From: Mika Westerberg <mika.westerberg@linux.intel.com>
To: Lukas Wunner <lukas@wunner.de>
Cc: Bjorn Helgaas <bhelgaas@google.com>,
"Rafael J. Wysocki" <rjw@rjwysocki.net>,
Len Brown <lenb@kernel.org>, Keith Busch <keith.busch@intel.com>,
Alex Williamson <alex.williamson@redhat.com>,
Alexandru Gagniuc <mr.nuke.me@gmail.com>,
linux-acpi@vger.kernel.org, linux-pci@vger.kernel.org
Subject: Re: [PATCH 2/3] PCI: Do not poll for PME if the device is in D3cold
Date: Thu, 6 Jun 2019 14:21:23 +0300 [thread overview]
Message-ID: <20190606112123.GX2781@lahna.fi.intel.com> (raw)
In-Reply-To: <20190605190557.mbklbuq4fgwbi3wp@wunner.de>
On Wed, Jun 05, 2019 at 09:05:57PM +0200, Lukas Wunner wrote:
> On Wed, Jun 05, 2019 at 05:58:19PM +0300, Mika Westerberg wrote:
> > PME polling does not take into account that a device that is directly
> > connected to the host bridge may go into D3cold as well. This leads to a
> > situation where the PME poll thread reads from a config space of a
> > device that is in D3cold and gets incorrect information because the
> > config space is not accessible.
> >
> > Here is an example from Intel Ice Lake system where two PCIe root ports
> > are in D3cold (I've instrumented the kernel to log the PMCSR register
> > contents):
> >
> > [ 62.971442] pcieport 0000:00:07.1: Check PME status, PMCSR=0xffff
> > [ 62.971504] pcieport 0000:00:07.0: Check PME status, PMCSR=0xffff
> >
> > Since 0xffff is interpreted so that PME is pending, the root ports will
> > be runtime resumed. This repeats over and over again essentially
> > blocking all runtime power management.
> >
> > Prevent this from happening by checking whether the device is in D3cold
> > before its PME status is read.
>
> There's more broken here. The below patch fixes a PME polling race
> and should also fix the issue you're witnessing, could you verify that?
It fixes the issue but I needed to tune it a bit ->
> The patch has been rotting on my development branch for several months,
> I just didn't get around to posting it, my apologies.
Better late than never :)
> -- >8 --
> Subject: [PATCH] PCI / PM: Fix race on PME polling
>
> Since commit df17e62e5bff ("PCI: Add support for polling PME state on
> suspended legacy PCI devices"), the work item pci_pme_list_scan() polls
> the PME status flag of devices and wakes them up if the bit is set.
>
> The function performs a check whether a device's upstream bridge is in
> D0 for otherwise the device is inaccessible, rendering PME polling
> impossible. However the check is racy because it is performed before
> polling the device. If the upstream bridge runtime suspends to D3hot
> after pci_pme_list_scan() checks its power state and before it invokes
> pci_pme_wakeup(), the latter will read the PMCSR as "all ones" and
> mistake it for a set PME status flag. I am seeing this race play out as
> a Thunderbolt controller going to D3cold and occasionally immediately
> going to D0 again because PM polling was performed at just the wrong
> time.
>
> Avoid by checking for an "all ones" PMCSR in pci_check_pme_status().
>
> Fixes: 58ff463396ad ("PCI PM: Add function for checking PME status of devices")
> Signed-off-by: Lukas Wunner <lukas@wunner.de>
> Cc: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
> Cc: stable@vger.kernel.org # v2.6.34+
> ---
> drivers/pci/pci.c | 2 ++
> 1 file changed, 2 insertions(+)
>
> diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c
> index b98a564..2e05348 100644
> --- a/drivers/pci/pci.c
> +++ b/drivers/pci/pci.c
> @@ -1753,6 +1753,8 @@ bool pci_check_pme_status(struct pci_dev *dev)
> pci_read_config_word(dev, pmcsr_pos, &pmcsr);
> if (!(pmcsr & PCI_PM_CTRL_PME_STATUS))
> return false;
> + if (pmcsr == ~0)
<- Here I needed to do
if (pmcsr == (u16)~0)
I think it is because pmcsr is u16 so we end up comparing:
0xffff == 0xffffffff
> + return false;
>
> /* Clear PME status. */
> pmcsr |= PCI_PM_CTRL_PME_STATUS;
> --
> 2.20.1
next prev parent reply other threads:[~2019-06-06 11:21 UTC|newest]
Thread overview: 23+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-06-05 14:58 [PATCH 0/3] PCI: Power management improvements Mika Westerberg
2019-06-05 14:58 ` [PATCH 1/3] PCI: Add missing link delays required by the PCIe spec Mika Westerberg
2019-06-06 8:40 ` Rafael J. Wysocki
2019-06-06 11:24 ` Mika Westerberg
2019-06-05 14:58 ` [PATCH 2/3] PCI: Do not poll for PME if the device is in D3cold Mika Westerberg
2019-06-05 19:05 ` Lukas Wunner
2019-06-06 11:21 ` Mika Westerberg [this message]
2019-06-09 18:38 ` Lukas Wunner
2019-06-10 11:35 ` Rafael J. Wysocki
2019-06-05 14:58 ` [PATCH 3/3] PCI / ACPI: Handle sibling devices sharing power resources Mika Westerberg
2019-06-06 8:54 ` Rafael J. Wysocki
2019-06-06 11:26 ` Mika Westerberg
2019-06-06 13:44 ` Mika Westerberg
2019-06-06 14:08 ` Rafael J. Wysocki
2019-06-06 14:17 ` Mika Westerberg
2019-06-06 14:27 ` Rafael J. Wysocki
2019-06-06 14:36 ` Mika Westerberg
2019-06-12 22:38 ` Rafael J. Wysocki
2019-06-13 12:52 ` Mika Westerberg
2019-06-13 13:51 ` Rafael J. Wysocki
2019-06-13 14:27 ` Mika Westerberg
2019-06-09 18:58 ` Lukas Wunner
2019-06-10 10:57 ` Rafael J. Wysocki
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=20190606112123.GX2781@lahna.fi.intel.com \
--to=mika.westerberg@linux.intel.com \
--cc=alex.williamson@redhat.com \
--cc=bhelgaas@google.com \
--cc=keith.busch@intel.com \
--cc=lenb@kernel.org \
--cc=linux-acpi@vger.kernel.org \
--cc=linux-pci@vger.kernel.org \
--cc=lukas@wunner.de \
--cc=mr.nuke.me@gmail.com \
--cc=rjw@rjwysocki.net \
/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).