1. 07 11月, 2022 1 次提交
    • H
      [Restore PR] Remove hard code of PADDLE_WITH_CUDA (#47630) · 908a381d
      HongyuJia 提交于
      * move cudnn hardcode outside GetExpectedKernelType
      
      * add header file
      
      * debug
      
      * update interpreter_util with hardcode
      
      * update interpreter_util headerfile
      
      * solve activation hardcode
      
      * debug with CI
      
      * add mkldnn_op_list header file
      
      * temporarily uncomment mkldnn
      
      * temporarily uncomment mkldnn
      
      * delete sequence_softmax cudnn hardcode
      
      * add hardcode to data_transfer.cc
      
      * update data_transfer headerfile
      
      * try fix segment fault
      
      * update cudnn&miopen_helper
      
      * reset HasAttr of DygraphExctnCtx
      
      * debug, this commit should pass all CI
      
      * debug should pass CI, temporarily disable activation
      
      * debug should pass CI
      
      * fix default_attr=nullptr bug
      
      * clean debug code
      
      * Call SetDnnFallback function in the base class
      
      * activation fallback to plain kernel
      
      * fix default GetExpectedKernelType find wrong kernel
      
      * search cudnn kernel instead of fallback
      
      * fix cudnn_handle bug
      
      * remove tanh use_cudnn
      
      * restore tanh use_cudnn
      
      * debug tanh
      
      * fix tanh bug
      
      * delete activation cudnn kernel
      
      * polish code
      908a381d
  2. 02 11月, 2022 1 次提交
  3. 01 11月, 2022 1 次提交
    • H
      [Kernel Selection] Remove hard code of PADDLE_WITH_CUDA (#47325) · f9134045
      HongyuJia 提交于
      * move cudnn hardcode outside GetExpectedKernelType
      
      * add header file
      
      * debug
      
      * update interpreter_util with hardcode
      
      * update interpreter_util headerfile
      
      * solve activation hardcode
      
      * debug with CI
      
      * add mkldnn_op_list header file
      
      * temporarily uncomment mkldnn
      
      * temporarily uncomment mkldnn
      
      * delete sequence_softmax cudnn hardcode
      
      * add hardcode to data_transfer.cc
      
      * update data_transfer headerfile
      
      * try fix segment fault
      
      * update cudnn&miopen_helper
      
      * reset HasAttr of DygraphExctnCtx
      
      * debug, this commit should pass all CI
      
      * debug should pass CI, temporarily disable activation
      
      * debug should pass CI
      
      * fix default_attr=nullptr bug
      
      * clean debug code
      f9134045
  4. 25 10月, 2022 1 次提交
  5. 21 10月, 2022 1 次提交
  6. 19 10月, 2022 1 次提交
  7. 11 10月, 2022 1 次提交
  8. 30 9月, 2022 1 次提交
    • S
      support pure bfloat16 for more ops (#46364) · b7b231a6
      sneaxiy 提交于
      * support pure bfloat16
      
      * support bf16 linear
      
      * update PR to pass CI
      
      * tiny fix where_grad_kernel.cu
      
      * add bfloat16 to selu_grad to pass CI
      
      * fix selu grad compilation error
      b7b231a6
  9. 28 9月, 2022 1 次提交
    • C
      Remove the declaration of using Tensor in framework/tensor.h (#46432) · e12a905e
      Chen Weihang 提交于
      * remove needless using tensor
      
      * remove needless using tensor
      
      * resolve conflict
      
      * replace tensor using
      
      * fix format error
      
      * revert needless changing
      
      * fix rocm and npu compile error
      
      * fix cinn compile error
      
      * fix format error
      
      * fix mkldnn format error
      
      * fix mkldnn format error
      
      * fix cinn compile error
      
      * fix cinn compile error
      
      * fix cinn compile error
      
      * resolve conflict
      e12a905e
  10. 16 9月, 2022 1 次提交
    • S
      Support broadcast elementwise operators with int64 index type (#45741) · 20b5bf84
      sneaxiy 提交于
      * support int64 non-broadcast
      
