linux-kernel - Re: [PATCH v4 2/2] PCI/PM: Fix pci_pm_suspend

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20220423150132.GA1552054@bhelgaas>
Date:   Sat, 23 Apr 2022 10:01:32 -0500
From:   Bjorn Helgaas <helgaas@...nel.org>
To:     "Jingar, Rajvi" <rajvi.jingar@...el.com>
Cc:     "bhelgaas@...gle.com" <bhelgaas@...gle.com>,
        "david.e.box@...ux.intel.com" <david.e.box@...ux.intel.com>,
        "linux-pci@...r.kernel.org" <linux-pci@...r.kernel.org>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
        "linux-pm@...r.kernel.org" <linux-pm@...r.kernel.org>,
        "Wysocki, Rafael J" <rafael.j.wysocki@...el.com>,
        Kai-Heng Feng <kai.heng.feng@...onical.com>,
        "mika.westerberg@...ux.intel.com" <mika.westerberg@...ux.intel.com>,
        "koba.ko@...onical.com" <koba.ko@...onical.com>,
        "baolu.lu@...ux.intel.com" <baolu.lu@...ux.intel.com>,
        "sathyanarayanan.kuppuswamy@...ux.intel.com" 
        <sathyanarayanan.kuppuswamy@...ux.intel.com>,
        Russell Currey <ruscur@...sell.cc>,
        Oliver O'Halloran <oohall@...il.com>,
        "linuxppc-dev@...ts.ozlabs.org" <linuxppc-dev@...ts.ozlabs.org>
Subject: Re: [PATCH v4 2/2] PCI/PM: Fix pci_pm_suspend_noirq() to disable PTM

On Sat, Apr 23, 2022 at 12:43:14AM +0000, Jingar, Rajvi wrote:
> > -----Original Message-----
> > From: Bjorn Helgaas <helgaas@...nel.org>
> > On Thu, Apr 14, 2022 at 07:54:02PM +0200, Rafael J. Wysocki wrote:
> > > On 3/25/2022 8:50 PM, Rajvi Jingar wrote:
> > > > For the PCIe devices (like nvme) that do not go into D3 state still need to
> > > > disable PTM on PCIe root ports to allow the port to enter a lower-power PM
> > > > state and the SoC to reach a lower-power idle state as a whole. Move the
> > > > pci_disable_ptm() out of pci_prepare_to_sleep() as this code path is not
> > > > followed for devices that do not go into D3. This patch fixes the issue
> > > > seen on Dell XPS 9300 with Ice Lake CPU and Dell Precision 5530 with Coffee
> > > > Lake CPU platforms to get improved residency in low power idle states.
> > > >
> > > > Fixes: a697f072f5da ("PCI: Disable PTM during suspend to save power")
> > > > Signed-off-by: Rajvi Jingar <rajvi.jingar@...el.com>
> > > > Suggested-by: David E. Box <david.e.box@...ux.intel.com>
> > > > ---
> > > >   drivers/pci/pci-driver.c | 10 ++++++++++
> > > >   drivers/pci/pci.c        | 10 ----------
> > > >   2 files changed, 10 insertions(+), 10 deletions(-)
> > > >
> > > > diff --git a/drivers/pci/pci-driver.c b/drivers/pci/pci-driver.c
> > > > index 8b55a90126a2..ab733374a260 100644
> > > > --- a/drivers/pci/pci-driver.c
> > > > +++ b/drivers/pci/pci-driver.c
> > > > @@ -847,6 +847,16 @@ static int pci_pm_suspend_noirq(struct device *dev)
> > > >   	if (!pci_dev->state_saved) {
> > > >   		pci_save_state(pci_dev);
> > > > +		/*
> > > > +		 * There are systems (for example, Intel mobile chips since
> > Coffee
> > > > +		 * Lake) where the power drawn while suspended can be
> > significantly
> > > > +		 * reduced by disabling PTM on PCIe root ports as this allows the
> > > > +		 * port to enter a lower-power PM state and the SoC to reach a
> > > > +		 * lower-power idle state as a whole.
> > > > +		 */
> > > > +		if (pci_pcie_type(pci_dev) == PCI_EXP_TYPE_ROOT_PORT)
> > > > +			pci_disable_ptm(pci_dev);
> > 
> > Why is disabling PTM dependent on pci_dev->state_saved?  The point of
> > this is to change the behavior of the device, and it seems like we
> > want to do that regardless of whether the driver has used
> > pci_save_state().
> 
> Because we use the saved state to restore PTM on the root port. 
> And it's under this condition that the root port state gets saved.

Yes, I understand that pci_restore_ptm_state() depends on a previous
call to pci_save_ptm_state().

The point I'm trying to make is that pci_disable_ptm() changes the
state of the device, and that state change should not depend on
whether the driver has used pci_save_state().

When we're putting a device into a low-power state, I think we want to
disable PTM *always*, no matter what the driver did.  And I think we
want to do it for all devices, not just Root Ports.

Bjorn