cpumask: Switch from inline to __always_inline

On recent (v6.6+) builds with Clang (based on Clang 18.0.0) and certain
configurations [0], I'm finding that (lack of) inlining decisions may
lead to section mismatch warnings like the following:

  WARNING: modpost: vmlinux.o: section mismatch in reference:
  cpumask_andnot (section: .text) ->
  cpuhp_bringup_cpus_parallel.tmp_mask (section: .init.data) ERROR:
  modpost: Section mismatches detected.

or more confusingly:

  WARNING: modpost: vmlinux: section mismatch in reference:
  cpumask_andnot+0x5f (section: .text) -> efi_systab_phys (section:
  .init.data)

The first warning makes a little sense, because
cpuhp_bringup_cpus_parallel() (an __init function) calls
cpumask_andnot() on tmp_mask (an __initdata symbol). If the compiler
doesn't inline cpumask_andnot(), this may appear like a mismatch.

The second warning makes less sense, but might be because efi_systab_phys
and cpuhp_bringup_cpus_parallel.tmp_mask are laid out near each other,
and the latter isn't a proper C symbol definition.

In any case, it seems a reasonable solution to suggest more strongly to
the compiler that these cpumask macros *must* be inlined, as 'inline' is
just a recommendation.

This change has been previously proposed in the past as:

  Subject: [PATCH 1/3] bitmap: switch from inline to __always_inline
  https://lore.kernel.org/all/20221027043810.350460-2-yury.norov@gmail.com/

But the change has been split up, to separately justify the cpumask
changes (which drive my work) and the bitmap/const optimizations (that
Yury separately proposed for other reasons). This ends up as somewhere
between a "rebase" and "rewrite" -- I had to rewrite most of the patch.

According to bloat-o-meter, vmlinux decreases minimally in size (-0.00%
to -0.01%, depending on the version of GCC or Clang and .config in
question) with this series of changes:

gcc 13.2.0, x86_64_defconfig
-3005 bytes, Before=21944501, After=21941496, chg -0.01%

clang 16.0.6, x86_64_defconfig
-105 bytes, Before=22571692, After=22571587, chg -0.00%

gcc 9.5.0, x86_64_defconfig
-1771 bytes, Before=21557598, After=21555827, chg -0.01%

clang 18.0_pre516547 (ChromiumOS toolchain), x86_64_defconfig
-191 bytes, Before=22615339, After=22615148, chg -0.00%

clang 18.0_pre516547 (ChromiumOS toolchain), based on ChromiumOS config + gcov
-979 bytes, Before=76294783, After=76293804, chg -0.00%

[0] CONFIG_HOTPLUG_PARALLEL=y ('select'ed for x86 as of [1]) and
    CONFIG_GCOV_PROFILE_ALL.

[1] commit 0c7ffa32dbd6 ("x86/smpboot/64: Implement
    arch_cpuhp_init_parallel_bringup() and enable it")

Co-developed-by: Brian Norris <briannorris@chromium.org>
Signed-off-by: Brian Norris <briannorris@chromium.org>
Reviewed-by: Kees Cook <kees@kernel.org>
Reviewed-by: Nathan Chancellor <nathan@kernel.org>
Signed-off-by: Yury Norov <yury.norov@gmail.com>

1 files changed, 112 insertions, 100 deletions

diff --git a/include/linux/cpumask.h b/include/linux/cpumask.h
index 801a7e524113..9e3f3c96607d 100644
--- a/include/linux/cpumask.h
+++ b/include/linux/cpumask.h
@@ -30,7 +30,7 @@
 extern unsigned int nr_cpu_ids;
 #endif
 
