Merge remote-tracking branch 'origin/develop' into miopen_downstream-dynamic_reduction_pr

b725e3fc · Chao Liu · df0d6810 · f3acd251 · b725e3fc · b725e3fc
Commit b725e3fc authored Sep 21, 2021 by Chao Liu
4 changed files
--- a/host/host_tensor/include/host_conv_bwd_weight.hpp
+++ b/host/host_tensor/include/host_conv_bwd_weight.hpp
--- a/host/host_tensor/include/host_gemm.hpp
+++ b/host/host_tensor/include/host_gemm.hpp
--- a/host/solver/include/conv_tunable_fwd_v4r4_xdlops_nchw_kcyx_nkhw.hpp
+++ b/host/solver/include/conv_tunable_fwd_v4r4_xdlops_nchw_kcyx_nkhw.hpp
@@ -9,8 +9,8 @@ struct tunable_dyn_conv_fwd_v4r4_xdlops_nchw_kcyx_nkhw
    int NPerBlock;
    int KPerBlock;

-    int MPerWave;
-    int NPerWave;
+    int MPerXDL;
+    int NPerXDL;
    int K1;

    int MRepeat;
@@ -45,8 +45,8 @@ static tunable_dyn_conv_fwd_v4r4_xdlops_nchw_kcyx_nkhw
        128,                      // MPerBlock,
        128,                      // NPerBlock,
        4,                        // KPerBlock,
-        32,                       // MPerWave,
-        32,                       // NPerWave,
+        32,                       // MPerXDL,
+        32,                       // NPerXDL,
        4,                        // K1,
        2,                        // MRepeat,
        2,                        // NRepeat,

--- a/script/run.sh
+++ b/script/run.sh