[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20250805155428.GU26511@ziepe.ca>
Date: Tue, 5 Aug 2025 12:54:28 -0300
From: Jason Gunthorpe <jgg@...pe.ca>
To: "Aneesh Kumar K.V" <aneesh.kumar@...nel.org>
Cc: linux-coco@...ts.linux.dev, kvmarm@...ts.linux.dev,
linux-pci@...r.kernel.org, linux-kernel@...r.kernel.org,
aik@....com, lukas@...ner.de, Samuel Ortiz <sameo@...osinc.com>,
Xu Yilun <yilun.xu@...ux.intel.com>,
Suzuki K Poulose <Suzuki.Poulose@....com>,
Steven Price <steven.price@....com>,
Catalin Marinas <catalin.marinas@....com>,
Marc Zyngier <maz@...nel.org>, Will Deacon <will@...nel.org>,
Oliver Upton <oliver.upton@...ux.dev>
Subject: Re: [RFC PATCH v1 04/38] tsm: Support DMA Allocation from private
memory
On Mon, Aug 04, 2025 at 12:28:33PM +0530, Aneesh Kumar K.V wrote:
> Note that we don’t update the phys_addr_t to set the top
> bit. For reference:
>
> tlb_addr = slot_addr(pool->start, index) + offset;
This seems unfortunate.
So you end up with the private IPA space having shared pages in it,
so *sometimes* you have to force the unencrypted bit?
Seems to me we should insist the phys_addr is cannonised before
reaching the dma API. Ie that the swiotlb/etc code will set the right
IPA bit.
Jason
Powered by blists - more mailing lists