-static inline void set_nr_cpu_ids(unsigned int nr)
+static __always_inline void set_nr_cpu_ids(unsigned int nr)
 {
 #if (NR_CPUS == 1) || defined(CONFIG_FORCE_NR_CPUS)
 	WARN_ON(nr != nr_cpu_ids);
@@ -149,7 +149,7 @@ static __always_inline unsigned int cpumask_check(unsigned int cpu)
  *
  * Return: >= nr_cpu_ids if no cpus set.
  */
-static inline unsigned int cpumask_first(const struct cpumask *srcp)
+static __always_inline unsigned int cpumask_first(const struct cpumask *srcp)
 {
 	return find_first_bit(cpumask_bits(srcp), small_cpumask_bits);
 }
@@ -160,7 +160,7 @@ static inline unsigned int cpumask_first(const struct cpumask *srcp)
  *
  * Return: >= nr_cpu_ids if all cpus are set.
  */
-static inline unsigned int cpumask_first_zero(const struct cpumask *srcp)
+static __always_inline unsigned int cpumask_first_zero(const struct cpumask *srcp)
 {
 	return find_first_zero_bit(cpumask_bits(srcp), small_cpumask_bits);
 }
@@ -172,7 +172,7 @@ static inline unsigned int cpumask_first_zero(const struct cpumask *srcp)
  *
  * Return: >= nr_cpu_ids if no cpus set in both.  See also cpumask_next_and().
  */
-static inline
+static __always_inline
 unsigned int cpumask_first_and(const struct cpumask *srcp1, const struct cpumask *srcp2)
 {
 	return find_first_and_bit(cpumask_bits(srcp1), cpumask_bits(srcp2), small_cpumask_bits);
@@ -186,7 +186,7 @@ unsigned int cpumask_first_and(const struct cpumask *srcp1, const struct cpumask
  *
  * Return: >= nr_cpu_ids if no cpus set in all.
  */
-static inline
+static __always_inline
 unsigned int cpumask_first_and_and(const struct cpumask *srcp1,
 				   const struct cpumask *srcp2,
 				   const struct cpumask *srcp3)
@@ -201,7 +201,7 @@ unsigned int cpumask_first_and_and(const struct cpumask *srcp1,
  *
  * Return:	>= nr_cpumask_bits if no CPUs set.
  */
-static inline unsigned int cpumask_last(const struct cpumask *srcp)
+static __always_inline unsigned int cpumask_last(const struct cpumask *srcp)
 {
 	return find_last_bit(cpumask_bits(srcp), small_cpumask_bits);
 }
@@ -213,7 +213,7 @@ static inline unsigned int cpumask_last(const struct cpumask *srcp)
  *
  * Return: >= nr_cpu_ids if no further cpus set.
  */
-static inline
+static __always_inline
 unsigned int cpumask_next(int n, const struct cpumask *srcp)
 {
 	/* -1 is a legal arg here. */
@@ -229,7 +229,8 @@ unsigned int cpumask_next(int n, const struct cpumask *srcp)
  *
  * Return: >= nr_cpu_ids if no further cpus unset.
  */
-static inline unsigned int cpumask_next_zero(int n, const struct cpumask *srcp)
+static __always_inline
+unsigned int cpumask_next_zero(int n, const struct cpumask *srcp)
 {
 	/* -1 is a legal arg here. */
 	if (n != -1)
@@ -239,18 +240,21 @@ static inline unsigned int cpumask_next_zero(int n, const struct cpumask *srcp)
 
 #if NR_CPUS == 1
 /* Uniprocessor: there is only one valid CPU */
-static inline unsigned int cpumask_local_spread(unsigned int i, int node)
+static __always_inline
+unsigned int cpumask_local_spread(unsigned int i, int node)
 {
 	return 0;
 }
 
-static inline unsigned int cpumask_any_and_distribute(const struct cpumask *src1p,
-						      const struct cpumask *src2p)
+static __always_inline
+unsigned int cpumask_any_and_distribute(const struct cpumask *src1p,
+					const struct cpumask *src2p)
 {
 	return cpumask_first_and(src1p, src2p);
 }
 
-static inline unsigned int cpumask_any_distribute(const struct cpumask *srcp)
+static __always_inline
+unsigned int cpumask_any_distribute(const struct cpumask *srcp)
 {
 	return cpumask_first(srcp);
 }
@@ -269,9 +273,9 @@ unsigned int cpumask_any_distribute(const struct cpumask *srcp);
  *
  * Return: >= nr_cpu_ids if no further cpus set in both.
  */
-static inline
+static __always_inline
 unsigned int cpumask_next_and(int n, const struct cpumask *src1p,
-		     const struct cpumask *src2p)
+			      const struct cpumask *src2p)
 {
 	/* -1 is a legal arg here. */
 	if (n != -1)
@@ -291,7 +295,7 @@ unsigned int cpumask_next_and(int n, const struct cpumask *src1p,
 	for_each_set_bit(cpu, cpumask_bits(mask), small_cpumask_bits)
 
