GitLab

* Use an internal cuda stream for CopyDataFromTo

* small fix white space

* Fix to compile

* Make stream optional in copydata for compile

* fix lint issue

* Update cub functions to use internal stream

* Lint check

* Update CopyTo/CopyFrom/CopyFromTo to use internal stream

* Address comments

* Fix backward CUDA stream

* Avoid overloading CopyFromTo()

* Minor comment update

* Overload copydatafromto in cuda device api
Co-authored-by: xiny <xiny@nvidia.com>

* enable turn on/off libxsmm at runtime by adding a global config and related API
Co-authored-by: Ubuntu <ubuntu@ip-172-31-19-194.ap-northeast-1.compute.internal>

* Change CUDA_MAX_NUM_THREADS to 256

* change the configuration of grid

* enable use for weighted neighbor sampler and biased random walk

* add unit tests

* fix for mxnet/tf

* fix typo

* add weighted sampling without replacement (A-Chao)

* improve Algorithm A-Chao with block-wise prefix sum

* correctly fill out_idxs

* implement weighted sampling with replacement

* small fix

* merge host-side code of weighted/uniform sampling

* enable unit tests for cuda weighted sampling

* move thrust/cub wrapper to the cmake file

* update docs accordingly

* fix linting

* fix linting

* fix unit test

* Bump external CUB/Thrust versions

* Fix code style and update description of algorithm design

* [Feature] GPU support weighted graph neighbor sampling
commit by pengqirong(OPPO)

* merge pengqirong's implementation

* revert the change to cub and thrust

* fix linting

* use DeviceSegmentedSort for better performance

* add more comments

* add necessary notes

* add necessary notes

* resolve some comments

* define THRUST_CUB_WRAPPED_NAMESPACE

* fix doc
Co-authored-by: 彭齐荣 <657017034@qq.com>

* [Feature] extend sort_csr/csc_by_tag to edge

* fix test ffailure in tensorflow

* refine sorting by edges

* fix docstring

* remove unnecessary mem
Co-authored-by: Xin Yao <xiny@nvidia.com>

* * Added missing specializations for `__half` of `DLDataTypeTraits`, `IndexSelect`, `Full`, `Scatter_`, `CSRGetData`, `CSRMM`, `CSRSum`, `IndexSelectCPUFromGPU`
* Fixed casting issue in `_LinearSearchKernel` that was preventing it from supporting `__half`
* Added `#if`'d out specializations of `CSRGEMM`, `CSRGEAM`, and `Xgeam`, which would require functions that aren't currently provided by cublas

* * Added more specific error messages for unimplemented FP16 specializations of Xgeam, CSRGEMM, and CSRGEAM

* * Added missing instantiation of DLDataTypeTraits<__half>::dtype

* * Fixed linter error
* Added clearer comment explaining why the cast to long long is necessary

* * Worked around a compile error in some particular setup, where __half can't be constructed on the host side

* * Fixed linter formatting errors

* * Changes to comments as recommended

* * Made recommended changes to logging errors in FP16 specializations
* Also changed the existing Xgeam function for unsupported data types from LOG(INFO) to LOG(FATAL)

* Add uva by default to embedding

* More updates

* Update optimizer

* Add new uva functions

* Expose new pinned memory function

* Add unit tests

* Update formatting

* Fix unit test

* Handle auto UVA case when training is on CPU

* Allow per-embedding decisions for whether to use UVA

* Address spares_optim.py comments

* Remove unused templates

* Update unit test

* Use dgl allocate memory for pinning

* allow automatically unpin

* workaround for d2h copy with a different dtype

* fix linting

* update error message

* update copyright
Co-authored-by: Xin Yao <xiny@nvidia.com>
Co-authored-by: Minjie Wang <wmjlyjemaine@gmail.com>

* Fix a cub compile error for CUDA 11.5

* Fix comparison of integer expressions of different signedness in coo_sort.cu file

* Fix comparison of integer expressions of different signedness in cuda_compact_graph.cu file

* Remove never referenced variable in spmm.cu

* Fix comparison of integer expressions of different signedness in rowwise_pick.h file

* Fix comparison of integer expressions of different signedness in choice.cc file

