2020-09-21 18:58:49

by Jerry Snitselaar

Subject: kdump boot failing with IVRS checksum failure

Hello Joerg,

We are seeing a kdump kernel boot failure in test on an HP DL325 Gen10
and it was tracked down to 387caf0b759a ("iommu/amd: Treat per-device
exclusion ranges as r/w unity-mapped regions"). Reproduced on 5.9-rc5
and goes away with revert of the commit. There is a follow on commit
that depends on this that was reverted as well 2ca6b6dc8512 ("iommu/amd:
Remove unused variable"). I'm working on getting system access and want
to see what the IVRS table looks like, but thought I'd give you heads
up.

Regards,
Jerry

2020-09-24 09:27:42

by Joerg Roedel

Subject: Re: kdump boot failing with IVRS checksum failure

Hi Jerry,

On Mon, Sep 21, 2020 at 11:56:42AM -0700, Jerry Snitselaar wrote:
> We are seeing a kdump kernel boot failure in test on an HP DL325 Gen10
> and it was tracked down to 387caf0b759a ("iommu/amd: Treat per-device
> exclusion ranges as r/w unity-mapped regions"). Reproduced on 5.9-rc5
> and goes away with revert of the commit. There is a follow on commit
> that depends on this that was reverted as well 2ca6b6dc8512 ("iommu/amd:
> Remove unused variable"). I'm working on getting system access and want
> to see what the IVRS table looks like, but thought I'd give you heads
> up.

Thanks for looking into this, we really need to find the root-cause to
avoid the revert.

Thanks,

Joerg