GitLab

Added an Accuracy checker to the tools directory.  Currently compares ONNX FP32 models against ORT CPUEP

Currently, while copying a host buffer to the device, it first registers/maps the host buffer pointer to address space of the device.

If the host buffer has been allocated by the hipHostMalloc then, it is implicitly registered to the device's address space, and no need to register again. This PR adds a check for the same.

Gives better clarity to which argument is throwing an error, especially in cases with nested IF statements in the network.

instancenorm parser always creates literal of type float which would fail in type check while creating binary ops if model is fp16.

* Add in changes for onnx Mod operator

Initial operator for mod implementation and test cases for integer and floating based types.

Need to use fmod from stdlib for floating point types. half_float::half thankfully is specced to the use the existing std::fmod() call when looking at the half.hpp implementation.

fmod_flag should mirror the onnx fmod attribute. Right now using a floating point type without setting that on the user side to true will result in an exception.

Ref ticket #1283 

* Add is_supported to the target
* Add get_target_assignments
* Rename assignment to target_assignments
* Add ref target header to test
* Add fpga target
* Make context const in compute

* Add performance check per commit

C++ API is not printing thrown exception string. this improves on it.

Remove specific person name from deb created packages and move toward a general maintainer id/email

Dynamic shape handling in shape object

Bug 1: create_literal was using back_inserter to copy vector with already allocated size, causing double the size of literal.
Fix 1 : not use back_inserter
Bug 2: Input param to model can be from operation that has multiple output, in that case name of the input param would contain : e.g. input_1:0
Fix 2: Look for : and take substring

Depends on #1199

Adds ONNX parser functionality for dynamic input shapes.
Uses options parameter in parse_onnx()

Changes to operator includes:

removed some includes that were not used
included argument.hpp where clang-tidy wanted it

Fix min_test.onnx generation as well as add a proper check to the parse program vs the expect program.
Adding this in to fix test converge for the min case.

Reduce header inclusion in op headers

This will ensure that migraphx.h can be included from a C compiler, and check that the C API can be called. This includes stdbool.h which is needed when using bool from C.

Enable the migraphx-dev package when using make|rbuild package

bitbucket needs a port that some servers do not make available. Move the Blaze dependency from a bitbucket to a github source repo.

* Only run __syncthreads when there is data to preload
* Improve loops
* Add const attribute to improve optimizations

Show the number of operators and per operator avg time in summary...

Summary:
gpu::gemm: 8.738ms / 73 = 0.119699ms, 64%
gpu::triadd_layernorm: 0.831381ms / 24 = 0.0346409ms, 7%

Improve the assembly dump to track where certain instruction come from.

Added is_supported and get_target_assignments methods to the target and program, respectively, to eventually support multi-target compilation and execution.

Initial sketch for changes to shape to handle dynamic dimensions

Instead of just unsqueezing to an axis of 1 a step can be set to use instead. So instead of unsqueezing {3, 12} to {3, 1, 12} a step of 2 will unsqeeze to {3, 2, 6} instead

*In the verification tests, check that saving and reloading the program is the same program. This also fixes serialization to always load instructions in the same order. There is also fixes for deconv and quant_conv which didn't save the solution id, and was broken for serialization.

* Add softmax kernel

This reorders the transposes across slice to improve horizontal fusion for contiguous. This also improves eliminate_contiguous to remove contiguous better across splits.

* Add mlir c api

* Formatting

* Create a type attribute

* Formatting

* Parse module

* Formatting

* Add mlir dump function

* Add test case

* Formatting

* Fix tidy issues

* Update mlit version

* Update to newer mlir

* Format

* Move mlir to the gpu and update the test

* Formatting

* Fix bug when appending module

* Format

* Remove old cmake flag

* Update message

* Add return

* Format

* Add mlir_compile

* Format

* Register dialect

* Handle unsinged integers

* Dont provide output for return instruction

* Format

* Add code to insert memrefs

* Format

* Add mlir verification

* Formatting

* Enable pointwise_fusion

* Disable eliminate_data_type

* Set kernal name

* Format

* Fix device name

* Formatting

* Fix output arg

* Format

* Updates

* Upate hash

* Add fuse_mlir pass

* Format

* Add fuse mlir

* Format

* Update mlir

* Sort parameter names

* Format

* Reenable disabled passes

* Remove old mlir conv

* Remove asym default padding

* Add more verbose tracing

* Format

* Fix compilation errors

* Format

* Whitelist operators

* Format

* Add namespace

* Format

* Update triple

* Format

* Use func dialect

* Format

* Use func.return

* Format

* Upgrade mlir version

* Add comment

* Handle symetrical padding

* Format

* Cleanup debug output

* Format

* List failed tests

* Move mlir compile to jit pipeline

* Format

* Update version

* Add source locations

* Format

* Correctly add module

* Format

* Update failed tests

* Fix failures when mlir is disabled

* Format

* Update mlir version

* Check type for fp32

* Format

* Remove failed test

* Update mlir in driver

* Tidy fixes

* Foramt

* Tidy fixes

* Format

* Fix const

* Remove from requirements

* Fix cmake version

* Fix tidy warning

* Use another ifdef

* Fix tidy

* Other tidy fix

* Format

* Update hash

* Add missing license files

* Format

* Format

* Fix fnction name

This is an extension to insert_module_instructions, but instead of just inserting from a module, it can insert a range or a vector of instructions.

should be --fp16 , not --fp16ref

Allows PyTorch converted version of SSD-resnet34 to work

 Compiles significantly faster than constructing all the objects. It also reduces recompiles as well.

This PR only adds an example using MIOpen Calls.

Add an example using rocBLAS Calls

This PR only adds an example using HIP kernel.

* Add function to get a module tree
* Get parent module in the pass manager

One-line fix to register the op miopen_fusion. This error was causing loading of compiled model files (*.mxr) to fail.