All Projects → NVIDIA → Cuda Gdb

NVIDIA / Cuda Gdb

Licence: other
CUDA GDB

Programming Languages

c
50402 projects - #5 most used programming language
	     README for CUDA-GDB release

This is CUDA-GDB, the NVIDIA CUDA source-level debugger for Linux and Mac OS, based on GDB-7.6, the GNU source-level debugger.

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

CUDA-GDB BUILD INSTRUCTIONS (example only, adjust as needed)

First, make sure that libtermcap and other required dependent packages are installed (try sudo yum install ncurses-devel). The configure command will issue an error if some packages are missing.

Please note that the libexpat development headers must be present if cuda-gdb is to be used for cross-platform debugging.

Issue the following commands to build cuda-gdb: ./configure --program-prefix=cuda-
--enable-cuda
--enable-targets="x86_64-apple-darwin,x86_64-unknown-linux-gnu,
arm-elf-linux-gnu,m68k-unknown-linux-gnu"
CFLAGS='-I/usr/local/cuda/include'
LDFLAGS='-lpthread' make

USING CUDA-GDB

All standard GDB commands could 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 http://docs.nvidia.com/cuda/cuda-gdb/

REPORTING BUGS

Send e-mail to [email protected] or fill an issue report form at NVIDIA registered developers portal.

Note that the project description data, including the texts, logos, images, and/or trademarks, for each open source project belongs to its rightful owner. If you wish to add or remove any projects, please contact us at [email protected].