• Chris Sullivan's avatar
    Preallocate intermediate buffers (#1231) · 0599a628
    Chris Sullivan authored
    * Utilize GPUMemoryManager/Allocator for preallocation of intermediate tensor buffer memory.
    
    * Formatting.
    
    * Merge with master required rework of memory due to CFE pass. Moved function memory pool allocation to pass as a result.
    
    * Formatting.
    
    * Added pass source files.
    
    * Updated tests to account for new assert check. All GPUAllocators should be deconstructed before allocation is made in GPUMemoryManager.
    
    * GPUAllocator::close() can be used to close the allocator prior to destruction
    
    * Removed open allocators. Replaced check with inspection of pass::MemoryManager node list.
    
    * Formatting.
    
    * Rename m_memory_buffers -> m_tensor_memory_buffers. Use full path to static alignment variable.
    
    * FunctionMemoryReservation -> TensorMemoryReservation. Only return true in pass if reservation is made (bug fix).
    
    * Moved static compilation mutex.
    
    * Update external function with new pass name.
    
    * GPU_ExternalFunction: Add s_memory_pool_alignment, remove optimize_and_assemble method.
    0599a628
Name
Last commit
Last update
..
files Loading commit data...
models Loading commit data...
ref_generators Loading commit data...
util Loading commit data...
CMakeLists.txt Loading commit data...
algebraic_simplification.cpp Loading commit data...
all_close_f.cpp Loading commit data...
assertion.cpp Loading commit data...
autodiff.in.cpp Loading commit data...
backend_api.cpp Loading commit data...
backend_debug_api.cpp Loading commit data...
backend_performance.cpp Loading commit data...
backend_test.in.cpp Loading commit data...
build_graph.cpp Loading commit data...
builder.cpp Loading commit data...
builder_autobroadcast.cpp Loading commit data...
codegen.cpp Loading commit data...
constant_folding.cpp Loading commit data...
convolution_test.in.cpp Loading commit data...
copy.cpp Loading commit data...
core_fusion.cpp Loading commit data...
cpio.cpp Loading commit data...
cpu_fusion.cpp Loading commit data...
cpu_test.cpp Loading commit data...
cse.cpp Loading commit data...
cudnn.cpp Loading commit data...
distributed.cpp Loading commit data...
element_type.cpp Loading commit data...
file_util.cpp Loading commit data...
gpu_test.cpp Loading commit data...
graph_partition.cpp Loading commit data...
includes.cpp Loading commit data...
inliner.cpp Loading commit data...
input_output_assign.cpp Loading commit data...
main.cpp Loading commit data...
mkldnn.cpp Loading commit data...
nop_elimination.cpp Loading commit data...
op.cpp Loading commit data...
pass_liveness.cpp Loading commit data...
pass_manager.cpp Loading commit data...
pass_memory_layout.cpp Loading commit data...
pattern.cpp Loading commit data...
reshape_elimination.cpp Loading commit data...
serialize.cpp Loading commit data...
shape.cpp Loading commit data...
tensor.cpp Loading commit data...
type_prop.cpp Loading commit data...
update_reference.sh Loading commit data...
util.cpp Loading commit data...
uuid.cpp Loading commit data...
zero_dim_tensor_elimination.cpp Loading commit data...