netdev - Re: Regression caused by commit 7bb05b85bc2d ("r8169: don't use MSI-X on RTL8106e")

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-Id: <AC238433-78C1-4FA7-894E-82CC123A7F89@canonical.com>
Date:   Wed, 12 Sep 2018 13:57:31 +0800
From:   Kai-Heng Feng <kai.heng.feng@...onical.com>
To:     Jian-Hong Pan <jian-hong@...lessm.com>
Cc:     Thomas Gleixner <tglx@...utronix.de>,
        Linux Netdev List <netdev@...r.kernel.org>,
        Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
        Daniel Drake <drake@...lessm.com>
Subject: Re: Regression caused by commit 7bb05b85bc2d ("r8169: don't use MSI-X
 on RTL8106e")

at 12:56, Jian-Hong Pan <jian-hong@...lessm.com> wrote:

> 2018-09-12 11:42 GMT+08:00 Kai-Heng Feng <kai.heng.feng@...onical.com>:
>> Hi Jian-Hong,
>>
>> There's a Dell machine with RTL8106e stops to work after S3 since the  
>> commit
>> introduced.
>> So I am wondering if it's possible to revert the commit and use
>> DMI/subsystem id based quirk table?
>>
>> It's because of commit bc976233a872 ("genirq/msi, x86/vector: Prevent
>> reservation mode for non maskable MSI") cleared the reservation mode,  
>> and I
>> can see this after S3:
>>
>> [   94.872838] do_IRQ: 3.33 No irq handler for vector
>>
>> If the device uses MSI-X instead of MSI, the issue doesn't happen  
>> because of
>> reservation mode.
>
> Interesting!  Opposite symptom!
> Could you help try the patch
> https://marc.info/?l=linux-pci&m=153629858601668&w=4 with and without
> reverting the commit?

Same issue after applying this patch. MSI-X works, MSI doesn't work.

>
> If the patch does not work, another suggestion: You can try falling
> back to only PCI_IRQ_LEGACY.

This device is capable of using MSI-X, I don't think falls back to use  
legacy is a good idea.
Instead, using a quirk table should be more appropriate.

Kai-Heng

>
> Regards,
> Jian-Hong Pan
>
>> Hi Thomas,
>>
>> Is it something should be handled by x86 BIOS? Because I don't see this
>> issue when I use Suspend-to-Idle, which doesn't use BIOS to do suspend.
>>
>> Kai-Heng