x86/nmi/64: Make the "NMI executing" variable more consistent

Currently, "NMI executing" is one the first time an outermost
NMI hits repeat_nmi and zero thereafter.  Change it to be zero
each time for consistency.

This is intended to help NMI handling fail harder if it's buggy.

Signed-off-by: Andy Lutomirski <luto@kernel.org>
Reviewed-by: Steven Rostedt <rostedt@goodmis.org>
Cc: Borislav Petkov <bp@suse.de>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@kernel.org>

1 files changed, 5 insertions, 6 deletions

diff --git a/arch/x86/entry/entry_64.S b/arch/x86/entry/entry_64.S
index 0fb52526e452..5422bd20bdf4 100644
--- a/arch/x86/entry/entry_64.S
+++ b/arch/x86/entry/entry_64.S
@@ -1441,8 +1441,8 @@ first_nmi:
         /* Restore rdx. */
         movq    (%rsp), %rdx
 
-        /* Set "NMI executing" on the stack. */
-        pushq   $1
+        /* Make room for "NMI executing". */
+        pushq   $0
 
         /* Leave room for the "iret" frame */
         subq    $(5*8), %rsp
@@ -1467,11 +1467,10 @@ repeat_nmi:
          * RSP is pointing to "outermost RIP".  gsbase is unknown, but, if
          * we're repeating an NMI, gsbase has the same value that it had on
          * the first iteration.  paranoid_entry will load the kernel
-         * gsbase if needed before we call do_nmi.
-         *
-         * Set "NMI executing" in case we came back here via IRET.
+         * gsbase if needed before we call do_nmi.  "NMI executing"
+         * is zero.
          */
-        movq    $1, 10*8(%rsp)
+        movq    $1, 10*8(%rsp)          /* Set "NMI executing". */
 
         /*
          * Copy the "outermost" frame to the "iret" frame.  NMIs that nest