Commit graph

7 commits

Author SHA1 Message Date
cloudhan
2748f38362
Drop hip_add_library (#13406)
Switching to use CMake's builtin hip language support.
2022-10-25 12:57:48 +08:00
cloudhan
51ac6617f5
Fix warnings and enable dev mode for ROCm CI (#13223)
Fix warnings and enable dev mode for ROCm CI:

* Fix ROCm headers complaining "This file is deprecated. Use the header file from ..."
* Disable warning signed and unsigned compare for kernel explorer
* Fix unused and nondiscard warnings
* Enable dev mode for ROCm CI
* Walkaround error "unknown warning option '-Wno-nonnull-compare'" in kernel explorer by using '-Wno-unknown-warning-option' to ignore the unknown option
* Fix error "unused parameter 'mask'"
* Fix warning "instantiation of variable 'onnxruntime::rocm::Consts<float>::One' required here, but no definition is available", etc. Fixed by using C++17's inline (implied by constexpr) static initialization.
* Remove unused variable
* Add the missing `override` specifier
2022-10-07 09:45:01 +08:00
cloudhan
72076b1eb2
Update ROCm CI to use HIP LANGUAGE (#13214)
Update for ROCm CI before reland tunable GEMM #12853. This PR also update
composable kernel to use CMakes's HIP language support so that we can
mix C/C++ compiler with HIP compiler instead of locking to hip-clang
2022-10-05 16:15:16 +08:00
cloudhan
a24b41d92e
Move all TunableOp related falicilities to EP level directory (#12857)
Some Ops in EP directory instead of contrib_ops directory will
require TunableOp. We will also need to add EP level session tuning
options for it. So move those code all at once.

Also remove duplicated utility functions.
2022-09-23 11:10:19 +08:00
cloudhan
f39354d7cb
Add composable kernel GEMM baseline for kernel explorer (#12364)
* Split GemmBase RocBlasGemm

* Add composable kernel GEMM baseline

* Make linter happy

* Address review comment

* Update bert cases with batchsize

* Adjust includes to fix IWYU lint

* Only builds and links used ck kernels to improve building time

* Remove warmup run on SelectImpl

* Add comment to utility function

* Mute cpplint

* Make RocBlasGemm<T>::SelectImpl semantically correct

* Add reduced basic test cases for ck gemm

* More robust gemm testing

* Fix warnings

* Fix grammar
2022-08-04 17:32:20 -07:00
cloudhan
a0074ba9bc
Add baseline gemm for kernel explorer (#12050)
Use rocblasGemmHelper gemm wrapper from ORT and profile for bert param size only.
2022-07-20 13:49:26 +08:00
cloudhan
785f74979b
Rework cmake for kernel_explorer (#12079)
Improve CMake for deep integration with ORT, so that we can easily hook ort function of microbenchmarking purpose.
2022-07-13 15:43:32 +08:00