Merge branch 'develop' into barkocot/image-to-column-with-groups

docs/sphinx/requirements.in

 rocm-docs-core>=0.20.0
-sphinxcontrib-bibtex==2.5.0
+sphinxcontrib-bibtex==2.6.1

docs/sphinx/requirements.txt

@@ -139,7 +139,7 @@ sphinx-notfound-page==0.8.3
     # via rocm-docs-core
 sphinxcontrib-applehelp==1.0.4
     # via sphinx
-sphinxcontrib-bibtex==2.5.0
+sphinxcontrib-bibtex==2.6.1
     # via -r requirements.in
 sphinxcontrib-devhelp==1.0.2
     # via sphinx

example/27_layernorm/run_layernorm_example.inc

@@ -8,8 +8,8 @@ int run_groupnorm_example()
 {
     bool time_kernel = false;
 
-    ck::index_t M      = 1024;
-    ck::index_t N      = 1024;
+    ck::index_t M = 1024;
+    ck::index_t N = 1024;
 
     Tensor<XDataType> x({M, N});
     Tensor<GammaDataType> gamma({N});
@@ -44,9 +44,9 @@ int run_groupnorm_example()
         {0, 1},
         std::vector<ck::index_t>{y.mDesc.GetStrides().begin(), y.mDesc.GetStrides().end()},
         std::vector<ck::index_t>{save_mean.mDesc.GetStrides().begin(),
-                                    save_mean.mDesc.GetStrides().end()},
+                                 save_mean.mDesc.GetStrides().end()},
         std::vector<ck::index_t>{save_mean.mDesc.GetStrides().begin(),
-                                    save_mean.mDesc.GetStrides().end()},
+                                 save_mean.mDesc.GetStrides().end()},
         {1},
         1e-4,
         x_dev.GetDeviceBuffer(),

example/42_groupnorm/run_groupnorm_example.inc

@@ -65,9 +65,9 @@ int run_groupnorm_example(int argc, char* argv[])
         {0, 0, 0, C, 1},
         std::vector<ck::index_t>{y.mDesc.GetStrides().begin(), y.mDesc.GetStrides().end()},
         std::vector<ck::index_t>{save_mean.mDesc.GetStrides().begin(),
-                                    save_mean.mDesc.GetStrides().end()},
+                                 save_mean.mDesc.GetStrides().end()},
         std::vector<ck::index_t>{save_mean.mDesc.GetStrides().begin(),
-                                    save_mean.mDesc.GetStrides().end()},
+                                 save_mean.mDesc.GetStrides().end()},
         {1, 2, 4}, // reduction dimension: [H, W, C]
         1e-6,
         x_dev.GetDeviceBuffer(),

include/ck/utility/type_convert.hpp

@@ -100,6 +100,8 @@ template <>
 inline __host__ __device__ f8_t type_convert<f8_t, float>(float x)
 {
 #if defined(__gfx940__) || defined(__gfx941__) || defined(__gfx942__)
+    float max_fp8 = 240.0f;
+    x             = x > max_fp8 ? max_fp8 : (x < -max_fp8 ? -max_fp8 : x);
     union
     {
         float fval;

profiler/include/profiler/profile_gemm_impl.hpp

@@ -75,8 +75,8 @@ int profile_gemm_impl(int do_verification,
         b_k_n.GenerateTensorValue(GeneratorTensor_2<BDataType>{-5, 5});
         break;
     default:
-        a_m_k.GenerateTensorValue(GeneratorTensor_3<ADataType>{0.0, 1.0});
-        b_k_n.GenerateTensorValue(GeneratorTensor_3<BDataType>{-0.5, 0.5});
+        a_m_k.GenerateTensorValue(GeneratorTensor_3<ADataType>{0.0, 0.1});
+        b_k_n.GenerateTensorValue(GeneratorTensor_3<BDataType>{-0.05, 0.05});
     }
 
     using AElementOp = ck::tensor_operation::element_wise::PassThrough;