linux-kernel - Re: [PATCH v2 0/3] amd : iommu : Initial IOMMU support for SNP

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <20200924105200.GP27174@8bytes.org>
Date:   Thu, 24 Sep 2020 12:52:00 +0200
From:   Joerg Roedel <joro@...tes.org>
To:     Suravee Suthikulpanit <suravee.suthikulpanit@....com>
Cc:     linux-kernel@...r.kernel.org, iommu@...ts.linux-foundation.org,
        Jon.Grimm@....com, brijesh.singh@....com
Subject: Re: [PATCH v2 0/3] amd : iommu : Initial IOMMU support for SNP

On Wed, Sep 23, 2020 at 12:13:44PM +0000, Suravee Suthikulpanit wrote:
> Suravee Suthikulpanit (3):
>   iommu: amd: Use 4K page for completion wait write-back semaphore
>   iommu: amd: Add support for RMP_PAGE_FAULT and RMP_HW_ERR
>   iommu: amd: Re-purpose Exclusion range registers to support SNP CWWB

Applied, thanks. I am slightly concerned about the re-purposing of the
exclusion-range registers based on a feature bit being set. This makes
the hardware incompatible to older IOMMU drivers which do not check the
FEATURE_SNP bit.

It will probably work in this case, as the firmware on systems with
IOMMU-SNP support will not declare exclusion ranges at all and
exclusion-ranges in the IOMMU hardware have been a bad idea since
forever, but it would have been nicer if hardware actually
provided a bit to enable this behavior.

Regards,

	Joerg