Date:   Fri, 12 Nov 2021 15:43:31 +0200
From:   Vladimir Oltean <olteanv@gmail.com>
To:     Joakim Zhang <qiangqing.zhang@nxp.com>
Cc:     "peppe.cavallaro@st.com" <peppe.cavallaro@st.com>,
        "alexandre.torgue@foss.st.com" <alexandre.torgue@foss.st.com>,
        "joabreu@synopsys.com" <joabreu@synopsys.com>,
        "davem@davemloft.net" <davem@davemloft.net>,
        "kuba@kernel.org" <kuba@kernel.org>,
        dl-linux-imx <linux-imx@nxp.com>,
        "netdev@vger.kernel.org" <netdev@vger.kernel.org>,
        "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH net] net: stmmac: fix system hang caused by
 eee_ctrl_timer during suspend/resume
Message-ID: <20211112134331.initwvks6ytlfi4x@skbuf>
References: <20210908074335.4662-1-qiangqing.zhang@nxp.com>
 <20211111151612.i7mnye5c3trfptda@skbuf>
 <DB8PR04MB6795FCF1B1E5E0B03CA49B90E6959@DB8PR04MB6795.eurprd04.prod.outlook.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <DB8PR04MB6795FCF1B1E5E0B03CA49B90E6959@DB8PR04MB6795.eurprd04.prod.outlook.com>
Precedence: bulk

On Fri, Nov 12, 2021 at 01:21:57AM +0000, Joakim Zhang wrote:
> Hi Vladimir,
>
> > Hi Joakim,
> >
> > This patch looks strange to me. You are basically saying that the LPI timer for
> > MAC-level EEE is clocked from the clk_ptp_ref clock?! Are you sure this is
> > correct? I thought this clock is only used for the PTP timestamping counter.
> > Maybe the clock definitions in imx8mp.dtsi are not correct?
>
> No, MAC-level EEE is not clocked from the clk_ptp_ref clock, this clock is for PTP. To fix this issue,
> I can only move pm_runtime_force_suspend() into noirq suspend stage. As commit message said,
> "postpone clocks management from suspend stage to noirq suspend stage", that means I move all
> the clocks management into noirq suspend stage, including PTP clock, it's should be harmless for PTP
> function. Do you find any regression with this patch?

No, I didn't find any regression, I was just frustrated yesterday that a
bugfix patch I wanted to send to "stable" on the handling of clk_ptp_ref
is conflicting with your (apparently unnecessary) movement of this clock
to the noirq suspend stage. I'm better now.