Re: [RFC PATCH 0/2] netdev: Add tracepoint to network/driver interface

[Koki Sanagi <sanagi.koki@jp.fujitsu.com>]

(2010/04/09 20:04), Neil Horman wrote:
> On Fri, Apr 09, 2010 at 04:37:53PM +0900, Koki Sanagi wrote:
>> These patches add tracepoints to network/driver interface.
>>
>> These tracepoints are helpful to investigate whether a packet passes or not.
>> For example, when Heart Beat is disconnected, that information is helpful
>> to investigate the cause is whether driver/device side or not.
>>
>> An output is below.
>>
>>    sshd-2443  [001] 68238.415621: netdev_start_xmit: dev=eth3 skbaddr=f3db5138 len=114
>> <idle>-0     [001] 68238.417058: netdev_receive_skb: dev=eth3 skbaddr=f3c81540 len=52
>> <idle>-0     [001] 68238.704363: netdev_receive_skb: dev=eth3 skbaddr=f3c81540 len=100
>>    sshd-2443  [001] 68238.705459: netdev_start_xmit: dev=eth3 skbaddr=f3db5138 len=114
>> <idle>-0     [001] 68238.706891: netdev_receive_skb: dev=eth3 skbaddr=f3c81540 len=52
>> <idle>-0     [001] 68238.878736: netdev_receive_skb: dev=eth3 skbaddr=f3c81540 len=100
>>    sshd-2443  [001] 68238.880361: netdev_start_xmit: dev=eth3 skbaddr=f3db5138 len=114
>>
>> As other use case I have, we can get throughput per interface with some sort of
>> perf scripts. I plan to create it.
>>
>> Thanks
>> Koki Sanagi
>>
> You can get a reasonable estimate of per-interface throughput using ethtool or
> even ifconfig in a script.  What are the tracepoints needed for that?  Don't get
> me wrong, I think these tracepoints could have some potential use thats not
> covered by other tools, I just don't see the above as a conclusive reason to add
> them.

Yeah, a script using ethtool or ifconfig can do same thing. But it must be poll
driven. By contrast this implement using tracepoint is event-driven.
It means that a record leaves on memory certainly whenever kernel crashes.
It is useful to investigate network record from dump.
This is superior to ethtool or ifconfig scripts.