      * support broadcast case for int64 index
      
      * fix bug
      
      * support more Arity
      
      * remove some codes
      
      * upgrade patchelf to v0.15.0 to pass CI build
      
      * fix bug
      
      * fix patchelf installation
      
      * add debug flags
      
      * remove useless codes
      
      * fix viterbi_decode and set_value op uts
      
      * remove always enable int64
      20b5bf84
  11. 05 9月, 2022 2 次提交
  12. 12 8月, 2022 1 次提交
    • S
      [geometric]Add paddle.geometric.send_ue_recv API (#43174) · 615b15a3
      Siming Dai 提交于
      * add init file
      
      * add op definition and infermeta
      
      * add kernel definition funcs
      
      * add broadcast infer shape
      
      * add gpu forward kernel
      
      * delete SUB and DIV
      
      * add x_grad
      
      * add template
      
      * add e_grad for min and max
      
      * fix small bug
      
      * temp commit
      
      * temp commit
      
      * add e_grad for sum and mean
      
      * fix some compile bug
      
      * fix compile bugs
      
      * fix compile problem
      
      * add sum forward unittest
      
      * fix broadcast error, add kernel sig, register e_grad, change unit test
      
      * fix grad
      
      * add temp grad fix
      
      * temp commit
      
      * add min max unittest
      
      * add max, min unittest, fix mul bug
      
      * add cpu forward sum and mean
      
      * add forward min max, fix mean unittest
      
      * add cpu backward min max
      
      * fix code-style
      
      * add backward sum mean
      
      * fix rocm ci
      
      * set uniitest timeout
      
      * fix bug of x broadcast to e, gpu grad
      
      * fix bug of x broadcast to e, cpu grad
      
      * rename BOOST_GET_CONST macro
      
      * fix rocm ci
      
      * mv graph_send_e_recv to graph_send_ue_recv
      
      * move out_size to IntArray
      
      * add eager op test
      
      * fix max pool type bug, add unittest for api
      
      * revise api doc
      
      * add fp16 for atomic min and max, add unittest
      
      * add unittest
      
      * add fp16 support for graph_send_recv
      
      * fix unittest fp16 bug
      
      * change OutSizeTensor to Out_size
      
      * move E to Y
      
      * add copyright, fix comment
      
      * review code
      
      * fix thread block size
      
      * fix thread block size
      
      * change api attribute name: pool_type to reduce_op, compute_type to message_op
      
      * change api attribute name, move pool_type to reduce_op, move compute_type to message_op
      615b15a3
  13. 01 8月, 2022 1 次提交
    • L
      unify gpu context (#44740) · 86763023
      Leo Chen 提交于
      * remove cudaDeviceContext
      
      * remove more template
      
      * fix rocm compile
      
      * remove alias name CUDADeviceContext
      
      * fix compile
      
      * fix tests
      
      * revert changes
      86763023
  14. 28 7月, 2022 1 次提交
  15. 27 7月, 2022 1 次提交
  16. 19 7月, 2022 1 次提交
  17. 14 7月, 2022 2 次提交
    • L
      refine allocation cmake (#44241) · dc5a0420
      Leo Chen 提交于
      * build into one static library
      
      * move memory/detail to memory/allocation
      
      * fix bug
      
      * fix profiler
      
      * fix framework_proto
      
      * fix deps
      
      * fix inference compilation
      
      * fix rocm compile
      
      * follow comments
      
      * fix buddy_allocator_test
      dc5a0420
    • W
      Compilation optimization (#44242) · 4baf0dbe
      wanghuancoder 提交于
      * Compilation optimization
      4baf0dbe
  18. 26 6月, 2022 1 次提交
  19. 24 6月, 2022 1 次提交
    • C
      record memory and op supplement info (#43550) · 8dd0a3b9
      chenjian 提交于
      * record memory and op supplement info
      
      * update
      
      * update
      
      * fix a bug
      
      * fix memory recording
      
      * fix a bug
      
      * update
      
      * update
      
      * fix a bug
      
      * update
      
      * fix a bug
      
      * fix a bug
      
      * fix a bug
      
      * Revert "fix a bug"
      
      This reverts commit c1d4df52762ba9ae7c7e27cd2ba4fc3a7ed9c7a5.
      