 #if NR_CPUS == 1
-static inline
+static __always_inline
 unsigned int cpumask_next_wrap(int n, const struct cpumask *mask, int start, bool wrap)
 {
 	cpumask_check(start);
@@ -394,7 +398,7 @@ unsigned int __pure cpumask_next_wrap(int n, const struct cpumask *mask, int sta
  * Often used to find any cpu but smp_processor_id() in a mask.
  * Return: >= nr_cpu_ids if no cpus set.
  */
-static inline
+static __always_inline
 unsigned int cpumask_any_but(const struct cpumask *mask, unsigned int cpu)
 {
 	unsigned int i;
@@ -414,7 +418,7 @@ unsigned int cpumask_any_but(const struct cpumask *mask, unsigned int cpu)
  *
  * Returns >= nr_cpu_ids if no cpus set.
  */
-static inline
+static __always_inline
 unsigned int cpumask_any_and_but(const struct cpumask *mask1,
 				 const struct cpumask *mask2,
 				 unsigned int cpu)
@@ -436,7 +440,8 @@ unsigned int cpumask_any_and_but(const struct cpumask *mask1,
  *
  * Return: >= nr_cpu_ids if such cpu doesn't exist.
  */
-static inline unsigned int cpumask_nth(unsigned int cpu, const struct cpumask *srcp)
+static __always_inline
+unsigned int cpumask_nth(unsigned int cpu, const struct cpumask *srcp)
 {
 	return find_nth_bit(cpumask_bits(srcp), small_cpumask_bits, cpumask_check(cpu));
 }
@@ -449,7 +454,7 @@ static inline unsigned int cpumask_nth(unsigned int cpu, const struct cpumask *s
  *
  * Return: >= nr_cpu_ids if such cpu doesn't exist.
  */
-static inline
+static __always_inline
 unsigned int cpumask_nth_and(unsigned int cpu, const struct cpumask *srcp1,
 							const struct cpumask *srcp2)
 {
@@ -465,7 +470,7 @@ unsigned int cpumask_nth_and(unsigned int cpu, const struct cpumask *srcp1,
  *
  * Return: >= nr_cpu_ids if such cpu doesn't exist.
  */
-static inline
+static __always_inline
 unsigned int cpumask_nth_andnot(unsigned int cpu, const struct cpumask *srcp1,
 							const struct cpumask *srcp2)
 {
@@ -508,12 +513,14 @@ unsigned int cpumask_nth_and_andnot(unsigned int cpu, const struct cpumask *srcp
  * @cpu: cpu number (< nr_cpu_ids)
  * @dstp: the cpumask pointer
  */
-static __always_inline void cpumask_set_cpu(unsigned int cpu, struct cpumask *dstp)
+static __always_inline
+void cpumask_set_cpu(unsigned int cpu, struct cpumask *dstp)
 {
 	set_bit(cpumask_check(cpu), cpumask_bits(dstp));
 }
 
