Commit 8e2c4f8
authored
Update Paddle submodule to latest develop (PaddlePaddle#2266)
Co-authored-by: tianshuo78520a <[email protected]>1 parent 2f6b467 commit 8e2c4f8
1 file changed
+1
-1
lines changed- paddle/cinn/hlir/framework/pir/fusion_info.cc+8-4
- paddle/fluid/distributed/collective/deep_ep_xpu/deep_ep.cpp+6-4
- paddle/phi/core/allocator.h+4-1
- paddle/phi/core/memory/allocation/allocator.cc+20-4
- paddle/phi/core/memory/allocation/allocator.h+30-11
- paddle/phi/core/memory/mem_utils.cc+2-2
- paddle/phi/core/memory/mem_utils.h+2-1
- paddle/phi/core/memory/mem_visitor.h+2-2
- paddle/phi/core/mixed_vector.cc+31-2
- paddle/phi/core/mixed_vector.h+17-4
- paddle/phi/kernels/legacy/gpu/batched_gemm.h+28
- paddle/phi/kernels/xpu/pool_grad_kernel.cc+17-19
- paddle/phi/kernels/xpu/pool_kernel.cc+15-21
- python/paddle/base/dygraph/math_op_patch.py+1
- python/paddle/base/dygraph/tensor_patch_methods.py+3
- python/paddle/base/layers/math_op_patch.py+1
- python/paddle/device/cuda/__init__.py+19-2
- python/paddle/device/cuda/memory_analyzer.py+36-19
- python/paddle/pir/math_op_patch.py+1
- python/unittest_py/requirements.txt+1-1
- test/dygraph_to_static/test_tensor_attr_consistency.py+1
- test/xpu/test_collective_deep_ep_alltoall_intranode.py+506
- test/xpu/test_low_latency_all2all.py+302
- test/xpu/test_low_latency_utils.py+121
0 commit comments