[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20250822132138.GD1311579@nvidia.com>
Date: Fri, 22 Aug 2025 10:21:38 -0300
From: Jason Gunthorpe <jgg@...dia.com>
To: "Tian, Kevin" <kevin.tian@...el.com>
Cc: Nicolin Chen <nicolinc@...dia.com>,
"robin.murphy@....com" <robin.murphy@....com>,
"joro@...tes.org" <joro@...tes.org>,
"bhelgaas@...gle.com" <bhelgaas@...gle.com>,
"will@...nel.org" <will@...nel.org>,
"robin.clark@....qualcomm.com" <robin.clark@....qualcomm.com>,
"yong.wu@...iatek.com" <yong.wu@...iatek.com>,
"matthias.bgg@...il.com" <matthias.bgg@...il.com>,
"angelogioacchino.delregno@...labora.com" <angelogioacchino.delregno@...labora.com>,
"thierry.reding@...il.com" <thierry.reding@...il.com>,
"vdumpa@...dia.com" <vdumpa@...dia.com>,
"jonathanh@...dia.com" <jonathanh@...dia.com>,
"rafael@...nel.org" <rafael@...nel.org>,
"lenb@...nel.org" <lenb@...nel.org>,
"Liu, Yi L" <yi.l.liu@...el.com>,
"baolu.lu@...ux.intel.com" <baolu.lu@...ux.intel.com>,
"linux-arm-kernel@...ts.infradead.org" <linux-arm-kernel@...ts.infradead.org>,
"iommu@...ts.linux.dev" <iommu@...ts.linux.dev>,
"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
"linux-arm-msm@...r.kernel.org" <linux-arm-msm@...r.kernel.org>,
"linux-mediatek@...ts.infradead.org" <linux-mediatek@...ts.infradead.org>,
"linux-tegra@...r.kernel.org" <linux-tegra@...r.kernel.org>,
"linux-acpi@...r.kernel.org" <linux-acpi@...r.kernel.org>,
"linux-pci@...r.kernel.org" <linux-pci@...r.kernel.org>,
"patches@...ts.linux.dev" <patches@...ts.linux.dev>,
"Jaroszynski, Piotr" <pjaroszynski@...dia.com>,
"Sethi, Vikram" <vsethi@...dia.com>,
"helgaas@...nel.org" <helgaas@...nel.org>,
"etzhao1900@...il.com" <etzhao1900@...il.com>
Subject: Re: [PATCH v3 3/5] iommu: Add iommu_get_domain_for_dev_locked()
helper
On Fri, Aug 22, 2025 at 09:45:28AM +0000, Tian, Kevin wrote:
> Thanks for the background. btw is it technically impossible or just
> not worth of the extra software complexity for no value?
Unsupported by HW. It cannot mix S2 and S1 formats on
SSIDs. If SSID is in use then all must be S1.
> e.g. if maintaining two page tables (S1/S2 formats) with exact same
> mappings, does SMMU allow smooth transition between two modes w/o
> breaking in-fly DMAs?
It could and there is SW support for that..
> but probably keeping two page tables in-sync in transition is
> already a problem w/o proper locking in map/unmap...
Yes, plus the doubling the memory. It is not worthwile there is really
very little advantange to the S2 format.
Jason
Powered by blists - more mailing lists