You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
To generate executables to run on nVidia GPUs you also need an installation of [CUDA >= 8.1](https://developer.nvidia.com/cuda-toolkit) and to use a [compatible GCC version](https://docs.nvidia.com/cuda/).
10
+
11
+
Build TCode following the instructions below:
4
12
1. clone the git repository: `git clone https://github.com/MultithreadCorner/TCode.git`
5
13
2. go to TCode directory: `cd TCode`
6
14
3. create a build directory: `mkdir build`
7
-
4. cd build
8
-
5.`cmake -DHYDRA_INCLUDE_DIR=<path-to-hydra>../`
9
-
6.`make -j8`
15
+
4.go to build directory cd `build`
16
+
5.run cmake specifying the path to Hydra: `cmake -DHYDRA_INCLUDE_DIR='path-to-hydra'../`
17
+
6.compile (all backends): `make -j8`
10
18
11
-
Separate executablesfor CPP, TBB, OMP and CUDA backend will be produced (if available in your machine).
19
+
Several executables, separate for each available backend (CPP, TBB, OMP and CUDA) will be generated.
0 commit comments