Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Content-Type: text/plain;
        charset=us-ascii
Mime-Version: 1.0 (Mac OS X Mail 13.0 \(3601\))
Subject: Re: [PATCH v2 2/2] PCI: pciehp: Prevent deadlock on disconnect
From:   Kai-Heng Feng <kai.heng.feng@canonical.com>
In-Reply-To: <20190923082832.GD2773@lahna.fi.intel.com>
Date:   Fri, 18 Oct 2019 15:10:32 +0800
Cc:     AceLan Kao <acelan.kao@canonical.com>,
        Lukas Wunner <lukas@wunner.de>,
        Bjorn Helgaas <bhelgaas@google.com>,
        "Rafael J. Wysocki" <rjw@rjwysocki.net>,
        Keith Busch <keith.busch@intel.com>,
        Andy Shevchenko <andriy.shevchenko@linux.intel.com>,
        Frederick Lawler <fred@fredlawl.com>,
        "Gustavo A . R . Silva" <gustavo@embeddedor.com>,
        Sinan Kaya <okaya@kernel.org>,
        Linux PCI <linux-pci@vger.kernel.org>,
        lkml <linux-kernel@vger.kernel.org>
Content-Transfer-Encoding: 8BIT
Message-Id: <0DDD8724-5331-495C-BB20-3754B1E667D3@canonical.com>
References: <20190812143133.75319-1-mika.westerberg@linux.intel.com>
 <20190812143133.75319-2-mika.westerberg@linux.intel.com>
 <20190923053403.jdjw6ed3sub6iuou@wunner.de>
 <20190923081237.GB2773@lahna.fi.intel.com>
 <20190923082832.GD2773@lahna.fi.intel.com>
To:     Mika Westerberg <mika.westerberg@linux.intel.com>
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

Hi Mika,

> On Sep 23, 2019, at 16:28, Mika Westerberg <mika.westerberg@linux.intel.com> wrote:
> 
> On Mon, Sep 23, 2019 at 11:12:42AM +0300, Mika Westerberg wrote:
>> Regarding suggestion of unbinding PCI drivers without
>> pci_lock_rescan_remove() hold, I haven't looked it too closely but I
>> think we need to take that lock anyway because when we are unbinding a
>> hotplug driver it is supposed to remove the hierarchy below touching the
>> shared structures, possibly concurrently. Unfortunately there is no
>> documentation what data pci_lock_rescan_remove() actually protects so
>> first one needs to understand that. I think one way to clean up this is
>> to use finer grained locking (with documented lock ordering) for PCI bus
>> structures that can be accessed simultaneusly by different threads. But
>> that is not a simple task.
> 
> Now that I looked more closely, I realized it actually is not supposed
> to remove the hierarchy below so indeed it might be possible to do that
> without taking pci_lock_rescan_remove().

This series fixes S3 resume hang when native PCIe TBT is connected to TBT dock.
Is there going to be a v3 of this series?

Anyway, please collect my tested-by tag,
Tested-by: Kai-Heng Feng <kai.heng.feng@canonical.com>

Kai-Heng