LKML Archive on lore.kernel.org
 help / color / Atom feed
From: Pingfan Liu <kernelfans@gmail.com>
To: rjw@rjwysocki.net
Cc: linux-kernel@vger.kernel.org,
	Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	Grygorii Strashko <grygorii.strashko@ti.com>,
	Christoph Hellwig <hch@infradead.org>,
	Bjorn Helgaas <helgaas@kernel.org>,
	Dave Young <dyoung@redhat.com>,
	linux-pci@vger.kernel.org, linuxppc-dev@lists.ozlabs.org,
	linux-pm@vger.kernel.org
Subject: Re: [PATCHv3 0/4] drivers/base: bugfix for supplier<-consumer ordering in device_kset
Date: Thu, 5 Jul 2018 10:44:57 +0800
Message-ID: <CAFgQCTt7VfqM=UyCnvNFxrSw8Z6cUtAi3HUwR4_xPAc03SgHjQ@mail.gmail.com> (raw)
In-Reply-To: <6281446.GoJLz6Hq6C@aspire.rjw.lan>

On Wed, Jul 4, 2018 at 6:23 PM Rafael J. Wysocki <rjw@rjwysocki.net> wrote:
>
> On Wednesday, July 4, 2018 4:47:07 AM CEST Pingfan Liu wrote:
> > On Tue, Jul 3, 2018 at 10:36 PM Rafael J. Wysocki <rjw@rjwysocki.net> wrote:
> > >
> > > On Tuesday, July 3, 2018 8:50:38 AM CEST Pingfan Liu wrote:
> > > > commit 52cdbdd49853 ("driver core: correct device's shutdown order")
> > > > places an assumption of supplier<-consumer order on the process of probe.
> > > > But it turns out to break down the parent <- child order in some scene.
> > > > E.g in pci, a bridge is enabled by pci core, and behind it, the devices
> > > > have been probed. Then comes the bridge's module, which enables extra
> > > > feature(such as hotplug) on this bridge.
> > >
> > > So what *exactly* does happen in that case?
> > >
> > I saw the  shpc_probe() is called on the bridge, although the probing
> > failed on that bare-metal. But if it success, then it will enable the
> > hotplug feature on the bridge.
>
> I don't understand what you are saying here, sorry.
>
On the system, I observe the following:
[    2.114986] devices_kset: Moving 0004:00:00.0 to end of list
<---pcie port drive's probe, but it failed
[    2.115192] devices_kset: Moving 0004:01:00.0 to end of list
[    2.115591] devices_kset: Moving 0004:02:02.0 to end of list
[    2.115923] devices_kset: Moving 0004:02:0a.0 to end of list
[    2.116141] devices_kset: Moving 0004:02:0b.0 to end of list
[    2.116358] devices_kset: Moving 0004:02:0c.0 to end of list
[    3.181860] devices_kset: Moving 0004:03:00.0 to end of list
<---the ata disk controller which sits behind the bridge
[   10.267081] devices_kset: Moving 0004:00:00.0 to end of list
 <---shpc_probe() on this bridge, failed too.

As you can the the parent device "0004:00:00.0" is moved twice, and
finally, it is after the "0004:03:00.0", this will break the
"parent<-child" order in devices_kset. This is caused by the code
really_probe()->devices_kset_move_last(). Apparently, it makes
assumption that child device's probing comes after its parent's. But
it does not stand up in the case.

> device_reorder_to_tail() walks the entire device hierarchy below the target
> and moves all of the children in there *after* their parents.
>
As described, the bug is not related with device_reorder_to_tail(), it
is related with really_probe()->devices_kset_move_last(). So [2/4]
uses different method to achieve the "parent<-child" and
"supplier<-consumer" order. The [3/4] clean up some code in
device_reorder_to_tail(), since I need to revert the commit.

> How can it break "the parent <- child order" then?
>
As described, it does not, just not be in use any longer.