      * fix a bug
      
      * fix format
      
      * fix
      8dd0a3b9
  20. 15 6月, 2022 1 次提交
  21. 05 6月, 2022 1 次提交
  22. 04 6月, 2022 1 次提交
  23. 02 6月, 2022 2 次提交
  24. 01 6月, 2022 1 次提交
  25. 30 5月, 2022 1 次提交
  26. 27 5月, 2022 1 次提交
  27. 10 5月, 2022 2 次提交
  28. 05 5月, 2022 1 次提交
    • R
      Print memory peak message for UT (#42092) · 28375ca4
      Ruibiao Chen 提交于
      * Add peak memory log for CI
      
      * Change VLOG to std::cout
      
      * Move print code to test_runner.py and paddle_gtest_main.cc
      
      * Fix typo
      
      * Fix conflicts
      
      * Updata message format
      
      * Fix CI errors
      
      * Add FLAGS_enable_gpu_memory_usage_log
      
      * Fix CI errors
      28375ca4
  29. 12 4月, 2022 1 次提交
  30. 09 4月, 2022 1 次提交
  31. 03 4月, 2022 1 次提交
    • F
      add maximum limit for grid of index_select (#41127) · af8d2482
      FlyingQianMM 提交于
      * limit grid dim for index select
      
      * mv LimitGridDim into gpu_launch_config.h
      
      * fix conflicts
      
      * fix conflicts
      
      * fix code style
      
      * set block to 256
      
      * fix grid setting
      
      * set dtype of block_dim to unsigned int
      af8d2482
  32. 01 4月, 2022 1 次提交
    • W
      [Eager] Support pinned (#41035) · f3270fc8
      wanghuancoder 提交于
      * support pinned, test=develop
      
      * support async_write, test=develop
      
      * refine, test=develop
      
      * refine, test=develop
      
      * refine, test=develop
      
      * refine,test=develop
      
      * refine, test=develop
      
      * refine, test=develop
      
      * refine, test=develop
      
      * refine, test=develop
      f3270fc8
  33. 30 3月, 2022 1 次提交
    • F
      Add new APIs for GPU memory monitoring (max_memory_allocated,... · afe02e9d
      From00 提交于
      Add new APIs for GPU memory monitoring (max_memory_allocated, max_memory_reserved, memory_allocated, memory_reserved) (#38657)
      
      * Add new API memory_reserved
      
      * Add memory_allocated, max_memory_reserved and max_memory_allocater
      
      * Fix CI error
      
      * Fix CI error
      
      * Enhance UT
      
      * Add FLAGS_memory_stats_opt
      
      * Add STATS macro functions
      
      * Add StatAllocator
      
      * Fix CI errors
      
      * Add UT
      
      * Fix CI errors
      afe02e9d
  34. 25 3月, 2022 1 次提交
  35. 07 3月, 2022 1 次提交
    • M
      cuBlasLt Epilogue To Fuse Linear + ReLU|GeLU (#39437) · 2a3d9eca
      Ming-Xu Huang 提交于
      * Added cuBlasLtHandle_t to device context.
      
      * Added fused_gemm_epilogue op.
      
      1. Added fused_gemm_epilogue op to leverage cuBlastLt Epilogue.
      2. Support fusion Act(X*Y + bias), X'dims >=2 and Y'dims shoule be 2.
      2. Act currently only be supported ReLU. (Will add GeLU in the future).
      
      * Added UT to fused_gemm_epilogue op.
      
      * Added LinearAct Pattern
      
      1. Added LinearAct into graph_pattern_detector.* to define (2.)'s
      pattern.
      2. LinearAct is used to detect act(element_add(matmul_v2(x, w), bias)).
      3. act currently only support ReLU (Will support GeLU in the future).
      
