On Thu, Apr 23, 2015 at 1:49 AM, Andy Lutomirski <luto@xxxxxxxxxxxxxx> wrote:
>
> I'm curious whether we can somehow end up in the kernel without a
> sensible SS. What happens if we have SS = 0?
>
> Try this on for size:
>
> 1. Wine process does syscall
> 2. Context switch to any other task
> 3. Interrupt (software or hardware), which loads SS with ss0, which is
> 0 on x86_64.
> 4. Context switch back to Wine.
> 5. sysretl
>
> Would fixing this be as simple as changing this code in
> arch/x86/kernel/process.c:
>
> __visible DEFINE_PER_CPU_SHARED_ALIGNED(struct tss_struct, cpu_tss) = {
> .x86_tss = {
> .sp0 = TOP_OF_INIT_STACK,
> #ifdef CONFIG_X86_32
> .ss0 = __KERNEL_DS,
>
> by moving the ifdef down a line? Even if that fixed it, it would be
> extremely fragile, but IMO it would be a good change to make
> regardless (i.e. the kernel's SS would be less unpredictable).
Confirmed with KVM on VMX: we can definitely end up in the kernel with SS == 0.
I don't know whether changing ss0 would be a good idea, though. It
would be cleaner, but it could slow down interrupt processing:
interrupt delivery would have to do an extra GDT load.
Food for thought: wouldn't this mean that we have a bug on sysretq
too? If we're in the kernel with SS == 0, we do sysretq, and then
user code does a far jump to 32-bit code, then we end up with a bogus
SS. Maybe we don't care, and reloading SS on every sysretq would
suck. We could fix it up in a kind of evil way: in do_stack_segment,
we could detect that we had SS == __USER_DS, in which case #SS should
be impossible, and just return without signalling the process. IRET
would fix up the attributes.
We just might need a stable fix, though -- I wonder if there's any bad
interaction with opportunistic sysret in 4.0. Maybe we should
benchmark ss0 = __KERNEL_DS and try it after all.
--Andy
--
To unsubscribe from this list: send the line "unsubscribe linux-tip-commits" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
|