Re: [PATCH] Fix non-lazy GS handling in sys_vm86()

[Ingo Molnar <mingo@elte.hu>]

* Lubomir Rintel <lkundrak@v3.sk> wrote:

> This fixes a stack corruption panic or null dereference oops due to
> a bad GS in resume_userspace() when returning from sys_vm86() and calling
> lockdep_sys_exit().
> 
> Only a problem when CONFIG_LOCKDEP and CONFIG_CC_STACKPROTECTOR enabled.
> 
> Signed-off-by: Lubomir Rintel <lkundrak@v3.sk>
> ---
>  arch/x86/kernel/vm86_32.c |   13 +++++++------
>  1 files changed, 7 insertions(+), 6 deletions(-)
> 
> diff --git a/arch/x86/kernel/vm86_32.c b/arch/x86/kernel/vm86_32.c
> index d7ac84e..177d976 100644
> --- a/arch/x86/kernel/vm86_32.c
> +++ b/arch/x86/kernel/vm86_32.c
> @@ -287,10 +287,9 @@ static void do_sys_vm86(struct kernel_vm86_struct *info, struct task_struct *tsk
>  	info->regs.pt.ds = 0;
>  	info->regs.pt.es = 0;
>  	info->regs.pt.fs = 0;
> -
> -/* we are clearing gs later just before "jmp resume_userspace",
> - * because it is not saved/restored.
> - */
> +#ifndef CONFIG_X86_32_LAZY_GS
> +	info->regs.pt.gs = 0;
> +#endif
>  
>  /*
>   * The flags register is also special: we cannot trust that the user
> @@ -343,10 +342,12 @@ static void do_sys_vm86(struct kernel_vm86_struct *info, struct task_struct *tsk
>  	__asm__ __volatile__(
>  		"movl %0,%%esp\n\t"
>  		"movl %1,%%ebp\n\t"
> -		"mov  %2, %%gs\n\t"
> +#ifdef CONFIG_X86_32_LAZY_GS
> +		"mov  $0, %%gs\n\t"
> +#endif

That bit looks rather untested - i dont think there's an immediate 
constant instruction variant for segment register moves ...

	Ingo