HCC2 - V 0.5-4
hcc2: Heterogeneous Compiler Collection (Version 2).
This is README.md for https:/github.com/ROCM-Developer-Tools/hcc2 . This is the base repository for HCC2, Use this for issues, documentation, packaging, examples, build.
HCC2 is an experimental PROTOTYPE that is intended to support multiple programming models including OpenMP 4.5+, C++ parallel extentions (original HCC), HIP, and cuda clang. It supports offloading to multiple GPU acceleration targets(multi-target). It also supports different host platforms such as AMD64, PPC64LE, and AARCH64. (multi-platform).
The bin directory of this repository contains a README and build scripts needed to build and install HCC2. However, we recommend that you install from the debian or rpm packages described below.
Attention Users! Use this repository for issues. Do not put issues in the source code repositories. Before creating an issue, you may want to see the developers list of TODOs. See link below.
Table of contents
- Copyright and Disclaimer
- Software License Agreement
- TODOs List of TODOs for this release
Copyright and Disclaimer Copyright (c) 2017 ADVANCED MICRO DEVICES, INC.
AMD is granting you permission to use this software and documentation (if any) (collectively, the �Materials�) pursuant to the terms and conditions of the Software License Agreement included with the Materials. If you do not have a copy of the Software License Agreement, contact your AMD representative for a copy.
You agree that you will not reverse engineer or decompile the Materials, in whole or in part, except for example code which is provided in source code form and as allowed by applicable law.
WARRANTY DISCLAIMER: THE SOFTWARE IS PROVIDED "AS IS" WITHOUT WARRANTY OF ANY KIND. AMD DISCLAIMS ALL WARRANTIES, EXPRESS, IMPLIED, OR STATUTORY, INCLUDING BUT NOT LIMITED TO THE IMPLIED WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE, TITLE, NON-INFRINGEMENT, THAT THE SOFTWARE WILL RUN UNINTERRUPTED OR ERROR- FREE OR WARRANTIES ARISING FROM CUSTOM OF TRADE OR COURSE OF USAGE. THE ENTIRE RISK ASSOCIATED WITH THE USE OF THE SOFTWARE IS ASSUMED BY YOU. Some jurisdictions do not allow the exclusion of implied warranties, so the above exclusion may not apply to You.
LIMITATION OF LIABILITY AND INDEMNIFICATION: AMD AND ITS LICENSORS WILL NOT, UNDER ANY CIRCUMSTANCES BE LIABLE TO YOU FOR ANY PUNITIVE, DIRECT, INCIDENTAL, INDIRECT, SPECIAL OR CONSEQUENTIAL DAMAGES ARISING FROM USE OF THE SOFTWARE OR THIS AGREEMENT EVEN IF AMD AND ITS LICENSORS HAVE BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES. In no event shall AMD's total liability to You for all damages, losses, and causes of action (whether in contract, tort (including negligence) or otherwise) exceed the amount of $100 USD. You agree to defend, indemnify and hold harmless AMD and its licensors, and any of their directors, officers, employees, affiliates or agents from and against any and all loss, damage, liability and other expenses (including reasonable attorneys' fees), resulting from Your use of the Software or violation of the terms and conditions of this Agreement.
U.S. GOVERNMENT RESTRICTED RIGHTS: The Materials are provided with "RESTRICTED RIGHTS." Use, duplication, or disclosure by the Government is subject to the restrictions as set forth in FAR 52.227-14 and DFAR252.227-7013, et seq., or its successor. Use of the Materials by the Government constitutes acknowledgement of AMD's proprietary rights in them.
EXPORT RESTRICTIONS: The Materials may be subject to export restrictions as stated in the Software License Agreement.
On Ubuntu 16.04 LTS (xenial), run these commands:
wget https://github.com/ROCm-Developer-Tools/hcc2/releases/download/rel_0.5-4/hcc2_0.5-4_amd64.deb sudo dpkg -P hcc2 sudo dpkg -P libamdgcn sudo dpkg -P amdcloc sudo dpkg -P mymcpu sudo dpkg -i hcc2_0.5-4_amd64.deb
The "dpkg -P" commands are used to delete previous versions of hcc2, libamdgcn, amdcloc, and mymcpu which may conflict with the installation. If these are not installed it is ok to just let the "dpkg -P" commands fail.
HCC2 does not conflict with the production HCC. There is no reason to delete HCC to use HCC2. The HCC2 bin directory (which includes the standard clang and llvm binaries) is not intended to be in your PATH for typical operation.
For rpm-based Linux distributions, use this rpm
wget https://github.com/ROCm-Developer-Tools/hcc2/releases/download/rel_0.5-4/hcc2-0.5-4.x86_64.rpm sudo rpm -i hcc2-0.5-4.x86_64.rpm
No root Install
The current packages are built without listing dependencies. By default, they install their content to the release directory /opt/rocm/hcc2_0.X-Y and then a symbolic link is created at /opt/rocm/hcc2 to the release directory. This requires root access. You can use the --prefix option of the rpm install command to change the location so as not to require root access.
wget https://github.com/ROCm-Developer-Tools/hcc2/releases/download/rel_0.5-4/hcc2-0.5-4.x86_64.rpm mkdir -p $HOME/rocm/hcc2 rpm -i hcc2-0.5-4.x86_64.rpm --prefix=$HOME/rocm/hcc2
Then permanently set the environment variable HCC2 to $HOME/rocm/hcc2. For example in .bash_profile add the command export HCC2=$HOME/rocm/hcc2
Build and install from sources is possible. However, the source build for HCC2 is complex for three reasons. - Many repos are required . There is a script to ensure you have all repos and checkout the correct branch. - Requires that both the Cuda SDK and ROCm are installed regardless of which graphic cards you have. - It is a bootstrapped build. The built and installed LLVM compiler is used to build library components. For details on the source build see [README](bin/README). ## HCC2 Limitations <A NAME="Limitations"> See the release notes in github. Here are some limitations.
- target teams distribute reduce does not work
- Dwarf debugging is turned off for GPUs. -g will turn on host level debugging only.
- Some simd constructs fail to vectorize on both host and GPUs.
- There are debug versions of the runtime libraries. However, these use printf on the device which currently print when the kernel terminates. So it is not a very useful debug feature if the GPU kernel crashes.