Merge branch 'develop' into codegen_hiprtc

script/process_qa_data.sh

@@ -52,3 +52,19 @@ file=./perf_fmha_bwd_gfx90a.log
 if [ -e "$file" ]; then
     python3 process_perf_data.py perf_fmha_bwd_gfx90a.log
 fi
+file=./perf_gemm_basic_gfx942.log
+if [ -e "$file" ]; then
+    python3 process_perf_data.py perf_gemm_basic_gfx942.log
+fi
+file=./perf_gemm_basic_gfx90a.log
+if [ -e "$file" ]; then
+    python3 process_perf_data.py perf_gemm_basic_gfx90a.log
+fi
+file=./perf_gemm_mem_pipeline_gfx942.log
+if [ -e "$file" ]; then
+    python3 process_perf_data.py perf_gemm_mem_pipeline_gfx942.log
+fi
+file=./perf_gemm_mem_pipeline_gfx90a.log
+if [ -e "$file" ]; then
+    python3 process_perf_data.py perf_gemm_mem_pipeline_gfx90a.log
+fi

test/grouped_convnd_bwd_weight/test_grouped_convnd_bwd_weight.cpp

 // SPDX-License-Identifier: MIT
-// Copyright (c) 2018-2024, Advanced Micro Devices, Inc. All rights reserved.
+// Copyright (c) 2018-2025, Advanced Micro Devices, Inc. All rights reserved.
 
 #include <cstdlib>
 #include <iostream>
@@ -43,7 +43,6 @@ class TestGroupedConvndBwdWeight : public ::testing::Test
                 return true;
             }
         }
-
         if(ck::is_gfx11_supported() || ck::is_gfx12_supported())
         {
             // on gfx11 only support for 3d is implemented
@@ -143,19 +142,23 @@ using KernelTypes2d = ::testing::Types<
     std::tuple<float, float, float, GNHWC, GKYXC, GNHWK, ck::Number<2>>,
     std::tuple<ck::half_t, ck::half_t, ck::half_t, GNHWC, GKYXC, GNHWK, ck::Number<2>>,
     std::tuple<ck::bhalf_t, float, ck::bhalf_t, GNHWC, GKYXC, GNHWK, ck::Number<2>>,
+    std::tuple<ck::bhalf_t, ck::bhalf_t, ck::bhalf_t, NHWGC, GKYXC, NHWGK, ck::Number<2>>,
     std::tuple<float, float, float, NHWGC, GKYXC, NHWGK, ck::Number<2>>,
     std::tuple<ck::half_t, ck::half_t, ck::half_t, NHWGC, GKYXC, NHWGK, ck::Number<2>>,
     std::tuple<ck::bhalf_t, float, ck::bhalf_t, NHWGC, GKYXC, NHWGK, ck::Number<2>>,
+    std::tuple<ck::bhalf_t, ck::bhalf_t, ck::bhalf_t, NGCHW, GKYXC, NGKHW, ck::Number<2>>,
     std::tuple<ck::half_t, ck::half_t, ck::half_t, NGCHW, GKYXC, NGKHW, ck::Number<2>>>;
 using KernelTypes3d = ::testing::Types<
     std::tuple<float, float, float, GNDHWC, GKZYXC, GNDHWK, ck::Number<3>>,
     std::tuple<ck::half_t, ck::half_t, ck::half_t, GNDHWC, GKZYXC, GNDHWK, ck::Number<3>>,
     std::tuple<ck::bhalf_t, float, ck::bhalf_t, GNDHWC, GKZYXC, GNDHWK, ck::Number<3>>,
     std::tuple<int8_t, int8_t, int8_t, GNDHWC, GKZYXC, GNDHWK, ck::Number<3>>,
+    std::tuple<ck::bhalf_t, ck::bhalf_t, ck::bhalf_t, NDHWGC, GKZYXC, NDHWGK, ck::Number<3>>,
     std::tuple<float, float, float, NDHWGC, GKZYXC, NDHWGK, ck::Number<3>>,
     std::tuple<ck::half_t, ck::half_t, ck::half_t, NDHWGC, GKZYXC, NDHWGK, ck::Number<3>>,
     std::tuple<ck::bhalf_t, float, ck::bhalf_t, NDHWGC, GKZYXC, NDHWGK, ck::Number<3>>,
     std::tuple<int8_t, int8_t, int8_t, NDHWGC, GKZYXC, NDHWGK, ck::Number<3>>,
+    std::tuple<ck::bhalf_t, ck::bhalf_t, ck::bhalf_t, NGCDHW, GKZYXC, NGKDHW, ck::Number<3>>,
     std::tuple<ck::half_t, ck::half_t, ck::half_t, NGCDHW, GKZYXC, NGKDHW, ck::Number<3>>>;
 
 TYPED_TEST_SUITE(TestGroupedConvndBwdWeight1d, KernelTypes1d);
@@ -179,6 +182,8 @@ TYPED_TEST(TestGroupedConvndBwdWeight2d, Test2D)
     this->conv_params.clear();
     this->conv_params.push_back(
         {2, 2, 64, 128, 256, {1, 1}, {7, 7}, {2, 2}, {1, 1}, {0, 0}, {0, 0}});
+    this->conv_params.push_back({2, 2, 64, 3, 3, {1, 1}, {7, 7}, {1, 1}, {1, 1}, {0, 0}, {0, 0}});
+    this->conv_params.push_back({2, 2, 64, 5, 5, {1, 1}, {7, 7}, {1, 1}, {1, 1}, {0, 0}, {0, 0}});
     this->conv_params.push_back(
         {2, 2, 4, 128, 256, {3, 3}, {14, 14}, {1, 1}, {1, 1}, {1, 1}, {1, 1}});
     this->conv_params.push_back(