From: Jean-Philippe Brucker <jean-philippe@linaro.org>
To: Jason Gunthorpe <jgg@ziepe.ca>
Cc: mark.rutland@arm.com, linux-pci@vger.kernel.org,
linux-mm@kvack.org, will@kernel.org,
Dimitri Sivanich <sivanich@sgi.com>,
catalin.marinas@arm.com, zhangfei.gao@linaro.org,
devicetree@vger.kernel.org, kevin.tian@intel.com,
Arnd Bergmann <arnd@arndb.de>,
robh+dt@kernel.org, linux-arm-kernel@lists.infradead.org,
Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
iommu@lists.linux-foundation.org,
Andrew Morton <akpm@linux-foundation.org>,
robin.murphy@arm.com, christian.koenig@amd.com
Subject: Re: [PATCH v4 01/26] mm/mmu_notifiers: pass private data down to alloc_notifier()
Date: Fri, 28 Feb 2020 16:04:27 +0100 [thread overview]
Message-ID: <20200228150427.GF2156@myrica> (raw)
In-Reply-To: <20200228144844.GQ31668@ziepe.ca>
On Fri, Feb 28, 2020 at 10:48:44AM -0400, Jason Gunthorpe wrote:
> On Fri, Feb 28, 2020 at 03:39:35PM +0100, Jean-Philippe Brucker wrote:
> > > > + list_for_each_entry_rcu(bond, &io_mm->devices, mm_head) {
> > > > + /*
> > > > + * To ensure that we observe the initialization of io_mm fields
> > > > + * by io_mm_finalize() before the registration of this bond to
> > > > + * the list by io_mm_attach(), introduce an address dependency
> > > > + * between bond and io_mm. It pairs with the smp_store_release()
> > > > + * from list_add_rcu().
> > > > + */
> > > > + io_mm = rcu_dereference(bond->io_mm);
> > >
> > > A rcu_dereference isn't need here, just a normal derference is fine.
> >
> > bond->io_mm is annotated with __rcu (for iommu_sva_get_pasid_generic(),
> > which does bond->io_mm under rcu_read_lock())
>
> I'm surprised the bond->io_mm can change over the lifetime of the
> bond memory..
The normal lifetime of the bond is between device driver calls to bind()
and unbind(). If the mm exits early, though, we clear bond->io_mm. The
bond is then stale but can only be freed when the device driver releases
it with unbind().
>
> > > > If io_mm->ctx and io_mm->ops are already valid before the
> > > > mmu notifier is published, then we don't need that stuff.
> > >
> > > So, this trickyness with RCU is not a bad reason to introduce the priv
> > > scheme, maybe explain it in the commit message?
> >
> > Ok, I've added this to the commit message:
> >
> > The IOMMU SVA module, which attaches an mm to multiple devices,
> > exemplifies this situation. In essence it does:
> >
> > mmu_notifier_get()
> > alloc_notifier()
> > A = kzalloc()
> > /* MMU notifier is published */
> > A->ctx = ctx; // (1)
> > device->A = A;
> > list_add_rcu(device, A->devices); // (2)
> >
> > The invalidate notifier, which may start running before A is fully
> > initialized at (1), does the following:
> >
> > io_mm_invalidate(A)
> > list_for_each_entry_rcu(device, A->devices)
> > A = device->A; // (3)
>
> I would drop the work around from the decription, it is enough to say
> that the line below needs to observe (1) after (2) and this is
> trivially achieved by moving (1) to before publishing the notifier so
> the core MM locking can be used.
Ok, will do
Thanks,
Jean
next prev parent reply other threads:[~2020-02-28 15:04 UTC|newest]
Thread overview: 70+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-02-24 18:23 [PATCH v4 00/26] iommu: Shared Virtual Addressing and SMMUv3 support Jean-Philippe Brucker
2020-02-24 18:23 ` [PATCH v4 01/26] mm/mmu_notifiers: pass private data down to alloc_notifier() Jean-Philippe Brucker
2020-02-24 19:00 ` Jason Gunthorpe
2020-02-25 9:24 ` Jean-Philippe Brucker
2020-02-25 14:08 ` Jason Gunthorpe
2020-02-28 14:39 ` Jean-Philippe Brucker
2020-02-28 14:48 ` Jason Gunthorpe
2020-02-28 15:04 ` Jean-Philippe Brucker [this message]
2020-02-28 15:13 ` Jason Gunthorpe
2020-03-06 9:56 ` Jean-Philippe Brucker
2020-03-06 13:09 ` Jason Gunthorpe
2020-03-06 14:35 ` Jean-Philippe Brucker
2020-03-06 14:52 ` Jason Gunthorpe
2020-03-06 16:15 ` Jean-Philippe Brucker
2020-03-06 17:42 ` Jason Gunthorpe
2020-03-13 18:49 ` Jean-Philippe Brucker
2020-03-13 19:13 ` Jason Gunthorpe
2020-03-16 15:46 ` Christoph Hellwig
2020-03-17 18:40 ` Jason Gunthorpe
2020-03-05 16:36 ` Christoph Hellwig
