Re: [PATCH] ACPI / LPSS: Don't skip late system PM ops for hibernate on BYT/CHT

From: Hans de Goede <hdegoede@redhat.com>
To: "Robert R. Howell" <RHowell@uwyo.edu>,
	"Rafael J. Wysocki" <rafael@kernel.org>
Cc: Kai-Heng Feng <kai.heng.feng@canonical.com>,
	"rjw@rjwysocki.net" <rjw@rjwysocki.net>,
	"lenb@kernel.org" <lenb@kernel.org>,
	"linux-acpi@vger.kernel.org" <linux-acpi@vger.kernel.org>,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH] ACPI / LPSS: Don't skip late system PM ops for hibernate on BYT/CHT
Date: Tue, 14 May 2019 12:10:59 +0200	[thread overview]
Message-ID: <0b3eaf00-1a28-441b-e1fd-dd57e118f7a3@redhat.com> (raw)
In-Reply-To: <cc1dacb5-ef0f-61c4-6187-99ee1da55d92@uwyo.edu>

Hi,

On 5/9/19 8:09 PM, Robert R. Howell wrote:
> Hi Hans
> 
> On 5/9/19 2:50 AM, Hans de Goede wrote:
> 
>>
>> Hi,
>>
>> On 09-05-19 06:24, Robert R. Howell wrote:
>>> On 4/30/19 8:39 AM, Hans de Goede wrote:
>>>>
> 
>>>>
>>>> I've just tried to reproduce the "Error i2c_dw_xfer call while suspended" error
>>>> on suspend/resume on my own T100TA and I could not reproduce this.
>>>>
>>>> Can you try without the BT keyboard paired and waking up from suspend using the
>>>> tablet part's power-button ?
>>>>
>>>> Also do you still have the scripts to rmmod some modules before suspend ?
>>>>
>>>
>>> The T100TA keyboard is actually a hardwired connection rather than Bluetooth but I
>>> did physically disconnect the keyboard, and also unpaired all the actual Bluetooth
>>> devices (such as the mouse) and then powered down the T100TA bluetooth adapter.
>>> When I suspend, then resume using the tablet power button, I still get the
>>> i2c_dw_xfererror error during the resume.  But whatever causes this error isn't fatal,
>>> in the sense that after resume the sound and other i2c functions do still work OK.
>>>
>>> While I always get this i2c_dw_xfer error on resume from suspend or hibernation on the T100TA,
>>> I also have a T100TAM and curiously, it NEVER shows that error -- although all the
>>> other suspend and hibernate behavior seems similar.  I'm not sure if the following could
>>> be the difference, but the T100TA uses an i2c connected ATML1000 touchscreen controller
>>> while the T100TAM uses an i2c connected SIS0817 touchscreen controller.  Other than that
>>> the hardware seems almost identical.
>>
>> I've been testing on an actual T100TA, with the ATML1000 touchscreen controller.
>>
>> Maybe it is a difference in BIOS version, my T100TA is running the latest BIOS, what
>> is the output of:
>>
>> cat /sys/class/dmi/id/bios_version /sys/class/dmi/id/bios_date
>>
> On the T100TA which shows the i2c_dw_xfer error the above cat reports:
> T100TA.307
> 05/09/2014

My T100TA has version T100TA.314, date 08/13/2015

> while the T100TA which does NOT show the i2c_dw_xfer error reports:
> T100TAM.205
> 07/25/2014
>>
>> Also do you perhaps have a microsd card inserted?  (I'm trying to figure out the
>> different between our setups so that I can hopefully reproduce the issue myself).
>>
> I do have a microsd card inserted in both the T100TA and the T100TAM.

Ok, so I tried reproducing the suspend/resume problem on my T100TA with a microsd
inserted and mounted and I still cannot reproduce the problem. So either you are
hitting a BIOS bug which since has been fixed, or the suspend/resume
problem is caused by you removing the DPM_FLAG_SMART_SUSPEND flag.

Regards,

Hans