linux-kernel - Re: [PATCH] iommu/arm-smmu-qcom: Enable use of all SMR groups when running bare-metal

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: Windows password security audit tool. GUI, reports in PDF.

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

Message-ID: <aQCguL11h6oa5Y4C@linaro.org>
Date: Tue, 28 Oct 2025 11:53:44 +0100
From: Stephan Gerhold <stephan.gerhold@...aro.org>
To: Robin Murphy <robin.murphy@....com>
Cc: Will Deacon <will@...nel.org>, Joerg Roedel <joro@...tes.org>,
	Rob Clark <robin.clark@....qualcomm.com>,
	Manivannan Sadhasivam <mani@...nel.org>,
	Johan Hovold <johan@...nel.org>,
	Bjorn Andersson <andersson@...nel.org>, iommu@...ts.linux.dev,
	linux-arm-msm@...r.kernel.org, linux-arm-kernel@...ts.infradead.org,
	linux-kernel@...r.kernel.org
Subject: Re: [PATCH] iommu/arm-smmu-qcom: Enable use of all SMR groups when
 running bare-metal

Hi Robin,

On Fri, Sep 19, 2025 at 08:51:01AM +0100, Will Deacon wrote:
> On Wed, Sep 17, 2025 at 09:16:46PM +0200, Stephan Gerhold wrote:
> > I realize it is weird to allow non-architectural features like this, but
> > I haven't found any indication that the additional SMRs work any
> > different from the standard ones. The SMMU spec seems to reserve space
> > for up to 256 SMRs in the address space and the register bits, as if it
> > was intended to be extended like this later. That's also why it works
> > correctly without any changes in arm-smmu.c: the bit masks used there
> > already allow up to 256 SMRs.
> > 
> > What do you think?
> 
> Although it's all pretty ugly, I think we really only have two choices:
> 
>   - Teach the core driver code about all this and use an rmr-like scheme
>     to leave the upper SMRs in bypass
> 
>   - Hack it in the impl code as per your patch
> 
> The latter option is probably the most pragmatic (especially considering
> the need to handle the virtualised case differently) but I'd like to see
> what Robin thinks.

Do you have any thoughts for this?

The stream match conflicts that happen without this patch still exist,
I've been testing this patch for quite some time now and have never
noticed any issues.

Thanks,
Stephan