diff mbox

PCI: Revert "PCI: Add runtime PM support for PCIe ports"

Message ID 2678604.V2vzbbOmVS@aspire.rjw.lan (mailing list archive)
State New, archived
Delegated to: Bjorn Helgaas
Headers show

Commit Message

Rafael J. Wysocki Dec. 30, 2016, 1:40 a.m. UTC 
On Friday, December 30, 2016 01:45:53 AM Kilian Singer wrote:
> The touchpad and builtin keyboard of laptop fail.
> But also connected usb keyboard and usb mouse fail.
> 
> In both cases mouse pointer is still moveable.

It may just be too late to turn the ports "on" once they have been suspended at
least once on your system.

Please check if the appended patch makes any difference.

Thanks,
Rafael


---
 drivers/pci/pcie/portdrv_pci.c |    2 --
 1 file changed, 2 deletions(-)


--
To unsubscribe from this list: send the line "unsubscribe linux-pci" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Comments

Rafael J. Wysocki Dec. 30, 2016, 1:50 a.m. UTC | #1 
On Friday, December 30, 2016 02:40:45 AM Rafael J. Wysocki wrote:
> On Friday, December 30, 2016 01:45:53 AM Kilian Singer wrote:
> > The touchpad and builtin keyboard of laptop fail.
> > But also connected usb keyboard and usb mouse fail.
> > 
> > In both cases mouse pointer is still moveable.
> 
> It may just be too late to turn the ports "on" once they have been suspended at
> least once on your system.
> 
> Please check if the appended patch makes any difference.

Actually, please first check if booting with pci_port_pm=off in the kernel
command line makes any difference.

Thanks,
Rafael

--
To unsubscribe from this list: send the line "unsubscribe linux-pci" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Rafael J. Wysocki Dec. 30, 2016, 1:52 a.m. UTC | #2 
On Friday, December 30, 2016 02:50:45 AM Rafael J. Wysocki wrote:
> On Friday, December 30, 2016 02:40:45 AM Rafael J. Wysocki wrote:
> > On Friday, December 30, 2016 01:45:53 AM Kilian Singer wrote:
> > > The touchpad and builtin keyboard of laptop fail.
> > > But also connected usb keyboard and usb mouse fail.
> > > 
> > > In both cases mouse pointer is still moveable.
> > 
> > It may just be too late to turn the ports "on" once they have been suspended at
> > least once on your system.
> > 
> > Please check if the appended patch makes any difference.
> 
> Actually, please first check if booting with pci_port_pm=off in the kernel
> command line makes any difference.

The command line option should be pcie_port_pm=off ("e" was missing), sorry.

Thanks,
Rafael

--
To unsubscribe from this list: send the line "unsubscribe linux-pci" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Kilian Singer Dec. 30, 2016, 1:37 p.m. UTC | #3 
Yes,
the pci_port_pm=off
fixes both the firefox issue and the lock screen issue. Also suspend/resume work.
Tested on 4.9

----- Original Message -----
From: "Rafael J. Wysocki" <rjw@rjwysocki.net>
To: "Kilian Singer" <kilian.singer@quantumtechnology.info>
Cc: "Lukas Wunner" <lukas@wunner.de>, "Bjorn Helgaas" <helgaas@kernel.org>, "linux-pci" <linux-pci@vger.kernel.org>, "Mika Westerberg" <mika.westerberg@linux.intel.com>
Sent: Friday, December 30, 2016 2:52:59 AM
Subject: Re: PCI: Revert "PCI: Add runtime PM support for PCIe ports"

On Friday, December 30, 2016 02:50:45 AM Rafael J. Wysocki wrote:
> On Friday, December 30, 2016 02:40:45 AM Rafael J. Wysocki wrote:
> > On Friday, December 30, 2016 01:45:53 AM Kilian Singer wrote:
> > > The touchpad and builtin keyboard of laptop fail.
> > > But also connected usb keyboard and usb mouse fail.
> > > 
> > > In both cases mouse pointer is still moveable.
> > 
> > It may just be too late to turn the ports "on" once they have been suspended at
> > least once on your system.
> > 
> > Please check if the appended patch makes any difference.
> 
> Actually, please first check if booting with pci_port_pm=off in the kernel
> command line makes any difference.

The command line option should be pcie_port_pm=off ("e" was missing), sorry.

Thanks,
Rafael
--
To unsubscribe from this list: send the line "unsubscribe linux-pci" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Kilian Singer Dec. 30, 2016, 1:59 p.m. UTC | #4 
The proposed patch alone did not change the issue.


----- Original Message -----
From: "Kilian Singer" <kilian.singer@quantumtechnology.info>
To: "Rafael J. Wysocki" <rjw@rjwysocki.net>
Cc: "Lukas Wunner" <lukas@wunner.de>, "Bjorn Helgaas" <helgaas@kernel.org>, "linux-pci" <linux-pci@vger.kernel.org>, "Mika Westerberg" <mika.westerberg@linux.intel.com>
Sent: Friday, December 30, 2016 2:37:17 PM
Subject: Re: PCI: Revert "PCI: Add runtime PM support for PCIe ports"

Yes,
the pci_port_pm=off
fixes both the firefox issue and the lock screen issue. Also suspend/resume work.
Tested on 4.9

----- Original Message -----
From: "Rafael J. Wysocki" <rjw@rjwysocki.net>
To: "Kilian Singer" <kilian.singer@quantumtechnology.info>
Cc: "Lukas Wunner" <lukas@wunner.de>, "Bjorn Helgaas" <helgaas@kernel.org>, "linux-pci" <linux-pci@vger.kernel.org>, "Mika Westerberg" <mika.westerberg@linux.intel.com>
Sent: Friday, December 30, 2016 2:52:59 AM
Subject: Re: PCI: Revert "PCI: Add runtime PM support for PCIe ports"

On Friday, December 30, 2016 02:50:45 AM Rafael J. Wysocki wrote:
> On Friday, December 30, 2016 02:40:45 AM Rafael J. Wysocki wrote:
> > On Friday, December 30, 2016 01:45:53 AM Kilian Singer wrote:
> > > The touchpad and builtin keyboard of laptop fail.
> > > But also connected usb keyboard and usb mouse fail.
> > > 
> > > In both cases mouse pointer is still moveable.
> > 
> > It may just be too late to turn the ports "on" once they have been suspended at
> > least once on your system.
> > 
> > Please check if the appended patch makes any difference.
> 
> Actually, please first check if booting with pci_port_pm=off in the kernel
> command line makes any difference.

