Messages in this thread |  | | | Date | Mon, 21 Mar 2011 20:55:21 +0300 | | From | Cyrill Gorcunov <> | | Subject | Re: [PATCH] x86, UV: Fix NMI handler for UV platforms |
| |
On 03/21/2011 08:48 PM, Cyrill Gorcunov wrote: ... >>> Another option might be to add pre-nmi notifier chain, which of course >>> not much differ from platform ops but I guess platform ops stands mostly >>> for one-shot events while chain might be more flexible. Ie I mean something >>> like >>> >>> if (notify_pre_die(DIE_NMI, "nmi", regs, 0, 2, SIGINT) == NOTIFY_STOP) >>> return; >> >> You still need to process both chains in order to handle the case where both >> hw_perf & the SGI BMC raise NMIs at about the same time. >> >> --- jack > > yes, but I meant to simply call this chain before the regular notify_die. Anyway > it would look ugly as hell too. >
And if I'm not missing something kgdb still might call IPI inside NMI handler which is look somehow strange to me...
-- Cyrill
|  |