* Remove never referenced variable col_data in spat_op_impl_coo.cc

* Remove never referenced variable allowed in global_uniform.cc

* Fix comparison of integer expressions of different signedness in graph.cc

* Fix comparison of integer expressions of different signedness in graph_apis.cc

* Fix the un-used ctx variable in ndarray_partition.cc file for cpu only build

* Fix comparison of integer expressions of different signedness in libra_partition.cc

* Fix comparison of integer expressions of different signedness in graph_op.cc
Co-authored-by: Triston Cao <tristonc@nvidia.com>
Co-authored-by: Quan (Andy) Gan <coin2028@hotmail.com>

* Disable non-atomic atomic operations

* Improve error message

* Make error message more friendly

* Wrap all CUDA runtime API/CUB calls with macro

* remove the usage of explicit cudaMalloc in favor of AllocWorkspace

* fix typo
Co-authored-by: Israt Nisa <neesha295@gmail.com>

* * Added specialization of cublasGemm function for `__half` type, to try to address https://github.com/dmlc/dgl/issues/3988



* * Added USE_FP16 guard

* * Added test cases to test_segment_mm, to test newly-added FP16 specialization of cublasGemm

* * Replaced for loop in test_segment_mm with pytest.mark.parametrize, as recommended
Co-authored-by: Xin Yao <xiny@nvidia.com>

* * Added support for common operations on FP16 (`half` or `__half`) for older GPU architectures
* Fixed an issue with previous check for FP16 support

* * Removing FP16 type checks, since they should no longer be needed

* * Fixed AtomicAdd to be atomic for `float` and `double` for old GPU architectures.  Unfortunately, it seems that atomicCAS for unsigned short seems to be unavailable until architecture 70, so half will have to stay non-atomic on old GPUs.

* * Fixed non-atomic version of `AtomicAdd<half>` for older GPUs to return old value instead value of new

Co-authored-by: Xin Yao <xiny@nvidia.com>

Co-authored-by: nv-dlasalle <63612878+nv-dlasalle@users.noreply.github.com>
Co-authored-by: Israt Nisa <neesha295@gmail.com>

* Enable FP16 for GPU builds in CI

* Limit default GPU archs to pascal and above

* Disable FP16 dispatching for cuda architectures less than 60

* Fix linting

* Fix typos

* Change the curand_init parameter

* Change the curand_init parameter

* commit

* commit

* change the curandState and launch dim of CSRRowwiseSample kernel

* commit

* keep  _CSRRowWiseSampleReplaceKernel in sync
Co-authored-by: nv-dlasalle <63612878+nv-dlasalle@users.noreply.github.com>

* remove unnecessary induced vertices in EdgeSubgraph

* add unit test

* Based on issue #3436. Improving _SegmentCopyKernel s GPU utilization by switching to nonzero based thread assignment

* fixing lint issues

* Update cub for cuda 11.5 compatibility (#3468)

* fixing type mismatch

* tx guaranteed to be smaller than nnz. Hence removing last check

* minor: updating comment

* adding three unit tests for csr slice method to cover some corner cases

* timing repeatkernel

* clean

* clean

* clean

* updating _SegmentMaskColKernel

* Working on requests: removing sorted array check and adding comments to utility functions

* fixing lint issue

* Optimizing disjoint union kernel

* Trying to resolve compilation issue on CI

* [EMPTY] Relevant commit message here

* applying revision requests on cpu/disjoint_union.cc

* removing unnecessary casts

* remove extra space
Co-authored-by: Abdurrahman Yasar <ayasar@nvidia.com>
Co-authored-by: nv-dlasalle <63612878+nv-dlasalle@users.noreply.github.com>
Co-authored-by: Jinjing Zhou <VoVAllen@users.noreply.github.com>
Co-authored-by: Quan (Andy) Gan <coin2028@hotmail.com>

* Change the curand_init parameter

* Change the curand_init parameter

* commit

* commit
Co-authored-by: nv-dlasalle <63612878+nv-dlasalle@users.noreply.github.com>

* split files

* fix

* WIP: TypedLinear and new RelGraphConv

* wip

* further simplify RGCN

* a bunch of tweak for performance; add basic cpu support

* update on segmm

* wip: segment.cu

* new backward kernel works

* fix a bunch of bugs in kernel; leave idx_a for future

* add nn test for typed_linear

* rgcn nn test

* bugfix in corner case; update RGCN README

* doc

* fix cpp lint

* fix lint

* fix ut

* wip: hgtconv; presorted flag for rgcn

* hgt code and ut; WIP: some fix on reorder graph

* better typed linear init

* fix ut

* fix lint; add docstring

* fixes

* fix

* more fixes

* update

* oops

* lint?

