Merge pull request #258 from electronic-structure/develop

Develop
electronic-structure · Sep 12, 2018 · 765d1ce · 765d1ce
2 parents fc1c648 + 1db9cd5
commit 765d1ce
Show file tree

Hide file tree

Showing 2 changed files with 10 additions and 10 deletions.
diff --git a/src/Hamiltonian/set_lapw_h_o.hpp b/src/Hamiltonian/set_lapw_h_o.hpp
@@ -241,10 +241,10 @@ inline void Hamiltonian::set_fv_h_o<GPU, electronic_structure_method_t::full_pot
     acc::copyout(o__.at<CPU>(), o__.ld(), o__.at<GPU>(), o__.ld(), kp__->num_gkvec_row(), kp__->num_gkvec_col());
 
     double tval = t1.stop();
-    if (kp__->comm().rank() == 0 && ctx_.control().print_performance_) {
-        DUMP("effective zgemm performance: %12.6f GFlops",
-             2 * 8e-9 * kp__->num_gkvec() * kp__->num_gkvec() * unit_cell_.mt_aw_basis_size() / tval);
-    }
+    //if (kp__->comm().rank() == 0 && ctx_.control().print_performance_) {
+    //    DUMP("effective zgemm performance: %12.6f GFlops",
+    //         2 * 8e-9 * kp__->num_gkvec() * kp__->num_gkvec() * unit_cell_.mt_aw_basis_size() / tval);
+    //}
 
     /* add interstitial contributon */
     set_fv_h_o_it(kp__, h__, o__);

diff --git a/src/Unit_cell/atom.hpp b/src/Unit_cell/atom.hpp
@@ -262,15 +262,15 @@ class Atom
             t1.stop();
 
             result.allocate(memory_t::device);
-            utils::timer t2("sirius::Atom::generate_radial_integrals|inner");
+            //utils::timer t2("sirius::Atom::generate_radial_integrals|inner");
             spline_inner_product_gpu_v3(idx_ri.at<GPU>(), (int)idx_ri.size(1), nmtp, rgrid.x().at<GPU>(),
                                         rgrid.dx().at<GPU>(), rf_coef.at<GPU>(), vrf_coef.at<GPU>(), result.at<GPU>());
             acc::sync();
-            if (type().parameters().control().print_performance_) {
-                double tval = t2.stop();
-                DUMP("spline GPU integration performance: %12.6f GFlops",
-                     1e-9 * double(idx_ri.size(1)) * nmtp * 85 / tval);
-            }
+            //if (type().parameters().control().print_performance_) {
+            //    double tval = t2.stop();
+            //    DUMP("spline GPU integration performance: %12.6f GFlops",
+            //         1e-9 * double(idx_ri.size(1)) * nmtp * 85 / tval);
+            //}
             result.copy<memory_t::device, memory_t::host>();
             result.deallocate(memory_t::device);
 #else