-static __always_inline void __cpumask_set_cpu(unsigned int cpu, struct cpumask *dstp)
+static __always_inline
+void __cpumask_set_cpu(unsigned int cpu, struct cpumask *dstp)
 {
 	__set_bit(cpumask_check(cpu), cpumask_bits(dstp));
 }
@@ -557,7 +564,8 @@ static __always_inline void __cpumask_assign_cpu(int cpu, struct cpumask *dstp,
  *
  * Return: true if @cpu is set in @cpumask, else returns false
  */
-static __always_inline bool cpumask_test_cpu(int cpu, const struct cpumask *cpumask)
+static __always_inline
+bool cpumask_test_cpu(int cpu, const struct cpumask *cpumask)
 {
 	return test_bit(cpumask_check(cpu), cpumask_bits((cpumask)));
 }
@@ -571,7 +579,8 @@ static __always_inline bool cpumask_test_cpu(int cpu, const struct cpumask *cpum
  *
  * Return: true if @cpu is set in old bitmap of @cpumask, else returns false
  */
-static __always_inline bool cpumask_test_and_set_cpu(int cpu, struct cpumask *cpumask)
+static __always_inline
+bool cpumask_test_and_set_cpu(int cpu, struct cpumask *cpumask)
 {
 	return test_and_set_bit(cpumask_check(cpu), cpumask_bits(cpumask));
 }
@@ -585,7 +594,8 @@ static __always_inline bool cpumask_test_and_set_cpu(int cpu, struct cpumask *cp
  *
  * Return: true if @cpu is set in old bitmap of @cpumask, else returns false
  */
-static __always_inline bool cpumask_test_and_clear_cpu(int cpu, struct cpumask *cpumask)
+static __always_inline
+bool cpumask_test_and_clear_cpu(int cpu, struct cpumask *cpumask)
 {
 	return test_and_clear_bit(cpumask_check(cpu), cpumask_bits(cpumask));
 }
@@ -594,7 +604,7 @@ static __always_inline bool cpumask_test_and_clear_cpu(int cpu, struct cpumask *
  * cpumask_setall - set all cpus (< nr_cpu_ids) in a cpumask
  * @dstp: the cpumask pointer
  */
-static inline void cpumask_setall(struct cpumask *dstp)
+static __always_inline void cpumask_setall(struct cpumask *dstp)
 {
 	if (small_const_nbits(small_cpumask_bits)) {
 		cpumask_bits(dstp)[0] = BITMAP_LAST_WORD_MASK(nr_cpumask_bits);
@@ -607,7 +617,7 @@ static inline void cpumask_setall(struct cpumask *dstp)
  * cpumask_clear - clear all cpus (< nr_cpu_ids) in a cpumask
  * @dstp: the cpumask pointer
  */
-static inline void cpumask_clear(struct cpumask *dstp)
+static __always_inline void cpumask_clear(struct cpumask *dstp)
 {
 	bitmap_zero(cpumask_bits(dstp), large_cpumask_bits);
 }
@@ -620,9 +630,9 @@ static inline void cpumask_clear(struct cpumask *dstp)
  *
  * Return: false if *@dstp is empty, else returns true
  */
-static inline bool cpumask_and(struct cpumask *dstp,
-			       const struct cpumask *src1p,
-			       const struct cpumask *src2p)
+static __always_inline
+bool cpumask_and(struct cpumask *dstp, const struct cpumask *src1p,
+		 const struct cpumask *src2p)
 {
 	return bitmap_and(cpumask_bits(dstp), cpumask_bits(src1p),
 				       cpumask_bits(src2p), small_cpumask_bits);
@@ -634,8 +644,9 @@ static inline bool cpumask_and(struct cpumask *dstp,
  * @src1p: the first input
  * @src2p: the second input
  */
-static inline void cpumask_or(struct cpumask *dstp, const struct cpumask *src1p,
-			      const struct cpumask *src2p)
+static __always_inline
+void cpumask_or(struct cpumask *dstp, const struct cpumask *src1p,
+		const struct cpumask *src2p)
 {
 	bitmap_or(cpumask_bits(dstp), cpumask_bits(src1p),
 				      cpumask_bits(src2p), small_cpumask_bits);
@@ -647,9 +658,9 @@ static inline void cpumask_or(struct cpumask *dstp, const struct cpumask *src1p,
  * @src1p: the first input
  * @src2p: the second input
  */
-static inline void cpumask_xor(struct cpumask *dstp,
-			       const struct cpumask *src1p,
-			       const struct cpumask *src2p)
+static __always_inline
+void cpumask_xor(struct cpumask *dstp, const struct cpumask *src1p,
+		 const struct cpumask *src2p)
 {
 	bitmap_xor(cpumask_bits(dstp), cpumask_bits(src1p),
 				       cpumask_bits(src2p), small_cpumask_bits);
@@ -663,9 +674,9 @@ static inline void cpumask_xor(struct cpumask *dstp,
