[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <9775216.kg7gRTdaXr@kreacher>
Date: Wed, 18 Sep 2019 23:42:44 +0200
From: "Rafael J. Wysocki" <rjw@...ysocki.net>
To: Mario Limonciello <mario.limonciello@...l.com>,
Keith Busch <kbusch@...nel.org>
Cc: Jens Axboe <axboe@...com>, Christoph Hellwig <hch@....de>,
Sagi Grimberg <sagi@...mberg.me>,
linux-nvme@...ts.infradead.org,
LKML <linux-kernel@...r.kernel.org>,
Ryan Hong <Ryan.Hong@...l.com>, Crag Wang <Crag.Wang@...l.com>,
sjg@...gle.com, Jared Dominguez <jared.dominguez@...l.com>,
Linux PCI <linux-pci@...r.kernel.org>,
Linux PM <linux-pm@...r.kernel.org>
Subject: Re: [PATCH] nvme-pci: Save PCI state before putting drive into deepest state
On Wednesday, September 18, 2019 11:31:19 PM CEST Rafael J. Wysocki wrote:
> On Thursday, September 12, 2019 1:42:33 AM CEST Mario Limonciello wrote:
> > The action of saving the PCI state will cause numerous PCI configuration
> > space reads which depending upon the vendor implementation may cause
> > the drive to exit the deepest NVMe state.
> >
> > In these cases ASPM will typically resolve the PCIe link state and APST
> > may resolve the NVMe power state. However it has also been observed
> > that this register access after quiesced will cause PC10 failure
> > on some device combinations.
> >
> > To resolve this, move the PCI state saving to before SetFeatures has been
> > called. This has been proven to resolve the issue across a 5000 sample
> > test on previously failing disk/system combinations.
>
> This sounds reasonable to me, but it would be nice to CC that to linux-pm
> and/or linux-pci too.
>
> > Signed-off-by: Mario Limonciello <mario.limonciello@...l.com>
> > ---
> > drivers/nvme/host/pci.c | 13 +++++++------
> > 1 file changed, 7 insertions(+), 6 deletions(-)
> >
> > diff --git a/drivers/nvme/host/pci.c b/drivers/nvme/host/pci.c
> > index 732d5b6..9b3fed4 100644
> > --- a/drivers/nvme/host/pci.c
> > +++ b/drivers/nvme/host/pci.c
> > @@ -2894,6 +2894,13 @@ static int nvme_suspend(struct device *dev)
> > if (ret < 0)
> > goto unfreeze;
> >
> > + /*
> > + * A saved state prevents pci pm from generically controlling the
> > + * device's power. If we're using protocol specific settings, we don't
> > + * want pci interfering.
> > + */
> > + pci_save_state(pdev);
> > +
> > ret = nvme_set_power_state(ctrl, ctrl->npss);
> > if (ret < 0)
> > goto unfreeze;
> > @@ -2908,12 +2915,6 @@ static int nvme_suspend(struct device *dev)
>
> This is the case in which the PCI layer is expected to put the device into
> D3, so you need
>
> pdev->state_saved = 0;
>
> at this point, because you have saved the config space already.
>
> > ret = 0;
> > goto unfreeze;
>
> And here you don't need to jump to "unfreeze" any more.
BTW, doing nvme_dev_disable() before nvme_unfreeze() looks odd to me.
Maybe it would be better to do "unfreeze" and then "disable" in this case?
>
> > }
> > - /*
> > - * A saved state prevents pci pm from generically controlling the
> > - * device's power. If we're using protocol specific settings, we don't
> > - * want pci interfering.
> > - */
> > - pci_save_state(pdev);
> > unfreeze:
> > nvme_unfreeze(ctrl);
> > return ret;
> >
>
>
>
>
>
Powered by blists - more mailing lists