sh -c"echo deb [arch=amd64 signed-by=/etc/apt/trusted.gpg.d/rocm-keyring.gpg] $DEB_ROCM_REPO focal main > /etc/apt/sources.list.d/rocm.list"&&\
sh -c"echo deb [arch=amd64 signed-by=/etc/apt/trusted.gpg.d/rocm-keyring.gpg] $DEB_ROCM_REPO focal main > /etc/apt/sources.list.d/rocm.list"&&\
sh -c'echo deb [arch=amd64 signed-by=/etc/apt/trusted.gpg.d/rocm-keyring.gpg] https://repo.radeon.com/amdgpu/$ROCMVERSION/ubuntu focal main > /etc/apt/sources.list.d/amdgpu.list';\
sh -c'echo deb [arch=amd64 signed-by=/etc/apt/trusted.gpg.d/rocm-keyring.gpg] https://repo.radeon.com/amdgpu/$ROCMVERSION/ubuntu focal main > /etc/apt/sources.list.d/amdgpu.list';\
sh -c'echo deb [arch=amd64 trusted=yes] http://compute-artifactory.amd.com/artifactory/list/rocm-release-archive-20.04-deb/ 6.3.0.1 rel-5 > /etc/apt/sources.list.d/rocm-build.list'&&\
amdgpu-repo --amdgpu-build=2033700;\
fi
fi
RUN sh -c"echo deb http://mirrors.kernel.org/ubuntu focal main universe | tee -a /etc/apt/sources.list"
RUN sh -c"echo deb http://mirrors.kernel.org/ubuntu focal main universe | tee -a /etc/apt/sources.list"&&\
RUNamdgpu-install -y--usecase=rocm --no-dkms
amdgpu-install -y--usecase=rocm --no-dkms
## Sccache binary built from source for ROCm, only install if CK_SCCACHE is defined
## Sccache binary built from source for ROCm, only install if CK_SCCACHE is defined
> The published documentation is available at [Composable Kernel](https://rocm.docs.amd.com/projects/composable_kernel/en/latest/) in an organized, easy-to-read format, with search and a table of contents. The documentation source files reside in the `docs` folder of this repository. As with all ROCm projects, the documentation is open source. For more information on contributing to the documentation, see [Contribute to ROCm documentation](https://rocm.docs.amd.com/en/latest/contribute/contributing.html).
The Composable Kernel (CK) library provides a programming model for writing performance-critical
The Composable Kernel (CK) library provides a programming model for writing performance-critical
kernels for machine learning workloads across multiple architectures (GPUs, CPUs, etc.). The CK library
kernels for machine learning workloads across multiple architectures (GPUs, CPUs, etc.). The CK library
uses general purpose kernel languages, such as HIP C++.
uses general purpose kernel languages, such as HIP C++.
...
@@ -23,23 +26,15 @@ The current CK library is structured into four layers:
...
@@ -23,23 +26,15 @@ The current CK library is structured into four layers:
## General information
## General information
To build our documentation locally, use the following code:
By default, `-j` launches one thread per CPU core, which can cause the build to run out of memory and
Note the `-j` option for building with multiple threads in parallel, which speeds up the build significantly.
crash. In such cases, you can reduce the number of threads to 32 by using `-j32`.
However, `-j` launches unlimited number of threads, which can cause the build to run out of memory and
crash. On average, you should expect each thread to use ~2Gb of RAM.
Depending on the number of CPU cores and the amount of RAM on your system, you may want to
limit the number of threads. For example, if you have a 128-core CPU and 128 Gb of RAM it's advisable to use `-j32`.
Additional cmake flags can be used to significantly speed-up the build:
Additional cmake flags can be used to significantly speed-up the build:
...
@@ -151,6 +153,10 @@ Additional cmake flags can be used to significantly speed-up the build:
...
@@ -151,6 +153,10 @@ Additional cmake flags can be used to significantly speed-up the build:
`batched_gemm_multi_d_dl`. These instances are useful on architectures like the NAVI2x, as most
`batched_gemm_multi_d_dl`. These instances are useful on architectures like the NAVI2x, as most
other platforms have faster instances, such as `xdl` or `wmma`, available.
other platforms have faster instances, such as `xdl` or `wmma`, available.
*`CK_USE_FP8_ON_UNSUPPORTED_ARCH` (default is OFF) must be set to ON in order to build instances,
such as `gemm_universal`, `gemm_universal_streamk` and `gemm_multiply_multiply` for fp8 data type for GPU targets which do not have native support for fp8 data type, such as gfx908 or gfx90a. These instances are useful on
architectures like the MI100/MI200 for the functional support only.
## Using sccache for building
## Using sccache for building
The default CK Docker images come with a pre-installed version of sccache, which supports clang
The default CK Docker images come with a pre-installed version of sccache, which supports clang
This tutorial demonstrates how to implement matrix multiplication using Composable Kernel (CK)
This tutorial demonstrates how to implement matrix multiplication using Composable Kernel (CK) wrapper. We present the base version of GEMM without most of the available optimizations; however, it's worth noting that CK has kernels with different optimizations.
wrapper. We present the base version of GEMM without most of the available optimizations; however,
it's worth noting that CK has kernels with different optimizations.
To implement these optimizations, you can use the CK wrapper or directly use available instances in
To implement these optimizations, you can use the CK wrapper or directly use available instances in CK. You can also refer to the [optimized GEMM example](https://github.com/ROCm/composable_kernel/blob/develop/client_example/25_wrapper/wrapper_optimized_gemm.cpp), that uses CK wrapper based on the [`gridwise_gemm_xdlops_v2r3`](https://github.com/ROCm/composable_kernel/blob/develop/include/ck/tensor_operation/gpu/grid/gridwise_gemm_xdlops_v2r3.hpp) implementation.
