[PATCHv3 0/6] Crashdump Accepting Active IOMMU
ddutile at redhat.com
Tue Apr 8 11:42:56 PDT 2014
On 04/08/2014 12:14 PM, David Woodhouse wrote:
> On Mon, 2014-04-07 at 16:43 -0400, Don Dutile wrote:
>> Additionally, a tidbit of information like "some servers force NMI's
>> on DMAR faults,
>> and cause a system reset, thereby, preventing a kdump to occur"
>> should have been included as one reason to stop DMAR faults from
>> occurring on kexec-boot,
>> in addition to the fact that a flood of them can lock up a system.
> How about allocating a physical scratch page, and setting up a mapping
> for each device such that *every* virtual address (apart from those
> listed in RMRRs, perhaps) is mapped to that same scratch page?
> That way you avoid the faults, but you also avoid stray DMA to parts of
> the system that you don't want to get corrupted.
+1... more isolation as second kernel booting sounds good.
More information about the kexec