linux-kernel - Re: rx_packets/bytes stats for error frames

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:   Wed, 14 Mar 2018 10:57:43 +0100
From:   Jakob Unterwurzacher <jakob.unterwurzacher@...obroma-systems.com>
To:     Wolfgang Grandegger <wg@...ndegger.com>,
        Marc Kleine-Budde <mkl@...gutronix.de>
Cc:     Martin Elshuber <martin.elshuber@...obroma-systems.com>,
        Philipp Tomsich <philipp.tomsich@...obroma-systems.com>,
        linux-can@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: rx_packets/bytes stats for error frames

On 14.03.18 10:46, Wolfgang Grandegger wrote:
>>> We do count them, as errors!
>>>
>>> This is what happens when you transmit a single CAN frame with nothing
>>> connected: "TX errors" shoots up but "RX packets" stays zero.
>>
>> This is handled not consistent in the existing CAN drivers. In flexcan
>> all and c_can (all but rx overflow) are counted as rx_packets and
>> rx_bytes. (I haven't looked at the other drivers.)
>>
>> I tend to count the error frames as ordinary frames.
> 
> +1, I think we should count the packets and bytes delivered to the 
> network layer.

Documentation/ABI/testing/sysfs-class-net-statistics says:

> What:		/sys/class/<iface>/statistics/rx_packets
> Date:		April 2005
> KernelVersion:	2.6.12
> Contact:	netdev@...r.kernel.org
> Description:
> 		Indicates the total number of good packets received by this
> 		network device.

I considered error frames not "good packets", but I can change that for 
v3, no problem.

Best regards,
Jakob