GitLab

First try to compile a call to rocblas-beta API; doesn't compile.  Note the update to rocm version in Dockerfile.  May require a new Docker image build.

* 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>

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

This prevents multiple adds.

* Use cosine to compute half sin

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.

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.

* Add implicit_conversion

Need this for when we debug and use MIGRAPHX_TRACE_EVAL() to show tuples
Without this we break when reading our buffer due to the use of visit()
This came up as part of #1283 debugging.

Update dialect registration interface
Update 2nd build pipeline call and use full arch name

Merging #1391 caused an extra adjust allocation pass for GPU targets. This removes that merge error.

Disabling it untill int8 fix is in mainline from MIOpen and also so that QA tests could run migraphx-driver and unittests from MIGraphX.

Can be enabled via environment variable MIGRAPHX_ENABLE_NHWC

Local Threads of multiples 32 were introduced in #1348
But LocalThreads that are not multiple of 64 are causing correctness issues.

Upgraded Dockerfiles and fixed tidy issues to make Ubuntu 20.04 and ROCm 5.3.0 the default 

updated GPU pad to now use JIT version.
added range functions for JIT kernels.

Fixes an observed regression error on certain Frozen Protobuf models due to PR 1280

Reiterate the assertion on the standard shape but relax it for the multibroadcast ops deliberately inserted to explicit the broadcast.

Refactor dynamic compute
- add a compute_output_shape object that implicitly converts to a new dyn_output or shape object
- dyn_output object can handle computing the static output shape of an operator given the input arguments shapes
  change an operator's compute function to argument compute(const dyn_output& dyn_out, std::vector<argument> args) to 
  use dyn_output object

Dynamic ref unary functions
-  Included these changes to have an example of the refactored dynamic compute being used
-  Changes to unary base class to handle dynamic shapes
-  Changed elu and leaky_relu to use unary base class and pointwise JIT

* use find2.0 for the convolution
Co-authored-by: Vasilii Filippov <DrizztDoUrden@users.noreply.github.com>
Co-authored-by: Chris Austen <causten@users.noreply.github.com>

* Enable non-standard shape
* Use perfdb for non xdlops
* Fix transpose+broadcast strides
Co-authored-by: jungpark-mlir <jungwook.park@amd.com>

Rewrites the TF batch norm like operators to other MIGX operators
Removes the code related to batch_norm_inference

Stream sync changes and associated API level changes

optimize the softmax operator

Adds two methods for the custom_ops virtual class.

bool runs_on_offload_target(), if the custom op runs directly on the gpu then it should be set to true. in this case, custom op expects its parameters to reside in GPU memory and writes output to the GPU memory. If it is set to false then, custom op expects it's parameter to reside on the host and puts back the result into the host memory.

output_alias, if output of the custom op is aliasing the input buffer. i.e. interpreting the same input buffer with differnet shape and strides.

Update as_vector() in C++ API to handle non-standard shapes. It required exposing element_index to space_index conversion method for the shape class.

Improvements/Additions to be made:

changes for the quant_convolution,
changes for the deconvolution,
Macros for MIOpen status checks

test_gpu_pack_int8_args fails on gfx908 machine, because it doesn't set compute_fp32 flag correctly. This PR fixes the test such that it checks for the device-name, and rocblas-versions and sets this flag accordingly.

Implement operator for CPU and GPU implementations

* Remove device functions
* Update tests

This PR allows for other values of epsilon to be matched when finding layernorm. Similarly, the calculation now uses the variable for epsilon.

Compute mean and variance in same reduction
Set block size to numbers divisible by 32 instead powers of 2
Global is also set exactly instead of being divisible by block size
More exact matching of global/local can help get rid of branching/loops
Reduce vectors first before doing dpp_reduce
Explicitly vectorize array operators since the compiler doesnt always vectorize them
Still uses old for loop when its computing at compile-time since the reinterpret_cast nor the all the vector types is supported

* fix typo for add_sigmoid