* temporarily revert - will fix in another PR

* more fixes

* skipping mxnet test

* address comments

* fix DDP

* fix edge dataloader exclusion problems

* stupid bug

* fix

* use_uvm option

* fix

* fixes

* fixes

* fixes

* fixes

* add evaluation for cluster gcn and ddp

* stupid bug again

* fixes

* move sanity checks to only support DGLGraphs

* pytorch lightning compatibility fixes

* remove

* poke

* more fixes

* fix

* fix

* disable test

* docstrings

* why is it getting a memory leak?

* fix

* update

* updates and temporarily disable forkingpickler

* update

* fix?

* fix?

* oops

* oops

* fix

* lint

* huh

* uh

* update

* fix

* made it memory efficient

* refine exclude interface

* fix tutorial

* fix tutorial

* fix graph duplication in CPU dataloader workers

* lint

* lint

* Revert "lint"

This reverts commit 805484dd553695111b5fb37f2125214a6b7276e9.

* Revert "lint"

This reverts commit 0bce411b2b415c2ab770343949404498436dc8b2.

* Revert "fix graph duplication in CPU dataloader workers"

This reverts commit 9e3a8cf34c175d3093c773f6bb023b155f2bd27f.
Co-authored-by: xiny <xiny@nvidia.com>
Co-authored-by: Jinjing Zhou <VoVAllen@users.noreply.github.com>

* Based on issue #3436. Improving _SegmentCopyKernel s GPU utilization by switching to nonzero based thread assignment

* fixing lint issues

* Update cub for cuda 11.5 compatibility (#3468)

* fixing type mismatch

* tx guaranteed to be smaller than nnz. Hence removing last check

* minor: updating comment

* adding three unit tests for csr slice method to cover some corner cases

* timing repeatkernel

* clean

* clean

* clean

* updating _SegmentMaskColKernel

* Working on requests: removing sorted array check and adding comments to utility functions

* fixing lint issue
Co-authored-by: Abdurrahman Yasar <ayasar@nvidia.com>
Co-authored-by: nv-dlasalle <63612878+nv-dlasalle@users.noreply.github.com>
Co-authored-by: Jinjing Zhou <VoVAllen@users.noreply.github.com>

* init

* init

* working cublasGemm

* benchmark high-mem/low-mem, err gather_mm output

* cuda kernel for bmm like kernel

* removed cpu copy for E_per_Rel

* benchmark code from Minjie

* fixed cublas results in gathermm sorted

* use GPU shared mem in unsorted gather mm

* minor

* Added an optimal version of gather_mm_unsorted

* lint

* init gather_mm_scatter

* cublas transpose added

* fixed h_offset for multiple rel

* backward unittest

* cublas support to transpose W

* adding missed file

* forgot to add header file

* lint

* lint

* cleanup

* lint

* docstring

* lint

* added unittest

* lint

* lint

* unittest

* changed err type

* skip cpu test

* skip CPU code

* move in-len loop inside

* lint

* added check different dim length for B

* w_per_len is optional now

* moved gather_mm to pytorch/backend with backward support

* removed a_/b_trans support

* transpose op inside GEMM call

* removed out alloc from API, changed W 2D to 3D

* Added se_gather_mm, Separate API for sortedE

* Fixed gather_mm (unsorted) user interface

* unsorted gmm backward + separate CAPI for un/sorted A

* typecast to float to support atomicAdd

* lint typecast

* lint

* added gather_mm_scatter

* minor

* const

* design changes

* Added idx_a, idx_b support gmm_scatter

* dgl doc

* lint

* adding gather_mm in ops

* lint

* lint

* minor

* removed benchmark files

* minor

* empty commit
Co-authored-by: Israt Nisa <nisisrat@amazon.com>

* [feature] edge softmax refact.

