11859 need swapgs mitigation
Reviewed by: Robert Mustacchi <rm@fingolfin.org>
Reviewed by: Dan McDonald <danmcd@joyent.com>
Reviewed by: Jerry Jelinek <jerry.jelinek@joyent.com>
Reviewed by: Robert Mustacchi <rm@fingolfin.org>

*** 21,31 ****
  
  /*
   * Copyright 2007 Sun Microsystems, Inc.  All rights reserved.
   * Use is subject to license terms.
   *
!  * Copyright 2018 Joyent, Inc.
   */
  
  /*
   * Debugger entry and exit for both master and slave CPUs. kdi_idthdl.s contains
   * the IDT stubs that drop into here (mainly via kdi_cmnint).
--- 21,31 ----
  
  /*
   * Copyright 2007 Sun Microsystems, Inc.  All rights reserved.
   * Use is subject to license terms.
   *
!  * Copyright 2019 Joyent, Inc.
   */
  
  /*
   * Debugger entry and exit for both master and slave CPUs. kdi_idthdl.s contains
   * the IDT stubs that drop into here (mainly via kdi_cmnint).
*** 269,278 ****
--- 269,281 ----
           * done a swapgs.  All is not lost, as the kernel can divine the correct
           * value for us.  Note that the previous GSBASE is saved in the
           * KDI_SAVE_REGS macro to prevent a usermode process's GSBASE from being
           * blown away.  On the hypervisor, we don't need to do this, since it's
           * ensured we're on our requested kernel GSBASE already.
+          *
+          * No need to worry about swapgs speculation here as it's unconditional
+          * and via wrmsr anyway.
           */
          subq    $10, %rsp
          sgdt    (%rsp)
          movq    2(%rsp), %rdi   /* gdt base now in %rdi */
          addq    $10, %rsp