From: Logan Gunthorpe <logang@deltatee.com>
To: Eric Dumazet <eric.dumazet@gmail.com>,
Bjorn Helgaas <bhelgaas@google.com>
Cc: linux-kernel <linux-kernel@vger.kernel.org>,
"Eric Dumazet" <edumazet@google.com>,
"Dan Williams" <dan.j.williams@intel.com>,
"Ira Weiny" <ira.weiny@intel.com>,
"Christoph Hellwig" <hch@lst.de>,
"Greg Kroah-Hartman" <gregkh@linuxfoundation.org>,
"Jérôme Glisse" <jglisse@redhat.com>,
"Rafael J. Wysocki" <rafael@kernel.org>,
linux-pci@vger.kernel.org
Subject: Re: [PATCH] PCI/P2PDMA: finish RCU conversion of pdev->p2pdma
Date: Mon, 5 Jul 2021 10:11:55 -0600 [thread overview]
Message-ID: <986ec985-a42c-9107-564d-5878a669388b@deltatee.com> (raw)
In-Reply-To: <20210701095621.3129283-1-eric.dumazet@gmail.com>
On 2021-07-01 3:56 a.m., Eric Dumazet wrote:
> From: Eric Dumazet <edumazet@google.com>
>
> While looking at pci_alloc_p2pmem() I found rcu protection
> was not properly applied there, as pdev->p2pdma was
> potentially read multiple times.
>
> I decided to fix pci_alloc_p2pmem(), add __rcu qualifier
> to p2pdma field of struct pci_dev, and fix all
> other accesses to this field with proper rcu verbs.
>
> Fixes: 1570175abd16 ("PCI/P2PDMA: track pgmap references per resource, not globally")
> Signed-off-by: Eric Dumazet <edumazet@google.com>
> Cc: Dan Williams <dan.j.williams@intel.com>
> Cc: Ira Weiny <ira.weiny@intel.com>
> Cc: Logan Gunthorpe <logang@deltatee.com>
> Cc: Bjorn Helgaas <bhelgaas@google.com>
> Cc: Christoph Hellwig <hch@lst.de>
> Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
> Cc: "Jérôme Glisse" <jglisse@redhat.com>
> Cc: "Rafael J. Wysocki" <rafael@kernel.org>
> Cc: linux-pci@vger.kernel.org
Reviewed-by: Logan Gunthorpe <logang@deltatee.com>
For history, though, Dan had originally suggested the full RCU
protection is not necessary and it was only a barrier to force the NULL
check on teardown to resolve:
https://lore.kernel.org/nvdimm/CAPcyv4jZiK+OHjwNqDARv4g326AQZx7N_Lmxj1Zux_bX1T2CLQ@mail.gmail.com/
Things may have changed since then and other uses might be racing with
the teardown, so having it marked __rcu and fully protected sounds like
a good idea to me.
Logan
prev parent reply other threads:[~2021-07-05 16:12 UTC|newest]
Thread overview: 4+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-07-01 9:56 [PATCH] PCI/P2PDMA: finish RCU conversion of pdev->p2pdma Eric Dumazet
2021-07-01 14:25 ` Christoph Hellwig
2021-07-01 22:09 ` Bjorn Helgaas
2021-07-05 16:11 ` Logan Gunthorpe [this message]
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=986ec985-a42c-9107-564d-5878a669388b@deltatee.com \
--to=logang@deltatee.com \
--cc=bhelgaas@google.com \
--cc=dan.j.williams@intel.com \
--cc=edumazet@google.com \
--cc=eric.dumazet@gmail.com \
--cc=gregkh@linuxfoundation.org \
--cc=hch@lst.de \
--cc=ira.weiny@intel.com \
--cc=jglisse@redhat.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-pci@vger.kernel.org \
--cc=rafael@kernel.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).