linux-kernel - Re: [RFC PATCH 2/2] serial: 8250

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: <a97eeee3-07a0-5ca6-edc4-f5550ca4acbf@huawei.com>
Date:   Mon, 26 Feb 2018 10:54:06 +0000
From:   John Garry <john.garry@...wei.com>
To:     Andy Shevchenko <andriy.shevchenko@...ux.intel.com>,
        <gregkh@...uxfoundation.org>, <slaby@...e.com>,
        <p.zabel@...gutronix.de>, <heiko@...ech.de>,
        <ed.blake@...drel.com>, <jhogan@...nel.org>
CC:     <linux-serial@...r.kernel.org>, <linux-kernel@...r.kernel.org>,
        <linuxarm@...wei.com>
Subject: Re: [RFC PATCH 2/2] serial: 8250_dw: support polling mode

On 23/02/2018 17:35, Andy Shevchenko wrote:
> On Fri, 2018-02-23 at 02:42 +0800, John Garry wrote:
>> It would be useful to make this driver support some
>> 8250-compatible devices which have no interrupt line.
>>
>> For these, we allow for no interrupt, and will fallback on
>> polling mode.
>>
>> Note: the 8250 dt bindings state that "interrupts"
>> is a required property:
>> "interrupts : should contain uart interrupt."
>>
>> But the 8250_of.c driver can live without it. So
>> this patch is going this way also.
>
> It should be documented in the binding.

Agreed. I find the wording a bit ambigious in bindings/serial/8250.txt 
and snps-dw-apb-uart.txt:

Required properties:
[...]
- interrupts : should contain uart interrupt.

It uses the word "should", and not "must". Maybe "should" means "must"...

>
>>  	if (irq < 0) {
>> -		if (irq != -EPROBE_DEFER)
>> -			dev_err(dev, "cannot get irq\n");
>> -		return irq;
>> +		if (irq == -EPROBE_DEFER)
>> +			return irq;
>> +		dev_warn(dev, "cannot get irq, using polling
>> mode\n");
>> +		irq = 0;
>
> NO_IRQ _is_ 0. You need to do something like
>
> if (irq < 0) }
>  ... leave existing code ...
> }
>
> if (!irq)
>  dev_warn(, "Use polling mode\n");

OK

>

Thanks,
John