linux-kernel - Re: [PATCH] hwmon: Relax name attribute validation for new APIs

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:   Tue, 31 Jan 2017 03:53:22 -0800
From:   Guenter Roeck <linux@...ck-us.net>
To:     Jean Delvare <jdelvare@...e.de>
Cc:     Hardware Monitoring <linux-hwmon@...r.kernel.org>,
        Dmitry Torokhov <dmitry.torokhov@...il.com>,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH] hwmon: Relax name attribute validation for new APIs

On 01/31/2017 01:07 AM, Jean Delvare wrote:
> Hi Guenter, Dmitry,
>
> On Fri, 27 Jan 2017 19:49:49 -0800, Guenter Roeck wrote:
>> While invalid name attributes are really not desirable and do mess up
>> libsensors, enforcing valid names has the detrimental effect of driving
>> users away from using the new hardware monitoring API, especially those
>> registering name attributes violating the ABI restrictions. Another
>> undesirable side effect is that this violation and the resulting error
>> may only be discovered some time after a conversion to the new API,
>> which in turn may trigger a revert of that conversion.
>>
>> To solve the problem, relax validation and only issue a warning instead
>> of returning an error if a name attribute violating the ABI is provided.
>> This lets callers continue to provide invalid name attributes while
>> notifying them about it.
>>
>> Many thanks are due to Dmitry Torokhov for the idea.
>>
>> Signed-off-by: Guenter Roeck <linux@...ck-us.net>
>> ---
>>  drivers/hwmon/hwmon.c | 5 +++--
>>  1 file changed, 3 insertions(+), 2 deletions(-)
>>
>> diff --git a/drivers/hwmon/hwmon.c b/drivers/hwmon/hwmon.c
>> index affff8195fff..53c54a81f7ad 100644
>> --- a/drivers/hwmon/hwmon.c
>> +++ b/drivers/hwmon/hwmon.c
>> @@ -544,9 +544,10 @@ __hwmon_device_register(struct device *dev, const char *name, void *drvdata,
>>  	struct device *hdev;
>>  	int i, j, err, id;
>>
>> -	/* Do not accept invalid characters in hwmon name attribute */
>> +	/* Complain about invalid characters in hwmon name attribute */
>>  	if (name && (!strlen(name) || strpbrk(name, "-* \t\n")))
>> -		return ERR_PTR(-EINVAL);
>> +		dev_warn(dev, "hwmon: '%s' is not a valid name attribute\n",
>> +			 name);
>
> May I suggest adding ", please fix"?
>

Ok, will do.

>>
>>  	id = ida_simple_get(&hwmon_ida, 0, 0, GFP_KERNEL);
>>  	if (id < 0)
>
> Reviewed-by: Jean Delvare <jdelvare@...e.de>
>
> Do I understand correctly that in the long run we will make it a fatal
> error again?
>

Hopefully, after there are no more drivers to convert, and after all converted
drivers fixed the problem.

Thanks,
Guenter