[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <f21aeb41-1648-24b8-01a5-143b6c99960d@amd.com>
Date: Wed, 29 Jun 2022 18:54:34 +0700
From: "Suthikulpanit, Suravee" <suravee.suthikulpanit@....com>
To: Joerg Roedel <joro@...tes.org>
Cc: linux-kernel@...r.kernel.org, iommu@...ts.linux-foundation.org,
robin.murphy@....com, ashish.kalra@....com, vasant.hegde@....com,
thomas.lendacky@....com
Subject: Re: [PATCH v3 7/7] iommu/amd: Do not support IOMMUv2 APIs when SNP is
enabled
On 6/23/2022 3:23 PM, Joerg Roedel wrote:
> On Wed, Jun 22, 2022 at 12:11:31PM -0500, Suravee Suthikulpanit wrote:
>> bool amd_iommu_v2_supported(void)
>> {
>> - return amd_iommu_v2_present;
>> + /*
>> + * Since DTE[Mode]=0 is prohibited on SNP-enabled system
>> + * (i.e. EFR[SNPSup]=1), IOMMUv2 page table cannot be used without
>> + * setting up IOMMUv1 page table.
>> + */
>> + return amd_iommu_v2_present && !amd_iommu_snp_en;
>
> IOMMU_v2 APIs could actually be supported with GIOV and IOMMUv2
> page-tables in-use, no?
>
We can support IOMMUv2 iff the v1 table is also setup (i.e. DTE[Mode] != 0).
Currently w/ IOMMU_v2 APIs, the IOMMU sets the mode to zero. Therefore, we
cannot support this use case.
Best Regards,
Suravee
Powered by blists - more mailing lists