Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Date:   Thu, 15 Jun 2023 09:39:09 -0500
From:   Nishanth Menon <nm@ti.com>
To:     Francesco Dolcini <francesco@dolcini.it>
CC:     Benjamin Bara <bbara93@gmail.com>, Wolfram Sang <wsa@kernel.org>,
        Lee Jones <lee@kernel.org>, <rafael.j.wysocki@intel.com>,
        <dmitry.osipenko@collabora.com>, <peterz@infradead.org>,
        <jonathanh@nvidia.com>, <richard.leitner@linux.dev>,
        <treding@nvidia.com>, <linux-kernel@vger.kernel.org>,
        <linux-i2c@vger.kernel.org>, <linux-tegra@vger.kernel.org>,
        Benjamin Bara <benjamin.bara@skidata.com>,
        <stable@vger.kernel.org>
Subject: Re: [PATCH v6 1/5] kernel/reboot: emergency_restart: set correct
 system_state
Message-ID: <20230615143909.fjr25wghwrtepk6g@unwieldy>
References: <20230327-tegra-pmic-reboot-v6-0-af44a4cd82e9@skidata.com>
 <20230327-tegra-pmic-reboot-v6-1-af44a4cd82e9@skidata.com>
 <20230615000650.coyphnwdai7smww7@unblended>
 <ZIsQPutiZWACawec@francesco-nb.int.toradex.com>
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Disposition: inline
In-Reply-To: <ZIsQPutiZWACawec@francesco-nb.int.toradex.com>
Precedence: bulk

On 15:21-20230615, Francesco Dolcini wrote:
> > > 
> > > -- 
> > > 2.34.1
> > > 
> > 
> > Tested-by: Nishanth Menon <nm@ti.com>
> > 
> > This in addition to a deeper bug in our driver seems to have helped
> > resolve a report we had been looking at. Tested on beagleplay platform
> > 
> > https://lore.kernel.org/all/ZGeHMjlnob2GFyHF@francesco-nb.int.toradex.com/
> 
> Is this patch going to fix the RCU warning I reported on that email or
> it is just part of a more complex solution?

From what I see, It is part of the solution.
Problem happens as follows for us:
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/drivers/firmware/ti_sci.c#n421

When i2c is not that frequently used, runtime pm disables the power
domain on our platform. As part of reset or power-off, when i2c is
invoked, it ends up calling into the firmware handler which (no
surprise), attempts to do the wrong thing (and rightly flagged by RCU).

We are in the middle of trying various combinations out to ensure we
are'nt messing things up.

-- 
Regards,
Nishanth Menon
Key (0xDDB5849D1736249D) / Fingerprint: F8A2 8693 54EB 8232 17A3  1A34 DDB5 849D 1736 249D