Re: [PATCH -next] x86/mm/pat: fix a data race in cpa_inc_4k_install

[Borislav Petkov]

On Mon, Jan 20, 2020 at 11:12:00PM -0500, Qian Cai wrote:
> diff --git a/arch/x86/mm/pat/set_memory.c b/arch/x86/mm/pat/set_memory.c
> index 20823392f4f2..31e4a73ae70e 100644
> --- a/arch/x86/mm/pat/set_memory.c
> +++ b/arch/x86/mm/pat/set_memory.c
> @@ -128,7 +128,7 @@ static inline void cpa_inc_2m_checked(void)
>  
>  static inline void cpa_inc_4k_install(void)
>  {
> -	cpa_4k_install++;
> +	WRITE_ONCE(cpa_4k_install, READ_ONCE(cpa_4k_install) + 1);
>  }
>  
>  static inline void cpa_inc_lp_sameprot(int level)
> -- 

Fix a data race, says your subject?

If it had to be honest, it probably should say "Make the code ugly
because the next tool can't handle it".

Frankly, I'm not a fan of all this "change the kernel to fix the tool"
attitude.

-- 
Regards/Gruss,
    Boris.

https://people.kernel.org/tglx/notes-about-netiquette