Re: query: ethtool --show-nfc mask interpretation

[Ben Hutchings <ben@decadent.org.uk>]

[-- Attachment #1: Type: text/plain, Size: 1392 bytes --]

On Fri, 2014-07-04 at 21:43 +0530, Govindarajulu Varadarajan wrote:
> I was working on ethtool --show-nfc implementation for enic.
> 
> While displaying, ethtool displays the 1's compliment of
> ethtool_rx_flow_spec->m_u.tcp_ip4_spec.psrc

The code in ethtool was originally written to work with the n-tuple
ethtool operations (which no longer exist) in which the mask fields
specify bits to be ignored in the packet.  For NFC rule operations, the
semantics are the opposite: the mask fields specify bits to be matched
in the packet.  So ethtool inverts the mask fields as necessary.

> I could not understand how to interpret the mask output
> 
> For eg.
> [root@a163 linux-next]# ethtool -n enp9s0
> 8 RX rings available
> Total 2 rules
> 
> Filter: 0
>          Rule Type: TCP over IPv4
>          Src IP addr: 10.65.79.1 mask: 255.255.0.0
>          Dest IP addr: 10.106.186.163 mask: 255.255.255.255
>          TOS: 0x0 mask: 0x0
>          Src port: 51331 mask: 0xff
>          Dest port: 22 mask: 0xffff
>          Action: Direct to queue 4
> 
> 
> The mask for src port is 0x00ff. Does that mean 8 bits in lsb should match and
> the other 8 bits in msb should be ignored? Or is it the other way?

It means the 8 least significant bits are ignored.

Ben.

-- 
Ben Hutchings
Quantity is no substitute for quality, but it's the only one we've got.

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 811 bytes --]