Warptiling bench

vkarihal · Feb 20, 2023 · 3659fe4 · 3659fe4
1 parent 49b3e8a
commit 3659fe4
Show file tree

Hide file tree

Showing 2 changed files with 4 additions and 3 deletions.
diff --git a/README.md b/README.md
@@ -22,8 +22,9 @@ GFLOPs at matrix size 4092x4092:
 | 8: Avoid Bank Conflicts (Offset)    |    11646.9 | 78.7%                            |
 | 7: Avoid Bank Conflicts (Linearize) |    11923.9 | 80.6%                            |
 | 6: Vectorized Mem Access            |    12088.9 | 81.7%                            |
-| 9: Autotuning                       |    12717.4 | 86.0%                            |
-| 0: cuBLAS                           |    14792.5 | 100.0%                           |
+| 9: Autotuning                       |    12715.3 | 85.9%                            |
+| 10: Warptiling                      |    12828.0   | 86.7%                            |
+| 0: cuBLAS                           |    14797.8 | 100.0%                           |
 <!-- benchmark_results -->
 
 ## Setup

diff --git a/gen_benchmark_results.sh b/gen_benchmark_results.sh
@@ -8,7 +8,7 @@ set -euo pipefail
 
 mkdir -p benchmark_results
 
-for kernel in {0..9}; do
+for kernel in {0..10}; do
     echo ""
     ./build/sgemm $kernel | tee "benchmark_results/${kernel}_output.txt"
     sleep 2