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] [day] [month] [year] [list]
Date:   Thu, 6 Sep 2018 06:33:07 +0000
From:   Minas Harutyunyan <Minas.Harutyunyan@...opsys.com>
To:     Stefan Wahren <stefan.wahren@...e.com>,
        Minas Harutyunyan <Minas.Harutyunyan@...opsys.com>
CC:     Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
        "linux-usb@...r.kernel.org" <linux-usb@...r.kernel.org>,
        Doug Anderson <dianders@...omium.org>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH RFC] usb: dwc2: power off during shutdown

Hi Stefan,

On 9/5/2018 10:42 PM, Stefan Wahren wrote:
> Hi Minas,
> 
>> Minas Harutyunyan <Minas.Harutyunyan@...opsys.com> hat am 5. September 2018 um 10:37 geschrieben:
>>
>>
>> Hi Stefan,
>>
>> On 9/1/2018 1:55 PM, Stefan Wahren wrote:
>>> Currently USB consumes a lot of power after shutting down a Raspberry Pi 3
>>> (example setup with Ethernet and a keyboard connected). So power off
>>> USB on shutdown.
>>>
>>> Measured on Raspberry Pi 3B with 4.19rc1/multi_v7_defconfig (HDMI, Ethernet
>>> and a keyboard connected)
>>>
>>> Before patch: 2.450 W
>>> After patch:  2.090 W
>>>
>>
>> Could you please elaborate. Power measurements done after shutdown?
> 
> I did the following:
> 
> sudo halt
> wait until "System halted" appears on the screen
> "measured" the current value displayed on my PSU
> 
> You are right that not the same as poweroff. But in even in poweroff the board consumes 0.6 W, but in this case USB is powered down.
> 
>> Why
>> your setup continue consume power after shutdown?
> 
> The BCM2835 SoC has a GPU and a ARM core. Linux is running only on the ARM core. The GPU is usually running a close source firmware, which is started at first during boot und runs all the time.
> 
>> Maybe its
>> sleep/hibernation mode?
> 
> There isn't much implementation of power management for BCM2835 yet. So i wanted to start with USB.
> Even for sleep/hibernation the consumption is much too high. Because the system consumes in idle 1.6 W.
> 
>>
>> dwc2 core in your platform which type of power optimization supports:
>> partial power down, hibernation or extended hibernation? Please provide
>> me GHWCFG4 register value of your core.
> 
> AFAIK all RPis use the same USB IP core:
> 
> GHWCFG4 = 0x1ff00020

Ok, your core doesn't support any power optimization mode.


> 
> Stefan
> 

1. Could you please explain what is real/physical cause of power saving 
after your patch: VBUS off, PHY clock stop or any other?
2. Can USB detect remote wake-up signaling from connected 
keyboard/ethernet before and after your patch?

Thanks,
Minas

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