Messages in this thread | | | Date | Sun, 8 Jan 2012 09:48:44 +0100 | From | Ingo Molnar <> | Subject | Re: [PATCH 0/6 v3] [GIT PULL] x86: Workaround for NMI iret woes |
| |
* Steven Rostedt <rostedt@goodmis.org> wrote:
> > Ingo, > > I updated the patches to reflect your comments and ran them through > all my tests again. Below is the full diff between v2 and v3. > > I renamed nmi_preprocess() and nmi_postprocess() to > nmi_nesting_preprocess() and nmi_nesting_postprocess() to at least > note that this has to do with nmi nesting. I also commented the > fact that i386 may loop back to the preprocess. > > I had to move the debug_usage functions to debugreg.h, because > smp.h includes processor.h where making the inc/dec inline > caused include header issues. > > -- Steve > > Please pull the latest tip/x86/core-3 tree, which can be found at: > > git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace.git > tip/x86/core-3 > > Head SHA1: 42181186ad4db986fcaa40ca95c6e407e9e79372 > > > Linus Torvalds (1): > x86: Do not schedule while still in NMI context > > Steven Rostedt (5): > x86: Document the NMI handler about not using paranoid_exit > x86: Add workaround to NMI iret woes > x86: Keep current stack in NMI breakpoints > x86: Allow NMIs to hit breakpoints in i386 > x86: Add counter when debug stack is used with interrupts enabled > > ---- > arch/x86/include/asm/debugreg.h | 22 ++++ > arch/x86/include/asm/desc.h | 12 ++ > arch/x86/kernel/cpu/common.c | 24 +++++ > arch/x86/kernel/entry_64.S | 218 +++++++++++++++++++++++++++++++++------ > arch/x86/kernel/head_64.S | 4 + > arch/x86/kernel/nmi.c | 102 ++++++++++++++++++ > arch/x86/kernel/traps.c | 20 ++++ > 7 files changed, 369 insertions(+), 33 deletions(-)
Pulled, thanks a lot Steve! I've ran 500 randconfig iterations and these bits seem fine.
Thanks,
Ingo
| |