Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.32 as permitted sender) client-ip=23.128.96.32;
Date:   Mon, 4 Dec 2023 12:11:46 -0800
In-Reply-To: <20231204195055.GA2692119@nvidia.com>
Mime-Version: 1.0
References: <20231202091211.13376-1-yan.y.zhao@intel.com> <ZW4Fx2U80L1PJKlh@google.com>
 <20231204173028.GJ1493156@nvidia.com> <ZW4nCUS9VDk0DycG@google.com> <20231204195055.GA2692119@nvidia.com>
Message-ID: <ZW4ygoqOq2JpXml3@google.com>
Subject: Re: [RFC PATCH 00/42] Sharing KVM TDP to IOMMU
From:   Sean Christopherson <seanjc@google.com>
To:     Jason Gunthorpe <jgg@nvidia.com>
Cc:     Yan Zhao <yan.y.zhao@intel.com>, iommu@lists.linux.dev,
        kvm@vger.kernel.org, linux-kernel@vger.kernel.org,
        alex.williamson@redhat.com, pbonzini@redhat.com, joro@8bytes.org,
        will@kernel.org, robin.murphy@arm.com, kevin.tian@intel.com,
        baolu.lu@linux.intel.com, dwmw2@infradead.org, yi.l.liu@intel.com
Content-Type: text/plain; charset="us-ascii"
Precedence: bulk

On Mon, Dec 04, 2023, Jason Gunthorpe wrote:
> On Mon, Dec 04, 2023 at 11:22:49AM -0800, Sean Christopherson wrote:
> > I'm not suggesting full blown mirroring, all I'm suggesting is a fire-and-forget
> > notifier for KVM to tell IOMMUFD "I've faulted in GFN A, you might want to do the
> > same".
> 
> If we say the only thing this works with is the memfd version of KVM,

That's likely a big "if", as guest_memfd is not and will not be a wholesale
replacement of VMA-based guest memory, at least not in the forseeable future.
I would be quite surprised if the target use cases for this could be moved to
guest_memfd without losing required functionality.

> could we design the memfd stuff to not have the same challenges with
> mirroring as normal VMAs? 

What challenges in particular are you concerned about?  And maybe also define
"mirroring"?  E.g. ensuring that the CPU and IOMMU page tables are synchronized
is very different than ensuring that the IOMMU page tables can only map memory
that is mappable by the guest, i.e. that KVM can map into the CPU page tables.