Thanks and regards,
Pingfan

  reply index

Thread overview: 51+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-07-03  6:50 Pingfan Liu
2018-07-03  6:50 ` [PATCHv3 1/4] drivers/base: fold the routine of device's shutdown into a func Pingfan Liu
2018-07-03  6:50 ` [PATCHv3 2/4] drivers/base: utilize device tree info to shutdown devices Pingfan Liu
2018-07-03  7:51   ` Lukas Wunner
2018-07-03  9:26     ` Pingfan Liu
2018-07-04  3:10       ` Pingfan Liu
2018-07-03 10:58   ` Andy Shevchenko
2018-07-03 17:03     ` Pavel Tatashin
2018-07-04 17:04   ` kbuild test robot
2018-07-05 10:11   ` Rafael J. Wysocki
2018-07-06  3:02     ` Pingfan Liu
2018-07-06  9:53       ` Rafael J. Wysocki
2018-07-07  4:02         ` Pingfan Liu
2018-07-06 10:00       ` [PATCH] driver core: Drop devices_kset_move_last() call from really_probe() Rafael J. Wysocki
2018-07-09 13:57         ` Bjorn Helgaas
2018-07-09 21:35           ` Rafael J. Wysocki
2018-07-09 22:06             ` Bjorn Helgaas
2018-07-10  6:19               ` Kishon Vijay Abraham I
2018-07-10 10:32                 ` Rafael J. Wysocki
2018-07-10 10:29               ` Rafael J. Wysocki
2018-07-10  6:33         ` Pingfan Liu
2018-07-10 11:35         ` [PATCH] driver core: Partially revert "driver core: correct device's shutdown order" Rafael J. Wysocki
2018-07-10 12:22           ` Kishon Vijay Abraham I
2018-07-10 12:38             ` Rafael J. Wysocki
2018-07-10 12:51           ` [PATCH v2] " Rafael J. Wysocki
2018-07-10 12:59             ` Greg Kroah-Hartman
2018-07-10 15:40               ` Rafael J. Wysocki
2018-07-10 15:47                 ` Greg Kroah-Hartman
2018-07-10 19:13                   ` Kishon Vijay Abraham I
2018-07-03  6:50 ` [PATCHv3 3/4] drivers/base: clean up the usage of devices_kset_move_last() Pingfan Liu
2018-07-03 14:26   ` Rafael J. Wysocki
2018-07-04  4:40     ` Pingfan Liu
2018-07-04 10:17       ` Rafael J. Wysocki
2018-07-05  2:32         ` Pingfan Liu
2018-07-03  6:50 ` [PATCHv3 4/4] Revert "driver core: correct device's shutdown order" Pingfan Liu
2018-07-03 14:35 ` [PATCHv3 0/4] drivers/base: bugfix for supplier<-consumer ordering in device_kset Rafael J. Wysocki
2018-07-04  2:47   ` Pingfan Liu
2018-07-04 10:21     ` Rafael J. Wysocki
2018-07-05  2:44       ` Pingfan Liu [this message]
2018-07-05  9:18         ` Rafael J. Wysocki
2018-07-06  8:36           ` Lukas Wunner
2018-07-06  8:47             ` Rafael J. Wysocki
2018-07-06 13:55               ` Pingfan Liu
2018-07-07  4:24                 ` Pingfan Liu
2018-07-08  8:25                   ` Rafael J. Wysocki
2018-07-09  6:48                     ` Pingfan Liu
2018-07-09  7:48                       ` Rafael J. Wysocki
2018-07-09  8:40                         ` Pingfan Liu
2018-07-09  8:58                           ` Rafael J. Wysocki
2018-07-06 10:02             ` Kishon Vijay Abraham I
2018-07-06 13:52             ` Pingfan Liu

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to='CAFgQCTt7VfqM=UyCnvNFxrSw8Z6cUtAi3HUwR4_xPAc03SgHjQ@mail.gmail.com' \
    --to=kernelfans@gmail.com \
    --cc=dyoung@redhat.com \
    --cc=gregkh@linuxfoundation.org \
    --cc=grygorii.strashko@ti.com \
    --cc=hch@infradead.org \
    --cc=helgaas@kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-pci@vger.kernel.org \
    --cc=linux-pm@vger.kernel.org \
    --cc=linuxppc-dev@lists.ozlabs.org \
    --cc=rjw@rjwysocki.net \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link

LKML Archive on lore.kernel.org

Archives are clonable:
	git clone --mirror https://lore.kernel.org/lkml/0 lkml/git/0.git
	git clone --mirror https://lore.kernel.org/lkml/1 lkml/git/1.git
	git clone --mirror https://lore.kernel.org/lkml/2 lkml/git/2.git
	git clone --mirror https://lore.kernel.org/lkml/3 lkml/git/3.git
	git clone --mirror https://lore.kernel.org/lkml/4 lkml/git/4.git
	git clone --mirror https://lore.kernel.org/lkml/5 lkml/git/5.git
	git clone --mirror https://lore.kernel.org/lkml/6 lkml/git/6.git
	git clone --mirror https://lore.kernel.org/lkml/7 lkml/git/7.git
	git clone --mirror https://lore.kernel.org/lkml/8 lkml/git/8.git
	git clone --mirror https://lore.kernel.org/lkml/9 lkml/git/9.git

	# If you have public-inbox 1.1+ installed, you may
	# initialize and index your mirror using the following commands:
	public-inbox-init -V2 lkml lkml/ https://lore.kernel.org/lkml \
		linux-kernel@vger.kernel.org
	public-inbox-index lkml

Example config snippet for mirrors

Newsgroup available over NNTP:
	nntp://nntp.lore.kernel.org/org.kernel.vger.linux-kernel


AGPL code for this site: git clone https://public-inbox.org/public-inbox.git