The code in this repo is now quite out of date compared to master llvm plus all the RISC-V patches that were added here have now been upstreamed.
However, the build instructions themselves are still a good quick-start guide.
I've updated the commands below to clone llvm from the official llvm repository, not from here, and to account for RISC-V now having been upgraded to non-experimental status.
This repository adds support for various aspects of the RISC-V instruction set to the Clang C/C++ compiler and LLVM back end.
The "master" branch follows official LLVM.
The "dev" branch at the moment contains mostly patches developed by the LowRISC project that have not yet been upstreamed to the official LLVM repository. It is updated from the master branch regularly. As more patches are committed to upstream LLVM the differences will become smaller, but the intention is to add other patches here so people can conveniently use them in advance of them being accepted into upstream.
As well as this repository, you will need the RISC-V gcc toolchain. If you don't have RISC-V hardware then you will want to have QEMU to run your programs.
The following bash commands have been tested on a fresh Ubuntu install (e.g. Amazon EC2 Ubuntu 16.04 LTS ami-0e32ec5bc225539f5 and 18.04 LTS ami-0bbe6b35405ecebdb) and are known to result in a working RISC-V LLVM.
If you already have recent RISC-V gnu tools and/or qemu in your PATH then you can omit the steps to download and build those.
At the moment the clang driver does not successfully find the gnu linker and libraries so you need to explicitly use gcc to link your object files. This should be fixed soon.
By default clang is generating rv32imac/rv64imac code but you can override this, and in particular if you have floating point hardware you can add "-march=rv32gc" (or rv64gc) to the clang command line to get FPU instructions with a soft float ABI (scalar FP function arguments and results passed in integer registers).
sudo apt-get update
sudo apt-get -y dist-upgrade
The dist-upgrade can pop up a text GUI and ask questions (for example I've seen them about updating grub), which eats input already pasted, so it's best to wait for the result from the above two lines by themselves before proceeding.
sudo apt-get -y install \
binutils build-essential libtool texinfo \
gzip zip unzip patchutils curl git \
make cmake ninja-build automake bison flex gperf \
grep sed gawk python bc \
zlib1g-dev libexpat1-dev libmpc-dev \
libglib2.0-dev libfdt-dev libpixman-1-dev
mkdir riscv
cd riscv
mkdir _install
export PATH=`pwd`/_install/bin:$PATH
hash -r
# gcc, binutils, newlib
git clone --recursive https://github.com/riscv/riscv-gnu-toolchain
pushd riscv-gnu-toolchain
./configure --prefix=`pwd`/../_install --enable-multilib
make -j`nproc`
# qemu
make -j`nproc` build-qemu
popd
# LLVM
git clone https://github.com/llvm/llvm-project.git riscv-llvm
#git clone https://github.com/sifive/riscv-llvm
pushd riscv-llvm
ln -s ../../clang llvm/tools || true
mkdir _build
cd _build
cmake -G Ninja -DCMAKE_BUILD_TYPE="Release" \
-DBUILD_SHARED_LIBS=True -DLLVM_USE_SPLIT_DWARF=True \
-DCMAKE_INSTALL_PREFIX="../../_install" \
-DLLVM_OPTIMIZED_TABLEGEN=True -DLLVM_BUILD_TESTS=False \
-DDEFAULT_SYSROOT="../../_install/riscv64-unknown-elf" \
-DLLVM_DEFAULT_TARGET_TRIPLE="riscv64-unknown-elf" \
-DLLVM_TARGETS_TO_BUILD="RISCV" \
../llvm
cmake --build . --target install
popd
# Sanity test your new RISC-V LLVM
cat >hello.c <<END
#include <stdio.h>
int main(){
printf("Hello RISCV!\n");
return 0;
}
END
# 32 bit
clang -O -c hello.c --target=riscv32
riscv64-unknown-elf-gcc hello.o -o hello -march=rv32imac -mabi=ilp32
qemu-riscv32 hello
# 64 bit
clang -O -c hello.c
riscv64-unknown-elf-gcc hello.o -o hello -march=rv64imac -mabi=lp64
qemu-riscv64 hello