Merge tag 'kcsan.2024.07.12a' of git://git.kernel.org/pub/scm/linux/kernel/git/paulmck/linux-rcu

Pull KCSAN updates from Paul McKenney:

 - improve the documentation for the new __data_racy type qualifier
   to the data_race() macro's kernel-doc header and to the LKMM's
   access-marking documentation

 - add missing MODULE_DESCRIPTION

* tag 'kcsan.2024.07.12a' of git://git.kernel.org/pub/scm/linux/kernel/git/paulmck/linux-rcu:
  kcsan: Add missing MODULE_DESCRIPTION() macro
  kcsan: Add example to data_race() kerneldoc header

Author: torvalds

include/linux/compiler.h

@@ -194,9 +194,17 @@ void ftrace_likely_update(struct ftrace_likely_data *f, int val,
  * This data_race() macro is useful for situations in which data races
  * should be forgiven.  One example is diagnostic code that accesses
  * shared variables but is not a part of the core synchronization design.
+ * For example, if accesses to a given variable are protected by a lock,
+ * except for diagnostic code, then the accesses under the lock should
+ * be plain C-language accesses and those in the diagnostic code should
+ * use data_race().  This way, KCSAN will complain if buggy lockless
+ * accesses to that variable are introduced, even if the buggy accesses
+ * are protected by READ_ONCE() or WRITE_ONCE().
  *
  * This macro *does not* affect normal code generation, but is a hint
- * to tooling that data races here are to be ignored.
+ * to tooling that data races here are to be ignored.  If the access must
+ * be atomic *and* KCSAN should ignore the access, use both data_race()
+ * and READ_ONCE(), for example, data_race(READ_ONCE(x)).
  */
 #define data_race(expr)							\
 ({									\

kernel/kcsan/kcsan_test.c

@@ -1620,5 +1620,6 @@ static struct kunit_suite kcsan_test_suite = {
 
 kunit_test_suites(&kcsan_test_suite);
 
+MODULE_DESCRIPTION("KCSAN test suite");
 MODULE_LICENSE("GPL v2");
 MODULE_AUTHOR("Marco Elver <[email protected]>");

tools/memory-model/Documentation/access-marking.txt

@@ -25,6 +25,11 @@ The Linux kernel provides the following access-marking options:
 4.	WRITE_ONCE(), for example, "WRITE_ONCE(a, b);"
 	The various forms of atomic_set() also fit in here.
 
+5.	__data_racy, for example "int __data_racy a;"
+
+6.	KCSAN's negative-marking assertions, ASSERT_EXCLUSIVE_ACCESS()
+	and ASSERT_EXCLUSIVE_WRITER(), are described in the
+	"ACCESS-DOCUMENTATION OPTIONS" section below.
 
 These may be used in combination, as shown in this admittedly improbable
 example:
@@ -206,6 +211,23 @@ because doing otherwise prevents KCSAN from detecting violations of your
 code's synchronization rules.
 
 
+Use of __data_racy
+------------------
+
+Adding the __data_racy type qualifier to the declaration of a variable
+causes KCSAN to treat all accesses to that variable as if they were
+enclosed by data_race().  However, __data_racy does not affect the
+compiler, though one could imagine hardened kernel builds treating the
+__data_racy type qualifier as if it was the volatile keyword.
+
+Note well that __data_racy is subject to the same pointer-declaration
+rules as are other type qualifiers such as const and volatile.
+For example:
+
+	int __data_racy *p; // Pointer to data-racy data.
+	int *__data_racy p; // Data-racy pointer to non-data-racy data.
+
+
 ACCESS-DOCUMENTATION OPTIONS
 ============================
 
@@ -343,7 +365,7 @@ as follows:
 
 Because foo is read locklessly, all accesses are marked.  The purpose
 of the ASSERT_EXCLUSIVE_WRITER() is to allow KCSAN to check for a buggy
-concurrent lockless write.
+concurrent write, whether marked or not.
 
 
 Lock-Protected Writes With Heuristic Lockless Reads