6.6-stable review patch. If anyone has any objections, please let me know.
------------------
From: Uros Bizjak ubizjak@gmail.com
[ Upstream commit f730fd535fc51573f982fad629f2fc6b4a0cde2f ]
Guard functions in local_lock.h are defined using DEFINE_GUARD() and DEFINE_LOCK_GUARD_1() macros having lock type defined as pointer in the percpu address space. The functions, defined by these macros return value in generic address space, causing:
cleanup.h:157:18: error: return from pointer to non-enclosed address space
and
cleanup.h:214:18: error: return from pointer to non-enclosed address space
when strict percpu checks are enabled.
Add explicit casts to remove address space of the returned pointer.
Found by GCC's named address space checks.
Fixes: e4ab322fbaaa ("cleanup: Add conditional guard support") Signed-off-by: Uros Bizjak ubizjak@gmail.com Signed-off-by: Peter Zijlstra (Intel) peterz@infradead.org Link: https://lkml.kernel.org/r/20240819074124.143565-1-ubizjak@gmail.com Signed-off-by: Sasha Levin sashal@kernel.org --- include/linux/cleanup.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/include/linux/cleanup.h b/include/linux/cleanup.h index f0c6d1d45e67..64b8600eb8c0 100644 --- a/include/linux/cleanup.h +++ b/include/linux/cleanup.h @@ -123,7 +123,7 @@ static __maybe_unused const bool class_##_name##_is_conditional = _is_cond __DEFINE_CLASS_IS_CONDITIONAL(_name, false); \ DEFINE_CLASS(_name, _type, if (_T) { _unlock; }, ({ _lock; _T; }), _type _T); \ static inline void * class_##_name##_lock_ptr(class_##_name##_t *_T) \ - { return *_T; } + { return (void *)(__force unsigned long)*_T; }
#define DEFINE_GUARD_COND(_name, _ext, _condlock) \ __DEFINE_CLASS_IS_CONDITIONAL(_name##_ext, true); \ @@ -204,7 +204,7 @@ static inline void class_##_name##_destructor(class_##_name##_t *_T) \ \ static inline void *class_##_name##_lock_ptr(class_##_name##_t *_T) \ { \ - return _T->lock; \ + return (void *)(__force unsigned long)_T->lock; \ }