linux-kernel - Re: [PATCH v5 6/8] usb: misc: onboard_dev: add support for non-hub devices

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: <Zd93JZTlN4BCxWm7@google.com>
Date: Wed, 28 Feb 2024 18:10:45 +0000
From: Matthias Kaehlcke <mka@...omium.org>
To: Javier Carrasco <javier.carrasco@...fvision.net>
Cc: Liam Girdwood <lgirdwood@...il.com>, Mark Brown <broonie@...nel.org>,
	Rob Herring <robh+dt@...nel.org>,
	Krzysztof Kozlowski <krzysztof.kozlowski+dt@...aro.org>,
	Conor Dooley <conor+dt@...nel.org>,
	Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
	Helen Koike <helen.koike@...labora.com>,
	Maarten Lankhorst <maarten.lankhorst@...ux.intel.com>,
	Maxime Ripard <mripard@...nel.org>,
	Thomas Zimmermann <tzimmermann@...e.de>,
	David Airlie <airlied@...il.com>, Daniel Vetter <daniel@...ll.ch>,
	Catalin Marinas <catalin.marinas@....com>,
	Will Deacon <will@...nel.org>, Russell King <linux@...linux.org.uk>,
	linux-sound@...r.kernel.org, devicetree@...r.kernel.org,
	linux-kernel@...r.kernel.org, linux-usb@...r.kernel.org,
	dri-devel@...ts.freedesktop.org,
	linux-arm-kernel@...ts.infradead.org
Subject: Re: [PATCH v5 6/8] usb: misc: onboard_dev: add support for non-hub
 devices

On Wed, Feb 28, 2024 at 02:51:33PM +0100, Javier Carrasco wrote:
> Most of the functionality this driver provides can be used by non-hub
> devices as well.
> 
> To account for the hub-specific code, add a flag to the device data
> structure and check its value for hub-specific code.
> 
> The 'always_powered_in_supend' attribute is only available for hub
> devices, keeping the driver's default behavior for non-hub devices (keep
> on in suspend).
> 
> Signed-off-by: Javier Carrasco <javier.carrasco@...fvision.net>
> ---
>  drivers/usb/misc/onboard_usb_dev.c | 25 +++++++++++++++++++++++--
>  drivers/usb/misc/onboard_usb_dev.h | 10 ++++++++++
>  2 files changed, 33 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/usb/misc/onboard_usb_dev.c b/drivers/usb/misc/onboard_usb_dev.c
> index e1779bd2d126..df0ed172c7ec 100644
> --- a/drivers/usb/misc/onboard_usb_dev.c
> +++ b/drivers/usb/misc/onboard_usb_dev.c
> @@ -132,7 +132,8 @@ static int __maybe_unused onboard_dev_suspend(struct device *dev)
>  	struct usbdev_node *node;
>  	bool power_off = true;
>  
> -	if (onboard_dev->always_powered_in_suspend)
> +	if (onboard_dev->always_powered_in_suspend &&
> +	    !onboard_dev->pdata->is_hub)
>  		return 0;

With this non-hub devices would always be powered down, since
'always_powerd_in_suspend' is not set for them. This should be:

  if (!onboard_dev->pdata->is_hub ||
       onboard_dev->always_powered_in_suspend)

Checking for the (non-)hub status first is clearer IMO, also it avoids
an unneccessary check of 'always_powered' for non-hub devices.

Without code context: for hubs there can be multiple device tree nodes
for the same physical hub chip (e.g. one for the USB2 and another for
the USB3 part). I suppose this could also be the case for non-hub
devices. For hubs there is the 'peer-hub' device tree property to
establish a link between the two USB devices, as a result the onboard
driver only creates a single platform device (which is desired,
otherwise two platform devices would be in charge for power sequencing
the same phyiscal device. For non-hub devices there is currently no such
link. In many cases I expect there will be just one DT entry even though
the device has multiple USB interfaces, but it could happen and would
actually be a more accurate representation.

General support is already there (the code dealing with 'peer-hub'), but
we'd have to come up with a suitable name. 'peer-device' is the first
thing that comes to my mind, but there might be better options. If such
a generic property is added then we should deprecate 'peer-hub', but
maintain backwards compatibility.