		     README for CUDA-GDB release

This is CUDA-GDB, the NVIDIA CUDA source-level debugger for Linux,
based on GDB-16.3, the GNU source-level debugger.

For more information about gdb, please refer to the gdb/README file or
check the GDB home page at https://www.gnu.org/software/gdb

CUDA-GDB BUILD INSTRUCTIONS (x86_64 Linux only, adjust as needed)
=================================================================

First, make sure that the required library dependencies are installed
on your system.
Install the required development packages by running the commands:
On Ubuntu:
    sudo apt-get update && \
    sudo apt-get install libncurses-dev libgmp-dev libmpfr-dev \
        libreadline-dev bison flex texinfo python3-dev
On RHEL/CentOS:
    sudo yum install ncurses-devel gmp-devel mpfr-devel \
        readline-devel bison flex texinfo python3-devel
The configure command will print an error if required packages are missing.

Please note that the libexpat development package must be installed if CUDA-GDB
is to be used for cross-platform debugging.

Note that the value of the CUDA_HOME variable should match the actual CUDA
Toolkit installation directory.

Run the following commands to build CUDA-GDB:
    CUDA_HOME="${CUDA_HOME:-/usr/local/cuda}" \
    CUDA_INC_FLAGS="-I${CUDA_HOME}/include/ -I${CUDA_HOME}/extras/Debugger/include/" \
    ./configure --program-prefix=cuda- \
        --enable-cuda \
        --enable-targets=x86_64-unknown-linux-gnu,arm-elf-linux-gnu,m68k-unknown-linux-gnu \
        CFLAGS="${CUDA_INC_FLAGS}" \
        CXXFLAGS="${CUDA_INC_FLAGS}" \
        LDFLAGS="-lpthread"
    make -j$(nproc)


USING CUDA-GDB
==============
All standard GDB commands can be used both for CPU and GPU code debugging.
In addition to that, CUDA-GDB provides CUDA-specific command families like 
"info cuda ..." to query GPU states, "cuda .." to control debugger focus on
GPU and "[get|set] cuda .." to alter/query CUDA debugger configuration.

For more information please check https://docs.nvidia.com/cuda/cuda-gdb/


REPORTING BUGS
==============
Use any of the following options.
1. Send e-mail to cuda-debugger-bugs@nvidia.com.
2. File an issue report at the NVIDIA registered developers portal:
   https://developer.nvidia.com/cuda-gdb
3. Post a new topic at the NVIDIA CUDA-GDB forum:
   https://forums.developer.nvidia.com/c/developer-tools/cuda-developer-tools/cuda-gdb
4. File an issue report at the NVIDIA CUDA-GDB Github:
   https://github.com/NVIDIA/cuda-gdb/issues
