GitLab

Avoids double global loads.  Strided loops are unrolled which lets store results in array which compiler will use registers for since the index access is constant.   Updated to handle large reductions so which results with a better stable diffusion result

* deprecate HCC

* Add driver flag "--exhaustive-tune" to enable tuning, add support for the same in C/C++ and python API

Add dynamic shape support to slice operator.

First draft of this feature doesn't support ops slicing non-fixed, dynamic axes. Resulting shape in such cases is not guaranteed.* Also, onnx parsing doesn't support any arguments other than "axes".

Currently, we default to device 0 when loading programs. Updating this to use hipGetDevice to set the device for the loaded program.

Expands on the documentation and corrects default option documentation error.

* Add serialization of tuples and optional types

Using add_instruction for the neg op was causing issues on replace_instruction. Changed to use insert_instruction.  Tests and added a new one that is failing without the change.

* add dynamic shape support to concat operator.  Includes new op_shape_test and ref_ops_test cases

dyn shape support for Where operator.  Includes shape test, ref_ops test, onx_test. 

* Fuse layernorm with different patterns
* Only match when using the last axis
Co-authored-by: kahmed10 <15948690+kahmed10@users.noreply.github.com>
Co-authored-by: kahmed10 <15948690+kahmed10@users.noreply.github.com>

Refactors memory coloring to only handle allocation instructions. It also handles allocations for tuple shapes.

* Implement dynamic shapes for scatterND operators.

Dynamic shape support for gathernd op.

Allows migraphx to inline the IF operator when we run into an IF that can be evaluated at compile time, thus avoiding us injecting IF and just inserting the instructions directly.

upgrade to ROCm 5.4.2 in CI

Added CMakeFlag for hipRTC. MIGRAPHX_USE_HIPRTC.
Added stages in Jenkins for hipRTC.
Fixes for some of the pending issues from hipRTC.

* Add general optimize pass
* Fuse gemm multiplies by scalar
* Handle zero epsilon

Dynamic shape support for gather op.

* Create the shape correctly in python printout

supports upper and lower trilu
not able to support dynamic k values
not able to support negative k values

Adds support for parsing dynamic ONNX gemm bias input C

This prevents multiple adds.

Extends ONNX Gemm parser to handle dynamic input shapes
Limits ONNX Gemm parsing to 2D input tensors for A and B inputs
As per the ONNX specifications
Changed Gemm ONNX tests to 2D input versions
Add onnx_verify tests for Gemm
Parsing ONNX Gemm links to more than one operator, checking that it produces the correct result

Extends reshape to handle the case of a single non-fixed dynamic_dimension

Extends pad operator to handle dynamic input shapes
Only handles computing the shape for adding constant padding to a dynamic shape
- adds the padding to the min, max, and opt values (unless opt is 0, where it keeps it 0)
- does not handle reflect padding with dynamic shapes

This PR resolves the bug addressed in #1496. 

Extends parse_matmul.hpp to handle dynamic input shapes
Does not support broadcasting of the outer dimensions for dynamic shapes at this time

* Use cosine to compute half sin

Fixes ONNX parsing of convolution to handle dynamic broadcasting of bias input

JIT implementation of the gather operator
Added a few more unit tests to this one as well since I saw some odd behavior during bring up.

Implements dynamic shapes in reduce_op and all its child operator classes (reduce_max etc.)

* Print python code

Implements the operator==(dynamic_dimension, size_t) functions

HIP had change in previous rocm releases to use --offload-arch instead of --cuda-gpu-arch.

This should be backwards compatbile. hipRTC also supports --offload-arch.

Extends dot MIGX operator to handle dynamic input shapes
Only allow dot between two dynamic shapes that have exactly matching outer dimensions
Inner dimensions must also match correspondingly
Updates dot related tests
Change check_shapes to use shape.ndim()
ONNX parsers for GEMM and MatMult will be updated in a separate PR