On Tue, Jan 9, 2018 at 8:03 PM, Andi Kleen <a...@firstfloor.org> wrote: > From: Andi Kleen <a...@linux.intel.com> > > We clear all the non argument registers for 64bit SYSCALLs > to minimize any risk of bad speculation using user values. > > So far unused argument registers still leak. To be addressed > in future patches. > > Signed-off-by: Andi Kleen <a...@linux.intel.com> > --- > arch/x86/entry/entry_64.S | 9 +++++++++ > 1 file changed, 9 insertions(+) > > diff --git a/arch/x86/entry/entry_64.S b/arch/x86/entry/entry_64.S > index bbdfbdd817d6..632081fd7086 100644 > --- a/arch/x86/entry/entry_64.S > +++ b/arch/x86/entry/entry_64.S > @@ -236,6 +236,14 @@ GLOBAL(entry_SYSCALL_64_after_hwframe) > pushq %r11 /* pt_regs->r11 */ > sub $(6*8), %rsp > SAVE_EXTRA_REGS > + /* Sanitize registers against speculation attacks */ > + /* r10 is cleared later, arguments are handled in san_args* */ > + CLEAR_R11_TO_R15
Don't need to explicitly clear R11 here. It is clobbered with current_task. > +#ifndef CONFIG_FRAME_POINTER > + xor %ebp, %ebp > +#endif > + xor %ebx, %ebx > + xor %ecx, %ecx > > UNWIND_HINT_REGS extra=0 > > @@ -263,6 +271,7 @@ entry_SYSCALL_64_fastpath: > #endif > ja 1f /* return -ENOSYS (already in > pt_regs->ax) */ > movq %r10, %rcx > + xor %r10, %r10 RCX is already clear, so xchgq %r10, %rcx will be simpler. -- Brian Gerst
