WARNING - OLD ARCHIVES

This is an archived copy of the Xen.org mailing list, which we have preserved to ensure that existing links to archives are not broken. The live archive, which contains the latest emails, can be found at http://lists.xen.org/
   
 
 
Xen 
 
Home Products Support Community News
 
   
 

xen-devel

Re: [Xen-devel] Re: [RFC PATCH] set current_state to D0 in register_slot

On Tue, 2011-03-08 at 14:50 +0000, Ian Campbell wrote:
> On Fri, 2011-03-04 at 18:42 +0000, Jesse Barnes wrote:
> > On Mon, 28 Feb 2011 16:20:11 +0000
> > Stefano Stabellini <stefano.stabellini@xxxxxxxxxxxxx> wrote:
> > 
> > > Hi all,
> > > if a device doesn't support power management (pm_cap == 0) but it is
> > > acpi_pci_power_manageable() because there is a _PS0 method declared for
> > > it and _EJ0 is also declared for the slot then nobody is going to set
> > > current_state = PCI_D0 for this device.  This is what I think it is
> > > happening:
> > > 
> > > 
> > > pci_enable_device
> > >     |
> > > __pci_enable_device_flags
> > > /* here we do not set current_state because !pm_cap */
> > >     |
> > > do_pci_enable_device
> > >     |
> > > pci_set_power_state
> > >     |
> > > __pci_start_power_transition
> > >     |
> > > pci_platform_power_transition
> > > /* platform_pci_power_manageable() calls acpi_pci_power_manageable that
> > >  * returns true */
> > >     |
> > > platform_pci_set_power_state
> > > /* acpi_pci_set_power_state gets called and does nothing because the
> > >  * acpi device has _EJ0, see the comment "If the ACPI device has _EJ0,
> > >  * ignore the device" */
> > > 
> > > 
> > > at this point if we refer to the commit message that introduced the
> > > comment above (10b3dcae0f275e2546e55303d64ddbb58cec7599), it is up to
> > > the hotplug driver to set the state to D0.
> > > However AFAICT the pci hotplug driver never does, in fact
> > > drivers/pci/hotplug/acpiphp_glue.c:register_slot sets the slot flags to
> > > (SLOT_ENABLED | SLOT_POWEREDON) but it does not set the pci device
> > > current state to PCI_D0.
> > > 
> > > So my proposed fix is also to set current_state = PCI_D0 in
> > > register_slot.
> > > Comments are very welcome.
> > > 
> > > 
> > > Signed-off-by: Stefano Stabellini <stefano.stabellini@xxxxxxxxxxxxx>
> > 
> > Yeah, looks fine.  ACPIPHP is happy for the attention. :)
> 
> This helps for coldplug of devices which lack the PCI CFG space power
> management capability but doesn't help with hotplugging such devices.
> 
> Is it fair to assume that any PCI device in a slot which has just been
> powered on will be in D0?
> 
> e.g. Stefano's patch plus the following seem to fix the crashes I've
> been having with 82599 VFs for both hot and cold plug.
> 
> Ian.
> 
> acpiphp: set each new function to state D0 after powering on a slot.
> 
> Signed-off-by: Ian Campbell <ian.campbell@xxxxxxxxxx>

That version did it too soon and/or to devices which were already known.
I think this one is more like it:

8<------

>From 5cb5d576e7177762abe29cfba3f616c48e619145 Mon Sep 17 00:00:00 2001
From: Ian Campbell <ian.campbell@xxxxxxxxxx>
Date: Tue, 8 Mar 2011 16:16:34 +0000
Subject: [PATCH] acpiphp: assume device is in state D0 after powering on a slot.

Devices which do not support PCI configuration space based power
management may not otherwise be enabled.

Signed-off-by: Ian Campbell <ian.campbell@xxxxxxxxxx>
---
 drivers/pci/hotplug/acpiphp_glue.c |    7 +++++++
 1 files changed, 7 insertions(+), 0 deletions(-)

diff --git a/drivers/pci/hotplug/acpiphp_glue.c 
b/drivers/pci/hotplug/acpiphp_glue.c
index e610cfe..a502fef 100644
--- a/drivers/pci/hotplug/acpiphp_glue.c
+++ b/drivers/pci/hotplug/acpiphp_glue.c
@@ -827,6 +827,13 @@ static int __ref enable_device(struct acpiphp_slot *slot)
        acpiphp_set_hpp_values(bus);
        acpiphp_set_acpi_region(slot);
        pci_enable_bridges(bus);
+
+       list_for_each_entry(dev, &bus->devices, bus_list) {
+               /* Assume that newly added devices are powered on already. */
+               if (!dev->is_added)
+                       dev->current_state = PCI_D0;
+       }
+
        pci_bus_add_devices(bus);
 
        list_for_each_entry(func, &slot->funcs, sibling) {
-- 
1.5.6.5



-- 
Ian Campbell
Current Noise: Behemoth - Daimonos

Unix is a Registered Bell of AT&T Trademark Laboratories.
                -- Donn Seeley


_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-devel