[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <6703a552-8d23-6136-c0b8-c68845d00aa8@prolan.hu>
Date: Fri, 2 Sep 2022 09:35:56 +0200
From: Csókás Bence <csokas.bence@...lan.hu>
To: Florian Fainelli <f.fainelli@...il.com>, <netdev@...r.kernel.org>
CC: Richard Cochran <richardcochran@...il.com>,
"David S. Miller" <davem@...emloft.net>,
Jakub Kicinski <kuba@...nel.org>,
Francesco Dolcini <francesco.dolcini@...adex.com>,
Andrew Lunn <andrew@...n.ch>, <kernel@...gutronix.de>,
Marc Kleine-Budde <mkl@...gutronix.de>
Subject: Re: [PATCH v2] net: fec: Use a spinlock to guard `fep->ptp_clk_on`
On 2022. 09. 01. 18:26, Florian Fainelli wrote:
>
>> schedule_delayed_work(&fep->time_keep, HZ);
>> }
>> @@ -599,8 +593,6 @@ void fec_ptp_init(struct platform_device *pdev,
>> int irq_idx)
>> }
>> fep->ptp_inc = NSEC_PER_SEC / fep->cycle_speed;
>> - spin_lock_init(&fep->tmreg_lock);
>
> This change needs to be kept as there is no other code in the driver
> that would initialize the tmreg_lock otherwise. Try building a kernel
> with spinlock debugging enabled and you should see it barf with an
> incorrect spinlock bad magic.
`fec_ptp_init()` is called from `fec_probe()`, which init's the spinlock:
> @@ -3907,7 +3908,7 @@ fec_probe(struct platform_device *pdev)
> }
>
> fep->ptp_clk_on = false;
> - mutex_init(&fep->ptp_clk_mutex);
> + spin_lock_init(&fep->tmreg_lock);
Bence
Powered by blists - more mailing lists