The command line option should be pcie_port_pm=off ("e" was missing), sorry.

Thanks,
Rafael
--
To unsubscribe from this list: send the line "unsubscribe linux-pci" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Rafael J. Wysocki Dec. 30, 2016, 2:44 p.m. UTC | #5 
On Friday, December 30, 2016 02:59:53 PM Kilian Singer wrote:
> The proposed patch alone did not change the issue.

I guess you mean https://patchwork.kernel.org/patch/9491693/ which probably
means that the runtime PM of the PCIe ports is enabled by user space anyway.

Thanks,
Rafael

--
To unsubscribe from this list: send the line "unsubscribe linux-pci" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Rafael J. Wysocki Dec. 30, 2016, 2:47 p.m. UTC | #6 
On Friday, December 30, 2016 02:37:17 PM Kilian Singer wrote:
> Yes,
> the pci_port_pm=off
> fixes both the firefox issue and the lock screen issue. Also suspend/resume work.
> Tested on 4.9

OK, thanks!

Please use that as a manual workaround for the time being.

I looked at the acpidump attached to the BZ entry, but nothing jumped up at me
immediately.  I'll let Mika take care of this going forward when he's back.

Thanks,
Rafael

--
To unsubscribe from this list: send the line "unsubscribe linux-pci" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Mika Westerberg Jan. 2, 2017, 12:22 p.m. UTC | #7 
On Fri, Dec 30, 2016 at 03:47:31PM +0100, Rafael J. Wysocki wrote:
> On Friday, December 30, 2016 02:37:17 PM Kilian Singer wrote:
> > Yes,
> > the pci_port_pm=off
> > fixes both the firefox issue and the lock screen issue. Also suspend/resume work.
> > Tested on 4.9
> 
> OK, thanks!
> 
> Please use that as a manual workaround for the time being.
> 
> I looked at the acpidump attached to the BZ entry, but nothing jumped up at me
> immediately.  I'll let Mika take care of this going forward when he's back.

Thanks Rafael and Lucas for the help. I'm now back from my vacation so I
can start investigating this as well.

Kilian, can you attach full output of 'sudo lspci -vv' to the bug? The
one in the comments is pretty hard to read.
--
To unsubscribe from this list: send the line "unsubscribe linux-pci" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Kilian Singer Jan. 3, 2017, 5:12 p.m. UTC | #8 
Just attached the output to the bug report as attachment.

----- Original Message -----
From: "Mika Westerberg" <mika.westerberg@linux.intel.com>
To: "Rafael J. Wysocki" <rjw@rjwysocki.net>
Cc: "Kilian Singer" <kilian.singer@quantumtechnology.info>, "Lukas Wunner" <lukas@wunner.de>, "Bjorn Helgaas" <helgaas@kernel.org>, "linux-pci" <linux-pci@vger.kernel.org>
Sent: Monday, January 2, 2017 1:22:28 PM
Subject: Re: PCI: Revert "PCI: Add runtime PM support for PCIe ports"

On Fri, Dec 30, 2016 at 03:47:31PM +0100, Rafael J. Wysocki wrote:
> On Friday, December 30, 2016 02:37:17 PM Kilian Singer wrote:
> > Yes,
> > the pci_port_pm=off
> > fixes both the firefox issue and the lock screen issue. Also suspend/resume work.
> > Tested on 4.9
> 
> OK, thanks!
> 
> Please use that as a manual workaround for the time being.
> 
> I looked at the acpidump attached to the BZ entry, but nothing jumped up at me
> immediately.  I'll let Mika take care of this going forward when he's back.

Thanks Rafael and Lucas for the help. I'm now back from my vacation so I
can start investigating this as well.

Kilian, can you attach full output of 'sudo lspci -vv' to the bug? The
one in the comments is pretty hard to read.
--
To unsubscribe from this list: send the line "unsubscribe linux-pci" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
diff mbox

Patch

Index: linux-pm/drivers/pci/pcie/portdrv_pci.c
===================================================================
--- linux-pm.orig/drivers/pci/pcie/portdrv_pci.c
+++ linux-pm/drivers/pci/pcie/portdrv_pci.c
@@ -160,7 +160,6 @@  static int pcie_portdrv_probe(struct pci
 		pm_runtime_use_autosuspend(&dev->dev);
 		pm_runtime_mark_last_busy(&dev->dev);
 		pm_runtime_put_autosuspend(&dev->dev);
-		pm_runtime_allow(&dev->dev);
 	}
 
 	return 0;
@@ -169,7 +168,6 @@  static int pcie_portdrv_probe(struct pci
 static void pcie_portdrv_remove(struct pci_dev *dev)
 {
 	if (pci_bridge_d3_possible(dev)) {
-		pm_runtime_forbid(&dev->dev);
 		pm_runtime_get_noresume(&dev->dev);
 		pm_runtime_dont_use_autosuspend(&dev->dev);
 	}