[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20250908182404.GH789684@nvidia.com>
Date: Mon, 8 Sep 2025 15:24:04 -0300
From: Jason Gunthorpe <jgg@...dia.com>
To: Nicolin Chen <nicolinc@...dia.com>
Cc: will@...nel.org, robin.murphy@....com, joro@...tes.org,
jean-philippe@...aro.org, miko.lenczewski@....com,
balbirs@...dia.com, peterz@...radead.org, smostafa@...gle.com,
kevin.tian@...el.com, praan@...gle.com, zhangzekun11@...wei.com,
linux-arm-kernel@...ts.infradead.org, iommu@...ts.linux.dev,
linux-kernel@...r.kernel.org, patches@...ts.linux.dev
Subject: Re: [PATCH rfcv1 7/8] iommu/arm-smmu-v3: Add arm_smmu_invs based
arm_smmu_domain_inv_range()
On Mon, Sep 08, 2025 at 11:19:40AM -0700, Nicolin Chen wrote:
> On Mon, Sep 08, 2025 at 12:39:11PM -0300, Jason Gunthorpe wrote:
> > On Sat, Sep 06, 2025 at 01:12:33AM -0700, Nicolin Chen wrote:
> >
> > > I know that performance-wise, this piece will be a quick respin,
> > > as the attach side releases the lock very fast. It still looks
> > > a bit complicated. And practically, it would respin even if the
> > > attachment removes a non-PCI device, right?
> >
> > If you are paying the cost of taking the lock then it should become
> > fully locked and consistent.
>
> Well, the point is that the reader doesn't know if an ATS entry
> is getting removed, and it can only speculate by looking at the
> full list.
It doesn't care. It knows if it has to get a full lock or not, that's
it.
> So, would it be better to just always take the read lock, while
> applying the ATS condition to the writer side:
No, the whole optimization is to avoid read side locking on the fairly
common no-ATS case.
Always taking the lock destroys that.
Jsaon
Powered by blists - more mailing lists