From: Keith Busch <kbusch@kernel.org>
To: Sagi Grimberg <sagi@grimberg.me>
Cc: Edmund Nadolski <edmund.nadolski@intel.com>,
Christoph Hellwig <hch@lst.de>,
linux-nvme@lists.infradead.org,
James Smart <james.smart@broadcom.com>
Subject: Re: [PATCHv2 6/6] nvme: Wait for reset state when required
Date: Fri, 20 Sep 2019 12:26:37 -0600 [thread overview]
Message-ID: <20190920182637.GD97854@C02WT3WMHTD6> (raw)
In-Reply-To: <debdee44-22a1-0361-9d20-4af6a11dab0d@grimberg.me>
On Fri, Sep 20, 2019 at 11:13:06AM -0700, Sagi Grimberg wrote:
> On 9/19/19 2:34 PM, kbusch@kernel.org wrote:
> > static void nvme_reset_done(struct pci_dev *pdev)
> > {
> > struct nvme_dev *dev = pci_get_drvdata(pdev);
> > - nvme_reset_ctrl_sync(&dev->ctrl);
> > +
> > + if (!nvme_reset_continue(&dev->ctrl))
> > + flush_work(&dev->ctrl.reset_work);
>
> Is nvme_reset_continue allowed to fail here?
> As I see it, the flush must complete regardless doesn't it?
I was thinking a surprise hot-removal, but no, the pci driver is holding
the device lock across the prepare/done sequence, so I don't think it's
possible for this reset_continue() to fail to schedule.
> > }
> > static void nvme_shutdown(struct pci_dev *pdev)
> > {
> > struct nvme_dev *dev = pci_get_drvdata(pdev);
> > - nvme_dev_disable(dev, true);
> > + nvme_disable_prepare_reset(dev, true);
> > }
> > /*
> > @@ -2898,7 +2920,7 @@ static int nvme_resume(struct device *dev)
> > if (ndev->last_ps == U32_MAX ||
> > nvme_set_power_state(ctrl, ndev->last_ps) != 0)
> > - nvme_reset_ctrl(ctrl);
> > + return nvme_reset_continue(&ndev->ctrl);
>
> Is the controller allowed not to be in RESETTING in this stage?
> do we need a WARN_ON if it is?
This one, though, I think a hot removal can cause this to fail since the
device lock isn't held from suspend all the way through resume, so we
could be in the DELETING or DEAD state.
_______________________________________________
Linux-nvme mailing list
Linux-nvme@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-nvme
next prev parent reply other threads:[~2019-09-20 18:26 UTC|newest]
Thread overview: 41+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-09-19 21:34 [PATCHv2 0/6] nvme: double reset prevention kbusch
2019-09-19 21:34 ` [PATCHv2 1/6] nvme-pci: Free tagset if no IO queues kbusch
2019-09-20 17:49 ` Sagi Grimberg
2019-09-20 17:53 ` Keith Busch
2019-09-20 17:56 ` Sagi Grimberg
2019-09-27 21:33 ` Christoph Hellwig
2019-09-19 21:34 ` [PATCHv2 2/6] nvme: Remove ADMIN_ONLY state kbusch
2019-09-20 17:53 ` Sagi Grimberg
2019-09-27 21:35 ` Christoph Hellwig
2019-09-19 21:34 ` [PATCHv2 3/6] nvme: Restart request timers in resetting state kbusch
2019-09-20 17:53 ` Sagi Grimberg
2019-09-27 21:36 ` Christoph Hellwig
2019-09-19 21:34 ` [PATCHv2 4/6] nvme: Introduce nvme_reset_continue kbusch
2019-09-20 18:15 ` Sagi Grimberg
2019-09-27 21:37 ` Christoph Hellwig
2019-09-27 21:44 ` Christoph Hellwig
2019-09-19 21:34 ` [PATCHv2 5/6] nvme: Prevent resets during paused states kbusch
2019-09-20 18:03 ` Sagi Grimberg
2019-09-20 18:05 ` Sagi Grimberg
2019-09-20 18:08 ` Keith Busch
2019-09-20 18:14 ` Sagi Grimberg
2019-09-27 21:41 ` Christoph Hellwig
2019-09-28 0:23 ` Keith Busch
[not found] ` <CGME20191010120317uscas1p1e4483ca19dbb0e550c413c18c5928537@uscas1p1.samsung.com>
2019-10-10 12:03 ` Judy Brock
2019-10-10 13:59 ` Keith Busch
2019-10-10 14:03 ` Christoph Hellwig
2019-10-11 0:51 ` Judy Brock
2019-10-11 14:23 ` Keith Busch
2019-09-19 21:34 ` [PATCHv2 6/6] nvme: Wait for reset state when required kbusch
2019-09-20 18:13 ` Sagi Grimberg
2019-09-20 18:26 ` Keith Busch [this message]
2019-09-20 19:29 ` Keith Busch
2019-09-20 20:49 ` Sagi Grimberg
2019-09-20 21:06 ` Keith Busch
2019-09-27 21:43 ` Christoph Hellwig
2019-09-20 17:51 ` [PATCHv2 0/6] nvme: double reset prevention Sagi Grimberg
2019-09-20 17:54 ` Keith Busch
2019-09-23 22:09 ` Sagi Grimberg
2019-09-24 15:07 ` Keith Busch
2019-09-24 18:07 ` Sagi Grimberg
2019-09-24 18:12 ` Keith Busch
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=20190920182637.GD97854@C02WT3WMHTD6 \
--to=kbusch@kernel.org \
--cc=edmund.nadolski@intel.com \
--cc=hch@lst.de \
--cc=james.smart@broadcom.com \
--cc=linux-nvme@lists.infradead.org \
--cc=sagi@grimberg.me \
/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).