arm64: atomic: always inline the assembly

The __lse_*() and __ll_sc_*() atomic implementations are marked as
inline rather than __always_inline, permitting a compiler to generate
out-of-line versions, which may be instrumented.

We marked the atomic wrappers as __always_inline in commit:

  c35a824c31834d94 ("arm64: make atomic helpers __always_inline")

... but did not think to do the same for the underlying implementations.

If the compiler were to out-of-line an LSE or LL/SC atomic, this could
break noinstr code. Ensure this doesn't happen by marking the underlying
implementations as __always_inline.

There should be no functional change as a result of this patch.

Signed-off-by: Mark Rutland <mark.rutland@arm.com>
Cc: Will Deacon <will@kernel.org>
Link: https://lore.kernel.org/r/20220817155914.3975112-3-mark.rutland@arm.com
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>

2 files changed, 38 insertions, 26 deletions

diff --git a/arch/arm64/include/asm/atomic_ll_sc.h b/arch/arm64/include/asm/atomic_ll_sc.h
index 906e2d8c254c..0890e4f568fb 100644
--- a/arch/arm64/include/asm/atomic_ll_sc.h
+++ b/arch/arm64/include/asm/atomic_ll_sc.h
@@ -23,7 +23,7 @@
  */
 
 #define ATOMIC_OP(op, asm_op, constraint)				\
-static inline void							\
+static __always_inline void						\
 __ll_sc_atomic_##op(int i, atomic_t *v)					\
 {									\
 	unsigned long tmp;						\
@@ -40,7 +40,7 @@ __ll_sc_atomic_##op(int i, atomic_t *v)					\
 }
 
 #define ATOMIC_OP_RETURN(name, mb, acq, rel, cl, op, asm_op, constraint)\
-static inline int							\
+static __always_inline int						\
 __ll_sc_atomic_##op##_return##name(int i, atomic_t *v)			\
 {									\
 	unsigned long tmp;						\
@@ -61,7 +61,7 @@ __ll_sc_atomic_##op##_return##name(int i, atomic_t *v)			\
 }
 
 #define ATOMIC_FETCH_OP(name, mb, acq, rel, cl, op, asm_op, constraint) \
-static inline int							\
+static __always_inline int						\
 __ll_sc_atomic_fetch_##op##name(int i, atomic_t *v)			\
 {									\
 	unsigned long tmp;						\
@@ -119,7 +119,7 @@ ATOMIC_OPS(andnot, bic, )
 #undef ATOMIC_OP
 
 #define ATOMIC64_OP(op, asm_op, constraint)				\
-static inline void							\
+static __always_inline void						\
 __ll_sc_atomic64_##op(s64 i, atomic64_t *v)				\
 {									\
 	s64 result;							\
@@ -136,7 +136,7 @@ __ll_sc_atomic64_##op(s64 i, atomic64_t *v)				\
 }
 
 #define ATOMIC64_OP_RETURN(name, mb, acq, rel, cl, op, asm_op, constraint)\
-static inline long							\
+static __always_inline long						\
 __ll_sc_atomic64_##op##_return##name(s64 i, atomic64_t *v)		\
 {									\
 	s64 result;							\
@@ -157,7 +157,7 @@ __ll_sc_atomic64_##op##_return##name(s64 i, atomic64_t *v)		\
 }
 
 #define ATOMIC64_FETCH_OP(name, mb, acq, rel, cl, op, asm_op, constraint)\
-static inline long							\
+static __always_inline long						\
 __ll_sc_atomic64_fetch_##op##name(s64 i, atomic64_t *v)			\
 {									\
 	s64 result, val;						\
@@ -214,7 +214,7 @@ ATOMIC64_OPS(andnot, bic, )
 #undef ATOMIC64_OP_RETURN
 #undef ATOMIC64_OP
 
-static inline s64
+static __always_inline s64
 __ll_sc_atomic64_dec_if_positive(atomic64_t *v)
 {
 	s64 result;
@@ -237,7 +237,7 @@ __ll_sc_atomic64_dec_if_positive(atomic64_t *v)
 }
 
