[PATCH 3/4] Revert "x86: disable IOMMUs on kernel crash"

[Eric W. Biederman]

Bernhard Walle <bernhard at bwalle.de> writes:

> Am 03.04.10 19:49, schrieb Eric W. Biederman:
>> Not a problem.  We require a lot of things of the kdump kernel,
>> and it is immediately apparent in a basic sanity test.
>
> Also, in most cases (for example: distribution kernels), the kdump
> kernel nowadays is identical to the running kernel. So, if the running
> kernel has IOMMU support, the kdump kernel also has.

I have been rather pleasantly surprised at how well that works.  Although
I am still glad I insisted that using the same kernel version not be a
hard requirement.  It allowed us to specify a good solid interface.

How to cope with GART in that situation without having to manually specify
iommu=off in the configuration I find a more compelling question.

I expect I will have a look at that in the coming months if someone doesn't get
there before I do.

Eric