Rebase the PR #1520 to ROCm repo. (#1574)

* Implement hiprtc for codegen tests

* Introduce gemm_softmax_gemm to codegen.

* Fix codegen build issues.

* Address PR comments.

* Separate ck_host lib and gemm_softmax_gemm into different PR.

* Fix cmake.

* Replace ENV variable with CMake option for toggling hipRTC in codegen
tests.

* Address PR comments.

* fix clang format

* Add missing header in magic_division.hpp

* - Workaround for hipRTC content wrapper
- Move descriptor for gemm_softmax_gemm to different branch

* Fix formatting.

* Revert "Fix formatting."

This reverts commit b5209eaef4.

* formatting fix

* fixed header guard issues

* updated header guards

* updated data_type for new types

* fixed redefinition error

* Add codegen test for batched_gemm_softmax_gemm.

Signed-off-by: Mirza Halilcevic <mirza.halilcevic@amd.com>

* formatting fix

---------

Signed-off-by: Mirza Halilcevic <mirza.halilcevic@amd.com>
Co-authored-by: Dino Musić <dino.music@htecgroup.com>
Co-authored-by: Mirza Halilcevic <mirza.halilcevic@htecgroup.com>
Co-authored-by: Po Yen Chen <PoYen.Chen@amd.com>
Co-authored-by: arai713 <67439843+arai713@users.noreply.github.com>
Co-authored-by: Astha Rai <astha.rai713@gmail.com>
Co-authored-by: Mirza Halilcevic <mirza.halilcevic@amd.com>

include/ck/utility/math_v2.hpp

@@ -19,7 +19,7 @@ extern "C" __device__ float __ocml_native_recip_f32(float);
 #endif
 
 // math functions for the host,  some are implemented by calling C++ std functions
-#ifndef CK_CODE_GEN_RTC
+#if !defined(__HIPCC_RTC__) || !defined(CK_CODE_GEN_RTC)
 static inline __host__ float abs(float x) { return std::abs(x); };
 
 static inline __host__ double abs(double x) { return std::abs(x); };
@@ -924,5 +924,23 @@ inline __device__ double expm1<double>(double x)
     return expm1(x);
 };
 
+template <typename T>
+inline __device__ T cos(T x)
+{
+    return ck::type_convert<T>(cosf(ck::type_convert<float>(x)));
+};
+
+template <>
+inline __device__ float cos<float>(float x)
+{
+    return cosf(x);
+};
+
+template <>
+inline __device__ double cos<double>(double x)
+{
+    return cos(x);
+};
+
 } // namespace math
 } // namespace ck