Re: [PATCH 0/2] CPU Package temperarure thermal driver

[Eduardo Valentin <eduardo.valentin@ti.com>]

[-- Attachment #1: Type: text/plain, Size: 3975 bytes --]

On 07-05-2013 14:57, Srinivas Pandruvada wrote:
> This driver register CPU digital temperature package level sensor as a
> thermal zone with two user mode configurable trip points. Once
> the trip point is violated, user mode can receive notification via thermal
> notification mechanism and can take any action to control temperature.


So, you have an IRQ that will be translated into a thermal event to
userland. Userland takes care of cooling the device. Is that correct?

> 
> Background:
> This set of changes were done to coretemp driver and posted to lm_sensors
> mailing list on 04/04/2013. This was reviewed by Guenter Roeck from lm-sensors
> and Zhang Rui (thermal maintainer). They were in agreement not to add notification
> mechanism to coretemp driver but use thermal sysfs.
> Guenter Roeck suggested to use approach like "db8500_thermal driver in drivers/thermal".
> So resubmitting the driver as a thermal zone driver.
> Previous discussion link:
> http://comments.gmane.org/gmane.linux.drivers.sensors/32182
> " 
> This is clear that there is reluctance in adding thresholds in coretemp sysfs,
> during previous attempts. Probably because of lake of use cases.
> But this time use case may be more compelling.
> 
> We have many small form factor devices like ultrabooks, slate PCs in the market.
> Unfortunately these devices reach maximum temperature with relatively less
> workloads, causing BIOS to do thermal throttling. There are real performance
> issues due to aggressive BIOS action to control thermals and also thermal breakdown
> in some cases.
> 
> Even the most expensive laptops, don't have correct ACPI thermal configuration,
> so that kernel thermal driver can act. In some case even the trip point is higher
> than critical temperature setting.
> 

So, this driver is meant for ACPI capable systems, but with incorrect
ACPI thermal configuration. Is my understanding correct? Does it apply
to other types of systems?



> Intel has developed several drivers, which can be used to cool the system very efficiently.
> They include RAPL based cooling driver, Powerclamp driver and P state driver.
> To utilize these cooling device a closed loop user mode program is required, which
> will utilize these method and dynamically compensate for high CPU temperatures,
> without relying on any configuration data.
> One such solution is developed is "Linux thermal daemon". More details can be
> obtained from 
> "https://github.com/01org/thermal_daemon/blob/master/ThermalDaemon_Introduction.pdf".
> This daemon polls for cpu temperature and apply compensation once the CPU reach target
> temperature. 
> 
> This polling can be mostly avoided, by getting notification for the temperature, where
> it needs to wake up and get ready for apply compensation. In most of the normal use 
> cases, there may not be any threshold events. So very minimal number of user space
> notification for thermal thresholds.
> "
> 

IMO, interrupt based TDM can be extended to non ACPI systems in most
cases. This is based on the fact that most temperature sensors for
embedded systems provide int high/low thresholds for triggering IRQs.
Not only Bandgaps but also I2C devices. Thus, avoiding polling is a
common target.

If the design can be generalized enough to cope with improving the
framework, I believe is a better way to go.

> 
> Srinivas Pandruvada (2):
>   x86, mcheck, therm_throt: Process package thresholds
>   Thermal: CPU Package temperature thermal
> 
>  arch/x86/include/asm/mce.h               |   7 +
>  arch/x86/kernel/cpu/mcheck/therm_throt.c |  63 ++-
>  drivers/thermal/Kconfig                  |  12 +
>  drivers/thermal/Makefile                 |   2 +-
>  drivers/thermal/x86_pkg_temp_thermal.c   | 633 +++++++++++++++++++++++++++++++
>  5 files changed, 712 insertions(+), 5 deletions(-)
>  create mode 100644 drivers/thermal/x86_pkg_temp_thermal.c
> 



[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 295 bytes --]