GitLab

* Improve applicable batched gemms for bert

As described in #1196, the ONNX mean parser does not work correctly for integral types. This update fixes the issue by handling integral types separately, where summation is performed before division. Additional test cases have also been added for handling integral types.

Fuse layernorm and added triadd_layernorm fusion.  This is a prep performance booster

Reverts the dyn_data struct change
Should get around the ambiguous braced initialization list error

Expose add_literal method in C/C++ api

Add compile tests for gpu math functions

Helps avoid dangling references. This also deprecates the constructors that didnt take a lifetime annotation since its ambiguous the lifetime.

Add ref and gpu implementations for ONNX op GatherND

Resolves #1032

* expose get_queue method

Implements the ReverseSequence ONNX operator as a parser.

This parser can only handle a constant sequence_lens input. This is the same as what is handled for TensorRT as far as I can tell.
We could handle a variable sequence_lens input; that would require ref and GPU implementations of the operator.
The ONNX backend tests are disabled because this does not handle variable sequence_lens.

Refactored the reference implementation of pooling to something like what was done for roialign. Moved the reference implementation of pooling from targets/ref/lowering.cpp to pooling.hpp.
Removed cpu_pooling, instead using reference pooling in pooling.hpp
Added reference implementation of Lp Norm pooling and the global version
Added tests for the Lp Norm Pooling

There is significant improvement on larger tensors with half almost 50% faster:

lens: [1024, 384, 768]
gpu::code_object[code_object=13832,symbol_name=kernel,global=39321600,local=256,]: 1.16685ms
gpu::reduce_sum[axes={2}]: 1.73126ms
Also for non-trivial layouts this can sometimes be over 2x faster:

lens: [64, 1024, 768, 4]
gpu::code_object[code_object=13832,symbol_name=kernel,global=39321600,local=256,]: 1.1706ms
gpu::reduce_sum[axes={1}]: 2.63375ms
Of course if the stride becomes larger this speed improvement diminishes due to poor memory access patterns. A lane_reduce instead of a block_reduce is needed for such type of kernels. I plan to address that in a future PR.

Finally, this also includes a MIGRAPHX_GPU_DUMP_ASM env variable which will print out the assembly when the kernel compiles.

Issue 1127 Updates the math.hpp header file to perform overloads of various standard functions (ops) for the hip half2 type. The half2 type is two 16-bit floats packed into a 32-bit number and therefore the overloads act on vectors of sizes that are multiples of 2. They are invoked in runtime compilation any time one of the ops is called on a tensor declared with the data type shape::half_type.

Defined new template, made instances of the template for those math operations that the hip library contains, added verify tests for the sqrt operator for three cases:

tensor size not divisible by 2
tensor size divisible by 2 but not by 4
tensor size divisible by 4

Change the "scatter" struct and op to a base/child set of three: scatter_none, scatter_add, scatter_mul to mirror Onnx' ScatterElements op. and its three reduction options. (Onnx Scatter op is deprecated and is equivalent to scatter_none.)

Provides both a reference op. and update to Onnx parsing. Tests updated and new test case added.

* Fix comparisons in migraphx::value class

Adds following API binding and tests to python :

add_return
add_instruction
add_parameter
create_module.

* Fix and change doc CMakeLists
1. Fix include directory location with hange from #1088
2. Create a DoxygenWarningLog.txt file in <build_dir>/doc/doxygen
3. Move compiled html or pdf files to <build_dir>/doc/[pdf, html]

Documentation update for valid targets

Follow up to #1128

This adds the infrastructure so we can compile everything in parallel, whereas before only pointwise kernels were compiled in parallel. This will also directly integrate with lowering and the gpu-driver. The kernels for pointwise and roialign are using this infrastructure. Scatternd is not since it does require standard shape.

This also makes it easier to add new runtime compiled kernels in the future.

* Handle string literal in construction
* Improve get_default with vector

This creates a custom op which has name() and compute_shape() methods. 

* LpNormalization ONNX parser

Add exclusive and reverse modes to gpu implementation of prefix_scan_sum, which completes support for ONNX op CumSum

The get_context may change in the future(when we support multi-targets) so make this experimental for now.