  *
  * Return: false if *@dstp is empty, else returns true
  */
-static inline bool cpumask_andnot(struct cpumask *dstp,
-				  const struct cpumask *src1p,
-				  const struct cpumask *src2p)
+static __always_inline
+bool cpumask_andnot(struct cpumask *dstp, const struct cpumask *src1p,
+		    const struct cpumask *src2p)
 {
 	return bitmap_andnot(cpumask_bits(dstp), cpumask_bits(src1p),
 					  cpumask_bits(src2p), small_cpumask_bits);
@@ -678,8 +689,8 @@ static inline bool cpumask_andnot(struct cpumask *dstp,
  *
  * Return: true if the cpumasks are equal, false if not
  */
-static inline bool cpumask_equal(const struct cpumask *src1p,
-				const struct cpumask *src2p)
+static __always_inline
+bool cpumask_equal(const struct cpumask *src1p, const struct cpumask *src2p)
 {
 	return bitmap_equal(cpumask_bits(src1p), cpumask_bits(src2p),
 						 small_cpumask_bits);
@@ -694,9 +705,9 @@ static inline bool cpumask_equal(const struct cpumask *src1p,
  * Return: true if first cpumask ORed with second cpumask == third cpumask,
  *	   otherwise false
  */
-static inline bool cpumask_or_equal(const struct cpumask *src1p,
-				    const struct cpumask *src2p,
-				    const struct cpumask *src3p)
+static __always_inline
+bool cpumask_or_equal(const struct cpumask *src1p, const struct cpumask *src2p,
+		      const struct cpumask *src3p)
 {
 	return bitmap_or_equal(cpumask_bits(src1p), cpumask_bits(src2p),
 			       cpumask_bits(src3p), small_cpumask_bits);
@@ -710,8 +721,8 @@ static inline bool cpumask_or_equal(const struct cpumask *src1p,
  * Return: true if first cpumask ANDed with second cpumask is non-empty,
  *	   otherwise false
  */
-static inline bool cpumask_intersects(const struct cpumask *src1p,
-				     const struct cpumask *src2p)
+static __always_inline
+bool cpumask_intersects(const struct cpumask *src1p, const struct cpumask *src2p)
 {
 	return bitmap_intersects(cpumask_bits(src1p), cpumask_bits(src2p),
 						      small_cpumask_bits);
@@ -724,8 +735,8 @@ static inline bool cpumask_intersects(const struct cpumask *src1p,
  *
  * Return: true if *@src1p is a subset of *@src2p, else returns false
  */
-static inline bool cpumask_subset(const struct cpumask *src1p,
-				 const struct cpumask *src2p)
+static __always_inline
+bool cpumask_subset(const struct cpumask *src1p, const struct cpumask *src2p)
 {
 	return bitmap_subset(cpumask_bits(src1p), cpumask_bits(src2p),
 						  small_cpumask_bits);
@@ -737,7 +748,7 @@ static inline bool cpumask_subset(const struct cpumask *src1p,
  *
  * Return: true if srcp is empty (has no bits set), else false
  */
-static inline bool cpumask_empty(const struct cpumask *srcp)
+static __always_inline bool cpumask_empty(const struct cpumask *srcp)
 {
 	return bitmap_empty(cpumask_bits(srcp), small_cpumask_bits);
 }
@@ -748,7 +759,7 @@ static inline bool cpumask_empty(const struct cpumask *srcp)
  *
  * Return: true if srcp is full (has all bits set), else false
  */
-static inline bool cpumask_full(const struct cpumask *srcp)
+static __always_inline bool cpumask_full(const struct cpumask *srcp)
 {
 	return bitmap_full(cpumask_bits(srcp), nr_cpumask_bits);
 }
@@ -759,7 +770,7 @@ static inline bool cpumask_full(const struct cpumask *srcp)
  *
  * Return: count of bits set in *srcp
  */
-static inline unsigned int cpumask_weight(const struct cpumask *srcp)
+static __always_inline unsigned int cpumask_weight(const struct cpumask *srcp)
 {
 	return bitmap_weight(cpumask_bits(srcp), small_cpumask_bits);
 }
@@ -771,8 +782,8 @@ static inline unsigned int cpumask_weight(const struct cpumask *srcp)
  *
  * Return: count of bits set in both *srcp1 and *srcp2
  */
-static inline unsigned int cpumask_weight_and(const struct cpumask *srcp1,
-						const struct cpumask *srcp2)
+static __always_inline
+unsigned int cpumask_weight_and(const struct cpumask *srcp1, const struct cpumask *srcp2)
