Re: [patch part-II V2 11/13] x86/speculation/mds: Mark mds_user_clear_cpu_buffers() __always_inline

[Alexandre Chartre]

On 3/8/20 11:24 PM, Thomas Gleixner wrote:
> Prevent the compiler from uninlining and creating traceable/probable
> functions as this is invoked _after_ context tracking switched to
> CONTEXT_USER and rcu idle.
>
> Signed-off-by: Thomas Gleixner <tglx@xxxxxxxxxxxxx>
> ---
>   arch/x86/include/asm/nospec-branch.h |    4 ++--
>   1 file changed, 2 insertions(+), 2 deletions(-)
>
> --- a/arch/x86/include/asm/nospec-branch.h
> +++ b/arch/x86/include/asm/nospec-branch.h
> @@ -319,7 +319,7 @@ DECLARE_STATIC_KEY_FALSE(mds_idle_clear)
>    * combination with microcode which triggers a CPU buffer flush when the
>    * instruction is executed.
>    */
> -static inline void mds_clear_cpu_buffers(void)
> +static __always_inline void mds_clear_cpu_buffers(void)
>   {
>   	static const u16 ds = __KERNEL_DS;
>   
> @@ -340,7 +340,7 @@ static inline void mds_clear_cpu_buffers
>    *
>    * Clear CPU buffers if the corresponding static key is enabled
>    */
> -static inline void mds_user_clear_cpu_buffers(void)
> +static __always_inline void mds_user_clear_cpu_buffers(void)
>   {
>   	if (static_branch_likely(&mds_user_clear))
>   		mds_clear_cpu_buffers();

Reviewed-by: Alexandre Chartre <alexandre.chartre@xxxxxxxxxx>

I am just wondering if it would be worth defining a new function attribute to
identify functions which shouldn't be trace/probe more clearly. For example:

#define no_trace_and_probe __always_inline

static no_trace_and_probe void mds_user_clear_cpu_buffers(void)
{
        ...
}

I am just concerned that overtime we might forgot that a function is defined
__always_inline just because it shouldn't be traced/probed.

alex.