Quote:
Originally Posted by bsquared
Very nice!
I haven't been able to build it yet. My knowledge of automake/autoconf is very limited, but trying to use these tools just throws a bunch of errors about missing files. Any advice on building this for linux?
|
There's some instructions in README.dev.
This is what I use
Code:
autoreconf -i
./configure --enable-gpu=61 --with-cuda=/usr/local/cuda CC=gcc-9 -with-cgbn-include=/home/five/Projects/CGBN/include/cgbn
make -j8
Replace `--enable_gpu=61` with a different compute capability if you need one (and apply the change from
#27 if you need SM_70) but you already seem to have that figured out.
Quote:
Originally Posted by chris2be8
I'm trying to build it on Linux, but have not had much luck. I copied the process I used to build the previous version:
What version of CUDA Toolkit and runtime is needed to support CGBN? And where is cgbn.h on your system?
|
I ran `git clone https://github.com/NVlabs/CGBN.git` under /home/five/Projects
then added `-with-cgbn-include=/home/five/Projects/CGBN/include/cgbn` to the list of options I pass to `./configure`