 {
 	return bitmap_weight_and(cpumask_bits(srcp1), cpumask_bits(srcp2), small_cpumask_bits);
 }
@@ -784,8 +795,9 @@ static inline unsigned int cpumask_weight_and(const struct cpumask *srcp1,
  *
  * Return: count of bits set in both *srcp1 and *srcp2
  */
-static inline unsigned int cpumask_weight_andnot(const struct cpumask *srcp1,
-						const struct cpumask *srcp2)
+static __always_inline
+unsigned int cpumask_weight_andnot(const struct cpumask *srcp1,
+				   const struct cpumask *srcp2)
 {
 	return bitmap_weight_andnot(cpumask_bits(srcp1), cpumask_bits(srcp2), small_cpumask_bits);
 }
@@ -796,8 +808,8 @@ static inline unsigned int cpumask_weight_andnot(const struct cpumask *srcp1,
  * @srcp: the input to shift
  * @n: the number of bits to shift by
  */
-static inline void cpumask_shift_right(struct cpumask *dstp,
-				       const struct cpumask *srcp, int n)
+static __always_inline
+void cpumask_shift_right(struct cpumask *dstp, const struct cpumask *srcp, int n)
 {
 	bitmap_shift_right(cpumask_bits(dstp), cpumask_bits(srcp), n,
 					       small_cpumask_bits);
@@ -809,8 +821,8 @@ static inline void cpumask_shift_right(struct cpumask *dstp,
  * @srcp: the input to shift
  * @n: the number of bits to shift by
  */
-static inline void cpumask_shift_left(struct cpumask *dstp,
-				      const struct cpumask *srcp, int n)
+static __always_inline
+void cpumask_shift_left(struct cpumask *dstp, const struct cpumask *srcp, int n)
 {
 	bitmap_shift_left(cpumask_bits(dstp), cpumask_bits(srcp), n,
 					      nr_cpumask_bits);
@@ -821,8 +833,8 @@ static inline void cpumask_shift_left(struct cpumask *dstp,
  * @dstp: the result
  * @srcp: the input cpumask
  */
-static inline void cpumask_copy(struct cpumask *dstp,
-				const struct cpumask *srcp)
+static __always_inline
+void cpumask_copy(struct cpumask *dstp, const struct cpumask *srcp)
 {
 	bitmap_copy(cpumask_bits(dstp), cpumask_bits(srcp), large_cpumask_bits);
 }
@@ -858,8 +870,8 @@ static inline void cpumask_copy(struct cpumask *dstp,
  *
  * Return: -errno, or 0 for success.
  */
-static inline int cpumask_parse_user(const char __user *buf, int len,
-				     struct cpumask *dstp)
+static __always_inline
+int cpumask_parse_user(const char __user *buf, int len, struct cpumask *dstp)
 {
 	return bitmap_parse_user(buf, len, cpumask_bits(dstp), nr_cpumask_bits);
 }
@@ -872,8 +884,8 @@ static inline int cpumask_parse_user(const char __user *buf, int len,
  *
  * Return: -errno, or 0 for success.
  */
-static inline int cpumask_parselist_user(const char __user *buf, int len,
-				     struct cpumask *dstp)
+static __always_inline
+int cpumask_parselist_user(const char __user *buf, int len, struct cpumask *dstp)
 {
 	return bitmap_parselist_user(buf, len, cpumask_bits(dstp),
 				     nr_cpumask_bits);
@@ -886,7 +898,7 @@ static inline int cpumask_parselist_user(const char __user *buf, int len,
  *
  * Return: -errno, or 0 for success.
  */
-static inline int cpumask_parse(const char *buf, struct cpumask *dstp)
+static __always_inline int cpumask_parse(const char *buf, struct cpumask *dstp)
 {
 	return bitmap_parse(buf, UINT_MAX, cpumask_bits(dstp), nr_cpumask_bits);
 }
@@ -898,7 +910,7 @@ static inline int cpumask_parse(const char *buf, struct cpumask *dstp)
  *
  * Return: -errno, or 0 for success.
  */
-static inline int cpulist_parse(const char *buf, struct cpumask *dstp)
+static __always_inline int cpulist_parse(const char *buf, struct cpumask *dstp)
 {
 	return bitmap_parselist(buf, cpumask_bits(dstp), nr_cpumask_bits);
 }
@@ -908,7 +920,7 @@ static inline int cpulist_parse(const char *buf, struct cpumask *dstp)
  *
  * Return: size to allocate for a &struct cpumask in bytes
  */
-static inline unsigned int cpumask_size(void)
+static __always_inline unsigned int cpumask_size(void)
 {
 	return bitmap_size(large_cpumask_bits);
 }
@@ -920,7 +932,7 @@ static inline unsigned int cpumask_size(void)
 
