From bae67e6a355047e360c30391588c2076913f86fa Mon Sep 17 00:00:00 2001 From: negargoli93 Date: Sat, 12 May 2018 17:11:34 -0700 Subject: mma added --- src/cuda-sim/instructions.cc | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'src/cuda-sim') diff --git a/src/cuda-sim/instructions.cc b/src/cuda-sim/instructions.cc index 7903343..bddaf8c 100644 --- a/src/cuda-sim/instructions.cc +++ b/src/cuda-sim/instructions.cc @@ -1565,7 +1565,7 @@ void mma_impl( const ptx_instruction *pI, core_t *core, warp_inst_t inst ) for (i=0;i<16;i++){ for(j=0;j<16;j++){ for(k=0;k<16;k++){ - matrix_d[i][j].f32=matrix_d[i][j].f32+matrix_a[i][k].f32*matrix_b[j][k].f32; + matrix_d[i][j].f32=matrix_d[i][j].f32+matrix_a[i][k].f32*matrix_b[k][j].f32; } matrix_d[i][j].f32+=matrix_c[i][j].f32; } -- cgit v1.3