From: Yicong Yang <yangyicong@hisilicon.com>
To: Bjorn Helgaas <helgaas@kernel.org>,
Mika Westerberg <mika.westerberg@linux.intel.com>
Cc: Bjorn Helgaas <bhelgaas@google.com>,
Kalle Valo <kvalo@codeaurora.org>, <linux-pci@vger.kernel.org>
Subject: Re: [PATCH] PCI: Make pcie_find_root_port() work for PCIe root ports as well
Date: Wed, 1 Jul 2020 09:53:51 +0800 [thread overview]
Message-ID: <c4282c55-8312-53a0-d9e6-4818b9206c1f@hisilicon.com> (raw)
In-Reply-To: <20200630220107.GA3489322@bjorn-Precision-5520>
Hi Bjorn,
On 2020/7/1 6:01, Bjorn Helgaas wrote:
> On Mon, Jun 22, 2020 at 07:12:48PM +0300, Mika Westerberg wrote:
>> Commit 6ae72bfa656e ("PCI: Unify pcie_find_root_port() and
>> pci_find_pcie_root_port()") unified the root port finding functionality
>> into a single function but missed the fact that the passed in device may
>> already be a root port. This causes the kernel to block power management
>> of PCIe hierarchies in recent systems because ->bridge_d3 started to
>> return false for such ports after the commit in question.
>>
>> Fixes: 6ae72bfa656e ("PCI: Unify pcie_find_root_port() and pci_find_pcie_root_port()")
>> Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com>
>> Cc: stable@vger.kernel.org
>> ---
>> include/linux/pci.h | 7 ++++++-
>> 1 file changed, 6 insertions(+), 1 deletion(-)
>>
>> diff --git a/include/linux/pci.h b/include/linux/pci.h
>> index c79d83304e52..c17c24f5eeed 100644
>> --- a/include/linux/pci.h
>> +++ b/include/linux/pci.h
>> @@ -2169,8 +2169,13 @@ static inline int pci_pcie_type(const struct pci_dev *dev)
>> */
>> static inline struct pci_dev *pcie_find_root_port(struct pci_dev *dev)
>> {
>> - struct pci_dev *bridge = pci_upstream_bridge(dev);
>> + struct pci_dev *bridge;
>>
>> + /* If dev is already root port */
>> + if (pci_pcie_type(dev) == PCI_EXP_TYPE_ROOT_PORT)
>> + return dev;
>> +
>> + bridge = pci_upstream_bridge(dev);
>> while (bridge) {
>> if (pci_pcie_type(bridge) == PCI_EXP_TYPE_ROOT_PORT)
>> return bridge;
> I applied the patch below, which is slightly simplified but I think
> still equivalent, to for-linus for v5.8. Let me know if it's not.
>
> I dropped the stable tag because 6ae72bfa656e was merged for v5.8-rc1,
> and I assume v5.7 works correctly so it doesn't need any change.
>
>
> commit 5396956cc7c6 ("PCI: Make pcie_find_root_port() work for Root Ports")
> Author: Mika Westerberg <mika.westerberg@linux.intel.com>
> Date: Mon Jun 22 19:12:48 2020 +0300
>
> PCI: Make pcie_find_root_port() work for Root Ports
>
> Commit 6ae72bfa656e ("PCI: Unify pcie_find_root_port() and
> pci_find_pcie_root_port()") broke acpi_pci_bridge_d3() because calling
> pcie_find_root_port() on a Root Port returned NULL when it should return
> the Root Port, which in turn broke power management of PCIe hierarchies.
>
> Rework pcie_find_root_port() so it returns its argument when it is already
> a Root Port.
>
> [bhelgaas: test device only once, test for PCIe]
> Fixes: 6ae72bfa656e ("PCI: Unify pcie_find_root_port() and pci_find_pcie_root_port()")
> Link: https://lore.kernel.org/r/20200622161248.51099-1-mika.westerberg@linux.intel.com
> Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com>
> Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
>
> diff --git a/include/linux/pci.h b/include/linux/pci.h
> index c79d83304e52..34c1c4f45288 100644
> --- a/include/linux/pci.h
> +++ b/include/linux/pci.h
> @@ -2169,12 +2169,11 @@ static inline int pci_pcie_type(const struct pci_dev *dev)
> */
> static inline struct pci_dev *pcie_find_root_port(struct pci_dev *dev)
> {
> - struct pci_dev *bridge = pci_upstream_bridge(dev);
> -
> - while (bridge) {
> - if (pci_pcie_type(bridge) == PCI_EXP_TYPE_ROOT_PORT)
> - return bridge;
> - bridge = pci_upstream_bridge(bridge);
> + while (dev) {
> + if (pci_is_pcie(dev) &&
> + pci_pcie_type(dev) == PCI_EXP_TYPE_ROOT_PORT)
> + return dev;
> + dev = pci_upstream_bridge(dev);
> }
>
We may have some problems here, as after pcie_find_root_port() called, *dev will
be either root port or NULL but users may want it unchanged. One such usage is
in acpi_pci_bridge_d3(), drivers/pci/pci-acpi.c, *dev is used as origin
after called this.
So we should use a temporary point to *dev rather than directly modify it.
Thanks,
Yicong
> return NULL;
> .
>
next prev parent reply other threads:[~2020-07-01 1:53 UTC|newest]
Thread overview: 6+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-06-22 16:12 [PATCH] PCI: Make pcie_find_root_port() work for PCIe root ports as well Mika Westerberg
2020-06-30 22:01 ` Bjorn Helgaas
2020-07-01 1:53 ` Yicong Yang [this message]
2020-07-01 10:15 ` Mika Westerberg
2020-07-01 12:47 ` Yicong Yang
2020-07-01 10:14 ` Mika Westerberg
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=c4282c55-8312-53a0-d9e6-4818b9206c1f@hisilicon.com \
--to=yangyicong@hisilicon.com \
--cc=bhelgaas@google.com \
--cc=helgaas@kernel.org \
--cc=kvalo@codeaurora.org \
--cc=linux-pci@vger.kernel.org \
--cc=mika.westerberg@linux.intel.com \
/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
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).