 #define __CMPXCHG_CASE(w, sfx, name, sz, mb, acq, rel, cl, constraint)	\
-static inline u##sz							\
+static __always_inline u##sz						\
 __ll_sc__cmpxchg_case_##name##sz(volatile void *ptr,			\
 					 unsigned long old,		\
 					 u##sz new)			\
@@ -295,7 +295,7 @@ __CMPXCHG_CASE( ,  ,  mb_, 64, dmb ish,  , l, "memory", L)
 #undef __CMPXCHG_CASE
 
 #define __CMPXCHG_DBL(name, mb, rel, cl)				\
-static inline long							\
+static __always_inline long						\
 __ll_sc__cmpxchg_double##name(unsigned long old1,			\
 				      unsigned long old2,		\
 				      unsigned long new1,		\
diff --git a/arch/arm64/include/asm/atomic_lse.h b/arch/arm64/include/asm/atomic_lse.h
index 5d460f6b7675..52075e93de6c 100644
--- a/arch/arm64/include/asm/atomic_lse.h
+++ b/arch/arm64/include/asm/atomic_lse.h
@@ -11,7 +11,8 @@
 #define __ASM_ATOMIC_LSE_H
 
 #define ATOMIC_OP(op, asm_op)						\
-static inline void __lse_atomic_##op(int i, atomic_t *v)		\
+static __always_inline void						\
+__lse_atomic_##op(int i, atomic_t *v)					\
 {									\
 	asm volatile(							\
 	__LSE_PREAMBLE							\
@@ -25,7 +26,7 @@ ATOMIC_OP(or, stset)
 ATOMIC_OP(xor, steor)
 ATOMIC_OP(add, stadd)
 
-static inline void __lse_atomic_sub(int i, atomic_t *v)
+static __always_inline void __lse_atomic_sub(int i, atomic_t *v)
 {
 	__lse_atomic_add(-i, v);
 }
@@ -33,7 +34,8 @@ static inline void __lse_atomic_sub(int i, atomic_t *v)
 #undef ATOMIC_OP
 
 #define ATOMIC_FETCH_OP(name, mb, op, asm_op, cl...)			\
-static inline int __lse_atomic_fetch_##op##name(int i, atomic_t *v)	\
+static __always_inline int						\
+__lse_atomic_fetch_##op##name(int i, atomic_t *v)			\
 {									\
 	int old;							\
 									\
@@ -63,7 +65,8 @@ ATOMIC_FETCH_OPS(add, ldadd)
 #undef ATOMIC_FETCH_OPS
 
 #define ATOMIC_FETCH_OP_SUB(name)					\
-static inline int __lse_atomic_fetch_sub##name(int i, atomic_t *v)	\
+static __always_inline int						\
+__lse_atomic_fetch_sub##name(int i, atomic_t *v)			\
 {									\
 	return __lse_atomic_fetch_add##name(-i, v);			\
 }
@@ -76,12 +79,14 @@ ATOMIC_FETCH_OP_SUB(        )
 #undef ATOMIC_FETCH_OP_SUB
 
 #define ATOMIC_OP_ADD_SUB_RETURN(name)					\
-static inline int __lse_atomic_add_return##name(int i, atomic_t *v)	\
+static __always_inline int						\
+__lse_atomic_add_return##name(int i, atomic_t *v)			\
 {									\
 	return __lse_atomic_fetch_add##name(i, v) + i;			\
 }									\
 									\
-static inline int __lse_atomic_sub_return##name(int i, atomic_t *v)	\
+static __always_inline int						\
+__lse_atomic_sub_return##name(int i, atomic_t *v)			\
 {									\
 	return __lse_atomic_fetch_sub(i, v) - i;			\
 }
@@ -93,13 +98,14 @@ ATOMIC_OP_ADD_SUB_RETURN(        )
 
 #undef ATOMIC_OP_ADD_SUB_RETURN
 
-static inline void __lse_atomic_and(int i, atomic_t *v)
+static __always_inline void __lse_atomic_and(int i, atomic_t *v)
 {
 	return __lse_atomic_andnot(~i, v);
 }
 
