On Fri, Feb 1, 2019 at 12:54 PM Chang S. Bae <chang.seok....@intel.com> wrote:
>
> Copy real FS/GSBASE values instead of approximation when FSGSBASE is
> enabled.
>
> Factoring out to save_fsgs() does not result in the same behavior because
> save_base_legacy() does not copy FS/GSBASE when the index is zero.
>
> Signed-off-by: Chang S. Bae <chang.seok....@intel.com>
> Cc: Andy Lutomirski <l...@kernel.org>
> Cc: H. Peter Anvin <h...@zytor.com>
> Cc: Thomas Gleixner <t...@linutronix.de>
> Cc: Ingo Molnar <mi...@kernel.org>
> Cc: Andi Kleen <a...@linux.intel.com>
> ---
> arch/x86/kernel/process_64.c | 12 +++++++++---
> 1 file changed, 9 insertions(+), 3 deletions(-)
>
> diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c
> index d8ade9530fdb..648e43b58c69 100644
> --- a/arch/x86/kernel/process_64.c
> +++ b/arch/x86/kernel/process_64.c
> @@ -477,10 +477,16 @@ int copy_thread_tls(unsigned long clone_flags, unsigned
> long sp,
> p->thread.sp = (unsigned long) fork_frame;
> p->thread.io_bitmap_ptr = NULL;
>
> - savesegment(gs, p->thread.gsindex);
> - p->thread.gsbase = p->thread.gsindex ? 0 : me->thread.gsbase;
> savesegment(fs, p->thread.fsindex);
> - p->thread.fsbase = p->thread.fsindex ? 0 : me->thread.fsbase;
> + savesegment(gs, p->thread.gsindex);
> + if (static_cpu_has(X86_FEATURE_FSGSBASE)) {
> + p->thread.fsbase = rdfsbase();
> + p->thread.gsbase = __rdgsbase_inactive();
> + } else {
> + /* save_base_legacy() does not set base when index is zero. */
After looking at this a bit, I propose that we just clean this up all
the way. Can't this whole mess be changed to:
save_fsgs(me);
p->thread.fsindex = me->thread.fsindex;
p->thread.fsbase = me->thread.fsbase;
p->thread.gsindex = me->thread.gsindex;
p->thread.gsbase = me->thread.gsbase;
This will avoid all of the horrible tracing through the logic to
figure out why the code is correct.
Sure, it'll be a few cycles slower with FSGSBASE, but this isn't
really a fast path, and if we ever really care, we can optimize it
later.
