diff options
author | Cedric Nugteren <web@cedricnugteren.nl> | 2018-09-16 20:01:18 +0200 |
---|---|---|
committer | Cedric Nugteren <web@cedricnugteren.nl> | 2018-09-16 20:01:18 +0200 |
commit | 83ba3d4b7ba3a9cb5fbd2c1ad2bb14b2addd39fb (patch) | |
tree | 58900a63158d08e76342b46372fcc59015b4d3ca /src/tuning/kernels | |
parent | b7d833901213d03fe5e7f10c15741f55c6c1eb54 (diff) | |
parent | c163868e1822a97750b4380f0d9cdd38369f9f0b (diff) |
Merge branch 'master' into convgemm_multi_kernel
Diffstat (limited to 'src/tuning/kernels')
-rw-r--r-- | src/tuning/kernels/xgemm.cpp | 4 | ||||
-rw-r--r-- | src/tuning/kernels/xgemm.hpp | 2 |
2 files changed, 6 insertions, 0 deletions
diff --git a/src/tuning/kernels/xgemm.cpp b/src/tuning/kernels/xgemm.cpp index 75e776e6..dd907ba4 100644 --- a/src/tuning/kernels/xgemm.cpp +++ b/src/tuning/kernels/xgemm.cpp @@ -33,9 +33,13 @@ void StartVariation(int argc, char *argv[]) { // Main function (not within the clblast namespace) int main(int argc, char *argv[]) { + printf("* (1/4) Tuning main GEMM kernel (GEMMK == 0) for fixed set of parameters\n\n"); StartVariation<1>(argc, argv); + printf("* (2/4) Tuning main GEMM kernel (GEMMK == 0) for random parameters out of larger set\n\n"); StartVariation<2>(argc, argv); + printf("* (3/4) Tuning secondary GEMM kernel (GEMMK == 1) for fixed set of parameters\n\n"); StartVariation<11>(argc, argv); + printf("* (4/4) Tuning secondary GEMM kernel (GEMMK == 1) for random parameters out of larger set\n\n"); StartVariation<12>(argc, argv); return 0; } diff --git a/src/tuning/kernels/xgemm.hpp b/src/tuning/kernels/xgemm.hpp index 9a538c1b..fa1bb6ec 100644 --- a/src/tuning/kernels/xgemm.hpp +++ b/src/tuning/kernels/xgemm.hpp @@ -50,6 +50,8 @@ TunerSettings XgemmGetTunerSettings(const int V, const Arguments<T> &args) { settings.sources += #include "../src/kernels/level3/xgemm_part1.opencl" #include "../src/kernels/level3/xgemm_part2.opencl" + ; + settings.sources += #include "../src/kernels/level3/xgemm_part3.opencl" #include "../src/kernels/level3/xgemm_part4.opencl" ; |