From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Return-Path: Subject: Re: PCI: Revert "PCI: Add runtime PM support for PCIe ports" To: "Rafael J. Wysocki" , Bjorn Helgaas References: <20161228161816.GA19653@bhelgaas-glaptop.roam.corp.google.com> <20170103163130.GC9763@al> <20170103181221.GB25398@bhelgaas-glaptop.roam.corp.google.com> <3913750.OA6IItgEjC@aspire.rjw.lan> Cc: Peter Wu , Lukas Wunner , Mika Westerberg , linux-pci , Alex Deucher , Dave Airlie From: Kilian Singer Message-ID: <3ce9e85d-381c-1c5d-b01d-65b8cf64e1ae@quantumtechnology.info> Date: Tue, 3 Jan 2017 22:52:48 +0100 MIME-Version: 1.0 In-Reply-To: <3913750.OA6IItgEjC@aspire.rjw.lan> Content-Type: text/plain; charset=windows-1252 List-ID: I have not checked if suspend/resume broken. Which patch should I check suspend resume for. When I wrote about firefox and lockscreen issue resolved I forgot to check about the suspend/resume. On 03-Jan-17 22:38, Rafael J. Wysocki wrote: > On Tuesday, January 03, 2017 12:12:21 PM Bjorn Helgaas wrote: >> On Tue, Jan 03, 2017 at 05:31:30PM +0100, Peter Wu wrote: >>> On Tue, Jan 03, 2017 at 05:11:23PM +0100, Lukas Wunner wrote: >>>> [cc += Dave Airlie: >>>> >>>> Dave, we're about to lose support for newer Optimus laptops which use >>>> _PR3 to cut power to the discrete GPU because Bjorn Helgaas has queued >>>> a commit on his for-linus branch to remove runtime PM for PCIe ports. >>>> This fixes a regression on Kilian Singer's laptop on which locking the >>>> screen breaks USB and PS/2 input devices: Mouse movements are still >>>> visible, but button or key presses no longer have any effect. The GPU >>>> is powered down upon locking the screen and the current theory is that >>>> this causes the issues.] >>> (+cc Alex: this might affect amdgpu/radeon too.] >>> >>> Bjorn, please reconsider the rpm patch. Reverting support would >>> introduce other regressions (see issues below) and make future >>> Thunderbolt work harder (according to Lukas). If Kilian's laptop has >>> issues, what about a "temporary" quirk? >> As I mentioned at the beginning, the outcome I'm hoping for is a patch >> that fixes Kilian's laptop while preserving the runtime PM support. >> >> As I also mentioned at the beginning, preserving the runtime PM >> support at the expense of breaking Kilian's laptop is not one of the >> options. > But the revert doesn't really help. > > It doesn't fix system suspend/resume on that laptop, which also breaks when > PCIe ports PM is enabled on it. > > If you really want to use a sledgehammer approach here (which I don't recommend, > but that's your call), you can change the initial value of pci_bridge_d3_disable to > "true" (and update the pcie_ports_pm= command line to take "on" in case > someone wants to enable the feature). That at least will take care of the > regression entirely and not just partly. > > Thanks, > Rafael >