linux-kernel - Re: [PATCH v3 02/11] pwm: imx: remove ipg clock

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20161123094305.75f1eb7a@bbrezillon>
Date:   Wed, 23 Nov 2016 09:43:05 +0100
From:   Boris Brezillon <boris.brezillon@...e-electrons.com>
To:     Stefan Agner <stefan@...er.ch>
Cc:     Lukasz Majewski <l.majewski@...ess.pl>,
        Thierry Reding <thierry.reding@...il.com>,
        Sascha Hauer <s.hauer@...gutronix.de>,
        linux-pwm@...r.kernel.org, linux-kernel@...r.kernel.org,
        Fabio Estevam <fabio.estevam@....com>,
        Fabio Estevam <festevam@...il.com>,
        Lothar Wassmann <LW@...o-electronics.de>,
        Bhuvanchandra DV <bhuvanchandra.dv@...adex.com>,
        kernel@...gutronix.de, Philipp Zabel <p.zabel@...gutronix.de>
Subject: Re: [PATCH v3 02/11] pwm: imx: remove ipg clock

On Tue, 22 Nov 2016 13:04:11 -0800
Stefan Agner <stefan@...er.ch> wrote:

> On 2016-11-01 00:10, Lukasz Majewski wrote:
> > From: Sascha Hauer <s.hauer@...gutronix.de>
> > 
> > The use of the ipg clock was introduced with commit 7b27c160c681
> > ("pwm: i.MX: fix clock lookup").
> > In the commit message it was claimed that the ipg clock is enabled for
> > register accesses. This is true for the ->config() callback, but not
> > for the ->set_enable() callback. Given that the ipg clock is not
> > consistently enabled for all register accesses we can assume that either
> > it is not required at all or that the current code does not work.
> > Remove the ipg clock code for now so that it's no longer in the way of
> > refactoring the driver.
> > 
> > Signed-off-by: Sascha Hauer <s.hauer@...gutronix.de>
> > Cc: Philipp Zabel <p.zabel@...gutronix.de>  
> 
> I have to NACK here, sorry guys.
> 
> Just tested this on a i.MX 7, the kernel freezes in imx_pwm_config, I
> guess that is where the code accesses a register first.
> 
> The i.MX 7 DT (imx7s.dtsi) specifies the same clock for ipg and per, but
> it seems that this clock is crucial for register access on i.MX 7:
> 
> clocks = <&clks IMX7D_PWM1_ROOT_CLK>,
>          <&clks IMX7D_PWM1_ROOT_CLK>;
> clock-names = "ipg", "per";          
> 
> So since the "per" clock is the same in the i.MX 7 case, imx_pwm_enable
> worked...
> 
> I agree that the old code is a bit weird, especially that we get the
> clock in imx_pwm_enable. It seems that all device trees specify a "ipg"
> clock, so I guess we can get the clock at probe time for all variants of
> this IP and just enable it on peripheral access...

Or, we patch the code to take the per clk before accessing PWM regs,
and release it once we're done.
AFAIU, the IPG clock is only supposed to be enabled when the PWM takes
its sources from the IPG channel, it has nothing to do we register
accesses. If this is correct, then doing what you suggest implies
abusing the IPG clk meaning.

> 
> --
> Stefan
> 
> 
> > ---
> > [commit message text refactored by Lukasz Majewski <l.majewski@...ess.pl>]
> > ---
> > Changes for v3:
> > - New patch
> > ---
> >  drivers/pwm/pwm-imx.c | 19 +------------------
> >  1 file changed, 1 insertion(+), 18 deletions(-)
> > 
> > diff --git a/drivers/pwm/pwm-imx.c b/drivers/pwm/pwm-imx.c
> > index d600fd5..70609ef2 100644
> > --- a/drivers/pwm/pwm-imx.c
> > +++ b/drivers/pwm/pwm-imx.c
> > @@ -49,7 +49,6 @@
> >  
> >  struct imx_chip {
> >  	struct clk	*clk_per;
> > -	struct clk	*clk_ipg;
> >  
> >  	void __iomem	*mmio_base;
> >  
> > @@ -204,17 +203,8 @@ static int imx_pwm_config(struct pwm_chip *chip,
> >  		struct pwm_device *pwm, int duty_ns, int period_ns)
> >  {
> >  	struct imx_chip *imx = to_imx_chip(chip);
> > -	int ret;
> > -
> > -	ret = clk_prepare_enable(imx->clk_ipg);
> > -	if (ret)
> > -		return ret;
> >  
> > -	ret = imx->config(chip, pwm, duty_ns, period_ns);
> > -
> > -	clk_disable_unprepare(imx->clk_ipg);
> > -
> > -	return ret;
> > +	return imx->config(chip, pwm, duty_ns, period_ns);
> >  }
> >  
> >  static int imx_pwm_enable(struct pwm_chip *chip, struct pwm_device *pwm)
> > @@ -293,13 +283,6 @@ static int imx_pwm_probe(struct platform_device *pdev)
> >  		return PTR_ERR(imx->clk_per);
> >  	}
> >  
> > -	imx->clk_ipg = devm_clk_get(&pdev->dev, "ipg");
> > -	if (IS_ERR(imx->clk_ipg)) {
> > -		dev_err(&pdev->dev, "getting ipg clock failed with %ld\n",
> > -				PTR_ERR(imx->clk_ipg));
> > -		return PTR_ERR(imx->clk_ipg);
> > -	}
> > -
> >  	imx->chip.ops = &imx_pwm_ops;
> >  	imx->chip.dev = &pdev->dev;
> >  	imx->chip.base = -1;