* delete file

* fix backward and cmake version

* fix backward

* format function

* fix setting

* refix

* refix

* refix

* refix

* refix

* refix

* refix

* refix

* refix

* refix

* refix

* refix

* add cuda kernel for backward and rename some function

* add benchmark for edge_softmax

* fix format

* remove cuda_backwrd

* fix code format and add comment for op on CPU

* fix lint
Co-authored-by: Jinjing Zhou <VoVAllen@users.noreply.github.com>

* implement pin_memory/unpin_memory/is_pinned for dgl.graph

* update python docstring

* update c++ docstring

* add test

* fix the broken UnifiedTensor

* XPU_SWITCH for kDLCPUPinned

* a rough version ready for testing

* eliminate extra context parameter for pin/unpin

* update train_sampling

* fix linting

* fix typo

* multi-gpu uva sampling case

* disable new format materialization for pinned graphs

* update python doc for pin_memory_

* fix unit test

* UVA sampling for link prediction

* dispatch most csr ops

* update graphsage example to combine uva sampling and UnifiedTensor

* update graphsage example to combine uva sampling and UnifiedTensor

* update graphsage example to combine uva sampling and UnifiedTensor

* update doc

* update examples

* change unitgraph and heterograph's PinMemory to in-place

* update examples for multi-gpu uva sampling

* update doc

* fix linting

* fix cpu build

* fix is_pinned for DistGraph

* fix is_pinned for DistGraph

* update graphsage unsupervised example

* update doc for gpu sampling

* update some check for sampling device switching

* fix linting

* adapt for new dataloader

* fix linting

* fix

* fix some name issue

* adjust device check

* add unit test for uva sampling & fix some zero_copy bug

* fix linting

* update num_threads in graphsage examples
Co-authored-by: Quan (Andy) Gan <coin2028@hotmail.com>
Co-authored-by: Jinjing Zhou <VoVAllen@users.noreply.github.com>

* implement pin_memory/unpin_memory/is_pinned for dgl.graph

* update python docstring

* update c++ docstring

* add test

* fix the broken UnifiedTensor

* eliminate extra context parameter for pin/unpin

* fix linting

* fix typo

* disable new format materialization for pinned graphs

* update python doc for pin_memory_

* fix unit test

* update doc

* change unitgraph and heterograph's PinMemory to in-place

* update comments for NDArray's PinMemory_ and PinData

* update doc
Co-authored-by: Jinjing Zhou <VoVAllen@users.noreply.github.com>

* oops

* test

* fix GPU global negative sampling code

* Update negative_sampling.cu

* Pass the std:min argument's type, to avoid the compilation error.

* Update parallel_for.h

* Update negative_sampling.cc
Co-authored-by: Quan (Andy) Gan <coin2028@hotmail.com>

* first commit

* a bunch of fixes

* add unique

* lint

* lint

* lint

* address comments

* Update negative_sampler.py

* fix

* description

* address comments and fix

* fix

* replace unique with replace

* test pylint

* Update negative_sampler.py

[Feature] Add CUDA support for `min` and `max` reducer in heterogeneous API for unary message functions (#3566)

* CUDA support max/min reducer on forward pass

* docstring

* concised UpdateGradMinMax_hetero

* reorganized UpdateGradMinMax_hetero

* CUDA kernels for max/min reducer

* variable name

* lint check

* changed CUDA 2D thread mapping to 1D

* removed legacy cusparse for min/max reducer

* git CI issue

* restarting git CI

* adding namespace std
Co-authored-by: Israt Nisa <nisisrat@amazon.com>
Co-authored-by: Quan (Andy) Gan <coin2028@hotmail.com>

* added distgnn plus libra codebase

* Dist application codes

* added comments in partition code. changed the interface of partitioning call.

* updated readme

* create libra partitioning branch for the PR

* removed disgnn files for first PR

* updated kernel.cc

* added libra_partition.cc and moved libra code from kernel.cc to libra_partition.cc

* fixed lint error; merged libra2dgl.py and main_Libra.py to libra_partition.py; added graphsage/distgnn folder and partition script.

* removed libra2dgl.py

* fixed the lint error and cleaned the code.

* revisions due to PR comments. added distgnn/tools contains partitions routines

* update 2 PR revision I

* fixed errors; also improved the runtime by 10x.

* fixed minor lint error

* fixed some more lints

* PR revision II changed the interface of libra partition function

* rewrite docstring
Co-authored-by: Quan (Andy) Gan <coin2028@hotmail.com>