 bool alloc_cpumask_var_node(cpumask_var_t *mask, gfp_t flags, int node);
 
-static inline
+static __always_inline
 bool zalloc_cpumask_var_node(cpumask_var_t *mask, gfp_t flags, int node)
 {
 	return alloc_cpumask_var_node(mask, flags | __GFP_ZERO, node);
@@ -938,13 +950,13 @@ bool zalloc_cpumask_var_node(cpumask_var_t *mask, gfp_t flags, int node)
  *
  * Return: %true if allocation succeeded, %false if not
  */
-static inline
+static __always_inline
 bool alloc_cpumask_var(cpumask_var_t *mask, gfp_t flags)
 {
 	return alloc_cpumask_var_node(mask, flags, NUMA_NO_NODE);
 }
 
-static inline
+static __always_inline
 bool zalloc_cpumask_var(cpumask_var_t *mask, gfp_t flags)
 {
 	return alloc_cpumask_var(mask, flags | __GFP_ZERO);
@@ -954,7 +966,7 @@ void alloc_bootmem_cpumask_var(cpumask_var_t *mask);
 void free_cpumask_var(cpumask_var_t mask);
 void free_bootmem_cpumask_var(cpumask_var_t mask);
 
-static inline bool cpumask_available(cpumask_var_t mask)
+static __always_inline bool cpumask_available(cpumask_var_t mask)
 {
 	return mask != NULL;
 }
@@ -964,43 +976,43 @@ static inline bool cpumask_available(cpumask_var_t mask)
 #define this_cpu_cpumask_var_ptr(x) this_cpu_ptr(x)
 #define __cpumask_var_read_mostly
 
-static inline bool alloc_cpumask_var(cpumask_var_t *mask, gfp_t flags)
+static __always_inline bool alloc_cpumask_var(cpumask_var_t *mask, gfp_t flags)
 {
 	return true;
 }
 
-static inline bool alloc_cpumask_var_node(cpumask_var_t *mask, gfp_t flags,
+static __always_inline bool alloc_cpumask_var_node(cpumask_var_t *mask, gfp_t flags,
 					  int node)
 {
 	return true;
 }
 
-static inline bool zalloc_cpumask_var(cpumask_var_t *mask, gfp_t flags)
+static __always_inline bool zalloc_cpumask_var(cpumask_var_t *mask, gfp_t flags)
 {
 	cpumask_clear(*mask);
 	return true;
 }
 
-static inline bool zalloc_cpumask_var_node(cpumask_var_t *mask, gfp_t flags,
+static __always_inline bool zalloc_cpumask_var_node(cpumask_var_t *mask, gfp_t flags,
 					  int node)
 {
 	cpumask_clear(*mask);
 	return true;
 }
 
-static inline void alloc_bootmem_cpumask_var(cpumask_var_t *mask)
+static __always_inline void alloc_bootmem_cpumask_var(cpumask_var_t *mask)
 {
 }
 
-static inline void free_cpumask_var(cpumask_var_t mask)
+static __always_inline void free_cpumask_var(cpumask_var_t mask)
 {
 }
 
-static inline void free_bootmem_cpumask_var(cpumask_var_t mask)
+static __always_inline void free_bootmem_cpumask_var(cpumask_var_t mask)
 {
 }
 
-static inline bool cpumask_available(cpumask_var_t mask)
+static __always_inline bool cpumask_available(cpumask_var_t mask)
 {
 	return true;
 }
@@ -1058,7 +1070,7 @@ void set_cpu_online(unsigned int cpu, bool online);
 	((struct cpumask *)(1 ? (bitmap)				\
 			    : (void *)sizeof(__check_is_bitmap(bitmap))))
 
-static inline int __check_is_bitmap(const unsigned long *bitmap)
+static __always_inline int __check_is_bitmap(const unsigned long *bitmap)
 {
 	return 1;
 }
@@ -1073,7 +1085,7 @@ static inline int __check_is_bitmap(const unsigned long *bitmap)
 extern const unsigned long
 	cpu_bit_bitmap[BITS_PER_LONG+1][BITS_TO_LONGS(NR_CPUS)];
 
-static inline const struct cpumask *get_cpu_mask(unsigned int cpu)
+static __always_inline const struct cpumask *get_cpu_mask(unsigned int cpu)
 {
 	const unsigned long *p = cpu_bit_bitmap[1 + cpu % BITS_PER_LONG];
 	p -= cpu / BITS_PER_LONG;
@@ -1100,32 +1112,32 @@ static __always_inline unsigned int num_online_cpus(void)
 #define num_present_cpus()	cpumask_weight(cpu_present_mask)
 #define num_active_cpus()	cpumask_weight(cpu_active_mask)
 