2020-02-24 18:23 ` [PATCH v4 02/26] iommu/sva: Manage process address spaces Jean-Philippe Brucker
2020-02-26 12:35 ` Jonathan Cameron
2020-02-28 14:43 ` Jean-Philippe Brucker
2020-02-28 16:26 ` Jonathan Cameron
2020-02-26 19:13 ` Jacob Pan
2020-02-28 14:40 ` Jean-Philippe Brucker
2020-02-28 14:57 ` Jason Gunthorpe
2020-02-24 18:23 ` [PATCH v4 03/26] iommu: Add a page fault handler Jean-Philippe Brucker
2020-02-25 3:30 ` Xu Zaibo
2020-02-25 9:25 ` Jean-Philippe Brucker
2020-02-26 3:05 ` Xu Zaibo
2020-02-26 13:59 ` Jonathan Cameron
2020-02-28 14:44 ` Jean-Philippe Brucker
2020-02-24 18:23 ` [PATCH v4 04/26] iommu/sva: Search mm by PASID Jean-Philippe Brucker
2020-02-24 18:23 ` [PATCH v4 05/26] iommu/iopf: Handle mm faults Jean-Philippe Brucker
2020-02-24 18:23 ` [PATCH v4 06/26] iommu/sva: Register page fault handler Jean-Philippe Brucker
2020-02-26 19:39 ` Jacob Pan
2020-02-28 14:44 ` Jean-Philippe Brucker
2020-02-24 18:23 ` [PATCH v4 07/26] arm64: mm: Pin down ASIDs for sharing mm with devices Jean-Philippe Brucker
2020-02-27 17:43 ` Jonathan Cameron
2020-03-04 14:10 ` Jean-Philippe Brucker
2020-02-24 18:23 ` [PATCH v4 08/26] iommu/io-pgtable-arm: Move some definitions to a header Jean-Philippe Brucker
2020-02-24 18:23 ` [PATCH v4 09/26] iommu/arm-smmu-v3: Manage ASIDs with xarray Jean-Philippe Brucker
2020-02-24 18:23 ` [PATCH v4 10/26] arm64: cpufeature: Export symbol read_sanitised_ftr_reg() Jean-Philippe Brucker
2020-02-24 18:23 ` [PATCH v4 11/26] iommu/arm-smmu-v3: Share process page tables Jean-Philippe Brucker
2020-02-24 18:23 ` [PATCH v4 12/26] iommu/arm-smmu-v3: Seize private ASID Jean-Philippe Brucker
2020-02-24 18:23 ` [PATCH v4 13/26] iommu/arm-smmu-v3: Add support for VHE Jean-Philippe Brucker
2020-02-24 18:23 ` [PATCH v4 14/26] iommu/arm-smmu-v3: Enable broadcast TLB maintenance Jean-Philippe Brucker
2020-02-24 18:23 ` [PATCH v4 15/26] iommu/arm-smmu-v3: Add SVA feature checking Jean-Philippe Brucker
2020-02-24 18:23 ` [PATCH v4 16/26] iommu/arm-smmu-v3: Add dev_to_master() helper Jean-Philippe Brucker
2020-02-24 18:23 ` [PATCH v4 17/26] iommu/arm-smmu-v3: Implement mm operations Jean-Philippe Brucker
2020-02-24 18:23 ` [PATCH v4 18/26] iommu/arm-smmu-v3: Hook up ATC invalidation to mm ops Jean-Philippe Brucker
2020-02-24 18:23 ` [PATCH v4 19/26] iommu/arm-smmu-v3: Add support for Hardware Translation Table Update Jean-Philippe Brucker
2020-02-24 18:23 ` [PATCH v4 20/26] iommu/arm-smmu-v3: Maintain a SID->device structure Jean-Philippe Brucker
2020-02-24 18:23 ` [PATCH v4 21/26] iommu/arm-smmu-v3: Ratelimit event dump Jean-Philippe Brucker
2021-05-28 8:09 ` Aaro Koskinen
2021-05-28 16:25 ` Jean-Philippe Brucker
2020-02-24 18:23 ` [PATCH v4 22/26] dt-bindings: document stall property for IOMMU masters Jean-Philippe Brucker
2020-02-24 18:23 ` [PATCH v4 23/26] iommu/arm-smmu-v3: Add stall support for platform devices Jean-Philippe Brucker
2020-02-26 8:44 ` Xu Zaibo
2020-03-04 14:09 ` Jean-Philippe Brucker
2020-02-27 18:17 ` Jonathan Cameron
2020-03-04 14:08 ` Jean-Philippe Brucker
2020-03-09 10:48 ` Jonathan Cameron
2020-02-24 18:23 ` [PATCH v4 24/26] PCI/ATS: Add PRI stubs Jean-Philippe Brucker
2020-02-27 20:55 ` Bjorn Helgaas
2020-02-24 18:24 ` [PATCH v4 25/26] PCI/ATS: Export symbols of PRI functions Jean-Philippe Brucker
2020-02-27 20:55 ` Bjorn Helgaas
2020-02-24 18:24 ` [PATCH v4 26/26] iommu/arm-smmu-v3: Add support for PRI Jean-Philippe Brucker
2020-02-27 18:22 ` [PATCH v4 00/26] iommu: Shared Virtual Addressing and SMMUv3 support Jonathan Cameron
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20200228150427.GF2156@myrica \
--to=jean-philippe@linaro.org \
--cc=akpm@linux-foundation.org \
--cc=arnd@arndb.de \
--cc=catalin.marinas@arm.com \
--cc=christian.koenig@amd.com \
--cc=devicetree@vger.kernel.org \
--cc=gregkh@linuxfoundation.org \
--cc=iommu@lists.linux-foundation.org \
--cc=jgg@ziepe.ca \
--cc=kevin.tian@intel.com \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=linux-mm@kvack.org \
--cc=linux-pci@vger.kernel.org \
--cc=mark.rutland@arm.com \
--cc=robh+dt@kernel.org \
--cc=robin.murphy@arm.com \
--cc=sivanich@sgi.com \
--cc=will@kernel.org \
--cc=zhangfei.gao@linaro.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).