linux-kernel - Re: [PATCH 2/2 v4] tty/serial/8250: use mctrl

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] [day] [month] [year] [list]

Date:   Wed, 5 Jun 2019 11:35:01 +0200
From:   Stefan Roese <sr@...x.de>
To:     Andy Shevchenko <andriy.shevchenko@...ux.intel.com>
Cc:     linux-serial@...r.kernel.org, linux-kernel@...r.kernel.org,
        Yegor Yefremov <yegorslists@...glemail.com>,
        Mika Westerberg <mika.westerberg@...ux.intel.com>,
        Giulio Benetti <giulio.benetti@...ronovasrl.com>,
        Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
        Johan Hovold <johan@...nel.org>
Subject: Re: [PATCH 2/2 v4] tty/serial/8250: use mctrl_gpio helpers

On 04.06.19 18:52, Andy Shevchenko wrote:
> On Mon, Jun 03, 2019 at 10:33:32AM +0200, Stefan Roese wrote:
>> From: Yegor Yefremov <yegorslists@...glemail.com>
>>
>> This patch permits the usage for GPIOs to control
>> the CTS/RTS/DTR/DSR/DCD/RI signals.
> 
>> +	if (up->gpios) {
> 
>> +		mctrl_gpio_set(up->gpios, mctrl_gpio);
>> +	}
> 
> ...
> 
>> +	if (up->gpios) {
> 
>> +		mctrl_gpio = mctrl_gpio_get_outputs(up->gpios, &mctrl_gpio);
> 
>> +	}
> 
> ...
> 
>> +			gpios = mctrl_gpio_init(&uart->port, 0);
>> +			if (IS_ERR(gpios)) {
>> +				if (PTR_ERR(gpios) != -ENOSYS)
>> +					return PTR_ERR(gpios);
>> +			}
> 
> ...
> 
>> +	if (IS_ERR_OR_NULL(mctrl_gpio_to_gpiod(up->gpios,
>> +						UART_GPIO_RTS))) {
> 
>> +	}
> 
> ...
> 
>> -	if (termios->c_cflag & CRTSCTS && up->port.flags & UPF_HARD_FLOW) {
>> +	if (termios->c_cflag & CRTSCTS && up->port.flags & UPF_HARD_FLOW
>> +		&& IS_ERR_OR_NULL(mctrl_gpio_to_gpiod(up->gpios,
>> +							UART_GPIO_RTS))) {
> 
>> }
> 
> ...
> 
>> +	if (up->gpios)
>> +		mctrl_gpio_disable_ms(up->gpios);
> 
> ...
> 
>> +	if (up->gpios)
>> +		mctrl_gpio_enable_ms(up->gpios);
> 
> ...
> 
>> +	if (up->gpios)
>> +		return mctrl_gpio_get(up->gpios, &ret);
> 
> 
> Can we rather make this mimic the gpiod_get_optional() API?
> 
> So, if we get an error, it's an error, otherwise with NULL pointer the
> operations goes to be no-op.
> 
> [IS_ERR_OR_NULL() -> IS_ERR(), if (up->gpios) -> /dev/null, etc]

So you want me to drop all "if (up->gpios)" checks? I can do this in
some cases (e.g. serial8250_disable_ms()). But I would like to keep
it in other cases, like serial8250_out_MCR(), where this check prevents
some unnecessary code execution in the "non-gpios mode" (and vice-versa).

Would this be acceptable?

BTW: Regarding the OMAP specific code: I'm not the author of this code
and I don't have access to such hardware to do some tests here. But
changing IS_ERR_OR_NULL() -> IS_ERR() in this OMAP code does not
seem correct. IIUTC, these "if" clauses are extended here by
IS_ERR_OR_NULL(mctrl_gpio_to_gpiod()) to check if the GPIO's are not
enabled / used. Currently this will probably break, since when called
with "gpios == NULL", mctrl_gpio_to_gpiod() will crash [1].

If you don't object (or have other suggestions), I'll change this to
use "up->gpios == 0" instead. This seems to be what the original author
wanted to achieve.

Okay?

Thanks,
Stefan

[1]

struct gpio_desc *mctrl_gpio_to_gpiod(struct mctrl_gpios *gpios,
				      enum mctrl_gpio_idx gidx)
{
	return gpios->gpio[gidx];
}