linux-kernel - Re: [PATCH v4 REPOST 1/5] of: Add descriptions of thermtrip properties to Tegra PMC bindings

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]

Date:	Wed, 12 Nov 2014 14:07:51 +0200
From:	Mikko Perttunen <mikko.perttunen@...si.fi>
To:	Alexandre Courbot <acourbot@...dia.com>, swarren@...dotorg.org,
	thierry.reding@...il.com, gnurou@...il.com
CC:	linux-kernel@...r.kernel.org, linux-arm-kernel@...ts.infradead.org,
	linux-tegra@...r.kernel.org, wni@...dia.com,
	Mikko Perttunen <mperttunen@...dia.com>
Subject: Re: [PATCH v4 REPOST 1/5] of: Add descriptions of thermtrip properties
 to Tegra PMC bindings

On 11/11/2014 08:37 AM, Alexandre Courbot wrote:
> On 11/10/2014 10:12 PM, Mikko Perttunen wrote:
>> From: Mikko Perttunen <mperttunen@...dia.com>
>>
>> Hardware-triggered thermal reset requires configuring the I2C
>> reset procedure. This configuration is read from the device tree,
>> so document the relevant properties in the binding documentation.
>>
>> Signed-off-by: Mikko Perttunen <mperttunen@...dia.com>
>> Reviewed-by: Wei Ni <wni@...dia.com>
>> Tested-by: Wei Ni <wni@...dia.com>
>> ---
>>   .../bindings/arm/tegra/nvidia,tegra20-pmc.txt      | 24
>> ++++++++++++++++++++++
>>   1 file changed, 24 insertions(+)
>>
>> diff --git
>> a/Documentation/devicetree/bindings/arm/tegra/nvidia,tegra20-pmc.txt
>> b/Documentation/devicetree/bindings/arm/tegra/nvidia,tegra20-pmc.txt
>> index 68ac65f..dc13fb0 100644
>> --- a/Documentation/devicetree/bindings/arm/tegra/nvidia,tegra20-pmc.txt
>> +++ b/Documentation/devicetree/bindings/arm/tegra/nvidia,tegra20-pmc.txt
>> @@ -47,6 +47,21 @@ Required properties when nvidia,suspend-mode=<0>:
>>     sleep mode, the warm boot code will restore some PLLs, clocks and
>> then
>>     bring up CPU0 for resuming the system.
>>
>> +Hardware-triggered thermal reset:
>> +On Tegra30, Tegra114 and Tegra124, if the 'i2c-thermtrip' subnode
>> exists,
>> +hardware-triggered thermal reset will be enabled.
>> +
>> +Required properties for hardware-triggered thermal reset (inside
>> 'i2c-thermtrip'):
>> +- nvidia,i2c-bus : Phandle to I2C bus containing the PMU
>> +- nvidia,bus-addr : Bus address of the PMU on the I2C bus
>> +- nvidia,reg-addr : I2C register address to write poweroff command to
>> +- nvidia,reg-data : Poweroff command to write to PMU
>
> This binding is taking two different routes to provide values to the
> driver:
>
> 1) It uses a phandle for i2c-bus (which must then be provided by another
> binding implemented in the two following patches)
>
> 2) It uses direct values for bus-addr, reg-addr and reg-data.
>
> Do we need to use both approaches? bus-addr could just as well be
> obtained through a phandle to the i2c device and reading its reg
> property. From this phandle you could also go back up to the bus, making
> the i2c-bus property unnecessary. reg-addr and reg-data cannot be
> specified that way, obviously.

This was in fact how I used to implement this, but Stephen or Thierry 
pointed out that the reg property actually might not contain the correct 
address (I think because the PMIC could have multiple addresses, and the 
one in DT might not be the one that accepts the reset command).

The workaround for that was to either add this integer property for 
bus-addr or add a new PMIC API for querying. I went for this as it is 
much simpler.

>
> Actually I think I'd prefer to see i2c-bus become an integer property
> instead of a phandle, because at the end of the day it is a value field
> of a particular register and the reference is only used to retrieve that
> value. It is not like we are actually going to call functions on the bus
> instance or change its state. And for the single purpose of retrieving
> that value, this binding requires the addition of a new property on the
> bus node that will probably never be used for something else.

And this was how I used to implement this even earlier, but Thierry 
objected to that since it was duplicating information :)

>
> We can also imagine that some design may not have a PMIC device in the
> DT but still want to use the thermal reset feature. In this case, it is
> again preferable to use direct values.
>

I don't see how that'd change anything; the hardware feature still 
requires an I2C device to send a message to. If there's an I2C device on 
one of the SoC buses, you probably should have it in the DT..

Cheers,
Mikko

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/