      * Added FuseGemmEpiloguePass
      
      1, Added FuseGemmEpiloguePass to handle nn.Linear + Act{ReLU}
      fusion (GeLU will be supported in the future).
      2. Only support matmul_v2 from nn.Linear.
      
      * Added pybind to BuildStrageter.fuse_gemm_epilogue_.
      
      * Added UT for fuse_gemm_epilogue_pass.
      
      * GeLU support and EpilogueSingleton
      
      1. Added GeLU support to fused_gemm_epilogue op.
      2. Added EpilogueSingleton to cache auxiliary pointer.
      3. Added related UTs.
      
      * Rename cublaslt_epilogue_opto gemm_epilogue_op.*.
      
      * Added both train and infer pattern to LinearAct.
      
      1. Added support of fwd graph with grap_ops linking to LinearAct.
      2. Added related changes to fuse_gemm_epilogue_pass for above
      modification.
      
      * Changed CUDA requirement from 11.4 to 11.6 for fuse_gemm_epilogue_pass.
      
      * Added identity activation support to gemm_epilogue_op.
      
      * Added Linear Fusion (matmul_v2 + ele_add)
      
      1. Added matmul_v2 + ele_add pattern to LinearActPattern.
      2. Added matmul_v2 + ele_add support to fuse_gemm_epilogue_pass.
      
      * Rename gemm_epilogue_op.* to fused_gemm_epilogue_op.*
      
      * Add fused_gemm_epilogue_grad op.
      
      1. Added fused_gemm_epilogue_grad to support backward epilogue fusion.
      
      * Add UTs to fused_gemm_epilogue_grad_op.
      
      * Change attribute name in fused_gemm_epilogue_grad_op for clearing.
      
      * Allow DX and DBias be dispensable to fused_gemm_epilogue_grad op.
      
      * Added ElementwiseAdd+Matmul+Act graph pattern detection.
      
      * Fuse backward of Linear( Act(x))
      
      1. Added backward fusion pass to Linear( Act(x)).
      2. Added backward fusion pass to Linear(x).
      
      * Added UTs to backward fusion of Linear(Act(x)).
      
      * Complete document of arguments to fused_gemm_epilogue_op.
      
      * Made arguments of some functions pass by reference.
      
      * Modify code with review comments.
      
      1. Made arguments of some function pass by reference.
      2. Removed redundant code.
      3. Followed Google code style to change code.
      
      * Made 'const' code style be consistent
      
      * Fixed random seed of python UTs.
      
      * Set Compiling constrains to cuBlasLt
      
      1. Require CUDA 11.6+
      2. Remove fuse_gemm_epilogue related tests when CUDA < 11.6.
      
      * Code Reivew from Paddle
      
      1. Changed arguments name is_first_gemm to without_x_gradient for
      clearing.
      2. Applied PADDLE_THROW in fused_gemm_epilogue_op.
      
      * Remove EpilogueSingleton
      
      1. Applied ReserveSpace to replace Epilogue for passing auxiliary
      pointers between FWD and BWD.
      
      * Fix a logical error and enhance UTs.
      
      1. Added act op count checking in UTs.
      2. Fix issue to fuse backward or ReLU(Linear(X)).
      3. TODO: solve GELU fusion issues.
      
      * Fix Linear and GeLU fusion issues.
      
      1. Modified graph_detech_pattern to fit with both linear wiht gelu or
      relu.
      2. Modified data range in Uts to allow negative values.
      
      * Removed fused_gemm_epilogue_op.h.
      
      * Rename namespace pten to phi.
      
      * Rename name of arguments in fused_gemm_epilogue_op
      
      1. bias -> Bias.
      2. out -> Out.
      3. reserve_space -> ReserveSpace.
      
      * Change EpiloguePassActivationCache as local variable.
      
      1. Removed singleton in EpiloguePassActivationCache.
      2. Made EpiloguePassActivationCache as an argument to each pass
      functions.
      2a3d9eca
  36. 02 3月, 2022 1 次提交