Add refs and links to RAJA Perf and a couple other updates.

Users find looking at RAJA Perf as a source of RAJA usage examples instructive.
Hopefully, these modest changes will help them navigate to that project.

README.md

@@ -94,7 +94,10 @@ The [**RAJA Performance Suite**](https://github.com/LLNL/RAJAPerf) contains
 a collection of loop kernels implemented in multiple RAJA and non-RAJA
 variants. We use it to monitor and assess RAJA performance on different
 platforms using a variety of compilers. Many major compiler vendors use the
-Suite to improve their support of abstractions like RAJA. 
+Suite to improve their support of abstractions like RAJA. **The RAJA
+Performance Suite is an excellent source of examples of RAJA usage where you
+can compare RAJA and non-RAJA variants of a variety of different kernels and
+RAJA back-ends.**
 
 The [**RAJA Proxies**](https://github.com/LLNL/RAJAProxies) repository 
 contains RAJA versions of several important HPC proxy applications.

docs/sphinx/user_guide/getting_started.rst

@@ -28,7 +28,7 @@ in :ref:`configopt-label`.
 To build RAJA and use its most basic features, you will need:
 
 - C++ compiler with C++14 support
-- `CMake <https://cmake.org/>`_ version 3.23 or greater when building the HIP back-end, and version 3.20 or greater otherwise.
+- `CMake <https://cmake.org/>`_ version 3.23 or greater.
 
 
 ==================
@@ -371,3 +371,11 @@ be located in the ``<build-dir>/test`` directory.
 For an overview of all the main RAJA features, see :ref:`features-label`.
 A full tutorial with a variety of examples showing how to use RAJA features
 can be found in :ref:`tutorial-label`.
+
+.. important:: The `RAJA Performance Suite <https://github.com/LLNL/RAJAPerf>`
+               is an excellent source of RAJA usage examples. The Suite
+               contains many numerical kernels, each of which is implemented
+               in a variety of RAJA and non-RAJA variants in OpenMP, CUDA, HIP,
+               SYCL, etc. Comparing different variants of these kernels is
+               instructive to understand how to use RAJA features and how they
+               work.

docs/sphinx/user_guide/tutorial.rst

@@ -449,3 +449,11 @@ Other RAJA Features and Usage Examples
 
    tutorial/halo-exchange.rst
    tutorial/matrix_multiply.rst
+
+.. important:: The `RAJA Performance Suite <https://github.com/LLNL/RAJAPerf>`
+               is an excellent source of RAJA usage examples. The Suite
+               contains many numerical kernels, each of which is implemented
+               in a variety of RAJA and non-RAJA variants in OpenMP, CUDA, HIP,
+               SYCL, etc. Comparing different variants of these kernels is
+               instructive to understand how to use RAJA features and how they
+               work.