linux-kernel - Re: [PATCH 1/4] Introduce regmap infrastructure over Maxim/Dalas OneWire (W1) bus

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:   Fri, 26 May 2017 18:50:25 +0300
From:   "Alex A. Mihaylov" <minimumlaw@...bler.ru>
To:     Mark Brown <broonie@...nel.org>
Cc:     Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
        Sebastian Reichel <sre@...nel.org>,
        Evgeniy Polyakov <zbr@...emap.net>,
        linux-kernel@...r.kernel.org, linux-pm@...r.kernel.org
Subject: Re: [PATCH 1/4] Introduce regmap infrastructure over Maxim/Dalas
 OneWire (W1) bus


> This looks mostly fine, a couple of small things and like I said in
> reply to Greg please use subject lines matching the style for the
> subsystem - this makes it a lot easier for people to identify relevant
> patches.
>
>> +	int ret = -ENODEV;
>> +
>> +
>> +	if (reg > 255)
>> +		return -EINVAL;
>> +
>> +	mutex_lock(&sl->master->bus_mutex);
>> +	if (!w1_reset_select_slave(sl)) {
>> +		w1_write_8(sl->master, W1_CMD_READ_DATA);
>> +		w1_write_8(sl->master, reg);
>> +		*val = w1_read_8(sl->master);
>> +		ret = 0;
>> +	}
>> +	mutex_unlock(&sl->master->bus_mutex);
> This is a bit confusing with how -ENODEV is generated - move the
> assignment into the if statement so it doesn't look like we're silently
> ignoring errors unless you look back to the top of the function.
Ok. I set default return value to -ENODEV. W1 (OneWire) Bus periodically 
scan for connected devices. Typical time between scans about 60 sec. 
This period W1 slave device can present in kernel device list, but will 
physically disconnected.

Only w1_reset_select_slave(sl) can say me about device still physically 
accessible. Only on success w1_reset_select_slave return zero. I think 
code, like

if (!w1_reset_select_slave(sl)) {
   [...]
   ret =0;
} else
   ret = -ENODEV;

not good.

>> +static struct regmap_bus regmap_w1_bus_a8_v16 = {
>> +	.reg_read = w1_reg_a8_v16_read,
>> +	.reg_write = w1_reg_a8_v16_write,
>> +};
> It'd be clearer to just have all all these structs at the end of the set
> of functions rather than scattered about randomly.
Ok. I move this structs down in next version of patches.

I hope that the next edition will not contain such a large number of 
registration errors. Sorry for all.