linux-kernel - Re: [PATCH v1 01/13] genirq/msi: Store the IOMMU IOVA directly in msi_desc instead of iommu

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 for Android: free password hash cracker in your pocket

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

Message-ID: <20250213133320.5a031289@DESKTOP-0403QTC.>
Date: Thu, 13 Feb 2025 13:33:20 -0800
From: Jacob Pan <jacob.pan@...ux.microsoft.com>
To: Nicolin Chen <nicolinc@...dia.com>
Cc: <jgg@...dia.com>, <kevin.tian@...el.com>, <tglx@...utronix.de>,
 <maz@...nel.org>, <joro@...tes.org>, <will@...nel.org>,
 <robin.murphy@....com>, <shuah@...nel.org>, <iommu@...ts.linux.dev>,
 <linux-kernel@...r.kernel.org>, <linux-arm-kernel@...ts.infradead.org>,
 <linux-kselftest@...r.kernel.org>, <eric.auger@...hat.com>,
 <baolu.lu@...ux.intel.com>, <yi.l.liu@...el.com>, <yury.norov@...il.com>,
 <patches@...ts.linux.dev>, jacob.pan@...ux.microsoft.com
Subject: Re: [PATCH v1 01/13] genirq/msi: Store the IOMMU IOVA directly in
 msi_desc instead of iommu_cookie

Hi Nicolin,

On Thu, 13 Feb 2025 13:02:27 -0800
Nicolin Chen <nicolinc@...dia.com> wrote:

> On Thu, Feb 13, 2025 at 12:28:49PM -0800, Jacob Pan wrote:
> > Hi Nicolin,
> > 
> > On Sat, 8 Feb 2025 01:02:34 -0800
> > Nicolin Chen <nicolinc@...dia.com> wrote:
> >   
> > > -static inline void msi_desc_set_iommu_cookie(struct msi_desc
> > > *desc,
> > > -					     const void
> > > *iommu_cookie) +/**
> > > + * iommu_dma_compose_msi_msg() - Apply translation to an MSI
> > > message
> > > + * @desc: MSI descriptor prepared by iommu_dma_prepare_msi()
> > > + * @msg: MSI message containing target physical address
> > > + */  
> > Is it IOVA not PA?  
> 
> This is moved from dma-iommu.c so we didn't change that.
> 
> And I think it's correct to say "target physical address" as the
> irqchip driver does pass in a PA via @msg.
> 
> Then iommu_dma_compose_msi_msg() kind of reverse-translates that,
> overwriting the msg with the "IOVA" from @desc.
> 
It is just confusing that the msg can be IOVA, not always PA as the
comment says. I see it gets deleted anyway in the next patch :)