linux-kernel - Re: [PATCH v3] platform/x86/hp: Avoid spurious wakeup on HP ProOne 440

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: <c1f1ab87-eda2-4570-ab00-1114d0bdade0@rowland.harvard.edu>
Date: Thu, 12 Sep 2024 11:06:06 -0400
From: Alan Stern <stern@...land.harvard.edu>
To: Kai-Heng Feng <kai.heng.feng@...onical.com>
Cc: Mathias Nyman <mathias.nyman@...ux.intel.com>, hdegoede@...hat.com,
	ilpo.jarvinen@...ux.intel.com, gregkh@...uxfoundation.org,
	jorge.lopez2@...com, acelan.kao@...onical.com,
	platform-driver-x86@...r.kernel.org, linux-kernel@...r.kernel.org,
	linux-usb@...r.kernel.org, kaihengfeng@...il.com
Subject: Re: [PATCH v3] platform/x86/hp: Avoid spurious wakeup on HP ProOne
 440

On Thu, Sep 12, 2024 at 02:28:15PM +0800, Kai-Heng Feng wrote:
> On Tue, Sep 10, 2024 at 9:13 PM Alan Stern <stern@...land.harvard.edu> wrote:
> > It should be possible for this to work.  Just make the interrupt
> > handler skip calling usb_hcd_resume_root_hub() if wakeup is not enabled
> > for the root hub getting the port-status change.  When the root hub
> > resumes as part of the system resume later on, the hub driver will check
> > and see that a connect change occurred.
> 
> This can work. But should the change be made in
> usb_hcd_resume_root_hub() or by the caller?
> The issue can potentially happen to all USB controllers, not just xHCI.

True.  However, we need to make sure that remote wakeup continues to 
work properly.  This means that the handler should skip calling 
usb_hcd_resume_root_hub() (when the root hub is suspended with wakeup = 
0) for port connect/disconnect changes or for port overcurrent changes.  
But it should _not_ skip calling usb_hcd_resume_root_hub() for port 
resume events (i.e., wakeup requests received from downstream).

usb_hcd_resume_root_hub() does not have enough information to know the 
reason for the resume; only the interrupt handler does.

Have you been following the discussion in this other email thread?

https://lore.kernel.org/linux-usb/20240906030548.845115-1-duanchenghao@kylinos.cn/

It seems very similar to the problem you are trying to fix.

Alan Stern