On Wed, 8 May 2019, Chang S. Bae wrote:
> From: Andy Lutomirski <l...@kernel.org>
>
> With the new FSGSBASE instructions, we can efficiently read and write
> the FSBASE and GSBASE in __switch_to(). Use that capability to preserve
> the full state.
>
> This will enable user code to do whatever it wants with the new
> instructions without any kernel-induced gotchas. (There can still be
> architectural gotchas: movl %gs,%eax; movl %eax,%gs may change GSBASE
> if WRGSBASE was used, but users are expected to read the CPU manual
> before doing things like that.)
>
> This is a considerable speedup. It seems to save about 100 cycles
> per context switch compared to the baseline 4.6-rc1 behavior on my
> Skylake laptop.
>
> [ chang: 5~10% performance improvements were seen by a context switch
> benchmark that ran threads with different FS/GSBASE values (to the
> baseline 4.16). Minor edit on the changelog. ]
>
> Signed-off-by: Andy Lutomirski <l...@kernel.org>
> Signed-off-by: Chang S. Bae <chang.seok....@intel.com>
> Reviewed-by: Andi Kleen <a...@linux.intel.com>
> Cc: H. Peter Anvin <h...@zytor.com>
> Cc: Thomas Gleixner <t...@linutronix.de>
> Cc: Ingo Molnar <mi...@kernel.org>
> ---
> arch/x86/kernel/process_64.c | 34 ++++++++++++++++++++++++++++------
> 1 file changed, 28 insertions(+), 6 deletions(-)
>
> diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c
> index 17421c3..e2089c9 100644
> --- a/arch/x86/kernel/process_64.c
> +++ b/arch/x86/kernel/process_64.c
> @@ -239,8 +239,18 @@ static __always_inline void save_fsgs(struct task_struct
> *task)
> {
> savesegment(fs, task->thread.fsindex);
> savesegment(gs, task->thread.gsindex);
> - save_base_legacy(task, task->thread.fsindex, FS);
> - save_base_legacy(task, task->thread.gsindex, GS);
> + if (static_cpu_has(X86_FEATURE_FSGSBASE)) {
> + /*
> + * If FSGSBASE is enabled, we can't make any useful guesses
> + * about the base, and user code expects us to save the current
> + * value. Fortunately, reading the base directly is efficient.
> + */
> + task->thread.fsbase = rdfsbase();
> + task->thread.gsbase = __rdgsbase_inactive();
This explodes when called from copy_thread_tls() when an interrupt hits
after switching GS in __rdgsbase_inactive(). Wrapped it with
local_irq_save/restore().
Oh well.
Thanks,
tglx
