Re: [PATCH] x86_64, asm: Work around AMD SYSRET SS descriptor attribute issue

[Denys Vlasenko]

On Fri, Apr 24, 2015 at 4:18 AM, Andy Lutomirski <luto@xxxxxxxxxxxxxx> wrote:
> On Thu, Apr 23, 2015 at 7:15 PM, Andy Lutomirski <luto@xxxxxxxxxx> wrote:
> Even if the issue affects SYSRETQ, it could be that we don't care.  If
> the extent of the info leak is whether we context switched during a
> 64-bit syscall to a non-syscall context, then this is basically
> uninteresting.  In that case, we could either ignore the 64-bit issue
> entirely or fix it the other way: force SS to NULL on context switch
> (much faster, I presume) and fix it up before SYSRETL as Denys
> suggested.
>
> We clearly don't have a spate of crashes in programs that do SYSCALL
> from a 64-bit CS and then far jump/return to a 32-bit CS without first
> reloading SS, since this bug has been here forever.  I agree that the
> issue is ugly (if it exists in the first place), but maybe we don't
> need to fix it.

If you feel that concerned by the speed impact,
you can also disable this fix for !CONFIG_IA32_EMULATION kernels.
If kernel builder declared they don't want 32-bit userspace,
they won't do any ridiculous "I will temporarily
switch to 32-bit mode in 64-bit tasks" stuff either.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/