On 6/8/2016 6:31 PM, Rafael J. Wysocki wrote:

On Sun, May 29, 2016 at 12:01 AM, Sinan Kaya [off-list ref] wrote:

quoted

The device tree code checks for the presence of a reset driver and calls
the of_reset function pointer by looking up the reset driver as a module.

ACPI defines _RST method to perform device level reset. After the _RST
method is executed, the OS can resume using the device. _RST method is
expected to stop DMA transfers and IRQs.

This patch checks the presence of _RST method and calls the _RST
method when reset is requested.

A little bit of misinformation here. The current code is checking the presence
during probe time. If the presence of _RST method is required then probe is
aborted. Otherwise, probe will complete execution.

When reset call is to be executed, presence of _RST method is no longer checked.
Instead, the method is directly called. 

I was talking about the contribution of this patch as both here. I'll clarify
the commit message. 

You could check if _RST is present at probe time and store the ACPI
handle of it instead of the HID pointer.

This way you wouldn't need to repeat that check every time reset is used.

Based on the requirement that the code can be executed without the presence
of _RST method for development purposes, I'm hesitant to use the handle of the
reset method as a gating factor.

-- 
Sinan Kaya
Qualcomm Technologies, Inc. on behalf of Qualcomm Innovation Center, Inc.
Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum, a Linux Foundation Collaborative Project