HI Dmitry: 1. You mean to let all devices ignore skipping reset/sleep part of device initialization? 2. The test team found that some old firmware will have errors (invalid report etc...), so ELAN can only ensure that the new device can meet the newer parts. Thanks jingle -----Original Message----- From: 'Dmitry Torokhov' [mailto:dmitry.torokhov@xxxxxxxxx] Sent: Friday, March 05, 2021 9:31 AM To: jingle Cc: 'linux-kernel'; 'linux-input'; 'phoenix'; 'dave.wang'; 'josh.chen' Subject: Re: [PATCH] Input: elan_i2c - Reduce the resume time for new dev ices Hi Jingle, On Fri, Mar 05, 2021 at 09:24:05AM +0800, jingle wrote: > HI Dmitry: > > In this case (in the newer parts behavior regarding need to reset > after powering them on), it is consistent with the original driver > behavior with any new or old device (be called > data->ops->initialize(client) : usleep(100) , etc.. , because this > times "data->quirks" is equal 0 at probe state.) You misunderstood my question. I was asking what specifically, if anything, was changed in the firmware to allow skipping reset/sleep part of device initialization on newer parts during resume process. Because of there were no specific changes I would say let's not do a quirk and change the driver to skip reset on resume. Thanks. -- Dmitry
