commit details
    Browse the repository at this point in the history

  113. rm prints in cuda batch gmress

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    7e3c151 View commit details
    Browse the repository at this point in the history

  114. fixed typos

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    bddf39b View commit details
    Browse the repository at this point in the history

  115. added case 0 for batch_bicgstab template

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    2aa47b8 View commit details
    Browse the repository at this point in the history

  116. added case 0 for batch cg

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    d7cbd0c View commit details
    Browse the repository at this point in the history

  117. rm n_shared templating for batch_gmres

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    6a00f2e View commit details
    Browse the repository at this point in the history

  118. rm time print

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    6e03daf View commit details
    Browse the repository at this point in the history

  119. temp: commented out templating in batch gmres

    @pratikvn
    Phuong Nguyen authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    ca819d3 View commit details
    Browse the repository at this point in the history

  120. added large test for batch cg

    @pratikvn
    Phuong Nguyen authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    d29044e View commit details
    Browse the repository at this point in the history

  121. cuda: added n_shared template for batch_cg and batch_bicgstab

    @pratikvn
    Phuong Nguyen authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    cdcd35c View commit details
    Browse the repository at this point in the history

  122. fixed raw_free mem issue

    @pratikvn
    Phuong Nguyen authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    f71c98e View commit details
    Browse the repository at this point in the history

  123. cuda: cleaning

    @pratikvn
    Phuong Nguyen authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    0affb43 View commit details
    Browse the repository at this point in the history

  124. add pele/sundials matrices

    @pratikvn
    pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    0abd4e1 View commit details
    Browse the repository at this point in the history

  125. enabled tests

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    362445b View commit details
    Browse the repository at this point in the history

  126. changed nreps batched-solver.cpp

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    f150acd View commit details
    Browse the repository at this point in the history

  127. added check prec_storage != 0

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    26ab22a View commit details
    Browse the repository at this point in the history

  128. changes batched-solver-from-files to print only time in msec

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    103143e View commit details
    Browse the repository at this point in the history

  129. optimized work-group selection

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    1602c9d View commit details
    Browse the repository at this point in the history

  130. added test scripts

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    6d71332 View commit details
    Browse the repository at this point in the history

  131. renamed get_group_size func, and rewrite 1 0 into true false

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    292bc07 View commit details
    Browse the repository at this point in the history

  132. formatted

    @phu0ngng @pratikvn
    phu0ngng authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    94623a6 View commit details
    Browse the repository at this point in the history

  133. added gmres with n_shared templating

    @pratikvn
    Phuong Nguyen authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    b39b740 View commit details
    Browse the repository at this point in the history

  134. changed alignment to 2

    @pratikvn
    Phuong Nguyen authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    487d7b1 View commit details
    Browse the repository at this point in the history

  135. added a large test case for batch_gmres

    @pratikvn
    Phuong Nguyen authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    1f81ab6 View commit details
    Browse the repository at this point in the history

  136. Format files 

    Co-authored-by: Phuong Nguyen <phu0ngng@users.noreply.github.com>
    @ginkgo-bot @phu0ngng @pratikvn
    2 people authored and pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    a5654c9 View commit details
    Browse the repository at this point in the history

  137. rebase fixes

    @pratikvn
    pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    8e01e7f View commit details
    Browse the repository at this point in the history

  138. remove sundials mats

    @pratikvn
    pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    2396572 View commit details
    Browse the repository at this point in the history

  139. prefix_sum rename

    @pratikvn
    pratikvn committed Oct 15, 2023
    Configuration menu
    Copy the full SHA
    ebee316 View commit details
    Browse the repository at this point in the history

Commits on Oct 16, 2023

  1. stream and dpcpp fixes

    @pratikvn
    pratikvn committed Oct 16, 2023
    Configuration menu
    Copy the full SHA
    e3f2bc7 View commit details
    Browse the repository at this point in the history