-static inline bool cpu_online(unsigned int cpu)
+static __always_inline bool cpu_online(unsigned int cpu)
 {
 	return cpumask_test_cpu(cpu, cpu_online_mask);
 }
 
-static inline bool cpu_enabled(unsigned int cpu)
+static __always_inline bool cpu_enabled(unsigned int cpu)
 {
 	return cpumask_test_cpu(cpu, cpu_enabled_mask);
 }
 
-static inline bool cpu_possible(unsigned int cpu)
+static __always_inline bool cpu_possible(unsigned int cpu)
 {
 	return cpumask_test_cpu(cpu, cpu_possible_mask);
 }
 
-static inline bool cpu_present(unsigned int cpu)
+static __always_inline bool cpu_present(unsigned int cpu)
 {
 	return cpumask_test_cpu(cpu, cpu_present_mask);
 }
 
-static inline bool cpu_active(unsigned int cpu)
+static __always_inline bool cpu_active(unsigned int cpu)
 {
 	return cpumask_test_cpu(cpu, cpu_active_mask);
 }
 
-static inline bool cpu_dying(unsigned int cpu)
+static __always_inline bool cpu_dying(unsigned int cpu)
 {
 	return cpumask_test_cpu(cpu, cpu_dying_mask);
 }
@@ -1138,32 +1150,32 @@ static inline bool cpu_dying(unsigned int cpu)
 #define num_present_cpus()	1U
 #define num_active_cpus()	1U
 
-static inline bool cpu_online(unsigned int cpu)
+static __always_inline bool cpu_online(unsigned int cpu)
 {
 	return cpu == 0;
 }
 
-static inline bool cpu_possible(unsigned int cpu)
+static __always_inline bool cpu_possible(unsigned int cpu)
 {
 	return cpu == 0;
 }
 
-static inline bool cpu_enabled(unsigned int cpu)
+static __always_inline bool cpu_enabled(unsigned int cpu)
 {
 	return cpu == 0;
 }
 
-static inline bool cpu_present(unsigned int cpu)
+static __always_inline bool cpu_present(unsigned int cpu)
 {
 	return cpu == 0;
 }
 
-static inline bool cpu_active(unsigned int cpu)
+static __always_inline bool cpu_active(unsigned int cpu)
 {
 	return cpu == 0;
 }
 
-static inline bool cpu_dying(unsigned int cpu)
+static __always_inline bool cpu_dying(unsigned int cpu)
 {
 	return false;
 }
@@ -1197,7 +1209,7 @@ static inline bool cpu_dying(unsigned int cpu)
  * Return: the length of the (null-terminated) @buf string, zero if
  * nothing is copied.
  */
-static inline ssize_t
+static __always_inline ssize_t
 cpumap_print_to_pagebuf(bool list, char *buf, const struct cpumask *mask)
 {
 	return bitmap_print_to_pagebuf(list, buf, cpumask_bits(mask),
@@ -1220,9 +1232,9 @@ cpumap_print_to_pagebuf(bool list, char *buf, const struct cpumask *mask)
  * Return: the length of how many bytes have been copied, excluding
  * terminating '\0'.
  */
-static inline ssize_t
-cpumap_print_bitmask_to_buf(char *buf, const struct cpumask *mask,
-		loff_t off, size_t count)
+static __always_inline
+ssize_t cpumap_print_bitmask_to_buf(char *buf, const struct cpumask *mask,
+				    loff_t off, size_t count)
 {
 	return bitmap_print_bitmask_to_buf(buf, cpumask_bits(mask),
 				   nr_cpu_ids, off, count) - 1;
@@ -1242,9 +1254,9 @@ cpumap_print_bitmask_to_buf(char *buf, const struct cpumask *mask,
  * Return: the length of how many bytes have been copied, excluding
  * terminating '\0'.
  */
-static inline ssize_t
-cpumap_print_list_to_buf(char *buf, const struct cpumask *mask,
-		loff_t off, size_t count)
+static __always_inline
+ssize_t cpumap_print_list_to_buf(char *buf, const struct cpumask *mask,
+				 loff_t off, size_t count)
 {
 	return bitmap_print_list_to_buf(buf, cpumask_bits(mask),
 				   nr_cpu_ids, off, count) - 1;