linux-kernel - Re: [PATCH v4 2/9] usb: dwc3: Execute GCTL Core Soft Reset while switch modes

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]

Date:   Tue, 29 Oct 2019 11:09:47 +0200
From:   Felipe Balbi <balbi@...nel.org>
To:     John Stultz <john.stultz@...aro.org>,
        lkml <linux-kernel@...r.kernel.org>
Cc:     Yu Chen <chenyu56@...wei.com>,
        Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
        Rob Herring <robh+dt@...nel.org>,
        Mark Rutland <mark.rutland@....com>,
        ShuFan Lee <shufan_lee@...htek.com>,
        Heikki Krogerus <heikki.krogerus@...ux.intel.com>,
        Suzuki K Poulose <suzuki.poulose@....com>,
        Chunfeng Yun <chunfeng.yun@...iatek.com>,
        Hans de Goede <hdegoede@...hat.com>,
        Andy Shevchenko <andy.shevchenko@...il.com>,
        Jun Li <lijun.kernel@...il.com>,
        Valentin Schneider <valentin.schneider@....com>,
        Jack Pham <jackp@...eaurora.org>, linux-usb@...r.kernel.org,
        devicetree@...r.kernel.org, John Stultz <john.stultz@...aro.org>
Subject: Re: [PATCH v4 2/9] usb: dwc3: Execute GCTL Core Soft Reset while switch modes


Hi,

John Stultz <john.stultz@...aro.org> writes:
> From: Yu Chen <chenyu56@...wei.com>
>
> On the HiKey960, we need to do a GCTL soft reset when
> switching modes.
>
> Jack Pham also noted that in the Synopsys databook it
> mentions performing a GCTL CoreSoftReset when changing the
> PrtCapDir between device & host modes.
>
> So this patch always does a GCTL Core Soft Reset when
> changing the mode.
>
> Cc: Greg Kroah-Hartman <gregkh@...uxfoundation.org>
> Cc: Rob Herring <robh+dt@...nel.org>
> Cc: Mark Rutland <mark.rutland@....com>
> CC: ShuFan Lee <shufan_lee@...htek.com>
> Cc: Heikki Krogerus <heikki.krogerus@...ux.intel.com>
> Cc: Suzuki K Poulose <suzuki.poulose@....com>
> Cc: Chunfeng Yun <chunfeng.yun@...iatek.com>
> Cc: Yu Chen <chenyu56@...wei.com>
> Cc: Felipe Balbi <balbi@...nel.org>
> Cc: Hans de Goede <hdegoede@...hat.com>
> Cc: Andy Shevchenko <andy.shevchenko@...il.com>
> Cc: Jun Li <lijun.kernel@...il.com>
> Cc: Valentin Schneider <valentin.schneider@....com>
> Cc: Jack Pham <jackp@...eaurora.org>
> Cc: linux-usb@...r.kernel.org
> Cc: devicetree@...r.kernel.org
> Signed-off-by: Yu Chen <chenyu56@...wei.com>
> Signed-off-by: John Stultz <john.stultz@...aro.org>
> ---
> v3: Remove quirk conditional, as Jack Pham noted the
>     Synopsis databook states this should be done generally.
>     Also, at Jacks' suggestion, make the reset call before
>     changing the prtcap direction.
> ---
>  drivers/usb/dwc3/core.c | 16 ++++++++++++++++
>  1 file changed, 16 insertions(+)
>
> diff --git a/drivers/usb/dwc3/core.c b/drivers/usb/dwc3/core.c
> index 999ce5e84d3c..a039e35ec7ad 100644
> --- a/drivers/usb/dwc3/core.c
> +++ b/drivers/usb/dwc3/core.c
> @@ -112,6 +112,19 @@ void dwc3_set_prtcap(struct dwc3 *dwc, u32 mode)
>  	dwc->current_dr_role = mode;
>  }
>  
> +static void dwc3_gctl_core_soft_reset(struct dwc3 *dwc)
> +{
> +	u32 reg;
> +
> +	reg = dwc3_readl(dwc->regs, DWC3_GCTL);
> +	reg |= DWC3_GCTL_CORESOFTRESET;
> +	dwc3_writel(dwc->regs, DWC3_GCTL, reg);
> +
> +	reg = dwc3_readl(dwc->regs, DWC3_GCTL);
> +	reg &= ~DWC3_GCTL_CORESOFTRESET;
> +	dwc3_writel(dwc->regs, DWC3_GCTL, reg);
> +}
> +
>  static void __dwc3_set_mode(struct work_struct *work)
>  {
>  	struct dwc3 *dwc = work_to_dwc(work);
> @@ -154,6 +167,9 @@ static void __dwc3_set_mode(struct work_struct *work)
>  
>  	spin_lock_irqsave(&dwc->lock, flags);
>  
> +	/* Execute a GCTL Core Soft Reset when switch mode */
> +	dwc3_gctl_core_soft_reset(dwc);
> +

This is totally unnecessary. We have several platforms supporting dual
role *without* this trick. The only reason why the databook mentions a
reset is because some registers are shadowed, meaning that they share
the same physical space and just appear as different things for SW. The
reason being that Synopsys wanted to reduce the area of the IP and
decided to shadow registers which are mutually exclusive.

-- 
balbi

Download attachment "signature.asc" of type "application/pgp-signature" (833 bytes)