Re: [PATCH 1/2] ACPI / scan: Make namespace scanning and trimming mutually exclusive

[Steven Newbury <steve@snewbury.org.uk>]

On Sat,   2 Feb 2013, 20:18:28 GMT, Rafael J. Wysocki <rjw@sisk.pl> wrote:

> On Saturday, February 02, 2013 11:58:41 AM Steven Newbury wrote:
> > On Sat, 2013-01-26 at 15:19 -0800, Yinghai Lu wrote:
> > > On Sat, Jan 26, 2013 at 2:41 PM, Rafael J. Wysocki <rjw@sisk.pl>
> > > wrote:
> > > > From: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
> > > > 
> > > > There is no guarantee that acpi_bus_scan() and acpi_bus_trim() will
> > > > not be run in parallel for the same scope of the ACPI namespace,
> > > > which may lead to a great deal of confusion, so introduce a new
> > > > mutex to prevent that from happening.
> > > > 
> > > > Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
> > > 
> > > Acked-by: Yinghai Lu <yinghai@kernel.org>
> > > 
> > > Steven,
> > > 
> > > Can you apply this one to for-pci-res-alloc to check if racing with
> > > docking hotplug/eject
> > > still happen?
> > > or wait one or two days after i rebase that branch.
> > 
> > Tried merging with linux-pm/bleeding-edge, same behaviour:
> > 
> > [ 3589.013578] ACPI: \_SB_.PCI0.PCIE.GDCK: undocking
> > [ 3589.585356] vgaarb: device changed decodes:
> > PCI:0000:00:02.0,olddecodes=none,decodes=io+mem:owns=none [
> > 3589.585422] ACPI: Delete PCI Interrupt Routing Table for 0000:04 [
> > 3589.585426] pci 0000:03:08.0: Oops, 'acpi_handle' corrupt [
> > 3589.585446] pci_bus 0000:04: busn_res: [bus 04] is released
> > 
> > 03:08.0 PCI bridge: PLX Technology, Inc. PEX8112 x1 Lane PCI
> > Express-to-PCI Bridge (rev aa)
> > 04:00.0 VGA compatible controller: Advanced Micro Devices [AMD] nee ATI
> > Manhattan [Mobility Radeon HD 5430 Series]
> > 04:00.1 Audio device: Advanced Micro Devices [AMD] nee ATI Cedar HDMI
> > Audio [Radeon HD 5400/6300 Series]
> 
> That's because of a bug in the dock code I believe.
> 
> If you're willing to test patches, I can try to cook up something to
> debug/fix this.
Sure, I'm always willing to test patches.