Re: [PATCH resend v2] PCI: pciehp: Use down_read/write_nested(reset_lock) to fix lockdep errors

[Bjorn Helgaas <helgaas@kernel.org>]

On Wed, Jan 12, 2022 at 01:39:07PM +0100, Hans de Goede wrote:
> Hi,
> 
> On 1/12/22 09:28, Lukas Wunner wrote:
> > On Tue, Jan 11, 2022 at 11:14:47AM -0600, Bjorn Helgaas wrote:
> >> On Fri, Dec 17, 2021 at 03:17:09PM +0100, Hans de Goede wrote:
> >>> Use down_read_nested() and down_write_nested() when taking the
> >>> ctrl->reset_lock rw-sem, passing the number of PCIe hotplug controllers in
> >>> the path to the PCI root bus as lock subclass parameter. This fixes the
> >>> following false-positive lockdep report when unplugging a Lenovo X1C8 from
> >>> a Lenovo 2nd gen TB3 dock:
> > [...]
> >> Applied to pci/hotplug for v5.17, thanks, Hans!
> > 
> > I've realized only now that Hans reported this issue already in August 2020
> > and opened a bugzilla for it:
> > 
> > https://bugzilla.kernel.org/show_bug.cgi?id=208855
> 
> Ah I completely forgot about having filed that bug, good catch.
> 
> > The status can now be set to RESOLVED FIXED.  I don't have permission
> > to do that but perhaps either of you, Bjorn or Hans, has?
> 
> I have added a comment and closed the bug now. Note that you can email
> the kernel.org admins with your bugzilla login + a friendly requests
> to give you some more bugzilla rights. I did that a while ago when
> I hit similar issues doing triage of bugzilla.kernel.org bugs.
> 
> > Also, the commit could optionally be amended with a Link: tag to that
> > bugzilla entry.
> 
> There isn't really any new info in the bugzilla though, so I guess
> the commit is fine as is. With that said if Bjorn wants to add it
> that is fine too of course.

Thanks, I added it.  It does have a link to Ted's original email,
which includes a complete dmesg log.

Of course, that makes the commit referenced by
https://bugzilla.kernel.org/show_bug.cgi?id=208855#c2 obsolete, but
that happens anyway because I often rebase to add things like this.

Bjorn