Messages in this thread |  | | | Subject | Re: [PATCH -v3 3/6] x86, NMI, Rewrite NMI handler | | From | Peter Zijlstra <> | | Date | Tue, 12 Oct 2010 16:15:27 +0200 |
| |
On Tue, 2010-10-12 at 15:51 +0200, Andi Kleen wrote: > > No, its not making sense, there's only one event source - the NMI, it > > doesn't make any sense what so ever to then artificially split it in > > two. > > I can see that "unknown NMI" is a different kind of event than > "NMI event triggered" > > e.g. the debugger would only hook into "unknown NMI"
Sure, I'm not arguing about DIE_NMI vs DIE_NMI_UNKNOWN. I simply don't see the point of DIE_NMI_IPI.
|  |