[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <9b7fac69-265e-52f2-21e4-83d9da0f257b@amd.com>
Date: Thu, 29 Sep 2022 09:08:26 +0200
From: Michal Simek <michal.simek@....com>
To: Laurent Pinchart <laurent.pinchart@...asonboard.com>,
Sai Krishna Potthuri <lakshmi.sai.krishna.potthuri@...inx.com>,
"Kannan, Arun Balaji" <arun.balaji.kannan@....com>,
"Mutthareddyvari, Jyotheeswar Reddy"
<jyotheeswar.reddy.mutthareddyvari@....com>
CC: Linus Walleij <linus.walleij@...aro.org>,
Rob Herring <robh+dt@...nel.org>,
Michal Simek <michal.simek@...inx.com>,
"Krzysztof Kozlowski" <krzysztof.kozlowski+dt@...aro.org>,
<linux-arm-kernel@...ts.infradead.org>,
<linux-kernel@...r.kernel.org>, <devicetree@...r.kernel.org>,
<linux-gpio@...r.kernel.org>, <git@...inx.com>,
<saikrishna12468@...il.com>
Subject: Re: [PATCH 0/4] pinctrl: pinctrl-zynqmp: Add tri-state configuration
support
Hi Laurent,
On 9/28/22 18:42, Laurent Pinchart wrote:
>
> On Wed, Sep 28, 2022 at 06:58:10PM +0300, Laurent Pinchart wrote:
>> Hi Sai,
>>
>> On Fri, Jun 17, 2022 at 04:16:55PM +0530, Sai Krishna Potthuri wrote:
>>> This series update the Xilinx firmware, ZynqMP dt-binding and ZynqMP
>>> pinctrl driver to handle 'output-enable' and 'bias-high-impedance'
>>> configurations. As part of these configurations, ZynqMP pinctrl driver
>>> takes care of pin tri-state setting.
>>> Also fix the kernel doc warning in ZynqMP pinctrl driver.
>>
>> I'm afraid this causes a regression :-( With this series applied, boot
>> breaks with the following message being printed to the serial console:
>>
>> Received exception
>> MSR: 0x200, EAR: 0xFF180198, EDR: 0x0, ESR: 0x64
>>
>> I've traced that to the probe of the UART, when it calls into the
>> firmware to set pin MIO18 to high impedance. According to v1.7 of the
>> ZynqMP registers reference (UG1087), there is no register at address
>> 0xFF180198.
>>
>> I am using the VCU TRD 2021.1 for testing. Does this series require a
>> firmware update ? If so backward compatibility needs to be preserved.
>> It's very late in the v6.0-rc cycle for a fix, a revert may be best at
>> this point, to give us time to fix the issue properly.
>
> I've now tested the VCU TRD 2022.1 (which AFAIK is the latest available
> version), and the problem doesn't occue then. It thus seems this depends
> on a firmware update, which is impractical at best for all old designs
> :-(
That's correct observation. Supporting these two properties requires newer pmufw
or that message is received.
I will let Arun and Jyotheeswar to comment it. I don't think there is a way to
detect which firmware has implementation for it available.
Thanks,
Michal
Powered by blists - more mailing lists