Re: [PATCH v2 2/3] kcsan: mark !__SANITIZE_THREAD__ stub __always_inline

[Brendan Jackman <jackmanb@google.com>]

On Mon Dec 15, 2025 at 12:11 PM UTC, Marco Elver wrote:
> On Mon, 15 Dec 2025 at 11:12, Brendan Jackman <jackmanb@google.com> wrote:
>>
>> The x86 instrumented bitops in
>> include/asm-generic/bitops/instrumented-non-atomic.h are
>> KCSAN-instrumented via explicit calls to instrument_* functions from
>> include/linux/instrumented.h.
>>
>> This bitops are used from noinstr code in __sev_es_nmi_complete(). This
>> code avoids noinstr violations by disabling __SANITIZE_THREAD__ etc for
>> the compilation unit.
>>
>> However, when GCOV is enabled, there can still be violations caused by
>> the stub versions of these functions, since coverage instrumentation is
>> injected that causes them to be out-of-lined.
>>
>> Fix this by just applying __always_inline.
>>
>> Signed-off-by: Brendan Jackman <jackmanb@google.com>
>> ---
>>  include/linux/kcsan-checks.h | 2 +-
>>  1 file changed, 1 insertion(+), 1 deletion(-)
>>
>> diff --git a/include/linux/kcsan-checks.h b/include/linux/kcsan-checks.h
>> index 92f3843d9ebb8177432bb4eccc151ea66d3dcbb7..cabb2ae46bdc0963bd89533777cab586ab4d5a1b 100644
>> --- a/include/linux/kcsan-checks.h
>> +++ b/include/linux/kcsan-checks.h
>> @@ -226,7 +226,7 @@ static inline void kcsan_end_scoped_access(struct kcsan_scoped_access *sa) { }
>>  #define __kcsan_disable_current kcsan_disable_current
>>  #define __kcsan_enable_current kcsan_enable_current_nowarn
>>  #else /* __SANITIZE_THREAD__ */
>> -static inline void kcsan_check_access(const volatile void *ptr, size_t size,
>> +static __always_inline void kcsan_check_access(const volatile void *ptr, size_t size,
>>                                       int type) { }
>>  static inline void __kcsan_enable_current(void)  { }
>>  static inline void __kcsan_disable_current(void) { }
>
> It wouldn't be wrong to apply __always_inline to these 2 stub
> functions as well, but I think it's fair if you just limit this to the
> ones used from <linux/instrumented.h>. Either way, please
> double-check.

I was thinking here that it's a bug to call these from noinstr code,
regardless of your ambient instrumentation settings.

But yeah, on second thoughts... says who? I don't think that _has_ to be
a bug, we could totally mark them __always_inline.