The Linux Kernel Mailing List
 help / color / mirror / Atom feed
From: Marco Elver <elver@google.com>
To: elver@google.com, "Paul E. McKenney" <paulmck@kernel.org>
Cc: kasan-dev@googlegroups.com, linux-kernel@vger.kernel.org,
	kernel test robot <lkp@intel.com>
Subject: [PATCH -rcu] kcsan: Turn barrier instrumentation into macros
Date: Sat,  4 Dec 2021 13:57:03 +0100	[thread overview]
Message-ID: <20211204125703.3344454-1-elver@google.com> (raw)

Some architectures use barriers in 'extern inline' functions, from which
we should not refer to static inline functions.

For example, building Alpha with gcc and W=1 shows:

./include/asm-generic/barrier.h:70:30: warning: 'kcsan_rmb' is static but used in inline function 'pmd_offset' which is not static
   70 | #define smp_rmb()       do { kcsan_rmb(); __smp_rmb(); } while (0)
      |                              ^~~~~~~~~
./arch/alpha/include/asm/pgtable.h:293:9: note: in expansion of macro 'smp_rmb'
  293 |         smp_rmb(); /* see above */
      |         ^~~~~~~

Which seems to warn about 6.7.4#3 of the C standard:
  "An inline definition of a function with external linkage shall not
   contain a definition of a modifiable object with static or thread
   storage duration, and shall not contain a reference to an identifier
   with internal linkage."

Fix it by turning barrier instrumentation into macros, which matches
definitions in <asm/barrier.h>.

Perhaps we can revert this change in future, when there are no more
'extern inline' users left.

Link: https://lkml.kernel.org/r/202112041334.X44uWZXf-lkp@intel.com
Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: Marco Elver <elver@google.com>
---
 include/linux/kcsan-checks.h | 24 +++++++++++++-----------
 1 file changed, 13 insertions(+), 11 deletions(-)

diff --git a/include/linux/kcsan-checks.h b/include/linux/kcsan-checks.h
index 9d2c869167f2..92f3843d9ebb 100644
--- a/include/linux/kcsan-checks.h
+++ b/include/linux/kcsan-checks.h
@@ -241,28 +241,30 @@ static inline void __kcsan_disable_current(void) { }
  * disabled with the __no_kcsan function attribute.
  *
  * Also see definition of __tsan_atomic_signal_fence() in kernel/kcsan/core.c.
+ *
+ * These are all macros, like <asm/barrier.h>, since some architectures use them
+ * in non-static inline functions.
  */
 #define __KCSAN_BARRIER_TO_SIGNAL_FENCE(name)					\
-	static __always_inline void kcsan_##name(void)				\
-	{									\
+	do {									\
 		barrier();							\
 		__atomic_signal_fence(__KCSAN_BARRIER_TO_SIGNAL_FENCE_##name);	\
 		barrier();							\
-	}
-__KCSAN_BARRIER_TO_SIGNAL_FENCE(mb)
-__KCSAN_BARRIER_TO_SIGNAL_FENCE(wmb)
-__KCSAN_BARRIER_TO_SIGNAL_FENCE(rmb)
-__KCSAN_BARRIER_TO_SIGNAL_FENCE(release)
+	} while (0)
+#define kcsan_mb()	__KCSAN_BARRIER_TO_SIGNAL_FENCE(mb)
+#define kcsan_wmb()	__KCSAN_BARRIER_TO_SIGNAL_FENCE(wmb)
+#define kcsan_rmb()	__KCSAN_BARRIER_TO_SIGNAL_FENCE(rmb)
+#define kcsan_release()	__KCSAN_BARRIER_TO_SIGNAL_FENCE(release)
 #elif defined(CONFIG_KCSAN_WEAK_MEMORY) && defined(__KCSAN_INSTRUMENT_BARRIERS__)
 #define kcsan_mb	__kcsan_mb
 #define kcsan_wmb	__kcsan_wmb
 #define kcsan_rmb	__kcsan_rmb
 #define kcsan_release	__kcsan_release
 #else /* CONFIG_KCSAN_WEAK_MEMORY && ... */
-static inline void kcsan_mb(void)		{ }
-static inline void kcsan_wmb(void)		{ }
-static inline void kcsan_rmb(void)		{ }
-static inline void kcsan_release(void)		{ }
+#define kcsan_mb()	do { } while (0)
+#define kcsan_wmb()	do { } while (0)
+#define kcsan_rmb()	do { } while (0)
+#define kcsan_release()	do { } while (0)
 #endif /* CONFIG_KCSAN_WEAK_MEMORY && ... */
 
 /**
-- 
2.34.1.400.ga245620fadb-goog


             reply	other threads:[~2021-12-04 12:59 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-12-04 12:57 Marco Elver [this message]
2021-12-04 15:12 ` [PATCH -rcu] kcsan: Turn barrier instrumentation into macros Paul E. McKenney

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20211204125703.3344454-1-elver@google.com \
    --to=elver@google.com \
    --cc=kasan-dev@googlegroups.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=lkp@intel.com \
    --cc=paulmck@kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox