Add README.rst for KVM selftest runner and explain how to use the
runner.

Signed-off-by: Vipin Sharma <vipinsh@google.com>
---
 tools/testing/selftests/kvm/.gitignore        |  1 +
 tools/testing/selftests/kvm/runner/README.rst | 95 +++++++++++++++++++
 2 files changed, 96 insertions(+)
 create mode 100644 tools/testing/selftests/kvm/runner/README.rst

diff --git a/tools/testing/selftests/kvm/.gitignore b/tools/testing/selftests/kvm/.gitignore
index 548d435bde2f..83aa2fe01bac 100644
--- a/tools/testing/selftests/kvm/.gitignore
+++ b/tools/testing/selftests/kvm/.gitignore
@@ -4,6 +4,7 @@
 !*.c
 !*.h
 !*.py
+!*.rst
 !*.S
 !*.sh
 !*.test
diff --git a/tools/testing/selftests/kvm/runner/README.rst b/tools/testing/selftests/kvm/runner/README.rst
new file mode 100644
index 000000000000..e264c78ace68
--- /dev/null
+++ b/tools/testing/selftests/kvm/runner/README.rst
@@ -0,0 +1,95 @@
+KVM Selftest Runner
+===================
+
+Execute KVM selftests with high configurability. The runner supports parallel
+execution, hierarchical output storage, console output control, and execution
+status reporting.
+
+Compatibility with KVM Selftests
+===============================
+
+The runner acts as a non-intrusive wrapper around the existing KVM selftests.
+It does not modify the underlying test binaries or the way they are built.
+You can still run KVM selftests using traditional methods without any conflict:
+
+- Execute binaries directly: ``./dirty_log_perf_test -v 4``
+- Use the standard kselftest framework: ``make -C tools/testing/selftests TARGETS=kvm run_tests``
+
+Comparison with kselftest framework
+===================================
+
+While the standard ``kselftest`` framework provides basic execution and reporting,
+this runner is designed for more advanced testing scenarios and offers several
+advantages:
+
+- **Flexible Test Selection**: Unlike some ``kselftest`` options that may use a
+  fixed ``suite:test`` syntax for selection, this runner leverages the
+  filesystem. You can group tests into directories (acting as suites) or
+  provide a list of specific test files, offering a more intuitive way to
+  organize and execute subsets of tests.
+- **Parallel Execution**: Run tests concurrently using the ``-j`` flag to
+  significantly reduce total execution time.
+- **Granular Output Control**: Control exactly what gets printed to the console
+  for passed, failed, skipped, or timed-out tests.
+- **Hierarchical Logging**: Automatically save the stdout and stderr of every
+  test into a structured directory tree, making it easy to debug failures in
+  large test runs.
+- **Out-of-Tree Support**: Easily run test binaries located in a different
+  directory using the ``-p`` flag.
+
+Generate Default Tests
+======================
+
+Generate the default test case files using the provided make target::
+
+  # make tests_install
+
+This creates the ``testcases_default_gen`` directory containing ``default.test``
+files. Each KVM selftest has a directory with a ``default.test`` file. This file
+contains the executable path relative to the KVM selftest root directory
+(``/tools/testing/selftests/kvm``). For example, the ``dirty_log_perf_test``
+entry looks like::
+
+  # cat testcases_default_gen/dirty_log_perf_test/default.test
+  dirty_log_perf_test
+
+In above testcase, the runner executes ``dirty_log_perf_test``. Testcase files
+can provide extra arguments to the test::
+
+  # cat tests/dirty_log_perf_test/2slot_5vcpu_10iter.test
+  dirty_log_perf_test -x 2 -v 5 -i 10
+
+In this case, the runner executes ``dirty_log_perf_test`` with the specified
+options.
+
+Examples
+========
+
+Display all options::
+
+  # python3 runner -h
+
+Run all default tests::
+
+  # python3 runner -d testcases_default_gen
+
+Run tests in parallel::
+
+  # python3 runner -d testcases_default_gen -j 40
+
+Print only passed test status and failed test stderr::
+
+  # python3 runner -d testcases_default_gen --print-passed status \
+  --print-failed stderr
+
+Run test binaries from a different directory (e.g., out-of-tree builds)::
+
+  # python3 runner -d testcases_default_gen -p /path/to/binaries
+
+Save all test outputs (stdout, stderr, status) to a directory::
+
+  # python3 runner -d testcases_default_gen -o test_outputs
+
+Run specific testcase files::
+
+  # python3 runner -t tests/dirty_log_perf_test/2slot_5vcpu_10iter.test
-- 
2.53.0.1118.gaef5881109-goog