Re: [Patch v3 12/13] x86/speculation: Protect non-dumpable processes against Spectre v2 attack

[Waiman Long]

On 10/17/2018 01:59 PM, Tim Chen wrote:
> Mark the non-dumpable processes with TIF_STIBP flag so they will
> use STIBP and IBPB defenses against Spectre v2 attack from
> processes in user space.
>
> Signed-off-by: Tim Chen <tim.c.chen@xxxxxxxxxxxxxxx>
> ---
>  arch/x86/kernel/cpu/bugs.c | 21 +++++++++++++++++++++
>  1 file changed, 21 insertions(+)
>
> diff --git a/arch/x86/kernel/cpu/bugs.c b/arch/x86/kernel/cpu/bugs.c
> index 1d317f2..cc77b9e 100644
> --- a/arch/x86/kernel/cpu/bugs.c
> +++ b/arch/x86/kernel/cpu/bugs.c
> @@ -14,6 +14,7 @@
>  #include <linux/module.h>
>  #include <linux/nospec.h>
>  #include <linux/prctl.h>
> +#include <linux/sched/coredump.h>
>  
>  #include <asm/spec-ctrl.h>
>  #include <asm/cmdline.h>
> @@ -773,6 +774,26 @@ int arch_prctl_spec_ctrl_set(struct task_struct *task, unsigned long which,
>  	}
>  }
>  
> +void arch_set_dumpable(struct task_struct *tsk, unsigned int value)
> +{
> +	bool update;
> +
> +	if (!static_branch_unlikely(&spectre_v2_app_lite))
> +		return;
> +	if (!static_cpu_has(X86_FEATURE_STIBP))
> +		return;
> +	if (spectre_v2_app2app_enabled == SPECTRE_V2_APP2APP_NONE)
> +		return;

The third if above seems to be a subset of the first one. Do you need to
do the check one more time?

Cheers,
Longman

> +
> +	if (tsk->mm && value != SUID_DUMP_USER)
> +		update = !test_and_set_tsk_thread_flag(tsk, TIF_STIBP);
> +	else
> +		update = test_and_clear_tsk_thread_flag(tsk, TIF_STIBP);
> +
> +	if (tsk == current && update)
> +		speculation_ctrl_update_current();
> +}
> +
>  #ifdef CONFIG_SECCOMP
>  void arch_seccomp_spec_mitigate(struct task_struct *task)
>  {