lkml.org 
[lkml]   [2012]   [Nov]   [1]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
Patch in this message
/
SubjectRe: [PATCH] [PATCH] x86: Don't clobber top of pt_regs in nested NMI
From
Date
On Thu, 2012-11-01 at 19:53 +0000, Jan Beulich wrote:
> >>> Steven Rostedt <rostedt@goodmis.org> 11/01/12 2:04 AM >>>
> >On Mon, 2012-10-01 at 17:29 -0700, Salman Qazi wrote:
> >> @@ -1826,12 +1832,15 @@ repeat_nmi:
> >> * is benign for the non-repeat case, where 1 was pushed just above
> >> * to this very stack slot).
> >> */
> >> - movq $1, 5*8(%rsp)
> >> + movq $1, 10*8(%rsp)
> >>
> >> /* Make another copy, this one may be modified by nested NMIs */
> >> + addq $(10*8), %rsp
> >
> >This breaks the CFI magic.
> >
> >> .rept 5
> >> - pushq_cfi 4*8(%rsp)
> >> + pushq_cfi -6*8(%rsp)
> >> .endr
> >> + subq $(5*8), %rsp
> >
> >So does this.
> >
> >This needs to be annotated correctly before I can push it out. But the
> >good news is, I stressed tested this change, and it all works out.
> >
> >Jan, can you help out here?
>
> There doesn't appear to be anything special about these adjustments, so I
> don't see what help would be required here - it ought to be the normal use
> of CFI_ADJUST_CFA_OFFSET that needs adding.

Even the simple CFI adjustments look like magic to me :-) OK, I'll
update the patch and send it out. I'll Cc you in case I screw up even
the most simple case ;-)

-- Steve




\
 
 \ /
  Last update: 2012-11-01 22:21    [W:0.081 / U:0.112 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site