 #define ATOMIC_FETCH_OP_AND(name, mb, cl...)				\
-static inline int __lse_atomic_fetch_and##name(int i, atomic_t *v)	\
+static __always_inline int						\
+__lse_atomic_fetch_and##name(int i, atomic_t *v)			\
 {									\
 	return __lse_atomic_fetch_andnot##name(~i, v);			\
 }
@@ -112,7 +118,8 @@ ATOMIC_FETCH_OP_AND(        , al, "memory")
 #undef ATOMIC_FETCH_OP_AND
 
 #define ATOMIC64_OP(op, asm_op)						\
-static inline void __lse_atomic64_##op(s64 i, atomic64_t *v)		\
+static __always_inline void						\
+__lse_atomic64_##op(s64 i, atomic64_t *v)				\
 {									\
 	asm volatile(							\
 	__LSE_PREAMBLE							\
@@ -126,7 +133,7 @@ ATOMIC64_OP(or, stset)
 ATOMIC64_OP(xor, steor)
 ATOMIC64_OP(add, stadd)
 
-static inline void __lse_atomic64_sub(s64 i, atomic64_t *v)
+static __always_inline void __lse_atomic64_sub(s64 i, atomic64_t *v)
 {
 	__lse_atomic64_add(-i, v);
 }
@@ -134,7 +141,8 @@ static inline void __lse_atomic64_sub(s64 i, atomic64_t *v)
 #undef ATOMIC64_OP
 
 #define ATOMIC64_FETCH_OP(name, mb, op, asm_op, cl...)			\
-static inline long __lse_atomic64_fetch_##op##name(s64 i, atomic64_t *v)\
+static __always_inline long						\
+__lse_atomic64_fetch_##op##name(s64 i, atomic64_t *v)			\
 {									\
 	s64 old;							\
 									\
@@ -164,7 +172,8 @@ ATOMIC64_FETCH_OPS(add, ldadd)
 #undef ATOMIC64_FETCH_OPS
 
 #define ATOMIC64_FETCH_OP_SUB(name)					\
-static inline long __lse_atomic64_fetch_sub##name(s64 i, atomic64_t *v)	\
+static __always_inline long						\
+__lse_atomic64_fetch_sub##name(s64 i, atomic64_t *v)			\
 {									\
 	return __lse_atomic64_fetch_add##name(-i, v);			\
 }
@@ -177,12 +186,14 @@ ATOMIC64_FETCH_OP_SUB(        )
 #undef ATOMIC64_FETCH_OP_SUB
 
 #define ATOMIC64_OP_ADD_SUB_RETURN(name)				\
-static inline long __lse_atomic64_add_return##name(s64 i, atomic64_t *v)\
+static __always_inline long						\
+__lse_atomic64_add_return##name(s64 i, atomic64_t *v)			\
 {									\
 	return __lse_atomic64_fetch_add##name(i, v) + i;		\
 }									\
 									\
-static inline long __lse_atomic64_sub_return##name(s64 i, atomic64_t *v)\
+static __always_inline long						\
+__lse_atomic64_sub_return##name(s64 i, atomic64_t *v)			\
 {									\
 	return __lse_atomic64_fetch_sub##name(i, v) - i;		\
 }
@@ -194,13 +205,14 @@ ATOMIC64_OP_ADD_SUB_RETURN(        )
 
 #undef ATOMIC64_OP_ADD_SUB_RETURN
 
-static inline void __lse_atomic64_and(s64 i, atomic64_t *v)
+static __always_inline void __lse_atomic64_and(s64 i, atomic64_t *v)
 {
 	return __lse_atomic64_andnot(~i, v);
 }
 
 #define ATOMIC64_FETCH_OP_AND(name, mb, cl...)				\
-static inline long __lse_atomic64_fetch_and##name(s64 i, atomic64_t *v)	\
+static __always_inline long						\
+__lse_atomic64_fetch_and##name(s64 i, atomic64_t *v)			\
 {									\
 	return __lse_atomic64_fetch_andnot##name(~i, v);		\
 }
@@ -212,7 +224,7 @@ ATOMIC64_FETCH_OP_AND(        , al, "memory")
 
 #undef ATOMIC64_FETCH_OP_AND
 
-static inline s64 __lse_atomic64_dec_if_positive(atomic64_t *v)
+static __always_inline s64 __lse_atomic64_dec_if_positive(atomic64_t *v)
 {
 	unsigned long tmp;